The New York Times has a Twitter problem. More specifically, Twitter’s problems are now the New York Times’ problems. In this news-filled week, marked by several scoops from Times reporters, the paper has found itself on the receiving end of voluminous criticism for two of its stories: one revealing that the whistle-blower who has triggered impeachment proceedings against Donald Trump is a CIA officer, and another in which a reporter inaccurately framed a story as being about how “swing voters” had reacted to the prospect of impeachment.
I will address both in due course, but stepping back just a bit, this has been something of a pattern for the Times. Its op-ed page, in particular columnists Bret Stephens and Bari Weiss, has been the butt of Twitter memes and jokes since the 2016 election. But in recent months, it’s been the news side that has found itself checked and double-checked by tweets.
Though executive editor Dean Baquet told his staff that he won’t allow the paper to be edited by tweets (in a town hall held in the aftermath of another Twitter explosion, which I previously wrote about), responding to Trump’s favorite social-media platform is taking up more and more of the Times’ mental space.
It isn’t all Twitter’s fault, of course. The Times can be ham-handed in the way it responds to reader outrage, and the whistle-blower story is a good example. Hours after his complaint was released on Thursday by the House Intelligence Committee, many on Twitter echoed the concerns raised by the whistle-blower’s attorneys in the original report: “Publishing details about the whistleblower will only lead to identification of someone, whether our client or the wrong person, as the whistle-blower.… This will place this individual in a much more dangerous situation, not only in their professional world but also their possible personal safety.”
Related: “I had to rewind this to believe what I was hearing”: The media’s missteps in the Ukraine scandal
Initially, Baquet responded in a way that emphasized the news value of the scoop and not the potential risks of outing the individual: “The role of the whistle-blower, including his credibility and his place in the government, is essential to understanding one of the most important issues facing the country,” he said in a statement that was included in the piece. Yet, the most salient point he had to add about whether the Times had endangered the whistle-blower was appended only much later on Thursday night: “We also understand that the White House already knew he was a C.I.A. officer.” The lede of the main story was subsequently updated to indicate that the White House had been aware a CIA agent had lodged a whistle-blower complaint before that information was public.
The additional information barely slowed the criticism on Twitter— in part because, during a fast-breaking story, the torrent of tweets and updates from the Times can be hard to track. That means readers are often reacting to old or mistaken information, or simply are not aware of new developments in an active reporting process.
(The Times could do better on that front, too. The current version of the story on the Web as of this writing bears little resemblance to the version originally posted, and yet the only sign of that to readers is a time stamp: “Updated Sept. 27, 2019, 9:53 a.m. ET.”)
“The main identifying information in the piece,” Baquet told me in an interview on Friday, “was the fact that he worked for the CIA. It is pretty clear that the White House knew he worked for the CIA.” He also admits that they got a little lucky, adding, “By the way, I wasn’t absolutely sure of that when I made the judgment. I strongly suspected it, given that he ended up going to the IG [inspector general] for the intelligence community.
“The main reason I decided to say he was CIA—as did the Post and the Journal, by the way—is that this was important information for people to assess his veracity. He was not, as the White House described him, a political hack. He was an employee of a nonpolitical agency.”
On Friday morning, one Times reader—one of many who reached out to me via Twitter (where my DMs are open), and a journalist himself—was upset with the paper calling the story “a hugely important issue.” When I pointed out the new statement from Baquet, he said he hadn’t seen it, adding, “If true that would change things in my view.”
The outpouring of 280-character rage at another story on Wednesday, about how voters were reacting to the opening of impeachment proceedings, was likewise rooted in Twitter’s encouragement of superficial debates.
This one was started by a tweet (since deleted) from national correspondent Sabrina Tavernise, one of three bylined reporters on the piece. It read “I talked with six swing voters today. Impeachment repelled every one of them.”
The tweet drew fire from liberals who have been pushing for impeachment of Trump and railing against the argument that it would alienate moderate voters. But it also attracted more substantive criticism, most notably from Matt McDermott, a pollster, who interpreted Tavernise’s tweet as summarizing the story—“No, the New York Times did not talk to six ‘swing voters’ about impeachment”—pointing to people quoted in it, like Trisha Hope, who said she had been to twenty-three Trump rallies. “She’s a Trump fanatic,” McDermott said, “not a swing voter.”
The problem, though, is that the story never purported to be about “swing voters,” just “voters” of all partisan stripes. I did a quick analysis of how the twelve voters quoted by name were described, and it came down to five Republicans, five Democrats, one independent — the only voter quoted by name who Tavernise had interviewed — and one whose affiliation was not specified. Of these, three might be considered “swing voters”; all had voted for Trump in 2016, two had voted for Obama in 2008, and one was described as “wavering” on voting for Trump in 2020.
Julie Bloom, deputy national editor at the Times, said in a statement, “The piece was not focused on swing voters. We are proud of the story we published capturing the voices of voters across America and across the political spectrum.” She also offered this context for Tavernise’s tweet: “She has been talking to many voters since before the inaugural and has regularly checked back in with them. In her tweet she described her reporting and what she found. While she spoke to six swing voters, other reporters…talked to many people, and not all of those interviews made it into the story.”
There were ten reporters credited on the piece and it notes that they spoke to “more than two dozen voters across the country.”The voter interviewed by Tavernise was Donna Burgraff, who is described as an independent who voted for Trump in 2016 after voting Obama in 2008. As McDermott noted, Burgraff had previously been quoted in the Times saying she planned to vote Republican in the 2018 midterms.
Though the Twitter criticism of the individual story was misplaced, it still raised some valid critiques of Times political coverage. For instance, putting forward a dozen random voters as representative of the entire public doesn’t offer much value to readers and is a newspaper habit that’s worth retiring. Each political era creates a voter type that is subsequently fetishized by reporters—people found in New Hampshire diners, “soccer moms,” or, today, the “white working-class” voters who switched from Obama to Trump. Their numbers and influence alike have been exaggerated. And there’s a definitional issue of what a “swing voter” even is: Is it someone who’s unsure how they’ll vote, like the lone “waverer”? Or does someone like Burgraff, who voted Democrat twelve years ago and has consistently voted Republican since, still count?
Given that Twitter is terrible for the public discourse—it promotes shallow, hasty, extremist discussion that soon becomes polluted by misinformation—perhaps neither we nor the Times should see it as a good platform for answering those questions or for offering feedback to one of the most important newsrooms in America.
I asked Baquet how valuable he finds the feedback he gets from Twitter to be. He was typically diplomatic. “Twitter provides important feedback. Of course there are people on Twitter who don’t like the Times and are not criticizing us in good faith. But there are also many people who provide thoughtful feedback, and who criticize us because they want us to do better.”
That’s gracious, bordering on non-commital. But it may be time for those of us who want the Times to do better to stop relying on a platform as flawed as Twitter to deliver our critiques.
Correction: an earlier version of this story referred inaccurately to an editor at the Times, Julie Bloom. She is deputy national editor, not deputy editor for politics.
Related: How I missed the Ukraine story
Editors note: CJR has appointed its own outside public editors for four vital news outlets — The New York Times, The Washington Post, CNN and MSNBC — that currently lack any public ombudsman. You can reach them at email@example.com. (Any messages will be treated as off-the-record unless otherwise agreed.)Gabriel Snyder is a contributing editor to CJR.