Dara Massicot is a military analyst at the Carnegie Endowment for International Peace, who previously worked for the Department of Defense and the Rand Corporation. Learn more about Ukraine’s defense tech sector in this recent piece in American Purpose.

You can watch the full conversation below:

All episodes of the Frankly Fukuyama podcast are available here.

Podcast production by Ringo Harrison.

Connect with us!

X: @FukuyamaFrancis & @JoinPersuasion

YouTube: Frankly Fukuyama & Persuasion

We’re delighted to feature this article as part of our series on Liberal Virtues and Values.

That liberalism is under threat is now a cliché—yet this has done nothing to stem the global resurgence of illiberalism. Part of the problem is that liberalism is often considered too “thin” to win over the allegiance of citizens, and that liberals are too afraid of speaking in moral terms. Liberalism’s opponents, by contrast, speak to people’s passions and deepest moral sentiments.

This series, made possible with the generous support of the John Templeton Foundation, aims to change that narrative. In podcast conversations and long-form pieces, we’ll feature content making the case that liberalism has its own distinctive set of virtues and values that are capable not only of responding to the dissatisfaction that drives authoritarianism, but also of restoring faith in liberalism as an ideology worth believing in—and defending—on its own terms.

It’s no coincidence, it seems to me, that the decline of liberal democracy, as a fact and value, has succeeded the decline of liberal education as a fact and value. If we are ever to revive the first, an essential step will be to resurrect the second. The two “liberals,” after all, are the same. They refer to political liberty, as understood by ancient Athens, republican Rome, the American Founders: not libertarian freedom from individual constraint but collective self-government by civic equals. Its opposite is tyranny, arbitrary rule by a single will, a dispensation we’re becoming more familiar with than we had ever thought we’d be.

Liberal education is that form of education that prepares individuals for the exercise of political liberty—in other words, for citizenship. (Its opposite, in Aristotle’s account, is servile education, that which aims at mere utility, the performance of an economic function.) For generations, its importance was a governing idea in American higher education. In 1945, to pick a single milestone, Harvard published what became a widely influential volume, General Education in a Free Society (known from its color as the Redbook)—a pedagogical program, as the war neared its end, for the emerging era of mass political participation. “A republic, if you can keep it,” said Benjamin Franklin, and liberal education, which the Founders also championed, is part of how you keep it.

But citizenship, too, is a concept in long-term decline (along with republic, for that matter). On campus, as a goal of education, it has given way to mere utility, salaried servility, veiled, at selective schools, beneath the drapery of “social justice,” the language of changing the world, which bids young people be not citizens but activists.

Yet to imagine oneself as an activist is, in important respects, the reverse of regarding oneself as a citizen. The two entail divergent aims, virtues, attitudes about this country that we share. An activist is a soldier in a social or cultural war. A citizen is a member of a political community, a group of individuals who recognize that they have responsibilities to one another. Activism divides: us versus them, the good guys and the bad guys. Citizenship unites: we speak of “fellow citizens” or “fellow Americans.” Activists see those who oppose them as enemies to be defeated and, ideally, eliminated, if only through reeducation (though also, more and more, through violence). Citizenship demands toleration, the acknowledgment that even those you hate the most possess an equal share with you in the political collective: an equal right to speak, vote, advocate, educate, organize, assemble, and, if elected, govern. Activists say, go away; citizens say, we’re all in this together, dammit.

What would it look like to restore the idea of citizenship to the center of undergraduate education? What does liberal education entail? Two things: what to teach and how.

The what is named in Harvard’s title: “general education.” In other words, that which every student needs to know, regardless of what they major—that is, specialize—in. It seems obvious to me that that which every student needs to know as a citizen (and, a fortiori, at elite schools, a future leader) of this country is American, Western, and, to a lesser extent, global literature, philosophy, and (something often overlooked in core curricula) history. These compose the basic grammar of our common thought and culture, including, or especially, our lazy hot-take half-thought and our derivative-schlock popular culture. How much can’t you understand, of our collective imagination, if you do not know the story of the Exodus; of our political contentions, if you haven’t studied the development of the concept of equality in Western thought; of international affairs, if you aren’t familiar on at least a basic level with the history of European nationalism?

It’s incredible, to start with, that one can graduate from college in this country (that one can graduate from high school) without having taken a class in the documents and ideas of the American founding. To a mandatory course on those I’d add a second term on American political thought in the 19th and 20th centuries, full years each on American literature, American history, Western literature, Western philosophy, European history, and global history, plus two terms (from a menu of options) on non-Western civilizations.

Subscribe now

Yes, it’s a lot—16 semesters in the curriculum I just sketched—and that seems right, if we are going to graduate individuals who actually know a damn thing beyond their specialty, who can effectively evaluate whatever nonsense shows up in their feeds. One third to one half of an undergraduate career, to speak a little less prescriptively, should be devoted to general education. If that prevents students from double-majoring, so be it. There’s too much of that already anyway (largely since the withering of general education has left specialization as the only part of college students understand). If it means that engineering departments can no longer mandate 20 courses in their majors, good. They need to play better with others. As for the objection that students won’t or can’t read even single books, let alone a dozen courses-worth or more of weighty tomes, I call bullshit. If we can demand that students study chemistry and physics, French or Spanish, algebra, geometry, and trigonometry, let alone calculus, before they even get to college, we can insist that they read. And if they cannot get through college without reading a lot, or even be admitted if they haven’t read a lot already, then high schools will insist on it, as well.

A larger point here is that liberals need to reclaim the great books, and the humanities more broadly, from both the right and left: from the know-nothing populists who think that “Western civilization” means “I’m better than you” and the tenured avengers who use “dead white male” as a term of abuse. It’s been many years since liberals have even had a theory of culture, have engaged with culture as an object of thought or recognized its relevance to the political (many years since the heyday of Partisan Review). I remember hearing Al Franken, then an Air America host, insisting that the only poems he was interested in had five lines and started “There once was a.” That’s about the level of seriousness with respect to culture that liberals have been operating on since, I’d say, the 1970s.

A second point is that the university must recognize that the fields proper to general education do not exist for the same purposes, and should not be conducted in the same manner, as the natural and social sciences. The latter, at least in their mature form, are native to the modern research university, that factory of facts, with its positivist conception of knowledge, its quantitative methods, the kinds of questions that it asks. But philosophy and literary inquiry predate the research university by many centuries. They arose to ask a different set of questions, ones about the nature and meaning of existence, which cannot be answered empirically (or, in any final way, at all). It was only once they migrated into the modern university that those traditions became “the humanities,” a set of academic disciplines, and they’ve become denatured from needing to conform to scientific standards—the reason that, whether produced by non-academics or by academics writing in non-academic modes, a great deal of the best philosophy (often in the guise of “thought”) and certainly of the best literary criticism (and music, art, and theater criticism) has continued to take place outside the university.

But the purpose of general education is precisely to equip students to address those fundamental human questions—questions of value, not fact—which are perforce the fundamental political ones. That is why the study of literature (to turn to something dear to me as a former professor of English) is integral to liberal education. Even individuals and institutions committed to civic education and core texts (and there is a growing movement toward both, especially at public universities in states like Florida and North Carolina) do not appear to grasp this. Beyond Homer, Greek tragedy, and Shakespeare, curricula lean largely to political philosophy—that is, to texts that seem more relevant to public questions.

That literature is equally relevant, if less obviously so, begins for me with a remark that a professor made in graduate school, a kind of gnomic aside, and that I’ve been pondering ever since. He said that stories are a form of knowledge. What form, exactly, is what I’ve been pondering, but this much is clear: it is a form that’s recognized by every culture we’re aware of. Every culture tells stories, and not just for entertainment. In every culture, stories are central to the transmission of collective understandings. They are a culture’s way of expressing its sense of how the world goes, and of how we should conduct ourselves within it.

Subscribe now

Yet that expression is rarely transparent. In the Old Testament, for example, replete though it is with commandments and precepts, the meanings of stories are far from obvious—think of the binding of Isaac—which is why they’ve given rise to a vast and ongoing interpretive discourse. Indeed, one of the books of the Old Testament, the most morally and existentially challenging of all, is itself concerned with its own interpretation. Job spends most of his story largely in the company of friends who prove exquisitely unhelpful, precisely since they keep proposing simple answers, trying to understand the meaning of what has happened to him. Then God shows up and tells him the answer, which is that there is no answer—not one, at least, that humans can grasp.

When we turn to the New Testament, we find that Jesus himself told stories, the kind of stories we call parables, which are likewise frequently distinguished by their open, enigmatic quality, their endless interpretability. And so it is with all the great teachers: with Socrates and the Buddha, with the Zen masters and the Hasidic masters. They do not offer extractable morals or lessons. They offer riddles and puzzles, enigmas and dilemmas, because that is what life does itself. And the same is true of great works of literature, texts so central to our own culture that we denote them by a term that once referred exclusively to Scripture: the canon. The Iliad does not offer readily statable meanings, and neither does Antigone or Hamlet or Moby-Dick or To the Lighthouse.

Literature offers us not answers, but vivid ways of posing questions. Unlike philosophy, with which it shares an interest in foundational perplexities, it deals not in abstractions but particulars. Unlike the social sciences, with which it shares a focus on psychology and sociology, it puts us in the midst of life. Unlike both, it gives us a plurality of points of view. When you read a work of exposition, not just philosophy or social science but commercial nonfiction, opinion writing, even personal essay, you know what its author believes, because it is the purpose of the thing to tell you. But the better a work of literature is, the harder to say what its author believes—famously, most difficult of all with Shakespeare. Great works of literature present divergent perspectives without choosing between them. They show us the validity of all: Antigone and Creon, Lear and Cordelia, Elizabeth and Darcy. This is an essential political idea, a liberal idea. The world isn’t heroes and villains, good guys and bad guys. It is people, who are always flawed and always have their claims.

We can say more. One of the limitations of the way we think about both public and private life today, I’ve come to feel, is that we invariably talk in terms of problems and solutions. Not that there aren’t problems, or that they don’t need solutions. But some things—again, the fundamental things—do not have solutions, which means it doesn’t help to think of them as problems. Call them, rather, tensions: the tension between equality and liberty, or prosperity and the environment, or a woman’s bodily autonomy and the interests of her unborn child. Or, in private life, between desire and fidelity, or between our love for our parents and our need to separate from them. These aren’t problems to be solved, or resolved. They are tensions to be endured. They are tragedies, in Hegel’s sense of tragedy: a conflict between two rights, two values.

And that is what literature shows us: these tensions, those tragedies. The point of studying Antigone is not to decide whether Antigone is right or Creon is right. It is to see that they both are. That sometimes there are not only not simple answers, but any answers at all. That this, at its most acute, is what life is. That we all exist within these tensions, those tragedies—as individuals and as political communities.

Students may not read, but they are soaking in stories: superhero movies, action movies, rom-coms, detective shows, shows about teens with special powers. Video games. Porn, which tells a kind of story, too (though the ending is always the same). As well as, in the realm of supposed nonfiction, conspiracy theories, partisan narratives, propaganda. In other words, bad stories: stories that offer easy answers, stories with good guys and bad guys, stories where the hero gets their every wish fulfilled, stories where it all works out in the end. All of which has political implications, because it shapes one’s expectations of the world. Another reason students need great stories is to guard themselves against the bad ones.

If all great literature is, as Harold Bloom insisted, wisdom literature, then what its study helps you to develop is, precisely, wisdom. And wisdom is different from skills or expertise or subject knowledge, as valuable as all of those are. It is such understanding as pertains to the conduct of life. And not only the individual life—the collective life, as well. That is why Plato spoke of the need for wise leaders and we speak of the wisdom of the American Founders, their grasp of human nature. Wisdom is a political virtue. It is the highest political virtue.

So much for the what of liberal education. Now for the how. When people talk about the college classroom as a training ground for citizenship, they mean the inculcation of a certain set of virtues and habits. You sit around the table wrestling with Locke or Dostoevsky, and you learn to have a civil conversation: to disagree respectfully, to listen, to consider arguments, to change your mind. As David Bromwich puts it, you learn how to be wrong. You develop humility, patience, tolerance, courage (the courage to speak up, to make waves, to risk looking foolish)—perhaps above all, a thirst for truth, even at the cost of self-regard.1

Though—to digress for a moment—it is a lot easier to teach students how to disagree if they actually do. I’m speaking not just of the ideological conformity, real or performed, that dominates selective institutions, but also of the social homogeneity that lies behind it. I love that schools like Harvard claim they want their student bodies to “look like America”; I only wish they meant it. Looking like America would entail enrolling 40% of their students from the white working class (as opposed, at elite schools, to virtually none). Imagine how quickly campus culture would change if institutions did that—and, before long, how the culture of the professional-managerial class would, and then how our politics would. We are polarized because we’re siloed. The domains we occupy are not just different; they are separate, non-communicating. We need to reconstruct a shared world, in Hannah Arendt’s sense: a sphere of mutual intelligibility, an arena of discourse not in which we all agree, but in which it’s possible to disagree. At present, we cannot even disagree, because to disagree you need to understand each other. A genuine commitment to civic education at elite schools would begin by gathering students together across relevant forms of difference—social, cultural, political.

In any case, it’s easy to talk about teaching for citizenship; it’s a lot more difficult to do it. Teaching well is hard. The kind of seminar-style teaching that general education involves—orchestrating conversations that are substantive and rigorous, that keep the focus on the text, that stay on point, that require students to respond to one another, not just wait until the other person stops so you can say your thing, that don’t degenerate into speechifying or (in lit class) group therapy—is much, much harder. It takes a lot of time to learn, and it continues to take a lot of time even after you’ve learned, since each new text you teach necessitates reverse-engineering a fresh set of questions, and a lot of the art involves asking questions: ones both specific and challenging, not ambiguous or vague or obvious, not “so what did you guys think?” or guess-what’s-in-my-head, questions that you don’t have settled answers to yourself. The guiding principle is this: whoever is doing the talking is doing the thinking. In a seminar, 90% of what comes out of the instructor’s mouth should be questions, and 90% of everything should come from the students.

Most faculty do not know how to do this. Most college teaching is mediocre at best and often far worse. This is not a guess or an impression (though I’ve seen enough of it myself). In The Amateur Hour, a history of college teaching in America, Jonathan Zimmerman lays out the gory details. We’ve had the same problems, for the same reasons, with the same failed solutions, since the emergence of the research university: professors neglecting instruction; enormous lecture courses (and tedious discussion sections); contingent and underqualified faculty; students feeling bored and cheated; resistance from faculty to supervision, evaluation, or change; innovations, often based on new technology, rolled out with large claims; and no improvement ever. The reason for this last is clear. Under the research model, faculty are incentivized to do a single thing only: create knowledge. Publish or perish. When good teaching happens, it happens by accident, and often at a cost to one’s career.

Subscribe now

Which means that if general education is going to be resuscitated—and undergraduate education in general improved, and academia despised a little less—colleges and universities need to start seeing themselves, to an extent they never have before, as teaching institutions. “Our scholarship is a professional enterprise,” Zimmerman writes. “But when it comes to teaching … [w]e are flying by the seat of our pants.” No more. Instead of being flung into the classroom with minimal preparation and expected to figure it out on their own, graduate students need to learn to teach in a concerted, informed, and organized way: with extensive training, repeated observation, regular feedback, iterated improvement, and continual support. Just like in K-12. And they need to continue to learn and improve once they get a position, just like in K-12.

Faculty should meet to talk about teaching on a regular basis: troubleshooting, sharing insights, discussing best practices. Master teachers ought to serve as mentors for less experienced ones, including through team-teaching. Some of the former should act, at least part of the time, as instructional coaches, going into classrooms and working with colleagues on specific areas of weakness. Doors should stay open, and observers should be free to enter class at any time. Professors must get used to being less defensive and territorial about their teaching. Faculties should see themselves as communities of mindful practitioners, striving together toward excellence.

But how are academics ever going to find the time for this, with all of the research they do? Well, that’s just the thing. I hate to say this now, when the government seems intent on killing scholarship and science altogether, but there is far more research done than anybody needs. Truckloads of articles, monographs, studies, much of it trivial, most of it essentially ignored. The research model was never meant to apply to more than a relatively small number of institutions, but given academia’s incentive structure, especially since the postwar funding boom, it has spread to nearly every corner of the enterprise. The lion’s share of significant work continues to be done at a few dozen schools, the ones with the money and prestige to attract the best people. Yet according to the latest count, there are 543 Carnegie-classified research institutions in the United States, including 187 R1s alone (in 1994, the latter numbered 59). And that is not to mention the 2670 other colleges and universities, which also maintain expectations, albeit at a lower level, for scholarly production. No wonder there is so much fraud in science, a replication crisis in the social sciences, and mountains of meaningless bullshit in the humanities. Most research should simply stop. Most if not all of a professor’s time, at the vast majority of schools, should be devoted to teaching.

Believe me, I know what I’m saying sounds crazy, including the part about turning over much of the curriculum to general education. My suggestions would necessitate enormous changes: structural, cultural, and, above all, psychological. Academics would have to learn to see themselves, and value themselves, in radically different terms, as members of a helping profession. It isn’t clear how we could even do these things with the existing professoriate, given the way they’ve been socialized and trained. A new faculty, somehow, would have to be raised, gradually and in part, perhaps, from outside the academy.

But now is the time to think big. With the recent Yale report and other scattered indications, it seems that academia is starting to acknowledge its disasters and its culpability for them. Incremental changes, though, will not suffice. More than once in our past, at moments when the sector recognized that it had ceased to serve the nation’s needs, higher education overhauled itself in fundamental ways: in the late 19th century, when Ivy League and other colleges transformed themselves into research universities and the fixed curriculum of Greek and Latin, in place since time immemorial, was discarded for the system of majors and electives; in the 1960s, when the old-boy, private-school arrangements in elite admissions were torn down in favor of meritocratic criteria, which remade the American leadership class. This is the kind of juncture where we find ourselves. This is the scale at which we need to act.

William Deresiewicz is an author, essayist, and critic. He is working on a historically situated memoir about being Jewish in modernity.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Jordan Bardella is often described as France’s answer to Donald Trump. Like the U.S. president, the leader of the hard-right National Rally (RN) has ridden a wave of anti-immigration sentiment, anger at globalized elites, and mistrust of multilateral institutions. He’s France’s most popular politician and looks the favorite to win the presidency next year. And he’s only 30 years old.

Born in 1995, Bardella was raised by a single mother in Saint-Denis, a bleak northern suburb of Paris. He is almost entirely of Italian stock, with an Algerian great-grandfather on his father’s side. He is also, like his party, firmly anti-immigration. But to those who cite his ancestry against him, Bardella has a pointed answer: his European forebears fully assimilated; when the cultural distance is greater, the journey is harder.

He became politically active at age 16, drawn to far-right leader Marine Le Pen, whom he felt spoke to people like him living in France’s gang-ridden banlieues and other areas the establishment had forgotten. Le Pen was then engaged in turning the rabble-rousing National Front—founded by her father, Jean-Marie Le Pen—into a true governing force by expelling its openly racist and antisemitic elements, and eventually Jean-Marie himself.

Persuasion events

After dropping out of Paris-Sorbonne University to focus on politics, Bardella rose rapidly through the ranks and caught Marine’s eye. A smooth talker, he became party spokesman. His big break came in 2019, when she picked the 23-year-old to lead the National Front—now rebranded the National Rally—into the European Parliament election. The gamble paid off: Bardella’s list narrowly beat the centrist bloc supporting President Emmanuel Macron. The RN was not just the main opposition: it was France’s largest party.

2022 was another pivotal year. After a second consecutive defeat to Macron in the presidential run-off, Le Pen decided to prepare more rigorously for the 2027 election by stepping back from day-to-day party business. She needed a loyalist to take over as RN president. No prizes for guessing who got the job. The division of labor suited both: he had the title and the platform, while she retained her position as head of the parliamentary group and overall strategic leader.

A series of twists over the past two years have turned Bardella into clear presidential material. In European elections in 2024, Bardella’s list achieved double the vote of Macron’s centrists. A humiliated Macron immediately called a snap general election and paid dearly for it. The RN and its allies made historic gains, placing it in a position to help topple governments in a hung parliament—which it has since done, twice.

Then came the legal shock. In March 2025, Marine Le Pen was convicted of embezzling EU funds, and judges declared her ineligible to run for president for five years. Although an appeal verdict is expected in July, the party has been preparing an alternative candidate. “Plan B for Bardella” became the inevitable shorthand.

Bardella rejects the phrase, insisting that “plan A has not yet run its course.” But his evident electoral appeal has already placed him on a presidential trajectory. In 2024, his autobiography became the political best-seller of the year. Book signings in towns up and down France drew long lines, confirming his star status. By the end of the year, his approval ratings had overtaken Le Pen’s.

Bardella’s politics can best be described as populism with French characteristics. In this sense, the comparison with Trump is appealing but misleading. They took opposite routes to the top: while Trump radicalized a mainstream party, Bardella helped moderate a once-pariah movement. Last year he withdrew from an American conservative event after Steve Bannon appeared to make a Nazi salute. Since then, Bardella has been at pains to condemn Trump’s trade offensive against Europe (“blackmail”) and Middle East policy (“totally erratic”).

This speaks to Bardella’s main asset—his ability to expand the RN’s reach into territory where it has historically been weak, starting with younger voters from working-class backgrounds with little prior engagement with politics. His social media presence—2.3 million followers on TikTok—reinforces the image of a man who grew up in the same suburban France as his audience. In behind-the-scenes footage, he is shown relaxed, sipping pastis, nervously preparing for a rally, or compulsively snacking on candy. It may be manufactured authenticity, but it hits home. As one young voter told Le Monde: “The others are a different world, they don’t speak the same language.”

More broadly, a recent IFOP survey suggests that among lower-income voters, Le Pen and Bardella perform equally well. But in richer categories, where the RN has struggled, he pulls markedly ahead of her. IFOP also asked respondents whether they would like various politicians to run in the next presidential election—a different question from voting intention, but still a measure of how they feel toward various figures. Among centrist and center-right voters, approval for a Bardella candidacy significantly outstrips that for Le Pen.

Subscribe now

The lesson is clear: where Marine Le Pen consolidated the RN’s electorate, Bardella is expanding it upward—toward wealthier voters and what remains of the mainstream right. That, in essence, is the Bardella effect.

Another key part of Bardellism has been wooing corporate France, which has long shunned the RN. The party’s 2024 program included lower taxes for companies and a bonfire of regulations. But the finances were far from sufficiently broken down, and the plan piled on new spending by scrapping a planned pension reform.

Against this backdrop, Bardella worked hard to arrange meetings with France’s main employers’ federation. The bosses, on their side, were eager to inject some fiscal reality into a program that, as it stands, would alarm the bond markets. After a private lunch, a besuited Bardella struck a reassuring note: “I believe in freedom of enterprise; we need to give back freedom to those who create and innovate.”

This charm offensive towards big business sits uneasily with Marine Le Pen’s more traditionalist line. Catering to RN voters who peeled away from the left, notably in France’s northern rust belt, she is not afraid of striking redistributive poses. Her party’s program still contains a “tax on financial fortune.” More recently, her deputies in France’s assembly backed a levy on “unproductive” assets—an attempt to revive an old wealth tax—and an equally controversial tax on multinationals’ profits in France. Neither was retained in the final budget, but both confirmed where Le Pen’s instincts lie when it comes to big business.

The tension with Bardellism extends to the taxation of oil company windfall profits during supply shocks such as the current conflict in the Gulf, a measure also championed by the left. Le Pen embraces it as “a matter of social justice.” Bardella, characteristically, cultivates strategic ambiguity: when pressed on the question, he simply said, “Why not?” The two-word answer sums up the studied vagueness he has developed into an art form.

None of this is easy to read. Are the two playing a double act to maximize their electoral reach? Both Bardella and Le Pen insist there is no daylight between them. But even if genuine divergence exists, the question remains: does Bardella have the weight, or the will, to reshape a party the Le Pen family has spent half a century building?

Whatever happens, the RN’s core principles remain. And they still put the party on a collision course not only with France’s need for fiscal stability, but with European realities. The RN’s flagship policy of “préférence nationale”—reserving jobs, housing and welfare for French nationals—runs against the EU’s foundational principle of equal treatment for all citizens of member states. Its suspicion of free trade and Bardella’s talk of “intelligent protectionism” jar with France’s deep integration into global supply chains. And the RN’s vision of a “Europe of nations” rests on the illusion that French companies can enjoy the benefits of the European single market while Paris picks and chooses which rules to follow.

Subscribe now

Bardella is no Trump. But neither is he Italy’s Giorgia Meloni, who set out to make a pragmatic peace with EU rules and multilateral institutions. Populism with French characteristics represents something new and untested: a burst of utopian nationalism in the place that invented the European project. In a country traditionally run by high-flying graduates, a university drop-out with no government experience in the Elysée would be just as disruptive as a Trump White House has been in the United States.

It is sure to send shock waves well beyond France’s borders.

Henri Astier is a London-based journalist who writes for French- and English-language publications. He writes the Substack Out of France.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

This article is part of an ongoing project by American Purpose at Persuasion on “The ‘Deep State’ and Its Discontents.” The series aims to analyze the modern administrative state and critique the political right’s radical attempts to dismantle it.

To receive future installments into your inbox—plus more great pieces by American Purpose and Francis Fukuyama’s blog—simply click on “Email preferences” below and make sure you toggle on the buttons for “American Purpose” and “Francis Fukuyama.”

Email preferences

Subscribe now

The end this week of the longest ever shutdown of U.S. government offices marks a new normal in polarized Washington. Closing federal doors is now a routine power play for Republicans and Democrats alike. Last year, a bitter fight over healthcare funding forced some 900,000 civil servants to stop work for 43 days. This year, the same impasse hit about 100,000 employees in the Department of Homeland Security for almost twice as long. Returning to work with no resolution of the underlying dispute promises more future shutdowns.

Thirty years ago, American voters viewed the stoppage of government as unacceptable. When the Republican Speaker of the House, Newt Gingrich, refused to pass funding bills unless President Bill Clinton agreed to steep budget cuts, the backlash helped propel Clinton to reelection in 1996. A chastened GOP did not pull the trigger on federal funding until 2013, when closing government down failed to stop the rollout of the Affordable Care Act.

Now, however, the public seems ready to go along with missing the full range of government services as long as a few important red lines are not crossed. These include delays in federal benefits and tax refunds, the closure of national parks, and prolonged disruption of airport security. The other big things done by the government in public health, regulation, scientific research, and a host of other fields don’t figure as punishable.

While Congress and the White House focus on shutdowns as a maneuver, the deeper damage of holding government hostage goes unnoticed.

First, shutdowns over the past decade have pulled attention away from exploding U.S. budget deficits. The clash between President Clinton and House Speaker Gingrich in 1996 over the nation’s financial future had a positive outcome. Both sides agreed that U.S. budget deficits were unsustainable—they just differed on the pace and distribution of cuts. The compromise that ended the shutdown put the country on a path to pay down the national debt and resulted in one year of actual surplus.

The national debt has grown from $5.7 trillion to $39 trillion since 2000. Budget confrontations, however, have nothing to do with bringing deficits under control. Instead, government shutdowns stem from a bitter divide over America’s racial, ethnic, and cultural identity. The funding of President Trump’s border wall brought the government to a standstill during his first term, while no-holds-barred immigration enforcement has done so during his second. The divide over national identity has marginalized voices of fiscal restraint on both sides of the aisle. Neither Democrats nor Republicans are ready to hit the brakes.

Second, the normalization of shutdowns has weakened Congress. Every decision not to fund the government marks a failure on Capitol Hill of give-and-take political bargaining. Surveys record steep drops in public approval of Congress from the mid-30s to the mid-teens after every shutdown, followed by partial recovery when government reopens. Nevertheless, hardliners feel empowered by voters to stick to their guns.

The electoral costs of refusing to find common ground may be low, but the price that Congress pays as an institution is high. Legislation with input across party lines creates institutional leverage. However, the Republican-led Senate and House have given unequivocal backing to a them-versus-us agenda set by the White House. As a result, President Trump exercises a degree of control over Congress that used to be unthinkable, undermining its classic role as a check on executive power.

