Wait, what? Are beloved pets in Ohio really at risk? In last week’s presidential debate, many Americans first heard a conspiracy theory that Haitian immigrants in Springfield, Ohio, were killing and eating people’s pet cats. During the nationally televised debate on Sept. 10, former President Donald Trump claimed:
In Springfield, [immigrants are] eating the dogs, the people that came in, they’re eating the cats…. They’re eating the pets of the people that live there.
The bizarre, dehumanizing claim – though debunked by Springfield officials – had been circulating on social media for a few days. Moderator David Muir informed Trump and the audience that ABC News had also fact-checked the rumors by speaking with city officials in Springfield. As Trump and his running mate JD Vance continued to amplify the story after the debate, the small Ohio city faced several threats against local schools and government offices.
Misinformation stories like this spread quickly – and not just in the United States. To consider the episode in a comparative context, Good Authority contributor Christopher Clary spoke with three political scientists researching the origins and impact of rumors and misinformation: Sumitra Badrinathan of American University, Simon Chauchard of the University Carlos III of Madrid, and Niloufer Siddiqui at the University at Albany, State University of New York.
Christopher Clary: Trump’s odd answer during the debate with VP Kamala Harris about immigrants eating pets in Ohio reminded me of false rumors about minorities in your recent research in India and Pakistan. What struck you about Trump’s claims and how it fits into political scientists’ thinking about misinformation and rumors?
Simon Chauchard: Several things, actually. The first is that this rumor targeted a minority group – and a rather specific one at that. Whether in the United States, in India, or elsewhere, minority groups who are extensively “othered” in public discourse become targets. The second is that rumors about outgroups or minorities typically touch on deeply emotional issues, or on topics that can quickly generate a maximum reaction. Americans love their pets, so a threat to cats is great fodder for a rumor.
A third point, of course, is that the amplification of the rumor before, during, and after the debate by candidates seeking the highest office of the land highlights the central and nefarious role that political elites tend to play in making rumors go viral. We can assume that malevolent people generate outlandish rumors about groups they don’t like pretty much all the time and everywhere, in the darkest corners of the internet or even just offline. These rumors have low virality and limited reach – until political elites in places as diverse as Western Europe, Sri Lanka, or Rwanda mention them, and the rumor becomes a headline.
And a final point that struck me is that this particular rumor is actually not as easy to dismiss or counter as its outlandishness would imply.
ABC News opted to fact-check Trump live at the debate. How should we think about that decision in the context of what we know about misinformation?
Sumitra Badrinathan: There’s plenty of evidence now that fact-checking and correcting misinformation is generally effective. Studies have found that fact-checking can help change misinformed beliefs, even for people with more entrenched views. And this holds true across different countries and contexts. It’s also the most widely used method for combating misinformation, both in research and in practice. Plus, fact-checking is more cost-effective than other approaches, like investing in education or digital literacy programs, so I think ABC’s decision to fact-check live during the debate was the right call.
We also know from political science research that it’s easier to change people’s minds when their views aren’t too deeply entrenched. So, kudos to ABC for fact-checking Trump live, while millions of Americans are watching, rather than waiting until after the debate. That said, fact-checking does have its downsides. It’s tough to reach everyone who initially read the misinformation on social media. And, crucially, not all content is easily fact-checkable in the first place.
Niloufer Siddiqui: In addition, the big problem with the type of vague and generic claim that Trump made (“immigrants are eating our pets”) is that there is no specific incident to refute. This makes the claim much more difficult to verify. Our research has found that this type of rumor is very common across multiple contexts. For example, in India, numerous rumors during the covid-19 epidemic claimed that Muslims were spitting in food in order to spread the illness and make Hindus sick. While one could say that there is no evidence of such an act, it is of course impossible to prove that no Muslim has ever spit in someone else’s food.
