How to tell fact from fiction and trust the news again

by Kim Lisagor Bisheff Illustrations by Joan Wong

Journalist Dan Mac Guill was working at his home office in Maryland last August when he got a news tip from a colleague: A photo of a Democratic congresswoman was circulating on Twitter. It appeared to show her at a press conference amid a group of armed terrorists. She was smiling.

The Twitter replies ranged from skepticism (“This is verifiable as a real photograph?”) to condemnation (“The enemy is here”) to something in between (“I blew it up. … If it is photoshop they did an amazing job”). Many comments were too hate-filled to bear repeating.

The reactions caught Mac Guill’s attention right away. “If you see people who seem to genuinely believe that a sitting member of Congress is or has been a terrorist, then that’s worth pursuing,” he says.

Mac Guill, who works for the fact-checking website Snopes, suspected that this was yet another digital misinformation attack against U.S. Rep. Ilhan Omar of Minnesota, who has been a frequent target of online trolls since she became one of the first two Muslim women elected to Congress, in 2018. Just the previous week, in fact, Snopes had debunked a photo caption that falsely claimed that Omar had attended a “jihad academy.” The photo, which appears to show a woman in a headscarf holding a rifle, was taken before Omar was born. But that didn’t stop it from gaining traction on social media.

Though Mac Guill was pretty sure the newer image was also a fake, he knew it would require research to settle the matter. “You can’t always make the assumption that what’s obvious to you is obvious to everybody else,” he says. “Especially if people have certain biases that they might not even be conscious of, they might look at that image and say, ‘Well, look at it; it’s clearly her, and she’s been caught.’ And then you have somebody else saying, ‘She’s a sitting member of Congress. There’s no way this is real.’ People approach this content from different starting points.”

So he got to work.

Ideas and memes like these can go viral very quickly, exacerbating the ideological divide between groups with opposing political viewpoints. As Republicans and Democrats increasingly consume news from partisan sources, an individual’s political affiliation has become a strong indicator of whom they trust and what information they identify as factual.

Rotarians strive to abide by The Four-Way Test. So when we read something inflammatory, what guides our decision to believe it? Do we trust what we read because it is the truth? Because it’s fair to all concerned? Or because it validates our existing worldview? Rotarians have an obligation to set aside partisan assumptions in pursuit of truth and fairness. A good start would be to acknowledge that we are all susceptible to misinformation. (In fact, studies have shown that the older we are, the more likely we are to be duped.) And we can choose to start listening to the experts who have been trying for decades to help us sort manipulation from satire, opinion from fact, and fiction from truth.

The history of debunking misinformation far predates this political era. Snopes has been at it for 25 years, since long before “fake news” was on the public’s radar. CEO David Mikkelson launched the website in 1995 to tackle urban legends. Some of those early myths seem harmless today — like the one about the Poltergeist curse, which claimed that several of the 1982 horror movie’s cast members had since died under suspicious circumstances, or the one that correlated Super Bowl wins with stock market performance. The intensity and frequency of misinformation spiked after 9/11, when the internet, which was itself just taking off, became flooded with conspiracies and hoaxes, and fact-checking became an increasingly serious endeavor.

The next big bump came with the rise of social media. Facebook and Twitter enabled fake news to travel farther and faster, and fact-checkers struggled to keep up. Over the years, Snopes has been joined by new fact-checking organizations, including FactCheck.org, PolitiFact, and similar endeavors worldwide.

In the months leading up to the 2016 U.S. presidential election, the misinformation circulating on social media had become intensely political and polarized. People on both sides of the political aisle had honed their social feeds to match their existing biases, and in doing so, they became prime targets for made-up posts that aimed to validate and reinforce those views.

As journalists and academics began researching the phenomena that contributed to the spread of false information through social networks, stories emerged about Russian misinformation factories where hired trolls used fake social media identities to spread lies online. Reporters found hundreds of self-proclaimed “news” websites, based in the United States and abroad, that were deliberately publishing and spreading phony stories. The search term “fake news” started trending on Google. It has been a hot topic ever since — thanks in part to the fact that it is now often deployed to describe news someone doesn’t like, rather than stories that are objectively not true.

