Nine years ago, in 2015, I gave up on fact-checking. For a year and a half, I’d spent every Friday debunking hoaxes, conspiracies and pranks for a tongue-in-cheek column called “What Was Fake on the Internet Last Week.”

Less was probably fake then than is fake now. (Donald Trump hadn’t even been elected yet!) But already, I felt mistruths had overrun the world’s collective feeds, and there was little I could do to correct them. The hoaxes had grown partisan, militarized. Believers embraced them like a religion. They committed to conspiracies and chain emails and strangers’ Facebook posts on faith, never asking for sources or evidence.

Against these forces, I argued then, what good is a fact check?

“At which point does society become utterly irrational?” My last column asked.

In some ways, I’m still grappling with those questions (… though with enormous respect, I want to note, for the fact-checkers still doing this thankless work). In the near-decade since I quit fact-checking myself, political discourse has only grown more fantastical. Haitian immigrants are eating pets. Democrats import gangs to torment small towns. Just last week, meteorologists faced threats from viewers convinced they “engineered” Hurricanes Helene and Milton.

These lies speak of a crisis so urgent — so existential — that meaningful solutions seem hard to fathom. So when I saw that PolitiFact founder Bill Adair wrote a new book about the causes and consequences of political misinformation, I was eager to hear his perspective on how we arrived at our current moment … and how fact checking, of all things, could possibly reverse it.

I remain a bit cynical, as you’ll see in the following interview. But I’m intrigued by many of Bill’s ideas, including the suggestion that fact-checking can deter lie-prone public figures, even if their constituents don’t universally trust mainstream fact-checkers.

So earlier this week, Bill and I caught up on Google Meet to chat about election misinformation, right-wing media and his new book, Beyond The Big Lie: The Epidemic of Political Lying, Why Republicans Do It More, and How It Could Burn Down Our Democracy. Despite that (grim!!!) title, I found this book an entertaining, fast-paced history of political lying, the fact-checking movement and ongoing efforts to shore up media access and literacy. 

You can buy Bill’s book anywhere, but if you buy it through this newsletter I will get a small commission. Please enjoy our (condensed and lightly edited) conversation!

First off — I’m delighted that you could make time for this, thank you. And we really couldn’t have timed it better. You’re based in North Carolina, and North Carolina has been absolutely swamped by hurricane lies and conspiracy theories over the past week and a half. I’d like to start there, if that works for you: What have you seen, in the discourse around Hurricanes Helene and Milton?

Bill: To give you some perspective, when I was a local news reporter, I covered hurricanes. So I’m no stranger to hurricanes becoming political. That’s happened for a long time — we can all remember Chris Christie hugging Obama and it being a big deal. I remember covering hurricanes in 2004 when President Bush was running for reelection, and there was speculation about whether it would help or hurt him in Florida. So hurricanes have always had politics involved. 

But what’s different this time is the level of misinformation and disinformation is greater than ever before. There have always been rumors during hurricanes, but I’ve never seen anything like this, where you have so many lies. And never before have I seen lies pushed by presidential candidates. That’s just completely unprecedented. It’s gotten to the point where you’ve had Republican politicians in North Carolina saying, “Cut it out.” It’s just amazing.

Then over the weekend, you have these poor disaster relief workers who are now getting harassed because of these lies. It’s really terrible, what’s happened, and it has big implications for lots of different things. One, what people believe about disaster relief — it undermines trust in government unfairly. But then it also does the same thing that all the lies about the 2020 election did, in the way that they undermined the work of elections officials and made a lot of people think, “Well, why do I want to work in an elections office?” Now, if you’re going to go to work in disaster relief, you’re going to get harassed about it. That’s the sort of toxic atmosphere that lying has created.

When we talk about the misinformation crisis, you hear a number of different theories for the root causes: maybe it’s media literacy, maybe it’s polarization, maybe it’s the failure of mainstream journalists or social platforms. Your emphasis on the lying itself strikes me as sort of unique. Can you tell me more about that? 

Bill: There’s definitely a media literacy core to it all — I think media literacy is huge. But that wouldn’t be a problem if politicians didn’t see the opportunity in lying — if they didn’t believe they could take advantage of a lie and get away with it. They speed on the highway because there are no state troopers, so to speak. 

So while addressing misinformation does involve this question of how we can get better information to people, it also has to do with how we can change the behavior of politicians. I do have some hope that we can, because this is essentially about incentives and disincentives.

I want to talk about those incentives and disincentives, but I want to stick with the media literacy piece for a beat longer. I’m curious to what extent you think people are swayed by corrective information at all. In other words, to what extent are people willing to change their mind about a lie after they’ve committed to believing it? 

Bill: That’s a little bit out of my expertise — I’m not really a behavioral person. I’m probably not the right person to ask that question. 

