Brett Kavanaugh—Yes, Really—Just Stood Up For the Internet as We Know It

Can Florida and Texas weaponize conservative paranoia over Big Tech’s alleged liberal bias to destroy social media and free speech as we know it? After nearly four hours of arguments on this question at the Supreme Court on Monday, the answer is disconcertingly unclear. The justices split along unusual lines as they grappled with the two laws at issue. A few seemed genuinely torn over the best approach to the whole mess before them. And while uncertainty is fine—judges aren’t gods—the stakes are too high for the court to mess this up.

What are those stakes? Well, Florida and Texas are seeking to subvert basic First Amendment principles to turn the most popular websites on earth into unusable quagmires of hate and extremism. If SCOTUS allows them to succeed, it would be a grave misfortune for free expression, free enterprise, and democracy itself.

The two laws at issue were inspired by Republican lawmakers’ conviction that social media platforms discriminate against conservative voices. (They don’t, but these lawmakers mistake a few anecdotal instances as irrefutable proof of a trend.) In response to outcry on the right, Florida Gov. Ron DeSantis and Texas Gov. Greg Abbott signed substantially similar legislation that limited platforms’ ability to moderate content posted by users. The Florida law forbids platforms from moderating any speech about political candidates, deplatforming a political candidate, or disfavoring any “journalistic enterprises.” It also imposes rigid requirements of “consistency” for all other content moderation. Texas’ law goes even further, barring platforms from making any editorial choices at all that are based on the “viewpoint” of the user.

Both laws require companies to notify users who still end up “censored” and allows them to appeal. Both let individual users sue and collect damages—which, in Florida, stretch up to $250,000 per day. Both define “content moderation” broadly to encompass removing, deprioritizing, or “shadow banning” posts; “deplatforming” a user; or affixing commentary on others’ posts. Both are limited to the largest companies on the internet.

These laws were crafted to sound innocuous. They are anything but. Their radical restrictions on content moderation would require platforms to host all manner of odious, revolting speech (including election subversion). Consider the implications of just one provision, Texas’ ban on “viewpoint discrimination.” Under this regime, Facebook could not remove a post spreading dangerous lies about voter fraud. YouTube could not remove a video celebrating white supremacist brutality. Instagram could not remove a photo promoting terrorist propaganda. No company could take down run-of-the-mill bigotry—your racist screeds, your antisemitic memes, the garbage that pollutes everyone’s experience on the internet. They couldn’t even deprioritize this content to shield users from it. Which means users would almost certainly flee by the millions as their daughter’s wedding pictures were suddenly replaced with KKK recruitment videos.

That’s why, in response to the new laws, the platforms sued in both states, arguing that the First Amendment protects their right to moderate content on their own websites. They beat the Florida law but not the Texas law—though SCOTUS halted it while they appealed. Their theory is persuasive. Every platform seeks to foster a certain kind of “community” by removing and deprioritizing certain speech. By exercising this “editorial discretion,” they are engaging in expression themselves. Choosing which speech to boost, obscure, or remove, the platforms say, is fundamentally expressive activity. In that sense, modern content moderation is indistinguishable from a newspaper’s right to publish or not publish a specific column. The Supreme Court has long held that publications—from newspapers to corporate newsletters—have a right to “editorial control and judgment.” Citizens United, meanwhile, clarified that the First Amendment grants the exact same rights to corporations as it does to individuals and media outlets. These established free speech principles, the social media companies claim, protect their own right to moderate others’ speech as they see fit.

Paul Clement, a conservative lawyer working for Big Tech, made these points eloquently on Monday. So did Solicitor General Elizabeth Prelogar, weighing in on the side of the platforms. On the other side, Florida Solicitor General Henry C. Whitaker (bumbling) and Texas Solicitor General Aaron Nielson (obtuse) did a wretched job defending their states’ laws. This disparity left the justices largely in conversation with themselves, a dialogue that revealed two poles and a mushy middle among the nine.

