What’s a Fact, Anyway?
The Weekend EssayJournalists put more stress on accuracy than ever before. The problem is, accuracy is a slippery idea.Illustration by Ricardo TomasEvery tribe has its myths, and journalists are no exception. In America, one common story goes like this: once, in the prelapsarian era before social media—or before smartphones, or the Internet—there was a time when journalists were trusted. Back then, everybody read muscular daily newspapers and watched straight-down-the-line TV reporting. When citizens had to make political decisions, a robust social contract with the media insured that they were well informed; even if they couldn’t always agree on what to do or whom to vote for, they could rely on a shared set of facts. But then something changed: people stopped paying attention to the news, or decided that they didn’t believe it anymore. They got distracted by podcasts, Facebook, and Twitch. They became ill-informed, and started to act against their best interests. The media decayed and fragmented, along with the nation. Opinion and news became indistinguishable, misinformation ran amok, and that is how we came to live in the post-truth world.This story is not unsupported. Trust in many institutions has fallen over the years, but in journalism it has plummeted. In 1972, Gallup began asking people in the United States, “How much trust and confidence do you have in the mass media?” The numbers used to be highly favorable—in 1976, more than seventy per cent said they had a lot, and very few reported having none—but in 2024 a plurality of respondents (thirty-six per cent) said that they had “none at all.” A survey of how highly Americans rated the ethics of various professions found that only a fifth believed journalists’ standards were “high” or “very high”: a better ranking than car salespeople and senators, but worse than bankers and chiropractors. (Almost half said that journalists’ standards were “low” or “very low.”) Increasing numbers of Americans report not following the news, or doing so via factually unreliable social or alternative media. In some parts of the world, the picture looks rosier—the Nordic states, outliers as ever, have far higher levels of public trust in journalsim—but one recent study of twenty-eight countries showed the balance of media trust and distrust to be neutral at best, and negative among the most developed nations.Meanwhile, large swaths of the industry have been on the retreat, racked by dwindling advertising revenue, hostile governments, and declining audiences. In the United States, the number of people employed in newsrooms dropped by more than a quarter between 2008 and 2020; thousands were laid off in 2024 alone. In 2022-23, the Committee to Protect Journalists recorded more journalists jailed worldwide than ever before. Technological change—most recently and acutely the mainstreaming of generative artificial intelligence—presents its own challenges, as does political polarization. In the United States, for example, the erosion of trust in journalism has been sharper and more consistent among Republican voters, and it can seem, in both parties, that many people turn to the news more for affirmation than for information. What passes for truth often depends on where one falls on the political spectrum: it’s no coincidence that Donald Trump’s venture into social media has “truth” in its name.This is a grim picture. But buried in the statistics is another story, one in which people worry about misinformation, even if they can’t agree on what to call it. “Fake news” is not a new concept, but many Americans now register displeasure with inaccurate or unverified information on social media, and a majority now think that somebody, perhaps even the federal government, should do something about it. There seems to be widespread recognition that bad facts are bad news—globally, fears of an “information war” are rising—and, despite endemic skepticism and distraction, there is an enduring thirst for reliable information. The question is, where can it be found, and how can its purveyors make themselves heard amid the noise?One answer has been to turn up the volume: to loudly pronounce that, in a fallen world, only we can provide accuracy. Networks such as NBC and the BBC have launched units dedicated to fact-checking what other people say—as opposed to checking their own work—and for the first Presidential debate of 2024 the New York Times tasked twenty-nine staffers with combing through the candidates’ statements in real time. Not every entrant in the accuracy economy is as large or well resourced: in the past decade, hundreds of small fact-checking Web sites have sprung up around the world, including many in countries, such as India, where press freedom is far from a guarantee. (Many of these sites received significant funding from Meta’s third-party fact-checking program, the end of which the company announced earlier this week.)Whether rooted in service or business strategy, this
Every tribe has its myths, and journalists are no exception. In America, one common story goes like this: once, in the prelapsarian era before social media—or before smartphones, or the Internet—there was a time when journalists were trusted. Back then, everybody read muscular daily newspapers and watched straight-down-the-line TV reporting. When citizens had to make political decisions, a robust social contract with the media insured that they were well informed; even if they couldn’t always agree on what to do or whom to vote for, they could rely on a shared set of facts. But then something changed: people stopped paying attention to the news, or decided that they didn’t believe it anymore. They got distracted by podcasts, Facebook, and Twitch. They became ill-informed, and started to act against their best interests. The media decayed and fragmented, along with the nation. Opinion and news became indistinguishable, misinformation ran amok, and that is how we came to live in the post-truth world.