Snopes is busy these days. The site now has a staff of 15, most of whom are experienced journalists, working in home offices spread across three U.S. time zones. They keep regular business hours and communicate virtually via Slack throughout the day. Because they understand the importance of transparency in establishing readers’ trust, they are open about their operations and editorial process.

The “Transparency” page on the Snopes website details that process, along with the organization’s standards for sources. “We attempt to use non-partisan information and data sources (e.g., peer-reviewed journals, government agency statistics) as much as possible, and to alert readers that information and data from sources such as political advocacy organizations and partisan think tanks should be regarded with skepticism,” it says. “Any published sources (both paper and digital) that we quote, link to, use as background information for, or otherwise reference in our fact checks are listed in the Sources section at the foot of each fact check article.”

Such transparency is consistent with the code of principles established by the International Fact-Checking Network, which maintains a list of 29 organizations that are in compliance. That list includes Snopes, whose website says it follows the network’s principles “because we think being transparent with readers is the coolest.”

When fact-checkers come across a suspicious photograph like the one of Omar, Mac Guill says, their first move is to take a step back and get an overview of the claim. “What exactly is the question that we are being asked?” he says. Is it: “Is this a real photograph? Does it show what it appears to show? What exactly does the image consist of? What do I actually need in order to come to a conclusion?”

Glancing at the photo, he noted that Omar was the only one smiling. “Without any fact-checking expertise, you can see that Omar is the only person in the room who is grinning ear to ear and appears to be very happy, whereas everyone else is looking very solemn or has their faces covered,” he says. “That is very clearly out of place. That doesn’t mean that it’s a fake, but it’s a clue.”

One of Mac Guill’s editors took a screenshot of the image and used Google to do a reverse image search. That turned up a photo of Omar taken by an Associated Press photographer in Washington, D.C., as she was walking to a meeting in the Capitol on 15 November 2018. Omar’s head and facial expression were a perfect match. “That gave me a bit of a head start,” Mac Guill says. “It made it clear to me that this image consists of two separate photographs, at least, sewn together using software.”

To establish the truth about the image, Mac Guill needed to find both originals, identify their sources, and gather enough information to put them into context. He took another screenshot of the suspicious photograph and did his own Google reverse image search. It didn’t take him long to find various images from a news conference with the same men sitting at the same table — without Omar. “You can fairly safely say at that stage, this is fairly solid evidence that her face was digitally added and superimposed on the original photograph, and it’s a fake.”

To eliminate all doubt, he tracked the source image to the websites where it had been published, and he quickly figured out that the original was a Reuters photo from a 2008 press conference. A person whose head was almost completely obscured by a headscarf sat in the position where Omar’s face had been superimposed. “So there you’ve got it,” Mac Guill says.

As fact-check detective work goes, this case was pretty straightforward, Mac Guill says. “Sometimes image searches can get complicated,” he says. If a suspicious image was a still shot taken from a video, for example, it can take hours to uncover the original source. “I personally really enjoy that part of it. There’s a sense of accomplishment when you’re able to trace something back to its origins.”

The manipulated Omar photo is an example of what experts call “fauxtography,” which has been one of the most visited categories on Snopes over the past year, according to the site’s vice president of operations, Vinny Green.

Another popular category is “junk news,” or phony stories that are designed to draw traffic by intentionally misleading readers. Malicious entrepreneurs learned long ago that they can generate website traffic by taking advantage of a human weakness: our tendency to react to information that triggers a strong emotional response. When we see something that makes us feel anger or fear, or something that validates an existing bias, we tend to respond to it without thinking. On social media, that means liking, sharing, “hearting,” angry-facing, retweeting — all before stopping to verify that the information we’re spreading is correct.

As the tricksters who create fauxtography and junk news become more sophisticated, consumers are more easily duped. That’s why “deepfakes,” videos that have been manipulated to make individuals appear to be doing or saying things they did not actually do or say, are becoming a major concern among fact-checkers. Along the same lines are political quote memes, those boxes of text that contain quippy quotes attributed to politicians. They’re tantalizingly shareable — and quite often wrong.

