The Undecided Nihilists
Campaigns do matter, actually
With election anxiety on high, an intraparty debate is returning to the fore: should campaigns focus on persuading undecided voters, or turning out likely supporters?
This is mostly a false choice, as swing voters and “base” voters are motivated by similar issues. But the Persuasion vs. Turnout debate persists, driven by the latter group’s denial that swing voters even exist.
There are two main reasons for swing voter denial: ignorance and malice.
OK, that’s a bit harsh. Two related worldviews animate the denial that swing voters exist:
living in a bubble without swing voters
not wanting candidates to appeal to the centrist impulses of swing voters
Swing voter denial is common in leftist outlets like The American Prospect, but has also gained prominence in mainstream liberal publications in recent weeks. Such denial, given the absurdity of a second Trump term, is both understandable for readers of those publications (who could be undecided?!?) and dangerous. We’ll examine two examples, from The Atlantic and The New Yorker. We’ll rebut the common charges against swing voters (that they are either rare, unknowable, or unpersuadable). And touch on why it matters for polarization and intraparty debates (if there are no swing voters to compete for, all competition happens within parties).
Bad Math in The Atlantic
In “The Undecided Voters Are Not Who You Think They Are,” Ron Brownstein notes that polls show one-sixth of voters but that “most strategists in both parties consider those numbers an illusion”.
When most people think about a voter still trying to make up their mind, they probably imagine a person who is highly likely to vote but uncertain whether to support Harris, Trump, or a third-party candidate. Both political parties, however, are more focused on a different—and much larger—group of undecideds: potential voters who are highly likely to support Harris or Trump, but unsure if they will vote at all.
Campaigns typically describe the first group of reliable but conflicted voters as persuadable; they frequently describe the second group as irregular voters. Persuadable voters get the most attention from the media, but campaigns recognize that irregular voters can loom much larger in the outcome—especially in presidential elections when more of them ultimately participate.
Brownstein estimates the parties place 4 to 7 percent of voters in the persuadable category. He then gets caught in an activist trap in the Persuasion vs Turnout debate: downplaying the impact of policy on centrist swing voters while overhyping the ability of leftist rhetoric to boost Democratic turnout.
(Sunrise Movement) sees a fundamental conflict between Harris’s attempts to reassure centrist swing voters, by emphasizing moderate positions on energy from fossil fuels and on the war in Gaza, and her need to activate more progressive young voters uncertain whether to vote at all.
This is wrong on multiple levels. First, polling is clear that young voters prioritize issues similarly to the rest of the electorate. As
shown in this rundown of the Harvard Youth Poll, young voters are concerned about prices. And just 6% of swing voters are “Very Liberal” compared to 74% either moderate or conservative.Second, there is a basic math function at play here: flipping a likely voter from red to blue is worth two votes, whereas making a nonvoter a voter is worth one vote.
Also, a quick check on the data: if appealing to far-left or far-right constituencies was the way to win swing districts, where are the examples of politicians winning on the other party’s turf without focusing on centrist swing voters?
Is it random that the ‘most bipartisan’ GOP members of Congress are Republicans like Don Bacon and Brian Fitzpatrick, representing districts Biden won? Or that the most moderate modern Republican governors represent deep blue states like Maryland and Massachusetts (and vice versa for Democratic governors in Louisiana and Kentucky)? Is it random that the 'most bipartisan’ Democrats in Congress are those winning Trump districts like Marie Gluesenkamp Perez, Mary Peltola, and Jared Golden?
Nihilist Bubble in The New Yorker
Such randomness was proposed in The New Yorker last month where Jay Caspian Kang proclaimed “The Unknowability of the Undecided Voter.” He claims that “after nearly a decade of inquiry into these undecided voters … we still don’t really know who these people are, despite the fact that we, as the press, have seemingly interviewed every single one of them, sometimes more than once.”
Kang is an excellent writer, but here makes two mistakes.
First, he undercounts the share of persuadable voters. As we have covered, volatility in the electorate is badly underrated. A fact reinforced by the highly rated Monmouth poll showing just 73% of voters are “Definitely” voting for either Trump or Harris. Roughly eight million voters flipped between the 2016 and 2020 elections. Not all of them have been interviewed!
Even taking Ron Brownstein’s estimate that 4 to 7 percent of likely voters are persuadable brings a count around ten million voters. That’s a lot!
Second, political science consistently demonstrates moderate candidates do better. There may not be a uniform ideological profile of a persuadable voter, but the direction is clear. And leading outlets muddying the waters (without citing evidence from political scientists or polling or elsewhere) ain’t great. Swing voters may not be uniform, but they are knowable.
After we pointed this out, Kang tweeted a rebuttal to our charges of nihilism:
Yes I don’t put much faith in opinion polling, not bc they ‘comply w a worldview,’ whatever that means, but bc they’re mostly bad data.
