The Yale Law Journal


Beyond the Public Square: Imagining Digital Democracy

16 Nov 2021

abstract.To create online spaces that do not merely replicate existing hierarchies and reinforce unequal distributions of social, economic, cultural, and political power, we must move beyond the simplistic cliché of the unregulated public square and commit to the hard work of designing for democracy.

When we say ‘public square,’ . . . we need to ask—who or what is this public? Who owns this space, what makes it public? . . . This is the essence of democracy: the ability to question power, and the power to do so.1 – Tom Wilkinson


In the 2017 case Packingham v. North Carolina, the Supreme Court declared that the internet is “the modern public square.”2 This claim has been repeated so often by politicians, tech industry leaders, civil libertarians, and scholars that it has achieved the status of conventional wisdom. Proponents of the claim point not only to the vast array of activities now conducted online, but also to the public’s increasing reliance on social-media platforms such as Facebook and Twitter for discussion and debate on matters of public importance. The public square, in this view, is presumed to be the quintessential site of democratic deliberation and civic participation—a physical “marketplace of ideas.”

The concept of the “digital public square” is often presented as both a descriptive and a normative assessment. The doctrinal and policy consequences that flow from the analogy are significant. The digital-public-square view chiefly emphasizes the principle of openness to all people and all ideas. Accordingly, adherents of the digital-public-square perspective tend to view restrictions and regulations of online forums as antidemocratic and censorious. This is particularly true of governmental attempts to exclude certain speakers or listeners, such as the North Carolina law invalidated in Packingham, which prohibited sex offenders from accessing commercial social-networking sites. A more expansive version of the digital public square maintains that exclusion by private actors, such as major social-media companies, is equally or even more detrimental to democratic deliberation than exclusion by government actors.3

But if the goal is to promote a space for democratic deliberation and to realize the values underlying the First Amendment, the public-square analogy is both misleading and misguided. First, the significant differences between social-media forums—which are privately owned, virtual, and operated for profit—and physical public squares have important consequences for free speech. Second, the extent to which social-media forums do resemble physical public squares is no cause for celebration. After all, the public square has historically tended to reinforce legal and social hierarchies of race, gender, class, and ability rather than foster radically democratic and inclusive
dialogue.4 In the United States in particular, the public square has frequently served as a site for the assertion of violent white male supremacy.5 Relatedly, focusing on the public square as a uniquely significant site for meaningful democratic discourse and debate obscures the importance of governmental forums and nonpublic spaces that generate democratic discourse, debate, and activism, including homes, schools, workplaces, bookstores, hair salons, and clubs.

These troubling realities of the public square are glossed over when the concept is invoked to criticize forums deemed insufficiently “open.” This increasingly includes social-media forums when they attempt to address online misinformation, abuse, and violations of their terms of service. These attempts are pejoratively framed as introducing regulation, interference, or censorship into the public square. But the public square, like all public spaces, has never been unregulated. It has always been selectively regulated, and in ways that tend to benefit more powerful members of society at the expense of less powerful members.

If we want online spaces that do not merely replicate existing hierarchies and reinforce radically unequal distributions of social, economic, cultural, and political power, we must move beyond the simplistic and corrosive cliché of the unregulated public square and commit to the hard work of designing for democracy. Once we acknowledge that no truly inclusive, democratic, and free arena for public discourse has ever existed in the United States, either online or off, we can acknowledge that no single space is likely to achieve all of these ends simultaneously. Instead of an idealized public square, we can envision the flourishing of multiple spaces—online and off, public and private—that provide the conditions necessary for free expression and democratic deliberation.

This Essay proceeds in three parts. Part I details the significant differences between the physical public square and dominant social-media platforms. Part II describes the disturbing antidemocratic similarities between the two. And Part III offers ways to think beyond the concept of the public square when designing democratic online spaces.

I. the disanalogy of the public square

In its most literal sense, the public square is a physical space, open to the public and usually managed by the government, where people gather. Famous public squares include the ancient Agora in Athens, the Piazza San Marco in Venice, and Times Square in New York City.6 The term “public square” can also refer to other publicly accessible and governmentally managed locations, such as parks and sidewalks.7

It is certainly true that people also gather on social media, but beyond that, the analogy to the physical public square is strained. The dominant social-media platforms, including Facebook, Twitter, and YouTube, are privately owned, operated for profit, and virtual. Despite marketing rhetoric that emphasizes inclusion, community, and communication, these platforms are designed to serve corporate, not public, interests. With few exceptions, this means that they are designed to extract as much attention and information from people as possible for commercial purposes. The prioritization of “engagement” above all else creates perverse incentives for harassment, invasions of privacy, and false information that can destroy reputations, lives, and democracy itself.

