For many years, conservative politicians and journalists have complained that the major social media and other tech platforms are biased against conservative speech and speakers. And while these criticisms are not new, they have become much louder and more insistent following the deplatforming of President Trump by the major social media platforms, following the attack on the US Capitol on January 6, 2021. What is the nature of these critiques, to what extent are they legitimate, and what ultimately lies behind them? These are the questions this chapter explores.

As a starting point, it is important to recognize that while their volume has increased since 2021, conservative claims of alleged political bias on the part of social media are not new. As far back as the 2016 presidential campaign, conservative commentators began accusing Facebook of a left-leaning bias in its selection of “trending” news articles.Footnote ¹ These claims were rooted in conservatives’ (probably accurate) perception that the employees and management of the major social media firms, who are mainly residents of the San Francisco Bay Area, tend to lean politically to the left. Whatever the legitimacy of these claims (a question addressed later), the irony of this, of course, is that then-candidate Donald Trump was far more effective at deploying social media, especially Twitter/X, to his political advantage than his Democratic rival in the presidential election, Hilary Clinton (though, to be fair, he was also more effective than his Republican rivals in the primary elections).

After Trump’s election, the loudest attacks on social media shifted to their data privacy practices, especially after the Cambridge Analytica scandal of 2018. That scandal arose when in the lead-up to the 2016 presidential election Cambridge Analytica, a political firm tied to Republican donors and to Stephen Bannon (who later became a senior adviser to President Trump), harvested private data on over 50 million Facebook users. The data was gathered by a Cambridge University scientist on the pretense that they were engaged in academic research, and then it was used by Cambridge Analytica to construct profiles of voters, which were in turn used by the firm to provide services to the 2016 presidential campaigns of Senator Ted Cruz and then Donald Trump.Footnote ² The fact that Facebook, albeit inadvertently, aided the Trump campaign in this fashion, as well as in other ways (notably by failing to block Russian manipulation of the electionFootnote ³), did not, however, have any impact on ongoing conservative claims of bias against them.

For example, during Mark Zuckerberg’s testimony before Congress in July 2020, Republican members of Congress repeatedly accused Facebook of disproportionately targeting conservative content for blocking, echoing long-standing similar claims made by a number of prominent Republican political leaders.Footnote ⁴ Soon thereafter, in early August, Facebook deleted a post by President Trump’s campaign linking to a video in which Trump had said that children were “virtually immune” from COVID-19, on the grounds that the post violated its policies against COVID misinformation (soon after this Twitter/X blocked the Trump campaign’s account for linking to the same video).Footnote ⁵ In response, the White House deputy national press secretary accused Facebook and other Silicon Valley firms of “flagrant bias against this president, where the rules are only enforced in one direction.”Footnote ⁶

Indeed, throughout the COVID-19 pandemic, conservative politicians and commentators attacked social media firms for blocking, or labeling as misinformation, what they perceived to be conservative views on the disease. One example is social media platforms’ response to the claim, heavily pushed by senior Trump Administration officials,Footnote ⁷ that COVID-19 originated from an accidental leak from a lab in Wuhan, China. Facebook originally decided to remove claims, though it later rescinded that decision;Footnote ⁸ and Twitter/X originally labeled such claims as misinformation.Footnote ⁹ When some, but not all, US government agencies expressed “low confidence” support for the lab-leak theory, conservative politicians and journalists claimed vindication and sharply attacked Facebook and Twitter/X for their earlier policies.Footnote ¹⁰ Similarly, when President Trump began publicly claiming that the antimalarial drug hydroxychloroquine was effective in treating COVID-19,Footnote ¹¹ Facebook and Twitter/X began blocking such claims as medical misinformationFootnote ¹² (they were probably correct to do so,Footnote ¹³ though definitive proof remains elusive). Conservative politicians duly attacked social media for their “censorship.”Footnote ¹⁴ Senator Ted Cruz of Texas, a prominent Republican, indeed went so far as to criticize both the Biden Administration and “Big Tech” for allegedly coordinating actions to block misinformation about COVID-19 vaccines.Footnote ¹⁵ And ultimately, a group of Republican politicians and activists, led by Missouri’s Attorney General, sued the Biden Administration on the theory that its efforts to pressure the platforms regarding COVID misinformation violated the First Amendment. The claimants originally won a resounding victory in the (very conservative) United States Court of Appeals for the Fifth Circuit.Footnote ¹⁶ Their lawsuit, however, was ultimately dismissed by the US Supreme Court on grounds that effectively rejected the claim that the Biden Administration was responsible for the platforms’ misinformation policies.Footnote ¹⁷

