In 2021, conservative lawmakers were concerned about so-called “Big Tech” censorship. Fresh in their minds was Twitter’s and Facebook’s suppression of the Hunter Biden laptop story, as well as those platforms’ decision, in the wake of the January 6 Capitol riot, to ban Donald Trump from their services.
In Florida and Texas, this concern translated into a pair of statutes that (among other things) heavily restrict social media platforms’ power to demote or block content. Gov. Ron DeSantis signed Florida’s SB 7072 into law in order to fight the liberal “Silicon Valley ideology” and hold “Big Tech censors” accountable. Supporting Texas’s HB 20, Gov. Greg Abbott complained that “social media companies … silence conservative viewpoints.” In both instances, all eyes were on the largest supposed “censors” and “silence[rs].” A service with fewer than 50 million monthly users and less than $100 million in annual revenue is not subject to either law.
When the statutes were challenged in court, the lawyers, district judges, and circuit judges who grappled with the cases took Florida and Texas at their word. They focused on whether the government may alter the balance of speech on the major social media platforms. Much litigation occurred; there were many twists and turns. Both district courts issued a preliminary injunction. On appeal, the Eleventh Circuit affirmed in pertinent part, while the Fifth Circuit reversed. At one point, the Supreme Court stepped in and granted emergency relief. All the while, though, the lion’s share of attention remained fixed on how the laws might affect “social-media behemoths like Twitter and Facebook,” as Florida put it in the first lines of its petition for certiorari.
The Supreme Court granted review in both cases—Moody v. NetChoice (SB 7072) and NetChoice v. Paxton (HB 20)—and last week it issued its decision. The ruling is at once a great surprise and a predictable outcome.
Lets start with the surprise. All nine justices reject how the cases were framed below. The plaintiffs, a pair of trade groups, seek to have the laws struck down in full—in all their applications and for everyone. The justices find that “neither Court of Appeals properly considered the facial nature of [the plaintiffs’] challenge.”
For a law to be facially invalid under the First Amendment, Justice Kagan explains for the Court, “a law’s unconstitutional applications” must be “substantial compared to its constitutional ones.” But the lower courts looked only at “how the [Florida and Texas] laws applied” to things like “Facebook’s News Feed and YouTube’s homepage.” They did not consider whether or how the laws govern the major platforms’ “other services, like direct messaging or events management.” Nor did they consider whether or how the laws cover services such as Gmail, Etsy, Venmo, or Uber. In short, the lower courts “did not address the full range of activities that the laws cover, and measure the constitutional against the unconstitutional applications.” The Supreme Court therefore vacated the judgments and remanded the cases, directing the lower courts to perform the proper inquiry.
Now for the predictable development. A majority of the Court leaves no doubt as to its views on the point that everyone fought about below. Six of the justices confirm that, under well-established First Amendment law, “expressive activity includes presenting a curated compilation of speech originally created by others.” And five justices agree that “the Fifth Circuit was wrong in concluding that Texas’s restrictions on the platforms’ selection, ordering, and labeling of third-party posts do not interfere with expression.” Texas, the five-justice majority says, “may not interfere with private actors’ speech to advance its own vision of ideological balance.”
The Fifth Circuit’s analysis rested, the majority concludes, “on a serious misunderstanding of First Amendment precedent and principle.” The majority proceeds to “explain why that is so” in an effort to “prevent the Fifth Circuit from repeating its errors,” on remand, “as to Facebook’s and YouTube’s main feeds.”
Along with Justice Kagan’s opinion for the Court, four concurring opinions, by Justice Barrett, Justice Jackson, Justice Thomas, and Justice Alito (joined by Justices Thomas and Gorsuch), will confront the lower-court judges on remand. These separate writings point out, in their varying ways, that the majority opinion leaves important matters unresolved. Justice Alito, in particular, is at pains to assure the lower courts that they continue to enjoy maximum discretion. The Court’s only binding holding, he observes, is that the plaintiffs “failed to prove that the Florida and Texas laws … are facially unconstitutional.” He all but invites the Fifth Circuit to reissue its original ruling, under the guise of some alternative theory, in defiance of the majority opinion’s guidance.
The NetChoice majority recounts a history of government attempts to “promote greater diversity of expression” through regulation.
So the parties will continue to litigate the NetChoice cases, hotly contesting the scope and substance of the Court’s decision.
