Rediscovering the Pleasures of Pluralism: The Potential of Digitally Mediated Civic Participation
by Lily L. Tsai and Alex Pentland (1)
The authors explore how AI can foster new forms of civic engagement and direct democracy at both local and national levels. They propose designing digital infrastructure to enable collective decision-making and reduce polarization by encouraging interactions across diverse perspectives and backgrounds.
I. What Pluralist Societies Need from Digital Civic Infrastructure
Human society developed when most collective decision-making was limited to small, geographically concentrated groups such as tribes or extended family groups. Discussions about community issues could take place among small numbers of people with similar concerns. As coordination across larger distances evolved, the costs of travel required representatives from each clan or smaller group to participate in deliberations and decision-making involving multiple local communities. Divergence in the interests of representatives and their constituents opened up opportunities for corruption and elite capture.
Technologies now enable very large numbers of people to communicate, coordinate, and make collective decisions on the same platform. We have new opportunities for digitally enabled civic participation and direct democracy that scale for both the smallest and largest groups of people. Quantitative experiments, sometimes including tens of millions of individuals, have examined inclusiveness and efficiency in decision-making via digital networks. Their findings suggest that large networks of nonexperts can make practical, productive decisions and engage in collective action under certain (1) conditions. (2) These conditions include shared knowledge among individuals and communities with similar concerns, and information about their recent actions and outcomes.
We will need to be careful to keep these technologies from reinforcing the nationalization of politics and political discourse that has taken place over the last few decades. Even before the onset of social media, citizens in the United States were increasingly turning away from both local politics and grassroots organizations that have historically tried to address the everyday problems in local neighborhoods—teenagers skipping school, empty storefronts, broken windows on the street. Instead of spending time reading the town paper and going to town hall meetings, today we are transfixed by who might run for president and what their family members do, what national politician is under investigation, and what case is being heard in the U.S. Supreme Court.
In the United States, this nationalization of politics has led to partisan “mega-identities.” (3) The cleavage between political parties has come to overlap with and reinforce social and economic cleavages. (4) Liliana Mason writes, “A single vote can now indicate a person’s partisan preference as well as his or her religion, race, ethnicity, gender, neighborhood, and favorite grocery store.” (5) We now have many reasons, not just a single one, to distance ourselves from those we see in the “other camp.” And not only do we disagree with their views, but we treat them like the bad guys. Polarization didn’t happen just because of social media.
“Even before the onset of social media, citizens in the United States were increasingly turning away from both local politics and grassroots organizations that have historically tried to address the everyday problems in local neighborhoods.”
But both ideological and affective polarization play out on platforms that have increased the speed and scale of communication and ramped up the emotional intensity of confrontation. Some people leave in search of digital spaces filled only with like-minded people. Others get drawn in and double down on conflict. Neither is good for negotiating differences and either coming to a compromise or agreeing to disagree. Both reactions further reinforce the walls between our partisan mega-identities.
We agree with those who think revitalizing place-based identities and local engagement may help to break down these walls. If we redirect people’s attention back to local politics and working with fellow community members on projects that have a tangible impact, perhaps we can restart a positive feedback loop and regenerate the crosscutting connections that were once used to moderate conflict.
But it also seems clear that reinforcing local community structures should not be the only approach. So many towns and neighborhoods in the United States are effectively mini-one-party states. With the demographic sorting that has taken place over the last 40 years, focusing our attention primarily on local politics and participation in local decision-making might inadvertently reinforce social and political blocs. We might want to “break up” these blocks with cross-cutting connections between localities. Many studies suggest that connecting communities can be a major source of innovation and change. (6)
“With the demographic sorting that has taken place over the last 40 years, focusing our attention primarily on local politics and participation in local decisionmaking might inadvertently reinforce social and political blocs. We might want to “break up” these blocks with crosscutting connections between localities.”