This story is not unsupported. Trust in many institutions has fallen over the years, but in journalism it has plummeted. In 1972, Gallup began asking people in the United States, “How much trust and confidence do you have in the mass media?” The numbers used to be highly favorable—in 1976, more than seventy per cent said they had a lot, and very few reported having none—but in 2024 a plurality of respondents (thirty-six per cent) said that they had “none at all.” A survey of how highly Americans rated the ethics of various professions found that only a fifth believed journalists’ standards were “high” or “very high”: a better ranking than car salespeople and senators, but worse than bankers and chiropractors. (Almost half said that journalists’ standards were “low” or “very low.”) Increasing numbers of Americans report not following the news, or doing so via factually unreliable social or alternative media. In some parts of the world, the picture looks rosier—the Nordic states, outliers as ever, have far higher levels of public trust in journalsim—but one recent study of twenty-eight countries showed the balance of media trust and distrust to be neutral at best, and negative among the most developed nations.
Meanwhile, large swaths of the industry have been on the retreat, racked by dwindling advertising revenue, hostile governments, and declining audiences. In the United States, the number of people employed in newsrooms dropped by more than a quarter between 2008 and 2020; thousands were laid off in 2024 alone. In 2022-23, the Committee to Protect Journalists recorded more journalists jailed worldwide than ever before. Technological change—most recently and acutely the mainstreaming of generative artificial intelligence—presents its own challenges, as does political polarization. In the United States, for example, the erosion of trust in journalism has been sharper and more consistent among Republican voters, and it can seem, in both parties, that many people turn to the news more for affirmation than for information. What passes for truth often depends on where one falls on the political spectrum: it’s no coincidence that Donald Trump’s venture into social media has “truth” in its name.
This is a grim picture. But buried in the statistics is another story, one in which people worry about misinformation, even if they can’t agree on what to call it. “Fake news” is not a new concept, but many Americans now register displeasure with inaccurate or unverified information on social media, and a majority now think that somebody, perhaps even the federal government, should do something about it. There seems to be widespread recognition that bad facts are bad news—globally, fears of an “information war” are rising—and, despite endemic skepticism and distraction, there is an enduring thirst for reliable information. The question is, where can it be found, and how can its purveyors make themselves heard amid the noise?
One answer has been to turn up the volume: to loudly pronounce that, in a fallen world, only we can provide accuracy. Networks such as NBC and the BBC have launched units dedicated to fact-checking what other people say—as opposed to checking their own work—and for the first Presidential debate of 2024 the New York Times tasked twenty-nine staffers with combing through the candidates’ statements in real time. Not every entrant in the accuracy economy is as large or well resourced: in the past decade, hundreds of small fact-checking Web sites have sprung up around the world, including many in countries, such as India, where press freedom is far from a guarantee. (Many of these sites received significant funding from Meta’s third-party fact-checking program, the end of which the company announced earlier this week.)
Whether rooted in service or business strategy, this sort of “political fact checking,” which spotlights specific claims and seeks to confirm or disprove them, is hardly neutral. Setting out to counter inaccuracies, it often takes the form of rebuttal, and necessarily entails editorial decisions about what to cover. (No outfit can scrutinize every statement by every politician, and none of them wants to.) An insistent focus on pointing out misinformation may even inflate the scale of the problem—to the benefit of what the writer Joseph Bernstein has labelled Big Disinfo, and to the detriment of a publication’s appeal to uncommitted readers. The sense that journalists are worth listening to has as much to do with how they go about their work, and with the uses to which readers can put it, as with what they actually say. The provision of facts does not, in itself, engender trust.
What is more certain is that, from time to time, every journalist, no matter how well meaning, gets something wrong, or misses the point. This is true even if—as Michael Schudson, a professor of journalism at Columbia University, has argued—American journalism is deeper, more analytical, and more investigative today than it was fifty years ago. If there ever was a golden age of journalism, we may be living in it. The trouble is, pointing out the mistakes of others is not enough. If people are to trust journalists, we need to earn it.
