Introduction
“[T]he First Amendment,” Justice Kagan declared in Moody v. NetChoice, LLC,1 one of the five platform First Amendment cases the Court decided this Term, “does not go on leave when social media are involved.”2 That is obviously true. But no one seriously contests the claim that the First Amendment applies to the internet.3 The era when people argued that law should not, or perhaps could not, reach into the vast realms of cyberspace is decades behind us.4 Justice Kagan’s truism therefore only raises the question — what does or should it mean for the First Amendment to apply to the complicated and ever-changing online world? This is the question that this Term’s platform cases raised, in various ways. And it is not a question that has any easy answers, as the cases themselves make clear.
Indeed, this Term’s platform cases all forced the Court to confront one of the most vexing issues in free speech law — namely, how to reconcile expressive conflicts among different participants in the public sphere. Just as is true of the offline public sphere, the digital public sphere depends upon the participation of many different kinds of actors: speakers and listeners, to be sure, but also the private companies that create the forums in which people speak and listen, and the government officials who speak on behalf of their constituents and set the terms on which the system as a whole operates. And while it will often be the case that the interests of speakers, listeners, government actors, and social media companies align, there will be many cases in which they don’t.5 When the interests of these differently situated parties come into conflict, whose interests should courts prioritize?
Historically, this is a question that the First Amendment cases have reached quite different answers to, at different times, and as applied to different sectors of the media. In the last few decades, however, the cases had reached a somewhat uneasy consensus that, except in the most exceptional circumstances, free speech values are best promoted by what Justice Douglas described in 1973 as a “laissez-faire regime” of speech regulation, in which the central principle of operation is, as he put it, “that Government shall keep its hands off the press.”6 In other words, courts have come to believe that the most reliable means of safeguarding freedom of speech is to stringently prohibit government interference with the expressive autonomy of the newspapers and television providers and parade organizers (and, now, social media platforms) that provide the basic infrastructure of the mass public sphere, even when those providers exercise that autonomy to suppress other people’s speech.
This highly libertarian conception of freedom of speech has always been contested, but in the early days of the internet’s development, it was embraced by many as an important means of protecting the explosion of private expression and invention then occurring on this new speech terrain against the conservative and repressive (and surely wrongheaded) impulses of government officials. This attitude was perhaps best exemplified by John Perry Barlow’s famous claims, in 1996, that the internet would be a space into which “all may enter without privilege or prejudice” and “anyone, anywhere may express his or her beliefs, no matter how singular, without fear of being coerced into silence or conformity,”7 and that the government had no legitimate role to play in regulating speech online.8
When it decided its first internet speech case, Reno v. ACLU,9 in 1997, the Court echoed Barlow’s libertarianism in holding that, because the technology of the then-developing internet enabled “any person with a phone line [to] become a town crier with a voice that resonates farther than it could from any soapbox,”10 courts should “in the absence of evidence to the contrary . . . presume that governmental regulation of the content of speech is more likely to interfere with the free exchange of ideas than to encourage it.”11 As the phrase “in the absence of evidence to the contrary” suggests, Reno in fact left many questions open about what the First Amendment actually meant for the developing internet. Nevertheless, at the time, and for many years thereafter, Reno was interpreted as erecting an “extraordinarily high barrier” against the governmental regulation of online speech — and was celebrated for that reason.12 Barry Steinhardt of the ACLU predicted that the case would “be the Brown v. Board of Education of cyberspace.”13
By the early 2000s, however, the consolidation of power and control over the new social spaces of the online world began to complicate the assumption upon which Reno relied — namely, that the best way to promote the free expression of ideas on the internet was by limiting governmental intervention. The rise of dominant search engine, digital marketplace, and social media companies provoked considerable fear across the political spectrum that these companies would use, and in fact were already using, their control over the online speech environment in ways that undermined both public welfare and diversity of expression online. In response, scholars and litigants began to push for a reconceptualization of what it means to protect First Amendment rights online.14
In suggesting a greater role for the government in the protection of speech online, reformers did not challenge the general libertarian framework of First Amendment law. Instead, they argued that the unique features of internet platforms — specifically, the fact that, unlike radio or television broadcasts or newspapers, platforms were spaces in which vast swaths of the public could and (increasingly) did participate — made the power they wielded over the public sphere unique, and justified a level of regulation that would not be appropriate for other kinds of privately owned speech forums.15 The claim, in other words, was that the very thing that Barlow highlighted about cyberspace — its tendency to allow all to enter “without privilege or prejudice” — mandated the opposite conclusion that Barlow reached. Instead, government intervention was particularly necessary in this context to preserve the precious but also fragile openness of the internet as an expressive space.
