Emergence

by Steven Johnson

  Much has been made of the fact that you can’t ever “win” at SimCity, but it’s probably more important to note that you don’t really “play” SimCity either, at least the way we talk about playing conventional games. Users grow their virtual cities, but the cities evolve in unpredictable ways, and control over the city’s eventual shape is always indirect. You can create commercial zones or build a highway, but there’s never a guarantee that the neighborhood will take off or the crime rate go down. (It’s far from random, of course—longtime players learn how to push their virtual citizens in certain directions.) For most people, the sight of their first digital town sprouting upscale neighborhoods and chronically depressed slums is downright eerie, as though the hard math of the digital computer had somehow generated a life-form, something more organic and fluid, somewhere between the rigid dictates of programming and pure randomness.

  How did Wright create this extraordinary illusion? By designing the game as an emergent system, a meshwork of cells that are connected to other cells, and that alter their behavior in response to the behavior of other cells in the network. A given city block in SimCity possesses a number of values—the price of the land, say, or its pollution level. As in a real-world city, these values change in response to the values of neighboring blocks; if the block to the west drops in value, and the eastern neighbor develops a higher crime rate, then the current block may well grow a little less valuable. (A sophisticated SimCity player might counter the decline by placing a police station within ten blocks of the depressed area.) The algorithms themselves are relatively simple—look at your neighbors’ state, and change your state accordingly—but the magic of the simulation occurs because the computer makes thousands of these calculations per second. Because each cell is influencing the behavior of other cells, changes appear to ripple through the entire system with a fluidity and definition that can only be described as lifelike.

  The resemblance to our ants and embryos is striking. Each block in SimCity obeys a set of rigid instructions governing its behavior, just as our cells consult the cheat sheet of our genes. But those instructions are dependent on the signals received from other blocks in the neighborhood, just as cells peer out through gap junctions to gauge the state of their neighbors. With only a handful of city blocks, the game is deathly boring and unconvincingly robotic. But with thousands of blocks, each responding to dozens of variables, the simulated cityscape comes to life, sprouting upscale boroughs and slums, besieged by virtual recessions and lifted by sudden booms. As with ant colonies, more is different. “Great cities are not like towns only larger,” Jane Jabobs writes. “They are not like suburbs only denser. They differ from towns and suburbs in basic ways.” She was writing, of course, about real-world cities, but she could just as easily have been talking about SimCity’s networked algorithms, or the teeming colonies of Arizona harvester ants.

  Economists and urban sociologists have also been experimenting with models that can simulate the ways that cities self-organize themselves over time. While actual cities are heavily shaped by top-down forces, such as zoning laws and planning commissions, scholars have long recognized that bottom-up forces play a critical role in city formation, creating distinct neighborhoods and other unplanned demographic clusters. In recent years, some of those theorists—not to mention a handful of mainstream economists—have developed more precise models that re-create the neighborhood-formation process with startling precision.

  The economist (and now New York Times editorialist) Paul Krugman’s 1995 lectures, “The Self-Organizing Economy”—published as a book the following year—include a remarkably simple mathematical model that can account for the “polycentric, plum-pudding pattern of the modern metropolis.” Building on the game-theory models that Thomas Schelling developed to explain how segregated cities can form, Krugman’s system assumes a simplified city made up only of businesses, each of which makes a decision about where to locate itself based on the location of other businesses. Some centripetal forces draw businesses closer to one another (because firms may want to share a customer base or other local services), and some centrifugal forces drive businesses farther apart (because firms compete for labor, land, and in some cases customers). Within that environment, Krugman’s model relies on two primary axioms:

  1. There must be a tension between centripetal and centrifugal forces, with neither too strong.

  2. The range of the centripetal forces must be shorter than that of the centrifugal forces: business must like to have other businesses nearby, but dislike having them a little way away. (A specialty store likes it when other stores move into its shopping mall, because they pull in more potential customers; it does not like it when stores move into a rival mall ten miles away.)

  “And that’s all that we need,” Krugman continues. “In any model meeting these criteria, any initial distribution of businesses across the landscape, no matter how even (or random), will spontaneously organize itself into a pattern with multiple, clearly separated business centers.”

  Krugman even provides a chart demonstrating the city’s self-organization in time—an image that captures the elegance of the model. Scatter a thousand businesses across this landscape at random, then turn on the clock and watch them shuffle around the space. Eventually, no matter what the initial configuration, the firms will gather into a series of distinct clusters evenly spaced from each other. There’s no rule for clustering that the businesses are directly obeying: their motives are strictly local. But those micromotives nevertheless combine to form macrobehavior, a higher order that exists on the level of the city itself. Local rules lead to global structure—but a structure that you wouldn’t necessarily predict from the rules.