In other cases, rumors may be unverifiable because they are about an individual’s private intentions. A good example here is the “love jihad” conspiracy theory that Muslim men in India marry Hindu women to convert them to Islam. Anti-Muslim groups in India – and the ruling BJP party – have repeated this claim. And Muslim men who married Hindu women have faced violent attacks.
Another example, in Pakistan, are rumors about blasphemy – a crime punishable by death. These rumors sometimes rely on claims that one individual made a blasphemous comment to someone else, so refuting the accusation relies on accepting someone’s word. This is important because if anti-minority rumors are not easy to fact-check, or if fact-checking efforts only address a subset of the claims made, factual verification and correction will have a limited impact.
Have you watched this post-debate interview with Republican vice presidential candidate JD Vance? Vance defended Trump’s answer, saying that while the specific claim might not be verified, the larger truth implied by the claim had merit. What stands out to you about Vance’s statement?
Niloufer: Vance’s answer was fascinating for a number of reasons. Perhaps most striking is his suggestion that there are gradations of truth – that a rumor can be partially true. He’s claiming that ultimately whether a specific rumor is true is irrelevant to the broader narrative that is being constructed. Vance also appears to suggest that political elites sometimes rely on these outlandish rumors (which we can never truly be sure are false) to draw attention to a broader issue – in this case, he pivots away from migrants eating cats to claim migrants are pushing up housing and health care costs in Ohio. We contend that the majority of anti-minority rumors propagated by elites in India and Pakistan rely on a mix of false, true, and unverifiable claims. And these claims aim to create a narrative targeting a particular minority or outgroup.
Sumitra: Yes, Vance defends Trump by suggesting that even if the specific claim isn’t verified, the larger “truth“ behind it still holds merit. This kind of misinformation is especially tricky to fact-check – and dangerous. It’s tricky because, as we’ve mentioned, it’s nearly impossible for fact-checkers to definitively say that something has never happened. Vance seems to understand this and uses it to his advantage.
In spreading this lie, Vance stereotypes an entire group, generalizing from one hypothetical incident to imply that, if it were true, the entire group must be inherently bad. This deepens polarization, which can be dangerous. And it shows how misinformation can potentially lead to real-world violence. As with our recent research showing that misinformation is tied to vigilantism in South Asia, some journalistic reports are already mentioning vigilante violence targeting Haitian immigrants in Ohio.
While we know it’s possible to correct misinformation to a certain extent, it’s much harder to reduce polarization or shift attitudes towards minorities, especially when such negative views become deeply ingrained.
I understand your work on misinformation and rumor in India and Pakistan is ongoing. Is there anything else about that work that seems relevant in light of this episode?
Sumitra: Our academic project identified that correcting misinformation can have downstream effects – these efforts can reduce support for vigilante violence. Stemming from this finding, we now want to explore the broader effects that correcting misinformation might have on other democratic norms.
One of our upcoming projects will examine the relationship between misinformation and support for protest, free speech, the rule of law, and democratic backsliding in general. Although the connection between misinformation and democracy seems obvious, there is still no solid causal evidence to prove the link. Another area we’re focusing on is finding ways to correct rumors that aren’t easily fact-checkable, because they lend themselves to deep stereotypes or narratives that vilify entire groups, in line with the more general anti-immigrant claims JD Vance pushed after the debate.
Simon: We are currently in the process of designing a field experiment in India to investigate how to reduce people’s belief in just these types of rumors – ones that are composed of half-truths or vague, generic, unverifiable statements. We are testing a new set of interventions that we call “narrative deconstruction” in which we challenge the underpinnings on which the set of falsehoods is based. By deconstruction, we mean providing information on (1) the supply side of disinformation, including how majoritarian elites profit from disparaging minority groups and stoking polarization; and (2) historical, factual information that seeks to dislodge longstanding but incorrect narratives that may fuel belief in present falsehoods.
We will keep Good Authority posted on this intervention and our research!
Stay up to date on all things politics and political science. Sign up for Good Authority’s weekly newsletter by entering your email address in the box below.