Unfortunately, since the 1970s, we have experienced a decline in the churches, fraternal organizations, unions, and other forms of associational life that once created crosscutting connections both within and across localities. (7) Over the last 50 years, people have shifted away from meeting and volunteering regularly in associations and informal gatherings such as support groups and Bible study groups, choosing instead to engage through “checkbook participation,” where we outsource and pay local and national organizations who hire professional staff to do the good works and advocate for the policies that we want. (8) We no longer have regular practice in meeting to have discussions about how to make a decision that affects everyone in a group to which we belong. We still “talk” online about policy and political issues, but our talking is not tied to a need to come to an agreement about something that our group needs to accomplish.
With technologies that can now facilitate discussion and decision-making among groups both big and small, we must ask if we can build new kinds of intermediating digital spaces, ones that provide perspective, attention, and action on shared rather than personal problems, while at the same time accommodating discussion and deliberation at local as well as national scales. Could we design digital civic infrastructure to enable collective decision-making and direct democracy for both local and national publics, while dampening the polarization that has come with the nationalization of politics?
Before we explore this question, it is important to note that online spaces for civic participation and democratic decision-making must be rigorously tested and regulated and/or operated by public actors to avoid unwanted and unintended effects. We have elsewhere proposed a framework to ensure that such platforms and technologies are designed and regulated according to democratic values and principles. (9)
But what do we need to think about from an ordinary citizen’s perspective if we want to design platforms that enable and encourage people to come together and make decisions that affect everyone in a local or national community? We would have to solve for two problems.
The first challenge is how to get people to want to talk and work with one another. The second is how, once people have gotten onto the platform, to get people to want to understand the needs and concerns of others and to consider these concerns as we make decisions that affect all of us.
How do we build digital infrastructure for civic participation that addresses these problems? We might need to create new kinds of mediated civic participation, ones that could make us more comfortable and more curious than discussing and deciding on difficult public issues in person.
II. Digital Intermediation and “Reserved” Civic Participation
Alexis de Tocqueville, often quoted in discussions about revitalizing American democracy, talked about town meetings as schools for teaching people how to “use and enjoy liberty.” (10)
Town meetings, as Tocqueville saw them, have become diminished. We no longer attend town meetings at the same rates. And when we do attend them, we certainly do not enjoy them.
Many online spaces for political discussions are no better. To paraphrase Audrey Tang, Facebook is like trying to have a political discussion in a nightclub. (11) Some online spaces are far worse, involving people or bots who take a “no-holds-barred” approach to the things they say, or worse, may actively seek to incite a virtual mob to go after you. Noisy, sprawling social media platforms like X (formerly Twitter), Facebook, and Reddit are far from the ideal of leafy, public squares and citizen assemblies where we can stand around, listening and thinking, do some people-watching, and maybe exchange some thoughts with someone standing next to us in a noncommittal way.
Given where things are with online political discussions, what would make participation in collective discussions and decision-making tolerable, if not enjoyable in the way Tocqueville praises? And could this be done at scale? We have never successfully envisioned a national public square where everyone in the country could be talking at the same time.
Even in Tocqueville’s time—and probably for as long as humans have discussed things in groups—people have lost their tempers and raised their voices. That’s why it was important to have the option of standing on the edges of the square and watching the spectacle, or to be able to retreat there if things got to be too much, without having to leave the premises completely. One could move a little further away, maybe strike up a side conversation under a tree, and passively keep an ear out to see if the public conversation might be worth tuning into again.
In today’s online platforms, we have lost our power to calibrate how close we are to the conversation. We can either choose to get on the platform or channel and be confronted with sometimes unhinged, unbridled commentary, or we must choose not to be on there at all.
“Given where things are with online political discussions, what would make participation in collective discussions and decision-making tolerable, if not enjoyable in the way Tocqueville praises? And could this be done at scale?”