While the COVID-19 pandemic (and responses to it) sharply stimulated the conservative war on social media and claims of political bias, it was only a beginning. The event that unquestionably took it to a new level (and, as we shall see, even brought some Supreme Court Justices into the mix) was the deplatforming of President Donald Trump by the major platforms in January of 2021. The background here is of course familiar. After losing reelection in the November 2020 presidential election, President Trump (supported by conservative news outlets such as Fox News and Breibart) began aggressively disseminating unsupported charges of widespread election fraud, claiming that he had in fact won the election. On January 6, 2021, Congress was scheduled to convene to count electoral votes, as required by the Twelfth Amendment to the US Constitution. On the same day, President Trump called for a rally of his supporters outside the White House. After President Trump addressed the rally, his supporters began moving toward the Capitol. Eventually, some of those supporters began to break through police barricades and physically attack Capitol police officers, and then broke into the Capitol itself. Hours of violence and mayhem followed, during which time President Trump continued to Tweet claims of election fraud, though he also did call for his supporters to remain peaceful. Eventually law enforcement was able to secure the Capitol, after which Trump uploaded posts onto social media which seemed to express support for the rioters. These posts were, shortly thereafter, taken down by Facebook and Twitter/X, and President Trump’s accounts were temporarily suspended. Ultimately, Congress reconvened, and certified Joe Biden’s election as President (though only after a number of Republican members of Congress challenged the election results).Footnote ¹⁸

Horrific as the events of that day were, for our purposes the key point is that social media platforms’ suspension of President Trump originally occurred on January 6 itself, immediately in the wake of the attack on the Capitol. And originally, those suspensions were temporary, in response to the seemingly continuing risk of violence. Soon thereafter, however, citing concerns about violence leading up to Biden’s inauguration on January 21, Facebook and Twitter/X made Trump’s deplatforming permanent.Footnote ¹⁹ Twitter/X ultimately reinstated Trump’s account after Elon Musk’s purchase of the platform in late 2022,Footnote ²⁰ and Facebook reduced its suspension to two years, in conformity with a recommendation by the Facebook Oversight BoardFootnote ²¹ (Trump was duly reinstated in January of 2023Footnote ²²).

The primary conservative response to these events was a deluge of criticism alleging that the major tech platforms were intentionally silencing conservative speakers up to and including the sitting President of the United States. Donald Trump, Jr., the President’s son, described the actions (on Twitter/X, ironically) as Orwellian.Footnote ²³ Similarly, then-Representative Devin Nunes, a prominent Trump supporter who later left Congress to run Trump’s new social media platform Truth Social, sent a letter to colleagues in the House of Representatives on January 12, 2021, stating that “Big Tech has launched an overwhelming offensive to deprive Americans of our freedom to communicate with each other,” with the goal of “transforming our digital space into a left-wing monoculture in which conservatives are harassed, ostracized, banned, deplatformed, and threatened with an array of other punishments.”Footnote ²⁴

Nor did the attacks originate only in Washington, DC. Republican Governor Ron DeSantis of Florida, in particular, has made criticisms of social media bias a central part of his “War on Woke,” inducing the Florida legislature to pass legislation (titled S.B. 7072) that restricts social media platforms’ power to moderate content on their platforms or remove users.Footnote ²⁵ While the provisions of the Florida law are complex, their primary thrust was to ban platforms from deplatforming candidates for political office (a rather obviously self-serving provision by the Florida legislature), limiting the visibility of posts about political candidates, and limiting “journalistic enterprises.” The legislature justified these steps by describing social media platforms as being equivalent to “public utilities” or “common carriers.” When he signed the bill, Governor DeSantis explained that its purpose was to ensure that “Big Tech” does not “discriminate in favor of the dominant Silicon Valley ideology,” or against conservative voices.Footnote ²⁶ When challenged in court, major portions of the Florida law were struck down by both a federal district court and the regional federal court of appeals (the Eleventh Circuit).Footnote ²⁷ How the Supreme Court ultimately resolved the case will be taken up in detail in Chapter 4 (spoiler alert: The Supreme Court failed for technical reasons to finally resolve the case but strongly endorsed the Eleventh Circuit’s basic legal reasoning).