The third way in which shutdowns do damage is that they devalue federal workers. Having to stop work highlights their facelessness and political vulnerability rather than their specialized skills and commitment to public service. Since more than two million civil servants are spread over 15 cabinet departments and 50 independent agencies, most Americans lack an overall picture of what they do, why they do it, or what the true impact of stopping their work is. Lack of knowledge feeds a stereotype of careerists as overpaid, overprotected, and able to withstand shutdowns because they eventually get their paychecks.

The president has deliberately devalued the federal workforce by making its reduction a priority domestic goal. Some 350,000 career employees have left government through retirement, resignation, and layoffs since January 2025. The White House has taken credit for this downsizing despite the loss of experience and expertise and the need to backtrack on hasty, mission-threatening personnel cuts in disaster relief, tobacco oversight, and other federal programs.

A singular focus on reducing the size of government leaves out America’s stake in attracting and retaining top-tier personnel in dozens of specialized fields. As I have learned firsthand from high-performing federal workers, talent means as much in government as it does in the private sector. Devaluing federal careers reduces their appeal to the very men and women who are needed to improve federal performance.

Holding non-partisan civil servants hostage is an act of self-destruction. The damage won’t stop unless Americans get serious about rejecting extremism and making democracy work.

John Yochelson, the former president of the Council on Competitiveness, is assembling and editing a collection of personal stories of high-performing federal workers to make them more relatable to the public.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Call me naive, but I think I had given up on hypocrisy as a defining feature of American politics—I thought we had gone all the way past that to open avarice and the unapologetic exercise of force—but I guess if there’s one constant in life, it’s that there will be a place in politics for hypocrisy. So there was something refreshingly quaint about Tucker Carlson’s recent break with Trump right at the moment when a wedge issue formed in the MAGA coalition and Carlson could start to position himself for a 2028 presidential run. The news cycle was duly roiled with Tucker’s discovery of principles, even as it was evident on slightly closer inspection that the principles all benefited Tucker in the long-term.

The ostensible reason for the break is legible enough. Carlson has been an advocate for America First policies for a long time. He criticized Trump’s killing of Qasem Soleimani in 2020 and reportedly advised Trump against an Iran strike at that time. He campaigned—avidly—for Trump in 2024 on the premise that Trump would keep America out of foreign wars, and the attacks on Venezuela and then on Iran seem to have registered for Carlson as a genuine shock, and then led to the kind of falling out with Trump that, short as the memories of these two are, will be hard to patch up again.

On April 6, responding to Trump’s “Open the fuckin’ Strait, you crazy bastards” posts on Truth Social, Carlson said on his show, “How dare you speak that way on Easter morning to the country? Who do you think you are? You’re tweeting out the f-word on Easter morning?”

On April 7, Trump, skipping over the theological bits, countered by saying “Tucker’s a low IQ person that has absolutely no idea what’s going on.”

Subscribe now

On April 20, Carlson hosted his brother, a longtime Republican operative, on the Tucker Carlson Show, and, in anguished and deeply religious terms, talked about his reasons for the break. “You and I and everyone else who supported him, we’re implicated in this for sure,” he said.

“We’ll be tormented by this for a long time, I mean I will be. And I want to say I’m sorry for misleading people. It was not intentional.”

It’s a little hard to believe that there were no warning signs of what sort of person Trump might prove to be—not calling a beauty pageant contestant “Miss Piggy,” not boasting that he likes to grab women by the pussy, not going out to see a movie on the day of his brother’s death, not exploiting his father’s dementia to bilk his relatives out of their inheritance, and not encouraging a mob to storm the U.S. Capitol. But Pastor Tucker was in a very Christian frame of mind. Yes, Trump had his character failings, but “there are tons of people of low character who outperform their character,” he declared.

And what a compelling, nay, biblical picture that makes—prodigal Tucker misled by his overly compassionate nature and his desire to avoid foreign wars having his Road to Pennsylvania Avenue (er, Damascus) moment over Trump’s “or you’ll be living in hell” post. By now fully ascending the pulpit, Tucker had it in him to offer a remarkably ecumenical message to all faiths. “No decent person mocks other people’s religions,” he said. “To mock other people’s faith is to mock the idea of faith itself.” Never mind that he has discussed the “Islamic cult” and the “Islamic problem” and in 2019 a guest on his show called Islam “the most hateful, intolerant religion in the world.” Tucker, even in the depths of his torment, was in a forgiving mood, forgiving even for such a sinner as himself, and in his grace and willingness to look beyond past trifles he appeared, yes, positively presidential.

I really mean this. The betting odds site Polymarket has JD Vance leading the Republican GOP field with 39%, followed distantly by Secretary of State Marco Rubio at 22%, and then Carlson all the way behind them at 6%. I would never ever give betting advice to any Persuasion reader, but if I did, I might well counsel laying some sweet cheddar on Carlson. Vance is giving major Jeb Bush vibes at the moment—linked to an America First point of view that he seems to at one point have felt deeply but knowing that if he breaks with Trump over Iran or anything else he loses virtually the entirety of his support. Rubio was never exactly in his element as a presidential candidate and seems to have found his ceiling as more of a backroom boy.

But Carlson is a survivor, and comes with a built-in audience, and knows how to work the media in a way that resembles Trump himself. He also speaks to what may be the single most stable voting bloc in American politics—Christian nationalism.

So Tucker really is one to watch—for her part, Marjorie Taylor Greene was virtually the first to hop on the bandwagon and declare on March 5, “I SUPPORT Tucker … Tucker would beat Trump if he ran for president”—and this break with Trump may well be the start of the Making of the President 2028, when he would be the first modern candidate to run on a Christian nationalist platform as well as the first-ever Dancing With the Stars contestant to reach the White House.

So, a pretty arresting narrative of repentance and redemption. But before you are too taken in by it, do consider this: Tucker could have broken with Trump a long time ago, and the timing of his break, when it finally happened, could not be more serendipitous to getting organized for 2028. As far back as 1999, he called Trump “the single most repulsive person on the planet.” In documents that emerged through discovery in 2023, he wrote privately of Trump, “I hate him passionately.” This was of course around the time that Tucker pushed the narrative that the voting systems had been hacked by Communists to perpetrate widespread fraud and that the January 6 Capitol attack was an FBI “false flag” operation, and, while campaigning for Trump in 2024, declared that he “always agreed with Trump’s policies” and called Trump’s survival in the Butler, PA assassination attempt “divine intervention.” So, yes, while Tucker is in torment mode, he does have some “splainin’” to do.

But, as is often the case in politics, hypocrisy is a superpower. Tucker has, as so often happens, changed his stance on every single major issue. Foreign interventions? He supported the Iraq War up until 2004. Protectionism? He was an economic libertarian until 2018. Strong borders? It wasn’t an issue for him until the Trump era. His biographer Jason Zengerle argues that Carlson’s slipperiness, and ability to always put himself first, encapsulates the “larger story of conservative media and conservative politics over the last thirty years.” When he has about-faced, it hasn’t always necessarily been for the best of reasons. His principled opposition to the Iraq War in 2004 was based above all on the observation that “Iraq is a crappy place filled with a bunch of, you know, semi-literate primitive monkeys” as well as concerns that it’s not worth invading a place where people “don’t use toilet paper or forks.”

And Carlson has an ability to not only reverse his positions but to actually voice two completely contradictory positions at the same time. In 2021, he strongly pushed Trump lawyer Sidney Powell’s allegation that Dominion Voting Machines had been hacked at the same time that he was (unfortunately, as it later turned out) writing internally that “Sidney Powell is lying by the way. I caught her. It’s insane.” That little fib—well, more the honest admission on the back end—is one reason why Fox ended up paying Dominion Voting Systems $787 million for defamation and Tucker lost his job. But Tucker, as he has so many times before, bounced back, hosting his own show without oversight from Fox.

Faith and patriotism are, in the end, the last refuge of a scoundrel—and they are formidable sanctuaries. You may have noticed that Tucker has been sometimes less than Christian-like in his behavior in the past, as for instance when on the show Bubba the Love Sponge he referred to individual women as “pigs” and “cunts,” assessed women in general as “extremely primitive,” and volunteered to give Martha Stewart’s daughter “the spanking she so clearly needs.” But that’s the beauty of cloaking oneself in faith. Repentance is the ultimate get-out-of-jail free card, and there is no reason why Carlson’s path to repentance can’t coincide with a White House run.

That really is the only meaningful fissure that can break the MAGA coalition. Epstein won’t do it; and a foreign policy excursion like Iran isn’t enough. But religion could do the trick. Evangelical voters, through a bit of deft theology and a vituperative hatred of Hillary Clinton, managed to support Donald Trump, but it was always an uneasy and sometimes comical alliance. Tucker speaks that language far better than anybody in the more Trump-y core of MAGA.

We had a glimpse of what the future of the Republican Party, if not the 2028 campaign, might look like in a contentious interview between Carlson and Mike Huckabee—formerly the preferred candidate of evangelicals and now ensconced in MAGA-dom as the U.S. Ambassador to Israel. This was before the decisive break with Trump, and Carlson frequently came across like a seventh grader on the border between a B and B+ in civics class when he, for instance, asked about public disapproval of the war in Iran: “We care deeply about [poll numbers] in what sense? If we’re ignoring it then in what sense do we care deeply about it?” Watching that had a rivetingly Shakespearean aspect to it, like viewing Macbeth wander into his sleeping king’s chamber and consider that the poll numbers gave him all the legitimacy needed to do the bloody deed.

But the conversation also took a turn. And I suppose that we all knew American politics would one day come to this, with both Carlson and Huckabee carefully parsing Genesis 15, and Huckabee articulating an interest in a Greater Israel—like, a much Greater Israel (“It would be fine if they took it all,” he said, redrawing the hypothetical map of Israel from the Nile to the Euphrates). Carlson, in the opposite vein, took a much more, er, New Testament view speculating on whether Jews “have dual loyalty” and claiming that Huckabee’s “priorities are very clear” in reference to being overly sympathetic to the Jewish state.

Subscribe now

What Huckabee likely had in mind was “dispensational premillennialism,” the story of the rapture, with all Jews moving to Israel to initiate the “end times” button on the console. And what Carlson had in mind was revealed at Charlie Kirk’s funeral in 2025, when he compared Kirk to Jesus Christ and envisioned the conspiracy leading to Jesus’ death: “Picture the scene in a lamp-lit room with a bunch of guys sitting around eating hummus thinking about, ‘What do we do about this guy telling the truth about us? We must make him stop talking!’”

Having given the Elders of Zion their appetizer course, Carlson continued that vein of thought into this year. No sooner had he washed his hands of Trump than he was wondering aloud who the shadowy cabal was behind world events. “What was Butler? What was Ryan Routh?” his brother asked in their conversation on Tucker’s show. Tucker did not miss a beat. “I know that those investigations have been stymied. Fact,” he said. At which point his brother upped the ante by segueing to Israeli-American businesswoman Miriam Adelson’s (alleged) campaign contribution to Trump and asking, “Why would someone who has obvious and demonstrated allegiance to a foreign power give Donald Trump $250 million dollars while he’s running for president?”

The premise here is that both the assassination attempts in Butler, PA and in Mar-a-Lago were orchestrated by the Adelson family over a delicious hummus course as a way to show that they could get Trump anytime they wanted, and it’s that plot (this is also the plot of the Don DeLillo novel Libra, by the way) that accounts for Trump’s sudden turn towards warmongering in his second term—a scenario, to Tucker’s mind, that fully exculpates him for having ever been hoodwinked by Trump.

In a wide-ranging interview with The New York Times this week, Carlson—in “just asking questions” mode—took that notion well past the point of absurdity. “One thing that has bothered me for many years is the fact that a lot of people in Trump’s immediate orbit have been hurt—and really hurt. Gone to prison, become unemployable, publicly shamed, have gotten cancer,” Carlson said. Based on the unenthusiastic tone in Trump’s voice when Trump told him about the plans for the Iran War, Carlson could conclude nothing else except that “Trump was more a hostage than a sovereign decision-maker in this.” As a conspiracy theory, it’s a neat reversal of QAnon—with Trump held hostage by a shadowy cabal and needing to be rescued by a white knight, and a white knight with a mega-popular TV show.

If I am a little less than impressed by Tucker’s turn towards Christian righteousness and not exactly thrilled by the advent of of anti-Semitic blood libels into American politics, that is not to say that I find him any less formidable for it. He is extremely dangerous and he could well win. He is adaptable and resilient. He speaks fluently to the Christian Nationalist wing of the Republican Party, but for all his faith-mongering, he will not get bogged down in pieties like previous religion-oriented candidates have been.

He is as chameleonic as they come, and, in this era, with the kind of politics we have, what else can his gift for hypocrisy and shape-shifting be but a superpower?

Sam Kahn is associate editor at Persuasion, writes the Substack Castalia, and edits The Republic of Letters.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Laurenz Guenther is a political economist at the Toulouse School of Economics and a Fellow at the Institute for European Policymaking at Bocconi University. His research and Substack focus on representation, populism, and immigration in Western democracies.

In this week’s conversation, Yascha Mounk and Laurenz Guenther discuss why there’s a massive representation gap between political elites and voters on cultural issues, how this explains the rise of populist parties like the AfD in Germany, and whether new parties can successfully occupy the economically left but socially conservative political space.

This transcript has been condensed and lightly edited for clarity.

Yascha Mounk: You managed to write one of these papers that goes viral quite quickly because it really shows something interesting. Often in the social sciences, the best kind of papers show what we all already kind of knew, but actually demonstrate it in a thorough and methodological way. What you show is that there is a significant gap in representation, particularly of representation on cultural issues, between political elites and ordinary people. To me, the most striking graph in this paper came from Germany, and it looked at attitudes about immigration among members of the Bundestag, among elected politicians, and ordinary voters—I believe in 2013. Tell us a little bit about that specific data and what it shows us.

Laurenz Guenther: What we have is answers to surveys of representative samples of citizens and also of parliamentarians. These two groups answer the exact same question, which enables comparability. The example that you mentioned is about immigration—to what extent immigration should be facilitated or restricted. What one can see in this graph is that Germans, of course, have heterogeneous preferences, but most Germans want to restrict immigration to Germany. What I show is the average response of the members of all the parties that were relevant at this point in time in Germany, and all of these parties—measured by the average position of the member—wanted to facilitate immigration. There was a huge mismatch, even in the direction, in the sense that most people wanted to go to the right on immigration, but all the parties wanted to go to the left.

We hope you’re enjoying the podcast! If you’re a paying subscriber, you can set up the premium feed on your favorite podcast app at writing.yaschamounk.com/listen. This will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!

Subscribe now

Set Up Podcast

If you have any questions or issues setting up the full podcast feed on a third-party app, please email leonora.barclay@persuasion.community

Mounk: One of the striking things about this is that this is before the rise of the Alternative for Germany, which I think was founded right about then, but was not yet represented in the Bundestag. The most right-leaning political party in the Bundestag was the Christian Democratic Union, which was led at the time by Angela Merkel. The view of the average parliamentarian—not just in the Bundestag, but in the Christian Democratic Party—was way to the left of where the average view in the population was. Is that right?

Guenther: Yes, this is right. All of these people who are right-leaning on immigration—from their perspective, it must have looked like everyone, even the supposed right-wing politicians, were much more left-wing. They had absolutely no representation on this topic. This is one of the ideas of this paper: this provided fertile ground for the AfD, which then subsequently also rose.

Mounk: Over half of the population didn’t really have their views represented in the Bundestag. What happens? The AfD was founded as a political party by these slightly stodgy economics professors who were really worried about the euro and opposed the single currency. But the longer the party existed, the more it focused on issues like immigration. By the time it managed to get elected to the Bundestag in 2017, its main focus really was on restricting immigration. What this paper strikingly shows is that there was all of this fertile ground in which it could fish for voters, because so many voters weren’t represented by the pre-existing political parties.

How general is this? That is a really striking data point about Germany. To what extent do you have similar data about other countries? To what extent can we generalize from the existence of this kind of cultural representation gap in the German case to the existence of similar cultural representation gaps in other countries in Europe or beyond?

Guenther: The aim of this paper is to do that more systematically—to look at other countries and other issues. This can be generalized across European countries, and across cultural issues. Looking at 27 European countries, I find the same patterns on cultural issues for all of them. By cultural issues, I mean immigration, but also issues like gender relations, punishment for criminals, assimilation, and multiculturalism. On all of these, voters are much more right-wing than the parliamentarians of their countries. On economic issues it is much more mixed and the gaps are smaller. Notably, there is a great article in the Financial Times where this analysis was also extended to the United States, and there you find a similar pattern.

Mounk: Tell us a little bit more about each side of this. On the cultural representation gap, what other metrics do you look at where the views of ordinary citizens tend to be quite far apart from the views of parliamentarians?

Guenther: Immigration is certainly one of the issues where you have one of the largest differences. Another issue where the difference is similarly large is punishment for criminals—should sentences for criminals become harsher? Here again, majorities in basically all European countries say that this should be done, but parliamentarians disagree. This is a directional disagreement: majorities of parliamentarians disagree with that position, while around 70% of the population say that it should be done. There are also big differences on gender relations and European unification. European unification, however, is not so directional. Parliamentarians seem to be very strongly in favor, and voters are also somewhat in favor, but apparently want a much slower unification.

Mounk: There is an interesting contrast here. On some issues, political elites feel really strongly about something, but they are actually going in the same direction as ordinary voters. But when talking about things like immigration in particular, and to some extent assimilation and how strong criminal sentences should be, they are not just far apart—they are going in opposite directions. The average view of a parliamentarian is that immigration is a good thing for the country. The average view of a voter is that immigration is a bad thing for the country. That feels like a more significant gap. Even if on a 1 to 10 scale the gap is three points on each of them—and I don’t know exactly what the gap is on your scale, I’m making these numbers up—if that three means that overall preferences go in different directions, that seems to matter more than if three means one set of people is very enthusiastic and the other is somewhat enthusiastic.

Guenther: In the public discourse, people sometimes don’t use that term, but this idea often comes up that there may be differences between parliamentarians and voters. It is always framed in the sense that parliamentarians are somewhat ahead and voters want everything to go a little bit more slowly—which, of course, adds a normative dimension, and that is a whole different issue. But it also does not really address these findings. Because, as you say, this is not just a matter of everyone in principle wanting the same thing, with some wanting it a bit more quickly and others a bit more slowly. It is really that parliamentarians and the average voter disagree on the goal—on where we should be heading as a society. That is indeed a very different thing.

Subscribe now

Mounk: Tell me a little bit about those economic metrics, because I think that especially on the left, a lot of people want to think the real disjuncture between voters and the people is on economic issues. People want redistribution, they want a robust welfare state, and then there are the evil political elites—probably from a much more affluent milieu, who have to find donations for political parties, especially in the United States, but also in Europe—who are moving in fancy circles and don’t want any redistribution. That is really where the gap lies. Your data seems to suggest that the story is much more complicated. There are obviously some gaps on various issues of economic policy as well, but they seem to be much smaller on average than on cultural issues. Tell us about the extent to which the views of ordinary people and of political elites match up on economic issues.

Guenther: On economics it depends a bit. On redistribution, there is an item that asks people about redistribution, and qualitatively this is indeed what people on the left would think: ordinary people are a bit more in favor of redistribution than parliamentarians. Notably, wealthier ordinary people—those above median income—have similar preferences to the average parliamentarian of their country, and the gap is driven by the poorer half of the population, who really want much more redistribution. That makes a lot of sense.

But this gap—and I would have to look up the exact numbers—is much smaller than on cultural issues, by a factor of roughly five. The gap on immigration is really about five times as large as the gap on redistribution. On other economic issues it depends a bit more. Generally speaking, it looks like people want a bit more redistribution than parliamentarians, and they want less state intervention.

One thing that is missing here is trade. In the dataset that I use, there are no questions on trade, so there may be a representation gap there—we don’t really know.

Mounk: We would expect political elites to be probably more pro-free trade and ordinary people probably less so. There is another way of thinking about this, which is a two-dimensional graph where you look on one dimension at where people are on social and cultural issues and on the other dimension where they are on economic issues. What tends to be well represented, particularly in two-party systems but to some extent even in systems of proportional representation, is the things that go on the axis from the bottom left to the top right. On the one hand, there are right-leaning political parties that want less redistribution and are reasonably restrictive on migration. On the other hand, there are left-leaning political parties that are quite open to immigration, or even want more immigration, and are also pro-redistribution.

On one side, there are libertarians who are very socially liberal but want less economic redistribution. Those tend to be overrepresented in elite political discourse, but they are actually a relatively small part of the population. There is a much larger part of the population that occupies what is often called the populist quadrant—people who actually want a reasonably high level of redistribution, who do not oppose the welfare state and are not libertarians on economic policy, but who are quite conservative on social issues.

How does your research intersect with that line of thinking? Do you think it is basically right that it is that last quadrant—the people who are socially quite conservative but economically and fiscally reasonably progressive—that are most underrepresented?

Guenther: If you think about these four quadrants, this is the group that is least represented. In this two-dimensional space, the parties in most countries fall along a diagonal. The more you condition on political knowledge or participation in politics, the more you get a strong correlation to that line.

Subscribe now

Mounk: People who are very engaged in politics, if they are economically liberal, are also likely to be socially liberal. But people who are not very interested in politics are going to have views that are more all over the place, which doesn’t necessarily mean they are less coherent. They are just not bundled in the way that our political system conditions us to bundle our views.

Guenther: For the general population, views are more dispersed—it looks more like a circle, like a two-dimensional normal distribution. What my paper shows is that this line of the politically active has actually shifted downward. If you think about that, you can see that the people who are socially conservative and economically left-wing are particularly far away from the closest party in many countries. There is no major party that bundles their views, which I would predict would be an opportunity for new parties.

The populists that are very successful in many countries usually supply this policy position for the upper-right quadrant—that is, for people who are socially conservative and also economically right-wing, even though they are often quite flexible on economics. If you plot them over time in this space, you can see how they move around somewhat to attract more voters from the left.

Mounk: The one thing that defines right-wing populism today in most contexts—perhaps excluding Argentina or certain other countries—is not economics. It is those social and cultural issues. The AfD is interesting in that sense because it was founded on an economically populist issue, but over time it really became defined by those social and cultural issues much more than economic ones.

The economic views of populist parties are a little bit all over the place in different countries. A party like the AfD, in part because of its roots, is probably mostly right-leaning on economic issues, though it certainly isn’t a radical libertarian party and doesn’t want to abolish the welfare state. But a party like the Rassemblement National in France is much more left-leaning on economic issues. They have promised in many ways to preserve the welfare state and opposed Emmanuel Macron’s pension reform, wanting to preserve those entitlements for people. Even Donald Trump plays a strange role in this. On the one hand, a lot of the economic policies he has passed have been quite right-leaning economically—huge tax benefits for the rich, much less for ordinary people. But when you look at how he distinguished himself from his Republican competitors in the 2016 primaries, it was in part by saying things like, perhaps the state does have a responsibility to make sure that everybody has access to healthcare, which was something that the other fifteen Republican candidates did not say.

Why is it so hard for populist parties, particularly in systems of proportional representation, to really appeal to that quadrant? There is something surprising about the fact that you don’t see more political movements move squarely into the space that is economically relatively left-wing and socially right-wing. We have seen an attempt at something like that in the Bündnis Sahra Wagenknecht in Germany—a party that for all kinds of reasons I personally don’t particularly like. They nearly got into the Bundestag in the last elections but fell just short by a few thousand votes and now seem to be falling apart. It seems to be hard for political parties to move into that space, even though that clearly is where a significant portion of the electorate is. Tell us a little bit about who Sahra Wagenknecht is and what this party is.

Guenther: Let me answer this in two steps, looking at two parties that may have moved into that space: the AfD and the Bündnis Sahra Wagenknecht.

The AfD was founded as an economically right-leaning party, largely driven by economics professors opposed to the European Union, and then became very anti-immigration during the refugee crisis. At that point it was probably already positioned, just because of its members, as an economically right-wing party, and it is difficult to move from there. This plays a big role: it is difficult to completely change your position over time, which I think is also one of the reasons why established parties don’t do that. To really capture this quadrant, a party would have to move so much that it risks a split. It would have to make much bigger moves than, for instance, Merkel did during the refugee crisis, which was already a big stress test for the party.

One thing worth mentioning briefly is that in Germany, the quadrant of economically right-wing and socially right-wing people is quite a bit larger than the quadrant of socially right-wing and economically left-wing people. The latter—the quadrant we are interested in—is still a big quadrant, larger than the libertarian quadrant. But if you want to be the socially conservative party and you are simply thinking about maximizing votes, it makes sense to first go after the people who are right-wing across the board, then perhaps move to the center, and then take the second quadrant. The AfD does seem to be trying to do something like that.

Now, turning to Sahra Wagenknecht. She was for a very long time one of the most prominent members of Die Linke—the left-wing party that is basically a descendant of the former Communist Party of East Germany. It is a minor party, polling around five to ten percent, that is very economically left-wing but also very socially progressive and liberal. Wagenknecht distanced herself from this party, largely driven by her views on cultural issues, in particular immigration, on which she was more conservative than much of the rest of her party. This ultimately led to a split in which she left the party together with a few other members of parliament and founded her own party, Bündnis Sahra Wagenknecht—Alliance Sahra Wagenknecht.

This party initially did very well in the polls and was seen as considerably more successful than the original left-wing party. In the election, however, they nearly failed to clear the 5% threshold required for parliamentary representation, while Die Linke did make it into parliament. In some sense, it was a respectable performance—it is very rare for new parties in Germany to clear that threshold, and it was the party’s first election. Now they seem to be falling apart.

My sense is that on cultural issues they were not right-wing enough to capture the quadrant of socially conservative, economically left-wing voters. I think in the beginning people expected Wagenknecht to be more conservative on social issues, and that position appeared to be significantly watered down over time, which I think was a problem for the party.

Mounk: That is interesting. She is also a very charismatic but polarizing figure, and there is always a tension between a political party really just being a personal vehicle for its leader. The party’s name, BSW, literally consists of her initials—Bündnis Sahra Wagenknecht, or Alliance Sahra Wagenknecht. She has now distanced herself a little from the party, and that probably means it is not going to survive. But it was an interesting attempt at trying to capture that quadrant of the electorate.

Let me present to you a reading of your paper which I know you deliberately don’t put in the paper, but which I think is how a lot of people have read it. Tell me to what extent, speaking as a private citizen rather than a scholar of social science, you agree with that interpretation.

Here is the first part. Here is the real explanation of populism. We have been having this debate for ten or fifteen years about why populism is rising around the world, and people point to this explanation and that one. The one thing nobody has really been talking about is whether the problem is simply that the established political parties aren’t listening to what voters want. You might agree with the views of the established parties, or you might agree with the views of voters. On many things, I’m probably closer to the views of members of a political elite, because having a PhD in political science, I’m virtually by definition a member of the political elite. But when we collectively become so distanced from what ordinary people think, they are going to get upset. What happened here is that over time, this cultural representation gap increased. People no longer listened to voters. Election after election, people said they wanted more controls on immigration. Elite political parties did not deliver on those preferences. Finally, voters rebelled by voting for the one party willing to give them what they want.

The second part is a more provocative statement. Perhaps one of the things that established political parties need to do in order to deal with the rise of populism is to actually listen to what people want and adopt some of those policy positions. Not every single established party needs to do this. In the German context, it probably makes sense for the Green Party to continue to be very pro-immigration, because its electorate is very pro-immigration—and that is the virtue of a system of proportional representation. But if the Christian Democrats or even the Social Democrats want to compete with the AfD for the many voters who have shifted towards it over the last few years—a party that now polls roughly equal with the Christian Democrats in first place, having grown from less than 5% of the vote in 2013, around the time of your paper, to around 23 to 25% in polls today—the straightforward thing to do is to get closer to what a lot of ordinary people want. Is that a plausible interpretation of your paper?

Guenther: I think that is a plausible interpretation. The first part stays relatively true to the paper. The previous literature has done a lot of very valuable work, and what I do is not a substitute for that—I’m not saying everyone else was wrong. I have the impression that the previous literature just didn’t really look at this specific part of the puzzle. It looked more at how the financial crisis contributed to making people vote for populist parties, what the loss of manufacturing jobs did, and so on. But there was less focus on the choices of mainstream parties and in particular how they would respond. That ties into the second question, which I also think is a fair interpretation.

