In a 1995 episode of Seinfeld, an extremely Seinfeldian series of events leads Jerry to a problem: He has to take a lie-detector test. Specifically, he has to beat a lie-detector test. He seeks advice from his friend George Costanza, whose personal flaws render him uniquely suited to the task of polygraph cheating. George initially rejects the idea that conscience-free lying can be taught (“It’s like saying to Pavarotti, ‘Teach me to sing like you!’”). But he relents. “It’s not a lie,” George says, with a melodramatic flourish, “if you believe it.”
The joke, in the episode, is at George’s expense: Only in his upside-down world would sociopathy amount to a moral. These days, though, his advice might as well be political theory. To participate in American politics is to navigate, every day, an avalanche of falsehoods—lies issued, with Costanza-like ease, from the highest levels of power. Fact-checking was a theme of Donald Trump’s first presidency. Journalists kept count of those first-term fictions—30,573 in all, per one count—guided by the optimism that checking the president’s words might also serve as a check on his power. In late 2020, when Trump claimed victory in the presidential election he had lost, scholars saw in his declaration the kind of propaganda typically found only in authoritarian regimes. They gave the fiction an epithet befitting its magnitude: “the Big Lie.”
But that term, with its sense of emergency, has gone the way so many other fact-checks have in this age of heedless lying: It lives, now, in that democratically awkward space between accuracy and irrelevance. In 2024, Trump was reelected despite the Big Lie—and perhaps because of it. His false assertions are not liabilities, it seems, but rather selling points for many of his voters. They are weapons of partisan warfare, disorienting perceived enemies (Democrats, members of the media) even as they foment broader forms of cynicism and mistrust.
Though Trump’s second presidency began on January 20, 2025, the start of the new Trump era effectively began on January 7, 2025. That was the day that Meta, the parent company of Facebook and Instagram, announced it would be ending its efforts to fact-check claims made on those platforms. (It also happened to be exactly four years and one day after the attack on the Capitol that emerged from the Big Lie.)
Meta’s pronouncement was the company’s “Latest Bow to Trump,” as an Associated Press headline summed it up. It was also a harbinger of a wider kind of concession. Fact-checking, while increasing as a need over the first 100 days of the second Trump administration, is waning as an enterprise. The lies are winning. The president is wielding them ever more brazenly. George Costanza, for all his idiocy, may also be a savant: It’s not a lie if you believe it.
For decades, American politics have relied on the same logic that polygraph machines do: that liars will feel some level of shame when they tell their lies, and that the shame will manifest—the quickened heartbeat, the pang of guilt—in the body. But the body politic is cheating the test with alarming ease. Some Americans believe the lies. Others refuse to. Some Americans recognize the lies’ falsity but have decided that some things—their own tribe, their vision for the country—are simply more important than truth. Regardless, the lies remain, unchecked by the old machinery. The polygraph is a measure of conscience. So, in its way, is democracy.
A century ago, in his classic book Public Opinion, the journalist Walter Lippmann laid out a bleak argument: One of the threats to the American experiment was American democracy itself. The work of self-government, Lippmann thought—even back then—asked far too much of its citizens. It asked too much of our minds.
Democracy is a task of data management; ours is premised on the idea that voters’ political decisions will be based on reliable information. But it is also a matter of psychology, and of cognition. The atomic unit of democracy is the human brain. Everything will come down to its capabilities, its vulnerabilities, its biases—for better and, definitely, for worse.
Public Opinion considers mass media and propaganda, and the role that emotion plays in political life. Lippmann observed the importance of media inputs well before media was part of the American vernacular. The information people rely on to do the work of citizenship—voting, arguing, shaping a shared future—is data. But those data are processed by notoriously fickle hardware. The data inform our brains’ impressions of the world: the images that Lippmann called “the pictures in our heads.” The pictures are subjective. They are malleable. And, perhaps most of all, they make little distinction between things that are true and things that are merely believed to be.
Lippmann was writing in the 1920s, not only during the early age of radio but also during a smaller kind of communications revolution: penny presses; mass-produced illustrations and photographs; advertising. He was reckoning with the beginning stages of the information environment that humans navigate today. As people consumed these media, he discerned that they would become reliant on images of the world rather than on the evidence provided by the world itself. They would become confused, he feared, by the preponderance of competing images. And the confusion would weaken them—making them susceptible to the advertisements, to all the stories, to information overwhelm. (To describe the effect of the images, Lippmann borrowed a term from the factory floor: They functioned, he argued, as “stereotypes”—a term he used not as an insult but as a simple description of images’ heuristic powers.)
