Fighting Disinformation is not Censorship
Labeling “disinformation” as “censorship” is a lazy and inaccurate, but effective, frame.
It’s a grim time to be a disinformation researcher. There are the legal battles, including Biden vs. Missouri, now before the US Supreme Court, in which the 5th Circuit ruled that the Biden administration violated the First Amendment by collaborating with social media companies and to identify demonstrably false and harmful information online. Elon Musk is suing the Center for Countering Hate, a British nonprofit that works to combat health, climate change, and antisemitic disinformation. The Gateway Pundit, a hyper-partisan conservative website, is suing Stanford and the University of Washington for identifying them as spreaders of election disinformation. And Representative Jim Jordan is systematically subpoenaing disinformation researchers from major universities, claiming that their concerted efforts to combat election and COVID-19 disinformation constitute a “censorship regime.”
These attacks are prompted by Republican claims that disinformation researchers are dangerous partisans attempting to censor conservative speech with the government’s help. The attacks have been successful, shutting down grant programs at the National Institute of Health, hampering the efforts of organizations like the Election Integrity Partnership, and contributing to a slowdown in anti-disinformation efforts at major social media platforms. While this frame has circulated in conservative circles for years, the current campaign is fueled by the right-wing Foundation for Freedom Online, whose website states that “disinformation studies is academic-speak for censorship.” The FFO is run by Mike Benz, a former State Department official who claims that efforts to combat disinformation are turning the US into what he calls “Greater West Pakistan,” driven by a desire to “override what people say on the internet, to stop people from talking about narratives that might mobilize into political action.” The FFO recently criticized the National Science Foundation for funding research on efforts to stop online misinformation.
I am a disinformation researcher. I’ve been studying internet communication for twenty years and disinformation since 2016. I do not want to “end the Constitution'' (another favorite phrase of Benz) nor do I have any desire to enact censorship. (I’m extremely opposed to banning books with LGBTQ+ characters in public libraries, for instance.) I also know all too well the harms of spreading incorrect and hateful information on the internet. Saying that COVID-19 is a government plot and that vaccines kill people is simply not true, and contributed to vaccine hesitancy, resulting in 300,000 unnecessary deaths. Claims that Joe Biden stole the 2020 election led to the 2021 assault on the Capitol. QAnon’s preposterous theory that Democratic elites are a cabal of Satanic pedophiles prompted an array of violent incidents. “Disinformation,” unlike “fake news,” has a specific meaning: false or misleading information strategically spread for political or ideological goals, to create harm, or for profit. Of course, many people spread false information online that they believe is true, but that doesn’t make the information any less false.
The problem is that there is simply far more online right-wing disinformation than left-wing, which academics call “ideological asymmetry.” There certainly is left-wing disinformation– our team considers the notorious Trump “pee tape” story to be disinformation– and there’s even more false content online that has nothing to do with politics at all. But it is dwarfed in the public eye by right-leaning content that is provably false. Copious studies have also found that conservative Americans are more likely to create and spread disinformation. There are a plethora of reasons for that, including the obvious– more right-wing disinformation means right-wing users are more likely to spread it–but right-leaning users are also more likely to spend time in “echo chambers” online. As well, right-leaning media outlets frequently source stories from partisan publications that don’t adhere to journalistic norms of fact-checking, while conservative American politicians like Donald Trump and Marjorie Taylor Greene frequently support, create, and promote disinformation.
Disinformation researchers and social platforms are fighting an uphill battle against false online content. There are virtually no US government regulations on what can be posted to social media, except things that are already illegal, such as child sexual abuse material and content created by foreign terrorist organizations. Social platforms are private companies who are not bound by the First Amendment, and have the right to regulate content in any matter they see fit. As a result, these companies’ internal Trust & Safety teams are responsible for moderating hundreds of millions of pieces of content posted from around the globe per day. This moderation is extraordinarily difficult and regularly angers people across the political spectrum, from arguments over breastfeeding videos on Instagram to whether Facebook should remove a poem comparing Ukrainian soldiers to Nazis. The government rarely intervenes in such cases; rather, it depends on such moderation. As Bennett Clifford at the George Washington Program on Extremism writes, “the U.S. government has effectively outsourced its online counterterrorism responsibilities to major social media companies.”
This can, unfortunately, lead to, “jawboning” - behind-the-scenes government efforts to pressure social media platforms into changing content moderation- which is more than a little shady. It’s very clear that a wide swath of government employees “repeatedly and routinely” jawboned employees at Meta, Twitter, Google, and TikTok, asking them to ban content, allow content, change internal policy, use or stop using various technologies, and so on. This is, to put it mildly, problematic. Former Facebook public policy leads Matt Perault and Katie Harbath argue that jawboning is both antidemocratic and makes it harder for platforms to make good policy, since pressure often happens during moral panics or in response to specific events. It’s also extremely important to remember that actors on both sides of the aisle engage in jawboning. You might strongly agree with the Biden administration’s bids to prevent Covid disinformation but strongly criticize social media platforms who were pressured by conservatives not to remove “Stop the Steal” content. More transparency across the board, or explicit regulations on what different government bodies can and can’t do, might move such backroom negotiations to a more level playing field.
However, there are many other instances in which the government works alongside private, public, and civil society organizations to combat terrorism, improve public health, and ensure the stability of national institutions like the Census– often in collaboration with academics. For example, the Department of Homeland Security, the FBI, and the National Counterterrorism Center regularly conduct threat briefings with social platforms to plan for and prevent terrorism and targeted violence. The Census Bureau has its own Trust and Safety team that works with social platforms and cybersecurity experts to help curtail the spread of false information which can hamper full citizen participation in the Census– something required by the US Constitution. The idea that these well-meaning and proven efforts somehow constitute “censorship” is not simply misleading, but entirely disingenuous.
Labeling “disinformation” as “censorship” is a lazy and inaccurate, but effective, frame. It demonizes academics, a favorite target of the right. It legitimizes right-leaning information that is incorrect but politically useful– such as the idea that drag queens “groom” children or that Hunter Biden sniffed cocaine at the White House. And it hampers well-meaning and time-consuming efforts to ensure that Americans are not inundated with incorrect and harmful messages every time they go online. While disinformation researchers are a diverse bunch, we are all deeply committed to improving American democracy and ensuring the fair flow of information online. Labeling work based on our research as “censorship” makes it far more difficult to achieve these worthy goals.