Not to be outdone by his fellow conservative, Republican Governor Greg Abbott of Texas soon followed suit. In particular, Abbott began making public statements regarding the need to adopt legislation that would prevent platforms from “silenc[ing] conservative speech and ideas.”Footnote ²⁸ And at Abbott’s urging (after a first, unsuccessful attempt), in September of 2021 a special session of the Texas legislature enacted HB 20. Like the Florida law, HB 20 insists that “social media platforms function as common carriers.”Footnote ²⁹ And like Florida, the Texas law limited social media content moderation policies. But in its impact on social media content moderation, Texas went well beyond Florida. Rather than merely protecting politicians and journalists, HB 20 forbids all censorship by social media platforms based on “the viewpoint of the user or another person,” “the viewpoint represented in the user’s expression,” or “a user’s geographic location” within Texas.Footnote ³⁰ In practice, then, the effects of HB 20 were intended to be enormously broad, effectively depriving social media firms of any meaningful editorial control over the platforms that they own. When this law was challenged in federal court, the regional federal court of appeals covering Texas (the Fifth Circuit) upheld HB 20, reversing a lower court decision and adopting reasoning that was essentially the opposite of the Eleventh Circuit’s.Footnote ³¹ But as we shall see in Chapter 4, the Supreme Court, on review, flatly rejected the Fifth Circuit’s legal reasoning as inconsistent with precedent and basic First Amendment principles.

That conservative politicians would want to invoke a “War on Woke” or a “War on Big Tech” to energize their base is hardly surprising. What is striking, however, is where the legal inspiration for their actions originated. It was Justice Clarence Thomas of the United States Supreme Court. A few months after President Biden took office on January 21, 2021, the Supreme Court dismissed a First Amendment case brought against President Trump based on his actions blocking certain users from posting on his Twitter/X account (the case was dismissed because Trump was no longer President, so no longer subject to the First Amendment). Justice Thomas agreed with the Court’s action, but then went on to write a lengthy screed (I say screed because the opinion had little or nothing to do with the case under review) raising concerns about social media’s power over free speech, and advocating legislative action regulating social media platforms as either “common carriers” or “places of public accommodation.”Footnote ³² His arguments were primarily historical and legal, but they were clearly motivated (he said as much) by the seeming anomaly that the First Amendment arguably prevented President Trump from controlling his Twitter/X account but did not prevent Twitter/X from entirely deplatforming Trump. Justice Thomas is of course a famously conservative member of the Supreme Court, and also one of its most overtly partisan members (his wife was prominently involved in efforts to overturn the 2020 presidential electionFootnote ³³). As such, his publicly stated views on this matter provide an important window into how broadly distrust of social media, and belief in its political bias, pervaded conservative circles in the aftermath of the January 6 attack and President Trump’s subsequent deplatforming.

In short, conservatives apparently strongly dislike and distrust the major social media platforms. But are they justified in doing so? That question is almost impossible to definitively resolve, and in truth the answer generally lies in the eyes of the beholder. What is relatively clear is that prior to Elon Musk’s purchase of Twitter/X in late 2022,Footnote ³⁴ there was little empirical or other strong evidence supporting conservative claims of social media bias (though to be fair, there was also no conclusive evidence refuting them).Footnote ³⁵ Indeed, press reports suggest the contrary, that social media platforms historically bent over backward to permit conservative content to remain online,Footnote ³⁶ and an internal Twitter/X study suggests that its recommendation algorithms also favored conservative content (except, apparently, in Germany).Footnote ³⁷