In Public Opinion, Lippmann diagnosed how readily propaganda could make its way into a nation that was officially at peace. He outlined how seamlessly the false messages could mingle with, and override, true ones. He argued that Americans’ unsteady relationship with information made our democracy inherently fragile. The philosopher John Dewey, alternately impressed and horrified by Public Opinion, called it “perhaps the most effective indictment of democracy” ever written.
Lippmann’s critiques of democracy have become only more relevant with age. The media environment of the 1920s already featured elements of information overload. The first months of the second Trump presidency, having brought a “flood the zone” approach to government, have lent new acuity to Lippmann’s warnings. The number of news stories alone has made it seem almost absurd to expect citizens to attempt the basic work of democracy: staying informed.
With every lie Trump has told, from the petty to the consequential, he has eroded people’s ability to trust the pictures in their heads. Every time he condemns the pollsters who document his waning public approval, he further erodes that trust. The tethers that anchor people to their president—and to the ground truths of their politics—fray just a bit more.
The people closest to Trump weaken those tethers as well. See, for example, the White House’s changing story about Kilmar Abrego Garcia, the Maryland man whom the administration forcibly deported to El Salvador in March. Administration officials initially called his deportation—effected without due process—an “administrative error.” Soon, though, and without providing credible evidence, White House Press Secretary Karoline Leavitt was accusing Abrego Garcia of being a gang member, a human trafficker, and a “foreign terrorist.” Earlier this month, after the Supreme Court ordered the administration to facilitate Abrego Garcia’s release from custody in El Salvador, the White House adviser Stephen Miller insisted that the administration had “won” the case, “clearly.”
It had not. The Court had rebuked the White House, unanimously. But political power can be narrative power as well. Falsehoods, issued repeatedly from the bully pulpit, threaten to become conventional wisdom, then clichés, then foregone conclusions. Attempts to challenge them, as crucial as those efforts are as matters of historical recordkeeping, take on a certain listlessness. For others to point out the truth is to do the right thing. It is also to bring paper straws to a gunfight.
In February, responding to Trump’s ask-neither-permission-nor-forgiveness approach to presidential power, the New York Times journalist Ezra Klein published an essay titled, simply, “Don’t Believe Him.” The president’s strategy, Klein argued, is to perform a level of power he doesn’t have in the hopes that the performance might become, eventually, reality. Trump “has always wanted to be king,” Klein wrote. “His plan this time is to first play king on TV. If we believe he is already king, we will be likelier to let him govern as a king.”
This is absolutely correct. It is also an encapsulation of the problem that Lippmann foresaw. The president, a creature—and in some sense a creation—of television, is keenly aware of the power of images. He avails himself of the insight that Lippmann had years before the TV would become a fact of many people’s lives. And Trump knows how much is at stake. The pictures we carry around with us, in our mind’s ever-revolving camera rolls, are much more than representations of the world as we understand it. The pictures are biases, too. They are assumptions and expectations. They are like brands, in their way: ever expandable, ever expendable. They can be shaped by lies as well as truths. Human brains have a hard time telling the difference.
Lippmann was a contemporary of Freud, whose nascent insights in psychology informed Lippmann’s theories of politics. Our minds make us what we are; they also make us, collectively, vulnerable to deceit. They are biased toward emotion over information. They tend to prefer the easy stories over the complicated ones. The pictures they hold might be informed by our interactions with physical reality, or by fantasy. Humans can try to separate the two—reality here, irreality there, stored in separate files—and can do so successfully. But the separation itself is work. And it is work made ever more taxing in a media environment where the human-generated lies mingle with the AI-generated ones, and where even the fact-checked news comes at people in endless feeds and floods.
In the flurry, people can lose control of the pictures in their heads. They can lose control of themselves. “For it is clear enough,” Lippmann wrote, “that under certain conditions men respond as powerfully to fictions as they do to realities, and that in many cases they help to create the very fictions to which they respond.”
Public Opinion—written not long after a once-in-a-century pandemic, and after the urgencies and contingencies of war had changed the way that Americans thought about truth itself—also contains insights from Lippmann’s experience with public relations. He had worked in the field, on behalf of the United States and its allies, during World War I. He had seen firsthand how easily information could be spun and edited and, all too often, simply manufactured out of thin air.
Modern advertising works, generally speaking, by creating problems rather than solving them. It manufactures desires among the public; it also manufactures, in the process, discontent. Our politics are doing the same work as they sell our nation back to us. Trump is, too. He manufactures problems—the “rigged” elections, the invasion of “illegals,” the “woke mob,” the horrors of “American carnage”—to sell us the solution: Trump himself.
