Dinesh D’Souza can always be counted on to have a bespoke conspiracy theory. In the trailer for his “2,000 Mules” “documentary,” faux static flickers across the screen over footage of what appears to be people voting at ballot drop boxes. A slow, mournful rendition of “The Star-Spangled Banner” plays in the background. And that’s when D’Souza, a pro-Trump conservative personality who was pardoned by the former president after being convicted of making illegal campaign contributions, drops the (totally unproven and false) bombshell: Thousands of “mules” stuffed ballot boxes to steal the election for President Biden.
This is D’Souza’s version of the Big Lie, the unfounded claim that widespread voter fraud plagued the 2020 presidential election and that President Joe Biden’s win was illegitimate. But it is not the only version — the Big Lie means different things to different people.
A third of Americans — and a majority of Republicans — believe the Big Lie, but the disparate theories include the flawed-but-plausible, the unlikely and the illogical. While no two Big Lie theories are identical in composition, they do share commonality in function. In-depth polling and academic research reveal belief in the Big Lie is incredibly resilient and incredibly influential, swaying the creation of new voting laws and shaping election campaigns for 2022 and beyond.
There’s a mountain of baseless overlapping claims piled up inside the stultifying biodome of the Big Lie: voters casting multiple ballots, dead people voting, ballot-counting machines flipping votes, foreign nations hacking systems to swap totals. The Big Lie is an à la carte conspiracy theory — a bit like QAnon in that respect — where adherents pick and choose what sounds right to them and disregard what doesn’t. Each individual who believes the Big Lie has their own suspicions about what took place, a personal recipe of different conspiracies to nourish their belief that the election was illegitimate. In right-wing chat groups on the messaging app Telegram, these theories are traded as casually as chats about the weather.
“What’s going to be any different this time?!” a user in a QAnon Telegram chat wrote about the midterm elections this past Tuesday evening. “Wait till 3am on election night and lots of mail in ballots and dead people voting.”
Brian Cates, a writer for the far-right publication The Epoch Times, was less morbid and more focused on the levers of power when he posted to his nearly 80,000 followers on Telegram earlier this week. “[W]hat happened on election night? Trump was surging to victory. They had to shut everything down and then spent THREE DAYS furiously ginning up enough fake votes to hand Biden the ‘win’ on Friday night.”
Every iteration of the Big Lie, though, is wrong. The ones in the darkest corner of the Internet? Wrong. The ones brought forward in lawsuits by the Trump campaign? Wrong. The ones already debunked by news sources? Still wrong. There is no evidence of widespread fraud in the 2020 election.
Still, polling gives us a glimpse of the most popular theories on the Big Lie menu. Last summer, a YouGov/CBS News poll asked voters who thought there had been widespread voter fraud and irregularities in the 2020 election exactly what they thought had happened. They were asked about various sources of voting and how much of the voter fraud came from those sources, either “a lot of it,” “some of it” or “hardly any or none.”
Seventy-seven percent said “a lot” of voter fraud and irregularities had come from ballots cast by mail, and 70 percent said a lot of it had come from voting machines or equipment that were manipulated, but just 22 percent said a lot of the fraud had come from ballots cast in person. Racism also appeared to inform a lot of thinking around the Big Lie: 72 percent said a lot of the fraud had come from ballots cast in major cities and urban areas, compared with 22 percent and 14 percent who said a lot of it had come from suburbs and rural areas, respectively. And 39 percent of those who believed voter fraud was widespread said “a lot” of fraud had come from ballots cast in Black communities, while 25 percent said so for white communities and 27 percent said so for voters in Hispanic communities.
The University of Massachusetts Amherst and YouGov fielded a similar poll in December, asking those who said Biden’s victory was probably or definitely illegitimate to choose from a list of options all the reasons they believed that to be true. In that group, 83 percent said they believed fraudulent ballots had been counted, 81 percent said they believed absentee ballots had been cast using dead people’s identities and 76 percent endorsed the theory that ineligible voters had cast ballots. Just 39 percent said they believed voting machines had been reprogrammed by foreign governments to change ballots from Trump to Biden.
“It’s important to note that people are not just picking everything,” said Alexander Theodoridis, a political science professor at the University of Massachusetts Amherst who fielded the survey. “Pure expressive responding would be like, ‘Yeah, everything was wrong with this election. It’s all this stuff.’ Whereas the way they did it suggests they do have at least some critical view of some of these [theories].”
The idea of expressive responding — the tendency for people to give answers to pollsters that reflect their feelings or values, but not necessarily their true beliefs — has been cited as a possible explanation for the high percentage of Republicans in polls who say they believe the election was stolen. Some Republicans, in other words, might say they think the election was stolen when what they really mean is they think the results of the election stink.
Theodoridis has used some polling tools to test how genuine belief in the Big Lie is. He and his colleague Lane Cuthbert used a research technique meant to reduce expressive responding that went something like this: Survey respondents were divided into two groups, given a list of statements and asked how many of the statements they agreed with, but not which ones. The list of statements were identical, except one group was given an extra statement that the researchers wanted to test on: “Joe Biden’s victory in the 2020 presidential election was legitimate.”
“Basically, if the average person in the control group chooses two out of four items and the average person in the treatment group chooses three out of the five items, you know that the only difference between those averages is the additional fifth item,” Theodoridis said.
Using this method, Theodoridis and Cuthbert found that only 28 percent of Republicans thought that Biden was legitimately elected. In their survey, 21 percent of Republicans said Biden’s victory was legitimate, and 8 percent said they weren’t sure. Theodoridis said this suggests that the Republicans in the survey were expressing a genuine belief.
But even if some or all of the Americans who say they believe the Big Lie don’t really mean it, the consequences of that expressed belief are very real. A forthcoming study based on a series of daily tracking surveys shows that even when those who believe in the Big Lie are confronted with evidence that counters it, their beliefs don’t waver. Kevin Arceneaux, a political science professor at Sciences Po, and Rory Truex, a professor of politics and international affairs at Princeton University, teamed up to run tracking surveys of U.S. registered voters from October 2020 through January 2021. As part of the surveys, the pair asked respondents who had identified Trump as the winner of the 2020 election about whether certain scenarios (such as Trump conceding, the Electoral College awarding more votes to Biden or Biden being sworn in) would change their minds. Of those registered voters, 45 percent said they would change their mind if Trump conceded, 31 percent said they would if Biden won the Electoral College, and 38 percent said they would accept the results if Biden was sworn in. But as those events occurred and Arceneaux and Truex continued their survey, belief in the Big Lie remained relatively unchanged.
“It’s not totally surprising — if you ask a hypothetical question, you get a hypothetical answer,” Arceneaux said. “I was surprised how much the aggregate results did not really move in response to actual events that happened.”
Arceneaux and Truex’s work revealed that not only is belief in the Big Lie tenacious, it also impacts voters’ views of politicians. When they asked Americans to compare hypothetical political candidates, Republican voters favored candidates who embraced the Big Lie by an average of 5.7 percentage points to candidates who accurately said Trump lost the election. This suggests that the Big Lie is not going anywhere soon and that it will have a meaningful sway on elections. Already we’ve witnessed the Big Lie being wielded as a campaign tool by Republican candidates across the country, demonstrating the power of this belief among the party’s voters.
And as polls continue to capture the millions of Americans who endorse the Big Lie, precisely what they believe matters less than how that belief influences their actions.