A. Private Versus Public Distinction

In the United States, the concept of the public square is closely identified with the concept of “public fora,” which occupy a “special position in terms of First Amendment protection because of their historic role as sites for discussion and
debate.”8 That special protection generally means that governmental restrictions on public forums are heavily disfavored as infringements on free-speech rights. In the 2014 case McCullen v. Coakley, the Court justified its laissez-faire approach by emphasizing the physical, face-to-face nature of interactions within public forums:

It is no accident that public streets and sidewalks have developed as venues for the exchange of ideas. Even today, they remain one of the few places where a speaker can be confident that he is not simply preaching to the choir. With respect to other means of communication, an individual confronted with an uncomfortable message can always turn the page, change the channel, or leave the Web site. Not so on public streets and sidewalks. There, a listener often encounters speech he might otherwise tune out.9

Notably, in McCullen, the Court contrasted websites to the public square.10 According to the Court, websites are similar to print publications and television and radio stations, which allow people to simply “tune out” speech, and dissimilar to physical spaces such as public streets and sidewalks, where people are forced to confront viewpoints with which they disagree.11

In the Packingham decision just three years later, however, the Court characterized the internet as “the modern public square.”12 While the Court reaffirmed that “a street or a park is a quintessential forum for the exercise of First Amendment rights” and that “[e]ven in the modern era, these places are still essential venues for public gatherings to celebrate some views, to protest others, or simply to learn and inquire,”13 Justice Kennedy declared that “cyberspace” was the new true home for free speech:

While in the past there may have been difficulty in identifying the most important places (in a spatial sense) for the exchange of views, today the answer is clear. It is cyberspace—the “vast democratic forums of the Internet” in general and social media in particular.14

Some observers understood this characterization as a signal that the Court had not only come to view physical public spaces and virtual forums as symbolically equivalent, but also that it planned to apply the First Amendment public-forum doctrine to social-media sites.15 This would mean that the private companies controlling those sites would be treated as state actors and therefore be severely limited in the actions that they could take to restrict either speakers or speech on their sites.

Such a move would have been a radical departure from traditional First Amendment doctrine and state-action doctrine more broadly. That is because the most widely used and influential social-media platforms are privately, not publicly, owned. The free-speech right protected by the First Amendment is a negative right that applies only to government action, and the state-action doctrine maintains a fundamental distinction between government and private actors. Private actors are not subject to the restraints of the First Amendment, except in the rare cases where they perform a “traditional, exclusive public function” or where the government compels or collaborates with private actors.16

Disappointing as it may have been to those who hoped the Supreme Court would expand or obliterate the state-action doctrine with regard to private forums that host speech, the Court made clear in Manhattan Community Access Corp. v. Halleck that “merely hosting speech by others is not a traditional, exclusive public function and does not alone transform private entities into state actors subject to First Amendment constraints.”17 In the majority opinion, Justice Kavanaugh noted that to hold otherwise would be to intrude upon a “robust sphere of individual liberty.”18 According to Kavanaugh, this would be “especially problematic in the speech context, because it could eviscerate certain private entities’ rights to exercise editorial control over speech and speakers on their properties or platforms.”19 In other words, protecting free speech in a private forum requires the exact opposite of what it takes to protect free speech in a public forum: private actors must be allowed to exercise their free-speech rights to counter, ignore, or exclude speech as they see fit, even where state actors would be restrained from doing so.

Though Halleck involved a cable channel, not a social-media platform, the case had clear implications for the application of the state-action doctrine to the internet. Until very recently, tech companies themselves contributed to the misperception of their platforms as public forums and of themselves as state actors by invoking First Amendment principles as a justification for failing to address misinformation, abuse, and harassment on their platforms.

Reliance on First Amendment law is “a common theme” of major technology platforms.20 For years, social-media companies have characterized their notoriously lax policies against harmful content and conduct as a virtuous defense of free speech and open debate, often creating the impression that their failure to intervene is compelled by the First Amendment.

But as the harmful consequences of misinformation, extremism, and surveillance facilitated by their platforms have become increasingly apparent, some of the most influential companies have begun to take belated, modest steps to address them.21 These steps include labeling false or misleading content,22 removing content that violates the sites’ policies,23 and banning certain users.24 Perhaps most notably, both Facebook and Twitter banned former President Donald Trump following the January 6, 2021 Capitol insurrection.25

These actions have been met with cries of censorship, many of them coming from the political right. Prominent conservative figures, including former President Trump, Senators Ted Cruz and Josh Hawley, and Florida Governor Ron DeSantis, have demanded that social-media companies be prohibited from making their own decisions about what kind of speakers and speech they wish to allow on their platforms.26 These demands are often framed as necessary to protect First Amendment rights.27