After his takeover, however, Musk released a selection of internal documents – the so-called Twitter Files – which purported to support conservative claims of bias.Footnote ³⁸ These documents consisted of communications among Twitter/X employees regarding some of their most important content-moderation decisions, including the deplatforming of President Trump. The small group of (carefully chosen) journalists to whom Musk released the material, including notably Matt Taibbi and Bari Weiss, duly argued that the files demonstrated a left-leaning bias on the part of Twitter/X in its content moderation decisions, just as Elon Musk had claimed.Footnote ³⁹ Unsurprisingly, conservative media outlets such as Fox News took up the call, treating the Twitter Files revelations as a major scandal undermining the legitimacy of the major social media platforms, especially Twitter/X.Footnote ⁴⁰ Other major media outlets such as CNN and NPR, however, expressed skepticism that the Twitter Files contained any significant new revelations or demonstrated bias, suggesting instead that the files “have largely not contained any revelatory information. So far, the files have failed to do much outside highlighting exactly how messy content moderation can be.”Footnote ⁴¹ In other words, nothing to see here folks.

So who is right? Gerard Baker, a conservative former Wall Street Journal editor-in-chief, may have had it right when he suggested that while the Twitter Files do not reveal any intentional misbehavior or bias on the part of Twitter/X decisionmakers (apart from a lack of transparency or honesty), it does reveal a form of groupthink as a result of the fact that the staff of Twitter/X (based in San Francisco) overwhelmingly share a particular, progressive worldview which does not partake of doubt. The result, Baker argues, is a good-faith willingness on the part of Twitter/X employees to suppress content with which they disagree, labeling it illegitimate misinformation.Footnote ⁴² Given empirical evidence provided by Taibbi (and cited by Weiss) that Twitter/X’s workforce is between 97 percent and 99 percent Democratic (based on campaign contributions at least),Footnote ⁴³ that story is an eminently plausible one that also does not rely on conspiracy theories (much to Musk’s disappointment, one imagines) or identifying “bad guys.”

There is, however, another possible, and not even inconsistent, explanation for why it is that high-profile content moderation seems to disfavor conservative over progressive voices. Let us start by considering what kind of content it is that the major social media platforms – we will focus on Facebook and Twitter/X here, though the others are not so different – prohibit. These rules are laid out in Facebook’s Community StandardsFootnote ⁴⁴ and in the Twitter/X Rules.Footnote ⁴⁵ Most fundamentally, both sets of standards/rules, for obvious and generally noncontroversial reasons, prohibit incitement of violence, including speech that has a serious possibility of leading to violence. Twitter/X also prohibits glorifying violence, and although Facebook’s Community Standards do not explicitly do the same, their prohibition on implicit calls for violence can easily be read to prohibit such glorification. Indeed, it was President Trump’s claimed violation of precisely these rules that lead to his deplatforming by both Twitter/X and Facebook. But the truth is that, in today’s world, calls for violence and glorification of violence are far more likely to emerge from the political right than the political left (in the 1960s and 1970s it would have been the opposite, with the radical left far more likely to call for violence). Of course, most mainstream conservatives do not call for or support political violence. But more radical elements of the political right such as the Proud Boys and the Oathkeepers quite explicitly do endorse violenceFootnote ⁴⁶ – and some of the more extreme Republican members of Congress, such as Marjorie Taylor Greene and Matt Gaetz, are not all that far apart from that position.Footnote ⁴⁷ Given these facts, it is hardly surprising that conservative figures are more likely to violate content moderation rules regarding violence than progressives.

Facebook and Twitter/X also prohibit hate speech (or what Twitter/X calls hateful conduct), which Facebook defines as direct attacks on people on the basis of their “race, ethnicity, national origin, disability, religious affiliation, caste, sexual orientation, sex, gender identity and serious disease”Footnote ⁴⁸ (Twitter/X’s rules are similar and encompass attacks “on the basis of race, ethnicity, national origin, caste, sexual orientation, gender, gender identity, religious affiliation, age, disability, or serious disease”Footnote ⁴⁹). The question one might ask oneself is who, today, is most likely to violate these rules. The answer is not always conservatives – people on the left certainly can and do engage in such attacks (notably based on religious affiliation, but also sometimes race). But it is also true that with the unfortunate, possible exception of antisemitism, it is not generally a part of mainstream progressive ideology to condemn individuals on the basis of such characteristics, while hostility to or exclusion of individuals based on sexual orientation and gender identity most certainly are embraced by elements of the mainstream political right – as illustrated by the fact that the official platform of the Texas Republican Party, adopted in 2024, flatly opposes “the teaching of sex education, sexual health, or sexual choice or identity in any public school in any grade whatsoever,”Footnote ⁵⁰ and one speaker at the 2023 Conservative Political Action Conference (CPAC) convention (historically the most prominent conservative political convention) stated that “transgenderism must be eradicated from public life entirely.”Footnote ⁵¹