The work of political theorist Bernardo Zacka illuminates the importance of “in-between spaces.” Zacka’s focus is on physical spaces, such as balconies on apartment buildings that overlook the street, and how they can enable what Zacka calls a “reserved sociability” that strengthens pluralism in democratic societies. (12) When the driver of a car starts honking at someone for taking their parking space, and others start yelling at both cars to stop making so much noise, and a shopkeeper comes over to suggest that one of them goes and finds another spot—spaces such as balconies allow people to come out, see what’s going on, see how the drivers and the shopkeeper are dealing with the conflict, and observe whether other spectators approve or disapprove with the efforts.
Balconies and other in-between spaces such as porches, stoops, courtyards, and lobbies are interesting to Zacka because they serve several functions that are important for living in a pluralist society. The first is a measure of safety and autonomy. A balcony allows you to engage with what is going on in public. But you are also at a remove. You have control over whether you want to withdraw into your own space. You can see the argument that’s going on in the street, you can even shout a suggestion, but you don’t have to engage physically with the disputants, and you don’t have to say anything if you don’t want to. (13)
These spaces also give you the perspective that comes with pulling back and having a bit of distance. You have a bird’s-eye view, and you decide what you watch, how long you watch, how long you reflect on it, or whether you pay attention at all. As Zacka notes, “balconies offer an unobstructed view over a range of activities, enabling eyes and ears to wander freely. In this regard, they are similar to benches at the edge of a square.” (14)
Digital intermediation could help offer this kind of access to what Zacka calls “the lively and diverse intermingling of strangers in the public arena” and what the urban planner Jane Jacobs described as the “great and exuberant richness of differences and possibilities, many of these unique and unpredictable—and all the more valuable because they are.” (15) And digital intermediation could give this access in ways that are under one’s control, at one’s leisure, protected by anonymity, and, as Tocqueville described, enjoyable. Zacka and Jacobs argue that the diversity of city life can be wonderful, but only when the city brings diverse people together without forcing them to be too close. (16) When forced too close, they may withdraw into their private spaces, forego interaction with others, and become less tolerant of diversity. (17)
These qualities of safety and anonymity, along with autonomy and control over one’s time, attention, and intensity of engagement, including the ability to step back from within the discussion for perspective on the whole, enable people to process diversity and disagreement on their own terms, learn to tolerate it, and perhaps even become curious enough to learn more about others, enter the space, and participate more actively. Too much intimacy—“too much information (TMI)”—can be off-putting, but online platforms for discussion and deliberation could give those who are curious the power to dip their toes in first, then slowly wade in as they become interested enough to find out about the needs and opinions of others, particularly if the platform enables them to maintain a kind of depersonalized distance from others.
To illustrate the opportunities for civic—and civil—participation that digital intermediation can create, we look at two examples in the next section: an AI-enabled platform for public engagement on school reform recently piloted in Romania, and the Polis platform, which has been used widely for collective deliberation around the world.
III. Examples
Example 1: AI-enabled Public Engagement in The School of Possibilities
Created by an international team of designers and researchers with students and technologists in Romania, and led by civic media scholar and designer Eric Gordon, The School of Possibilities1 (18) is an AI-enabled process for students to provide input and engage in policy debates about school reform. Through an app on their phone, students provide performance evaluations and policy input to chatbots representing key aspects of their educational experience, such as the principal or a teacher, as well as important objects for their learning, such as a textbook, a whiteboard, or a mobile phone. As they come across chatbots representing different characters and objects, they can tell the bot about their previous experiences with that character or object. Insights from the conversation are processed in real time, and school characters and objects pop up on a screen projected for all participants to see, to share what the platform and decision-makers are learning from listening to participant experiences and perspectives. (19)
Participants report that talking to the system’s chatbots is “freeing” and rewarding. In one example, a student who interacts with a “teacher” chatbot is asked to share their stories or memories of interacting with teachers. Based on what the student shares, the “teacher” bot might respond, “Sometimes I don’t effectively listen to what my students need.” The “teacher” then expresses curiosity and asks the student to share their ideas about how to be better in the future. Gordon writes, “There is no deception here, [the student] never believed that the chatbot was the voice (or ear) of the institution. But she accepted the fiction created by the chatbot as a comfortable user experience, and she trusted the process because the anonymous insights were immediately shareable with decision-makers.”