But this framing gets First Amendment doctrine backwards. While the First Amendment constrains the ability of the government to regulate speech, it protects the right of private actors to do so. The rights of free expression and association include the right not to speak and the right not to associate. Indeed, the First Amendment demands respect for these companies’ right to fact-check, label, remove, ban, and make other interventions. Providing additional or alternative information to false or misleading posts is classic “counterspeech,” a treasured First Amendment value famously identified by Justice Brandeis in Whitney v. California, a landmark free-speech case. According to Brandeis, “If there be time to expose through discussion the falsehood and fallacies, to avert the evil by the processes of education, the remedy to be applied is more speech, not enforced silence.”28 The First Amendment also protects the right to refuse to host content altogether, as the right to free speech includes both the right to speak and the right not to speak. As the Supreme Court held in West Virginia State Board of Education v. Barnette, “If there is any fixed star in our constitutional constellation, it is that no official, high or petty, can prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion, or force citizens to confess by word or act their faith therein.”29

The First Amendment also protects the right of association, including the right of private actors to choose with whom they wish to associate.30 And the Supreme Court has long recognized that private-property owners generally have the right to exclude individuals from their property as they see fit.31 These are rights that conservatives and libertarians have long championed as essential to private liberty and the free market.32 Collapsing the public/private distinction would not just have consequences for tech companies, but for all private actors, and it would lead to absurd and troubling results. Without this distinction, restaurants would arguably have no right to refuse service to belligerent customers, private clubs could not maintain membership qualifications, and homeowners could not eject unwelcome visitors.

It may be argued that social-media platforms differ from these examples in important respects: they often characterize their services as being as open to anyone, they provide for what many people consider to be essential functions of modern life, and they exert greater influence over public opinion than other private actors. But while access to the internet might plausibly be characterized as essential to modern life, the same is not true of access to any particular social-media forum. And while social-media forums may feel like public spaces, and the companies that own them may exploit this perception to their advantage, their relationship to the public is fundamentally commercial and contractual. As Astra Taylor writes, “They are commercial enterprises designed to maximize revenue, not defend political expression, preserve our collective heritage, or facilitate creativity, and the people who work there are private employees, not public servants.”33 There are very good reasons to demand better consumer protections against arbitrary treatment and predatory data extraction by tech companies, but it does not follow that such companies are state actors for the purposes of the First Amendment.

Concern about the unchecked power of social-media platforms to shape public discourse is certainly justifiable, and has been voiced by scholars across the political spectrum.34 But this observation is a reason to move further away from the analogy of the public square, not closer to it.

B. Physical Versus Virtual Space

A second obvious difference between the public square and the internet is that the former is a physical location where identifiable and observable individuals interact in person, whereas the latter is a virtual environment in which individuals rarely see each other face to face and are often completely anonymous and untraceable.35 The disinhibition, amplification, permanence, and captivity of virtual interactions has a significant impact on the dynamics of communication in real-world versus virtual spaces.36 While online communication expands the boundaries of public discourse in many ways, it also allows for destructive forms of abuse that chill expression and inhibit participation in public debate.

The fact that social-media forums do not provide face-to-face interactions in physical spaces can have many salutary effects on free speech and debate. Large gatherings such as marches and protests present risks of physical harm, may require travel or long hours of standing, and can be challenging for older individuals and those with disabilities.37 Organizing such events often requires time-consuming paperwork, intensive planning, and resources.38 In person, individuals may feel intimidated by speakers who are physically larger, aggressive, or armed.39 Some people need time and reflection to express themselves that are not always available in-person.40

In contrast, online communication can happen anywhere, including from the comfort of one’s own home. While virtual communication can pose risks to physical safety, it does not do so in the same immediate physical sense as in-person communication. Many social-media platforms have few or no barriers to entry or participation, flattening the hierarchy between elites and the general public and making it possible for people to contribute to or join discussions almost instantly. Online platforms also make it easier to conduct asynchronous conversations, allowing for increased possibilities to ponder, prepare, and revise one’s thoughts. And because online communication is global, interactions on social-media platforms often involve a much wider range of people and ideas than is possible in physical gatherings.

At the same time, many characteristics of virtual interactions negatively impact communication and debate. Chief among these is the “online disinhibition effect,” or “how people say and do things in cyberspace that they wouldn’t ordinarily say and do in the face-to-face world.”41 Numerous studies have demonstrated that people are more likely to engage in abusive behavior when they feel insulated from the consequences, including when their identity is concealed.42 This anonymity facilitates disinhibition, which can have positive effects on the free exchange of ideas, but can also encourage reckless and destructive behavior that many individuals would eschew in person.43 Anonymity also makes detection, intervention, and accountability with regard to such behavior much more difficult.