This paper is also interesting because it looks at a variable that can be easily influenced. We have all of these variables in mind when it comes to populism—lack of trust, slow economic growth, certain cultural characteristics, and so on. But these things are more or less given and very difficult to change. If you take the estimates seriously, winning back voters by increasing economic growth would require growth rates that are just completely unrealistic. What this paper—and several related papers—does is look at the positioning of parties, which is a variable that can be adjusted relatively easily. In that sense, I do agree that if mainstream parties want to win back voters and weaken populist parties, they have to move their political positions.

Importantly—and this is often a point of confusion—they need to deliver on policy outputs, not just rhetoric. There are some papers that show it backfires if parties merely shift their rhetoric. There was a famous speech by Starmer—“island of strangers” and so on—and the analysis of that suggests it probably backfired.

Mounk: The idea here being that if politicians say they have heard the voters and are going to do something about immigration, and then there is no action to follow up on it, voters conclude that the politicians are simply being hypocritical and cannot be trusted.

Guenther: The facts on the ground must change. In some sense, this is a chance—and maybe the last chance—for the mainstream parties, because populists are often not that good at delivering real results or delivering on their promises, at least in their first attempt. During the first Trump administration, for instance, there was no major deportation effort. The second time, Trump is delivering on that: immigration has fallen dramatically and deportations are proceeding at scale. He learned from the first term and delivered on those promises in the second.

Something similar may be happening in Europe. Often, in their first time in power, right-wing populists don’t really deliver. We have seen this with right-wing populists in Austria, with Meloni, and others. Mainstream parties therefore have a chance to deliver now, or even if populists come to power, to use that period to develop a new strategy and new policy positions. If mainstream parties then win the election after the populists leave power, they have another chance. But I do think it is an increasingly difficult situation for mainstream parties across Europe. They have one or two election cycles to deliver. If they don’t, populist parties could, in the medium term, simply outperform them and come to dominate European policymaking.

Mounk: What do you say to people who claim that when mainstream parties try to emulate populists, it actually only reinforces the populists? This is a strand of research that I have seen repeatedly, and more than anything else it seems to be conventional wisdom among a lot of political scientists—that there are these studies, sometimes somewhat dubious studies I think, that demonstrate that when mainstream political parties start to use the verbal register of populists or emulate some of their policy positions, voters are simply going to vote for the original. This is an argument you hear a lot. How convincing do you find it, and why does it sound like you don’t agree with it?

Guenther: This depends a lot on whether you deliver results or whether you do something else—which is why I made the rhetoric versus results distinction. There are a lot of studies on this and they show mixed results. Some find that people vote for the original populist party if a mainstream party moves toward the populists. An example is the analysis of the Starmer “island of strangers” speech that I alluded to before. Other papers show the opposite: that if a party moves to the right on immigration, voters shift toward that party and away from the populists. There is a study in Denmark that shows this, and we also have a study done in Germany just before the most recent election that finds that if the CDU moves to the right and fills the representation gap, the AfD loses voters and the CDU gains voters.

I do think it depends, and it depends most strongly on whether actual policy positions change. If the policy positions really change and the output is delivered, people do not simply vote for the original.

One thing that is often forgotten in this context is that right-wing populism in Germany is only in some sense a third wave. There were early right-wing populists just after the foundation of the German republic who were strong in the 1950s. Then there were the Republikaner—the Republicans—in Germany in the 1980s, who were in some sense similar to the AfD. In both cases, the CDU—the centre-right party—accommodated them, moved toward their positions, and in the 1950s under Adenauer even gave them ministerial posts. Later, in the context of the asylum wave from Yugoslavia in the late 1980s and early 1990s, immigration had become a major issue and people wanted less of it, and the constitution was actually changed to limit asylum immigration significantly.

Mounk: Your point is that it worked at the time—the Republikaner really were a kind of insurgent populist force in the 1980s and 1990s. They had some significant successes, ended up being represented in a number of state parliaments, but never quite made it into the Bundestag. What you are saying, I take it, is that the Christian Democrats moved to the right in order to limit the oxygen for that political party. Rather than people saying that this somehow reinforced their impression that the Republikaner were onto something and they should vote for them, they said that some of their concerns were being taken care of by the political party they traditionally voted for and moved back toward voting for the Christian Democrats.

On this telling, it was Angela Merkel’s very deliberate strategy of capturing the political center—and no longer covering the right of politics—that allowed this insurgent movement to establish itself as a permanent political force. There are two ways of thinking about this. Strauss, a very influential Bavarian prime minister, always used to say that between the Christian Democrats and the right, nothing should be able to fit—meaning that he always wanted to make sure that the Christian Democrats, as a democratic political party, would cover the flank on the right far enough to prevent any party to their right from establishing itself. Merkel’s approach was very different: she would make the Christian Democrats into the party of the political center. But of course, that raises the question of who deeply conservative people—those who don’t think of themselves as being in the political center—should be voting for.

One way I have sometimes put this about German politics—and I am obviously a German citizen—is the following. I have been politically socialized on the left and continue to think of myself as being on the left, even though I have significant criticisms of the shape that the left is taking in many countries today. If I look at Angela Merkel and find that I share a lot of her basic value coordinates—even if I think she made a lot of bad decisions—we have a problem, because there are a lot of people in Germany who are considerably more conservative than I am. If the leader of the most conservative democratic political party in Germany holds views that someone on the left feels broadly comfortable with, that may be fine as long as that person is in charge. But it leaves a huge segment of the German electorate without political representation. What happens? They go and vote for the AfD.

There is another line of argument, which is that left-wing political parties in particular have not made sufficiently large redistributive demands. There was a period in which the left and the right had very similar positions on economic policy. If you look at Germany today—the Social Democrats and the Christian Democrats—or at Britain—Labour and the Conservatives—they are not that far apart on a lot of economic policy issues. People therefore no longer invest significance in economic policy, and that is why cultural issues become so salient. Why are we debating immigration? Because if there are no deep ideological differences in economic policy, that is the only thing left to debate. On this argument, what is really explaining the rise of right-wing political parties is the convergence on economic policy, and one way to fight that—particularly for left-wing political parties—is to move to the left on economics, making that a more charged political issue and shifting the public debate away from issues like immigration. This is an argument that has been made by a number of political scientists and that I often hear in the debate. I imagine that you are skeptical about that line of argument as well.

Guenther: I am indeed skeptical about that line of argument. It leads to an interesting deeper question, which I will address first. I do believe that in principle the theoretical argument is correct. All else being equal, if left-wing parties moved further to the left, we would probably talk more about economic topics and, because people’s attention is limited, they would probably think a bit less about cultural issues. However, I would guess—and I don’t think we have particularly strong studies on the relative importance of this effect—that it is a relatively minor thing. That is my main criticism. I’m sure the effect probably exists; all kinds of effects exist. But how big is it?

I don’t think people will stop caring about immigration because talk shows discuss economics more, or because the policy options on offer are somewhat more distinct from each other. This belief is driven by two things. The first is simply talking to Germans. I am German, most of my friends are German, and I do a lot of interviews and survey work where people tell their stories. Immigration and issues related to it is such a huge issue, and touches on such fundamental fears about identity and belonging, that I think it will always be salient for these people as long as immigration is high or looks like it could increase further.

The second thing is that if you look at the immigration that people say they care about—specifically immigration from the Middle East and Africa—the actual numbers of arrivals from these groups are extremely highly correlated with how much people care about the issue. During the refugee crisis, for instance, people were asked how important immigration was to them, and that line tracks extremely closely with the actual number of asylum applications in Germany. This suggests that concern about immigration is tied to things that really happen on the ground.

Notably—and this is an argument that is often made, but I think it is close to a straw man—how much people care about immigration is not really predicted by overall immigration numbers, that is, by how many people come to Germany in total. Most people don’t really care about, say, a French student coming to Germany to study. It is about very specific groups, and the correlation with real-world arrivals of those groups is clearly there.

Mounk: You said a little while ago that perhaps the last chance for established political parties to win the race against the populists is not just to shift their rhetoric, but to actually show results. The problem is that especially in systems of proportional representation, this becomes more and more difficult the higher the share of parliamentary seats populists hold. In Germany now, it is barely imaginable that there will be a right-wing governing majority in the Bundestag that excludes left-wing political parties while also maintaining the Cordon Sanitaire—the Brandmauer—the separation between the traditional democratic parties and the new right-wing populists.

The last German government was a left-leaning coalition in which the Social Democrats and the Greens governed alongside the right-of-center Liberal Party, the FDP. Now there is a conservative chancellor in power who is in various ways more conservative than Angela Merkel, but his coalition partner is the Social Democrats, the left-of-center political party. Even though Merz was talking a great deal about the need to curb immigration during the election campaign, it is actually very difficult for him to deliver on those policies because his coalition partners in the Social Democrats are opposed to many of them.

Do you think realistically that traditional political parties are going to be able to rein in immigration to a sufficient extent to make voters feel that their preferences are being represented? Or do you think we are now in a structural situation in which the inability of mainstream political parties to deal with that issue is simply going to lead to their continued decline in vote share and to parties like the AfD continuing to grow?

Guenther: This depends a lot on the country. In Germany, also for historical reasons, the resistance to accommodation is particularly high, even though as we discussed, it did happen in past episodes. I would guess—and this has to be taken with a grain of salt—that the current coalition has the last real opportunity to deliver the legislation needed to genuinely reduce asylum immigration and pursue deportation efforts.

After that, I would guess the AfD will be stronger in the next election than in this one, and it will then be very difficult to find a coalition that addresses immigration effectively while leaving the AfD aside. If efforts don’t increase significantly—and I would guess they won’t—the next government will probably also fail to deliver, the AfD will be even stronger, and it will probably eventually become part of the governing coalition. That is my guess. We can check in perhaps nine or ten years.

For other countries, the picture is a bit different. Populists are participating in or supporting governments in several countries. In Sweden, for instance, with the support of the Sweden Democrats, immigration policy has changed quite significantly—net asylum immigration is now reportedly negative. It is possible. Germany tends to look mostly at itself, but in other countries, such as Denmark, these things have happened.

Mounk: Let me ask you about how this theory applies in some other contexts. We haven’t talked much about the United States. America actually, on a lot of the generic questions about immigration that are asked cross-nationally, continues to have more pro-immigrant attitudes than most European countries, which is interesting. The failure of the Biden administration to deal with the southern border with Mexico is clearly one of the big reasons why Donald Trump won reelection in 2024. It was even one of the reasons why new voting groups like Latinos shifted to the Republican Party. At the same time, as you said earlier, the current administration is in some sense very successful in its policies—it has reduced the number of immigrants coming to the United States very significantly. But it has done that using cruel and indiscriminate tactics that turn out to be very unpopular. The immigration enforcement agency is very unpopular, and Donald Trump’s approval ratings on immigration are quite low.

Is that an American specificity, where American attitudes about immigration are simply somewhat more permissive than European ones? Or do you think that if, say, Marine Le Pen—or if she is not allowed to run, Jordan Bardella—became the next president of France and effectively cracked down on immigration, deporting large numbers of undocumented people, they might face a backlash as well? Not just from people with more progressive attitudes who become activated by those policies, but perhaps from some of the swing voters who helped them get into power in the first place—as, according to polls, seems to be happening in the United States.

Guenther: You can always overdo it. The vast majority of Europeans want lower immigration levels, but you can also have levels that are too low. The amount of cruelty—for lack of a better word—that voters are willing to tolerate is certainly limited. I do think that voters will support severe actions to reduce immigration, but there is some limit. The threshold in Europe might be a bit higher than in the United States, so Europeans might be willing to accept more severe measures than Americans, but even there some limit exists. It is possible that populists, when they come to power, overdo it to some extent—one could argue that Trump somewhat overdid it.

Relatedly—and this is a slightly different point—it also depends on how efficiently these policies are carried out. Some degree of severity may be necessary to achieve a goal, but there is also cruelty that is arguably indiscriminate and serves no purpose in achieving the desired outcome. I would guess that because populists are considerably less experienced, they will tend to be less efficient. There will be problems when they come to power and pursue immigration policies. This is something that mainstream parties can exploit by shifting their positions but then arguing that they are the professional alternative—that they will pursue these goals in a more targeted and discriminate way, and that they won’t overdo it. I think that is one way that mainstream parties can make their case.

In the rest of this conversation, Yascha and Laurenz discuss whether journalists and politicians have as much impact on public thought as they think they do, to what extent self-censorship has increased in recent years, and the impact AI will have on the media. This part of the conversation is reserved for paying subscribers…

Read more

In his recent interview with the Financial Times, Poland’s prime minister, Donald Tusk, startled many by predicting that Russia might soon test NATO’s resolve to defend its allies. Yet what has ruffled feathers even more was his question about the “loyalty” of the United States as an ally.

Given Poland’s steadfastness over the past thirty years, it is easy to sympathize with Tusk. Regardless of who holds power in Warsaw, Poland has always taken its defense seriously. It has bought U.S.-made military equipment and welcomed U.S. investment—including in nuclear energy.

There is very little that the current administration has done in return for highly reliable allies such as Poland—notwithstanding the recent U.S.-brokered release of a Polish journalist held by Belarus. Polish exports to the United States face arbitrary tariffs. Washington has stopped almost all of its aid to Ukraine, whose victory remains an existential question for Poles. More recently, the U.S. president has picked NATO as a scapegoat for the failures of his ill-conceived war against Iran.

Yet Tusk is not a public intellectual, free to air musings about his country’s relationship with the United States. His task as prime minister is different. First, it is to ensure that Poland is ready for less-than-ideal scenarios in the U.S.-European relationship. Second, it is to help prevent the worst-case scenarios from materializing.

As a result, leaders like Tusk must resist the magnetic pull of the grotesque show that Donald Trump is putting on in the United States. It is easy to score political points at home by confronting Trump. Yet the further east one goes, the higher are the stakes of making the U.S.-Europe relationship a part of everyday political life. That does not imply rolling over in the face of the U.S. president’s antics. Quite the contrary: Europeans must be ready to push back whenever their interests are threatened, as they did when the Trump administration readied to take over Greenland.

Subscribe now

It is also conceivable that voicing questions about the U.S. commitment to Europe, as Tusk did, will help mobilize the European (or Polish) public to do more to prepare for scenarios in which Europe cannot rely on the United States. Yet it can also act as a self-fulfilling prophecy, as any unforced fights with Trump risk accelerating America’s drift away from the continent.

To be sure, the odds that the current presidency is just a fleeting episode in U.S. history, to be followed by a swing back to the post-war norm, are slim. Hope, moreover, is never a substitute for strategy. Nor should one believe in the mistaken trope of “transactionalism” as a way of placating Trump. As Ukrainians have learned from their critical minerals deal, mutually beneficial business relations are no guarantee of staying in the good graces of the U.S. administration. Similarly, billions spent by European militaries on future deliveries of F-35 fighters may or may not be justified on defense grounds. However, one should not live under the illusion that such purchases would prompt Trump’s United States to rush to the defense of Europe.

Rather, the reason for restraint in rhetorical and other confrontation with the United States is that adaptation to America’s reduced reliability and wild swings in its politics is a long-term project, requiring not just unprecedented investment in defense but also the building of new institutions and new political projects. That takes time, intellectual leadership, and a willingness to look dispassionately beyond the immediate horizon, instead of just reacting to the latest outrage or international crisis provoked by Trump.

Why make institutions a part of the equation? Well, if the current moment teaches anything of value to international relations theorists, it is that institutions—understood both as organizations and as rules of the game—matter. Personalist alignments, flattery, bribes—or “transactionalism”—are no foundation for a stable international system. Never mind the one-sided deals meant to mollify Trump or outright bribes offered by the likes of Qatar and UAE; the U.S. president has only one loyalty—and it is to himself.

And no matter how skillful the likes of Alexander Stubb, Giorgia Meloni, or Mark Rutte have been, their sweet talk has not fundamentally altered the course that the U.S. administration is on. It is only a matter of time until the glamour of King Charles’ recent visit to Washington will be superseded by yet another episode of the administration’s lashing out against the UK and its government.

As Europeans look ahead, seeking to put the relationship with the United States on a more durable footing, they will have no choice but to think about institutions and institution-building. That means building structures of political, defense, and economic cooperation that can deliver in the absence of U.S. leadership and that could perhaps even provide a deterrent against America’s predatory behavior in the future. It also means building institutions that America would want to be a part of, allowing them to constrain and moderate its behavior in future political cycles.

Subscribe now

The reason is simple. Institutions are sticky and they often survive under mercurial and unpredictable leaders. One tends to think, for example, of the post-war ecosystem of international organizations and treaties as hopelessly outdated—just waiting to be shattered by a Trump-like figure. Yet it should give one pause that ten years into Trump’s full-frontal assault, many of the relevant institutions still persist, weakened though they are.

Few in the Republican Party harbor any love of multilateral institutions; yet the United States has (for the most part) remained a part of the UN system, a member of NATO, and a party to countless international treaties and conventions.

In Washington, the International Monetary Fund and the World Bank are humming along, with a staff who may be a little confused about the new realities but who certainly do not feel under any existential threat. Sometimes, even Trump and his team seem to recognize the usefulness of the existing institutional architecture. More importantly perhaps, even for the Trump administration, it may not always be an appealing proposition to just smash things while being patently unable to come up with alternatives.

The Five Eyes alliance continues to function despite intelligence-sharing tensions. Meanwhile, the dollar-denominated international financial architecture appears resilient. And even if the United States withdrew from NATO tomorrow—reducing its own ability to project power globally through its European bases—the alliance would still be extremely useful to Europeans as a platform for ensuring interoperability, joint planning, and socializing allied militaries into fighting together.

Yet fundamental realities about the world have changed relative to the post-war era in which such international institutions, treaties, and alliances were being created. And unless those institutions adapt effectively, they will eventually be hollowed out and replaced by new ones. For all of the U.S. administration’s ham-fistedness, there is thus a grain of wisdom in the quest to build a “NATO 3.0,” or to create new structures of cooperation around critical supply chains.

It is an open question how effective the United States can be at institution building anytime soon, considering the magnitude of its domestic political crisis. But that provides an opening for others—perhaps for Mark Carney’s middle powers, perhaps for the EU, perhaps for China—to think about how the international arena could be structured by new rules and new political projects, sometimes working around an unreliable United States, rather than with it or under its leadership. Europeans must be at the forefront of that exercise, both at home (especially in updating and strengthening the EU) and in their engagement with the wider world.

Concrete moves underway include the expansion of the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP); the EU-Mercosur trade deal with Latin America; strengthening of the Canada–EU strategic partnership; and, in the context of European security, the Coalition of the Willing to support Ukraine and the Joint Expeditionary Force of Northern European nations.

“We have all the institutions we need,” one European diplomat claimed during a panel I moderated. “We just have to make the existing ones work.” But that betrays a deeply ahistorical view of our situation. States, governance structures, alliances, and institutions have changed constantly through human history—sometimes through war and chaos and sometimes through ordered, democratic change. The Holy Roman Empire, the dual monarchy of Austria and Hungary, Czechoslovakia, the Polish-Lithuanian Commonwealth, and the Soviet Union all seemed like basic, immutable artefacts—until they weren’t.

Tusk asked whether the United States is still a loyal ally. It’s a valid question, but not one for Poland’s prime minister to ask aloud. The right question—quieter, harder, and answerable only over years—is what institutions Europe could build that would make American loyalty matter less. In a few years, Donald Trump will be a memory. The institutional solutions that Europe and its partners come up with—or fail to come up with—in response to the change in America’s role in the world will shape the continent’s geopolitical future for generations.

Dalibor Rohac is a senior fellow at the American Enterprise Institute in Washington, D.C. and a contributing editor at American Purpose.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Earlier this year, a prominent company with millions of customers announced a major product upgrade—albeit with one little catch.

If this new product was released to the public, the company said, it could be used to disrupt—and perhaps destroy—civilizational infrastructure, from financial markets to transportation systems to power and water utilities.

But fear not! The company hastened to reassure the public that it had the situation under control. The company would decide, on its own terms, what the world needed to know, who should be called in to contain the problem, and how much gratitude the rest of us should feel for being spared a catastrophe we never knew was coming. No public accountability or government intervention required.

This, of course, is the story of Anthropic and its latest AI model.

Anthropic discovered that the model, known as Mythos, could autonomously identify zero-day vulnerabilities—that is, security flaws that software makers don’t know exist—across every major operating system and web browser. Some of the flaws Mythos found were decades old, overlooked and unnoticed by literally millions of human eyes. This was not an intended feature, but one that the AI seems to have picked up along the way, as Anthropic’s developers rushed to create a more powerful model with better reasoning and coding abilities.

Intentional or not, it introduced a substantial new danger to the world. In the wrong hands, Mythos could be a weapon fit for a supervillain—a cheat code for attacking the world’s most critical infrastructure.

And yet, the decision to build such an advanced model was not made by any external agency. No independent body evaluated it. No regulator was notified in advance.

And once the threat was identified, Anthropic decided—alone—what to do about it. After judging Mythos too dangerous for public release, Anthropic created a private consortium made up of handpicked partners like Amazon, Apple, Cisco, JPMorgan Chase, and Nvidia to fix the bugs and ensure Mythos’ safety.

Subscribe now

With that all worked out, they gave policymakers and the public a heads-up on their dangerous new product and the plan to contain it.

This is what passes for AI governance in 2026: a single company accidentally builds an entity powerful enough to pose an existential threat to the digital systems that power modern life, unilaterally decides how to deal with it, and then loops in everyone else.

Except of course, it’s not at all clear that they’re dealing with it: A few weeks after all this transpired, we learn that Mythos was, in fact, accessed by unauthorized users. Was catastrophe avoided, or merely delayed? We may yet find out.

Mythos is the clearest evidence yet that our system for developing, assessing, and disseminating powerful AI systems is dangerously dysfunctional.

As tempting as it is to blame this dysfunction on bad actors or rogue tech CEOs, I think it’s something deeper than that: a broken incentive structure. As careless as their actions may sometimes seem, AI developers aren’t being intentionally malevolent—they’re rationally operating within a system that rewards chasing progress now and worrying about consequences later.

The leading AI companies, armed with billions in capital, are all sprinting toward the same horizon with an imperative to cross the finish line first. They all have the same motivation: “If I don’t build it, someone else will.”

That logic co-exists with a genuine belief that AI may prove to be a transformative force for good, generating productivity in unimagined new ways and pointing the way forward for progress. AI’s potential benefits have been exhaustively documented—whether to address climate change or to enhance medicine or simply to widen our horizons—but at this stage in the AI era, we all have to acknowledge that AI is accompanied by myriad harms, from job loss to manipulative engagement to cognitive offloading to AI psychosis to AI-assisted suicide and murder.

The scale of these numerous challenges demands a response as wide and deep as our society. One self-interested company or a hand-picked corporate consortium can’t be trusted to get it right—the issue is far larger than that. The solution, should we get there, will require public understanding and engagement, and government oversight.

Subscribe now

To those that claim AI is too complex, too consequential, or too powerful to govern: you’re wrong. In reality, this argument is—at best—a shoddy defense of the broken incentive structure producing it.

Because AI is complex, we have a responsibility to comprehend it. And because AI is so consequential, we have a responsibility to govern it. Institutions, policymakers, and regulators have been understandably disoriented by the AI frenzy of the last few years, but now must rise above the noise and rewrite misaligned incentives. That means—yes—establishing a role for government in the AI sphere. Concerns about governmental efficacy are understandable, but government must be meaningfully engaged. There simply is no other manifestation of the will of the public.

We have governed consequential technologies before: automobiles, aviation, pharmaceuticals, nuclear energy, and more. Every one of these industries today operates inside a hard-won system of accountability—a system that took time to build but, crucially, did not kill innovation. It’s time to apply the same rules and accountability structures to AI, and with even more urgency, considering how quickly it is integrating into virtually every aspect of our society.

And the fact is, no meaningful federal regulation of AI currently exists. States have stepped up to fill the void, with 73 AI laws—ranging from protecting kids online to ensuring a human is in the loop when it comes to critical decisions like healthcare—enacted across 27 states in 2025. But states’ reach is increasingly limited, with Trump issuing an executive order in December directed against “excessive state regulation.” The tech industry, meanwhile, has worked to paralyze regulation at every turn, with AI companies pouring money into Super PACs to support tech-friendly candidates and block state regulatory laws.

So what could a meaningful regulatory structure actually look like, assuming the political will for it materialized? Let’s take Mythos as a test case.

Under a more rational governance framework, a tool with society-altering capabilities like that of AI would face mandatory pre-deployment testing by independent evaluators—not the company selling the product.

There would be standardized public reporting of risks, so that regulators, businesses, and users could make informed decisions rather than relying on what the developer chooses to disclose. There would be real whistleblower protections for employees inside AI labs who see something wrong and want to say so.

And if an AI product caused foreseeable harm after its release, the company that built and deployed it would bear legal responsibility. Liability is what aligns private incentives with public safety. It’s why cars have seatbelts and airbags—not because manufacturers wanted them, but because they knew they would pay the price for cutting corners and because insurers and legislators aggressively pushed safety measures. The same logic applies here.

These two principles—safety and transparency before deployment; and a genuine duty of care to the public—are key to establishing a framework for orienting policymakers, companies, and citizens towards what responsible AI actually requires.

Subscribe now

None of this is radical. It’s all standard with existing products. And all of it is overdue.

Mythos is just the latest and most egregious evidence that we cannot keep relying on the judgment of individual companies to stand in for the public accountability structures we’ve so far refused to build around AI. The next threat may not be discovered in time. Or it might come from a company more desperate to succeed in an incentive structure that rewards reckless behavior.

We’ve done this before. We have the tools. It’s time we reclaim our future with principles that will protect us, individually and collectively.

Julie Guirado is the Executive Director of the Center for Humane Technology and oversees its AI Roadmap.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

This week, Richard talks to Odd Arne Westad, author of The Coming Storm: Power, Conflict, and Warnings from History (Henry Holt).

Host: Richard Aldous

Producer: Bhaasita Athani

Follow Persuasion on X, LinkedIn, Instagram, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below. Bookstack listeners can take advantage of a 20% offer for paid Persuasion membership, with access to special columns and bonus Good Fight with Yascha Mounk podcast content.

20% off for Bookstack listeners!

Lant Pritchett is a development economist from Idaho. Having now thrice retired, he is currently a Visiting Professor at the London School of Economics in the School of Public Policy and the co-founder and Research Director of Labor Mobility Partnerships (LaMP).

In this week’s conversation, Yascha Mounk and Lant Pritchett discuss why the traditional foreign aid approach to development is fundamentally misguided, how countries actually achieve prosperity through organic national transformation, and whether the classic path to development remains viable in the 21st century.

Will you be in London on Sunday, September 6? I’ll be interviewing Francis Fukuyama about his life and thought to mark the publication of his memoir In the Realm of the Last Man at the Sekforde at 5pm. Find out more and book tickets here. Paying subscribers can access a code for free tickets here. —Yascha

This transcript has been condensed and lightly edited for clarity.

Yascha Mounk: When I talk to people who care a lot about economic development, they love to talk about inclusive development, about sustainable development. They tend to focus on how much more affluent countries should donate to less affluent countries, perhaps figuring out what the best intervention is—whether foreign aid should be spent on this kind of thing or on that kind of thing. Not to simplify, but you think that whole approach is basically wrong. Why is that?

Lant Pritchett: What I think of when I think of development is what we call “getting to Denmark.” There was a historical process whereby many countries—not just Western countries, but other countries—had a fourfold transformation. They had a transformation from a low productivity to a high productivity economy, and it was mostly broad-based. That is economic growth, and inclusive economic growth, if you want to add the adjective. They also went through a transformation of acquiring state capability—the ability of the public sector to do things that needed to be done, like regulation and providing certain services. They also went through a transformation from subject to citizen, to a polity that was based on responsiveness to the needs and wishes of the citizen rather than vice versa. They also went through this harder-to-describe transition of equality under the rule of law, whereby kith and kin and other identities became reduced in importance and everybody was treated equally.

