Steve Marcus / Reuters

After Democrats gained a House majority, causing most of them to celebrate the biggest check on Donald Trump’s power since he was elected, a tiny faction in the progressive coalition reacted in anger and frustration, fixating on races that would have made the Democrats’ “wave” even bigger: Beto O’Rourke in Texas, Andrew Gillum in Florida, Stacey Abrams in Georgia.

In all of these Democratic defeats, there was an easily identifiable group that voted overwhelmingly against the progressive candidate: Republicans. But members of this progressive faction did not lash out at Republicans. They instead directed their ire at another group, defined by race and sex. They lashed out at white women.

And their vitriol surprised many observers. It reached its apotheosis in a graphic tweet by the actress Heather Matarazzo. “What the fuck is wrong with you fellow white ladies???!!!!!” she demanded. “May you choke to death on the white supremacist patriarchal cock.” It’s hard to conjure a more counterproductive approach to intersectional feminism than for one white woman to shout insults at other white women based on the results of a single Senate race fought between two men. Yet that tweet was inspired by this widely circulated CNN image:

The conversation that chart provoked captures the distortions of identity politics, as its presently practiced by an influential faction of progressives. Overall, white women in 2018 split their votes evenly between Republicans and Democrats. That Ted Cruz improved substantially on national partisan voting patterns is explained by the composition of the Texas electorate—not by any generalized trait of white womanhood in America.

The votes of white women varied widely by region, religious identity, and educational background, among other salient variables.

To lash out at “white women” based on the CNN chart is to express hostility to the 39 percent of white women in Texas who voted for the Democrat yet get stereotyped with the rest of their cohort, while ignoring the 71 percent of white men, 39 percent of Latino men, 34 percent of Latino women, 16 percent of black men, and 4 percent of black women who voted for the Republican.

The principle at work: Let us judge them not on the content of their votes, but by the candidate who was backed by a majority of the people who share their skin color. In this way, Democrats turn on their own allies.

Some conservatives insist that performative, hyperbolic white-woman bashing is broadly representative of the Democratic Party and the political left. It is not. This rhetorical mode is widely seen as wrongheaded. In my experience, it elicits eye-rolling from most residents of deep-blue neighborhoods and from most Democrats in all racial groups. It is the work of a tiny, largely white, mostly privileged vanguard.

The extent to which that narrowness surprises you, as Wesley Yang once wrote in another context, “is a measure of how successfully the toxic rhetoric of warring elite cliques has gaslighted you into submitting to a narrative that is brazenly false.” Indeed, this mode of discourse is alienating to many who voted for Democrats, and obviously isn’t winning any converts, for reasons The Stranger’s Katie Herzog adeptly explains.

Still, this accusatory mode can’t simply be ignored. On social and digital media, where algorithms optimized for engagement boost views that are unusually anger-inducing, accusatory, and tribally divisive, it is overrepresented. The percentage of Democrats who buy its assumptions is tiny, yet it has the power to shape perceptions of the coalition, to derail its internal arguments, and to lead many astray about the truth.

As a non-Democrat, that final drawback is the one that bothers me the most. Set aside the moral case against disparaging groups in sweeping, stereotyped ways and its tactical foolishness. Blaming “white women” for progressive electoral losses causes people to lose touch with reality.


After all the ballots were cast in the 2018 midterm elections, the Pew Research Center reported on exit-poll results showing a gender gap in voting that was “at least as wide as at any point over the past two decades.”

Overall, 59 percent of women voters cast ballots for Democrats in House races. But that one-dimensional statistic is misleading in some ways. So exit polls broke down results based on race and education, too.

Among college-educated white women, 59 percent voted for Democrats. Among all white women, the split was 49 percent Democratic, 49 percent Republican. Ninety-two percent of black women and 88 percent of black men voted for Democratic Party candidates. The vast majority of black women voted for Democrats at all educational levels, at all incomes, and regardless of religious affiliation, making them the most reliable of all progressive constituencies. Race and gender are obviously factors that bear on elections and that demand our attention if we want to understand them fully.

