“The mistake involves grouping together into an all-or-nothing package entire sets of claims whose epistemological credentials are quite varied. It also often involves collapsing epistemic and moral concerns.”

In the following guest post, Eric Winsberg, Professor of Philosophy at the University of South Florida and British Academy Global Professor of History and Philosophy of Science at the University of Cambridge, discusses why he has become “extremely skeptical of recent philosophical work, both in social epistemology and in philosophy of science, around such topics as misinformation or fake news.”

This is the third in a series of weekly guest posts by different authors at Daily Nous this summer.

(Posts in this series will remain pinned to the top of the homepage for several days following initial publication.)

[Robert Longo, drawings from the series “Men in the Cities”]

Misinformation Mistakes
by Eric Winsberg

I recently reviewed a book by Paul Thagard called Falsehoods Fly: Why Misinformation Spreads and How to Stop It for the Journal of Value Inquiry. (It got some uptake here.) The main reason that I was eager to review this particular book is that I have become extremely skeptical of recent philosophical work, both in social epistemology and in philosophy of science, around such topics as misinformation or fake news. If readers are interested in the topic, I urge them to go and read the review before they join the discussion of this topic below. This isn’t just because I want the clicks. It’s because to understand and appreciate the root of my skepticism requires working through some of the (admittedly) tedious scrutiny that I subject some of Thagard’s claims to. The book covers a number of topics, including the war in Ukraine, inequality, and political conspiracy theories like Pizzagate, but I focus primarily on the chapters on the Covid-19 pandemic and on climate change, both because those are areas about which I have some (admittedly limited) expertise, and because those two topics share a pair of features that I think aggravates a wound in the misinformation discourse in philosophy.

What pair of features do the Covid-19 pandemic and climate change share? First, both subject areas are founded on certain well-credentialed propositions that are nonetheless doubted by a highly vocal minority. Covid-19 was a real virus that killed millions of people. Carbon dioxide is a heat trapping gas that has contributed to some of the observed warming of the planet over the last century and, if emissions are unabated, will continue to warm the planet into the future. Some of the people who doubt the truth of these propositions have bad intentions; some of them are simply confused.

If Thagard’s book is at all representative when it comes to claims about misinformation (and I think it is), people often want to do quite a bit more than condemn people’s denial of these epistemologically secure claims. This points us to the second feature that the Covid-19 and climate change share. In both subject areas, people like Thagard want to condemn people for denying claims that are not at all epistemologically secure. These include claims, for example, to the effect that SARS-CoV-2 has a natural zoonotic origin; that all the Covid-19 vaccines stopped the spread of the virus, and provided overwhelmingly more benefit than risk to everyone; that the models that made projections of the likely impact of adopting, or failing to adopt, various non-pharmaceutical interventions (NPI) always reflected the best available evidence; that climate change, if left unchecked, will kill a net 50,000,000 people from malnutrition, disease, and heat stress between the years 2030 and 2050.[1] My view is that all of these claims are uncertain and at least some will prove to be false. (If you are keeping score, the two I’m fairly confident are false are the one about the NPI models and the one that all the vaccines provided a net benefit to everyone, including previously infected, healthy children.)

In the review, I show that in discussing these topics, Thagard makes a very large number of factual mistakes. Indeed, Thagard often gives an argument that goes, with respect to a given claim C, “only sloppy, motivated reasoning could lead anyone to doubt C, because reasons x, y and z”, but where at least some of x, y and z are contradicted by matters of record. To give one example, Thagard claims that “[clinical] trials have found that vaccines, produced by Pfizer, Moderna, AstraZeneca, and other companies, are effective at preventing the spread and reducing the severity of COVID-19”. But while it’s obviously true that all of these trials showed that vaccines were effective at reducing the severity of Covid, not a single clinical trial ever showed that any Covid-19 vaccine was at all effective at preventing spread. Indeed, the balance of real-world evidence strongly suggests that they don’t. (If you doubt either of these claims, please see the review for the receipts.)

