The misinformation cope

Democrats are misunderstanding their losses with less-educated voters

A couple of weeks ago, Barack Obama spoke at a University of Chicago seminar on Disinformation and the Erosion of Democracy. And as you might expect from a speaker at a seminar on Disinformation and the Erosion of Democracy, he offered the view that misinformation plays a major role in shaping our current problems.

Barack Obama @BarackObama

In recent years, we've seen how quickly disinformation spreads, especially on social media. This has created real challenges for our democracy.

I acknowledge that misinformation exists and that it’s bad and that ideally everyone would be well informed. But I also think that the structural biases of media and academia exaggerate this problem.

Jeffrey Goldberg, editor-in-chief of The Atlantic, recently declared that “Disinformation is the Story of Our Age.” But this article is just hype for the same conference Obama spoke at, a conference organized in part by The Atlantic. The importance of improving the information ecosystem is a conveniently appealing message for a highbrow magazine owned by a wealthy philanthropist, but it’s really Obama I’m disappointed in.

Barack Obama was a great politician. But while he’s very influential among Democratic Party movers and shakers, he doesn’t like to talk about the hard-knuckled aspects of his own approach to politics. And I agree with Michael Slaby that the misinformation focus among Democrats feels like a self-exculpatory cope:

Democrats are also seemingly preparing to blame disinformation for an ongoing lack of clarity and imagination, for an unwillingness to see their culpability in losing hearts and minds and failing to win them back, for an inconsistent ability to deliver for people. Failure of leadership is never the answer. The Russians, Trump, McConnell. And now disinformation. Never a lack of imagination, vision, organizing, effective long-term investment, being valuable to people, empathy.

I wish Obama had instead said that there’s no evidence that conspiracy theories are becoming more prevalent, that deactivating Facebook makes people less knowledgable about politics, that poorly informed people have always been with us, and that one part of politics is delivering quality governing results while the other part is meeting people where they are, not pining for some alternate reality where they have totally different beliefs.

The seductive half-truth of misinformation

Like a lot of dangerous ideas, this desire to blame misinformation contains a kernel of truth.

Independents and less-educated people are less knowledgable about politics. And in the 2016 and 2020 cycles, Democrats picked up support from college graduates (largely women in 2016, joined by men in 2020) while losing support from non-graduates (largely white in 2016, joined by many Hispanics in 2020). This is to say that Democrats have gained a lot of ground with demographics that correlate with being well-informed while losing ground with demographics that correlate with being poorly informed.

It is certainly possible to interpret that trend as suggesting that some people have swung toward Trump because they are misinformed while others have swung away from him because they are well-informed.

But recall that 10 years ago, the late Rush Limbaugh had a running gag about “low-information voters” based on the (true!) observation that at the time, Republicans were meaningfully better-informed about basic political facts than Democrats. That’s because in addition to educational attainment, political knowledge correlates with age (young people are less informed than old people) and with gender (women are less informed than men). Ten years ago, there was a strong tendency for less-educated white people to vote Republican, but that was offset by Obama’s huge margins with Black and Latin working-class people. So in the 2012 alignment, Obama voters were not better educated than Romney voters, but women and young people did lean left while men and old people leaned right, which meant that Obama voters were worse-informed than Romney voters.

So if you agree with Rush Limbaugh that Obama won thanks to misinformation, then I’m happy to see you apply that to voters who’ve swung toward Trump. But I think that if you resist Rush’s interpretation of Obama’s popularity, you ought to resist that interpretation of Trump’s.

The fact is that Obama did much better than Joe Biden with two groups of people — poor white people and Hispanic women — who happen to have low levels of political information. But if you take progressive ideas seriously, these are people whose interests we want to be representing. It’s on us to win their votes, as Obama did, not to charge them with being insufficiently informed. This applies not just to general ignorance but also to so-called “fake news.” A good recent study by Charles Angelucci and Andrea Prat shows that all kinds of people are pretty good at telling real stories from fake ones, and “the starkest pattern about the ability of voters to identify major news stories is not the generalized death of truth or its ideological polarization but rather its unequal distribution along socioeconomic lines.”

Less-educated people are less knowledgeable and less media literate, and that’s not ideal. But Democrats need to read the correlation in the correct direction and try harder to appeal to their values, not write them off as too misinformed to be reached.

Disagreement isn’t disinformation

Here’s a recent Jazmine Ulloa piece in the New York Times:

Online disinformation hit Latino communities hard ahead of the 2020 presidential election.

It came in the form of videos, tweets and WhatsApp messages, YouTube videos and the rants of Spanish-language radio hosts. It included false reports of widespread violence on the streets of Democratic cities after the murder of George Floyd, QAnon conspiracy theories, and overblown claims of terrorists and criminals crossing the U.S.-Mexico border.

Are “reports of widespread violence on the streets … after the murder of George Floyd” really an example of “online disinformation”? I don’t think there is a technical definition of when violence becomes widespread. A number of stores in my neighborhood had their windows smashed and their inventories stolen. Across the country, the total damage from rioting was around $1 to 2 billion, comparable to the damage done in the LA Riots of 1992. Of course we’re talking about incidents that were spread across dozens of cities rather than concentrated in any one metropolis. So each individual riot was dramatically smaller than the one from 30 years ago, but the rioting was much more widespread.

Looking back on the politics of the riots, I think the thing that Republicans clearly got wrong was the implication that electing Joe Biden would lead to more riots — what happened instead was totally the opposite.