Abusive behavior often has a chilling effect on targeted individuals’ expression. This effect can be more severe online than in person due to the amplification afforded by the internet.44 The internet provides abusers not only with loudspeakers, but also with extensive opportunities to allow others to join in on their abuse.45 The internet and social media make it possible to “crowdsource” harassment, an aggregating effect that greatly increases the negative impact of harassing behavior. Social-media platforms in particular “give cyber harassment campaigns the ability to go viral, because they allow for ‘near instantaneous, widespread dissemination.’”46

The nature of the internet also extends the shelf life of abusive content. It is extremely difficult, if not impossible, for abusive or privacy-violating content to be removed once it has been posted online.47 Such permanence creates the possibility that targets of harassment may never regain peace of mind, resigning them to a constant state of anxiety or fear about the exposure of their private information: “[E]ntering a doxing victim’s name into a search engine may reveal her personal details and the abuse associated with the doxing attack for years.”48 As Danielle Keats Citron writes, “Harassing letters are eventually thrown away, and memories fade in time. The web, however, can make it impossible to forget about malicious posts. Search engines index content on the web and produce it instantaneously. Indexed posts have no built-in expiration date; neither does the suffering they cause.”49

Finally, the global reach of the internet can make abuse and harassment virtually impossible to escape. Unlike offline harassment, which may be restricted by geography or time, online harassment can manifest anywhere at any time: “[V]ictims’ pasts follow them, even as they move physically.”50 If, as is often the case, the online abuse is indexable by a major search engine, it is accessible to almost anyone (the target’s coworkers, fellow students, clients, children), almost anywhere (at the target’s place of work, her school, her home, her doctor’s office).51

The significant differences between physical and virtual spaces provide compelling reasons to design and govern them in different ways. Features of online interaction such as disinhibition, amplification, permanence, and captivity create powerful incentives for harassment and abuse, especially of vulnerable individuals and groups.52 Given the relatively unbounded nature of online forums as opposed to physical forums, and the increased potential for lasting and irreparable harm, social-media forums are justified in taking a more assertive role in moderating or curating content so as to avoid chilling expression and inhibiting participation in public debate.

The features that distinguish social-media platforms from the physical public square—being privately owned, operated for profit, and virtual—illustrate not only the limitations of the public-square analogy, but also the potential of private companies to engage in innovation and experimentation. Private companies, as nonstate actors, have tremendous freedom to design their platforms to encourage the values of free speech and democratic deliberation.

II. the problem with the public square

As discussed above, the internet differs from the public square in many ways. But no analogy is perfect, and social-media forums do resemble public squares in some ways, if not all. Those similarities are, however, cause for concern rather than celebration. Though it is undeniable that public squares have intermittently functioned as powerful sites of resistance—through protests against unfair economic and labor practices,53 marches for civil rights,54 and demonstrations for gender equality55—the public square has never truly been public: law and norms have always served the powerful at the expense of the vulnerable. Indeed, public squares have always served more to reinforce legal and social hierarchies than to facilitate open and inclusive democratic deliberation.56 For example, in the United States, the public square has played a key role in the assertion of violent white male supremacy.57 This is one reason why meaningful debate and organization for social change have often flourished in other spaces beyond the public square, including homes, schools, workplaces, hair salons, bookstores, clubs, and governmental forums.58

From the Greek agora to the city sidewalk, the physical public square has been a site of exclusion, hostility, surveillance, and silencing of vulnerable and minority populations, including women, nonwhite men, the poor, and the disabled.59 In the United States, slavery and segregation excluded Black people outright from public spaces;60 later, selectively enforced surveillance and stop-and-frisk practices deterred Black people from entering them.61 Women in early America were relegated to the “private sphere” of the home62 and deprived of significant opportunities to earn money,63 receive an education,64 or move freely in public without a male companion. Women in public spaces were targeted by antiprostitution and other public-decency laws,65 and today they continue to contend with pervasive street harassment and sexual assault.66 Vagrancy laws targeted the poor and unemployed for harassment and arrest,67 and hostile architecture has rendered many public spaces inaccessible to those without homes and those with disabilities.68 Public spaces are also heavily marked with symbols of whiteness, masculinity, and dominance, from the names of streets and parks to the subjects of statues and monuments.69

Indeed, white male supremacy is so fundamentally intertwined with the public square that attempts to challenge or remove symbols of racial and gendered hierarchy have been met with violence. The deadly 2017 “Unite the Right” rally in Charlottesville, which led to the deaths of three people and injured dozens, began as a protest against the removal of a statue of Robert E. Lee and the renaming of Lee Park as Emancipation Park.70 The rally evoked memories of the 1939 Nazi rally in Madison Square, where more than 20,000 people gathered in front of a giant portrait of George Washington flanked by swastikas.71 The 2017 rally, and several white-supremacist gatherings in public squares that preceded and followed it, was touted as a demonstration in support of free-speech rights. When the city of Charlottesville attempted to have the rally moved to a different location on the grounds of public safety, the American Civil Liberties Union (ACLU) stepped in to ensure that it took place in Emancipation Park, arguing that the far-right organizers’ “choice of location is critical to the message of the rally” and that holding the rally in another location “would dilute and alter” their message.72

