CAMBRIDGE – “I want to be a part of it – New York, New York,” Frank Sinatra sang of the city that has attracted so many of the world’s most ambitious people, from artists and performers to businesspeople and bankers. In a sense, this is not a difficult phenomenon to explain; metropolises like New York City, with their multicultural populations, multinational corporations, and multitude of talented individuals, are rife with opportunities. But the impact of large cities runs deeper than economic or even cultural power; cities can fundamentally change people’s lives – and even the people themselves.
In 2010, Geoffrey West, together with a team of researchers, discovered that several socioeconomic measures – both positive and negative – increase with the size of the local population. In other words, the larger the city, the higher the average wage, productivity level, number of patents per person, crime rate, prevalence of anxiety, and incidence of HIV.
In fact, when a city doubles in size, every measure of economic activity increases by about 15% per capita. That is why people move to the big city; indeed, it is why cities thrive.
This law remains constant across city sizes. And it is not unique. A growing body of evidence suggests that similar functions govern even more aspects of urban life than the research by West’s team indicated.
How can cities as ostensibly different as New York, with its towering profile, and Paris, characterized by wide boulevards, function so similarly? If, as Shakespeare suggested, a city is nothing but its people, the answer may lie in the characteristic patterns of connection, interaction, and exchange among residents.
HIV – indeed, any sexually transmitted disease – provides a particularly vivid example of the way that social networks shape urban life, as it spreads through linkages of sexual partners. Ideas – and the innovations that result from them – spread in a similar manner.
Just a few years ago, a broad investigation of these complex social networks would have been virtually impossible. After all, the available tools – isolated laboratory experiments and written questionnaires – were both imprecise and difficult to apply on a large scale.
The Internet has changed that. By enmeshing billions of people in seamless connectivity, online platforms have transformed the scope of social networks and provided new tools for researchers to investigate human interaction.
In fact, an entirely new field of study is emerging at the intersection of data analytics and sociology: computational social science. Using data collected online or through telecommunications networks – the wireless providers Orange and Ericsson, for example, have recently made some data available to researchers – it is now possible to address, in a scientific way, fundamental questions about human sociability.
A recent paper (of which one of us, Carlo Ratti, is a co-author) uses anonymized data from telecommunications networks across Europe to explore how human networks change with city size. The results are striking: in large cities, people not only walk faster (a tendency recorded since the 1960s), but they also make – and change – friends faster.
This phenomenon is likely rooted in the fact that, in accordance with West’s findings, the total number of human connections increases with city size. London’s eight million inhabitants regularly connect with almost twice the number of people as Cambridge’s 100,000 residents. This increasing exposure to people – and hence to ideas, activities, and even diseases – could explain the impact of city size on socioeconomic outcomes.
But another tendency is also consistent across cities of all sizes: people tend to build “villages” around themselves. This behavior is quantified as the networks’ “clustering coefficient” – that is, the probability that a person’s friends will also be friends with one another – and remains extraordinarily stable across metropolitan areas. Simply put, humans everywhere are naturally inclined to live within tight-knit communities.
Of course, this idea has been suggested before. The urbanist Jane Jacobs, for example, described the rich interactions occurring in New York City neighborhoods – what she called an “intricate ballet, in which the individual dancers and ensembles all have distinctive parts which miraculously reinforce each other.” What computational social science offers is the prospect of quantifying such observations and gaining insights that could shape the design of urban environments in the future.
The question is whether these insights could also unlock the power of human interactions in small towns, enabling them to access some of the social and economic advantages of a large city. In this sense, it is critical to recognize the fundamental difference between “urban villages” and their rural counterparts. In the latter, social networks are largely predetermined by family, proximity, or history. City dwellers, by contrast, can explore a wide variety of options to create custom-made villages according to their social, intellectual, or creative affinities.
Perhaps that is why Sinatra left his hometown of Hoboken, New Jersey. Only in a city like New York could he find the Rat Pack.