That is what development meant in the post-decolonization era, after World War II, as countries became independent from their colonial overlords. It meant this big fourfold transformation. Foreign aid can be modestly helpful with that, but it is not very central to it. The more you think about donors and what donations should be and what the right interventions are, the more you lose the plot.

In my papers, I show that if you get to what I call national development—this fourfold process—that is a machinery for producing good things. We are worried about whether people have clean water, decent housing, and all of these material things that are really good for people to have. But if you get to national development, you get to that, and vice versa.

There is a big question of whether what we talk about as foreign aid is aid to the process of national development, or whether it is aid to mitigating the worst consequences for human well-being of that lack of national development. The latter is not a strategy for the former. By and large, a lot of the development agencies lost the plot completely and became essentially charity organizations focused on mitigating the worst consequences of the fact that many countries had not acquired national development.

There is nothing wrong with that. Mitigating terrible things that happen to human beings because they live in a country that is underdeveloped is a good thing. But it is not development. Too much attention to the latter detracts attention from the core issue.

Mounk: Is the concern here simply that this money is being spent on things that don’t actually help to solve a problem in the long term—that the most effective use of that money would be to invest in things that actually help solve these underlying problems, and therefore make the country more affluent so there is not as much poverty or human desperation that you need to buffer with these donations? Or is it more profound than that—that those donor dollars in some ways make it harder for that process to take place, that they might in some complicated way backfire? Is the concern mostly about efforts wasted, or efforts that might in some complicated way impede the country that is supposedly being helped from solving its long-term problems?

Pritchett: The big problem is more about ideas than it is about the concrete use of money. I have what I call the “bird on the elephant” theory of development. Development agencies are spending all this money, financing projects—some of them economic projects like roads and bridges and power plants, and some of them charity-like projects funding health interventions and the like. But that is in some sense secondary to the fact that this creates a global discourse about development and about how countries can do development.

I am much more worried about the waste of effort on ideas. We have geniuses—truly stunning geniuses—devoting themselves to charity work as opposed to thinking about development strategy. Ideas are supremely important to the fate of nations, and the ideas that get transmitted via a global discourse of research and practice, to government officials, to people in power, to people who have influence—that is a huge deal. Obviously the most consequential thing that has happened in the last 50 years is the leadership of China changing its mind about what to do in China and how to make China a better place to live. That was, fundamentally, a change of ideas. By losing the plot on national development in favor of mitigation, we also draw the discourse, the research, and the ideas away from big questions: How do we get states to be more capable? What is the right sequencing of state capability and democracy? How does democracy interact with the creation of economic growth—does it impede it or not? The number of people in the world who can produce new, original, and correct ideas is very few, and drawing those people onto small issues is a huge loss.

I don’t think aid is often counterproductive. Some of my friends—Bill Easterly, and Angus Deaton—think that aid can sometimes foster mentalities and practices that impede development. I think the elephant is mostly neutral. But if the bird, who sits on top of the elephant and sees what is going on, warns the herd of impending dangers, and can provide a vision—if that gets messed up, then the whole elephant is diminished.

Mounk: Let’s take a step back. One of the intuitive things about what you’re saying is that the United States, the United Kingdom, and France didn’t become rich because some much richer countries said, we’re going to give you a bunch of development aid. These were mostly internal processes, obviously with an international component—trade with each other and so on. So it stands to reason that if we want to think about how countries like India or Kenya might become rich in the future, it would probably be by following some of the same kind of processes.

Now, a certain kind of progressive critic would respond by saying that there is an economic structure in the world and that these countries are in some ways interdependent. If you’re one of the most developed countries in the world, you can specialize in high-return services, for example. If you are much poorer than those countries, then you can’t follow the same development path as those that developed historically, because you occupy a different niche in the economy. The other difference, of course, is just the stage of development. A lot of the countries that grew rich in the 19th and early 20th century did so through industry and manufacturing. But nowadays most wealth is not created in factories—it’s created in the knowledge sector, in the service industries. Perhaps the same kind of path to development just isn’t open anymore. When I was in India, a lot of people were worried about whether the traditional path to development is still available in the 21st century.

What do you think about that? Can you broadly follow the same playbook that made the rich countries of the world rich in the past? Or do we actually need to look for a different path, either because of a change in the nature of the world economy or because of the different relative standing of the poorest countries in the world within it?

Pritchett: Your question brought up two very different strands that we shouldn’t conflate. One is the path being organic and being driven by a country’s own dynamic. If you take, say, Dieter Mikloski’s work, the key to all of this was creating an environment in which you can give it a go. That is distinct from the question of specifically what, economically, a country is going to do. The question of whether the path that Denmark took is open to Kenya has two very different components.

We hope you’re enjoying the podcast! If you’re a paying subscriber, you can set up the premium feed on your favorite podcast app at writing.yaschamounk.com/listen. This will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!

Subscribe now

Set Up Podcast

If you have any questions or issues setting up the full podcast feed on a third-party app, please email leonora.barclay@persuasion.community

One is the deeper question: can Kenya develop an organic dynamic within its polity, society, and economy such that the actors find a path for Kenya to become prosperous? That is the true endogenous process. The second question is, when Kenya organizes itself to find that path, will it look like the specific economic path that Denmark followed? The answer to the latter is certainly no—and no in both a good way and a more challenging way.

The first way the answer is no: what happened—and I wrote a paper a long time ago called “Divergence Big Time” that emphasized this—is that the rich countries in 1870 were not that advanced relative to the most lagging countries, because there is a floor. There is only so poor you can be. What has happened in the world is that the rich countries have done this through exponential growth. They have created the basic hockey stick graph.

The rich countries collectively grew at 2% a year for 120 years, and the power of compound exponential growth means they are manifold richer than they were in 1870. But in the process, they invented and discovered a great many things—in science, in practice, and in all kinds of domains—that weren’t available in the world in 1870 but are available in 1970 or 2026. This means the countries that enter this process later have the potential to grow incredibly faster than any rich country grew.

None of the countries that were rich in 1970 or 1980 were rich because they grew fast. They were rich because they grew steadily. But that created the possibility that later countries could accelerate their growth enormously. The fast-growing countries in the world are growing incredibly faster than they did historically. China, Korea, Vietnam—growing at 6% per capita wasn’t an option for Denmark, which just had to stay on 2%. That is the good news. Countries that, in whatever way, manage this endogenous process that creates an organic drive for prosperity can discover ways to grow incredibly faster than was available to the old countries that had to make their own way.

We have seen incredibly good news, and we don’t want to lose the plot on that. The years since World War II have been the best years for improvement of the material condition of humankind by a multiple of any prior period. I work some on education. The average person in the developing world had roughly two years of education in 1950 and has eight now. From whoever your mythic forebears were—Adam and Eve, let’s say—to 1950, humanity had accumulated two years of education. In just 60 years, it added six. Three times more education was added in 60 years than in all of human history combined. That is true of health. That is true of a whole range of things. Just fantastic progress.

Take child mortality. There were all kinds of countries in the world 50 years ago where basically one in five children was dying before the age of five—a rate of around 200 per thousand. Worldwide, that is now down to around 30. Many people in the world in 1950 were living in material conditions not that much different from ancient Greece or ancient Egypt. Even today, if you compare Egypt in AD 1—when Joseph and Mary went to visit—to some of the poorest African countries today, they are at roughly the same level of GDP. For a very long period of human history there was very little progress, and then it accelerated dramatically. From all of human history, we got to still one in five children dying around 1960, and now that figure has fallen enormously. If you look at the people with access to electric power, people with access to water and sanitation—it has just been an amazingly good run.

These questions always start from the assumption that the global order is preventing progress, and that is just surreal when you look at the world over the last 60 years. There has been amazing progress in many places on many things, and this isn’t just a matter of measuring GDP.

Mounk: One illustration of that from my own life: when I was an undergraduate in England in the early 2000s, there were big debates about the World Trade Organization. The shape of the debate, and I remember this vividly, was always whether this was going to screw over China and India—whether the World Trade Organization was just a smart way for the rich countries of the world to keep the poor countries from developing. Today, when you look at the critiques of letting China into the WTO, it is exactly the opposite. The question is whether America screwed itself over, or screwed its working class over, by letting China in—which led to a huge increase in wealth in China and rapid deindustrialization in parts of the US.

You can take seriously the idea that there were some mistakes made in how that was done and what impact it had on the working class in Michigan, Pennsylvania, and other places. But if you have to choose, from the standpoint of humanity, between those two scenarios, we have ended up with the much better one. We have lifted hundreds of millions of people out of poverty to genuine middle-class status in China. Perhaps we could have done that a little bit better, with less disruptive consequences for the already pretty affluent people in the United States. Perhaps that’s a red herring. But it is striking to what extent the progressive case against things like the WTO in 2000 has completely flipped on its head in terms of how we talk about it today.

Pritchett: One of my favorite colleagues and friends at the Kennedy School was Dani Rodrik. If you look at Dani Rodrik’s intellectual trajectory, his most recent book is about shared prosperity, which is hugely concerned about the prosperity of the middle class in America and how the global system hasn’t been good for it—versus his earlier concerns about the WTO and whether its trade structure was truly open to facilitating the growth of poor countries. It turns out we had the opposite problem from the one we thought we were going to have.

Absolute poverty in the world has just amazingly declined. Every discussion of the world and how it’s going needs to start from that factual basis. Hans Rosling wrote a book called Factfulness—the fact is, on nearly every measure of material well-being, things are just fantastically better. In many dimensions, by the way, better than we would have expected even from the economic growth that we got. Sometimes people say, you economists focus just on economics, and we really should have focused on these other things. But if you look at the expansion of schooling, for instance, schooling expanded not by less than we would have expected from the economic growth we got, but way more. Just getting kids into school was one of the most phenomenally transformative things in history, and it happened more than we would have expected given the growth, not less. All of development was focused on expanding schooling. So you cannot say that economists focused on growth at the expense of these other things. That is just not true.

Mounk: Let me push you on that—it’s a question I was going to pose in any case. There are a lot of people who are very skeptical of GDP as a metric. They think we can be incredibly rich and yet people are miserable. Society can be very rich in the aggregate, but only because a few people are incredibly rich and everybody else is incredibly poor. There are these standard examples that come up in conversation all the time. If I’m stuck in a traffic jam burning gas and not getting anywhere, that’s increasing GDP—so GDP is a really bad metric.

You believe that GDP is in fact a very good metric of human well-being and that it correlates very strongly with things that we care about more directly. When listeners next encounter somebody who says GDP is a terrible metric we shouldn’t care about, what should they respond?

Pritchett: I want to be clear: I don’t think GDP is a good metric of human well-being. GDP is a very good metric of the net production of a society, and that production creates the material basis for human well-being. GDP per capita is just factually very highly correlated with nearly everything we care about in terms of material well-being.

Subscribe now

But here is the response you should make. The relationship between GDP and most things we care about in terms of material well-being is concave—meaning those things get better as GDP gets better, starting from poor levels to middle levels. By the time you reach a GDP per capita of roughly $40,000, you have met most of the basics of material well-being, and hence the relationship flattens out. It doesn’t go away necessarily, but it does flatten out.

The real danger in the world discourse is that people unhappy sitting in traffic in Luxembourg—which I use as an example because the only time I went to Luxembourg I got caught in traffic—are right that their overall well-being is not that highly correlated with GDP, because their GDP is so high in the first place. But we shouldn’t extrapolate that back.

Mounk: To put this point a little more polemically: it is very easy—and I grew up in that kind of milieu, not particularly affluent, but among artists and musicians in Germany living in pleasant towns—for people to say that the important thing in life isn’t to drive a big car and go on a fancy holiday, that there are things much more important than material well-being. Of course, if the floor of your society is that you have a decent apartment with heating and running water and you eat three meals a day, then that is probably true. But if you are sitting in a mud hut in a developing country, that is most assuredly untrue.

Pritchett: One of the fundamental insights of economics is declining marginal utility—the more you have of something, the less important it is to you. If you say the West is suffering a paucity of purpose and people are unhappy because they don’t have some driving purpose, well, that is because they have already satisfied the purpose of having material things like a hot shower, a heated home, and the ability to go 300 miles in a few hours. Whereas most of humankind, for most of history and even today, is nowhere near that.

I am happy for German artists to have angst about what their purpose in life is and whether they are really happy and whether more material goods would make them happy—and it probably won’t matter that much to them. But don’t project that back onto Africa, or conclude that India doesn’t need growth, or Bangladesh doesn’t need growth, or that we don’t really need to worry about whether those countries grow.

The right response to the claim that GDP is a bad metric for improving the human condition is to ask: where are you on this spectrum of existing progress? Since you are likely having this conversation with people who are materially fantastically well off relative to any period in human history and relative to most of humanity, they should be sensitive to the fact that yes, that may be true for them, but it is not true for six billion other people on the planet. Don’t project your life and concerns onto theirs, because you have diminishing marginal utility precisely because you have so much.

Mounk: Let’s go to those poorer people in the world. You are pointing out that GDP per capita is very strongly correlated with the other things we care about—child mortality, life expectancy, education, and so on. What does that tell us about what we should aim for? Does it suggest that we should just aim to increase GDP per capita and assume the other things are likely to follow? Or could the association go the other way around—that what we need to do is have all of those specialized interventions to improve the local hospital and the local school, and that is what will then correlate with increases in GDP? Which way around do we read the correlation?

Pritchett: This depends on where you are in the spectrum. I have done a paper where I try to disentangle this question. One of the things that GDP does, by the way, is create a broader tax base. People often get engaged in a very strange discussion as if focusing on GDP growth means ignoring the need for government services. But you cannot have government services unless governments have revenue, and government revenues are mostly tax buoyant—meaning they grow more than proportionately with growth.

Mounk: So GDP growth actually is what enables you to have all those government services?

Pritchett: At one point I did the calculation: Ethiopia’s government spending per head is around $300 per person per year. What can you do with $300 per person per year? Ethiopia’s GDP per capita is so small that if the government started taking bigger proportions of it, they would be eating into expenditures on food. Sixty percent of a poor household’s budget is on food. You cannot just say Ethiopia’s government should fund all these specific interventions—how are they supposed to do it?

When I argue in favor of GDP per capita, it is not because GDP per capita exclusively funds private goods. It is also the basis for funding public goods and public service provision. Getting that mix right is complicated, but economists are not out there saying there should be no government and everything will take care of itself if we just have high GDP per capita. The argument is that high GDP per capita reflects a high productivity economy, and a high productivity economy creates the material basis for both private and public expenditures.

National development is about the process of creating both the material possibilities and the mechanisms for doing that—hence my fourfold definition of development. You could call me a growth fanatic, but I am not a growth-only person. You also need a decent government. In all of the empirical work, you can look at Equatorial Guinea—a few kleptocrats dominating a bunch of oil, with the rest of the economy completely disarticulated from that. Can you get to high GDP per capita and still have low levels of living? Yes, if all of that is being captured by relatively few people.

Mounk: You probably need a natural resource for that as well. It would be very hard to imagine a case where 0.1% of people have an incredibly productive company that doesn’t rely on natural resources and doesn’t rely on broader education and so on. It probably takes a somewhat special case—like kleptocratic control over natural resources.

Pritchett: The way I like to describe the goal—the way we should think about how countries can make progress—is what I call “inclusion into productivity.” The reason human beings are this amazing species is that we have learned how to cooperate to create value. A large part of the development process is getting more and more sophisticated ways in which people can cooperate over time and space to create value. When I look at a great big corporation with tens of thousands of employees, this is a mechanism of cooperation to create value—a cooperation across people with all kinds of different skills and contributions.

Subscribe now

The process of development is getting more and more people out of being engaged in activities where they work a plot of land and attempt to eke out a living in low cooperation, and into more and more sophisticated value chains. That is inclusion into productivity. What I am actually concerned about is the productivity of individuals—but they are going to be more productive not by being more separated from a sophisticated modern economy, but by being more embedded into it. That dynamic, which leads to indicators of inclusive growth, is how I think about the fundamental dynamic on the growth side. Growth should be a process of more and more of the population engaged in higher and higher productivity, with more and more people included into these mechanisms.

Mounk: To go back to those two different challenges I posed earlier: one is whether poor countries can still grow at the same speed that others did, and your answer is yes—in fact, a lot of them are growing more quickly. There is a strange thing in the premise of that argument, which somehow implies that we haven’t seen examples of poor countries developing very quickly. That is partially because people who think about this tend to drop successful countries out of their sample. They don’t remember that China in 1980 was an incredibly poor country. Today we no longer think about China as being part of that sample because it has grown out of it. That in itself is evidence that rapid development is possible. If China was able to do that, there shouldn’t be an in-principle reason why Kenya or India couldn’t achieve the same feat.

The other question is what that looks like in the 21st century, when some of the historical development path is no longer available. My understanding is that your answer is going to center on those four development factors. Why are those so hard to implement in places where they are not in place? Ideas are really important, as you said earlier. If only the right people adopted the right ideas, they should be able to put those four things into place. But clearly it is more complicated than that. It is not just that the rulers of these countries have never had the right ideas. There must be obstacles beyond that.

Pritchett: One of the obstacles is that the process of growth of inclusive productivity is a transformational process. A transformational process requires winners and relative losers. In many places, economic, political, and bureaucratic structures congeal. They congeal because there is a certain way in which a country produces value, and that way of producing value often produces certain ways in which the government extracts value. That can lock into an equilibrium that resists change rather than encourages it.

There are low-level equilibrium traps—ways in which countries get stuck. We already alluded to one of those. In resource economies, if you are relatively well endowed with certain point-source resources—and by point-source I mean not land, which is diffuse and requires a geographically distributed population, but oil or diamonds—you can see how the people who mine the diamonds and the people who control the country get embedded in a relationship where, as long as that elite bargain can survive off the extraction of value from diamonds, it is not looking for anything else. It is not looking to solve the problem we are talking about, which is how a country comes to be embedded in more and more sophisticated value chains. If an elite can generate a bargain that sustains itself politically over time with just diamonds, it is not looking to do anything else, and that becomes an obstacle.

Economies and countries get stuck in an elite bargain that is more worried about the threats from new industries creating new power structures than about stagnation around the parts of the productive structure they are already in. You can easily get stuck in the dynamic between economy, polity, and bureaucracy in which an elite bargain undermines the rule of law and undermines the expansion of new opportunities, in the interest of playing defense around where they are. Finding countries that can handle this dynamic of change is very hard. That is why we see so few transformational successes, even though the global order makes them possible.

Mounk: Is there some way to change that, to contribute to that? If economists at the moment are focused far too much on whether giving people in this village one kind of intervention or people in that village another kind of intervention will lead to a little bit more growth—and you think that is fundamentally the wrong question to ask—do we have better answers to how we can help a country like Nigeria, with a very fractious political system and an elite bargain that is very bad for the country, get out of that trap?

Pritchett: Many of your questions have two sophisticated halves to them. Let me dwell on the first half—the premise that people are wasting their time. I want to emphasize that they are not just focused on the wrong question. They are focused on ontologically the wrong question.

The word ontology is not one I like to use, but what it means here, at its fundamental roots, is this: the problem with the world is not poor people. The problem is that people are in poor places. If you are studying the dynamics of how to make people less poor, you have to be ontologically studying the characteristics of the system, not the people. The methods being deployed by economists to study how to make people better off are focused as if the person were the unit at which we should be studying this. That is ontologically wrong.

Mounk: Let’s stay with this half of the question before moving on to the other half. Take a step back and explain to us what the dominant paradigm in development economics has been for the last 20 years and how that dominant paradigm is particularly vulnerable to the critique you’re making. I take it you’re mostly talking about what are called RCTs. What is an RCT? How did it become so dominant? Why do you think that’s the wrong way to ask this question?

Pritchett: If you characterize what development actors are doing as carrying out interventions, then you can get obsessed with understanding whether the particular intervention you’re undertaking has a truly causal effect. We can’t just evaluate a project on a before-and-after basis—we need to really separate out the causal impact. For separating out the causal impact, doing a randomized controlled trial is the best way to do that.

The problem is, if I’m trying to do a randomized controlled trial, I need some group of individuals to receive the treatment and some individuals to be the control group. That means I have already ontologically assumed that the important causal effects are individuated, as opposed to being country-wide or market-wide phenomena. This is as plain as I can make it.

Mounk: Let me try to put this in plain terms to see whether I understand correctly. The idea is: I study Village A and Village B, I give Village A deworming medicine and Village B nothing, or perhaps some other kind of treatment. If what I’m trying to figure out is how to spend a hundred million dollars for a charity, that is a very reasonable question to ask. The problem you are pointing to is that what actually explains why both Village A and Village B are very poor—versus Village A and Village B in England being relatively affluent—is these country-level characteristics: in particular, whether they have managed to figure out the rule of law and all the other things we have been talking about. You are defining the interesting stuff out of the question if you are just comparing two villages within the same country that are meant to be as similar as possible, and then seeing whether this or that kind of treatment is going to marginally improve the lives of people in each of them.

Pritchett: Let me take what I consider a paradigm example—one that, when you encounter it, should make you realize something has gone badly wrong. There was a paper in Nature—the most highly reputable scientific journal in the world—about carrying out an experiment in Niger where a cash transfer was given to some people and not to others. Bundled with that cash transfer was a psychosocial intervention, and the study then looked at whether this psychosocial intervention independently and causally caused individuals in Niger to see their incomes rise.

This was Nature magazine’s characterization of what development economics was doing, with 11 prominent development economists as co-authors. When you see that, you should immediately think: this is madness. People in Niger are poor because they are in Niger. Niger, on every indicator of national development, is a basket case. If you are not fixing Niger, thinking that you are doing good in the world by making tiny tweaks to psychosocial interventions at the individual level, you are assuming that a large part of the low standard of living there must be a characteristic of the people in Niger. That is just wrong—wrong by orders of magnitude. People in Niger are poor because they are in Niger. If you allow a person in Niger to move to France, their income will converge to that of people in France almost instantaneously, because France is a high productivity place and Niger is a low productivity place.

If Niger as a country had national development at the level of France, all of these problems would be addressed, because national development is a machinery for endogenously identifying and solving problems. If you don’t have that, attempting to solve these problems at the individuated level is wildly, ontologically wrong. You are not taking into account that the outcomes individuals have are determined not by their own characteristics, but by the characteristics of the political, organizational, economic, and social system they live in.

Mounk: The other thing that strikes me about this study is the theory of action lurking in the background. I have no doubt that people who encounter psychosocial problems are less productive, and I am moderately optimistic that the right kind of psychosocial intervention might reduce those problems and therefore make people more productive. But how on earth are you going to deliver psychosocial intervention at scale in one of the poorest countries in the world? It is not going to happen because an army of therapists is flown in from Brooklyn, New York to treat everyone.

How do you actually reduce the amount of psychosocial problems that people have? How do you increase access to therapy and other support for people who do have serious psychosocial problems? By making the country a lot more affluent. If the country is a lot more affluent, perhaps your child doesn’t die at age three and you have fewer psychosocial problems to begin with. You are less depressed about the fact that your child just died. You will have the money to afford a therapist, or perhaps the government will have the money to fund a mental health service. You are asking the question from the wrong end.

Pritchett: Then you get obsessed with cost effectiveness, which means reducing both the numerator and the denominator. In this study, per capita income went from $1.80 a day to $1.85 a day. If you look at Vietnam, in 1990 it had a per capita income of about $1.80 a day—similar to where Niger is now—and it has since gone up to around $7.50 a day. That is development. None of that in Vietnam came from people becoming psychosocially more capable first, with everything else following from that.

Mounk: It’s not that USAID or the Ford Foundation sent an army of psychologists to Vietnam.

Pritchett: It was because the development actor and discourse helped Vietnam change its mind about its development strategy. Vietnam said: we can embody Vietnamese labor into global value chains in a way that is going to radically change the productive possibilities and the inclusion into productivity of Vietnamese people. They did it successfully, and extreme poverty disappeared in Vietnam right before our eyes through a process that involved almost no direct anti-poverty interventions.

So that is the first point: this isn’t just wrong, it is ontologically wrong. Ontologically wrong means you cannot fix it by doing slightly better experiments on this or that. It is looking at the wrong set of ways in which prosperity actually happens. Prosperity happens through the inclusion of individuals into productivity—a cooperative endeavor involving artifacts and institutions whereby large numbers of people can cooperate to create value: markets, governments, bureaucracies, and large organizations. Not whether I am a more go-get-em person. That was the first part.

Mounk: The second part then is: if RCTs are the wrong thing to look at, if that is not how we are going to make change, if we shouldn’t be obsessed with whether it is this intervention or that intervention at the level of the village or the town—but rather should be asking how we help countries choose successful development paths like Vietnam, paths that put in place the institutions needed to actually develop—how can development economists or other social scientists help with that?

Is the problem fundamentally that the people in Nigeria just don’t get it? Or is it, as you were saying earlier, that the people in charge have their own interests and want to preserve the privileges they have? They are worried that in the process of development they might be displaced from power, and so they would rather continue to be the kleptocratic elite in a relatively poor country than be displaced in a much more affluent one. Is development economics, or more broadly the field of social science, actually going to be able to help in that process? Or is that a hopeless endeavor?

Pritchett: It is not a hopeless endeavor, because after all, it has happened. The strange thing about saying it is a hopeless endeavor is that it has to ignore that Korea happened, and Taiwan happened, and Vietnam happened, and China happened, and Indonesia happened, and India happened.

Mounk: But did it happen because those places, for whatever reasons, had more favorable circumstances where they were more able to challenge their elites—or did we just get lucky that they put a few of the right reforms in place? Or did it happen in some way because social scientists helped guide them?

Pritchett: Yes, social scientists helped guide them, because social scientists created the data—like GDP per capita—to show that some countries have had really rapid growth by doing certain things and other countries have not. Deng Xiaoping didn’t just decide on moving towards a path of unleashing incentives for people to engage in private enterprise—perhaps too strong a word for what happened in China in 1978—in an intellectual vacuum. He looked around the world, brought in experts who had studied economics and the relative performance of economies, and was convinced that there was an alternative path from where China was to where China wanted to be.

The idea that ideas are irrelevant and that social science hasn’t had an impact is surreal. In the case of India, which I know relatively well—I have been going back and forth to India since 1992 and have spent time living in different cities there—there was a debate in India in 1991 about how to respond to an incipient macroeconomic crisis. That debate drew on decades of social science about the relative importance of market versus government-led development, and of import substitution versus export orientation. It drew on a body of empirical science that had been generated over time, and it led the country to do different things. You cannot act as if India decided what it did in the reform period in an intellectual vacuum. It is surreal to say that social science cannot in principle affect the way governments act, because it has demonstrably done so in specific instances. Maybe it has not always been as successful as one would hope, and let’s admit that oftentimes social science is pushing against direct material and political interests.

But let me answer your question. The answer is what I call full Trinity growth diagnostics. If development economics were focusing its time, effort, and capabilities on the development of full Trinity growth diagnostics, I think we would be in a radically better place to help countries with their organic, country-level development strategies.

What do I mean by full Trinity? For something to be successful, it has to be technically correct—it has to have a correct causal model of how, if actors undertake this action, working perhaps through a complex adaptive system, a particular outcome will result. It is not direct cause and effect, because markets are complicated and complex adaptive systems. But the technical question is: if I do this, will this be the result?

The second element of the Trinity is organizational. If I am asking a government to undertake actions to promote a given outcome, does the government have organizations capable of doing that? If I say I am going to do industrial policy to promote high-tech industry, do I have government organizations capable of identifying and promoting high-tech industry?

The third element is that it has to be politically supportable. I have to be able to assemble and sustain a political coalition to support the implementation of these actions.

The first step of this is a growth diagnostic—a process that I and colleagues at the Kennedy School developed over time—of asking, in a given place, what are the binding constraints to having more rapid growth than we are currently experiencing. The binding constraints matter because when we ask what is good for growth, we tend to produce a very long list. If you say “get to Denmark,” you can say Denmark has this and Denmark has that. But you cannot do everything that Denmark now does. You have to do the right thing. Identifying the right thing involves an analytical process of prioritizing among the many things you could do—asking which would have the biggest impact on the prospects for a rapid, sustained, and inclusive episode of growth. We have mechanisms for trying to adjudicate among contesting claims about what the binding constraint is.