“Because each participant can remain anonymous and their input can be separated from their personal identity, they can feel protected. This can make them feel freer to speak, or equally important, not to speak, and simply survey the comments of others as they appear on the collective screen.”
Gordon’s School of Possibilities is an example of an online environment that promotes reserved civic participation. Because each participant can remain anonymous and their input can be separated from their personal identity, they can feel protected. This can make them feel freer to speak, or equally important, not to speak, and simply survey the comments of others as they appear on the collective screen. Participants have control over whether they engage passively or actively, how much time they want to take to formulate their comments or reflect on what they are hearing, and “who” they talk to within the school. The real-time reporting of insights from all participants gives each participant a bird’s-eye perspective on the conversation as a whole.
In addition, the “active listening” of the AI chatbots, which are designed to reflect back what the participants tell them to make people feel heard, suggests that AI bots can also help to set conversational norms for more productive and civil discussions about public issues. Gordon reflects on how the responses of the chatbots were just as important as the ability to speak to the chatbots: “[I]t wasn’t just the interaction with chatbots, it was how the chatbots represented that they were listening. For example, the textbook says: ‘The information in me can be out of date.’ And the whiteboard says: ‘Sometimes I create anxiety when students are asked to solve problems on me.’” Gordon reports that students responded to this real-time listening from the chatbots with increased confidence that decision-makers and other participants would also be more likely to pay attention to their experiences with some degree of sympathy.
Finally, it is worth highlighting that the digital environment in The School of Possibilities makes participation a pleasurable and rewarding experience. Gordon notes that organizations “have long been eager to get more people involved in planning processes or policy debates through big public meetings or distributed surveys.” Instead of these formats, Gordon advocates for “civic play spaces” (20) and games that not only motivate people to participate but enable them to have fun doing so. These approaches, he argues, often result in more creative and free expression, and a greater sense of collective efficacy. (21) The School of Possibilities illustrates how engagement that is mediated by machine could be more engaging, safe, and fun than in-person modes of interaction.
Example 2: The Polis Platform
The Polis platform seeks to give participants a dynamic overview of the entire spectrum of opinion around a discussion topic and has been seen as a highly effective social media tool for direct and deliberative democracy. (22) It allows authorities or organizers to pose policy questions to the public and then uses statistical summarization to provide graphical feedback on what the population as a whole is saying.
Polis has been used in Austria to generate consensus on climate issues (2022), in Uruguay on a national referendum (2020–2021), in New Zealand to facilitate the development of government policy (2016–2019), in the Philippines to generate municipal policy (2020–present), in the U.S. to counteract polarization in a Kentucky town (2018), in the UK as part of a government polling effort (2020), and in Germany to develop a political party’s platform (2018). (23) In addition, it has been used for X’s Community Notes and by Anthropic to draft a publicly-sourced constitution for an AI system. (24) Taiwan’s deployment of Polis (divya-vtaiwan) is widely believed to be a particularly effective example of achieving popular consensus around contentious issues. (25)
Polis works by first putting forward a topic for discussion. Participants can then choose whether or not to provide a comment on the topic. The platform also gives participants a sample of comments to agree or disagree with. Unlike other platforms, users cannot reply to other users’ comments, making it difficult to engage in trolling. The process of looking at a range of comments made by others for upvoting and downvoting exposes people to others’ opinions, which has been shown to reliably promote “wisdom of the crowd” effects and better decision-making. (26) The upvote/ downvote mechanism creates a citation network, similar to citation networks used in scientific papers, patent applications, and legal decisions; upvotes and downvotes in Polis are analogous to citations.
Based on reactions to each comment, Polis generates a citation map, clustering the comments (and the participants) who have agreed with the same comments. This map visualizes the clusters—or opinion groups—thereby showing where there is agreement and where there is division.