Sure, I understand that. At the same time, I’m sure you’re familiar with the studies that have found, for instance, that fact-checking labels on social media aren’t consistently effective, or that fact checks actually cause some people to double-down on false beliefs. I have to imagine you have thoughts on that. 

Bill: Now I can actually correct you on that. That phenomenon is known as the backfire effect. It comes from a famous study by Brendan Nyhan and Jason Reifler, which showed that fact checks cause people to double down on false information. 

But other researchers have since replicated that study and found that people who get corrective information do actually absorb that information, and it does generally correct falsehoods. If you talk to Brendan Nyhan today, he’ll tell you, “Nope — the backfire effect doesn’t occur.” But Brendan will also tell you that the backfire effect has a sticky name, so people tend to believe it — and also, it sort of fits into our sense that people are stubborn.

Overwhelmingly, the social science research shows that fact checks actually are effective. Here’s where I do think there’s a need: The social science research is limited in that a lot of it is experimental. They get people to do things that simulate the real world. What that fails to do often — not all the studies, but a lot of them — is recognize that a whole bunch of people in the United States don’t get fact checks. They not only never see fact checks, but they are told every day that fact-checkers are a bunch of lefties who they can’t trust: They’re the mainstream media, they represent the Democratic Party and Lord knows what else.

Thank you for fact-checking me! That was very meta and I appreciate it. If that’s the case, though — fact-checking works, but it doesn’t reach folks on the right — how do you correct for that? How do you make sure that conservatives have access to, and also confidence in, credible fact checks? 

Bill: We need more fact-checking by conservative media organizations to reassure conservatives that fact-checking is a legitimate form of journalism. The Dispatch, for instance, is a great organization, and they do fact checks that are really solid. A friend pointed out to me over the weekend that The Daily Caller also did a good fact check about some of the lies in North Carolina. I wish there were others. There aren’t, that I know of. But I do think more fact-checking is the answer.

I’ve been criticized for this, and I can understand why. The criticism of me there is, “Well, how’s that working for you?” But I think we have to flood the zone. The reason that fact-checking isn’t more effective is there simply isn’t enough of it. To go back to my state trooper metaphor, we need to have more fact-checking to really have a deterrent effect on politicians. I do really believe that if you get enough state troopers out on the highway with radar guns, people are going to slow down. 

My team at Duke University did a study about a year and a half ago we called “fact deserts,” and what it found were there are giant swaths of the country that have no fact-checking. A U.S. senator or member of the House can go a year without ever being fact-checked. That’s not a good thing.

Now, does that mean everybody’s going to stop lying? No, Lord knows. Springfield, Ohio was a fact-checking success story. We knew the truth that Haitian immigrants were not eating dogs and cats, but Donald Trump decided to just keep lying — he has no shame. So if someone is going to go beyond the bounds of what have always been the acceptable traditions of American politics, at some point, you’ve just got to throw up your hands. But I do believe, in general, if we were to get lots more fact-checking, it would have a tremendous effect.

I so want to believe that’s true, as a concerned member of civil society and a journalist (!). But it’s a little difficult for me to wrap my head around, especially when we have myths like Springfield that are so durable and so impervious to corrective information. How do you know we haven’t already reached the “throw up your hands” stages? 

I don’t think the complete eradication of false beliefs has ever been a realistic goal. We’re not going to eradicate people believing in dumb stuff. People have always believed dumb stuff, and people will always believe it. That’s just the reality of humanness.

But I do think that it’s troubling. I don’t know what the latest poll shows, but I’m sure if we look at the percentage of people who still believe the lie about Springfield, we would bury our faces in our hands. [Note: The latest polling I uncovered is from a month ago, but it found that just over half of Trump supporters believe the pet-eating myth.] On the other hand, a whole lot of people know the facts thanks to some great fact-checking. And a whole bunch of responsible Republicans came out and said, “That’s wrong. Cut it out.”

At some point this is about aberrant behavior by political leaders, and I don’t think you can blame fact-checkers.

No, I’m certainly not blaming fact-checkers. I just wonder if the Overton window hasn’t already shifted so much that lying isn’t necessarily “aberrant” behavior anymore.

Bill: Well, that’s right. I think it has shifted, and I hope we can get people talking about what we need to do to shift things back. Our whole news media’s approach to covering politics has been historically based on evenness — this idea that both parties generally behave by the same rules and follow certain informal rules of engagement. They both want to pass a budget. They don’t want to shut down the government. When I started covering Congress in 1997, there was a respect for the body itself. 

And you do see a change from that, which the media hasn’t caught up to yet. I’m thinking about, in particular, how the media deals with this great disparity in lying between the Democratic Party, which lies some, and the Republican Party, which lies so much more. 

But that isn’t brand new, either. Newt Gingrich was a turning point there. Many people, when I started to ask about why Republicans lie more, cited Newt as a key figure in how the party’s culture has changed in recent years.  

