This dramatic print shows the destruction of Pennsylvania Hall, a large building that was constructed in 1837--38 at Sixth and Haines Streets in Philadelphia as a meeting place for local abolitionist (antislavery) groups. Dedication ceremonies began on May 14, 1838, and continued over several days in a climate of growing hostility from anti-abolitionist forces in the city. On the night of May 17, 1838, an anti-abolitionist mob stormed the hall and set it on fire. Fire companies refused to fight the blaze, and the building was completely destroyed. A large crowd looks on, as firefighters spray water on an adjoining building.

When a pro-slavery mob set fire to Pennsylvania Hall in downtown Philadelphia on the night of May 17 1838, there was no doubt that the arson was a threat to basic freedoms. Designed as a “Temple of Free Discussion,” the building was designed to house meetings focused on the abolition of slavery and women’s rights. Abolitionists had been forbidden from other buildings, so they decided to crowd-fund their own, organizing a network of a hundred thousand people to give money, materials, and labor to its construction. 

In Pennsylvania Hall’s first four days, the event space reportedly attracted thousands of visitors of all genders and races. Political tensions were already high. Just months earlier, the state of Pennsylvania had extended voting rights to white men without land while stripping the vote from Black citizens. On the fourth day after its opening, a mob set the building aflame. With fire companies refusing to protect the building (they saved only the adjoining ones), the meeting hall was soon reduced to rubble.

What can we learn from the Temple of Free Discussion in our search for flourishing digital spaces? And how can behavioral science help us make sense of these stories? The attack on Pennsylvania Hall was not an attack on individuals but on infrastructure. As the fire disrupted the physical capacity of groups to meet, the flames also did invisible damage to the diverse relationships that would otherwise have formed. To protect the social fabric from these invisible relational threats, the rights that we invoke need to be relational in turn — freedom of association and assembly. By focusing science on these collective rights and capabilities, we can support the kind of societies where people form relationships and take collective action free of intimidation and fear.

To protect the social fabric from relational threats, the rights that we invoke need to be relational in turn — freedom of association and assembly

While online communities and their technical infrastructure can’t easily be burned to the ground, digital groups frequently face attempts to deprive them of the capacity to communicate and assemble. While news articles often focus on design problems like shadow bans and de-platforming by tech firms, risks such as online harassment, disruption, and threats of violence can also prevent people from forming connections and collective action. In CAT Lab’s research for example, we have studied how online harassment deters people from joining communities in the first place. In the U.S. today, forty percent of Americans — well over a hundred million people — have faced some kind of online harassment, often for their politics or identity. A swarm of bots or restrictive platform design may not be as dramatic as a three-story building on fire, but its effects on relationships and the fabric of society can be even more widespread.

A swarm of bots and hostile commenters may not be as dramatic as a three-story building on fire, but its effects on relationships and a community can be even more widespread

As social and behavioral scientists, we think about people’s rights in terms of how people actually exercise those rights. To put it simply, anything that helps people exercise their freedoms is a positive contribution to rights, and anything that dissuades people from exercising them is a negative force on people’s rights, whether or not the intervention uses the tools of law (Citron & Penney 2018; McAdams 2015). At CAT lab, we contribute to the science of rights by studying the effects of policies that potentially reduce people’s freedoms, by trying to neutralize those negative effects. We also test ideas for increasing how people responsibly exercise their rights.

Freedom of Assembly in the Digital Age

How would a behavioral scientist think about the freedom of assembly? In legal theory, assembly is the right of groups to have autonomy — for people to form, meet, choose leaders, and manage themselves. While assembly rights have historically been debated for political and religious groups, these rights also apply to any group, even if a group holds views that governments find objectionable, so long as they don’t break the law in some other way (Inazu 2012 p 166). Online, this freedom applies to the full range of such groups, from a massive political discussion space on WhatsApp to a family chat circle — whether or not they exist as a legal entity or own a building.

Even if the law focuses on protecting groups from government restrictions and corporate union-busting, social and behavioral scientists can study a wider set of forces that grow, maintain, and threaten groups. For example, behavioral scientists can help explain how intimidation and threats dissuade people from joining an existing group or discourage them from organizing a new group. An example would be CAT Lab’s research that tests ways to reduce the chilling effects of harassment on the actual rates at which people participate in groups under attack. Our work to broaden inclusion in African Wikipedia communities also represents a contribution to freedom of assembly, since we set out to understand and reduce the barriers to participation in collaborative public goods.

Freedom of Association in Social Technologies

People also have relational rights that are independent of our connection to groups. In the book Creating Capabilities, the philosopher Martha Nussbaum tells the story of Vasanti, a woman in Ahmedabad in the state of Gujarat in India. In the story, Vasanti leaves an abusive husband, returning to her family before joining an NGO that helps her make a new start. Nussbaum writes, “When a woman can leave [a marriage], she doesn’t have to endure being beaten.” While the freedom of assembly made the NGO possible, Vasanti was alive thanks to association — her right both to form and to end relationships.

Because social technologies formalize relationships, tech companies currently exercise unprecedented power over people’s rights of association. They define who we meet, who we can contact, and the basic terms on which we relate to others. Companies enable stalking, hinder people from escaping abusive relationships, monitor our relationships to decide what kinds of relationships are acceptable, and put cost barriers on the right to meet others. For a social or behavioral scientist, the right of association is about more than the design of software. Threats, intimidation, and other forms of control can shrink a person’s social life as easily as an account ban.