  Krugman talks about his “plum pudding” polycentrism as a feature of the modern “edge city,” but his model might also explain an older convention: the formation of neighborhoods within a larger metropolitan unit. Neighborhoods are themselves polycentric structures, born of thousands of local interactions, shapes forming within the city’s larger shape. Like Gordon’s ant colonies, or the cells of a developing embryo, neighborhoods are patterns in time. No one wills them into existence single-handedly; they emerge by a kind of tacit consensus: the artists go here, the investment bankers here, Mexican-Americans here, gays and lesbians here. The great preponderance of city dwellers live by those laws, without any legal authority mandating that compliance. It is the sidewalk—the public space where interactions between neighbors are the most expressive and the most frequent—that helps us create those laws. In the popular democracy of neighborhood formation, we vote with our feet.

  *

  A friend of mine who moved to California a few years ago once remarked to me, with a straight face, “The class segregation in Los Angeles is not nearly as bad as you might think. You’d be surprised how many low-income areas I pass on the freeway when I’m driving into work.”

  It was one of those comments that reveals an entire weltanschauung. “It’s not ‘an encounter with the working class,’” I thundered back, “if you’re gazing down at them from the overpass.” But he had a point. In a dispersed, car-centric city like Los Angeles, highways are the connecting nodes, one of the few zones where the city’s different groups encounter each other—albeit at sixty-five miles an hour.

  Ever since Death and Life was first published in the early sixties, Jacobs-inspired critics have lambasted the dispersed communities of L.A. and Phoenix, and their even more anonymous descendants—the “edge cities” that have sprouted up around convenient freeway intersections or high-volume parking lots, the way towns once nestled up to harbors or major rivers. Progressive urbanists bemoaned the mallification of the American city, with vibrant public streets giving way to generic, private shopping complexes. The sidewalk carnivalesque that had so vividly been captured by Wordsworth and Baudelaire in the previous century seemed headed the way of the horse and buggy, and in each case, the culprit turned out to be the same: the automobile, which necessitate
d all the injuries of sprawl—mixed-use zoning, gated communities, deserted or nonexistent sidewalks.

  At the core of this lamentable transformation was the street itself, and the interactions between strangers that once took place on it. The brilliance of Death and Life was that Jacobs understood—before the sciences had even developed a vocabulary to describe it—that those interactions enabled cities to create emergent systems. She fought so passionately against urban planning that got people “off the streets” because she recognized that both the order and the vitality of working cities came from the loose, improvised assemblages of individuals who inhabited those streets. Cities, Jacobs understood, were created not by central planning commissions, but by the low-level actions of borderline strangers going about their business in public life. Metropolitan space may habitually be pictured in the form of skylines, but the real magic of city living comes from below.

  Part of that magic is the elemental human need of safety. Chapter 2 of Death and Life investigates the way dense urban settlements collectively “solve” the problem of making themselves safe, a solution that has everything to do with the local interactions of strangers sharing the public space of the sidewalks:

  Under the seeming disorder of the old city, wherever the old city is working successfully, is a marvelous order for maintaining the safety of the streets and the freedom of the city. It is a complex order. Its essence is intimacy of sidewalk use, bringing with it a constant succession of eyes. This order is all composed of movement and change… . The ballet of the good city sidewalk never repeats itself from place to place to place, and in any one place is always replete with new improvisations.

  After a long and wonderfully detailed portrait of one day’s choreography, Jacobs ends with one of the great passages in the history of cultural criticism:

  I have made the daily ballet of Hudson Street sound more frenetic than it is, because writing it telescopes it. In real life, it is not that way. In real life, to be sure, something is always going on, the ballet is never at a halt, but the general effect is peaceful and the general tenor even leisurely. People who know well such animated city streets will know how it is. I am afraid people who do not will always have it a little wrong in their heads—like the old prints of rhinoceroses made from travelers’ descriptions of the rhinoceroses.

  On Hudson Street, the same as in the North End of Boston or in any other animated neighborhoods of great cities, we are not innately more competent at keeping the sidewalks safe than are the people who try to live off the hostile truce of turf in a blind-eyed city. We are the lucky possessors of a city order that makes it relatively simple to keep the peace because there are plenty of eyes on the street. But there is nothing simple about that order itself, or the bewildering number of components that go into it. Most of those components are specialized in one way or another. They unite in their joint effect upon the sidewalk which is not specialized in the least. That is its strength.

  Again, we are back to the world of the ants: random local interactions leading to global order; specialized components creating an unspecialized intelligence; neighborhoods of individuals solving problems without any of those individuals realizing it. And safety is only part of the story: there are many “uses of sidewalks” in Death and Life, some of which we will encounter in later chapters.