The word propaganda, in Lippmann’s era, had not adopted the negative connotations it carries today. It was a term of politics borrowed from Catholic practice: Propaganda shared a root with propagation and suggested the straightforward act of sharing and spreading the faith. In the 1920s, it meant something akin to what today we might call straightforward “publicity.” But Lippmann’s studies of psychology had chastened him. Our minds, for all their attunement to the nuances of the physical world—the subtle shifts in light, the micro-expressions that move on the faces of other people—are not terribly adept at perceiving those distinctions through the filters of airwaves and screens.
On the contrary, all the inputs people encounter, by choice or by circumstance—the news reports, the novels, the films, the celebrities, the radio shows, the billboards, the histories, the satires, the amusements, the truths, the lies—tend to end up in the same place. The inputs influence, then continually edit, the pictures in our heads. Those pictures might be accurate appraisals. They might be delusions. They are nearly impossible to categorize. They are also totalizing. “Whatever we believe to be a true picture, we treat it as if it were the environment itself,” Lippmann observed. The insight might seem simple: Of course we believe what we see. But the opposite is true as well: We see what we believe.
For Lippmann, that meant that the information people rely on to form their mental images would be the lifeblood, or the death, of American democracy. Lippmann was a celebrated columnist, and he wrote Public Opinion as American newspapers were in the process of reforming. This was the era when objectivity, as a standard, was born. It was the age when reporters instituted standards of sourcing and validation. They were responding to the proliferation of information and misinformation, the advent of advertising, the establishment of public relations as a field and a career choice.
The papers were reacting to market pressures, essentially, by creating a new kind of commodity: information that had been collected, vetted, verified. This information would go out of its way to clarify what had been reported and what had been merely opined. The lines were not always entirely clear—but they were efforts to impose new modes of order.
Now, though, the lies are imposing the order. TV-news organizations, hosting candidate debates during the presidential election last year, deliberated over whether moderators should correct inaccuracies uttered on their air. Confusion on the matter led J. D. Vance, facing Tim Walz in a vice-presidential debate, to make his infamous complaint: “The rules were that you guys weren’t going to fact-check.” (He was reacting to CBS moderators’ efforts to clarify that many of the Haitian immigrants of Springfield, Ohio, whom Vance had previously referred to as “illegal,” indeed had legal status in the United States.)
Challenges to the rules have expanded far beyond the format of the televised debate. In early April, President Trump issued an executive order condemning Chris Krebs—who in Trump’s first term headed the Cybersecurity and Infrastructure Security Agency—on the grounds that Krebs had said that the 2020 presidential election was free and fair. The order is so blatant in its attempt to rewrite history that to call it Orwellian would be something of an insult to Big Brother. But it is Costanzan. It is authoritarian. What Donald Trump believes, the order suggests, becomes the truth. “When the president does it … that means that it is not illegal,” Richard Nixon claimed, a few years after his association with illegal activity ended his presidency. That tautology, in the age of Trump, is now a matter of judicial precedent. It is also the defining logic of Trump’s attempts to expand executive authority. “I’m a very honest person, and I believe it with all my heart,” Trump said recently in an interview with my colleagues Ashley Parker, Michael Scherer, and Jeffrey Goldberg. He was responding to their questions about why Trump continues to insist, falsely, that he won the 2020 election. “I believe it with fact—you know, more important than heart,” Trump said. “I believe it with fact.”
This is the Costanza principle at work. “Because I believe it” is neither a factual argument nor a legal one. But Trump is treating it as both. He is treating his preferred reality as the only one that can exist. He is behaving, in that respect, less like a president than like a king. He is acting as the kind of demagogue whom James Madison and other Founders feared when they warned about “the passions” and their corrosive effect on politics. Passions, for those men of the Enlightenment, fought against reason. They weakened people’s defenses against the seductions of emotional appeals. They could make people unable to tell the difference between the convenient story and the true one. “Facts don’t care about your feelings,” as the conservative commentator Ben Shapiro put it, is a good slogan, but it gets things wrong: The guiding principle of Trumpism is “Feelings don’t care about your facts.”
And once facts are discarded, anything can come in their place. Trump’s lack of accountability for his lies has expanded, in these early days of his second term, into a more comprehensive form of unchecked power. “The first time, I had two things to do—run the country and survive,” he told Parker and Scherer. “The second time, I run the country and the world.”
If Can he say that? was a broad theme of Trump’s first term, Can he do that? is the even graver theme of his second. The deportations. The tariffs. The dismantling of the civil service, of scientific research, of government records, of civil rights, of voting rights, of basic standards of due process: The president’s efforts to destabilize his own government from within—to defund agencies, to “purge” the civil service of people he views as insufficiently loyal—have not merely been escalations of the attempted power grabs he made in his first term. They have been direct assaults on the delicate balance of power: an executive laying siege to the legislative and judicial branches. Lippmann did not predict this turn of events, but he understood their consequences. Democracy, under the sway of lies, becomes a form of anarchy.