Vertical communities

The internet has changed what "community" means.

Photo by tabitha turner on Unsplash

“I’m a Hoosier, too,” she crowed. “Nobody has to be ashamed of being a Hoosier.” — Hazel Crosby, “Cat’s Cradle”

An interesting paper is making the rounds. At first glance, the result of “Public perceptions of local influence”, by Joshua Hochberg and Eitan Hersh, seems pretty minor and innocuous — most people can no longer name individuals who they think are influential in their community. Back in the 1950s, most people would name local businesspeople.

One reaction is: So what? Maybe big companies have just muscled out small businesses, so that now it doesn’t mean much to be a local businessperson (in fact, this is one of the authors’ main candidate explanations). But another, more ominous interpretation is that Americans have become disengaged from their local communities, hunkering down in their houses and failing to interact with the people around them. In fact, there is some pretty well-known evidence that this is happening — the political scientists Robert Putnam and Theda Skocpol have both documented a major decline in Americans’ participation in civic organizations, local clubs, etc. Maybe the reason we can’t name important community figures is because we don’t really have communities anymore at all.

But I’m not so pessimistic. In fact, I think the kind of communities we inhabit has simply changed. In the past, our communities were primarily horizontal — they were simply the people we lived close to on the surface of the Earth. Increasingly, though, new technology has enabled us to construct communities that I’ve decided to call vertical — groups of people united by identities, interests, and values rather than by physical proximity.

The age of horizontal communities

Think about the use of “community” to refer to identity groups — “the Jewish community”, “the LGBT community”, etc. When the WSJ prints a headline that’s offensive to Asian people, it’s “the Asian community” that gets angry. But who is that “Asian community”? It’s not a group of people living near to each other, working together, or belonging to the same church groups or bowling leagues. Nor is it simply “all Asian people” — the overwhelming majority of Asian people didn’t even see the headline. Instead, the “Asian community” referenced in the backlash against the WSJ headline is a set of Asian people who are engaged with issues like this — and who come together online to discuss it.

Some people get annoyed at this usage. An identity grouping isn’t a real community, they say. And for most of history, they’d have had a point. There have always been notional identities that connected us with faraway people — other Jews, other Christians, other fans of punk rock, other people who moved away from Indiana. But because of limited technology, we typically didn’t have much contact with those people. Vertical communities have always existed, but the methods for the transmission of ideas, the organization of institutions, and the enforcement of norms were all very weak and slow. Pilgrims and crusaders and itinerant clergymen could spread religious culture. Diasporas could spread ethnic consciousness. But this took a lot of time and effort, and the demands of local, horizontal communities would usually dominate.

There were, of course, exceptions. The Catholic Church maintained a geographically vast organization that cut across the boundaries of towns and kingdoms (and which eventually came into conflict with those more horizontal communities). And there’s also my personal favorite exception — the scientific community that arose in Europe in the 1500s and 1600s using written letters to keep in touch and to allocate credit for scientific work. But for most of history, most of the people you interacted with most of the time were the people who lived near to you — your horizontal community.

Horizontal communities can often be stifling and repressive, because they impose community norms on people with a diverse array of occupations, temperaments, and backgrounds. Sinclair Lewis’ novel Main Street is a great depiction of the ever-present, crushing conformity pressure of small American towns in the 1910s. But that social pressure was nothing compared to the pogroms, inquisitions, and genocides that enforced religious, cultural, and racial homogeneity in many of the world’s horizontal communities — and which still do, in some parts of the world.

When the people around you pressure you to be the same as them, you can use exit, voice, or loyalty — you can knuckle under and conform, you can fight back and rebel, or you can simply leave and find some place that you fit in better. A lot of immigration to the U.S. was driven by misfits looking for communities where they didn’t stick out as much. In the latter half of the 20th century, Americans themselves sorted into different parts of the country in order to create pockets of local political homogeneity.

In fact, our use of the word “community” to describe racial, religious, and sexuality groups is probably a relic of an ugly pattern of history, in which minorities were forced to live in circumscribed, segregated areas — Chinatowns, ghettos, the Castro in San Francisco — either by law, or by mass violence that made them unwelcome elsewhere. There was a time when you really could go walk into the LGBT community — just walk down Castro Street, and there they were. Those horizontal minority communities were created by social exclusion, but they also developed highly original cultures and cooperative local institutions that lead many people to revere their memory.

As the country became more tolerant and more mobile, the idea that LGBT people or Asian people writ large constitute a “community” became more notional; Chinese Americans moved out of Chinatowns and into the suburbs. At the same time these minority communities were dissipating, horizontal communities across the country (and much of the developed world) were also breaking up, as Putnam and Skocpol documented. Cars made people more mobile, TV and big houses let them enjoy leisure in solitude, and so on. New vertical communities sprang up in the form of subcultures, but these were generally subordinate to horizontal communities in the form of local “scenes”.

The internet let us escape to the vertical

But then the internet came along, and everything changed. Suddenly we stopped being isolated and started being social again, through the windows of our laptop screens and phone screens. There was a whole world of human interaction waiting there for us — forums, social media feeds, chat apps, online games, and so on. Suddenly we were surrounded by people all the time — or at least, their written words, and perhaps once in a while their pictures or videos or voices.