But then we have to supplement that by asking: of the things that should be priorities and would have a big impact on economic growth, which can we actually do? Saying that if you had Korea’s current capabilities you could do a certain thing is not very helpful if you have Niger’s capabilities. We have to ask which of the binding constraints can realistically be addressed. Then you have to have a realistic, positive model of the politics—which of these can the existing elite be persuaded to engage in? None of those questions poses an intellectual challenge anything like reconciling general relativity with quantum mechanics.

Mounk: But it’s an art rather than a science, I imagine. What might that look like concretely? Let’s say there is a new government in some country—genuinely reforming in certain ways, perhaps a little corrupt in others. They have just come to power, perhaps through election, perhaps in some other way. They can make some changes, but they cannot make a million changes at the same time. It is hard to drive change across sectors simultaneously, and there is a limited amount of money to invest because it is a very poor country. Should they invest in education or in the judiciary? Should the first big push be to reform the judiciary—so that who wins a dispute is driven more by the facts and creates stable expectations, rather than by bribes or family relations? Or should they start with a big push on schools, making sure teachers are competent and actually show up? I imagine that is the kind of choice a government might realistically face. How should you go about making that decision?

Pritchett: There are analytical tools for addressing this, and this is where we get into the mechanics of what a growth diagnostic would do. Partly what a growth diagnostic does is ask: if it were the case that a proposed action were the binding constraint to growth, what should we observe about the economy?

There are four or five things we look for. First, if we think X is a binding constraint—say, corruption—then when we see relaxations or improvements in corruption, we should see more growth. If I have my arm in a sling and the movement of my arm is constrained by that sling, then if I remove the sling, I should be able to move it a lot more. If I can’t, maybe it was the shoulder injury all along and there is a different cause. Changes in the constraint should cause changes in outcomes.

The second is what we call bypassing. If finance is a constraint and I could otherwise be productive and profitable, we should see firms actively engaged in creative ways of raising finance. We should see enterprises actively adapting around the constraint.

The third involves what we have given the quirky name of camels and hippos. Different industries in an economy are more or less intensive in a particular proposed constraint. If we say water is a constraint, then we should observe camels—animals well adapted to a lack of water—and we should not observe hippos. If we look at the economy and the industries that are thriving are ones that economize on the proposed binding constraint, that is evidence in favor of its being the binding constraint.

These are plausible, sophisticated ways of analyzing the current situation of a country. They can produce answers like: putting more kids through school really isn’t a binding constraint here, because we don’t see changes in that producing changes in growth, we don’t see firms desperately engaged in training because they are short on skills, and we don’t see labor-intensive, non-skill-intensive industries thriving because they are economizing on the lack of skill. That means we need other explanations.

This technique has been deployed and it produces interesting results—not clean RCT-looking results, nor should one expect that. To me, the key test of a growth diagnostic is whether it comes to different answers for different countries. If you come to the same answer for every country, you have an ideology. If every country you go to produces the answer that reforming the trade system would accelerate growth, you are an ideologist. Maybe you are right sometimes, but you are not always going to be right, because there are going to be societies in which that is not the most important binding constraint. When we do these growth diagnostics in different countries, we come to demonstrably different answers. The data and the evidence line up around different things.

This is a promising technique, but it has been radically under-invested in—because the resources in the world have been devoted to deworming studies rather than to creating a global community of practice that does technical growth diagnostics. More support to a global community of practice focused on honing the art and science of growth diagnostics would be a tremendous way the world could support the organic process of countries finding their own development paths. It would allow you to engage with countries and actors within countries in a radically different way than coming to them and saying: you lack these 50 things, all of which would be good if you did them—when you know they are going to be able to do two, maybe three.

In the rest of this conversation, Yascha and Lant discuss migration, how to build a fair asylum system, and what the destruction of USAID means for the future of development. This part of the conversation is reserved for paying subscribers…

Read more

When the first few months of the full-scale war in Ukraine passed and it became clear the conflict would last years rather than months, a popular subject in political commentary in Ukraine was the need to adopt the Israeli model of steel porcupine—a small state that defends itself by building vast defensive and offensive military capability, making any attack too expensive in money and human lives—maybe with a nuclear bomb or two down the line.

At the time, it seemed impossible—it takes too long and costs too much. But Kyiv has created its own steel porcupine.

A Strategic Decision

Four years down the line, instead of being a security recipient, Ukraine acts as a security donor, acquiring allies in unexpected corners of the world. Instead of closing down and becoming entirely self-reliant, Ukraine chose to open up and make other countries reliant on what it can offer—drone know-how in the Middle East and the U.S. bases there, a defensive wall between Europe and the threat from Russia, and any future benefits Europe will enjoy once the Ukrainian ballistic programme comes of age.

Instead of the traditional steel porcupine, Ukraine has developed an inverted form—it shoots quills not at its enemies but at its allies, injecting them with a protective layer of technology. Not many have noticed, however, that this ensures Ukraine receives not just protection from the allies to whom it now becomes more valuable, but also offers Ukraine a level of control unlike that of many other countries. Due to the nature of the arms business, Ukraine will have a say on who will or will not be allowed to use its technology, even when produced in joint ventures. Ukraine is advancing its defence deals very strategically, at exactly the right time and place for maximum effect. Considering how strategic its actions on the foreign affairs front have been, this seems to be a deliberate plan a long time in the making. If you already have a lot of responsibility, you might as well go and get yourself great power.

How Did We Get Here?

Without the war in Ukraine, drone warfare would still be in its infancy, military procurement would still consider tanks and infantry fighting vehicles a better investment than a drone wall, warships would still sail the Black Sea without a care in the world, and there would still be no defence against the full complement of U.S. and European missiles (Russians have since learned to counter HIMARS and developed responses to ATACMS as well). Estonia’s decision this month to suspend a €500 million infantry vehicle order and redirect the funds entirely to drones and air defence, citing lessons from Ukraine, shows how fast procurement thinking has shifted.

The powers that be decided not to right the wrong when Russia invaded Ukraine, and now large sections of their military technology are slowly turning obsolete. These same powers now have to adjust to the new world of warfare, and the only country that knows how it works is Ukraine.

Why Ukraine is Unique

Ukraine is uniquely positioned to throw a wide defensive tech net over its allies, and not just because it is the only active battlefield capable of testing and fine-tuning any weapon under the sun. Take Russia—it has similar access to battlefield conditions, and yet not many countries are queuing up to buy Russian-made modern drones. In fact, their most successful models are Iranian Shahed drones. Russia needed Iran to set up the production lines and train the operators—and this is the extent of their international drone programme.

Ukraine, in addition to being highly innovative and resourceful, also designed a procurement system almost entirely devoid of bureaucracy—arms developers work directly with military units, which means there is no lag between development and the front line. War is about the survival of the fittest—only the best military tech companies remain on the market. The decentralized drone development model bypasses the layer where traditional corruption lives—when a military unit orders directly from a developer, the middleman is removed entirely.

We also often forget that the Ukrainian military sector did not just appear out of nowhere—it was well established and indeed formed the basis for the Soviet arms industry.

Ukraine’s Yuzhnoye Design Bureau in Dnipro designed and built the SS-18 Satan—the most fearsome nuclear ICBM in the Soviet arsenal, capable of carrying ten independently targeted warheads across 10,000 kilometres while releasing decoys to defeat radar. After 1991, Ukrainian engineers continued maintaining the remaining Satans in Russian silos until 2014, when Kyiv stopped all cooperation. The stockpile is ageing without the maintenance that kept them viable, and nobody knows if they would fly or explode if launched. Russia has been attempting to replace the Satan with its own design, the RS-28 Sarmat, for the past 20 years—and despite having a working Ukrainian-designed model at their disposal, has achieved exactly one successful test launch out of six attempts, one of which destroyed the test silo entirely. Ukraine built it. Russia cannot even replicate it.

How the Arms Trade Actually Works

International arms deals are controlled first by national interests and only after that by financial considerations.

Since defence is a matter of national security, defence companies cannot just sign agreements with other companies or states—they need permission from their government. The same applies to the purchasing side: the country that owns the underlying technology, not the company producing military goods, decides which other country can purchase the specific equipment, even if it is manufactured elsewhere and another party foots the bill.

The U.S. International Traffic in Arms Regulations (ITAR) ensures only specific defence articles reach very specific markets. Those markets require authorization from the State Department, not merely a commercial agreement, to sell or even gift them to a third party. This is a foreign policy instrument as much as a trade one.

Almost the entire European defence industry is criss-crossed with U.S. technology and requires re-export licences as a result. In practice, the United States decides who can and who cannot buy weapons that contain its technology, regardless of where those weapons are manufactured or who paid for them. There are countries with their fingers in several defence pies, but none more so than the United States.

When former Warsaw Pact members began transferring their Soviet-era weapons to Ukraine at the start of the full-scale invasion, Russia objected on the grounds that the original re-export licences had belonged to the Soviet Union. The problem was resolved on the basis that Ukraine, as a Soviet successor state, held the same standing as Russia over equipment that had once been part of the shared arsenal.

Subscribe now

The system in place is essential to Ukraine’s ability to sell weapons. Every country that signs a weapons manufacturing or sales contract with Ukraine must guarantee that the technology will not end up in Russian hands, or the hands of any Russian-affiliated state, under any circumstances. That means there is a long list of countries banned from acquiring Ukrainian weapons.

Ukraine is, in effect, building its own version of ITAR. The same architecture that keeps its technology out of Russian hands also gives Kyiv a say in who gets to defend themselves with it. Every cooperation agreement embeds Ukrainian technology into another country’s defence architecture, and every embedded system requires a Ukrainian licence to transfer further. The United States spent decades getting to that position. Ukraine is acquiring significant leverage over international arms markets in a few short years. Kyiv is too smart not to realize what they are building.

Re-export Licence as Diplomacy

Ukrainian weapons systems are not just a financial mechanism and an alliance-building exercise. They offer Kyiv foreign policy instruments very few countries have. By being an undisputed leader in the battlefield drone ecosystem, Ukraine as a monopolist of sorts can send geopolitical signals through its defence cooperation agreements.

The ten-year defence cooperation agreements signed in the Middle East send a pointed message to Washington from both sides: Ukraine indicating it has serious leverage, and Middle Eastern countries demonstrating that the United States is not the only player in the area. Ukraine also now has leverage in other areas—it can make sure its new partners do not accept Russian ships laden with stolen Ukrainian grain, unlike Israel, which has accepted several shipments.

Zelenskyy’s meeting with President Ilham Aliyev in Gabala on April 25, his first visit to Azerbaijan since the full-scale invasion began, was held just 100 kilometres from Russia’s border. The six defence cooperation agreements signed there send an unmistakable message to Putin: Russia’s influence in the South Caucasus is finally over, while Kyiv’s has just begun.

A good example of what successful arms deals look like for Ukraine is its cooperation with Turkey. Ukraine supplies the engines that power Turkey’s most advanced combat drones—Baykar’s Akıncı and Kızılelma were developed with Ukrainian-made Ivchenko-Progress engines, earning the Kızılelma the nickname “a Turkish bird with a Ukrainian heart.” Turkey, in turn, supplies Ukraine with Bayraktar drones and Ada-class corvettes currently under construction for the Ukrainian navy, and is building a Baykar manufacturing plant outside Kyiv. It is therefore not surprising that Turkey used its considerable influence over the UN-recognized Libyan government to allow Ukraine to establish a military presence on the Libyan coast, from which Ukrainian naval drones have since hunted Russian shadow fleet vessels in the Mediterranean.

A different kind of example is Switzerland, which in 2022 refused re-export permissions for Swiss-made Gepard ammunition Germany was trying to send to Ukraine, blocked Denmark from transferring Swiss-made Piranha III armored vehicles, and blocked Spain from re-exporting two Swiss-made anti-aircraft guns. The Netherlands responded by stopping all Swiss arms purchases. The policy cost Switzerland its reputation as a reliable defence partner across Europe at precisely the moment Europe began the largest rearmament programme in its history.

Not Bad For a Country With No Cards

While an obvious take-away from the arms deals Kyiv is currently signing is that it has emerged as a global security provider, the real outcome is the permission architecture Ukraine is embedding into the global arms industry as we speak, and the power that architecture affords on the global scene. Ukraine is increasingly holding the strings to a global defence network that will operate without Washington’s permission.

Subscribe now

This also acts as the other side of security guarantees for Ukraine. The lamentable demise of the rules-based order and the loss of the United States as the global sheriff means no written guarantees will work. There is almost no scenario in which Ukraine would trust them. Ukraine can rely on its own military, and now it will also be able to rely on its defence industry being too valuable for allies to walk away from. By becoming indispensable to the global security landscape, Ukraine can spread its influence far wider than anything Russia can achieve.

While Israel turned itself inwards and made itself hard to kill, Ukraine turned outward and made itself too important to abandon. In the process, it is acquiring a lot of control over global security. It did not ask for this task, and it still has hard times ahead. But it found a way to protect itself by protecting others.

Ines Burrell is a geopolitical analyst and political risk consultant based in the UK. Born in the Baltics, with a degree in International Relations from the University of Exeter, she writes and gives live commentary on European security and Russia.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

• The inaugural meeting of the Intellectual Bootcamp, which will discuss Jonathan Haidt’s “Why the Past 10 Years of American Life Have Been Uniquely Stupid,” will take place TODAY Friday, May 1 at 12pm EST at this link.

Read more

On Wednesday afternoon, a man with a knife ran down a street in north London trying to stab Jews. Two were wounded. I want to tell you why this did not surprise me. But first, a story.

In my early twenties I spent a summer traveling: two months in Ghana on an internship with a small NGO, several weeks in Jerusalem at a religious seminary at the foot of the Western Wall, then home to London for the last year of my studies.

Before I left for Ghana, my family were appalled. Wasn’t it dangerous? What about the state of the hospitals? I went, and even though I eventually did spend a few days hospitalized, I never once felt unsafe.

When I told my Ghanaian colleagues I was going on to Israel, their faces paled. Suicide bombs. Rockets. Stay with us, they said, where it’s safe. But rarely have I felt more at home than when wandering through the cobbled streets of the Old City.

And when I told my teachers that I was flying back to London, they shook their heads. What about the antisemitism? The radicalization? But I came home to my mum’s cooking, to unarmed police, to quiet streets and football on the TV.

I tell the story as a subtle rebuke: nowhere is as bad as the news might suggest. It’s like the opening image of Carol Rumens’s “The Émigrée”:

The worst news I receive of it cannot break / my original view, the bright, filled paperweight.

Solid. Durable. Belonging to a person who knows the truth about their home and is holding on to it.

This week’s attack in Golders Green has shaken a community that was already shaking. The paperweight is cracking.

Golders Green is the closest thing London has to a Jewish neighborhood: kosher bakeries, multiple synagogues within ten minutes’ walk, the kind of place where you hear Talmud discussed on the bus. Two men, one in his seventies, one in his thirties, were stabbed by a man who ran down the road targeting Jews. Both survived. They were the latest casualties on a list that includes the two killed in Manchester last Yom Kippur, when an Islamic State-inspired attacker drove into worshippers outside a synagogue and stabbed his way towards the entrance.

Subscribe now

Between those two attacks, the months were not quiet. Arson at a synagogue in Harrow. An attempt in Finchley. An attack on a building formerly used by a Jewish charity in Hendon. Hatzola ambulances, the Jewish volunteer medical service, set on fire. In fact, the Community Security Trust recorded 3,700 antisemitic incidents in Britain in 2025, the second-highest figure on record.

Counter-terror officers are investigating links to Iranian proxies. A group calling itself HAYI has claimed responsibility for several of the arson attacks, and three Iranian nationals were charged last year under the National Security Act for surveillance of UK targets.

People are afraid. They are asking the questions quietly, and then aloud, in living rooms and synagogues: Are we doing now what German Jews did years ago? Are we ignoring the same warning signs?

Britain in 2026 is not Weimar Germany. The government has not been captured. Jews are not being stripped of citizenship by statute.

But we might be France. In 2012, a gunman murdered three children and a teacher at a Jewish school in Toulouse. In 2015, four shoppers were killed at a kosher supermarket in Paris. In 2017, Sarah Halimi, a Jewish retiree, was beaten and thrown from her apartment window by a neighbor shouting “Allahu akbar.” In 2018, an eighty-five-year-old Holocaust survivor was stabbed eleven times and set on fire in her Paris flat.

The French Jewish community has lost tens of thousands to emigration since the early 2000s, well over ten percent of its number—not under fascism but under a functioning republic, with courts and civil rights law and presidents who gave speeches denouncing the attacks. What is happening in Britain today looks less like Germany in 1933 and more like France 15 years ago: pressure, applied from multiple directions, inside a democracy that keeps expressing its concern.

The attack was shocking. But it was not a surprise.

What we have lived with for the past few years is the slow normalization of a particular kind of language. Zionism is not a political ideology with internal debates, it is a cancer. Its adherents are not citizens, they are a parasitic body to be excised.

But once a person is a parasite, the moral grammar of what one does to him changes. You do not negotiate with cancer. You eliminate it. In central London, marchers have carried placards describing Zionism as Nazism, in public, in daylight.

The “eternal Jew” of the propaganda, redrafted as the “eternal Zionist.”

Rabbi Lord Jonathan Sacks, the former Chief Rabbi of the Commonwealth, described antisemitism’s survival strategy as mutation: “In the Middle Ages, Jews were hated because of their religion. In the 19th and 20th centuries they were hated because of their race. Today they are hated because of their nation state.” The new antisemite, he observed, declares: “I’m not a racist. I have no problem with Jews or Judaism. I only have a problem with the State of Israel.”

But a Jew does not need to be a state actor to be a target. Mere affiliation is enough: a flag in a window, a cousin in Tel Aviv, a kippah worn without a keffiyeh. It marks you as unaligned, which is to say suspect.

Meanwhile, the chorus is the one we have heard for decades. Politicians find the attack “deeply concerning.” Cross-party statements arrive within the hour, sincere but impotent. A site visit, a photo opportunity, a BBC interview. We are asked, again, to mobilize around the fact of being hated.

But that is the trap any besieged minority faces.

Most of us are scared. Some are angry. Many are tired. Tired of being asked to rally around a Jewish identity that, for many, has thinned to a surname and a grandparent. We have spent a lifetime being told that being Jewish is incidental, a private flavoring in an otherwise universal British self. Now we are asked to be Jews in public at the moment many of us have the least Jewish vocabulary to draw on.

The temptation, in that vacuum, is to let the fight become the identity. To be a Jew because you stand against antisemitism. To be proud because you are not afraid. But what would any of us be if there were no antisemitism to fight? Would we still stand and declare ourselves proud? And if so, proud of what?

It is a negative theology. Judaism defined by what is done to it, by what it endures, by what it is not. It hollows from the inside.

In his book A Letter in the Scroll, Sacks warned that a community whose shared content is the experience of attack will, when the attack eases, find it has nothing left to share. This was not an argument against the fight. It was an argument about what must exist alongside it: a Judaism defined by more than external opposition, by what it affirms rather than simply what it endures.

Subscribe now

That, I think, is the real precipice in front of any minority under siege. Not the one usually offered, between flight and resistance. The deeper choice is whether we let ourselves be defined by our opposition to antisemitism.

In what may be the most quoted passage of that book, Sacks wrote: “I am a Jew because, knowing the story of my people, I hear their call to write the next chapter. I did not come from nowhere; I have a past, and if any past commands anyone this past commands me.”

Any community whose only language is its own defense will, in the end, have nothing left worth defending. Instead, the call to action is not a fight for our “right” to be Jews. It is to engage in our responsibility to “be them.” Light the candles on Friday night. Make our sabbaths feel different. Learn the Hebrew you always awkwardly stumbled through. Open the books our grandparents kept and rarely read aloud. Argue about the weekly Torah reading. Sing Jewish songs badly. Make blessings on the good that we have with our children.

To live in our own language with fluency and conviction, with pride and with joy, is more than half the battle.

A community that can do that has nothing to prove and everything to keep.

Matt Marks runs Tribe, the United Synagogue’s youth arm, advises the BNJC on Jewish life, and is a south-coast rabbinic chaplain with University Jewish Chaplaincy. He is a Sacks Scholar writing a PhD on Rabbi Lord Jonathan Sacks’ intellectual legacy.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

https://us06web.zoom.us/j/82853159147?pwd=Xi8TS3nHyZMSxAGPBrF1LViI2onrK3.1

Passcode: 129264

If you have trouble joining, please email events@persuasion.community

We will have introductions and then discuss Jonathan Haidt’s “Why the Past Ten Ye…

Read more

During a rally in 1966, a recently drafted man, Robert Watts, reportedly told friends, “If they ever make me carry a rifle the first man I want to get in my sights is L.B.J.”

For this, a federal court convicted Watts of “knowingly and willfully threatening the President.” The Supreme Court would later overturn that ruling in Watts v. United States, noting that Watts’ comment simply constituted “political hyperbole” and was therefore protected by the First Amendment.

Now imagine if Watts had written his message out in seashells.

This is the absurd situation we find ourselves in this week, following the Trump administration’s second indictment of former FBI Director James Comey, this time for a social media post depicting shells on a beach spelling out “86 47.”

The administration alleges that, in posting the photograph, Comey “knowingly and willfully” made a threat to kill President Trump under the same statute used against Watts in 1966, along with another statute that prohibits transmitting threats through interstate communication. They also allege in their indictment that “a reasonable recipient” would interpret the post “as a serious expression of an intent to do harm” to the president.

But there’s nothing reasonable or serious about this prosecution.

For one thing, the term “86” has been around since the 1930s, commonly used in restaurants and other contexts to mean “get rid of,” “throw out,” or “refuse service to.” When combined with the number 47, referring to our current 47th president, the message becomes clear: Get rid of Trump.

Subscribe now

To assume that “86” means “kill” or “assassinate” is, at best, uncharitable. There are obvious ways to “get rid of” a president without ending his life, like impeachment and removal from office. When the Chicago Sun-Times reported that NBA coach Jim Boylen was “eighty-sixed by the Bulls,” nobody thought it meant the Bulls’ front office had murdered him.

Even if they can somehow establish that “86” unambiguously means what they say it means, the prosecution still has their work cut out for them. Unless it can be proven that Comey himself seriously expressed an intent to kill the president, the phrase “86 47” would still be protected speech. The law is clear that merely wishing for someone’s death is and should be protected speech, as distasteful as it may be, absent more evidence proving intent to cause harm.

Comey isn’t even the first to use “86” in reference to presidents. In 2022, right-wing activist Jack Posobiec posted the same message (minus the seashells) referring to then-President Joe Biden. It’s a ubiquitous enough sentiment that Amazon has both “86 47” and “86 46” merchandise for sale. It’s absurd to argue that anyone who purchases or displays those products necessarily intends to threaten the president in question, as opposed to simply voicing opposition to him.

Even the timing of this indictment is comical. Comey’s “86 47” photograph was posted in May of 2025. It strains credulity to think Trump and his administration considered four digits written in seashells a threat against the president’s life, but failed to act on it for 11 months.

Simply put, this indictment lacks merit. Merely labeling disfavored speech a “threat” doesn’t magically make it so. To lose First Amendment protection, that expression must clear a very high bar—and for very good reason. 2003’s Virginia v. Black established that “true threats” are “those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals.”

Twenty years later, Counterman v. Colorado made clear that the government must prove the speaker either intended the statement as a threat or “consciously disregarded a substantial risk that his communications would be viewed as threatening violence.” That’s the constitutional floor. In Comey’s case, however, the government must meet the higher standard—proving he subjectively intended to make a threat—because that is what the statutes he’s being charged under require.

Some have also suggested that Comey’s post incited violence. But the indictment doesn’t charge him with that, likely because even the Department of Justice recognizes how untenable such a claim would be. Incitement is another narrow First Amendment exception, limited to speech intended and likely to provoke imminent unlawful action. Not only is there no evidence Comey was urging anyone to assassinate Trump, but it’s impossible to argue with a straight face that a viewer of his seashell photo would have immediately set out to locate the president and attempt to kill him as a result.

The law is structured the way it is to address specific, concrete harms while giving the most possible breathing room for free expression—particularly when it comes to criticizing the government. As the Supreme Court noted in Watts’ case, “the language of the political arena … is often vituperative, abusive, and inexact.” Any attempts to punish such expression, the ruling continued, must be interpreted “against the background of a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open, and that it may well include vehement, caustic, and sometimes unpleasantly sharp attacks on government and public officials.”

When considered in that context, a photograph of seashells on the beach becomes about as threatening to the president’s life as a salad.

Subscribe now

Lowering the bar for what counts as true threats and incitement would hand the government a dangerous tool to crack down on dissent under the pretext of preventing violence. It’s not hard to characterize a wide range of political rhetoric as menacing or a potential contributor to future violence: calling Trump administration officials fascist, labeling abortion murder, declaring that “All Cops Are Bastards,” or claiming an election is rigged, for instance. Without the First Amendment’s narrow and exacting standards, it would be all too easy for administrations of either party to criminalize a vast swath of political expression.

The reality is that this indictment is a flimsy excuse to use government authority to punish the president’s political enemies—something this administration is quite fond of and has been thoroughly documented doing. This is all the more reason to vehemently oppose this indictment and the retaliatory actions of the administration.

If political retribution for speaking out becomes the norm in American politics, then everyone’s right to free speech could be eighty-sixed.

Angel Eduardo is senior writer and editor at The Foundation for Individual Rights and Expression.

Aaron Terr is Director of Public Advocacy at the Foundation for Individual Rights and Expression.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Predicting Donald Trump’s political demise has typically been a fool’s errand. Some of my smartest friends have declared his definitive fall from grace again and again, only to be proven wrong each and every time.

If you watch MSNBC or listen to NPR, you may over the past decade have believed that Trump’s presidential campaign is a hopeless publicity stunt; that the Republican Party is about to turn on him because of the Access Hollywood tape; that he has no chance of winning against Hillary Clinton; that his presidency will be so chaotic that he’ll be forced to resign within his first year in office; that Robert Mueller’s investigation into his relationship with the Kremlin will result in his impeachment; that his mishandling of the COVID pandemic will make him toxic to voters; that his loss against Joe Biden has ended his career for good; that he is about to be impeached over the January 6 riot at the Capitol; that he is sure to lose the race for the Republican nomination against Ron DeSantis; that he is sure to lose his bid for reelection against Joe Biden; that he is sure to lose it against Kamala Harris; and so on.

Subscribe now

Ten years into Trump’s political career, the most avoidable mistake pundits can make is to underestimate his powers of survival and resurrection.

And yet, I have come to the tentative conclusion that this time may, finally, be different. For the past decade, Trump has dominated American politics like no other president in living memory; now, signs of that era coming to a close are suddenly multiplying. It is, as Saturday’s appalling assassination attempt on the president reminds us, impossible to see around the next historical corner. But it sure seems as though Trump’s hold over the country is finally slipping. This, to misquote Winston Churchill, no longer feels like the end of the beginning; it may be the beginning of the end.

A minority of Americans has always been drawn to Trump because of his most extreme actions and statements. They loved his coarseness, reveled in his taunts, and unhesitatingly embraced his radicalism. This group made up a significant share of his most devoted base—but it was never big enough to explain how he could have won two presidential elections.

Many of the voters who twice put Trump over the top have, all along, had a more conflicted view of him. Trump swore that he would make Americans far richer. He would cut taxes and curb inflation. The costs of health insurance would fall. There would be peace in the Middle East. The country would return to its former grandeur. It is not hard to see why those who were inclined to believe that he might actually turn these promises into reality, at least to some extent, found them to be very enticing.

Subscribe now

During his first term, Trump did celebrate some genuine successes, from Operation Warp Speed to the Abraham Accords. But when he predictably failed to bring about most of his outsized promises, he proved shrewd at making up excuses. He had only just taken power. The deep state was standing in his way. The “Russia hoax” had made it impossible for him to govern. The global pandemic had messed everything up. The share of Americans who were genuinely excited about Trump shrank rapidly towards the end of his first term; and yet, the thought that it might be worth giving him a second chance in 2024—even if he just delivered on some tiny fraction of his promises—lingered in the minds of a surprising number of voters.

But the fulfillment of promises can’t be deferred forever without voters starting to lose patience. As Viktor Orbán learned to his chagrin in Hungary, there comes a time when leaders are measured by their results rather than their rhetoric. And that time has now come for Donald Trump.

