Who Killed the Third Place?

In partnership with

In 1989, sociologist Ray Oldenburg released a book titled The Great Good Place: Cafés, Coffee Shops, Bookstores, Bars, Hair Salons and Other Hangouts at the Heart of a Community. In it, Oldenburg coined the term "third places" to describe the informal public spaces that serve as neutral ground for social interaction outside the home (first place) and workplace (second place). While the term is only a few decades old, third places are certainly not recent phenomena. The ancient Greeks had the agora, and the Chinese had their teahouses. Enlightenment Europe had its coffeehouses, while America’s revolution was planned in taverns.

These spaces function as the connective tissue of American civic life, providing meaning and community outside of economic activity. They exist outside of the social hierarchies that dominate other spheres of life. What Oldenburg called the “leveling principle” allow for us to meet each other as equals, as patrons of a hair salon or as competitors in a game of dominos played on a Brooklyn sidewalk.

But third places are dying. They have been for as long as the term has existed.

Why?

Putnam's Framework and Its Limitations

Released in the year 2000, Robert Putnam's Bowling Alone documented the decline in American associational life between 1980 and 1993. The name of the book comes from one of the case studies Putnam uses to make his point: in that time span, the number of bowlers in America increased by 10 percent while league bowling decreased by 40 percent. "More Americans are bowling today than ever before," Putnam observed, "but bowling in organized leagues has plummeted." This pattern extended across virtually every form of civic engagement: membership in parent-teacher associations dropped by more than half, participation in religious activities declined substantially, and union membership fell off a cliff.

Putnam attributed this decline to what he termed "the collapse of American community," arguing that "social capital"—defined as "the connections among individuals' social networks and the norms of reciprocity and trustworthiness that arise from them"—had eroded to dangerous levels.

These are the connections formed by third places.

Putnam’s analysis distinguished between two forms of social capital: bonding social capital, which reinforces exclusive identities within homogeneous groups, and bridging social capital, which connects across the in-group/out-group divide. As Putnam explained, "bonding social capital constitutes a kind of sociological superglue, whereas bridging social capital provides a sociological WD-40."

The disappearance of third places represented a loss of both forms of social capital. Neighborhood bars and bowling leagues provided bonding opportunities for people with shared backgrounds and interests. Civic organizations and religious institutions created bridging connections across class, ethnic, and ideological lines. And now those were dying.

Much analysis of the relationship between third places and social capital rightly identifies that there’s a causal relationship—but they often get its direction wrong. Many urbanists today advocate for the return of the third place as a means to increase social capital. But the death of the third place didn’t cause a decline in social capital; a decline in social capital caused the death of the third place.

The rec league, the book club, and the pickup basketball games have decreased significantly over the decades not because there isn’t a place for them, but because there aren’t people for them. The reality is that there are simply more “fun” things to do after work than to schlep a change of clothes across town, change, play a game or two of basketball, and go home and shower. Sure it may be a fulfilling way to connect with friends and stay healthy, but there is far less friction involved in sitting on your couch and scrolling through TikTok for a few hours.

Social Media vs Social Capital

The emergence of social media platforms illustrates how digital alternatives have inverted traditional patterns of social capital formation.

Social media excels at creating bridging social capital—connecting people with shared interests across cultural identities or geographical boundaries. However, it performs poorly at generating bonding capital. The shotgun blast of an social media feed is a dopaminergic powerhouse that makes it impossible to stay too long in one place to be able to bond deeply with any group. The algorithmic curation of content feeds a loyalty to platform while stripping away any loyalty to group or place. Where traditional third places incentivized groups to bond deeply over a given interest and learn more about each other in a non-economic context, digital platforms allow users to avoid these potentially uncomfortable or vulnerable encounters entirely.

Digital technologies excel at providing the “sociological WD-40” of bridging capital while eliminating the "sociological superglue" of bonding capital. It’s worth specifying that these two types of social capital are not oppositional; they are complementary. They build off of each other: bridging capital finds new groups to bond with, and bonding capital creates the trust that is necessary to bridge additional divides.