Many of the Unite the Right protestors showed up with weapons, as well as swastikas and Confederate flags, raising questions about the coherence of the ACLU’s defense of the protest as an exercise in free speech.73 The threat of violence is fundamentally incompatible with free expression. As one commentator describes it, the violence of such rallies “threaten[] public space, an amenity that is both scarce and necessary for democracy. . . . [T]he extremist alt-right is waging a campaign to shut down the public square, using both violence and intimidation, especially under open-carry

On January 20, 2020, more than 20,000 gun-rights activists flooded into Richmond, Virginia to protest gun laws passed by the newly Democratic legislature.75 The modest gun regulation measures that triggered this demonstration were adopted through a lawful democratic process and were supported by the majority of Virginians.76 This did not stop protesters from describing state legislators who voted for the measures as “tyrants;” one legislator who had been wrongly identified as the sponsor of a gun control measure was forced into hiding due to death threats.77 The protesters brought rifles, “Don’t Tread on Me” and militia flags, and a homemade guillotine inscribed with the words “[t]he penalty for treason is death.”78 Fear of violence led to the cancellation of planned counterprotests and Martin Luther King Day celebrations, drove lawmakers from the Capitol, and brought everyday activities in the city to a halt.”79

The event was described by many commentators as “peaceful.”80 But as journalist Jamelle Bouie observed, “that ‘peace’ can’t be separated from intimidation.”81 The demonstration in Richmond was a literal and symbolic takeover of the public square: a demonstration against free speech and democracy itself. A similar takeover occurred in Michigan in April 2020, when hundreds of armed, unmasked demonstrators crowded into the State House to protest Governor Gretchen Whitmer’s stay-at-home order during the early days of the COVID-19 pandemic.82 An even starker display of intimidation and anarchy unfolded on January 6, 2021, when far-right Trump supporters stormed the Capitol in an attempt to violently overturn the results of a democratic election.83

With regard to unchecked racism, sexism, and extremism, the digital public square does often indeed resemble the physical public square. Those same forces are potent online, causing similar intimidation and silencing. White male supremacism dominates the online landscape, from Facebook groups planning insurrections to Nazi propaganda on Twitter to QAnon videos on YouTube.84 Rampant sexist and racist harassment on these forums—including rape threats, revenge porn, doxing, and organized campaigns of racial hatred—silence women and minorities and push them out of public discourse.85 When social-media platforms fail to address these abuses in the name of “free speech,”86 the result is not open debate and democratic deliberation, but an exclusionary and elitist echo chamber.

Racist and sexist abuse chills speech and undermines the rights, liberties, and wellbeing of its targets. As Cynthia Grant Bowman observes, “[T]he continuation and near-general tolerance of street harassment . . . inflicts the most direct costs upon women, in the form of fear, emotional distress, feelings of disempowerment, and significant limitations upon their liberty, mobility, and hopes for equality.”87 According to a 2014 study on street harassment, women respond to street harassment by no longer visiting certain places alone; changing the way they walk, behave, or dress; foregoing outdoor activities; quitting jobs; and moving.88 Mari J. Matsuda writes that victims of racist speech have similarly had to “quit jobs, forgo education, leave their homes, avoid certain public places, curtail their own exercise of speech rights, and otherwise modify their behavior and demeanor.”89

As with offline abuse, online abuse and its silencing impact are unevenly distributed across society.90 Women, especially younger women, experience more severe forms of online abuse than other groups. A 2016 report by Data and Society found that “[y]ounger women are more likely to be stalked, sexually harassed, and harassed online over a long period of time compared to men and older women.”91 The more frequent and more serious forms of online abuse that women experience help explain why women self-censor and digitally disengage at greater rates than men.92 According to the Data and Society study, “younger women are most likely to self-censor to avoid potential online harassment: 41% of women ages 15 to 29 self-censor, compared with 33% of men of the same age group and 24% of internet users ages 30 and older (men and women).”93 Other studies have documented women’s greater likelihood of self-censorship in the face of online abuse.94 A survey of women journalists found that 63% of respondents had been threatened or harassed online at least once, and that approximately 40% said they avoided reporting certain stories as a result of this harassment.95