The immediate reasons for Trump’s travails lie in his ill-fated war with Iran. The contention that foolish “foreign entanglements” had repeatedly led America astray was central to his political persona from day one. In his second inaugural address, he announced that “we will measure our success not only by the battles we win, but also by the wars that we end, and perhaps most importantly, the wars we never get into.” This makes it especially damaging that he pursued a war of choice in Iran without bothering to make a coherent case for it to the American public or ensuring that there would be a real exit strategy. The one major promise that Trump actually honored in his first term was that he would start no new major wars; that too now looks like empty self-promotion.

The knock-on effects for Trump’s other areas of traditional strength have been brutal. Americans voted out Joe Biden’s Democrats in good part because of the persistently high level of inflation after the pandemic, which had been fueled by the administration’s generous stimulus programs. Now, Trump’s failure to anticipate that Iran would choose to block oil tanker traffic through the Strait of Hormuz has led to a renewed spike in inflation, putting the president’s approval ratings on inflation and the cost of living underwater by a remarkable 40 points.

Trump is also in trouble in some historic areas of strength that are less directly connected to the war in the Middle East. Most Americans grew furious with Biden’s inability to control immigration at the southern border. But in his second term, Trump has embraced a deportation policy that is so pointlessly cruel that, in many polls, a clear majority of Americans now disapproves of his handling of the issue.

The result is becoming increasingly clear in the data: Overall support for Trump is at or near record lows.

Trump has often been far more popular with the American public than his detractors cared to acknowledge; today, his approval ratings are genuinely dismal. Nearly 58 percent of Americans disapprove of Trump’s job performance (most of them strongly) while only 39 percent approve (most of them weakly), according to statistician Nate Silver’s polling tracker. His net approval is as low today as it was in the immediate aftermath of the January 6 assault.

In the past, Trump has been hated by liberals, seen as divisive among independents, and (the complaints of a small band of principled columnists notwithstanding) enjoyed popular support among conservatives. Declines in Trump’s poll numbers were usually precipitated by independents abandoning him. Today, Trump remains toxic among liberals, has come to be viewed negatively by most independents, and is newly divisive among conservatives.

Subscribe now

Trump’s ironclad grip over the Republican base is starting to loosen. In the past, conservative critics of Trump have usually complained that he has sold out the views and values associated with figures such as Ronald Reagan. Now, criticism of Trump within the conservative camp is for the first time being framed as a betrayal of the supposed values on which the MAGA movement was founded. Some of the biggest influencers on the American right, such as Candace Owens and Tucker Carlson, have recently expressed regret for supporting Trump. For the first time since 2016, his hold over the MAGA movement may actually be weakening.

The political costs from these developments are likely to compound over the course of the coming months. Betting markets give Democrats about a 6-to-1 edge to win the House of Representatives in the midterm elections this November; despite a daunting electoral map, they also have slightly more than even odds to take control of the Senate. If Trump’s party really does suffer a serious shellacking in the midterms, his inability to push major legislation through Congress and the impending end of his term will further weaken his control over his own party. With attention turning toward the 2028 primaries, the White House may suddenly see its power slipping away, as happened after the 2006 midterm elections, in which Democrats took control of both houses of Congress, rendering George W. Bush largely powerless during the last years of his term. Sooner than we can now imagine, Trump may come to be seen as a lame duck.

When Trump was reelected with a bigger, younger, and more diverse electorate in 2024, it seemed as though he might actually manage to impose his vision and his values on the country. In the first months of his second term, the administration was moving with impressive speed. Resistance to its ascendancy was conspicuous by its absence. It felt as though America might stand at a genuine tipping point.

The window of opportunity for Trump to reshape the country in a significant way was, I think, real. But he responded to the cultural excesses of the Democratic Party—and the broader progressive establishment with which it is increasingly associated in the public mind—with even more extreme cultural excesses of his own, provoking a broad counter-reaction which extended well beyond those who partook in resistance marches during his first term. Therefore, it now seems increasingly safe to say that he has squandered it. Trump’s second term will leave behind an America that is weakened, cheapened, and fractious; but it seems increasingly unlikely that he will leave behind an America shaped in his own image.

This is cause for optimism, an indication that America has proven to be more resistant to the appeal of authoritarian populism than many feared. It would take someone who is much more popular and disciplined than Trump to change the country in a fundamental way.

And yet it is far too early to celebrate. Trump will, after all, remain in office for another 32 months. That is enough time to do a lot of damage to democratic institutions, to engage in a great deal of corruption, and perhaps to start more reckless wars. In all likelihood, a President Trump who is starting to sense that the tide is turning against him will turn out to be more, not less, dangerous to the American republic—and the world.

Some danger will persist even after he leaves office. When demagogues leave office—even when they are booted from office in disgrace—it rarely spells the end of their movement. Brazil’s Jair Bolsonaro lost his bid for reelection and was imprisoned for trying to impede the peaceful transfer of power, and yet his son Flávio has close to even odds of becoming the next president of Brazil, according to prediction markets. Alberto Fujimori was hounded out of Peruvian politics due to massive corruption and human rights abuses nearly three decades ago, and yet his daughter may be about to lead the country.

In Brazil, Peru, and many other democracies around the world, voters may decide to give populist movements a second (or third or fourth) chance because they were so disillusioned with the hapless alternatives. Given that the popularity of the Democratic Party remains at record lows, it would be deeply naive to rule out a similar future for the United States.

Trump looks likely to start fading from American politics over the coming years. But the broader threat of Trumpism may well outlast its creator.

This piece was originally published by the Dispatch.

https://us06web.zoom.us/j/82853159147?pwd=Xi8TS3nHyZMSxAGPBrF1LViI2onrK3.1

Passcode: 129264

If you have trouble joining, please email events@persuasion.community

For the first session, we will have introductions and then discuss Jonathan Ha…

Read more

David Bromwich has taught literature at Yale University since 1988. His books include Hazlitt: the Mind of a Critic, The Intellectual Life of Edmund Burke, How Words Make Things Happen, and Politics by Other Means: Higher Education and Group Thinking.

In this week’s conversation, Yascha Mounk and David Bromwich discuss why Americans have lost faith in universities, how grade inflation and political conformity undermine academic credibility, and whether the opacity of elite admissions processes can be reformed.

Note: David Bromwich asked us to be clear that the views he expresses are his own and not those of any institution or group within an institution.

This transcript has been condensed and lightly edited for clarity.

Yascha Mounk: I’ve been hoping to have you on the podcast for a long time, and we have a good occasion now because you were on the faculty committee at Yale University that was tasked with trying to figure out why people have lost faith in higher education in significant numbers in the United States, and what universities—and particularly Yale—can do to regain the trust of the public.

What do the findings of the committee suggest lies at the heart of this loss of faith in universities in the United States?

David Bromwich: Several things. You use the word “faith.” The official name of the committee was Trust in Higher Education. But trust, as anyone who has studied moral philosophy or just speaks English knows, is mutual. It depends on a shared understanding of what the purpose is of this or that institution, this or that custom or ritual.

The understanding that was lost is how public higher education—and specifically liberal arts education—prepares you for life in a way that will serve students well in getting jobs, but also make them thoughtful citizens. There is some effect from the education that they wouldn’t get just from reading a lot of books or even watching a lot of television.

Some of the causes of waning trust that we looked at were the process of admissions. Yale is an elite school, so this is particularly dealing with the loss of trust in schools that have that sort of prestige. The price of the school is very high, although most people don’t pay the sticker price. That relates to a lack of transparency in things like how you pay, what tuition costs are, how it may be deferred, what loans are available, and how much tuition can be covered for people who aren’t rich.

Recently—and this overlaps with the work of the committee—an announcement came out about a month or two before our report was put into its final draft. The president of Yale, Maurie McInnis, announced that families making $200,000 a year or less, per household, would not have to pay any tuition. If you know how low on the scale of disposable income $200,000 for a whole household actually is, that’s not surprising. It’s not extraordinarily generous, but it is more generous than most people imagine the Ivy League schools are.

Mounk: For probably most of the households that send kids to Yale, that threshold is above their income. The median household income in the United States is something like $70,000 a year, so for the median American family that is able to get their child into Yale, they are going to have a free ride.

Bromwich: There is what we should call opacity rather than transparency. I like the 18th century word for it: publicity—making public, in a neutral sense, the criteria that are used by a university like Yale to accept or not accept students. This is all the more important because we’ve acquired an excess of bogus prestige from rejecting a lot of students. There are more and more applicants, more and more people think they might make it, but admissions are now lower than 5% at Yale and at places like Yale.

So how to account for the distrust? There seems to be a kind of false advertising in which the institution has been unconsciously indulging.

Mounk: The report is also quite explicit about the problem of grade inflation. In Yale, the median grade is now an A or an A minus, I believe. How is that contributing to the loss of trust in universities?

Bromwich: A student reporter writing for the campus newspaper, the Yale Daily News, found in an article published about a year and a half ago that the average grade given at Yale was in the A family—meaning As or A minuses—something like 70% of grades. As a teacher in the humanities (and we are usually charged with being the great culprits on grade inflation) I was shocked to discover that even in the natural sciences, grade inflation of this sort prevails. In order to distinguish among students—non-invidiously, but to discriminate—some much greater separation of degrees of distinction would seem desired.

That is one of the suggestions made in the final section of our report, which is called “Recommendations.” There are 20 recommendations, and one involves making it easier to calibrate how well students are doing against the cohort of people in that class. If a class is 80% A’s, your A in that class is going to count less than an A in a class where there are 30% A’s. That can seem like a small recommendation, but it may mean that students demand classes where distinction can show up. If they demand classes like that, they may gravitate more towards courses that have some real rigor and be a little less shameless about attending big lecture semi-gut courses in the social sciences or even the humanities that assure them of a good grade, thereby mingling with the respectable crowd of Ivy graduates.

Mounk: One topic that has been broadly discussed in the media, but also felt by many professors—including left-leaning professors at universities—is a sense of political conformity and a fear that one may, in some explicit or implicit way, be punished for expressing unpopular views. The report mentions that the number of students and faculty members at Yale who fear sharing a political view is quite substantial. How has the problem of political conformity contributed to a loss of trust in universities?

Bromwich: It is a reductive way of framing this, but not without its own revelations, that 90% or more of faculty at universities like this tend to be registered Democrats and to have, let’s just say, left-liberal politics. Why is that? That’s a long story. I have my own speculative explanations. But there’s no doubt about the enormous, sharp divide in the political culture of the United States right now. It’s a 51-49 politics, counting it simply as Democrat versus Republican, and it keeps going back and forth. So it’s about even, and what these two parties stand for is not quite clear very often, even to themselves. To have it all leaning with great imbalance on one side doesn’t seem an adequate representation, or an adequate preparation for society, among the students who go to college.

That’s not to say that there’s a great deal of political indoctrination that is just part of the ethic of campus life. I don’t think that’s true at Yale. I think it’s been overplayed as a problem about universities in general. But there’s no doubt that the actual disproportion of political tendencies among faculty members was a factor in creating more distrust.

We hope you’re enjoying the podcast! If you’re a paying subscriber, you can set up the premium feed on your favorite podcast app at writing.yaschamounk.com/listen. This will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!

Subscribe now

Set Up Podcast

If you have any questions or issues setting up the full podcast feed on a third-party app, please email leonora.barclay@persuasion.community

Then there’s the question of free inquiry and free speech—related to politics, but related to cultural and social issues too. Are there questions that could be important moral topics for discussion in a university that are pretty much off limits, or where it’s understood—with nobody having to say so—that there’s one right answer, one right response, one right side to take, so that students are, imperceptibly but nevertheless, discouraged from getting into animated discussions about these things in class or outside of class?

Mounk: Part of what makes students afraid to speak up in class is the threat of social media. If a student says something that a classmate finds offensive, that classmate may go on TikTok or some other social media platform, perhaps misrepresent what was said, and call it out—leaving that student with no friends, with nowhere to sit in the dining hall, and so on.

Broomwich: Students are wary of saying something controversial because it might be reported in such a way as to harm their reputations early on, and how that affects people’s willingness to speak is pretty easy to guess. One of the recommendations—and it is one I’m most proud of on behalf of this committee—is a no-gizmo classroom: no laptops, no iPhones, no recording. It was suggested that we propose Chatham House rules for all classes, which means you can report things that were said, but you must never report who said it. I don’t think we went that far, and I don’t think it’s necessary to go that far. But you want to stop well short of permissiveness towards creating gossip around comments made in class, either by a teacher or a student.

Mounk: I want to double-click on a few of these because there are a lot of interesting things in there. I’m struck by your first observation that there is a lot of opacity. I don’t think that opacity is by design exactly—in some areas perhaps more than others—but it has been created over time.

It’s true that a lot of the things that universities do, that supposedly are meant to serve worthy goals like equity, in fact have the result of favoring the people who know how to play the game. Personal statements are meant to give admissions officers a richer view of a personality and allow students to share when they’ve dealt with some kind of genuine hardship. In reality, it is often the most privileged students, from the most privileged backgrounds, who have the cultural knowledge to understand what you do and don’t say in a personal statement, and who have had the money to go volunteering on some wonderful project somewhere.

With the financial aid system, there are some really strong reasons why universities have embraced the models they have. The logic is that students who come from very rich households pay a lot, and students who don’t come from rich households are not charged any tuition, or are perhaps even given full living expenses. The top universities with big endowments are now very generous in that regard. But the result is that if you come from a genuinely underprivileged background, you may not know that. You may only have heard in general conversation that going to Yale now costs about $100,000 a year all in. So you may not even apply.

Not to mention that there are surely some parents who are smart about how they plan their finances and income—if you’re self-employed, you might frontload some losses to your business in the year your kid goes off to one of these schools, and because you really know the system, you’re able to come in under $200,000 that year, even if in the years before and after your kid attends college you’re way over $200,000 a year.

These are all examples in which a system—often designed quite earnestly, for example to make college accessible, which it now factually is for most people who are not very affluent—remains so opaque that people may not know it. If your kid gets into Yale, Harvard, Johns Hopkins, and so on, you do actually get tremendous aid. But the system is so opaque that people may not know it. How did that come about, and is it fixable?

Bromwich: I don’t know how it came about. As the scholarship process became more intricate, with more kids from public schools being admitted to elite schools—the Ivies, Chicago, and so on—I suppose they found it difficult to explain the intricacies. There is also the gradual advent of what came to be called the holistic approach. This is what Harvard got hit very hard for in the fairness in admissions case before the Supreme Court, because the holistic approach could be shown to disfavor Asian students on personality criteria that are very obscure. They kept receiving low grades on those criteria to compensate for their high SATs, so that the university wouldn’t have to admit too many.

Mounk: To expand on this for a moment, there is a dark history here. Those personality scores were explicitly introduced to keep down the number of Jews at these schools some 50 years ago, and now they are being used to keep down the number of Asian students. The figures were just remarkable. I believe it was on a five-point scale, and the personality of the average Asian applicant was determined to be more than a standard deviation lower than that of any other racial group. Basically, Harvard University’s institutional judgment is that, on average, Asian students have remarkably terrible personalities—and all of this is just part of this opaque system.

Bromwich: It’s all a cultural and racist cliché, but it fits all the old models. Black people are just more interesting, more animated, more lively than everyone else—they have personality. Asian people, on the other hand, are harder to read; they all seem to have, somehow, not enough personality. White people are somewhere in between. It was absurd what came out of that.

All these schools have been practicing some version of holistic admissions. That system depends on the spontaneous judgment and tact of the people working in admissions when reading an application. They should presumably be people who have been taught some basic rules: what are the good schools, what are the hard schools, what are the districts where, if a student has really done something exceptional, it means a lot—and so on. Nevertheless, that system, because it is so personal and so subjective, is liable to particular abuses. No doubt about it.

Subscribe now

A well-meaning democratic desire to present a welcoming face to students of all kinds is also largely responsible for some of the obscurity of the process. One recommendation from our report is that schools not advertise sticker price alone, but actually go into some of the intricacies, showing just what kind of chance an applicant has. But a subjective element remains. There will be students who feel prompted to try their chance at Yale, Harvard, Princeton, or elsewhere, but who, on objective measures such as the SAT, aren’t quite up there with the best students—students who do come from more privileged backgrounds but are more academically prepared for a difficult university.

Mounk: One of the interesting things that the report suggests is that over time the mission of the university has broadened explicitly, and with it the criteria for the kind of students you want to admit. The report suggests that the university should refocus its mission much more narrowly on the creation, preservation, and transmission of knowledge, rather than those vague goals of educating future leaders and all kinds of other things. Going with that, it suggests that academic merit should be the core criterion for admissions, in a way that evidently it is not always now.

How far should we go with that? Why shouldn’t we do what other top universities in most other countries do? The report acknowledges that in most other countries, universities like Yale and Harvard—as we have recognized in the last year during the Trump administration’s attacks on these universities—are private universities with large endowments, but they do actually receive a lot of public funding. One obvious way to preserve trust in them is to have a very transparent, clear academic metric for who should get in, whether that is a national entrance examination or a university-specific entrance exam—something where people are graded without knowledge of their identities. The people who come out on top are admitted. That is broadly speaking the system that most other democratic countries in the world use. Why not go the whole way and get rid of holistic admissions altogether?

Bromwich: I am probably closer to that view than some of my colleagues on this committee, but we went a distance towards it. As you mentioned, what these mission statements import is a hard thing to say. They started becoming inspirational—part of the brochures sent out to prospective applicants—about 30 years ago or so. For example, all of these elite universities will speak of their virtues as being a “second home.” The word “home” is used, I would say, even half as often as the word “community”—as if the university is a whole separate community. That leads to some fallacies about the sort of concord or comity that ought to exist among everybody, fallacies that are very hard to erase but should be erased. If you want what John Stuart Mill called the clash of ideas to happen in a university, that clash means friction, some abrasiveness—not necessarily wounded feelings, but surprise, startlement, shock, the feeling of being rubbed the wrong way in an argument, hearing for the first time an argument made well that you didn’t think you had to take seriously. You want that to be part of what goes on in universities.

As a University of Chicago philosophy professor once told me, we respect you here if you can defend your ideas. That’s a nice way of putting it—maybe a little less forbidding than “clash of ideas.” Students, by the time they are some way into university life, should be interested in defending their ideas. If they are going to, for example, a lecture or a political speech by a person whom they dislike, distrust, and are opposed to, they should have some pride in their ability to ask a hard question instead of shouting the person down. Why? Because it’s a university.

That seems to me to move towards wanting people who are qualified. Are the people who are qualified the people who score the highest on these merit tests? On the whole, I think that should be a guiding line. I believe it is one of the recommendations in our report: that a baseline be set explicitly—which is not there yet in Yale’s advertisements for potential applicants—that if your SAT score is below a certain threshold, and it’s going to be a pretty high one, you are not encouraged to apply because you’re probably only in for disappointment. They don’t do that yet. In fact, in previous outreach to potential applicants, they say don’t worry too much about SATs.

As you know, schools like Stanford, Berkeley, Yale, Columbia, and Harvard, around the time of the George Floyd protests and COVID in 2020-2021, abandoned the SAT for a while and then gradually took it back because it was so impractical not to have it be a factor at all. There shouldn’t be a hangdog attitude about using this test—it’s merely objective. It should be a good test, one that tells you something about the student.

Mounk: One of the interesting things about the SAT test is that it went test-optional in most of those universities, which of course meant that if you had a very good score, you would include it. But if you had a bad score—yet otherwise had a great experience volunteering in Ecuador, or teachers who were really pushing you—you would omit it. The effect of that in the incoming class was, in many universities, very negative: there was a serious drop-off in intellectual quality.

The other thing this speaks to is exactly the nexus of admissions and opacity, and perhaps—I don’t like the term too much—a form of mollycoddling that has just become the background hum of the American elite class. There is a very odd mix of a highly meritocratic culture, with outsized returns for getting into Yale or into the investment bank you want to work for, and a culture that is very reluctant to be explicit about criteria. You see that in admissions and you see that in grading.

Part of the advantage of not having minimum SAT cutoffs, let’s be clear, is that you can do all kinds of social engineering. That’s one reason why universities are reluctant to adopt them. But part of it is simply that it feels mean to tell people that if your SAT is below 1300, you probably don’t stand a chance of getting into Yale. So let’s not say that explicitly—even though in reality, I’m sure these admissions offices that claim to read every file holistically probably discard every application below certain criteria without looking at it for more than two seconds. By not writing it on the website, you are actually inviting lots of students to spend a great deal of time on these applications, to get their hopes up, to think that perhaps they are such a unique student that their wonderfully neglected talents will be recognized—only to be disappointed six months later. It’s a kind of fake niceness.

Subscribe now

That is related to grade inflation. I am a relatively soft teacher in the sense that I don’t want my students to be punished for taking my class rather than the class of some colleague of mine. Because grade inflation has become the practice, my grades are about as inflated. I try not to make them more inflated than those of my colleagues, but I don’t think they’re less inflated either. There is just no individual incentive to hold firm. But I think that’s also fake niceness. My students deserve to know whether the piece of writing they have handed in is poor, decent, good, or exceptional. At the moment we don’t have the signals to send them for them to know that about themselves. At the moment that feels nice, but it isn’t actually nice if we take intellectual development seriously. Universities should have a stronger sense of mission and be able to stand up against this better than they have. But it’s a downstream effect, I think, from a broader set of cultural attitudes that have become very widespread in the American professional-managerial class.

Bromwich: There is a well-meaning and rather innocent democratic idealism reflected here: we want all kinds of people, we don’t want to feel that we discriminate, we don’t want to feel we’re an aristocratic country. As the Ivy League—which was an aristocracy if anything was—began to become more democratic, at least in its surface presentation, holistic considerations began to take hold: find a student who has overcome an obstacle, a student who is a fascinating, strong personality and intellectually good enough. That kind of exception started to be made, and it is one form of diversity.

The word “diversity,” before Trump—who uses it to cover everything having to do with civil rights, gender, and whatever else—really meant devoting special attention to race, gender, ethnic background, immigration status, and cultural and ethnic issues. But diversity is also just wanting different types of people, and that impulse goes far back.

This is touched on briefly in the preface to our report. Anti-intellectualism—or, to put it more politely, non-intellectualism—has been a major strand of American life. It seems wrong for the academy to present itself as participating in this unprejudiced non-intellectualism, but there is a little bit of that too. It leads to unfortunate effects when you try to have a popular face while not really being a popular institution.

Mounk: I find the idea that you need to put together a diverse class very strange. I was an undergraduate in England at Cambridge, and the admissions system is so fragmented that there’s no chance of putting together a unified class for the whole university, because colleges do their own admissions per subject. There’s no real way of doing that. Yet the university orchestra somehow always had a second violinist. The idea in America is that you need to make sure that you have this kind of student and that kind of student. But there is a law of large numbers in statistics: if you admit the smartest students, you’re going to end up having one student who happens to be interested in music and one student who happens to be interested in sports. I just don’t buy the premise that active curation is necessary to create a diverse class. My class at Cambridge was every bit as diverse in terms of interests, talents, and the way students spent their free time as a class in America—without any constituency being able to say, we need to make sure we have somebody who can do this sport or that.

The report rightly calls out all kinds of special categories that continue to be given preferential treatment. These include athletes, the children of alumni, and—this is a minor point numerically, but one I find very striking—the children of faculty and staff, who explicitly get a leg up. That is really quite remarkable.

The one thing the report does not include is a treatment of race. That is partially because officially we no longer have affirmative action at Ivy League universities in the wake of recent Supreme Court judgments. But that is where the opacity comes back in. When you look at some of the amicus briefs that various universities wrote in the Supreme Court litigation about affirmative action, they said that the number of black students would decline radically if they were not able to practice affirmative action. In the case of Harvard, one of the calculations presented by Harvard’s side said that without affirmative action, the number of black students at the university would go down from about 14% to something like 2%. Yet after the Supreme Court outlawed affirmative action, the number didn’t budge—or it budged only a tiny bit. At Yale, it went from 14% before the Supreme Court judgment to 14% a couple of years later, and I think it has since fallen a little to 12%.

Bromwich: The desire to have the percentage of students in an entering class reflect the percentage in the population goes with, I think, a fallacy that Michael Oakeshott talks about in some of his essays on education, which he calls the reflection theory of culture. That is to say, the idea that higher education and universities should, in all possible respects, reflect the society they are meant to serve. But institutions are good at functioning for different purposes. If you’re thinking about an activity—call it an institution—such as classical music, or engineering at a high level of specialization, or the armed forces, the kind of abilities needed for one thing or another aren’t necessarily going to reflect the distribution in the population. Universities are a large institution and you should perhaps strive to make them as representative as possible. But the reflection theory also carries the trouble that it makes the university think it should reflect attitudes in society as well as populations. Again, I think that’s wrong.

Mounk: That brings us to the question of intellectual diversity. You alluded to this briefly earlier: the vast majority of faculty at the leading schools lean left, many of them registered Democrats. That in itself need not be a problem. But I do think it’s indicative of a deeper problem—that the debate at these universities lies, as I see it, between the identitarian left and the liberal left. There is probably a slight preponderance of the liberal left on most campuses, not necessarily in every department or every field. But everything that falls outside of that range is hard to formulate at those universities without experiencing significant pushback.

Some of the solutions to this problem I view somewhat critically. In an ideal world, we wouldn’t hire colleagues at our universities in order for them to be intellectually diverse. First, it’s very hard for one person to be intellectually diverse in themselves unless they hold a very incoherent view of the world. Second, it’s rather odd to have a colleague who is just there to have a different opinion. But the truth is that political criteria are currently a very large part of the selection process, especially in the social sciences and the humanities, where the quality of work certainly matters, but where people whose views fall too far outside the consensus in a particular department often aren’t even considered in the first place. There is an effective application of ideological criteria that is covert and sometimes not even self-conscious—it’s simply that this person falls outside a reasonable fold, and so they are not considered. That is a very effective implicit political filter.

Unless you are able to upend that, perhaps the only alternative is to sometimes hire people who fall outside that consensus. It’s very hard to know how to fix this. As in the case of a broader reluctance to state criteria explicitly, this too is downstream from a larger transformation of the professional-managerial class. A lot of the problem today is that the Democrats have simply become the party of the professional-managerial class. As professors, university administrators, and (to a significant extent) students are recruited from the ranks of the professional-managerial class and its offspring, there is always going to be some lean in that direction, as long as that is the nature of our political cleavage.

So how can universities improve on intellectual diversity and foster genuine debates on campus, in a way that doesn’t itself run counter to the principles of free inquiry and that avoids political litmus tests for faculty hiring?

Bromwich: It’s very hard. What you’re describing is a standoff between two desirable goods. On the one hand, the intellectual autonomy of departments: departments should be able to choose to hire, and then possibly to tenure, the scholars they consider the best in a given field. On the other hand, if you have very little variety of opinion in a field such as politics—where there is in fact great variety outside the academy—there is something lacking in the education students are going to get.

This is called, by people who do the kind of study this report was working at, the pipeline problem. The pipeline problem of having so few black people represented in the academy was addressed by affirmative action, but affirmative action is no longer constitutional and may have run its course in any case.

Where the left-liberal side got its strongest foothold was in certain departments in the humanities and social sciences, but also in the studies programs—black studies, ethnic studies of various kinds, immigration studies, gender studies, and so on.

Mounk: Is that where the left-liberal or the identitarian left got its foothold? My sense—and I know Harvard much better than Yale—is that at Harvard, the government department and the main faculties have a left-liberal predominance.

Bromwich: Correction taken. I meant consciously political and consciously left, but you’re right—that would better be described as identitarian.

In any case, if you want a pipeline going the other way, you can’t do it by political affirmative action. That is, as you said, a litmus test. But universities—Harvard, Yale, Princeton, Berkeley, Stanford, and so on—have created what they call centers or institutes, which aren’t quite academic programs, let alone full departments, but where people can have a permanent presence on campus and teach credit-bearing classes. It is imaginable that, at some places, if done with imagination and scruple, these centers could provide more room for people with academic training to hold views that inform their teaching without becoming dogma—or to hold views that are libertarian in ways that fall outside the usual range. That is one solution, but it is a standoff. It is a hell of a problem, how to change the composition of faculties without somehow compromising intellectual and departmental autonomy.