The New Yorker is known for its accuracy. At least, that’s what the magazine tells itself. In reality, it depends on whom you ask. A prospectus announcing the first issue of The New Yorker, in 1925, promised that:
True to this mandate, The New Yorker maintains an obsessive interest in facts, and it didn’t take long for the early editors to recognize the usefulness of at least basic stenography. In 1927, the magazine published a Profile of the poet Edna St. Vincent Millay, “America’s first starlet.” Its first sentence—“Edna Millay’s father was a stevedore on the wharves at Rockland, Maine”—sounded good, but, like plenty more in the piece, it wasn’t true: Millay’s father was a schoolteacher turned insurance agent. The embarrassment inspired The New Yorker’s founding editor, Harold Ross, to install a fact-checking department, and since then checkers have left their marks all over the magazine. John McPhee, in an essay published in 2009, depicted fact checking as a pursuit bordering on mania. To shore up one important but rickety anecdote of his concerning wartime nuclear reactors, a checker spent days placing calls that “ricocheted all over the United States: from Brookhaven to Bethesda, from La Jolla to Los Alamos.” The chase culminated, moments before the press deadline, in a call to a Florida police department, which the checker enlisted to track down a crucial witness who turned out to have gone to the mall. He called the checker from a phone booth, just in time to correct an error.
Not many pieces require such heroics: the reality is that fact checkers are busy people, who traffic only occasionally in the dark arts of deep research. Most facts can be checked fairly easily today, especially with the benefit of the Internet, but, since there are so many, a checker has to prioritize. (Merriam-Webster defines a fact as “a piece of information presented as having objective reality”; a long piece might contain thousands.) When a particular fact turns out to be sticky, persistence and attention, rather than any kind of special knowledge, are generally what’s needed. Checkers are not infallible, and their successes are mostly due to hard work and creativity. What is truly extraordinary about any fact-checking department is that it exists at all. At the time of writing, nearly thirty people work in Checking at The New Yorker, almost all of them full time. It is labor, at scale, that produces accuracy.
But the pursuit of accuracy—that is, confirming a fact is a fact—is only part of the project. With a few pragmatic exceptions, fact checkers go over everything published by the magazine, editing for balance, fairness, and context. Writers are asked to share their sourcing; checkers review both research materials (books, articles, e-mails, documents) and original reporting. For complex topics, and as needed for corroboration, they do their own independent research, perhaps conferring with relevant experts. And, unless there’s a good reason not to, they get in touch with every person and entity discussed in the story, and comb through what’s attributed to or said about them.
This fulfills a few functions. If the person coöperates, as they typically do, the checker can verify factual details, including those which appear within quotations. If they disagree with something, they get a chance to comment or correct the record. And if they are angry, well, it’s better to know before publication. (This step is also important in libel-proofing a piece: under U.S. law, a good-faith and thorough effort to discover the truth is a solid defense should you accidentally publish an actionable inaccuracy. No matter how solid your standards are, it doesn’t look great if you didn’t bother trying to contact the person involved.) Checkers may discover hidden holes in a story, or garner perspectives that ought to be included. Most important, perhaps, this process, which is a kind of re-reporting, gives a checker an opportunity to test the basis of each statement of fact and opinion: is the writer relying on personal experience, the pages of a half-remembered book, a thirdhand rumor, or years of study? The value of a source, whether a person or a document, comes down mostly to quality of information.
Depending on the stakes—are we talking about what I ate for breakfast, or the location at which I saw the crime occur?—and on what other sources are available, a checker may then need to seek corroboration or to do some reporting of their own. Finally, based on all this, they can decide what should be said, and how: after a fashion, they decide what the magazine believes to be true.
Truth, in this sense, is sought and rendered, something produced by rigorous inquiry and informed consideration rather than something discovered. Many points, tweaked as necessary, can be stated as fact, but often a checker will want to insert a citation (“according to . . .”) or present conflicting accounts. Tone matters, and can be used to indicate the level of confidence: a fairly made assertion may be one that leaves room for doubt. Falsehoods that reveal something about the speaker may have enough value to keep, so long as they are rebutted, but detail that is truly unverifiable, or clearly and unfixably specious, will typically be cut: deciding what not to publish can be just as important as deciding what to keep in. In the landscape of a story, fact checkers’ trails can often be followed along a path of hedges and denials, but an elegant fix may render the footwork behind it invisible. Dispiritingly, checking tends to be at its most visible when it goes wrong, as when, despite Der Spiegel’s dozens-strong research department, a star reporter there turned out a few years ago to have fabricated many of his stories. (Among other things, he pretended to have interviewed Colin Kaepernick’s parents, and described the administrator of Fergus Falls, a Minnesota town, as a lonely singleton who’d never been to the ocean. Residents responded with a photograph of the administrator and his girlfriend embracing on a beach.)