“I don’t like rioting” was a bad reason to vote for Trump and “Biden’s election will lead to more rioting” was a bad analysis. But it’s true that Trump effectively branded himself as the “rioting is bad” candidate and that many people were very disturbed by the rioting. Perhaps you disagree with them, but this is fundamentally not a factual question. And the fact that some liberals want to process concern about rioting as a form of disinformation suggests that voters who were very bothered by rioting were correct to assess that the Democratic Party’s leaders were less aligned with them on that question.

Voters are in some ways getting better-informed

This past Sunday was Easter, and I saw lots of people — including many influential political and media figures — tweet phrases like “He is risen!”

That’s because, according to the Bible, Jesus Christ was executed on a Friday and then rose from the dead two days later. The holiday of Easter commemorates this occurrence. And yet it’s not actually true that Jesus rose from the dead. Or to use the jargon of the modern fact-checker, experts say there is no evidence that Jesus was ever resurrected, and saying that dead people cannot be revived has a strong theoretical basis in science. But of course neither Twitter nor Facebook flagged these posts as misinformation, and none of the professional fact-check organizations or misinformation reporters complained about the fact that churches around the country were spreading bad facts about this.¹

Indeed, according to the General Social Survey, the false belief that the Bible is the literal word of God (the other options are that it’s divinely inspired or that it’s a book of fables) is extremely widespread in American society and is becoming increasingly aligned with partisanship.

Using the GSS Data Explorer, you can see that Biblical literalism correlates with being Hispanic, being a woman, and having a lower level of education. So at least some of the growing alignment of Biblical literalism with voting for the GOP is the same phenomenon as the tendency of non-college Hispanics (especially women) to switch to Trump in 2020. And indeed some analysts believe that vote-switching was specifically correlated with belief that the Bible is the literal word of God. Now does that mean that disinformation (i.e., the Bible) caused people to vote for Trump? I think that’s an odd way to think about it.

Equis Research has a much more straightforward take, which is that conservative Hispanics (especially younger ones and women) flipped to Trump.

These young conservative Latina Clinton-Trump voters very likely do have a lot of false beliefs — not only false beliefs about the Bible but also about American politics. But fundamentally their decision to start voting for the more conservative party seems a lot like Limbaugh’s low-information voters becoming better-informed about the structure of American politics. Democrats became more left-wing and more explicitly centered on Black identity appeals, while Trump toned down the most overtly anti-Hispanic rhetoric from his 2016 campaign, so some conservative Hispanic voters were dislodged from identity-based voting and voted for the more conservative candidate.

Another big problem for Democrats that Simon Bazelon wrote about recently is that the precipitous decline of ticket-splitting is making it way harder to pull off the old Tom Daschle/Jon Tester move of winning in a red state based on a strong personal brand and rural identity appeals. Lots of rural white working-class voters are coming to see that it doesn’t matter that Tester seems like a nice relatable guy if he’s still going to vote for all of Biden’s nominees.

Democrats are afflicted by self-misinformation

I think the basic thing that Democrats have been missing for years about Donald Trump’s electoral appeal is that Donald Trump in fact has very little electoral appeal.

Most of the stuff that Trump-haters say is terrible about Trump is in fact terrible, and most voters in fact do not like it. That’s why Trump ran behind congressional Republicans in 2016 and again in 2020, and it’s why Trump being off Twitter has been good for the GOP cause. The reason the fundamental electoral toxicity of Trump is invisible is that ever since 2012, the Democratic Party has shifted its positions on a bunch of issues to become more left-wing, and becoming more left-wing generally costs you votes. These two trends — people don’t like Trump and people don’t like the Democratic Party’s leftward evolution — have been broadly offsetting, though right now the GOP has the upper hand thanks to inflation.

But this leftward evolution is in turn invisible to a lot of progressives because they insist on viewing everything in terms of factions and personnel. Every few weeks Adam Jentleson and I have a version of this argument where he explains that the problem can’t be that Democrats got too left-wing because “the centrists” (which to him is a defined community of specific individuals) are still running the party.

Adam Jentleson 🎈 @AJentleson

It’s really incredible that anyone can still think centrists are some besieged faction when, after running the party for the past 40 years, they still set the entire legislative agenda for the first two years of Biden’s presidency and are now setting the midterm message for 2022. https://t.co/FGOL4ZULD6

Al From @alfromDLC

I remember when we built one in the Democratic Leadership Council. You can read the game plan in my book, “The New Democrats and the Return to Power.” Or view it in the documentary film, “Crashing the Party.” https://t.co/q1uv5drAAO

And that is absolutely true. Joe Biden — the factional candidate of moderate Democrats — defeated Elizabeth Warren and Bernie Sanders in the 2020 primary.

But it’s also true that during his first term, Barack Obama spoke sympathetically about his racist white grandmother, said he believed marriage is between a man and a woman, bragged about setting record deportation levels, downplayed the gun control issue, never proposed anything close to the multi-trillion Build Back Better initiative, didn’t try to block oil & gas leasing on public lands, and frequently annoyed Black intellectuals with his practice of respectability politics.

Maybe some or all of this change was for the better. But it was change, and I think if your view of why some people who voted for Obama later voted for Trump focuses on misinformation rather than shifts in actual issue positions and salience, then you are the one who is misinformed.

1. To be non-sectarian about this, I should say that until very recently I believed that the story of a large Jewish migration out of Egypt and into Palestine must be broadly true even if the supernatural elements were not. This is apparently not what researchers have found despite considerable effort to uncover evidence for it — the Hagaddah is full of misinformation!