It is thus again unsurprising that the major social media platforms’ hate speech policies are far more likely to impact conservative than progressive speakers. Is it nonetheless possible that employees of social media enforce their rules more strictly against conservative than progressive users, perhaps even inadvertently? For example, is it possible that a Twitter/X employee was more likely, prior to the platform’s purchase by Elon Musk, to label an attack based on sexual orientation as hate speech than to label one as such based on an individual being an Evangelical Christian?Footnote ⁵² Of course it is. But the problem is that simple statistics showing that more conservatives than progressives run afoul of hate speech policies – assuming such statistics exist – cannot demonstrate the existence of bias on the part of platforms, either conscious or unconscious.

Finally, consider one more example of a widely followed content moderation policy: bans on particular forms of dis- and misinformation, especially related to the COVID-19 pandemic. Facebook continues to enforce such a rule,Footnote ⁵³ and Twitter/X did so until Elon Musk’s purchase of the company in late 2022.Footnote ⁵⁴ In enforcing their rules, both platforms inevitably relied on outside health policy experts in determining what constituted misinformation, according to the prevailing scientific consensus. But as with hate speech, while COVID misinformation flowed from across the political spectrum, the tendency of political figures on the right to endorse such things as vaccine skepticism, or unproven treatments such as hydroxychloroquine and ivermectin, unsurprisingly lead to greater enforcement of this policy against conservative speakers. Of course, the reality is that enforcement of the anti-misinformation policy was spotty and sometimes grossly wrong – as most obviously was true of the decision to label the theory that COVID-19 originated from a lab leak in Wuhan, China as misinformation early in the pandemic. But as with the incitement and hate speech policies, the fact that the misinformation policy disproportionately affected conservatives proves nothing regarding alleged bias.

Indeed, a very recent study published in the leading scientific journal Nature provides strong empirical support for this explanation.Footnote ⁵⁵ Focusing on data regarding politically active Twitter/X users during the 2020 US presidential election, the authors of the study concluded that while pro-Trump conservative users were indeed more likely to be suspended than pro-Biden/liberal users, conservative users were also far more likely to link to what the authors call “low-quality news sites.” And strikingly, this result held even when Republican laypeople evaluated the quality of the relevant news. Furthermore, when the authors examined a broader set of datasets examining Twitter/X and Facebook users from 2016 to 2023 across sixteen different countries, the same results emerged. Of course, these empirical results do not rule out the possibility of platform bias, especially because the study did not examine reasons provided for individual suspensions; but it does raise doubts about such claims when they are based (as they generally are) only on anecdotal observations of platform actions against conservatives.

Furthermore, whatever the evidence (or lack thereof) of anti-conservative bias in the past, there is every reason to believe that, going forward, this issue has largely disappeared. Most obviously, Elon Musk’s purchase of Twitter/X in late 2022 has resulted in that platform, if anything, adopting an anti-progressive bias (as illustrated by its banning of the terms “cis” and “cisgender” in 2023Footnote ⁵⁶). More fundamentally, one of the coauthors of the Nature study, Professor David G. Rand of MIT, points out that in the current (late 2024) political environment, platforms are receiving far more public attacks for their alleged anti-conservative bias than for spreading misinformation, making it likely that moving forward (especially after Donald Trump’s victory in the 2024 presidential election), social media platforms will lean over backward to avoid blocking conservative content.Footnote ⁵⁷ And indeed, in an August 2024 letter from Mark Zuckerberg, the CEO of Meta (which owns both the Facebook and Instagram, as well as the related Threads, platforms), to Representative Jim Jordan, the Republican chair of the House Judiciary Committee, Zuckerberg apologized for yielding to pressures from the Biden Administration to block alleged COVID-19 disinformation. And in the same letter, Zuckerberg strongly suggested that moving forward, Meta platforms will work hard to avoid complying with such political pressure, as well as reducing the extent to which they demote content labeled as misinformation.Footnote ⁵⁸ Given that Zuckerberg and Musk, between themselves, fully control Facebook, Instagram, and Twitter/X, concerns about anti-conservative bias on the major social media platforms seem, going forward, to be baseless.