  The key here is that sidewalks are important not because they provide an environmentally sound alternative to freeways (though that is also the case) nor because walking is better exercise than driving (though that too is the case) nor because there’s something quaintly old-fashioned about pedestrian-centered towns (that is more a matter of fashion than empirical evidence). In fact, there’s nothing about the physical existence of sidewalks that matters to Jacobs. What matters is that they are the primary conduit for the flow of information between city residents. Neighbors learn from each other because they pass each other—and each other’s stores and dwellings—on the sidewalk. Sidewalks allow relatively high-bandwidth communication between total strangers, and they mix large numbers of individuals in random configurations. Without the sidewalks, cities would be like ants without a sense of smell, or a colony with too few worker ants. Sidewalks provide both the right kind and the right number of local interactions. They are the gap junctions of city life.

  This is one of those instances where thinking about a social problem using the conceptual tools of emergence sheds genuinely new light on the problem, and on the ways it has been approached in the past. Since Death and Life, the celebration of sidewalk culture has become an idée fixe of all left-leaning urbanists, an axiom as widely agreed upon as any in the liberal canon. But the irony is that many of the same critics who cited Jacobs as the initial warrior in the sidewalk crusade misunderstood the reasons why she had embraced the sidewalk in the first place. And that is because they saw the city as a kind of political theater, and not as an emergent system. The clash and contradiction of city streets—versus the antiseptic segregations of suburbia—became a virtue in and of itself, something that people should be “exposed to” for their own good. The logic was a kind of inverted rendition of the old bromides about kids watching too much television: if people were somehow deprived of the theatrical conflicts of city sidewalks, they’d all end up hollow men—or worse, Republicans.

  This turns out to be an aesthetic agenda wrapped up in a thin veil of politics. Some critics carried their paeans to sidewalk diversity to laughably condescending extremes. “Poor people have taught us so much about what we know about being fully alive in public,” Marshall Berman wrote in an early-eighties essay called “Take It to the Streets.” “[They’ve taught us] about how to move rhythmically and melodically down a street; about how to use color and ornamentation to say new things about ourselves, and to make new connections with the world; about how to bring out the rhetorical and theatrical powers of the English language in our everyday talk.” Paraphrase: Those poor people have so much rhythm!

  However much Berman might resist the idea, the very same morality play underlies my friend’s ode to L.A. freeway culture: both perspectives assume that seeing racial and economic diversity is intrinsically good for you, like some kind of political cardiovascular workout. From this perspective, what was laughable about my friend’s observation was the idea that he could truly take in the “melodic movements” or hear the “rhetorical” flourishes of South Central while driving on the highway. The exposure itself is assumed prima facie to be good for the soul. The only question is whether my friend was getting a big enough dosage from his car.

  This is all perfectly commendable, if a little patronizing, and for all I know we might indeed turn out to be more charitable and expansive people if we encountered more diversity on our streets. But that diet has nothing to do with the Jacobs understanding of sidewalks and their uses. According to the gospel of Death and Life, individuals only benefit indirectly from their sidewalk rituals: better sidewalks make better cities, which in turn improve the lives of the city dwellers. The value of the exchange between strangers lies in what it does for the superorganism of the city, not in what it does for the strangers themselves. The sidewalks exist to create the “complex order” of the city, not to make the citizens more well-rounded. Sidewalks work because they permit local interactions to create global order.

  From this angle, then, the problem with my friend’s sojourns on the Santa Monica Freeway—and indeed the problem with all car-centric cities—is that the potential for local interaction is so limited by the speed and the distance of the automobile that no higher-level order can emerge. For all we know, there may well be something psychologically broadening in gazing out over the slums from your Ford Explorer, but that experience will do nothing for the larger health of the city itself, because the information transmitted between agents is so famished and so fleeting. City life depends on the odd interaction between strangers that changes one individual’s behavior: the sudden swerve into the boutique you’ve never noticed before, or the decis
ion to move out of the neighborhood after you pass the hundredth dot-com kid on a cell phone. Encountering diversity does nothing for the global system of the city unless that encounter has a chance of altering your behavior. There has to be feedback between agents, cells that change in response to the changes in other cells. At sixty-five miles an hour, the information transmitted between agents is too limited for such subtle interactions, just as it would be in the ant world if a worker ant suddenly began to hurtle across the desert floor at ten times the speed of her neighbors.

  And so this is the ultimate lesson of Jacobs’s sidewalks, and of her way of thinking about cities as self-organizing systems. The information networks of sidewalk life are fine-grained enough to permit higher-level learning to emerge. The cars occupy a different scale from the sidewalks, and so the lines of communication between the two orders are necessarily finite. At highway speed, the only complex systems that form are between the cars themselves—in other words, between agents that operate on the same scale. Unlike the ballet of the pedestrian city, these are global patterns that would be familiar to any resident of Los Angeles. We call them traffic jams.