Source: Pew

Constant internet usage allowed us to organize a far greater percentage of our human interaction around vertical communities. It let us find the people we identified with and interact with them, rather than being forced to interact with whoever was close to us on the map. We could surround ourselves with other anime fans, or other Muslims, or other economists, or other trans people — and we did. What were once notional bonds of connection that existed mostly in our minds became Facebook groups and subreddits and loose networks of Twitter contacts. And those spaces developed their own norms, rules, customs, and institutions, because now, thanks to the internet, it was easy to do that.

To see how important these vertical communities are, just repeat the exercise of Hochberg and Hersh’s paper. I can’t tell you who important local business leaders in my district of San Francisco are. But I can easily identify Paul Krugman, Olivier Blanchard, Larry Summers, and Jason Furman as highly influential figures in my online economics community. I’m sure members of the online Asian community, or LGBT community, or Black community could rattle off their own lists.

The identity-based “communities” that people talk about are thus no longer simply shorthand for a notion of cultural or political affinity with distant people, or for a fading memory of segregated neighborhoods. They’re thriving online verticals — archipelagos of online spaces where people can go to talk about what it means to be gay, or Jewish, or Pakistani. And like the small towns of Sinclair Lewis’ day, these vertical communities have the ability to use social ostracism to punish those who deviate from consensus norms and political objectives.

At the same time, horizontal communities didn’t completely vanish. We still educate our children in physical space (more or less), meaning we still have to deal with other children’s parents in a local community. Local government policies rule many of the aspects of our lives that are still offline — food, public safety, housing, transport — and this means we have to go to city planning meetings and school board meetings and various other community forums to hash out our differences with people who don’t share our interests or our identities. We now live in a world where our communities exist in three dimensions — the familiar hodgepodge of local humanity in two dimensions, and our self-sorted online spaces in a third.

And this dichotomy presents an enormous challenge to our institutions.

Public goods and the clash of communities

You’ll notice that the vertical online communities I describe are not “network states”. My friend Balaji Srinivasan wrote a book in which he imagines countries being replaced by human networks linked together by the internet and cryptocurrency. But whether or not such vertically organized states will ever come about, they’re just not feasible for now. Because it’s horizontal organizations — nation-states — that still provide almost all of our essential public goods.

Economics views public goods as a core reason — perhaps the core reason — that large human organizations exist in the first place. Things like national defense, courts of law, property rights, product standards, infrastructure, scientific research, and so on require something like a government to administer them. And governments are still organized horizontally; they administer physical territory defined by lines on maps. Moving from the nation-state system to a network-state system would require us to implement private law for members of the various networks — just like the law in medieval France was different for a priest than for a peasant, the law in a world of network states would have to apply differently to two different people who passed each other on the street. And remember that the old French word for “private law” is “privilege”. We are very, very far from having any idea how to live productively and happily in a world like that.

So for now and for the foreseeable future, our public goods are provided locally, but our social interaction happens in the cloud. In theory, this could be a dangerous recipe.

Political economics teaches us that public goods are easier to provide when people have homogeneous preferences. If 90% of people would want to use a road, it’s going to be a lot easier to get that road built than if only 40% would want to use it, even if it passes a cost-benefit test in both cases. When some people want quiet streets and others want dense exciting neighborhoods, local development policy will be hotly contested. And so on. This is one reason why people generally think it’s easier for a city-state to get rich than a big country — it’s easier for Singapore to implement an effective national housing policy than China.

But preferences are also endogenous — they can be changed. And affinity groups — who you see as being “your people” — might be able to change preferences substantially. Some economists argue that “artificial states” with arbitrary post-colonial borders struggle to provide public goods because of the eternal infighting between disparate ethnicities that don’t really want to share a country with each other. Those same economists hypothesize that the U.S.’ inefficient and highly contested welfare state is dysfunctional because of racial resentment.

What if vertical communities exacerbate these divisions? What if we talk and socialize and cooperate and fall in love with the people from our online crowds, and grow terminally apart from the people next door? What if we begin to feel our primary allegiance is toward people who share our race or our religion or our interests, rather than toward the people who share our country and our city? What if we go to the PTA or the planning board meeting and discover a bunch of strangers we despise and disdain?

In such a world, how will government get anything done? How will we decide what roads to build, what housing to allow, what universities to fund, or how to reform the police? How can we build a country together with neighbors with whom we no longer share any sort of common bond?

In his 2020 book The Great Demographic Illusion, the sociologist Richard Alba expresses hope that many of the identitarian cleavages of the 2010s will become less severe as racial boundaries blur and racial consciousness shifts toward a new “mainstream”. He tells a hopeful anecdote of going to a funeral where a very diverse set of people all interact peacefully and cordially as they all come to pay their respects. But his hope is fundamentally based on horizontal homogeneity; it’s all offline. In their pockets, the people at that funeral had cell phones that increasingly define their social interactions. And each of those cell phones was a portal its own world — its own set of personalized vertical communities. Being in the same building might have meant coming together in 1990; in 2023, that proximity itself could be the illusion.

So I’m worried about the future of our public goods. I’m worried that the online space will fragment and degrade horizontal communities, but will never manage to fully replace them either. We need to find some way to get along with the human beings who live within physical proximity to us, even as we continue to spend much of our time online. Vertical communities must make their peace with horizontal communities…somehow.

---