Russia is again peddling misinformation in an attempt to upend our election. It’s an important news story, but there’s a larger threat that’s getting less coverage. In a New York University Stern School study, Paul Barrett shows that Russia’s contribution to our misinformation problem is nearly a distraction compared with what Americans themselves are providing. An Oxford University study of Facebook and Twitter shares during the 2018 midterm election reached the same conclusion, finding that a full fourth of the messages contained deceptive or inaccurate information — most of it homegrown.
Judging from what’s reported in the news, social media and far-right media outlets are the primary sources of domestic misinformation. And in fact, social media and far-right outlets are hotbeds of misinformation.
Nevertheless, there is a major source of misinformation that’s rarely mentioned in news stories — the news media themselves. A Columbia University’s Tow Center for Digital Journalism study found that “news organizations play a major role in propagating hoaxes, false claims, questionable rumors, and dubious viral content.” In another study, Harvard’s Yochai Benkler, Robert Faris, and Hal Roberts examined four million online messages transmitted or shared during the 2016 election campaign, including those from the websites of traditional news outlets like the New York Times and fake news sites like those of Russian operatives. The research team expected the Russian effort to be a major source of disinformation, and it was. But they discovered that traditional media were a larger contributor.
How could this possibly be true? How could America’s news outlets, which claim to be guardians of the truth, be a prime source of untruth?
The answer lies in journalists’ routines and their need to attract an audience. Donald Trump’s tweets are a case in point. Many of them include a false claim, which is typically said to be a problem that Twitter could solve if it deleted them. But the nature of the problem looks different when exposure to his tweets is examined. Direct exposure to Trump’s twitter feed accounts for a small portion of the total exposure to the content of his tweets. Americans typically learn of his tweets through the news media. Moreover, as a Media Matters study found, two-thirds of the media’s retweets of Trump’s false claims are passed along without noting that the claim is false. If Trump is patient zero, the news media are the superspreader.
News outlets defend the reporting of newsmakers’ deceptive claims by saying that their statements are “newsworthy.” It’s an argument not unlike that of social media platforms who cite “free speech” as justification for allowing false statements to stand. Such arguments ignore the way that our minds work. We tend to accept claims that conform to what we’d like to believe, whether true or not. And when we encounter such claims, we tend to embellish them, hold them in memory, share them with others.
The problem defies easy solution. The notion that social media platforms can thoroughly police their space ignores the enormity of that space, the gray line between fact and fiction, the value of unbridled free expression in a democratic society, the high error rate of algorithmic models, and more. News outlets are also not positioned to be an ironclad firewall against the spread of misinformation. Although news outlets are gatekeepers, they’re not positioned to act as moral police. If they were to attempt it, charges of media bias would rise to a steady roar and trust in the press would plummet. Yet, news outlets don’t have to be super spreaders. Many of Trump’s tweets, for example, don’t meet any defensible standard of newsworthiness. Titillating yes. Newsworthy no. And the fact that other news outlets are reporting a juicy Trump tweet doesn’t make it any more newsworthy.
Fact-checking will not compensate for lax journalistic standards. There’s no evidence to support the notion that news outlets’ fact checking puts an appreciable dent in our misinformation problem. The misinformed rarely fact check their beliefs and, when they do, are seldom persuaded that they’re wrong. The only effective way for news outlets to fight misinformation is to reduce their contribution. In a report commissioned by the New America Foundation, political scientists Brendan Nyhan and Jason Reifler proposed a set of reporting guidelines that would have that effect:
- Getting the story right the first time
- Minimizing repetition of false claims
- Making rapid corrections
- Relying on credible sources/Ignoring unreliable ones
- Not amplifying the problem
Election Day is a week away. Americans will be awash in misinformation in the closing days. News outlets can’t possibly fix that problem, but they can limit their contribution. A case in point is Joe Biden’s debate remark on the oil industry. It’s clear from Biden’s position papers and what he’s said on the campaign trail that he’s proposing a transition to clean energy rather than oil’s elimination. News stories that suggest otherwise confuse or mislead voters. The problem is endemic in campaign reporting, one that Kathleen Hall Jamieson describes as treating an atypical act as if it represents the whole. It highlights the conflict that journalists prize in their stories but confounds voters’ understanding of the choices they face.
Thomas E. Patterson is Bradlee Professor of Government & the Press at Harvard’s Kennedy School and author of the recently published Is the Republican Party Destroying Itself? Journalist’s Resource plans to post a new installment of his Election Beat 2020 series every week leading up to the 2020 U.S. election. Patterson can be contacted at email@example.com.
Yochai Benkler, Robert Faris, and Hal Roberts. Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics. New York: Oxford University Press, 2018.
Brendan Nyhan and Jason Reifler, “Misinformation and Fact-checking: Research Findings from Social Science,” New America Foundation, February, 2012.
Nahema Marchal, Lisa-Maria Neudert, Bence Kollanyi, and Philip N. Howard, Oxford University study of Facebook and Twitter shares during the 2018 midterm election Comprop Data Memo, Oxford University, November 1, 2018.
Daniel Kahneman, Thinking Fast and Slow, New York: Farrar, Straus and Giroux, 2011.