Polis thus exhibits many of the features that enable the reserved sociability or civic participation that can contribute to a healthy pluralist democracy. Polis ensures that participants focus on the content of comments and the state of the conversation, because there are no direct conversations or interactions with other participants. Discussion on the platform in this sense happens at a distance, like an observer sitting on a park bench or standing on a balcony.
Polis can also give participants autonomy and control over their terms of engagement with the discussion on the platform. Participants who are not yet ready to engage in more depth can be allowed to peruse the platform’s summarizations and mappings of the current state of the discussion, which provide the bird’s-eye view that supports reserved civic participation. They can follow the evolution of the conversation and watch how different groups of opinions may be merging and changing over time. They can decide when to enter their own comment, whether to upvote or downvote the comments of others, and how many other comments to view and evaluate.
Individuals can also zoom in if they would like. If they want to understand what a particularly large cluster of comments are focused on, or if they want to understand the outlier point out on the margin of the conversation, they can view the comments of others through the platform’s map of the conversation by hovering over points on the map.
Some have also argued that the Polis visualization of comments, as shaped by citations, seems to be helpful in promoting convergence of opinion, and is much like the visualizations that have proven very effective in domains such as finance. (27) Taiwan’s Minister of Digital Affairs, Audrey Tang, has declared, for example, that “if you show people the face of the crowd, and if you take away the reply button, then people stop wasting time on the divisive statements.” (28) MIT research has shown that there is reason to believe that the Polis-style approach could have a very significant impact on decreasing polarization. We use this type of approach to achieve significant increases in democratic attitudes among partisans in America, (29) and recent related MIT papers show convergence of opinion in financial decisions by providing users with a visualization of the range of opinion and action. (30)
IV. Designing Digital Civic Infrastructure for Pluralist Democracies
To conclude, we argue that online platforms and generative AI can offer a digital intermediation that creates spaces for civic participation with the qualities of reserved sociability—a kind of “reserved civic participation.” However, accomplishing this goal requires three features that are often missing in existing platforms.
First, identity authentication can ensure that everyone on the platform is a human entitled to join the platform…no bots, and maybe no commercial companies. It used to be that knowing someone’s identity was necessary to ensure that they had rights to participate. But much of the toxicity of current social media interactions is about identities as much as opinions. (31) Now technology enables us to separate one’s identity from one’s right to participate; you can be anonymous, but others can know you are a real person with legitimate standing in the community. Conversation can focus on opinions and ideas, not on a person’s identity.
The data and relationships that we co-generate with others in our economic and social interactions—our bank, doctor, government, friends—form our digital identity. These relationships can attest to who we are, and the multiplicity of these relationships means that no one actor can deny you the rights associated with your identity. In the digital realm, this type of community grounding of digital identity is well known to be one of the most secure methods of protecting citizens from malicious attack.
“The dual nature of anonymity and shared identity can allow a new reserved civic participation that can make us more comfortable interacting with fellow community members who can otherwise seem very different or foreign to us, which can perhaps also increase our tolerance of diversity.”
The same web of relationships that form our digital identity can help us find other people with similar concerns or experiences, allowing the formation of communities of concern that cut across communities defined by physical location. The data that help us find and engage with communities with common concerns are often extremely private, so any successful engagement infrastructure must give us rights over these co-generated data and relationships, including not just privacy rights but also ownership and use rights. With such rights, we are protected, free to be anonymous and to engage as strangers on online platforms, but also confident in the knowledge that we share parts of our identity with them. The dual nature of anonymity and shared identity can allow a new reserved civic participation that can make us more comfortable interacting with fellow community members who can otherwise seem very different or foreign to us, which can perhaps also increase our tolerance of diversity.
Second, the digital intermediation that could be offered by new types of online platforms and generative AI can help guide us to focus more on public issues and avoid the emotional reactions and personal biases that cloud our judgment. (32) Such platforms can summarize comments from large numbers of participants and visualize common themes, connections, and the distances among comments. The benefits of these functions go well beyond merely providing additional information for participants and authorities to consider.
Summarization and visualization of where participants agree and disagree is analogous to watching a town assembly while walking around the edges of the square or looking down on a heated discussion in the street from your balcony.