The reason that I tediously compared so many of Thagard’s claims to the best available evidence was to make a general point. And the general point is that a lot of the empirical claims that people fight over when it comes to policy decisions (like whether to enact mask mandates, impose lockdowns, mandate Covid-19 vaccines for children and young adults at university, or support various climate mitigation or adaptation measures) are actually hard to get to the bottom of, and, ultimately, uncertain. I worry that this is underappreciated by many of the people who work on misinformation, especially those (like Thagard) who think the solution to misinformation lies in censorship, regulation of social media companies, or even the application of consumer pressure on social media companies to do the regulation voluntarily.

I believe that many of the parties who contribute to this line of thinking are making a fundamental mistake. The mistake involves grouping together into an all-or-nothing package entire sets of claims whose epistemological credentials are quite varied. It also often involves collapsing epistemic and moral concerns. To give just one example, consider the claim that the Measles/Mumps/Rubella vaccine has not saved lives. We should all be extremely confident that this claim is false. And we should all be concerned that people who make this claim are guilty of moral wrongdoing. When these convictions kick in, there’s a natural moral-epistemic heuristic that leads many to assume that anyone who is even in the vicinity of these claims is equally factually wrong and equally morally reprehensible. Suddenly, the conviction isn’t just that MMR vaccines have saved lives, it’s that every single Covid-19 vaccine provides a net benefit to everyone, in any number of doses. Or, even, simply that there is incontrovertible evidence that climate change will take the form of a specific disaster by a specific time. We saw this over and over during the pandemic, where people equated skepticism about any particular Covid mitigation policy with “Covid denialism.”[2]

To some degree, I understand this moral-epistemic heuristic, and I’m sure that I sometimes employ it. But what I believe quite strongly is that it is not the job of philosophers, especially philosophers of science and social epistemologists, to try to find a philosophical foundation for this heuristic as if it were actually a secure way to reason. It’s not a good way to reason and there are no shortcuts for figuring out what policy-relevant facts are true by consulting the tribal affiliations of their exponents. And if Thagard’s book is any guide (and I think it is), this is exactly what much of the philosophical literature on misinformation is up to.

(Addendum: Justin specifically asked me to comment on the work of Cailin O’Connor and James Weatherall on misinformation, pointing me to this piece they wrote in the Boston Review. Justin noted that what they say about the pandemic is not that different than what I say. They wrote, for example “This legitimate uncertainty means that pundits and journalists who treat claims supporting hydroxychloroquine [and other controversial claims during the pandemic] as akin to typical misinformation are misdiagnosing the situation.” So, one thing I want to clear up is that I did not say, above, that all of the philosophical literature on misinformation is up to this. Indeed, O’Connor and Weatherall never say anything about what “typical misinformation” is.  And more importantly, they are not doing what Thagard and many other misinformationists are doing: purporting to provide simple tools for identifying misinformation and making recommendations for how it should be eliminated. It’s noteworthy that reviewers often comment that readers are likely to find this aspect of their book “disappointing”. I would actually argue that their work has little to do with “misinformation” as it is commonly understood, as it is instead simply a book about how falsehoods of any kind persist. But it is clear that Thagard doesn’t equate misinformation with falsehood. He thinks truths can be misinformation, and he thinks some misinformation is a particularly morally loaded kind of falsehood. And it is easy to see by perusing the philosophical literature on misinformation that most of the contributors think misinformation per se needs an analysis, and that this is a novel topic relative to the analysis of truth and falsehood that goes back at least to Aristotle.)

[1] See the review for evidence that Thagard condemns people either explicitly for denying these claims or for making value judgments that implicitly assume these claims are known with certainty.  (For example he condemns people for valuing corporate profits more than 50,000,000 lives, rather than considering the possibility that they consider the claim uncertain, or that they view the cost benefit analysis in different terms.)

[2] Indeed even in the last few days, “Philosopher Jersey Flight” on Twitter/X said my review was “like the people trying to argue that Covid wasn’t real”.