Another corrosive effect of this online abuse is that it interferes with women’s political participation, and thereby diminishes democratic discourse and governance. A recent United Nations (UN) Inter-Parliamentary Union report found that “[p]olitical abuse is a distinctly gendered phenomenon,” and that “sexism, harassment and violence against female politicians was a ‘phenomenon that knew no boundaries and exists to different degrees in every country.’”96 The consequences of this gendered harassment include not only reluctance to run for reelection, but also self-censorship while in office. Many of the targeted female politicians “self-censor, particularly when it comes to speaking up about women’s issues, which tend to generate the most aggression.” Some even “dispense with social media altogether, and in this way deprive themselves ‘of a forum in which to disseminate and debate their ideas.’”97

An array of empirical studies demonstrates that the targets of bigoted speech commonly experience not only psychological effects—lack of confidence, social anxiety, fear—but also physiological effects, such as increased heart rate and stress.98 This in turn can lead to targets censoring themselves as a means of avoiding these negative effects.99

In spaces where intimidation and abuse of vulnerable groups is rampant, a laissez-faire approach only reinforces status-quo inequalities in free expression and civic participation. The insistence on remedies such as counterspeech ignores how responding to harmful speech is a form of compelled labor (and speech) that depletes time, effort, and energy from other expressive activities. While counterspeech can be powerful, as it is when social-media platforms label false or misleading information (discussed above), it is often less effective and highly costly for private individuals with few resources and little influence. It can also backfire by exposing the individual to increased attention and humiliation. The valorization of counterspeech also ignores the problem of “unanswerable” speech, that is, speech that is resistant to the remedy of counterspeech.100 There is no counterspeech, for example, to the publication of a person’s nude image, the dissemination of a home address, or the disclosure of undocumented status.101

The negative impact of such chilling effects is not limited to the individuals who are targeted for abuse. This selective silencing impoverishes public discourse as a whole. Social-media sites “function as hosts for public conversations on a huge variety of social issues,” Alice Marwick observes, and so “[i]f women, people of color, and LGB[TQ+] internet users are shying away from contributing because of well-founded fears of retaliation, their voices will be missing from this important civic sphere.”102 Such a result directly undermines democracy, as Citron and others have argued: “An online discourse which systematically under-represents people—particularly women and people of color—cannot effectively process our various attitudes and convert them into truly democratic decisions.”103

III. beyond the public square

The question then becomes, what lessons can be learned from the failures of the physical public square? What is the theory of the public square that provides a way to normatively assess current structures and approaches to public interaction? What are the salient differences between offline and online discourse that might provide paths forward to greater democracy and free speech? What are the thoughtful, intentional design choices or commitments that can be made to create true sites of public discourse and rational deliberation?

Part of the problem with the concept of the public square is that it is invoked in imprecise and undertheorized ways. It is often used to describe places or activities in which contentious speech takes place, even though contentious speech alone is not a marker of democratic deliberation. For a more sophisticated view, we can look to the philosopher Jürgen Habermas, who offers an influential and comprehensive description and theory of the public sphere. For Habermas, the public sphere must be distinct from the state; it must be a “site for the production and circulation of discourses that can in principle be critical of the state.”104 It must also be distinct from the market and instead “a theater for debating and deliberating rather than for buying and selling.”105 The public sphere is a place accessible to all private citizens, where they can discuss matters of common interest in an unrestricted manner.106

Habermas’s theory of the public sphere is useful in many respects. It demonstrates how social-media forums cannot constitute a public sphere because they are not distinct from either the state or the market. Powerful government figures wield tremendous influence in social-media forums, and the forums themselves are for-profit operations of private businesses.

But Habermas’s theory also replicates the false claims of more popular views of the public square. As Douglas Kellner writes, “while the concept of the public sphere and democracy assume a liberal and populist celebration of diversity, tolerance, debate, and consensus, in actuality, the bourgeois public sphere was dominated by white, property-owning males.”107

One of Habermas’s most trenchant critics, Nancy Fraser, directly addressed limitations of the concept of the public sphere, given its elitist and exclusionary history:

Should we conclude that the very concept of the public sphere is a piece of bourgeois masculinist ideology, so thoroughly compromised that it can shed no genuinely critical light on the limits of actually existing democracy? Or, should we conclude, rather, that the public sphere was a good idea that unfortunately was not realized in practice but that retains some emancipatory force? In short, is the idea of the public sphere an instrument of domination or a utopian ideal? Well, perhaps both. But actually neither.108