But what sorts of generalizations do those figures justify? Certainly not these:

  • “It’s ‘the Year of the Woman,’ No Thanks to White Women” (Elle)
  • “Why Do White Women Keep Voting for the GOP and Against Their Own Interests?” (Vogue)
  • “The Betrayal of White Women Voters: In Pivotal State Races, They Still Backed the GOP” (Vox)

Alluding to GOP victors who won more votes from white women than did their rivals, Lyz Lenz, a writer at the Columbia Journalism Review, commented, “If any of your blue wave stories don’t include the fact that white women still uphold white supremacy then you aren’t doing your job.” Recall, only 49 percent of white women supported GOP House candidates; they were no less likely to support Democrats than Republicans. Even setting aside the assumed equivalence between voting GOP and upholding white supremacy—an equivalence no newspaper would or should assert as fact—this demands attributing to “white women” what fewer than half of them ostensibly did.

Shouldn’t coverage of the election be as precise as possible?

“Muting anyone who pulls a ,” Lenz declared in a follow-up tweet. Never mind that gets closer to the mark, or that the unqualified formulation that Lenz prefers makes it more difficult for readers to understand what happened.

Journalists aren’t doing their job when they lose sight of those people. Imagine a grocery-store clerk trying to organize a Democrat-aligned neighborhood group, or a first-time office seeker of color and a friend trying to help her gather signatures and raise funds. They see a group of white women, and wonder whether to invest any time in outreach. Which understanding better serves them? That “white women still uphold white supremacy”? Or that #NotAllWhiteWomen do? Surely it would be useful for them to know that the white women in question are more likely than not to reject Donald Trump and to support Democratic candidates if they are in California, or unmarried, or at a college-alumni event, or are Millennials.

#NotAllWhiteWomen would afford those Democrats a more nuanced grasp of reality that would prove hugely useful as they try to achieve their goals. Nevertheless, political discourse is rife with comments like this one:

White Women.

I swear to God.

Y’all are infuriating.

And before you #NotAllWhiteWomen me (which will get you blocked) tell me you are COMMITTED to flipping 10 of your alabaster friends from red to blue by 2020. It’s not enough to be “not racist.”

You must be *anti-racist.*

That we might agree is not enough. Emphasize our agreement and I will block you. You are responsible for the actions of people who share your skin color. But set aside that wrongheaded principle. Set aside the tone as irrelevant. Engage the substance of the admonition, as Jill Filipovic did.

She wrote this in a tweet thread:

I’m seeing a lot of calls for white women to come get other white women, and I agree … what that analysis misses is just how divided white women are, especially by education & location (rural vs. urban especially).

I look around my friends, family, and even acquaintances and I don’t really know women who voted for Trump or who support the GOP. That’s because my community is urban and educated (and diverse, but we’re talking about white women here). Maybe there’s a high school rando on FB?

Among the most retweeted and liked responses to Filipovic was this thread:

Don’t trust the white women who tell you they don’t know anyone who voted Trump because their friends are urban and educated. I can point you to 5 urban educated wealthy white women in their circle who are racist as fuck. Including them. Y’all keep acting like Trump isn’t straight out of Queens. Like f the “dapper” bigots supporting him don’t have Ivy League degrees & a place they keep in the city.

It’s not poor white people that are his base beloved. It’s the same rich white folks that drive past hunger daily. And white women that claim the problem is the lack of focus on white men? We know they’re a problem. We’ve been challenging them. The reason the focus is on white women is because they demand sisterhood and solidarity and stab WOC and their communities in the back. Repeatedly.

Y’all are passing around the tweets of @JillFilipovic like she wasn’t one of the primary triggers for #solidarityisforwhitewomen and I need you to ask yourself why so many white feminists are trying to pass the buck instead of step up and do the work.

In this telling, the educated, wealthy white women of New York City are surrounded by friends who voted Trump. In fact, Hillary Clinton won 86 percent of the vote in Manhattan, 79 percent of the vote in Brooklyn, and 75 percent of the vote in Queens; and she everywhere outperformed Trump among women with college degrees. The narrative blinds those who embrace it to uncontested facts.

As for Filipovic, one could hardly invent a character more credibly able to make such a claim. She holds undergraduate and graduate degrees from NYU. Her bio notes that she is “a journalist based in Nairobi and New York City, and the author of The H-Spot: The Feminist Pursuit of Happiness. A contributing opinion writer for the New York Times, she is also a weekly columnist for Cosmopolitan and CNN. Formerly a columnist for the Guardian and Cosmopolitan.com’s senior political writer.” Her profile fairly screams My personal bubble is devoid of Trump supporters.