But why?
It might seem like a silly question. And I have no illusions that I—who have believed from the jump that SB 7072 and HB 20 are folly—can convince Florida and Texas to consider it. But if only they would do so, they might start to wonder if their crusade has run its course. The online world looks rather different than it did in 2021, and the claim that “Big Tech censorship” poses a significant threat to “conservative viewpoints” seems flimsier than ever.
The most obvious change is that Elon Musk purchased Twitter, gutted its trust and safety team, and offered a “general amnesty” to accounts banned under the prior regime. Donald Trump’s account is back—though, tellingly, he seems no longer to want or need it. These days, the content moderation rules at what is now called X are ad hoc: the people who get banned are generally either personally offensive to Musk or mentally disturbed in the style of Kanye West. This is not the “free speech absolutism” that Musk sometimes, before the takeover, vowed to implement, but it seems to suit conservatives just fine. Indeed, some on the right celebrated Musk’s (short-lived) decision to ban liberal journalists who rubbed him the wrong way.
Facebook and YouTube, too, have loosened their rules (to howls of protest from progressives). Many Twitter users have migrated to Threads, Mastodon, or Bluesky. And many of us have sharply curtailed the time we spend on public-facing social media of any sort. Believe me, much of the best online conversation now occurs on the cozy web—private chat groups on Signal, WhatsApp, Discord, and Slack. Some contend that this fragmentation is a sign that mass social media’s brief heyday is drawing to a close. It might be that the age of centralized platforms—when it felt like Twitter, in particular, served as “a giant, asynchronous chat room for the world” (in the words of The Atlantic’s Ian Bogost)—was a passing phenomenon.
It is no longer possible (if it ever was) to think of any one social media platform as the “public square.” Public debate occurs on the internet as a whole. Conservatives correctly worry about their lack of artistic, cultural, and institutional clout; but they enjoy plenty of internet clout. On the internet, the mainstream media is struggling, while rightwing (or, at least, non-leftwing) media is thriving. The freshest outlets—e.g., Pirate Wires, The Free Press, Compact—are heterodox. The most interesting voices on Substack—e.g., N. S. Lyons, Paul Kingsnorth—are reactionary. Many journalists—e.g., Aaron Sibarium, Sanjana Friedman, Olivia Reingold, Julio Rosas—are doing serious investigative reporting from outside the liberal media bubble. The Daily Wire and PragerU create tons of conservative content, right down to children’s programming. Hillsdale College offers a lecture series that examines everything from the fall of the Roman Republic to chemistry through a traditionalist lens. There are more conservative podcasts than you can poke a stick at. Many of the most popular podcasts in the country are, if not on the right, immensely skeptical of the left.
Some conservatives worry that technology will eventually enable the government, corporations, or both to work together to impose mass censorship and construct a social credit system. That might one day become a real concern. But for now, we actually have something like the opposite problem. Our immediate task is to overcome the disorientation wrought by an information explosion. Viewpoints are not scarce; they are abundant. The internet allows you to immerse yourself in almost any ideology you want. That creates its own set of problems, but they’re not problems to do with online censorship and the need for online “anti-censorship” legislation. (Especially not a patchwork of state legislation coming out of places like Tallahassee or Austin.) To clamor for such legislation anyway is to embrace the kind of victim mentality that was until yesterday a specialty of the left.
The rise of generative AI, by the way, will pile disruption on top of disruption. Assuming we don’t lock in the power of the major firms through regulation, we are likely to end up with AI assistants that can adopt almost any given worldview. Heretical ideas—and, with some luck, fresh thinking—will proliferate. That’s good, and laws like SB 7072 and HB 20 won’t be needed to help things along. Meanwhile, spam and other synthetic content will flood online spaces. That’s bad, and such laws would make things worse. The NetChoice majority recounts a history of government attempts to “promote greater diversity of expression” through regulation. But “in case after case,” the majority reminds us, “the Court has barred the government from forcing a private speaker to present views it wished to spurn in order to rejigger the expressive realm.” However “imperfect the private marketplace of ideas, here [i]s a worse proposal—the government itself deciding when speech [i]s imbalanced, and then coercing speakers to provide more of some views or less of others.” Government intervention is indeed the worst proposal. And anyway, the online marketplace of ideas might not be so imperfect after all.