And the problem is they’re NOT ‘evidence backed’ lol. Anyone can generate a number. I find it much more ‘nihilist’ to believe that this ‘data’ about what you admit is an eclectic and odd bunch of voters is actionable. You can gain an inch on one front and lose two elsewhere.
Kang then doubled down with another piece in The New Yorker, Has the Presidential Election Become a Game of Random Chance?
The subtitle is “We have seemingly reached an end point in polarization, where any new developments short of swapping out a candidate wholesale will be met with indifference in the polls.” And if that is not nihilistic enough, here’s his summary:
Basically nobody knows what matters in this election, has no solid theories for what voters want, and whether what happens in the world can change minds.
But is it true that we have no idea what voters want?
The Senate campaign of Rep. Ruben Gallego is instructive. A springtime headline from The Washington Post read “The moderate reinvention of Ruben Gallego, Senate hopeful in Arizona.” An NBC News headline was even more specific: Ruben Gallego redefines himself … no longer describing himself as progressive. Like Trump, Gallego has been endorsed by the Arizona Police Association. He would “love to have” Kyrsten Sinema’s endorsement. He is calling for more Border Patrol funding.
The result? Polls have shown Gallego running far ahead of Harris, including a recent CBS News poll showing him twelve points ahead of Harris.
Semafor’s Benjy Sardin posited “Why is Ruben Gallego outrunning Kamala Harris by a mile in Arizona? One possible answer: He's spent two years pulling off the same pivot she's had just months to execute.“
Denying Depolarizers
We’ve decried “polarization porn” that overhypes how many persuadable voters are out there. But Kang takes it another step. When he denies the political science, the polling, and the case studies, he is denying that depolarization is possible. He is denying that there is something we can reliably do to practice democracy effectively. Denying that a Democrat can win a Senate seat in West Virginia, or an R+5 seat in Washington, or beat Freedom Caucus Chair Scott Perry in an R+5 seat.
And denying that taking leftist positions can make losing more likely, as Matthew Yglesias noted in reinforcing our critique highlighting the danger of such writing:
when a prestigious magazine like the New Yorker runs Jay Caspian Kang’s nihilism about the political significance of issue-positioning, that really does help persuade the New Yorker’s liberal audience that there is no cost to Democratic Party politicians saying and doing stuff that New Yorker readers agree with. If Kang is right about this, then of course it’s good for the New Yorker to be equipping its readers with that information. But I don’t think he’s right, and the aggregate impact of an editorial saying “Trump Is A Threat to Democracy” and an article saying “Issue-Positioning Doesn’t Matter” is just to exacerbate the threat to democracy.
Kang is the rightful heir to another writer for The New Yorker, Pauline Kael, who famously remarked after the 1972 election that:
"I live in a rather special world. I only know one person who voted for Nixon. Where they [Nixon's other supporters] are I don't know. They're outside my ken. But sometimes when I'm in a theater I can feel them."
Nixon won 48% of the New York City vote in 1972 (and 34% in Manhattan).
Sympathy for the Bubble
Kang is an excellent writer, with talents beyond just pen to paper (he’s also Emmy-nominated1). And seems like a perfectly nice guy. So I’d like to bring empathy to this critique. In Kang’s response to our original piece, he accused us of doing “a whole lot of pathologizing about my intentions.”
I think Kang fits into the bubble category of swing voter denial, not the willfully misleading category. It isn’t a pathology, it may just be part of living in a bubble. Kang lives in Berkeley, where Trump got 3% of the vote in 2016 (second lowest of any municipality in the nation). Trump came in third place behind Jill Stein! Kang grew up in Cambridge2, which placed just behind Berkeley in share of the vote for Trump (6%).
When Pauline Kael noted she only knew one Nixon voter, that became shorthand for the shrouded liberal elite bubble. But she could at least feel them in the theater. Nixon voters were ten times as prevalent in 1972 Manhattan as Trump voters are in present-day Berkeley. Kang’s biography reads like a Liberal Mad Libs. In addition to Cambridge and Berkeley, he’s lived in San Francisco and Los Angeles. He’s written for the New York Times and was a civil rights correspondent for Vice on HBO. He did undergrad at Bowdoin College, which enrolls fewer low-income students than almost any college in the country, despite an endowment of more than $1.5 million per student. He then got a Masters in Fine Arts from Columbia (cost per year: $82,829).
There is nothing wrong with that! I love getting a fresh copy of The New Yorker, and I’ve appreciated Kang’s work on other subjects (like gambling addiction and risk). The world needs all kinds of people, and he seems like a great one.
But our democracy does not need nihilism, which Camus said “is not only despair and negation, but above all the desire to despair and to negate.” There’s something about liberal bubbles - and The New Yorker - that embraces the idea that nothing matters.
Luckily, a whole bunch of voters and candidates know they’re wrong.
Updated 11/19/24, previously incorrectly said Oscar-winner
and Chapel Hill, NC