“By giving us control and autonomy about how and when we participate, digital intermediation can help cool our initial reaction to the conversation and create the emotional distance that enables us to activate our more analytical, logical reasoning mode.”
By giving us control and autonomy about how and when we participate, digital intermediation can help cool our initial reaction to the conversation and create the emotional distance that enables us to activate our more analytical, logical reasoning mode. It can also encourage deliberation by enabling asynchronous versions of listening and speaking. It provides not just space and perspective, but also the time to walk away and process before coming back to say something or listen more. Asynchronous forms of engagement can be synchronized with regular weekly or daily reports provided by the platform that serve something of the function that “talk of the town” columns in local newspapers used to provide.
The interactive intermediation that AI makes possible can also help establish the kinds of norms of expression and communication that we would normally use in arm ’s-length interactions with people we don’t know very well but have no reason to suspect of ill intent. As we see in The School of Possibilities, AI chatbots can model how to respond with interest and calm to the experiences and feelings voiced by other participants, and even more interestingly, they can establish these conversational norms even when participants understand that they are chatbots.
Third, digital intermediation on online platforms can draw us into online discussion and deliberation about public issues when they are designed to take advantage of our natural curiosity and playfulness. Our interests in people-watching, eavesdropping, and checking out what others are looking at are important aspects of what makes us human. We need to design digital platforms for civic engagement that help us satisfy these instincts that undergird and strengthen human sociality. These platforms could be used to improve both local and national public discussion and deliberation. Revitalizing place-based engagement can take place in parallel with creating new mechanisms for national public discourse—it doesn’t have to be either/or. At this point in our history, exploring multiple strategies for realizing e pluribus unum is critical.
As with physical town squares and public spaces, it remains important to ensure that digital public spaces for civic participation are owned and operated by the people, for the people, and of the people. As Mitchell Stevens notes, existing social media platforms are commercial spaces in the same way that malls and casinos are. We can interact with others from our community in malls and casinos, sometimes in ways we find rewarding, but commercial actors own those spaces, and the terms of our interactions are made to maximize their benefit, not ours. (33)
We must answer the same questions for online platforms that exist when we organize traditional in-person public assemblies and deliberations. How do we determine who “owns” the space—who has agenda-setting power and chooses the topics and questions for collective consideration
We have elsewhere enumerated the democratic values and principles to which we must remain sensitive if we want to design digital platforms that strengthen democratic societies—preserving human agency and autonomy so that AI and other technologies are not used to manipulate us; encouraging mutual respect; promoting equality and inclusiveness (issues that Tocqueville and his contemporaries passed over); and augmenting rather than substituting for the actions we take as citizens. (34) It is these principles that we have to keep in mind when we decide what role AI bots should play in suggesting topics for collective consideration, what data they use to do so, and what the overall governance of these platforms should be.
But making sure that we do not violate these values and principles is just the start. If we want vibrant civic life and democratic processes that encompass a rich variety of perspectives and backgrounds, and decisions that benefit from the “many different tastes, skills, needs, supplies, and bees in [our] bonnets” that Jacobs marveled at, (35) we need new environments that enable us to tolerate disagreement and difference, that depersonalize conflict, and that enable us to step away when we need to. We need digital civic infrastructure that enables people to watch, listen, and mull things over, not just express and speak. We need room and time to step back and process things. Digital technologies offer new opportunities for designing spaces for civic engagement that uphold democratic principles, that make us feel safe and protected and enable us to look out over the goings-on while giving us control over how much we want to participate. Only then can we fully use and enjoy our liberty again, in the way Tocqueville reported long ago.
Footnotes
(1) The authors would like to thank generous gifts from both Project Liberty and MIT Schwartzman College.