Stepping back from the uncertainties surrounding whether claims of past anti-conservative bias on the part of platforms have any validity (my own suspicion is that they do, but that the extent of such bias is vastly exaggerated), it is worth considering the underlying assumptions and bases of the public and legislative attacks on social media from conservative circles. At first cut, the reasoning behind the conservative attack seems straightforward. Social media has become the primary source of news and information, and the primary site for political debate and discourse, in this country and abroad since social media’s explosion in the early 2010s. Indeed, especially for younger people social media is often their sole source of news and political information. Furthermore, the social media industry is dominated by a handful of platforms, controlled by a handful of individuals – Mark Zuckerberg alone, with his control over both the Facebook and Instagram platforms, can impose his will on the availability of information, and access to discourse, for a huge percentage of the global population of all ages. If this kind of power is used to bias the debate in favor of particular viewpoints or perspectives, as conservatives claim is the case, then that will have a profoundly distorting effect on political discourse, and ultimately on democracy itself. It is therefore, this argument goes, perfectly legitimate for states like Florida and Texas to step in and prevent such abuses of power.

But now let us take a step back and consider the implications of this position for another very important, and in many ways more dominant, source of news and public discourse: Fox News. Fox News has for years been the most-watched cable network of any kind in the country – in particular, it is far ahead of its primary news competitors, MSNBC and CNN.Footnote ⁵⁹ Fox News is also the most important source of news and political commentary for Republicans, especially for older Republicans (which matters because older people are more likely to vote).Footnote ⁶⁰ As a consequence of the broad support, trust, and loyalty that Fox News enjoys with conservatives, its coverage has a significant (and from the point of view of progressives deleterious) impact on national politics.Footnote ⁶¹ And, of course, Fox News’s coverage and commentary famously takes a highly conservative slant, to the point sometimes of knowingly spreading falsehoods such as claims that the 2020 presidential election was stolen through electoral fraud, in order to please their conservative audience.Footnote ⁶² In short, the conservative bias of Fox News has at least as important an impact on public discourse in the United States as the alleged anti-conservative bias of social media platforms, and almost certainly a far greater one.

So do conservatives, or does anyone, believe that progressive commentators should have a right to appear on Fox News, and be given equal treatment compared to conservative commentators such as Tucker Carlson or Laura Ingraham? Of course not. Most people, whether conservative or progressive, surely agree that Fox News, as a media outlet, has a right to hold and spread its own chosen political opinions using its privately owned platform. Indeed, the right to such editorial discretion and control lies at the core of the freedoms protected by the Press Clause of the First Amendment to the US Constitution; and it also follows from the property rights Fox News holds over its assets. And exactly the same is true of other forms of media such as newspapers – the Wall Street Journal’s editorial pages are famously conservative, while the New York Times and Washington Post pages lean progressive – and radio stations such as those who host right-wing talk radio shows (think of Rush Limbaugh, Mark Levin, and Ben Shapiro).