Derek Thompson's analysis of declining social interaction demonstrates how this digital bonding has substituted for physical gathering without providing equivalent social benefits. Online communities offer easy paths to new interests, but they cannot replicate the complex negotiations, compromises, and mutual obligations that characterize successful civic life.

When Loneliness is a Choice

Putnam's bowlers did not become lonely because their leagues disbanded; they stopped participating in leagues because league bowling no longer provided sufficient social satisfaction relative to available alternatives. We don’t stop going to book club because there’s nowhere to host it; we just get busy with more interesting things in our lives. Technology initially aimed at increasing connection has quickly become a substitute for the traditional forms of engagement that social capital relies on.

If isolation causes loneliness, then creating new spaces for gathering should alleviate the problem. Community centers, public squares, and mixed-use developments become obvious solutions. But if loneliness causes isolation—if people withdraw from available social opportunities because those opportunities fail to provide adequate satisfaction—then building more gathering spaces addresses symptoms while ignoring underlying causes.

The infrastructure of community cannot be divorced from the cultural conditions that make community appealing. Third places flourished when they served functions that could not be easily replicated elsewhere—information exchange, entertainment, nourishment, faith. As these functions migrated to digital platforms, the spaces themselves became redundant regardless of their convenient locations or communal value.

Could AI Revive Third Places?

The emergence of artificial intelligence both threatens the acceleration of existing trends and presents an opportunity for reversal. A world with AI companions, co-workers, tutors, and entertainers can complete the trend toward total social isolation, or it can give us an opportunity to escape it.

Putnam noted that "social capital is often most easily created in opposition to something or someone else," making bonding capital easier to develop in the face of a trend which uncomfortably disrupts our social norms.

Consider one way AI might enhance social capital formation: Current community organizing efforts fail because they cannot efficiently identify compatible individuals, coordinate schedules, or sustain engagement over time. AI systems could analyze individual preferences, interests, and availability to assemble stable groups for ongoing interaction. Rather than replacing human relationships, such systems would serve as sophisticated matchmaking infrastructure for physical community formation.

Another way AI could incentivize social capital formation is by replacing some of the social capital reducing forms of entertainment in our lives. Many fear AI musicians, and there is good reason to be skeptical/concerned. But is it possible that in reaction, some people may begin attending live concerts with more frequency? When AI inevitably begins being used in movies and TV, is it possible that this may incentivize increased attendance of plays?

Don’t mistake these thoughts as blind optimism; I know that these aren’t the most likely outcomes.

But AI is making people uncomfortable. Many of us cringe at seeing an AI-created piece of art—where is its soul? Its flawed, deeply human view on the world shaped by its creator’s unique experiences? Where is its emotion? These feelings will not go away for many any time soon; there is an energy there waiting to be tapped. Just like today’s Zoomers engage in countercultural usage of cassette tapes and CD players over the more convenient streaming platforms, there is an opportunity here to promote a culture of human connection.

The Choice Between Substitution and Augmentation

The future trajectory of American community life depends on collective choices about how to deploy these emerging technologies. The path toward computational solipsism offers clear advantages: perfectly customized social experiences without the friction, compromise, and occasional disappointment inherent in human relationships. The path toward augmented community requires accepting that some goods can only be achieved through shared physical presence and mutual obligation, even when more “low-friction” or “fun” digital alternatives exist.

This choice cannot be made through policy prescriptions or urban planning initiatives. It emerges from millions of individual decisions about how to allocate our time and attention. The question is whether sufficient numbers of Americans can retain an appreciation for the particular satisfactions that genuine community provides—satisfactions that may be less immediately gratifying than digital alternatives but prove more durable and meaningful over extended periods.

The revival of third places, if it occurs, will require more than better design or smarter technology. It will demand recovery of the understanding that some human needs can only be satisfied through the patient work of building common life with others, despite the inefficiencies and imperfections such work entails.

Whether we still possess this understanding, or can recover it, remains to be seen.