Political figures are common targets for all forms of misinformation, which is why Snopes has increased its focus on political content in recent years. While reader interest in political stories used to drop off between presidential elections, Green says, “politics has never left the tip of our culture’s tongue in the past five years.” As the 2020 election season heats up, the number of political hoaxes and the demand for political fact-checking are likely to increase accordingly.

At Snopes, the process for fact-checking text-based content is similar to that for photos and videos. A staff member starts by trying to contact the source of the claim to ask for supporting documentation. They also contact individuals and organizations with direct knowledge of the subject. That reporting is backed up by research from news articles, journal articles, books, interview transcripts, and statistical sources, all of which are cited in the writer’s fact-checking story. At least one editor reviews the story and adds to the research as needed.

No matter how you define fake news or measure the political fallout, one major impact is clear: Its very existence has left readers disheartened and confused. A Pew Research Center study published in December 2016 found that 64 percent of adults said misinformation was causing “a great deal of confusion about the basic facts of current issues and events.” In a 2019 update, that number went up to 67 percent, and 68 percent of the Americans surveyed said that fake news has affected their confidence in government.

A 2019 report by the Knight Commission on Trust, Media and Democracy found that Americans have far less faith in their institutions — especially the media — than they did 50 years ago. It blames this “crisis of trust” on several factors, including the overwhelming number of information sources available online; the increasingly blurred line between news and opinion; declining news budgets; attacks by politicians on the media; and Americans’ inability to agree on what constitutes a fact.

“‘Filter bubbles’ make it possible for people to live in ‘echo chambers,’ exposed primarily to the information and opinions that are in accord with their own,” the report says. “One result of this technique is to provide users with content that reinforces their pre-existing views while isolating them from alternative views, contributing to political polarization and a fragmentation of the body politic. In turn, increasing political polarization encourages people to remain isolated in ever-more-separate ideological silos, offline as well as online.”

The problem is fixable, the report says, but it requires action by news organizations, tech companies — and us. Our main job: to learn how to consume media responsibly in this new media era. “My general advice to any news consumer or consumer of fact checks: Trust no one and nothing,” says Snopes managing editor Doreen Marchionni, a former Seattle Times editor.

If a news story or image seems scary or outrageous, that’s a red flag. If you see an image that doesn’t contain a link, be suspicious. If someone shares a picture of a tweet that doesn’t link to the actual tweet, it may be a fake. If an outlet publishing a story doesn’t have a protocol for running corrections or retractions of erroneous information, it might not be a trustworthy source.

“Start by looking for sound, primary data on the source of the stuff that you want to share,” Marchionni says. “See if you can find the original source of it.” Google unfamiliar stories and websites to see if they’ve been flagged as fakes. Use reverse image searches to find the earliest versions of suspicious images. Check independent, nonpartisan fact-checking websites for help with difficult cases.

In the meantime, resist the urge to share. “It is your civic responsibility and your civic duty to do the right thing by your [fellow] citizens. In this context, that means don’t share bad stuff,” Marchionni says. “Don’t share outrageous headlines and links unless you yourself know them to be true. If you can’t suss out the truth of the thing, then, by all means, check our website.”

But why should people trust Snopes? “Read up on our history. Look at the girth of our reporting across 25 years. Decide for yourself if you think we’re trustworthy,” Marchionni says. “I think we are, but basically the same rules apply when evaluating a potential meme by a white supremacist or evaluating a fact-checking organization that you look to in order to help you understand whether something’s true or not.”

Ultimately, the responsibility falls on each of us as consumers and sharers of news. “Misinformation has always been out there, since the dawn of humanity. What is different right now is social media,” Marchionni says. “It’s the act of sharing bad information that is creating this crisis we’re in.”

• Kim Lisagor Bisheff worked as a fact-checker in the late 1990s, when “fact-checking” was still a politically neutral term. Over the past 20 years, she has reported for newspapers, magazines, books, and websites. Bisheff has taught journalism at California Polytechnic State University in San Luis Obispo since 2004. She teaches multimedia journalism and public affairs reporting and gives talks to campus and community groups on news literacy and fact-checking.

• This story originally appeared in the February 2020 issue of The Rotarian magazine.