(2) Cameron Martel, Jennifer Allen, Gordon Pennycook, and David G. Rand, “Crowds Can Effectively Identify Misinformation at Scale,” Perspectives on Psychological Science 19, no. 2 (2024): 477–88, https://doi.org/10.1177/17456916231190388; P.M. Krafft, Erez Shmueli, Thomas L. Griffiths, Joshua B. Tenenbaum, and Alex Pentland, “Bayesian Collective Learning Emerges from Heuristic Social Learning,” Cognition 212 (July 2021):104469, https://doi.org/10.1016/j.cognition.2020.104469; Galen Pickard, Wei Pan, Iyad Rahwan, Manuel Cebrian, Riley Crane, Anmol Madan, and Alex Pentland, “Time-Critical Social Mobilization,” Science 334, no. 6055 (2011): 509–12, https://doi. org/10.1126/science.1205869.
(3) Lilliana Mason, Uncivil Agreement: How Politics Became Our Identity (Chicago: University of Chicago Press, 2018), 14.
(4) Matthew Levendusky, The Partisan Sort: How Liberals Became Democrats and Conservatives Became Republicans (Chicago: University of Chicago Press, 2009); Alan Abramowitz, “Partisan Polarization and the Rise of the Tea Party Movement,” in APSA 2011 Annual Meeting Paper (American Political Science Association, 2011), https://ssrn. com/abstract=1903153.
(5) Mason, “Uncivil Agreement,” 14.
(6) Wei Pan, Gourab Ghoshal, Coco Krumme, Manuel Cebrian, and Alex Pentland, “Urban Characteristics Attributable to Density-Driven Tie Formation,” Nature Communications 4, no. 1 (2013): 1961, https://doi.org/10.1038/ncomms2961; Shi Kai Chong, Mohsen Bahrami, Hao Chen, Selim Balcisoy, Burcin Bozkaya, and Alex Pentland, “Economic Outcomes Predicted by Diversity in Cities,” EPJ Data Science 9, no. 1 (2020): 17, https://doi.org/10.1140/epjds/s13688-020-00234-x; Eaman Jahani, Samuel P. Fraiberger, Michael Bailey, and Dean Eckles, “Long Ties, Disruptive Life Events, and Economic Prosperity,” PNAS 120, no. 28 (2023): e2211062120, https://doi.org/10.1073/ pnas.2211062120.
(7) Robert D. Putnam, Bowling Alone: The Collapse and Revival of American Community (London: Simon & Schuster, 2001).
(8) Theda Skocpol, “Associations Without Members,” The American Prospect 45 (1999): 66–74; Theda Skocpol, Diminished Democracy: From Membership to Management in American Civic Life (Norman: University of Oklahoma Press, 2003).
(9) Lily L. Tsai, Alex Pentland, Alia Braley, Nuole Chen, José Ramón Enríquez, and Anka Reuel, “Generative AI for Pro-Democracy Platforms,” An MIT Exploration of Generative AI (Boston: MIT Open Publishing Services, March 2024), https://doi.org/10.21428/ e4baedd9.5aaf489a.
(10) Alexis de Tocqueville, Democracy in America, Chapter 5, Part I.
(11) See, e.g., Bruno Kaufmann and Jonas Glatthard, “‘Humour over Rumour’: Lessons from Taiwan in Digital Democracy,” SWI swissinfo.ch, May 7, 2021, https://www.swissinfo.ch/ eng/politics/freedom-of-expression-humour-over-rumour-lessons-from-taiwan-indigital- democracy/46592080.
(12) See Bernardo Zacka and Duncan Bell, eds., Political Theory and Architecture (London: Bloomsbury, 2020). Stanford.
(13) Bernardo Zacka, “What’s in a Balcony? The In-Between as a Public Good,” in Political Theory and Architecture, eds. Bernardo Zacka and Duncan Bell (London: Bloomsbury, 2020), 83.
(14) Zacka, “What’s in a Balcony?,” 86.
(15) Jane Jacobs, The Death and Life of Great American Cities (New York: Random House, 1961; reprinted 1993).
(16) Zacka, “What’s in a Balcony?,” 88.
