Anonymity gets a bad rap in the Internet age. A veil of namelessness, critics say, loosens inhibitions and makes it easier for people to escape accountability for saying terrible things online.
And it’s true that anonymous trolls regularly dump their toxic views on social media and in the comments sections of news and opinion pieces. By trading in misinformation, disinformation, and scaremongering, they and their faceless ilk hurt democracy and compromise public health.
In 2019, Jeff Kosseff’s book “The Twenty-Six Words That Created the Internet” examined the boons and banes of Section 230 of the Communications Decency Act, which helps shield online companies from liability over content posted to their platforms. Section 230 is widely considered a foundational component of the Internet.
Kosseff is back with a follow-up, “The United States of Anonymous: How the First Amendment Shaped Online Speech.” This time, while acknowledging the harm that anonymous online speech causes, Kosseff reminds readers of the long, poorly understood, and underappreciated history of protecting anonymous expression in America. While the right to anonymity shouldn’t be absolute, Kosseff says, he makes a case for better protecting it against encroachments from powerful threats such as technology companies that impose real-name policies, invasive surveillance technologies, and data brokers.
Kosseff is an associate professor of cybersecurity law in the United States Naval Academy’s Cyber Science Department. His views do not represent those of the Naval Academy, the Navy, or the Defense Department. Our conversation has been condensed and edited.
Why is it important to protect anonymous speech in the digital age?
The ability to speak anonymously protects so many people who do not have the luxury of being able to speak freely under their real names. This includes many marginalized groups. Jillian York at the Electronic Frontier Foundation has looked at many proposals that require people to register their real names when speaking online, and she calls them the “White Man’s Gambit.” It’s a very good point. The idea that we all must use our real names is something that only people who have a fairly privileged position in society can support. As a tenured professor, it’s easy for me to say, “Oh yeah, I can use my real name.” But it’s different for many people, including employees complaining about their working conditions.
But doesn’t requiring people to use their real names increase the civility of online communication? Facebook insists that its real-name policy is essential for promoting safety, for example.
At first glance, it seems logical to say that a real-name policy would promote safety and civility. But I’ve been on Facebook for more than a decade now, and I can say that’s certainly not true. Although Facebook’s policy is well intentioned, members of marginalized groups who are worried about harassment repeatedly complain about having their accounts frozen or shut down because people are reporting them for not using their real names. Domestic violence victims also have good reasons for not using their real names.
Twitter and Reddit recognize these issues, and they don’t require real names. However, they prohibit impersonation. Too often, anonymity and pseudonymity are conflated with somehow being fraudulent or inauthentic.
How does the First Amendment protect anonymous speech?
The Supreme Court has recognized that the First Amendment provides strong — but not absolute — protections for people to speak anonymously and pseudonymously. For instance, this means that a law requiring political leaflets to include authors’ names will have a difficult time surviving a First Amendment challenge. Other courts have applied this right to the Internet and set a very high bar for plaintiffs to meet when subpoenaing identifying information of anonymous posters.
In your new book, you cite a 1999 case in which the Rev. Al Sharpton’s National Action Network filed an amicus brief defending the Ku Klux Klan’s right to hold a rally in New York while wearing masks and hoods. Given what the Klan stands for, why did Sharpton take this position?
When Al Sharpton supported the legal position of a group whose values are completely antithetical to his, he took an objective stance on the First Amendment and free speech protections. I can’t imagine it was an easy political position. But I think it was intellectually honest. Sharpton recognized that even if his support helps a group that he vehemently opposes, what matters most are the broader long-term consequences of eroding the First Amendment. I believe he made the right call.
The Washington Post recently revealed the identity of a woman who has been anonymously posting on the Libs of TikTok Twitter account, which shares anti-LGBTQ+ content that gets amplified by influential people like Joe Rogan, Glenn Greenwald, and Tucker Carlson. Some cheered the unmasking, while others criticized the reporting for going too far. What’s your take?
I devote a chapter to cases in which people thought that they were pseudonymous but they disclosed enough publicly available information that an observer could piece together their identities. The Post article suggests that Chaya Raichik was identified based on the name and phone number that she used on a domain registration. Because this information is publicly available, there are no legal barriers to a reporter using it to identify someone. Ultimately, people who want to be anonymous or pseudonymous online need to be aware of all the ways in which they might be identified based on public information. Whether the Post was correct to identify Raichik is more an issue of journalistic ethics than of law.
You recommend that Congress pass a strong national data protection law to protect anonymity from invasive new technologies and data brokers. What privacy safeguards should the law include?
We need to have a data protection law that says certain types of data use, sharing, and collection are off limits. Jurisdictions like San Francisco and Boston that have restricted law enforcement’s ability to use facial recognition technology are a good example of saying there are certain things we’re not going to tolerate.
States like California, Virginia, and Colorado have passed privacy/data protection laws that are a step in the right direction with the value they place on user access and choice. For example, they allow people to request that companies delete some of their data. But I’m not incredibly enthusiastic about the fairly significant burden the laws place on individuals to stop the shenanigans with their identifying information. I work in the field, and I couldn’t even tell you all the companies I would need to submit requests to.
Will a federal privacy law likely pass in the coming years?
I’ve been in Washington, D.C., for almost two decades. I see the lobbying up close and am highly skeptical that something like a strong national privacy law will get passed.
Evan Selinger is a professor of philosophy at the Rochester Institute of Technology and an affiliate scholar at Northeastern University’s Center for Law, Innovation, and Creativity. Follow him on Twitter @evanselinger.