Mounk: What about the student side of this? The report refers, at surprising length I thought, to the famous Halloween incident, in which there was an email from the office of intercultural affairs urging people to be very sensitive in the Halloween costumes they chose. There was then a response to that by Erika Christakis, who was—as it was then called—the master, along with her husband Nicholas Christakis, the associate master, of one of the residential houses at Yale. Her response suggested that students should be able to think for themselves about these kinds of things. There was a huge eruption of anger that did look rather like some form of cultural revolution—an effigy of Nicholas in the courtyard and so on. A subsection of students, but an influential subsection, saw it as their calling to impose a certain identitarian orthodoxy on campus. That included intimidating senior faculty members, as in this instance, and sometimes intimidating their fellow students. There is some discussion in the report about the norms that should be established so that people feel free to share their opinions in class without the fear of being canceled afterwards on social media or in other ways.

I have personally found that that moment has somewhat passed. For the last few years of teaching, I’ve been struck by the fact that for many students, these ideas are now the received wisdom of what they’ve been told in high school, middle school, and sometimes elementary school. It depends a little on where they grew up—if they grew up, for example, in the suburbs of a major metropolitan area like New York, Los Angeles, or Boston, these are largely just the ideas their teachers have always taught them. As a result, they tend to take them for granted. It’s the world they’ve grown up in. But they no longer think of themselves as bearers of a flame whose goal is to impose it on others. It’s more that this is what their teachers told them and, like most things their teachers told them, they assume it’s true—until somebody has a different opinion, which surprises them, and they find they want to talk about it. It feels somewhat less fraught.

What is your impression of campus culture at Yale—has it somewhat improved? And more broadly, how big is the challenge today and what are some possible solutions? One of the recommendations in the report is to remove electronics from the classroom. The report also entertains but ultimately dismisses the idea of a Chatham House rule prohibiting attribution of remarks made in class. Where do things stand and what can be done?

Bromwich: I agree with your impression that the pressure zone, so to speak, has lifted somewhat in the last two or three years. A person I’m very close to in this household, who is a psychologist, speculated that in the year 2020 this country had a nervous breakdown—the whole country had a nervous breakdown. COVID was an element of it, but so were the George Floyd protests and the disorders in cities. The Halloween event you mentioned came in 2015-16, so that is an earlier BLM moment, and it had been going on for a long time—really from Obama’s second term into recent days.

But I do agree that there is more tolerance and interest in the exchange of ideas, and more animated talk and susceptibility to humor in the classroom, as far as I am able to discern. That is a good thing. It means that younger teachers, who tend to be cautious because they worry about student evaluations, will not constantly develop twenty different tones of voice for saying that’s interesting—and may instead learn how to say, I don’t really think that’s true and here’s why, or that’s absolutely true but let’s go further with it. That kind of give and take in a classroom is very important.

One of the things you could hear from conservative students in the worse recent years—here at Yale, which is not as badly off as other places because it has a conservative political union and a conservative presence in the background, so those students don’t feel entirely isolated—is that sometimes in class, when they made a remark that showed their traditionalist sympathies, they would not get denounced or shamed by anyone in the class or by the teacher, but the teacher would be silent for thirty seconds and then move on to something else. That’s not good. A greater openness that reflects a community with divergent, somewhat disharmonizing points of view is a good thing, to the extent that it really is coming back. I think it may be.

Mounk: Is one challenge not explicitly mentioned in the report that there is a minority of faculty who really do abuse their power in the classroom to impose their ideological views? My impression is that most faculty members do not do that, and I certainly think most of my colleagues do not. But I am struck, speaking to students, that they very consistently bring up experiences of this.

I go out of my way, when I explain what I think a good essay is for my class, to make clear that I don’t care whether students agree with me or not. The last thing I want is for somebody to badly parrot my views back to me—that’s not going to earn a great grade. If you write something I agree with, I want to feel that it is a really strong, interesting representation of that view. If you write something I disagree with, the question is not whether I am going to come to agree with you—that is unlikely. What I want to feel is whether the pull of that argument in my mind is a little bit stronger after reading your essay than it was at the beginning. Have you made a case for a point of view I may happen to disagree with, where I find myself thinking: I still probably disagree, but I see the force of that—I see why somebody might believe that.

Because I am very explicit about this, students sometimes say that they feel they can say what they think in my class, because it is not always the case elsewhere. I always listen up at that. When is it not the case? I remember one recent instance in particular where a student said that in high school, a teacher had a set of views—actually a rather woke set of views, views the student largely shared—but the student chafed at the fact that any disagreement was clearly going to be punished.

Are there ways that universities need to reckon with that, even just in terms of training teachers and setting clear expectations? You don’t want to be too intrusive, and you certainly don’t want auditing of every grade, which would be open to abuse in all kinds of other ways. But even if it is one in twenty faculty members, that is an experience most undergraduates are going to have once or perhaps twice during their college career. Is that one of the things that damages trust in higher education, and something we need to deal with?

Bromwich: I don’t know if it is one in twenty teachers, but I agree that it goes against the ethic of teaching. If you have anything resembling a Hippocratic oath as a teacher of the liberal arts and sciences, it should be that a view that is well defended gets the respect of the teacher, and a view that is earnestly and genuinely meant but not well defended receives a response that is not crushing or personally harsh, but reasonable—an example of how to point out the fallacies or the loopholes in a rival argument. Teachers should be exemplary in that way and it should be part of their training. But what is involved in acquiring a PhD is so specialized in other ways that that element of pedagogy is often neglected.

Some of it is intuitive. Some of it is, as I’m sure you have experienced, getting to know what your classroom presence is like and what it’s like to teach different kinds of classes. I remember in my early days teaching at Princeton—I was in my 20s—being morbidly worried about what I thought of as dead time, what disc jockeys call dead air: nothing being said. If I asked a question and there was no answer for ten seconds, I got very anxious. But as a teacher you learn that if a question was well asked, you simply wait.

Mounk: The best piece of teaching advice I’ve ever gotten was in graduate school. The department put on a little meeting for people who were going to be teaching assistants for the first time, and the best piece of advice—which I follow very often—is this: if you ask a question and you look anxious that people might not answer it, and you look like you’re going to jump in yourself, students think they can just sit back and let the teacher do their thing. What you should do instead is lean back, look comfortable, and let an awkward silence arise. Some student always can’t bear the awkwardness and jumps in, and suddenly you have a real discussion. I don’t generally have trouble getting my students to talk in the classroom, but on the occasions when that happens, I do that. Sometimes I even call it out explicitly: I’m happy to sit here in awkward silence for as long as you like. When you do that, somebody always jumps in.

Bromwich: Sometimes you haven’t asked the question well, and that shouldn’t lead to too much self-consciousness either. You learn to listen to yourself and say, a few seconds later, you know what? I posed that badly. Let me try again. There was a philosophy professor at Princeton—modal logic and philosophy of language and other such things—David Lewis, who spoke perfect paragraphs and was an excellent lecturer. When he would walk back and forth on the stage and, in the middle of a sentence, realize he was about to talk nonsense, he would stop and say so, then go back and say the whole thing again. Teachers, in a quieter way, should be able to do that.

Let me read a passage from a British education master of the 19th century that is quoted by Oakeshott, because I think it speaks to exactly what we’re talking about—the kinds of habits and manners that go with thinking in the context of higher education:

A certain amount of knowledge you can indeed with average faculties acquire so as to retain, nor need you regret the hours you spend on much that is forgotten, for the shadow of lost knowledge at least protects you from many illusions. But you go to a great school not so much for knowledge as for arts and habits, for the habit of attention, for the art of expression, for the art of assuming at a moment’s notice a new intellectual position, for the art of entering quickly into another person’s thoughts, for the habit of submitting to censure and refutation, for the art of indicating assent or dissent in graduated terms.

That last part is obviously what has become most challenging for students in the last generation or so.

In the rest of this conversation, Yascha and David discuss why the number of students choosing humanities subjects is declining and how colleges and universities can regain public trust. This part of the conversation is reserved for paying subscribers…

Read more

As Donald Trump tries to strong-arm the media into compliance using federal agencies as his weapon, and to purge undesirable content everywhere from museums to universities, freedom of speech in America under the second Trump administration faces real and urgent dangers. Amid these dangers, charges of hypocrisy and opportunism are frequently leveled at those who advocated for free speech and intellectual tolerance during the era when censorious “woke” progressivism was at its height (roughly 2013-2021). The title of an episode of the progressive podcast “Cancel Me, Daddy” sums it up: “They Screamed ‘Cancel Culture’—Then Went Silent While Trump Gutted Free Speech.”

Others, such as writer David Klion in The Nation last June and, most recently, attorney and blogger Ken White in The UnPopulist, make a charge that goes beyond hypocrisy or silence: they argue that critics of “wokeness”—such as the signatories of the July 2020 “Letter on Justice and Open Debate” in Harper’s Magazine—bear at least some blame for the Trump administration’s war on speech.1 Such critics, it is argued, partly enabled Trump’s current actions with their hyperbolic claims of speech suppression by the left.

Persuasion events

Likewise, former Yale philosophy professor Jason Stanley, now at the University of Toronto, has accused the Foundation for Individual Rights and Expression (FIRE), the premier free-speech group that challenged speech suppression by the illiberal left at the height of “woke,” of fostering a “moral panic about leftism [in] universities” that supposedly cleared the way for the Trump administration’s assaults on the academy.

Is there any substance to these arguments?

First, the hypocrisy claim. It is very true that some vocal opponents of “woke” left-wing illiberalism have not acquitted themselves well under Trump. Political hacks like anti-woke activist Christopher Rufo have gleefully embraced right-wing “cancel culture,” and have been quite upfront about having double standards on the issue.

Before the 2024 election, I wrote about other anti-woke culture warriors who had gotten on the Trump train in the hope that a new Trump presidency would effectively combat far-left progressivism. Some have stayed on that train. Colin Wright, an evolutionary biologist who claims he was denied a position at Cornell University because of discrimination favoring racial minorities, has not only defended Trump’s widely criticized deportation policies, but has insisted that “for wokeness to go away—to salt the Earth of it—Republicans must win again in 2028.”

When you start talking about “salting the earth” to exterminate a viewpoint you dislike, maybe it’s time to consider the “are we the baddies?” meme.

A far more prominent—though more complicated—case is Bari Weiss, whose magazine, The Free Press, was recently acquired by Paramount under the new leadership of Trump-friendly tycoon David Ellison. Weiss herself was elevated to editor-in-chief at CBS News last October.

Despite Weiss’s stated commitment to fairness, independence, and truth-seeking, both Weiss and The Free Press have taken a largely accommodating stance toward the Trump administration. A lot of Free Press content still focuses on the absurdities and abuses of the “woke left,” including articles that revisit “cancel culture” episodes from a few years ago.

Much of its coverage of the Trump presidency, meanwhile, qualifies as out-and-out cheerleading or even attempts to push the administration further. Last year, the site published a manifesto by Rufo (co-signed by a few dozen mostly pro-Trump pundits) calling for a de facto White House takeover of American universities—with no reply or counterpoint from any Rufo (or Trump) critics, many of whom are also critics of the academic left. Rufo’s proposed “contract with the universities” would strip schools not only of public funds but of accreditation for failing to comply with the right’s nebulous ideological demands.

Perhaps more disturbingly, The Free Press has trodden softly on an issue indirectly related to Weiss’s position at CBS: the Trump administration’s use of the FCC’s powers to strong-arm broadcast media. (Its article on the subject focused on the agency’s excessive powers in general, only briefly mentioning the specific, and unique, Trump-era abuses.)

Meanwhile, the site’s criticism of the government—including immigration enforcement tactics that trample due process, and blatant abuse of federal prosecutorial powers to punish Trump’s enemies—tends to be tempered with soft-pedaling and “Democrats do it too” both-sidesing.

Subscribe now

I happen to agree that it’s important not to forget the real harms of left-wing illiberalism (more on that in a moment). But to act as if left-wing illiberalism is still the main threat to freedom of speech in America during the second Trump administration is an egregious failure to keep things in perspective. Weiss has been wrongly caricatured as a pro-Trump or right-wing propagandist, but it’s fair to say that she has failed to live up to her stated principle of holding both left and right accountable without fear or favor.

But does this reflect a more general trend among anti-woke “free speech warriors”?

Of the still-living signatories of the Harper’s Letter, I could find one, the writer Sarah Haider, who has emerged as cautiously pro-Trump. She hailed Trump’s victory in November 2024 and suggested that concerns about the harmful effects of his presidency were overblown. (There is no evidence that she has come around to a more critical view since; her recent posts on X are still in a “Democrats bad” vein.) Three or four others, including Weiss, can be counted in the “anti-anti” or “both sides” column, or punch predominantly left.

But such people are vastly outnumbered by those signatories who have been scathingly critical of the administration: Anne Applebaum, Garry Kasparov, Thomas Chatterton Williams, Jonathan Rauch, Nadine Strossen, Caitlin Flanagan, David Frum, Matt Yglesias, Damon Linker, Judith Shulevitz, Jesse Singal, Steven Pinker, and Persuasion’s own Yascha Mounk and Francis Fukuyama—and that’s far from a complete list. The “where are they now?” critique is manifestly wrong.

It is also worth noting that FIRE, the vanguard of opposition to “woke” speech-policing, has emerged as one of the strongest free speech advocates opposing Trump’s attempts to bully and muzzle critics. It filed a legal challenge to the administration’s move to deport noncitizen students who have expressed pro-Palestinian views, and backed Harvard’s lawsuit challenging the Trump administration’s cuts in research funds. It also represented, pro bono, Iowa pollster J. Ann Selzer, who had been sued for “fraud” by a Trump supporter claiming that she had misrepresented poll results to help Kamala Harris. And it has stood up against Trump’s own lawsuits against media organizations.

So, for the most part, the “free speech warriors” have acquitted themselves quite well.

That brings us to the second charge. Klion concedes that many signers of the Harper’s Letter have criticized Trump’s abuses—but he thinks they’re evading their own partial responsibility for our current predicament. What they did, according to Klion, was “build a broad elite consensus that has functioned mainly to legitimize Trump’s actions.”

White’s piece in The UnPopulist makes a more fundamental case against “the ‘free speech culture’ ethos” espoused by FIRE and the Harper’s Letter—an ethos that he believes absurdly inflates the threat from the progressive left and allows bad-faith actors on the right, including authoritarians like Trump and Elon Musk, to pose as free speech warriors challenging a “totalitarian” order.

Few people would deny that pro-free-speech arguments criticizing the left can be, and are, disingenuously misused by bad-faith actors on the right. White’s point, however, is that these were bad-faith arguments in the first place: in his view, “free speech culture” champions have relentlessly hyped a supposed left-wing threat from “relatively powerless people like students,” whose worst offenses are to respond too rudely or obnoxiously to speech they consider harmful. The result, he says, is to blur the lines between rudeness and repression, and to elevate a phantom menace that serves both to distract from and to excuse repressive measures by “antiwoke” government officials.

But this account fundamentally misunderstands, or misrepresents, what happened in American culture during the “Great Awokening.”

To begin with, one may quibble with the depiction of students as “powerless” when they have the institutional backing of schools. During the protests that followed the murder of George Floyd in the summer of 2020, Leslie Neal-Boylan, Dean of the University of Massachusetts-Lowell School of Nursing, was fired after a student’s Twitter “call-out” of her email which said, “BLACK LIVES MATTER, but also, EVERYONE’S LIFE MATTERS.” Several years earlier, Northwestern professor Laura Kipnis was put through two formal Title IX investigations for criticizing what she saw as overreactions by colleges to allegations of sexual misconduct.

What’s more, plenty of “cancel culture” episodes happened in non-academic settings, and involved people being fired, or forced out of their jobs, over very tame “problematic” opinions or trivial offenses against progressive racial etiquette.

Neal-Boylan was one of several people, including a school principal and a sports broadcast host, who lost their jobs in the wake of the 2020 protests for social media posts dissenting from the Black Lives Matter movement over its coercive tactics or its racial focus. Around the same time, the Journal of the American Medical Association fired a deputy editor over a podcast questioning the existence of “structural racism” in medicine. Philadelphia Inquirer editor Stan Wischnowski was pressured into resigning over an article on the damage from property destruction during riots, headlined “Buildings Matter Too.”

Subscribe now

Meanwhile, a senior curator at the San Francisco Museum of Modern Art stepped down under fire for “white supremacist beliefs” because he said that refusing to acquire works by white artists is “reverse discrimination.” A Denver yoga studio chain was driven out of business by a social media backlash after “callouts” from a few employees who felt that the management’s statement of solidarity with Black Lives Matter was “too little, too late,” and who dredged up other vague grievances.

The list goes on and on—and, contrary to White’s assertion that the backlash against “cancel culture” has protected primarily powerful people such as pundits or politicians, many of the casualties were working-class employees. A truck driver lost his job after being accused of making a “white supremacist” gesture at another driver. (Around that time, there was a trollish effort by the far right to popularize the idea that the “okay” symbol—an “O” formed with two fingers—is a badge of white-power loyalty.) In Portland, Oregon, two bakery employees were fired after a black activist made a video claiming a racist slight because the shop refused to serve her several minutes after closing time.

Even children got caught in the net: Two teenage boys in California were expelled from school after a goofy photo they had shared of themselves wearing dark green acne masks was mistaken for blackface.

It’s also worth noting that critiques of left-wing illiberalism did not come solely from conservatives or liberal centrists prone to “both-sidesing.” In 2023, PEN America, a veteran free-expression advocacy group that strongly supports racial and gender diversity, issued a report documenting the chilly climate in publishing created by speech-policing in the name of social justice—including books canceled by publishers or withdrawn by authors because of an online backlash against writers of the “wrong” identity depicting “marginalized” characters and cultures.

Of course, none of the cases described above were assaults on the First Amendment, since the government was not involved in censoring speech. But are “free speech culture” advocates either unable to make that distinction, or unwilling to admit that state repression is worse than non-state “cancel culture,” as their detractors suggest?

At the “Global Free Speech Summit” held at the University of Vanderbilt in Nashville, Tennessee last October, most speakers did agree that while it’s important to continue opposing cultural repression of unpopular speech, it is also important to acknowledge that repression by the government is worse. As Jonathan Rauch put it in one of the panels, “We do have to walk and chew gum.”

True, in some ways, the existence of left-wing illiberalism does complicate opposition to right-wing authoritarianism. Trump’s moves to yank funds from universities and browbeat them into bending to his ideological agenda may look a little less outrageous if one acknowledges (as does FIRE) that many of these universities have a not-so-great record on intellectual freedom themselves—and that some of the steps schools have pledged to take as a sign of compliance, such as more support for diverse ideas and civil discourse, are positive.

Likewise, the Trump administration’s war on “diversity, equity and inclusion” initiatives at schools and in workplaces may not seem so straightforwardly authoritarian or bigoted if one admits that many DEI programs have had strong elements of enforced ideological conformity and even compelled speech, sometimes of a creepy and intrusive kind. (One program developed by the Seattle Office of Civil Rights, for example, required white city employees to share “a time in the past two to three months when [they] caused harm to a person of color.”)

But should liberals avoid talking about these issues simply because it might give ammunition to the right? That doesn’t seem like a winning strategy. White thinks that free speech advocates who oppose left-wing cancel culture spout such blatantly hypocritical “bullshit” that they turn multitudes of people off the free speech cause. But surely it’s much more off-putting to evade or sanitize the truth—evasions plenty of people are savvy enough to notice. Given today’s free-for-all media ecosystem, liberal and centrist silence about the abuses of the illiberal left is not going to sweep those problems under the rug (even if that were desirable); it will only boost right-wing media that will happily report these problems—and, sometimes, misreport them in the service of their own agenda.

Subscribe now

There is no question that the populist right, including Trump, has exploited the issue of free speech and “cancel culture” for its own authoritarian power grab. (And no, it didn’t need any assistance from the Harper’s Letter, whose outlook was explicitly anti-Trumpian, or from FIRE, which has opposed speech suppression regardless of ideology since its inception in 1999.) It is equally true that plenty of anti-woke “free speech warriors” turned out to be unprincipled or opportunistic, cheering on the government when it attempted to impose ideological conformity on colleges or bully the media.

But plenty of others have been at the forefront of standing up to the Trump administration’s assaults on civil liberties. Seeking to pillory them for imaginary sins of complicity with the right is not just intolerant—it’s remarkably self-defeating.

Cathy Young is a writer at The Bulwark, a columnist for Newsday, and a contributing editor to Reason.

Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

Subscribe now

Luis Garicano is Professor of Public Policy at the London School of Economics.

In this week’s conversation, Yascha Mounk and Luis Garicano discuss the economic magnitude of AI’s transformative potential, whether artificial intelligence complements or replaces human workers, and why Silicon Valley predictions about automation consistently miss the mark.

This transcript has been condensed and lightly edited for clarity.

Yascha Mounk: There are many things I would love to talk to you about, but the topic I have been thinking about a lot is artificial intelligence. I have had conversations on this podcast about the technology itself with people like Geoffrey Hinton. I have discussed the dimension of existential risk with people like one of the co-authors of If Anybody Builds It, Everyone Dies. I have also thought about some of the broader public policy angles.

However, I have not yet had a conversation specifically about the economics of artificial intelligence. It would be really interesting to try to get a handle on those questions. We will focus particularly on the labor market, but before we get there: what, in general, do you expect the impact of AI to be? Is it going to be major, middling, or minor? Is it going to lead to the vast economic growth some are predicting, or is it going to really decimate the number of jobs out there for humans? Is this going to be an economically revolutionary time, or is it just one of many developments that are interesting but ultimately not that consequential?

Luis Garicano: I don’t have a crystal ball—anticipating things is always hard. But let me give you my best take based on what we see. It is clear that a lot of knowledge work, even if the technology stopped tomorrow, could be automated—a lot of knowledge tasks. It is already very clear that tasks which are routine, tasks that have to do with diagnosis, writing, crafting documents, doing research—the AI is already doing these perfectly. Coding work is really spectacular. In terms of whether it is going to be big, I think it is going to be huge. It is probably as big a revolution as the industrial revolution—that is a very likely thing—except that instead of automating physical work, it is for cognitive work. Everything points to a large impact, and also an accelerating one.

There were people who were doubting, people who were wondering if AI would be a big deal or not. I don’t think any of those people could still be doubting, given what we have observed in the last six or eight weeks. The explosion of new models, the way they work—Claude Code is really taking the world by storm. Everybody has noticed that software firms’ valuations are plummeting in the stock market, showing that people believe many functions, many verticals, and many software products that were accommodating one particular use case can be replaced by AI.

So yes, a big deal, and in many segments. On the question of growth: yes, if this is as big a deal as it appears, we will see big productivity growth and an acceleration—though not the kind of growth that many people in Silicon Valley predict, because most economists think in terms of O-rings and bottlenecks and weak links. Meaning: you can invent as many compounds to solve cancers as you want, but if you need to go through years of clinical trials and regulatory approvals, that is not going to suddenly accelerate massively. Those weak links will constrain growth everywhere.

On the question of labor: the evidence so far is that AI is more complementing than replacing. In the three areas where we expect the largest impacts, translators haven’t dropped—everybody thought translators were going to be decimated, translation seems like a solved problem, and yet the amount of translation work hasn’t dropped according to world labor statistics. Customer service agents: some people get let go, some get rehired to do different jobs; again, the BLS doesn’t see much. Even computer programmers—we are not seeing big drops. There were a couple of papers earlier in the year. Erik Brynjolfsson has a paper with co-authors called Canaries in a Coal Mine which was starting to see drops in more exposed segments for more junior employees, and we do see a bit of that. But there is a lot of discussion on whether that has to do with COVID and so on. For employment at the moment, it looks like AI assists more than it replaces.

It is clear that AI can do many tasks. My main quarrel with the Silicon Valley interpretation of things is the belief that if a computer can replace the tasks most easily done by a machine, then the job is gone. Jobs are more than their most automatable tasks—a radiologist, for instance, spends only 30% of his time looking at scans. The job of a radiologist is much more than just diagnosing scans.

Mounk: I was in a meeting with Sam Altman in, I believe, 2018—I barely knew who he was at that time. I remember him pointing outside the window of a hotel in Silicon Valley, saying that in three or five years there were going to be robots building homes there. None of that has materialized. There is a very real tendency of people in Silicon Valley not just to overpromise on the technology, but to underestimate the obstacles to real-world adoption of technology. Those obstacles are particularly evident, as you recently pointed out, in something like house construction, where the constraints are not actually the inability to build homes—we know how to build homes. It is regulatory approval, zoning laws, concerns about whether the nature of a neighborhood is going to change, and all of those kinds of things.

Garicano: Two points about the comments you made. One is about the Silicon Valley position. I am very surprised that they are not just hyping the technology—which I understand, because you want to sell enterprise subscriptions—but they are also hyping the risks of the technology, and all the time threatening people with extinction, saying AI will take all their jobs. I don’t see the point of this tactic. I can see that if you want to justify the valuations, they need to say that all these things are incredibly transformative—and they are transformative.

The other day, Mustafa Suleyman—the ex-DeepMind co-founder and Microsoft AI head—was saying to the FT that they are going to automate all white-collar work in 18 months. I was joking: does anyone really believe that Microsoft will actually get Outlook or Word to work properly in 18 months? I don’t think they can fix their two pieces of terrible software in 18 months. We all hate Outlook—we’ve hated it for 15 years, and I’d bet we’ll hate it in 18 months. So they’re talking about automating complete, complex jobs, and they cannot fix their own software. That’s just completely ridiculous.

Mounk: Before we dive into the substance of this, I would love for you to help us establish the premise you’re operating on. A lot of my listeners are tech-forward, and a lot are not. I still find many people in conversation who experimented with ChatGPT when it came out three or so years ago and have gone back to use it every now and again—perhaps using it instead of Google to search for certain things, or for a translation need, or for very specific tasks. They are still convinced that it hallucinates a lot, and they feel the limitations of what it can do are very strong.

Part of that, I think, is that the most commercially used ChatGPT products are not very good compared to some of the competitors now—in part because they route your requests sometimes to a really powerful model and sometimes to a not very powerful model at all. Part of it is that a lot of people use free versions of these AI tools, which are much less powerful than the ones for which you need to pay at least $20 a month. Part of it is that probably only a fraction of people who listen to this podcast have used tools like Claude Code.

We hope you’re enjoying the podcast! If you’re a paying subscriber, you can set up the premium feed on your favorite podcast app at writing.yaschamounk.com/listen. This will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!

Subscribe now

Set Up Podcast

If you have any questions or issues setting up the full podcast feed on a third-party app, please email leonora.barclay@persuasion.community

So, just to motivate what we’re talking about: when you say there has been tremendous progress over the last few months, and more broadly over the last few years, what are these tools able to do today? How are people using them in ways that are so different from what you might expect if you’re just using the free tier of ChatGPT?

Garicano: Let me give you a Claude Code example and a deep research example.

The Claude Code example is the following. What is interesting is that the machine can talk to you and it can deploy tools—it can put Python tools to work. Let me explain this in a very clear way. I did a paper: I was a member of the European Parliament, and after returning to academia, I wanted to do some research on how narratives work in the European Parliament—I wanted to show there are no trade-offs in the narratives. What I did was collect 46,000 speeches, all the speeches, downloaded them, put them in a spreadsheet. Each speech goes to ChatGPT through an API—which means it goes through a special pipe—gets processed, comes back into a spreadsheet, gets classified in certain ways, and then we analyze that classification with statistical tools. That took six months. It is a lot of work: getting each speech, sending it, bringing it back, and so on. I had done this for climate.

I then decided to use Claude Code to do all of that work—six months of work—for the topic of AI. How is the discourse in the Parliament evolving on AI? I told Claude Code—in text, no programming—here is my directory, where I had all these files, and I told it to write the entire same pipeline: get the speech, send it, classify it, analyze it, but instead of for climate, as in the original files, do it for AI. There are many Python programs involved, multiple programs I had to run. Six months of work. Six to ten hours later, there was a complete analysis by Claude Code—all the directories, all the tables, every single figure, from start to end. The difference is that you talk to it, but it can deploy all these tools, do all these things, go over the web, run code.