(17) Zacka, “What’s in a Balcony?,” 89, discussing Jacobs’s “The Uses of Sidewalks: Contact,” in The Death and Life of Great American Cities, 62–4.
(18) See “The School of Possibilities, https://www.schoolofpossibilities.org; Eric Gordon, Barbara Bulc, Tomás Guarna, Tijana Zderic, Bianca Stefania Baluta, and Nadina Pantea, Activating Values in Urban Transitions: A Novel Approach to Urban Innovation in Romania (OurCluj, Fondation Butnar, April 2022).
(19) Eric Gordon, Generative Listening, (manuscript in progress), 1.
(20) Eric Gordon and Gabriel Mugar, Meaningful Inefficiencies: Civic Design in an Age of Digital Expediency (Oxford: Oxford University Press, 2020).
(21). Gordon, Generative Listening, 2–3.
(22) Chris Horton, “The Simple but Ingenious System Taiwan Uses to Crowdsource Its Laws,” MIT Technology Review, August 21, 2018, https://www.technologyreview. com/2018/08/21/240284/the-simple-but-ingenious-system-taiwan-uses-to-crowdsource- its-laws/.
(23) “Featured Case Studies,” The Computational Democracy Project, accessed July 3, 2024, https://compdemocracy.org/case-studies/.
(24) Anthropic, “Collective Constitutional AI: Aligning a Language Model with Public Input,” October 17, 2023, https://www.anthropic.com/index/collective-constitutional-aialigning- a-language-modelwith-public-input; Carl Miller, “Elon Musk Embraces Twitter’s Radical Fact-Checking Experiment,” Wired, November 20, 2022, https://www. wired.com/story/elon-musk-embraces-twitters-radical-crowdsourcing-experiment/
(25) Horton, “The Simple but Ingenious System Taiwan Uses to Crowdsource Its Laws.”
(26) Abdullah Almaatouq, Alejandro Noriega-Campero, Abdulrahman Alotaibi, P. M. Krafft, Mehdi Moussaid, and Alex Pentland, “Adaptive Social Networks Promote the Wisdom of Crowds,” Proceedings of the National Academy of Sciences 117, no. 21 (2020): 11379–86, https://doi.org/10.1073/pnas.1917687117.
(27) Dhaval Adjodah, Yan Leng, Shi Kai Chong, P. M. Krafft, Esteban Moro, and Alex Pentland, “Accuracy-Risk Trade-Off Due to Social Learning in Crowd-Sourced Financial Predictions,” Entropy 23, no. 7 (2021): 801, https://doi.org/10.3390/e23070801.
(28) Chand Rajendra-Nicolucci and Ethan Zuckerman, An Illustrated Field Guide to Social Media (New York: Knight First Amendment Institute at Columbia University, 2021), 12, https://s3.amazonaws.com/kfai-documents/documents/43e1299387/Mapping_ Social_Media_compressed.pdf.
(29) Alia Braley, Gabriel S. Lenz, Dhaval Adjodah, Hossein Rahnama, and Alex Pentland, “Why Voters Who Value Democracy Participate in Democratic Backsliding,” Nature Human Behaviour 7, no. 8 (2023): 1282–93, https://doi.org/10.1038/s41562-023- 01594-w.
(30) Dhaval et. al., “Accuracy-Risk Trade-Off Due to Social Learning,” 801; Krafft et al., “Bayesian Collective Learning”; Abdullah et al., “Adaptive Social Networks Promote the Wisdom of Crowds.”
(31) We thank Mitchell Stevens for this observation.
(32) See, for example, Samara Klar, “Partisanship in a Social Setting,” American Journal of Political Science 58, no. 3 (2014): 687-704, https://doi.org/10.1111/ajps.12087. Depersonalization might also be good because it prevents you from being overly affected by the identities of those around you—if those around you have the same identity, you are more fixed in your beliefs.
(33) Mitchell L. Stevens, personal communication, May 21, 2024.
(34) Tsai et al., “Generative AI for Pro-Democracy Platforms.”
(35) Jacobs, The Death and Life of Great American Cities.