To be clear, the right to such editorial slant was not always the law. During the heyday of broadcast (not cable) television and radio, beginning in 1949, the Federal Communications Commission (FCC), an agency of the US government, enforced a set of rules called the Fairness Doctrine which required broadcasters to report news evenly. And in the 1960s the Supreme Court upheld the Fairness Doctrine against a First Amendment challenge.Footnote ⁶³ But the FCC repealed the Fairness Doctrine in 1987 (thereby enabling the rise of right-wing talk radio), and even when it was in place, the Fairness Doctrine was strictly limited to the broadcasting industry – the Supreme Court explicitly rejected efforts by the State of Florida to impose similar requirements on newspapers.Footnote ⁶⁴

The obvious question that arises, of course, is that if Fox News has a right to adopt a political slant, and so do newspapers and radio stations, why don’t Facebook or Twitter/X have the same rights, either under the First Amendment or as a matter of fairness? And from a more traditionally conservative perspective, one might ask why social media companies should not have the right to use their private property as they choose, and exclude from their property whomever and whatever they want. After all, President or not, Donald Trump has no right to invade my living room, or use my backyard, to hold a rally. And it should be noted that in a recent case pitting property rights against the free speech interests of labor organizers, the conservative majority of the US Supreme Court ruled resoundingly in favor of protecting property rights.Footnote ⁶⁵ So why, according to conservatives such as Justice Clarence Thomas (and his colleagues Justices Samuel Alito and Neil Gorsuch), are social media platforms differently situated?

To be fair, there are important differences between social media platforms and traditional media outlets such as Fox News and the New York Times. Most obviously, the latter primarily distribute content that they themselves have created or, in the case of advertisements and op eds, chosen; social media platforms, in contrast, obviously distribute primarily user-generated, third-party content. Because of this, the Florida and Texas legislatures argued (inspired by Justice Thomas) that social media are more like telephone companies than newspapers; and historically, telephone companies were regulated as common carriers, meaning that they had to serve all customers in a nondiscriminatory fashion. In other words, common carriers such as telephone companies (and railroads and inns and ferries) did not have a right to exclude customers or content of which they disapproved. And this was true regardless of the First Amendment, or the fact that the relevant firms were privately owned.

The conservative efforts to regulate social media thus are not without any plausible legal basis. But they are ultimately wrong. The full explanation for why the common carrier label is not a good fit for social media platforms will have to wait until Chapter 4, but at the outset it is important to note that common carrier status is very much the exception in our legal tradition – normally the assumption is that private actors can choose with whom to do business, subject (in modern times) to narrow antidiscrimination laws. And such labels were never applied to media outlets possessing First Amendment rights – which, as I will argue in Chapter 4, social media platforms do possess. Indeed, even at the height of the Fairness Doctrine era, the Supreme Court explicitly rejected an argument that television broadcast stations should be treated as common carriers.Footnote ⁶⁶

Ultimately, then, what conservative attacks on social media amount to is a claim that, unlike most owners of private property, social media firms for some reason have moral, and eventually legal, obligations to permit conservative speakers to access and use the platforms’ private property. On its face, such a claim is extraordinary. After all, in our political dialogue it is normally conservatives who defend the sacrosanctity of private property rights, fiercely resisting attempts to regulate such property via, for example, environmental regulation (no matter how strong the economic case is for such regulation). Aside from the common carrier argument, perhaps the conservative claim of a right to access platform private property might be justified if a there was a single social media platform possessing an absolute monopoly power to control public discourse; but no such monopoly exists. After all, there are several major social media platforms (Facebook, Twitter/X, Instagram, YouTube, and TikTok at a minimum) as well as numerous smaller platforms such as Gab, Parler, Telegram, and Reddit. Furthermore, the very fact that Donald Trump was able to create his own social media platform – Truth Social – from scratch following his deplatforming demonstrates that no single platform constitutes a bottleneck (or in the antitrust legal jargon, an “essential facility”) for public discourse.

Ultimately, then, one comes to suspect that the enormous conservative deviation from their general values where social media is concerned is simply a product of self-interest, not any form of principle (much the same is true of progressives, it might be added, as the next chapter will demonstrate). This suspicion tends to find support in the fact that during the lead-up to the 2024 presidential election, Elon Musk used his control over the Twitter/X platform to systematically favor conservative messages, without a peep of concern or protest from conservative crusaders for platform neutrality.Footnote ⁶⁷

In short, conservatives want to spread specific messages to their political base, and the large platforms are a cheap (indeed, free) and convenient tool for doing so. And if platform owners will not play ball, conservatives are happy to impose political pressure and legal obligations, cloaked in the name of free speech, to make them do so. Because after all, the ultimate stakes here are very high.