The second thing I would tell your listeners about, which they would very much enjoy using if they are not already, is the deep research tools. On the highest-end research frontier of these models, you ask it to research a question—for example, you might say: populism has been growing, and there are two explanations, cultural and economic. I want an in-depth literature review of all the evidence comparing those theories. I want you to spend a lot of time collecting hundreds of papers, classifying them, telling me the prevalence of the evidence, and writing a thorough research report. This is now better than what a research assistant could do over several months.

Those are two examples of what is possible at the higher end. Why are they useful for the world? Think of a lawyer. A transactions lawyer is essentially comparing a situation to existing precedents and existing case law—drafting, for example, an intellectual property agreement or a contract to buy a house. They go and find similar contracts, upload the relevant knowledge, and convert it into a new contract. If a law firm incorporates all its contracts into Claude Code and asks the system to use that knowledge to automate contract drafting, compliance, and verification, it is definitely able, right now and without any question, to handle that complete task.

Mounk: I started to use Claude Code about a month and a half ago. I have basically no coding background—I had a couple of group lessons of C++ in middle school, did a little bit of programming in statistical software in graduate school, but very limited. I was mostly a political theorist. I then took a few weeks of CS50, the famous online computer science course on edX—a very good course, and that was ten years ago. If you had set me an entry-level coding task, like programming a number guessing game, I would not have been able to do it. Now, with this tool, I have been able to program five different things that are of concrete use to me. It is just astonishing what it can do.

More broadly, some of the pitfalls that AI used to have until a few years ago are not really there anymore. When ChatGPT 3.5 launched, it didn’t have an extended thinking modality. It’s as though I ask you a challenging question and it’s part of a game show where if you don’t start answering within one second, or if you hesitate more than one second between any two words, you lose—that answer is not going to be very coherent. Now the systems, at the higher tier, talk to themselves and walk you through the process by which they attempt an answer. They try one answer, check whether it makes sense, and then say, no, actually I made a mistake, I should do this. By the time they give you output, they have thought through it in a much, much bigger way.

On the problem of hallucinations: I wrote a post on Substack about asking Claude to write a publishable paper of political theory. A number of senior colleagues in the field wrote to me after I published it saying it would absolutely have been published in a top journal if it had been submitted. I looked through some of the references—not every single one—and it was not hallucinating. It now knows, by and large, how to ensure that something actually exists, and it flags when it is uncertain. It told me: I have put in the page numbers for the canonical translations of Tocqueville—I’m not sure about those, please go and double-check them, I don’t have access to that full text. If I upload the PDF of that book, it will do it for me. So it knows what it knows and it knows what it doesn’t know. A lot of those problems have been fixed.

Now we go into the realm of economics. I don’t know whether we have reached superintelligence as defined by Dario Amodei—where we suddenly have the equivalent of a whole country of geniuses. But we certainly have the equivalent of a whole country of middle-class professionals. Suddenly the number of people who can competently draft a legal contract, and do so in ten seconds for very little money, is vastly larger than it used to be. So what does that do—first for growth? If economic growth was in some ways constrained by human capital, constrained by the number of well-trained people with access to a lot of knowledge able to carry out that work, that should mean we are going to see a real increase in economic growth. Or is it more complicated than that?

Garicano: The first-order approximation is that you have an increase in productivity and an increase in growth—that is a reasonable place to start. There are two or three caveats that I think are important in trying to figure out how big that increase is.

The first is organizations. The organization of work is intensely human. As you were hinting from my recent post on London housing: the reason 23 out of 25 boroughs of London are building zero housing this year—in 2025, there were zero housing starts—has nothing to do with technology. Giving them better technology is not going to solve the problems with the neighbors, with the NIMBYs, with the Greens, with the land regulation, the lawyers, and all the other things that stop construction that we already know about. So the first caveat is organizations and all-too-human obstacles, which mean that even when the technology is there, many other factors have to collaborate.

Subscribe now

There are entire sectors which have Baumol characteristics. Baumol had an observation in the 1960s—and maybe you have discussed this with your listeners—that a string quartet would still take one hour to play a Mozart piece, the same exact hour as it would have taken 200 years ago: four people, one hour, no productivity gain.

Mounk: This is a very old point because nowadays, no economist would talk about string quartets.

Garicano: This observation holds for a very large share of the economy. For hairdressers, cooks—technology doesn’t play any role. It’s not just that there are bottlenecks, but that productivity growth is very small because there is really no actual technology and no actual AI involved.

What is interesting is that in the sector of the economy that enjoys technological change, as prices drop, it is perfectly possible—and we will talk about demand elasticity in a moment—that people reach satiation and that sector becomes smaller. Think of agriculture: it became technologically fantastic, but it became smaller and smaller as it grew more productive, because people’s stomachs didn’t grow. The amount of workers employed went down. What that means is that the sector with the technological expansion reduces its size, and the other sector—the one with the violinists—expands its. As a result, the weighted average of growth depends not just on how much the productive sector is growing, but on the fact that the sector that is growing may itself be getting smaller.

Mounk: One way of thinking about this is that everything that can be automated suddenly becomes plentiful. To that extent, it might not fully show up in GDP figures, but it does fundamentally remake the world. When I think about the agriculture case: as a result of the successful mechanization of agriculture, that has become a much smaller part of the economy and we are paying vastly less for food than we used to. You will understand the technical details better than I do, but that sort of underplays the degree of that change in the way we track GDP.

What it does mean is that whereas for most of human history, even people in affluent countries—if you weren’t at the very top of the hierarchy—were deeply constrained in how much food they could consume, were malnourished as a result, and died earlier as a result, nowadays, if you are anywhere outside the bottom 20% of a medium-to-affluent country, food is not your primary expense. It is a significant expense if you like nice food and shop for nice things, but if all you want is to feed yourself on ramen and a few supplements in such a way that you avoid malnutrition, that is going to be a tiny part of your budget. That is a fundamental positive transformation of human life, even if it doesn’t fully show up in GDP figures.

Garicano: Economists like to talk about welfare as the sum of consumer and producer surplus. In this case, the consumer is enjoying the biggest gain. A lot of what happens with AI is that the gains are going to consumers and not showing up in GDP figures.

Let me give you an example. We have a dishwasher that is broken. We take a picture, upload it to ChatGPT and ask what is going on. It says this part is stuck and you should just remove it. We remove it. Our welfare has gone up—we are happier, we solved the problem. Now, there is a transaction that would have taken place—some person coming to our house to fix the dishwasher—that didn’t happen. The GDP would have been higher if that person had come and we had paid him. But our welfare increased. If we can diagnose our own illnesses, if we can assess whether our diet is good or bad without going to a dietitian, if we can do our own contracts—all of those things increase our welfare but do not show up in GDP. In fact, some of them could reduce GDP.

I was talking to a CEO from China who told me he thought a lot of the gains were being “smoked in the corridor.” I asked what he meant. He said he observed all his IT people becoming more productive—solving problems faster—but that it wasn’t showing up in better numbers at the end of the month. Each person in IT was more productive, but they were going home earlier or playing video games. Those are real gains that need not increase GDP.

The other thing I would mention is the difference between the short and the long run. Imagine there are two sectors, and sector A gets fully automated—let’s say lawyers, even if that example is imperfect because lawyers have a lot of regulatory protection and there are many contexts where you are required to use one. Imagine we no longer need any lawyers and we solve our legal problems ourselves. All the people in sector A that gets automated need to move to sector B. All the demand that is now consumer surplus—money we no longer need to spend on legal problems—can go and be spent on the other sector. But that reallocation doesn’t happen instantly. The capital has to be moved, the labor has to be relocated, the demand has to be redirected. There is a moment when GDP could be dropping, because we are not consuming legal services or dishwasher repairs, and the transition to new consumption patterns hasn’t yet occurred. In the meantime, capital is being written down, labor is being relocated, and there may not be sufficient demand either. All of that transition could definitely look nothing like smooth, continuous growth.

Mounk: I’m trying to figure out what the aggregate effect of these changes might be. On one hand, agriculture—historically a huge part of human activity—mostly gets automated, the number of people working in agriculture is now astonishingly low, output goes up a lot, and as a result prices go down a lot. Most of the consumer surplus is captured by consumers, and so it is a very good thing.

What I don’t fully understand is what actually provides the basis of ordinary people’s bargaining power. In the agricultural world, the answer is that the production of agricultural goods becomes very cheap, but it turns out that humans are necessary for running all kinds of other elements of the economy. There is a strong demand for human labor, and that is what allows people to continue to consume.

Now, if we get to a world—and this still sounds a little like science fiction, but I am trying to imagine the scenario—where AI can fully run agriculture without any humans, and can fully run the systems needed to manage agriculture, and can fully run the law firms needed to efficiently allocate capital to agriculture and ensure the most efficient firms are tilling the most land, it may be that there are still elements of a human economy where human work is needed. It may be that humans still prefer human teachers, or that humans continue to be required in medical decisions—perhaps because we don’t trust AI systems to make them, or perhaps because of regulatory obstacles to fully automating those decisions.

Subscribe now

But if all of the underlying productive processes that actually generate material wealth no longer require humans, is there a kind of perpetual motion in the circular economy of humans that is enough to sustain affluence on its own? Or does there need to be some relation back to material production for the whole construct to sustain itself? If all of the demand for human labor is produced by the fact that it is extremely expensive to look after old people, by regulations that prevent us from building houses, by the willingness of capital owners to pay a lot for housing because they need somewhere to live, and by some people continuing to be employed in human-facing roles because of regulation—is that actually enough to sustain affluence for human workers if all of the genuinely productive processes can be done by non-human workers?

Garicano: Let me break this down into a few parts. First, the satiation case we are discussing—where the sector gets smaller as it becomes more efficient—doesn’t necessarily have to be the case. In fact, in many sectors, as technology gets better and things become more efficient, the sector actually grows in size. This is called the Jevons effect, after William Stanley Jevons, an English economist who observed that machines using coal were getting more and more efficient and yet were consuming more coal rather than less. Why? Because as they got better, they were being used for so many more things that total coal consumption was going up. In many sectors—think about health, think about energy—as things get more and more efficient, it is unlikely that the sector as a whole will shrink. In fact, it is more likely that it could grow in size and demand more humans. The sectors most likely to grow when prices go down—those with the most elastic demand—would be things like health and energy, just to give two simple examples.

The second important point is the idea of complements, which you were clearly hinting at in your question. There are many situations where a human is needed at a bottleneck. Even if the first 99 tasks can be automated, if the 100th task requires a human, the 99 automated tasks are abundant but the scarcity is still the human—and the human is going to capture the rent and the labor income.

Mounk: That depends on the human being scarce. If that task requires a very high level of qualification and you need millions of humans to do it because they are so productive, then a lot of people are going to be in relatively decent employment. But if you only need seven people to do it and they have to be excellent, then those seven people are going to capture huge rents—some of that economic gain is going to go to them, but only to them.

Consider that something like 5% of the male workforce globally is employed as drivers. The rent from the need for human drivers is very broadly distributed—each of those drivers is probably not very affluent, but the rent for that activity is widely shared. Now say that ten people have to supervise all of the self-driving cars, and they have to be incredibly qualified with very few people able to do it. Perhaps they capture a lot of that rent, but that is only going to be ten people who get that money. Or say it needs a thousand people, but a million people are able to do that job—in that case, the wage for those thousand people is going to be really low, because any one of them can be fired and there are 999,000 waiting outside the door willing to take their position. So it depends a lot on those kinds of details.

Garicano: I am writing a book on this point—it is called Messy Jobs. The argument of Messy Jobs is that there is a big difference between a task and a job.

Geoffrey Hinton, whom you have had on your podcast, is famous for having said in 2016 that nobody should study radiology because radiology was just an expert system that could scan photos—and of course any expert system was going to be better, trained on hundreds of millions of breast cancer scans and perfect at detecting those cancers. The truth of the matter is that demand for radiologists has never been higher. Their salaries are growing, their numbers are growing, and it is the third highest-paying medical profession in the United States. Why? Because the task is very different from the job. The technologist imagines a radiologist just looking at scans. But only 30% of a radiologist’s time is spent looking at scans—they have to develop the diagnosis plan, talk to colleagues, talk to patients, and do many other things.

The first crucial obstacle to your dystopia is that automating parts of jobs—tasks—is not automating the job. I invite all your listeners to think of what they did today and consider which of those things could be replaced by a machine. I went to a workshop, had a job market seminar, had a meeting with colleagues, had students walking in, worked on a paper—and if you think through how many of those tasks you could replace with a machine, you will discover that many of them can’t be. The task we are doing right now—having a human conversation about something—can’t be. A job and a task are really very different things. Many aspects of a job can change without the whole bundle disappearing. It will get re-bundled, it will look different, but it will not go away.

There are specific reasons for that. One is the need to direct the AI. You cannot just let it do its thing. The AI is sycophantic—it tends to agree with what you say. If you want to direct it to the left, it says yes, left is great. If you want to direct it to the right, it says yes, right is the best, you’re the smartest. What you tell it is going to matter, and that means somebody is going to have to exercise judgment. Crucially, this is not a problem that is solved by AI getting smarter and smarter. Think of managing a family—everything you do in the morning with the kids, moving around, deciding. A lot of that is not automatable because a lot of the knowledge of what is going on is tacit. It is in your head. No machine can tell you whether the kid has to wear these boots or whether today is the day they need this or that particular thing. Authority is inherently human. Making difficult decisions is inherently human.

Being the consultant who does PowerPoint presentations—yes, that can be automated. But does the consultant only do PowerPoints, or does he or she go to the company, listen to the workers, figure out where the problems are, and determine how to improve things? A lot of that is tacit. So I would push back against the idea that entire jobs are going to be done autonomously.

You are right that self-driving cars passed the autonomy threshold—the cars can essentially drive themselves, which means the supply of drivers suddenly becomes infinite and the wage floor collapses. That is a good example, but it is an example where the task is very clearly defined and very repetitive. Is that the majority of jobs? The claim of Messy Jobs is that if you think about demand elasticity, many sectors will grow; if you think about complementarities, there are going to be crucial scarcities that humans can exploit—and there will be many such scarcities. This is without even getting to the question of demand for human services, about which I am actually not sure. I am not sure that people, when they are old, will necessarily want a person bossing them around asking “are we well today?”—I might prefer a robot taking care of me. So it is not obvious that human-ness in itself is always the preferred option.

Mounk: Including a lot of the more intimate tasks involved in elder care—would you rather have another human wipe your ass, or would you rather have a machine do it? You certainly want some human company. Once your ass is wiped, you would love to have a conversation with a human.

I have a middle position in these debates, and I want to push back on a couple of the things you said—though I am not coming from a maximalist position. I agree that a lot of the predictions that all jobs are going to be gone in two years are testimony to people who haven’t thought carefully about politics or the real world. But some of the examples you gave leave me a little less convinced.

To give one example: can AI outsource the management of a family? Part of family life is that you are negotiating between human beings, trying to come up with a plan together. Even if the AI can make a plan that is Pareto superior to whatever plan you would have arrived at, part of what it is to be a family is to make those plans together—to decide what you are doing today, and so on. On an emotional level, you might not be able to outsource those things. On a purely planning level, though, I think AI absolutely could handle the tasks you described—and in fact, many feminists would say that is precisely what they have been arguing for for a very long time, because it is often women who do the emotional labor and the second shift: keeping track of the fact that Timmy has to go to the dentist tomorrow and Tammy has to go to ballet the day after, and whether the dress she needs for ballet has already been washed. It would require an invasion of privacy—an AI that is party to all of these conversations and immediately notes down when Tammy says, don’t forget I need X or Y for my ballet practice next week. But can AI do all of those things? Absolutely. Could it, in fact, save some marriages in the process? Probably yes.

Garicano: Here is why I disagree. There are information processing tasks—and you are right that a lot of information processing tasks can be automated. We synthesize information, put it in a form that can be processed, and make a decision. But there are other tasks that have nothing to do with information processing. Your wife or your kid is upset—someone needs to talk to the kid, someone needs to understand why he is upset, and someone needs to decide: yes, the optimal plan from the perspective of the family was that you couldn’t stay home, but I listened to you and I decided that you are staying.

A lot of it is not information processing. You understand your kids. You understand what a look means—from your wife, from somebody else. When a look means yes, I will do it. When they say yes but in fact they mean no. There is a lot of tacit, local knowledge that goes into management, into family life, and into business. We are not just talking about politics or emotions—we are talking about interpersonal knowledge. You have known your wife for many years and you know when you can push and when she knows she can push. You might say the machine could know those things—I honestly don’t think it could.

Take the contractor: you know which electrician is reliable and which one played tricks on you last time. Can the AI know whether you can use some piece of leverage to get that electrician to show up on time? We are talking about a level of interpersonal and tacit knowledge that is extraordinary—and also, think about this: a lot of the tacit knowledge within jobs is knowledge that employees have that gives them power. They are not going to be happily sharing it with AI. You should know that my colleague so-and-so has this problem with the boss and never wants to work with him—that kind of thing is going to remain in the heads of humans.

So yes, information processing tasks can and will be automated. But a lot of what remains has to do not just with emotional and social skills, but with tacit knowledge and personal knowledge that the machine will probably never fully gain, because it cannot capture it.

Mounk: I have two different lines of questioning about this. The first is: if we move away from the extreme predictions, and if we recognize that advanced AI tools are clearly capable of doing a lot of the tasks involved in knowledge production, that presumably means some jobs are going to go away. The idea that AI is incompetent, that it can’t do any of those things, that it’s all hype and a bubble—we agree that is wrong. But I think we also agree, at the other end, that real-world frictions are very real. Jobs are messy because the world is messy, and therefore the idea that the moment Claude beats doctors on a bunch of stylized medical questions—which it more or less does now—we should expect there to be no doctors tomorrow, is naive and doesn’t understand the real world.

But what happens in the middle space? What happens if the demand for white-collar work is suddenly reduced by 25% or perhaps 30%? It doesn’t have to happen between today and tomorrow—it happens over the course of 10 or 20 years. You just see a continuing, gradual reduction in the demand for that kind of high-skilled work: as existing firms automate work away, as firms that are too stubborn or unable to do that are outcompeted by new entrants that are AI-native—in the same way that in many areas of the economy, it took internet-native companies to outcompete old ones before you really saw productivity gains come online.

That is going to be a significant process, and it is not going to happen all at once. In a way, that raises an equally troubling possibility: that the job market is going to slowly slump for an extended period, and that we face the famous, somewhat apocryphal, boiling frog scenario. If everybody lost their job over the course of two months, perhaps we would all organize and demand some way of being made whole. But if this shows up as decades in which the bargaining power of ordinary people diminishes gradually—because the demand for human labor just continues to fall in a messy, haphazard way—that could still be an incredibly painful period ahead for ordinary people.

Garicano: You are more or less describing my scenario of the transition between sector A and sector B. We know that during the Industrial Revolution, what was called the Engels’ Pause—roughly between 1790 and 1840—this was happening: wages were stagnating or dropping and workers were in trouble. Then GDP roughly doubled over the following decades to 1900. So yes, it could happen that over a period of time the transition is hard.

I would think instead about the combination of factors working in the other direction. First, there are sectors where nothing is going to happen because they are outside the reach of this technology entirely. Second, in the sectors with elastic demand, there will be enormous growth—think of medical scanning. If AI handles all the routine scans, perhaps we would all be getting whole-body scans every year or every six months. The demand is extremely elastic and the sector could grow much, much larger, with radiologists needed to oversee far more machines. I think this is true for many sectors. Third, within the sectors that are getting automated, there are still messy jobs—humans directing, judging, making decisions, setting direction. If you count all of this together, you don’t have a catastrophe. You have a transition that is more significant in some subsectors and less significant in others.

We will also discover entirely new sectors—who would have predicted TikTokers and Instagramers? If you add up the sectors where nothing is happening—from public sector jobs to arts, music, barbers, hairdressers, cooks, and pet care, which alone accounts for roughly 1% of the U.S. population and is of course entirely unaffected by AI—and then add the sectors with very elastic demand that are going to grow, like health and energy, and then add the messy jobs where even though some tasks are being automated the jobs continue—from managers to entrepreneurs—and then add the complementarities, there is also David Autor’s idea of the new middle class: think of a nurse who is empowered with a genius in a box, who can now diagnose really complicated illnesses, hold the patient’s hand, do all the other parts of the job, and solve more problems than ever before. Of course, as you said, then maybe everybody wants to be a nurse, and we have to think about the supply of nursing and other skilled trades. But when you add all of this up, you move away from the feeling that there is a cataclysmic change ahead, and more toward the view that yes, this is automation, yes, it is going to be a bigger revolution than what we have seen in the last 50 or 60 years, perhaps more similar to the Industrial Revolution—but no, it is not going to cause widespread, long-term unemployment. We are going to see new jobs we wouldn’t even think of today, from TikTokers and Instagramers to dog walkers. Who would have told you that you were going to be a podcaster?

Mounk: I don’t know if the vision of the future is that humans are going to be fine because we’re still going to be TikTokers and Instagrammers and dog walkers.

Garicano: No, I was not saying that. I was saying that the pet care sector is 1% of the population. It’s nurses and people who take care of the pets and all these other things.

Mounk: Let me ask you about the dog walkers. One interesting thing that has happened over the last ten years—which just shows how epistemically modest we should be about all of this—is that I remember all of the conversations about drivers losing their jobs, and how that was somehow linked in the conversation about populism to why the Midwest went for Trump. The proposed solution was that they should all learn to code. Now it turns out that AI is really good at coding. Meanwhile, because of a set of technical issues that ended up being harder to solve for a while—though mostly solved now—Waymo is very efficient and much safer than human drivers, yet there are still significant regulatory obstacles. The number of rides Waymo is offering is going up exponentially, but it is still a very small share of the market and most human drivers are still fine. This is going to take longer to play out than many people think.

But we are now in a world in which coders appear to be losing their jobs—though I understand the economic data on that is mixed—and in which knowledge workers are seemingly about to lose their jobs, while all of the manual trades are assumed to be safe. The plumbers are going to be fine. The dog walkers are going to be fine. Well, I watched, as many others did, the quite remarkable display by Chinese robots at the annual Chinese state television gala. The progress in their dexterity from a year ago to today is just astonishing. The ability to combine the manual dexterity of these machines with visual processing and understanding of the world is advancing very quickly as well.

I am personally waiting for the ChatGPT 3.5 moment in robotics. I don’t think it will take very long for there to be some consumer product that is actually usable—we are getting close to that. The applications in the industrial sector are likely to increase as well. Again, I don’t think that is going to happen tomorrow, and it will take time to be fully implemented in the economy.

But when we are talking about a timescale of decades—when you say that in 20 or 30 years, more and more knowledge work tasks are going to be automated because those skills can already be performed by AI, and perhaps it will take a long time for firms to reorganize and for new entrants to come in, but that is okay because perhaps we will all be in the pet care sector—well, that assumes that in 20 or 30 years we will still not have figured out household assistance robots. That if you are out at the office or doing whatever you do during the day, you cannot have a little robot walking your dog in your stead. Given the rate of progress of this technology, that seems to me like a pretty significant background assumption to be making.

Garicano: Physical AI—robotics, let’s say—is not that far off. What we have seen in the past is that capital is in what we call elastic supply: you can always invest more in capital, which means the rents on capital eventually get competed away and the robot gets sold at a competitive price. That means people can use robots for care. Remember, we have significant fertility problems and population growth problems when it comes to paying our pensions, and having robots could be a solution to all of that—it is like having more population growth.

In a world where those returns are competed away, we are back to consumer gains. The capital doesn’t earn extraordinary returns because there is an infinitely elastic supply of capital—more people can invest in making more robots. What is the scarce resource? It is going to be land, it is going to be energy, and it is going to be whatever human labor is still needed. That human labor could mean we work fewer hours, that we are able to enjoy more leisure, or that human labor is employed in a whole range of jobs which, as you rightly say, we cannot anticipate.

What we should not imagine is an economy that works without humans, because all value is generated for humans. What does the economy generate value for if nobody is buying the products? Value, by definition, is something that is worth more to humans than it costs to make. If there is no human who can buy things because they are all poor, there is no value. The way the economy works is that the return to capital gets pushed back down to the competitive return, and the rents get captured by the scarce resources—in this case, the complementary human labor that is still needed.

Nobody can anticipate what happens in 30 years. Both physical robotics and cognitive AI are going to represent a major revolution. I don’t think we should be thinking of this as an apocalypse. There are a lot of complementarities, a lot of scarcities that still favor human labor, and a lot of areas where this doesn’t really bind at all.

Mounk: Tell me a little bit about the state of the empirical literature. I understand that there is a real distinction between micro and macro studies—a real distinction between studies that look at the extent to which particular tasks can be automated and studies that look at how the overall job picture has changed.

When I look at the fields I know a little bit, I worry that the absence of change so far is an indication of what is yet to come, rather than an indication that AI won’t have a big impact. You mentioned translation earlier. Another thing I have been thinking about is index-making in the publishing industry. In all of those fields, there has been basically no change in the economic flows—so far as I can tell, my next book is going to be translated by human translators. Well, perhaps they don’t actually do it and privately send it to Claude and capture the consumer surplus by going on a nice vacation while they pretend to be working on the book. But in terms of the actual economic flows, nothing has really changed, and I don’t know how long that is going to continue to be the case.

It is very sticky and very complicated to change those processes. Somebody needs to be willing to fire all the translators and deal with the backlash—the agent saying the author doesn’t like the idea of AI doing the work, the risk of a newspaper story, the possibility that customers will be upset. There are all kinds of reasons to be risk-averse about being the first mover to make that change.

What I will tell you is that one of the things I have built for myself with Claude Code is a personalized translation tool, because I publish my articles—including some podcast transcripts—not just in English but also in German and French. It is not just better than the off-the-shelf tools; at this point it is better than all but the very best translators I have worked with. The very best translators—particularly in France, for whom I am deeply grateful—I think are still better. But 90-plus percent of the professional translators I have dealt with, people who have translated famous books by famous authors, are now significantly outperformed by it.

For now, if economists tell me that translators haven’t lost their jobs and none of this has changed that much, I believe it—I can see that. But given that AI has existed in its current form for only about three years, and that for two of those three years it really wasn’t yet at the level it is reaching now, and that people have not yet integrated these tools sufficiently into their processes—I would say: come back to me in 15 years and let’s see whether those translators still have jobs in the way they do today.

Garicano: Nobody is predicting that translators will still exist in their current form indefinitely. I said “so far, so good”—but perhaps like the person falling past the window. Jobs do go away. Newspapers went digital, and there were lots of people working in printing presses, paper, and all the associated industries, and all of that was automated away.

Mounk: Including my grandfather, whose job it was to, as a young man, to lay the newspaper letter by letter. He helped to manage the printing side.

Garicano: This is human history all the time. On the question of empirical evidence: the evidence up to now is positive. When randomized controlled trials have been conducted—giving AI to workers in a controlled setting—the results are consistent. In customer support, the most junior agents achieved performance similar to more senior ones. In writing tasks, the worst writers achieved performance similar to the better writers. When it was given to software programmers across three different tasks, the less skilled programmers were brought closer to the level of the better ones. Micro studies seem to be finding complementarities rather than substitution, consistently.

At the aggregate level, there is much more confusion and much less clarity. We don’t see big drops in demand. There are some canaries—as I mentioned from that paper, Canaries in the Coal Mine—some preliminary evidence that there may be drops in junior roles. When we think about the research task, the PowerPoint task, the Excel task—those are the obvious things to automate—we have to imagine that junior lawyers, junior consultants, and junior investment bankers will not be recruited as much, because you can do a research task without a junior person. Yet it turns out the McKinsey class this year is bigger than before. They keep hiring. So far, so good.

I agree with you that this is not a forecast of the future—I don’t mean to say that because we haven’t seen much yet, we won’t. That is not the point. The point is that there are indications that complementarities are important, that people who use AI produce better work, and that substitution is still limited. It is hard not to think that tasks involving basic PowerPoint work and research are going to be fully automated at some point. But I agree—we should not try to make this a 15-year forecast.

In the rest of this conversation, Yascha and Luis share advice for young people at the start of their careers, why AI won’t kill off bullshit jobs, and whether companies run by AI would be more successful than those run by messy, emotional humans. This part of the conversation is reserved for paying subscribers…

Read more