The Role of Platform Design in Relational Rights

Just as other people shape our uptake of freedoms, design can also enhance or threaten basic rights. In human-computer interaction, the term “affordances” refers to ways that technologies allow or disallow certain behaviors (Ronzhyn et al., 2023). Since platform design shapes how we’re able to engage (or disengage) with people and groups, design has direct impacts on relational rights. Two clear examples are association and visibility (Treem & Lombardi, 2012).

design can enhance or threaten basic rights

Designers use the term “association” to describe affordances that support people to establish connections between each other and content. Writing in 2012, Treem and Lombardi explain the fundamental concepts of design that support a wide range of social connections. Examples include the ability to:

  • form relationships, for example, follow people on BlueSky, friend people on Facebook, use hashtags to follow and participate in conversations on X, or join a Discord group or subreddit. 
  • dissolve relationships, by unfriending or unfollowing an individual or a group, muting people or hiding specific kinds of content, blocking people, or migrating a community to another platform 
  • respond and react to people and content, by liking, voting, or commenting on something someone has shared. 

To designers, “visibility” describes affordances that make behaviors, knowledge, preferences, and connections visible to others. These affordances shape the privacy of people’s relationships and also enable new relationships to form by making people aware of each other. Design features for visibility enable people to:

  • share content, like update a status on Facebook, make a post on Mastodon, create a video on TikTok, or edit a Wikipedia page.
  • create a profile to share information about yourself or your interests 
  • search, to find others to connect with 
  • be found, so that people can connect with you. 

Searchability and findability are often supported by algorithms, such as search systems, recommender systems, and algorithmic feeds that prioritize some information at the expense of others. When people debate the design of platforms, content moderation, and the role of algorithms in society, they are often debating questions about visibility and association.

Debates about design affordances of association and visibility are ultimately about relational rights. These affordances provide the infrastructure that allow people to hear others and be heard, to find and be found by others, and to form groups. As a result, tech companies’ decisions about relational affordances play powerful roles in people’s relational rights— defining who can convene and who can’t, who can be found and who will be hidden, and who has the right to end a relationship or protect themselves from harm.

Building Conversations about Rights around People and Human Flourishing

Because the future of human groups and relationships online needs protecting, we need to strengthen the bridges of understanding between social science and the law. Earlier this year, we held a convening at the Knight First Amendment Institute about strengthening the link between social scientists, technologists and the law— focusing on civil liberties as one of the main areas. We learned a lot about the pathways for understanding — and misunderstanding between our professions.

As social scientists, we tend to start conversations about basic freedoms with people and their lives—asking how to ensure that they can fully enjoy those rights. That can sometimes create conversational hang-ups with litigators and policymakers whose starting point is the law. This difference of starting-point has been especially tricky in conversations about the role of corporations. Legal strategy has also sometimes created hang-ups in conversations about whether freedom of expression alone is enough to protect fundamental rights without needing reference to rights of assembly or association.

For example, legal experts often point out that documents like the U.S. Constitution protect people from government actions, and that corporations aren’t bound by the same laws. As a result, they sometimes argue, there’s no point in asking questions about rights in private spaces. While we understand that people often misunderstand this aspect of U.S. law, that’s not a reason to leave the idea of human rights at the door in our relationship with tech companies (MacKinnon 2013). That’s the argument in a recent book on The Universal Declaration of Human Rights in the 21st Century, which points out that human rights provide a powerful tool to protect people from corporations as well as governments. When describing the need for human rights frameworks that apply to corporations, they point out “freedoms of expression, association, and assembly… increasingly are facilitated [by corporations] online and on mobile devices” (Brown 2016).

Legal experts also sometimes tell us that we should prioritize one right over the others as a matter of strategy—for example prioritizing expression above association. Initially, this confused us — why would lawyers choose to set aside whole areas of human experience and social science in the pursuit of a just society? We were especially confused by why so many discussions of freedom focus on individual rights when centuries of social science have revealed that relationships and groups are fundamental parts of a flourishing life.

As we have learned more about the law, we have learned that policymakers and litigators often have good reason to prioritize some rights over others in the quest to win cases and accomplish policy goals. In those situations, if a given right is more persuasive to courts or policymakers, lawyers will choose the most persuasive argument.

Those valuable strategic choices can also create misunderstandings with scientists when we don’t understand the underlying game. We have learned the value of centering early conversations on people’s experiences, their rights, and what it means for them to flourish—and then considering legal strategy—to create better bridges between science and the law.

Strengthening the Civic Fabric With Social Science and Design

Philadelphia’s Temple of Free Discussion is a powerful reminder of the influence that design and social forces exert over the exercise of fundamental rights. The building’s founders realized that they could strengthen democracy with an act of architecture that brought people together. Their opponents hoped to weaken the civic fabric by sending a message that meetings for equality would be met with violence. But while the mob destroyed the building, they lost the argument. By 1870, the fifteenth Amendment protected Black Americans’ right to vote and more than one state had extended voting to women— thanks to countless people exercising their rights of assembly and association. 

One hundred fifty years later, as society reckons with the problem of mob violence in our digital architectures, we believe the tools of social and behavioral science can help our relational rights of assembly and association continue to flourish.

Acknowledgments

We are grateful to Kushal Dev, Katy Glenn Bass, Kit Walsh, Jon Penney, Alex Abdo, and Ashley Gorski for conversations and resources that informed this post. We are also grateful to the Siegel Family Endowment for supporting our convening on Causality, Platforms, and the Law. All errors are our own.

(image source: Destruction by Fire of Pennsylvania Hall, May 17, 1838, by John Caspar Wild)

References