Perhaps unsurprisingly, these efforts to reconceive how the First Amendment applied to tech companies at first failed to gain widespread traction, given the doctrinal hegemony of the laissez-faire First Amendment. But then things got weird.16 Anxiety among conservative politicians that the tech giants were biased against them, only intensified by the decision by all of the major social media companies to kick President Donald Trump off their platforms after the invasion of the Capitol on January 6, 2021, led conservative jurists and politicians to take up the argument long associated with progressives: namely, that protecting freedom of speech required, rather than precluded, government action.17 These dynamics created a flood of litigation that attempted to secure constitutional rights of access to the internet, as well as laws that constrained the freedom of tech companies in the name of users’ freedom of speech.18 At the forefront of this movement were lawmakers in Texas and Florida, who in 2021 — following a roadmap helpfully outlined by Justice Thomas in a concurring opinion that year19 — enacted audacious new laws that imposed significant nondiscrimination, due process, and disclosure obligations on social media platforms doing business in those states.20
Ultimately this flood of activity, and the weird politics behind it, resulted in a number of lower court decisions that broke, to varying degrees, with the libertarian approach to the regulation of the internet that had dominated since Reno.21 It was in this context that the Court, after years of avoiding these kinds of disputes, agreed to hear five cases that all raised the question, in various ways, of how the First Amendment applies to social media platforms and, more specifically, whether the First Amendment protects or allows states to protect the right of ordinary persons to speak or listen online. This produced a trilogy of platform law decisions: Moody (argued the same day as, and ultimately consolidated with, NetChoice, L.L.C. v. Paxton22), which addressed the facial challenge that NetChoice, a trade association representing the largest social media platforms, brought against the Texas and Florida laws;23 Murthy v. Missouri,24 which asked the Court to decide whether members of the Biden Administration violated the First Amendment by pressuring a range of social media platforms to suppress users’ speech;25 and Lindke v. Freed26 (argued the same day as O’Connor-Ratcliff v. Garnier27), which addressed the question of whether government officials violate the First Amendment when they block private persons or delete their comments after they speak on officials’ social media pages.28
By agreeing to decide all these cases, the Court raised hopes (and fears) that the moment of reckoning for the project of internet exceptionalism had arrived.29 But it was not to be. On their face, the decisions in the platform trilogy seemed to pour cold water on the idea that the arrival of the internet would provide any respite from the laissez-faire, or what some have described as Lochnerian, First Amendment.30 What the cases made crystal clear was that there is nothing close to majority support on the Court for a doctrine of internet exceptionalism that permits the government, or requires it, to address structural inequalities in the public sphere or to protect users’ speech rights online. Instead, all three cases purported to simply apply the standard offline rules to the online context. And in all three cases, when the Court talked about the First Amendment (which, as we will discuss, was less often than you might expect), it did so in highly libertarian terms. One reading of these cases therefore is that they confirmed what has long been true: namely, the dominance of the laissez-faire First Amendment.
But on closer examination, the story is more complicated. Because even as the decisions from this Term proclaimed the centrality and obviousness of the laissez-faire First Amendment, in every case, the Court also refused to finally resolve the case before it. Instead, it either sent the case back to lower courts for further proceedings or dismissed the case on what sometimes appeared to be rather exaggerated procedural grounds.31 The Court in varying degrees decided not to decide, in other words. The result is that the practical consequences of these decisions are very hard to know, and the potential scope of the Court’s rulings remains wide open. It is therefore still possible that, despite the Court’s libertarian rhetoric, the platform trilogy might in fact leave a significant amount of room for legislatures and others to protect individual rights to speak and listen in the online public sphere.
The story is not yet written, however, and the path ahead is precarious. In order to avoid having to fully resolve the constitutional questions raised, the Court decided every single one of the cases in the platform trilogy on some basis other than substantive First Amendment law. Lindke turned on the question of state action,32 Murthy on the question of standing,33 and Moody on the rules for facial challenges.34 The Court may have thought that by framing these cases in this way, it preserved the space for further doctrinal development. What today’s Court appears to have forgotten that earlier Courts well understood, however, is that First Amendment rights must be protected both substantively and procedurally. A doctrine that fails to reckon with the ways procedural rules may hinder the vindication of substantive rights in practice will systematically under-protect those rights. This Term’s platform cases threaten to bring about this very future — the doctrinal maneuvers that the Court employed that left open space for decision about the substantive law later may, in practice, make it harder for speakers and listeners (especially less well-resourced ones) to actually vindicate their rights in court. In this respect, therefore, even if this Term’s cases were not ultimately as deregulatory as they could have been, they still recapitulated many of the ills associated with First Amendment Lochnerism in their failure “to consider the economic and social forces that as a practical matter shape the exercise of First Amendment rights.”35
This Comment proceeds in three parts. Part I discusses the Court’s changing view of how the First Amendment applies in cases involving conflicts between private speakers and the dividing line between the public and private spheres. What this history shows is that there is nothing natural or inevitable about the laissez-faire First Amendment and, in fact, that the laissez-faire commitments of First Amendment doctrine are far less absolute and simplistic than is commonly understood. Part II explores the challenges that all five of the platform cases the Court decided this Term posed to the doctrinal status quo and the Court’s complicated response to those challenges. Finally, Part III examines the decisions in the platform trilogy normatively. It argues that while the Court showed salutary humility and cautiousness in approaching its task of writing a First Amendment law for the platform era, it also showed insensitivity to the structural features that will determine how that law will be applied in practice.
Continue Reading in the Full PDF
* Assistant Professor of Law, Stanford Law School.
** Professor of Law & Herbert and Marjorie Fried Teaching Scholar, University of Chicago Law School. Our thanks to Ros Dixon, Marty Lederman, Nelson Tebbe, Rebecca Tushnet, Eugene Volokh, and Laura Weinrib, whose comments made this piece better, no question mark. Thanks also to the wonderful editors at the Harvard Law Review for all their careful and hard work on our piece.