Fraser emphasizes that one of the greatest flaws in the theory of the public sphere is its unitary focus. Given how the public sphere reinforces existing power structures, “where there is only a single, comprehensive public sphere . . . members of subordinated groups would have no arenas for deliberation among themselves about their needs, objectives, and strategies.”109 Instead, society should be arranged so as to “accommodate contestation among a plurality of competing publics” in order to “better promote the ideal of participatory parity.”110 As an example, Fraser points to what she refers to as the “U.S. feminist subaltern counterpublic, with its variegated array of journals, bookstores, publishing companies, film and video distribution networks, lecture series, research centers, academic programs, conferences, conventions, festivals, and local meeting places.”111 Jane Mansbridge suggests adding factories, which “unexpectedly brought workers together to share their experiences and black colleges that initiated the sit-ins of the civil rights movement.”112 These counterpublics “have deliberative uses even for members of dominant majorities, but are crucial for the marginalized as a protection against hegemonic discourse.”113

In place of an idealized, unitary public square, we can envision the flourishing of multiple spaces—online and off, public and private—that provide the conditions necessary for free expression and democratic deliberation for different groups with different needs. This vision would entail crafting law and policy to ensure that no single host or forum, or even single medium, dominates the shaping of public opinion. Given the current domination of internet communication, this will likely require a combination of robust antitrust interventions and Section 230 reforms to break the stranglehold of the tech industry on social discourse, as well as serious investments in traditional media, public education, universities, community centers, and small businesses to return to or become alternate sites of free expression and informed debate.

Given the complex power dynamics and social striations of American society, effective counterpublics must necessarily adopt nimble and innovative principles specific to various groups’ evolving needs in order to be effective. But a few general guidelines for the would-be architects of democratic counterpublics can be sketched. One is the rejection of the faux laissez-faire approach of the idealized public square in favor of an intentional commitment to designing for democracy. Here, principles of universal-design theory can be instructive,114 in particular what is known as the “curb-cut effect.”115

The idea of the curb-cut effect is, in essence, that measures designed to accommodate people with particular needs end up benefiting everyone. As the story goes, one evening in the 1970s in Berkeley, a group of disability advocates poured cement on a sidewalk to make a ramp so that people in wheelchairs could more easily navigate the daunting gap between the sidewalk and the street.116 This was a makeshift “curb cut,” and they were among the environmental accommodations mandated by the 1990 Americans with Disabilities Act.117 Angela Blackwell describes the “magnificent and unexpected thing” that happened when the “wall of exclusion came down” through curb cuts: “[E]verybody benefitted—not only people in wheelchairs. Parents pushing strollers headed straight for curb cuts. So did workers pushing heavy carts, business travelers wheeling luggage, even runners and skateboarders. . . . [N]ine out of 10 ‘unencumbered pedestrians’ go out of their way to use a curb cut.”118 What the curb-cut effect demonstrates, writes Blackwell, is that accommodation is not a zero-sum game: “[W]hen we create the circumstances that allow those who have been left behind to participate and contribute fully—everyone wins.”119 Spaces that accommodate a greater range of people are also, in many cases, superior spaces.

Creating these spaces will require what scholar Mari Matsuda calls “looking to the bottom”—that is, consulting the expertise of those who have been exploited and excluded.120 Matsuda explains that this involves not an abstract exercise of imagination, but actual collaboration with “grass roots philosophers who are uniquely able to relate theory to the concrete experience of oppression.”121

Looking to the bottom also provides a framework for dealing with what academics call “rivalrous goods”: “goods whose consumption by one consumer prevents simultaneous consumption by other consumers.”122 While it will be true in many cases that designing for the most vulnerable will lead to increased benefits for all, there will also be cases in which accommodating the vulnerable may lead to a reduction in benefits for more privileged groups. In those situations, a choice must be made whether to preserve the status-quo benefits of those with power or to shift the status quo in favor of those who have historically been excluded and exploited. A true commitment to public welfare and democratic deliberation requires choosing the latter:

When the law must choose among realities, the principle of equality requires that we look to see whose dignity is most at stake, whose point of view has historically been silenced and is in danger of being silenced again, and that, in the ordinary case, we choose that point of view as our interpretation.123

What this means in practice is that those who wish to create spaces of true democratic deliberation should abandon the position of false neutrality and step fully into the role of curator. Those who want their private spaces to become sites of democratic deliberation should reject the myth of the public square and embrace more creative and innovative models for encouraging free expression and critical thinking.