Can we talk about social media for a minute and how that has sort of supercharged the phenomenon of political lying? You single out Facebook in particular as a sort of ground zero for the spread of misinformation. Tell me about that. 

Bill: This was interesting. Facebook comes up a lot in the transcripts of the interviews of all the people whom the January 6 committee interviewed — the people who’d stormed the Capitol. One thing I noticed, as I read all the transcripts, was that the investigators asked the same questions over and over again: “How did you find out about the Stop the Steal rally?”“What social media did you use?” In most cases, they didn’t use Parler or Truth Social or these other more conservative platforms. Everyone used Facebook. Facebook was the common denominator. 

So I ended up focusing on a guy from Parkersburg, W.V., named Eric Barber — fascinating guy, wonderful guy. I really got to like him as I dug into his story and got to talking to him, to the extent that he would. He was very candid about how he was radicalized by Facebook. 

Barber had been a city council member and had gotten elected as a Democrat, but then he came to this vote where he realized that the only way he was going to survive was if he switched parties. So he became a Republican and, when he did, he started to get a little Trumpy in his comments on Facebook. He then started to get all these likes — and he also was fed this constant stream of conservative media clips from Fox and other sources. It’s really the algorithm that causes a lot of the problems. 

So after the 2020 election, when Barber saw Trump in a Facebook ad that said, “We need you to come to Washington” … he went! And he went with the idea of creating content for his Facebook account as much as he went to join Trump at the rally, because his whole world was through the prism of Facebook. I found that really fascinating.

That is fascinating. So many of the debates around social media radicalization involve platforms like YouTube or TikTok — we probably don’t talk about Facebook enough anymore. And Facebook is also notable in that so much of this activity plays out in closed groups, which is a growing trend across platforms. How do you get good information into these gated communities, whether on Facebook or Telegram or Discord or somewhere else? 

Bill: Yeah, that really can’t be done. I think there are legitimate privacy issues there. I mean, I don’t think fact-checkers should be breaking into buildings and listening to people’s meetings — nor is it appropriate for them to sneak into a 15-member Facebook group to make a correction.

However, when you’re in a place like India, and Prime Minister Narendra Modi sends a WhatsApp message that goes out to hundreds of thousands of people, I think it’s legitimate to monitor that and fact-check it. So at some point, journalists have to figure out what’s appropriate in this new world. 

Either way, those closed conversations can only be fact-checked if somebody inside the conversation sends the message to a journalist that says, “Hey, tell me if this is true.” That’s what fact-checkers have been doing with WhatsApp. But it’s not very effective.

How does generative AI impact the work of fact-checkers? We obviously hear a lot about its potential to supercharge misinformation. 

Bill: I think of it both ways. The bad guys are probably going to use it first at scale. I don’t think deep fakes have really clobbered us yet, but we saw cheap fakes take off a few years ago and we’re seeing more of them now. [Note: “Deep fake” describes an AI-generated video; a “cheap fake” typically involves real footage that has been manipulated or taken out of context, either with AI or other editing software.] We’re going to keep seeing more of those as AI makes them easier. So the bad guys will beat us to AI, because they’re resourceful and they have more money and more motivation.

What can the journalists do? We have a concept we’re testing at Duke called “Half-Baked Pizza.” The idea is that politicians repeat their lies in different places — they have talking points. So if a human fact-checker in one state does a fact-check on a talking point, we should be able to check that claim automatically the next time it comes up. But you still need a human to review it, because you want to make sure there’s no hallucinations in the new content. That’s what I liken it to a half-baked pizza. 

The benefit of using generative AI is that we can produce more pizzas and get them out to more people, including the people in big parts of America that don’t have access to fact checks. I think that’s where AI can combat the bad guys who are otherwise going to beat us to implementing it. 

I know we’re already over time here, so thank you for that! But I hoped you’d allow me one more question: We’re obviously bearing down on a presidential election in the U.S. that has been plagued with all kinds of lies and conspiracies. I’m curious what story lines you’re following as we approach election day. What keeps you up at night? What are you tracking?

I’m always interested in where people are getting information and the new places where they’re getting information. I don’t know where that is this year, but I’m keeping an eye on it. 

Thematically, I think one recurring story this election is that everything is that “simple is better.” Everything is so complicated, and the candidates who can offer simplicity do better. Unfortunately, that doesn’t always mean that they’re accurate, or that the media organizations that offer simplicity are superior, but I am struck every day by that reality: Our world is filled with complexity, and the people who seem to succeed are the ones who simplify things.

If I’m not mistaken, your book quotes Obama advisor Valerie Jarrett saying something similar: Democrats are always trying to explain things, to explain nuance, and that’s ultimately to their detriment. It’s not what sticks with people. 

Bill: Yes, and I thought that was right on the money. People don’t want nuance. They want something they can understand.

I don’t know if I’ll call that a great note to end on. But it does feel appropriate!