The COVID-19 pandemic would seem tailor-made for the world's fact-checking organizations to mitigate the spread of falsehoods at a time when bad information can cost lives. But as social media platforms have largely turned inward, deciding for themselves what is true or false, fact checkers have been relegated to quibbling over minutiae.
This failure to meet the moment comes as free-wheeling daily White House press briefings and other officials’ pronouncements demand objective, evidence-based information in response. Instead, fact checkers seem to have settled on reviewing the semantics of the coronavirus debate.
For example, PolitiFact assigned a score of “Half True” to Steve Bannon’s claim that Dr. Anthony Fauci had assured the American public back in January that the disease would not pose a major threat to Americans. Despite conceding that Fauci had made the statement attributed to him by Bannon, the fact-checking site rated the claim only Half True because Fauci had noted that things could change later on.
Similarly, the site rated as “False” a recent Facebook post that reproduced previous CDC guidance that Americans not wear masks. While again acknowledging that this was official CDC guidance until recently, PolitiFact nonetheless rated the post as simply False because that guidance has now been superseded, rather than assigning a rating such as “Outdated” to acknowledge that the information was previously correct.
The site also rated as “Half True” Donald Trump’s claim that new mask sterilization technology permitting up to 20 reuses of an N95 mask is equivalent to having 20 times the supply of masks. While it is correct that mask sterilization results in degradation and is suboptimal, the core of Trump’s argument -- that the practice results in, effectively, a 20-times increase in mask access during a health emergency -- is correct. Yet, while allowing Fauci’s caveat to render his statement as true, PolitiFact ignored Trump’s qualification that sterilization is “like” ordering 20 times more masks and treated his statement as a literal claim that reused respirators were identical to new ones.
In another contrast, when the governor of Wisconsin claimed that his state’s mitigation measures had saved 300 to 1,400 lives, PolitiFact rated this as “Mostly True,” citing theoretical models hypothesizing the number of deaths that would have occurred without those measures. Yet, the sheer number of uncertainties about the pandemic, including critical model inputs such as infection counts, means many of these models have proven inaccurate. A more accurate rating for this claim would have been “Unknown” since it is simply unknowable whether a failure to implement those measures, or only some of them, would have yielded a higher fatality rate in Wisconsin.
In the end, with fact-checking sites engaged in these pedantic debates, the weight of combating public health falsehoods that could cost lives has fallen to social media platforms. But instead of deferring to health officials, those platforms have largely taken it upon themselves to decide what constitutes “truth.”