Because The New Yorker, like many publications, trades on its reputation for accuracy, readers can easily feel cheated, even betrayed, when an error slips in. (Letter writers’ tendency to complain about failures of “The New Yorker’s vaunted fact-checking department” is surely a form of flattery.) This is not simply a commercial problem. “Trust,” as the philosopher Annette Baier once wrote, “is a fragile plant,” one that can easily be damaged by the kind of increased scrutiny or skepticism that a flaw, once noticed, tends to induce. After the roots have withered, they can be hard to restore.
Trust is something that journalists ask for. It’s also something that they give. Just as readers learn about the world from the news, so journalists learn from their sources. Another philosopher, Neil Levy, points out that “much—perhaps most—of what we know, we know on the basis of testimony, and (except in some unusual cases) we acquire knowledge through testimony only when we trust speakers.” Most information comes to us from other people, whom we choose, deliberately or not, to believe. As journalists, how and why we make those choices is particularly important. In the same essay, Levy writes that “trust makes us vulnerable,” and if this is the case for a writer it’s doubly so for the reader. No reporter can avoid that vulnerability, but mediating it is essential to earning and keeping the trust of a reader. Journalists choose whom to trust on their readers’ behalf.
For journalists, as for anyone, there are certain shortcuts to trustworthiness, including reputation, expertise, and transparency—the sharing of sources, for example, or the prompt correction of errors. Some of these shortcuts are more perilous than others. Various outfits, positioning themselves as neutral guides to the marketplace of ideas, now tout evaluations of news organizations’ trustworthiness, but relying on these requires trusting in the quality and objectivity of the evaluation. Official data is often taken at face value, but numbers can conceal motives: think of the dispute over how to count casualties in recent conflicts. Governments, meanwhile, may use their powers over information to suppress unfavorable narratives: laws originally aimed at misinformation, many enacted during the COVID-19 pandemic, can hinder free expression. The spectre of this phenomenon is fuelling a growing backlash in America and elsewhere.
Although some categories of information may come to be considered inherently trustworthy, these, too, are in flux. For decades, the technical difficulty of editing photographs and videos allowed them to be treated, by most people, as essentially incontrovertible. With the advent of A.I.-based editing software, footage and imagery have swiftly become much harder to credit. Similar tools are already used to spoof voices based on only seconds of recorded audio. For anyone, this might manifest in scams (your grandmother calls, but it’s not Grandma on the other end), but for a journalist it also puts source calls into question. Technologies of deception tend to be accompanied by ones of detection or verification—a battery of companies, for example, already promise that they can spot A.I.-manipulated imagery—but they’re often locked in an arms race, and they never achieve total accuracy. Though chatbots and A.I.-enabled search engines promise to help us with research (when a colleague “interviewed” ChatGPT, it told him, “I aim to provide information that is as neutral and unbiased as possible”), their inability to provide sourcing, and their tendency to hallucinate, looks more like a shortcut to nowhere, at least for now. The resulting problems extend far beyond media: election campaigns, in which subtle impressions can lead to big differences in voting behavior, feel increasingly vulnerable to deepfakes and other manipulations by inscrutable algorithms. Like everyone else, journalists have only just begun to grapple with the implications.
In such circumstances, it becomes difficult to know what is true, and, consequently, to make decisions. Good journalism offers a way through, but only if readers are willing to follow: trust and naïveté can feel uncomfortably close. Gaining and holding that trust is hard. But failure—the end point of the story of generational decay, of gold exchanged for dross—is not inevitable. Fact checking of the sort practiced at The New Yorker is highly specific and resource-intensive, and it’s only one potential solution. But any solution must acknowledge the messiness of truth, the requirements of attention, the way we squint to see more clearly. It must tell you to say what you mean, and know that you mean it. ♦
This is drawn from “Age of Doubt: Building Trust in a World of Misinformation.”