One model to consider is the academy, with its emphasis on scholarship, rigor, and norms of civil interaction.124 In its most noble form, the project of higher learning encourages critical reflection and intellectual evolution. Such a project requires discernment and evaluation, not the simplistic embrace of “both sides.” As Justice Felix Frankfurter wrote in Wieman v. Updegraff, democracy is built on “disciplined and responsible” public opinion, and “[i]t is the special task of teachers to foster those habits of open-mindedness and critical inquiry which alone make for responsible citizens, who, in turn, make possible an enlightened and effective public opinion.”125

Another potential model is the Enlightenment-Era, European salon. While the salon (like the academy) was by no means free of the trappings of wealth and privilege, the ideals it promoted were egalitarian, democratic, and revolutionary for their time.126 According to historian Dena Goodman, the salon played a key role in the development of the concept of the public sphere.127 Unlike the historical public sphere, however, salons were ruled by women. Known as “salonières,” these women created spaces in their homes for people of diverse backgrounds to debate politics, art, and culture in a safe and respectful setting.128 Salonières were influential Enlightenment figures who demonstrated “a respect for all opinions, a refusal to prejudge, and a distaste for orthodoxies of all kinds.”129 They invited a wide variety of individuals to participate in interactive, informal education outside of elite institutions:

Those participating in a salon understood that they were entering a space in which typical forms of valorization such as wealth, social status, or family lineage were not prioritized, or to a far lesser extent than in other forms of social interaction. What mattered most were the ideas and knowledge that could be gained from contemplation that benefited the collective. Erudition, wit, inventiveness, the ability to poetically capture an idea or elegantly communicate a concept, these were the cardinal virtues of the salon. Equally, the manner in which ideas were pursued was of great importance. Self-love and arrogance were discouraged for they signaled the wrong motivation for participation.130

It is worth imagining what social-media platforms might look like if they were modeled on the ideals of the academy or the salon, rather than the public square: a variety of semiprivate, secure spaces where members of the public could explore and discuss politics, art, and culture in safety and mutual respect.

They might look something like MetaFilter, a once-iconic weblog that has existed since 1999. On its “About” page, MetaFilter (MeFi for short) states: “Here you can expect thoughtful and varied discussions. Since 1999, we’ve been focused on fulfilling the web’s potential to bring people together and create genuine, vibrant, good-hearted community spaces.”131 Content on the site is diverse, ranging from news to art to politics, and includes a question-and-answer subsite called Ask MeFi.132 Unlike most social-media forums, MetaFilter restricts participation on the site in several ways.133 First, while anyone can view content on the site, individuals must register in order to post content. They must also pay a one-time, five-dollar fee to participate in the forum.134 Participants are also subjected to mandatory waiting periods between
posts.135 In naming MetaFilter one of the “50 Best Websites” of 2009, Time Magazine wrote that the subscription fee “ends up feeling like a feature rather than an impediment, because it manages to keep the site remarkably free of trolls, griefers and other anonymous jerks,” and that the site “has the public-spirited flavor of a small town or good university.”136

MetaFilter is much less well-known today. Its decline in prominence can be traced to 2012, when MetaFilter experienced a dramatic drop-off in traffic following the “Panda” update to Google’s indexing algorithm.137 The update was intended to promote high-quality content and downrank sites using search-optimization techniques or spam.138 But there was already a consensus that MetaFilter was a high-quality site. Therefore, it should not have suffered negative consequences from the update, raising the suspicion that Google was using its black-box powers to crush small communities.139

MetaFilter’s decline raised other, deeper issues about the changing nature of what we see and how we connect online. As Caitlin Dewey wrote in the Washington Post, “the most striking, prescient takeaway from the whole MetaFilter episode” is “the extent to which the modern Web does not incentivize quality.”140 Sites that deliberately set standards to encourage thoughtfulness, reflection, and consideration of community norms constantly lose out to sites that reward impulsivity, provocation, and narcissism.

But it does not have to be this way. Instead of funneling humanity into a handful of corporate-owned, cacophonous spaces, the internet could provide pathways into a multitude of settings designed to serve a diverse array of personal, cultural, and intellectual interests. As the late Congressman John Lewis wrote, democracy is not a state, but an act.141 It requires deliberate design and constant commitment, in online spaces as well as offline.


Cyndi Suarez writes that “the history of democracy is about the fight for the public square—who speaks and who doesn’t, whose issues matter and whose are marginalized, who can congregate and who is intimidated.”142 She asks, “What happens when the government does not step in to ensure peace and order, and may even instead stoke chaos and entropy? Can we rely on public opinion and counter-demonstrations, using constructive, unifying speech to counter destructive, segregating speech?”143 In the United States, calls to protect the “public square,” whether physical or virtual, are efforts to maintain its status-quo domination by the white, wealthy, and male. Rather than function as a site of free expression and democratic deliberation, the public square excludes and exploits women, nonwhite men, and other vulnerable groups.

If we move beyond the public square, we can imagine a multitude of spaces designed for reflection instead of performativity; accessibility instead of exclusion; and intellectual curiosity, humility, and empathy instead of ignorance, arrogance, and cruelty. We can imagine spaces designed for democracy.

The Author would like to thank Edwin Elliott, Anabelle Torek, and Megan Valent for their valuable research assistance and insights.