According to a recent survey, one third of Americans say they have never interacted with the people next door. The reasons why are profuse: more time spent socializing with “real” friends online, less free time overall, more spatial spread between neighbors.
But there may be another, more worrisome component to a lack of neighborly interaction: diversity. A working paper finds that New York City residents who live sandwiched between racial enclaves often forgo a polite door-knock when they’re irritated with their neighbors. Instead, they go straight to dialing 311, the city’s non-emergency service line.
Social scientists Joscha Legewie of New York University and Merlin Schaeffer of the Berlin Social Science Center used data from 7.7 million geocoded 311 calls made between 2009 and 2013, winnowing them down to complaint calls made about neighbors. These included grievances about blocked driveways, public drinking, use-of-space suspicions (say, an illegal Airbnb rental), and loud music, parties, or talking.
“Instead of going to your neighbor and asking them to turn the music down, you’re reaching out to an external authority to intervene,” says Legewie. “Probably because you don’t feel comfortable knocking on their door.”
The authors applied “edge-detection” algorithms (similar to what driverless cars rely on to not crash) to Census block data on race and ethnicity for all of New York City. That way, they could determine which neighborhoods were predominated by white, black, Hispanic, or Asian citizens, as well as where the lines between those communities were.
They found that some boundaries between enclaves were well-defined, with stark differences in racial make-up on either side. Others were “fuzzy”—more transitional, with a mix of residents of different races.
For example, the above map (from the study) shows a range of “edge intensities” in Crown Heights South. There, black residents predominate, while whites cluster together in about 24 city blocks. The authors write:
Edge intensity is low (no boundary) in areas without changes in ethno-racial composition such as the parts homogeneously populated by black residents or in the middle of the white enclave. Around the 24 city blocks occupied by white residents, however, we observe well-defined boundaries on the west side of the enclave and fuzzy boundaries on the north-east and south side.
Adjusting for overall 311 use and population—as well as Census data on crime, foreclosure rates, and how much residents moved around (all possible factors in neighborly conflict)—the authors found that “fuzzy boundaries” between groups produced the most complaint calls to 311.
In fact, people who lived in those liminal zones were more likely to call 311 than those who lived in truly integrated neighborhoods—that is, where racial and ethnic groups were mixed, without any boundaries. And “fuzzy boundary” residents were much more likely to have complaints about each other than those who lived along stark, clear-cut boundaries.
We might think that racially integrated neighborhoods encourage stronger, more trusting relationships between members of different groups—or what’s called the “contact theory.” But a large body of research suggests otherwise: Studies have shown that many neighborhoods share a sense of identity based on the exclusion of other races, an identity that gets rocked when “others” move in. “Group rank” also gets ambiguous when neighborhoods are more mixed: It’s less clear whose turf you’re on, and what behaviors are acceptable there.
In the early 2000s, the famed sociologist Robert Putnam found that, in a survey of 41 American communities, residents of ethnically diverse neighborhoods kept much more to themselves than those in more homogenous areas. Residents were less trusting, less altruistic, and less cooperative with one another, including with those of the same race. "People living in ethnically diverse settings appear to 'hunker down' —that is, to pull in like a turtle," Putnam wrote. Hence, they could be more likely to call the city than knock on the door when conflict arises.
Legewie’s new paper brings some nuance to these theories. “We wanted to move away from the general idea that diversity undermines community life, and actually look at the the conditions under which that might be the case and where it’s not,” he says.
There are limitations to the study, of course. Although 311 call data contains time and location information, it doesn’t include any information about the identity of the caller—age, race, or gender, for example. So it’s hard to say who is complaining about whom. Legewie also says that he hasn’t analyzed whether fuzzy boundaries between certain types of groups produced more 311 calls than others. And there’s still much more to learn about boundary areas themselves, and other factors that might be working within them—such as crime, or lots of moving in and out of homes.
Still, Legewie’s work presents a striking new angle on community tensions. It suggests it’s not so much diversity that produces the conflict, as much as it is the boundaries between people. Indeed, mixed neighborhoods that had no boundaries placed fewer calls than mixed areas sandwiched between two homogenous ones. More integrated communities would mean fewer boundaries of any kind. That could mean fewer complaint calls—and more friendly door-knocks.