Were she to dedicate 40 hours to trying to win voters for Democrats in 2020, canvassing in a heavily Latino neighborhood in any swing state would be a much better use of her time than talking to the educated urban white women she knows. Most anything else would be time better spent than that. Notice I am not complaining that she was forcefully challenged or that she was accused of being racist. The core problem is one of substance. The premise that she can be most effective in persuading voters by talking to the white women around her is simply mistaken.

Such errors are inevitable because the mode of discourse in question lends itself to domination by people who are not ultimately aiming for the truth. Insofar as they shape the Democratic understanding of the 2018 election and that coalition’s approach to 2020, they will add heat but subtract clarity.

Consider another social-media exchange––one that began when the institutional account of Vox.com tweeted: “The midterm election confirmed once again that black women show up for progressive candidates. But white women? Not so much.”

That caught the attention of my colleague Ron Brownstein. He replied that it is astonishing “how invariably critiques like this from left never mention the huge gap” between the voting behavior of college-educated white women and the very different behavior of white women without college degrees, especially since the former are indispensable to understanding Democratic gains in the 2018 midterms.

“As a liberal educated white woman, I’m aware of this,” one person retorted, “but honestly, white people in general need to get out of the way. Waving our hands in the air and yelling ‘But not me!’ isn’t the conversation we need to be having. Let’s talk to each other, not get defensive on Twitter.” She assumed Brownstein’s tweet was born of defensiveness.

To which someone else replied, “Thank you! Can black women get one second of praise without white people storming in with their ‘well actually’?” This person assumed Brownstein was denying black women their due.

Said another like-minded respondent, “It’s astonishing that Black women, college educated or not, historically vote for the good of the entire country. It’s not so astonishing that even some of the most liberal White people won’t credit Black women for their faithfulness and fortitude.”

Some background on Brownstein is useful here: He isn’t just a two-time Pulitzer finalist who has been covering national politics for 35 years; he is the rare sort of journalist who once won the Excellence in Media award from the National Council on Public Polls—an association of public-polling and media organizations whose mission is “to advance the understanding among politicians, the media, and general public of how polls are conducted and how to interpret poll results” and “to promote better understanding and reporting of public opinion polls.”

That doesn’t make him right in any given debate. But one could hardly invent a person who is more likely to be motivated by earnest interest in accurately conveying the most nuanced possible truth about voters and their motivations—it’s been among his recognized, award-winning professional obsessions for many years. He’s concerned with advancing clarity.

Yet even Brownstein, responding to a tweet from a site known for its wonky political coverage, is accused of discussing the subject in order to apportion social credit and blame, as his critics were doing, rather than to advance public understanding with facts. To such critics, discussing the degree to which white women did or did not vote for progressive candidates is not about adjudicating the truth of the matter. It is about scolding the privileged for their shortcomings and extolling the marginalized for unsung heroics. As a result, they mistook and mischaracterized truth seeking as racist villainy. And if a professional journalist like Brownstein can’t be granted the benefit of the doubt, what hope does an average person seeking to advance clarity have?

Ordinary people who want to understand the politics of the moment are confused as to why individuals would get credit or derive blame from how others who share their skin color voted. They can’t understand why the literal truth of “Not all women” is no justification for saying it (or why progressives must preemptively apologize for meekly clarifying a fact). They are less able to parse the truth, because of the opaque, ever-changing language codes of educational and cognitive elites.


Imagine if one were to say “Women are Democrats––59 percent of women cast ballots for Democrats in 2018 House races.” And imagine a critic’s reply: “Fewer than half of white women voted for Democrats, while a far higher percentage of black women voted for Democrats. Eliding and neglecting the distinction paints an incomplete, misleading picture.” Imagine retorting, “Stop it with your #NotAllWomen stuff.”

That would be silly. The intersectional insight that black women support progressive candidates at much higher rates than white women adds to our understanding of reality—as does the insight that college-educated white women support progressive candidates at very high rates and that many sorts of white women are highly reliable Democrats.

Going beyond gender adds useful information. So does going beyond race and gender.

That’s the irony: At the bottom, the mode of discourse critiqued in this article is a failure of intersectional thinking, and just a step removed from failures of intersectional thinking that its adherents find obvious and infuriating.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.