Why Would I Possibly Trust Bill Adair Or PolitiFact To Determine Which ‘Misinformation’ Should Be ‘Suppressed’?

If their performance covering the unrest in Kenosha is any indication, they have no grounds for stone-throwing

Nieman Lab recently ran a roundup of year-end predictions for 2024, and Bill Adair argued in one of them that “Fact-checking needs a reboot.” Adair, his bioline at the end of the article notes, “is founder of PolitiFact and the Knight Professor of the Practice of Journalism and Public Policy at Duke University.”

The first sentence argues that “Fact-checking is failing.” Adair continues: “The old way of publishing fact-checks — putting them on websites and promoting them through social media — isn’t getting them to the people who need them. It’s time to reimagine how fact-checkers publish and broadcast their work.”

You can read the article if you want to get his full argument. He makes some perfectly fair points about how, for example, state legislators are subjected to very little of the fact-checking that sometimes covers higher-profile politicians.

But I’d like to focus on one part that really jumped out at me:

After I founded PolitiFact in 2007, I often said that our goal wasn’t to change people’s minds or get politicians to stop lying — it was simply to inform democracy. In the last few years, I’ve changed my mind. “Informing democracy” is not enough in an age of rampant lies about elections and public health and climate. Fact-checkers need to be more assertive in getting truthful information to the audience that needs it.

In 2024, they will dream up new ways of getting the facts to the people who need them. Fact-checkers will be bold and think more like marketers trying to push content rather than publishers waiting for the audience to come to a website. They will experiment with new forms that target the people who are misinformed and push the content directly to them.

Another way they will innovate: They’ll get tech companies and social media platforms to expand the use of fact-checking data to suppress misinformation. My Duke team helped develop ClaimReview, a tagging system used by most of the world’s fact-checkers. Tech companies such as Google use it to identify fact-checks and highlight them in search results and news summaries. But this is just a start. ClaimReview and MediaReview, a sibling tagging system for fact-checks of videos and images, can be used more widely to suppress inaccurate content.

I suspect that by “in the last few years,” Adair means something like “since Donald Trump was elected.” After all, countless journalists have decided since then that they need to be a little bit more “assertive” about what they do — that they cannot merely “inform democracy.”

As I’ve argued before, from a certain perspective it all sounds well and good — until you dig down into the specifics, into how journalists actually execute on their new supposed imperative. In this case, it’s telling that a journalism professor would speak this breezily about “suppress[ing] misinformation” without any critical inquiry into what that might mean.

This article fits neatly into a recent obsession, in some liberal circles, with the idea that the United States can fact-check its way out of various social ills and political crises. This has brought with it some fairly ominous-seeming ideas about what a “fact” is and who gets to “check” it. In this case, Adair offers no hint as to how we should determine what crosses so far into “misinformation” that it should be suppressed. Not “corrected,” not “contextualized via a note from an editor” — suppressed!

Even though he doesn’t say so, it stands to reason that as anti-misinformation technology improves (or “improves”) and more misinformation (or “misinformation”) is suppressed, it’ll be the Bills Adair of the world and their PolitiFacts that get to determine what is sufficiently untrue as to warrant suppression. This raises an obvious question: How have they done so far?

One relatively easy way to evaluate the quality of fact-checking organizations is to judge their past performance. Specifically, their past performance on the most politicized, incendiary cases. If a fact-checking organization can keep a steady eye on its principles even in these instances, that’s a good sign that it is functioning well and can be trusted.

I continue to think that the chaos that unfolded in Kenosha, Wisconsin, in August 2020 was a vitally important test case for mainstream American journalism. To review: first, Jacob Blake was shot seven times by Kenosha police officer Rusten Sheskey on August 23, sparking both peaceful protests and rioting. In the chaos that ensued, Kyle Rittenhouse, a teenager from a nearby town in Illinois who styled himself a guardian of the increasingly dangerous streets, shot three individuals, two fatally, late in the day on August 25.

Despite all the outrage over Sheskey’s and Rittenhouse’s actions, the available evidence suggests that Sheskey’s use of potentially lethal force against Blake on August 23 was almost certainly justified, legally speaking, and that Rittenhouse almost certainly acted in self-defense (also legally speaking) as well, since the men he shot were pursuing him. Therefore, the Kenosha County District Attorney was correct when he announced he wouldn’t pursue any legal charges against Sheskey in early 2021; the Department of Justice was correct when it announced it wouldn’t pursue federal charges nine months after that; and later, a jury was correct to acquit Rittenhouse on charges that included intentional homicide.

I’m not going to rehash these cases in detail here, but to be honest I don’t think either of them is much of a close call, and I’ve written and said plenty about both for readers who are curious to know more: here’s what I wrote about the Rittenhouse case in the early days of that controversy; here’s an episode of Blocked and Reported Katie Herzog and I dedicated to carefully going through the Jacob Blake case; here’s a piece I wrote for Persuasion about how the not-guilty verdict shouldn’t have surprised anyone following the Rittenhouse case closely (and how badly journalists screwed it up); and here’s an episode of Honestly in which I discussed the verdict with Bari Weiss.

Why would I choose these particular events as a quality-control test for journalists and their outlets? Partly because it’s convenient for me — I know a fair amount about the events in question. But beyond that, these events constitute fairly clean means of assessing journalistic competence because 1) they were ultra-polarizing, meaning that during the heady days of 2020 and 2021 there was tremendous pressure for partisan actors and institutions to express the “correct” opinions about them (and there still is, to a lesser extent); 2) we have a lot of useful video about what actually happened in both instances; and 3) Blake’s shooting sparked two investigations (both the DA’s and an independent one by Noble Wray, a law enforcement veteran and former Barack Obama police-reform appointee, which the DA relied on in part to make its decision), while Rittenhouse’s led to a court case, and both of these processes produced yet more evidence.

In other words, journalists had (and have) a great number of social and political reasons to get Kenosha wrong, but few excuses to do so when it comes to the available evidence, which is not in short supply. It’s a low bar to clear and professional journalists should have been able to hurdle it from the start.

PolitiFact has basically failed this test.