At one pole, Justices Clarence Thomas and Samuel Alito defended the laws and disparaged the platforms as totalitarian bullies. Alito suggested that the phrase “content moderation” succumbed to an “Orwellian temptation to recategorize offensive conduct in seemingly bland terms,” dismissing it as a “euphemism for censors.” Thomas derided the platforms for “censoring, as far as I can tell,” adding, “I don’t know of any protected speech interests in censoring other speech.” (The court has always held that excluding a message is, itself, protected expression.) Thomas also implied that because they are so big, the companies at issue have somehow forfeited their First Amendment rights—a strange argument from the court’s proudest defender of corporations’ right to buy elections. Justice Neil Gorsuch leaned this direction as well.

At the other pole, Chief Justice John Roberts and Justice Brett Kavanaugh boiled the case down to this: The First Amendment prevents the government from censoring private companies; it does not prevent those companies from censoring their own users. Kavanaugh, despite his recent hard-right turn, has been a consistent champion of corporations’ right to host any speech they want, which also of course means excluding any speech they want. While serving on the D.C. Circuit, he wrote an opinion opposing net neutrality with broad language that foreshadowed these cases: The government, he asserted, cannot “tell Twitter or YouTube what videos to post; or tell Facebook or Google what content to favor.” On SCOTUS, he has carefully guarded the distinction between state censorship and private platform moderation. And on Monday, he directly responded to Alito’s ridiculous Orwell reference, reminding his colleague: “When I think of ‘Orwellian,’ I think of the state, not the private sector, not private individuals.”

Kavanaugh hit this theme about as hard as he could. “When the government excludes speech from the public square, that is obviously a violation of the First Amendment,” the justice said. “When a private individual or private entity makes decisions about what to include and what to exclude, that’s protected generally [as] editorial discretion.” He also rebutted Thomas’ suggestion that a company loses its free speech privileges when it reaches a certain size or popularity. Reading from one key precedent, Kavanaugh explained that “the concept that the government may restrict the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment.” To ice the point, he noted that previous generations complained that newspapers had unchecked power “to shape public opinion, and that that had led to abuses of bias and manipulation.” Yet, SCOTUS “said that wasn’t good enough” to let the government force newspapers to carry others’ speech. His implicit question: Why is the internet any different?

Everybody else sat in the middle. Notably, these ambivalent justices sounded fairly confident that these laws violate the First Amendment when applied to platforms that consist of pure speech, like Facebook and YouTube. They wondered, however, whether they also applied to speech in service of commerce—think Etsy or Uber—as well as interpersonal communications like Gmail. Justice Elena Kagan asked if the court could hold that the First Amendment protects “curated news feeds” but not actual services like “Venmo and Dropbox and Uber.” Those businesses are already forbidden from discriminating on the basis of race or sex, Kagan pointed out. Could a state add viewpoint to the list? (The Big Tech lawyer Clement said no, to the justice’s consternation.)

Justice Amy Coney Barrett candidly admitted that “this is a sprawling statute and it makes me a little bit nervous.” She told Prelogar the case contained “a bunch of land” and worried about its “implications” for future cases. Justice Ketanji Brown Jackson likewise complained about “a lot of indeterminacy in this set of facts,” noting that “we’re not quite sure who it covers.” Jackson sounded deeply skeptical that the Florida law, at least, was unconstitutional in “all of its applications,” raising the possibility that they might return the case to the trial court for more fact-finding and a narrower judgment. Jackson and Barrett were more receptive to the possibility that Texas’ law is more blatantly unconstitutional because it applies to only the big speech-oriented platforms.

These concerns about an overly broad or premature decision are understandable. But at the end of the day, Kavanaugh’s straightforward view of this dispute is clearly correct. There’s a specious appeal to the states’ argument that they are somehow vindicating free speech principles by forcing platforms to host more speech. But their argument makes no sense, because the First Amendment applies exclusively to the government. The Constitution prohibits the states from censoring speech; it doesn’t give them license to tell private companies what speech they’re obligated to host. Florida and Texas want to turn that rule on its head. The result would be bad law and terrible policy—and an internet dominated by rotten speech that nobody wants to hear.