Lessons From An Emerging “Science of Cities”
More than a half a century ago the urban scholar Jane Jacobs offered a scathing indictment of the “pseudo-science” of planning, which seemed, she said, “almost neurotic in its determination to imitate empiric failure and ignore empiric success.” In the presciently brilliant last chapter of The Death and Life of Great American Cities (1961) she urged planners and architects to re-assess their understanding of “the kind of problem a city is,” and apply the more inductive methods offered by the then-emerging sciences of complexity.
Although Jacobs’ work was widely acclaimed, the effort to put planning and urban design on a more usefully scientific footing has remained, at best, a work in progress. But in the last decade or so, a remarkable set of advances in related fields has begun to suggest the outlines of a more rigorous “science of cities” – one that, following Jacobs’ advice, goes beyond simple mechanical or statistical formulas, and relies more directly on the iterative, evidentiary methods of medical and biological sciences.
Recent development in urban economics offer a case in point. In her book Jacobs described the “intricate ballet” that occurs on urban sidewalks – the series of casual encounters that form the “small change” from which, as she observed, a city’s “wealth of public life may grow.” This wealth was literal as well as figurative: for Jacobs it included actual economic development. In her later economic works, she described the web-like network by which this wealth grew: an expanding interaction of contacts, knowledge, resources and creativity.
Jacobs’ work was not much noticed by economists at the time – but that has changed. Today prominent economists like Harvard’s Edward Glaeser have published noted works on “Jacobs Spillovers” – the transfers of knowledge and activity that help to explain the generation of wealth in cities like New York and London. Economists like ASU’s Jose Lobo are collaborating with physicists like Santa Fe Institute’s Geoffrey West to develop useful new insights into the complex dynamics of these urban processes, and new strategies to exploit them.
Related insights are emerging from other disciplines, shedding light on urban topics as diverse as transportation, resource use, human health and ecology. Scholars like the urban theorists Mike Batty and Sergio Porta, and the mathematician Nikos Salingaros, are describing these structures in precise mathematical and geometric terms. Although the advancements go far beyond (and began well before) Jacobs’ own work, her insights are getting notable credit. As West told me recently, “you could say we’re really just doing Jacobs with the math.”
Here are six of the most salient lessons from this emerging “science of cities:”
- Cities are spatial networks. We can readily see that the streets and pathways of a city form an inter-connected pattern in space, much like the veins and arteries of the body. Like biological networks, the networks of cities span from the largest scale (for example, highways) to the finest scales (for example, sidewalks and footpaths). To put it simply, cities are connectors.
- Cities are social networks. At any given time, a city is also a pattern of connections of people, comprising their social and economic interaction, exchange, creativity and enterprise. Importantly, these social connections must also exist in space. We must physically encounter others in space, or use proxy means to do so, such as the physical infrastructure of telecommunications networks.
- Cities are partially “scale-free”. The network structures of a city tend to be self-similar across scales, or “fractal”, in the same way that the trunks, branches and twigs of a tree are self-similar and fractal. This feature has important implications for the way a city functions.
- Cities are also partially “scale-dependent.” The people of a city are especially active within one particular scale: the scale of the human animal and its pedestrian-bounded activities. This feature also has important implications for the way a city functions – a subject of active research at present.
- Cities are partially decentralized, and partially generated by self-organizing agents. The people of a city do not passively reside entirely within structures planned and built “top-down” by experts. (When they do, these places prove to be remarkably unsatisfactory, as Jacobs observed.) Rather, they are co-creators and modifiers of the spaces of their city, often working “bottom-up.” Their actions follow on one another and react to one another, with the result that more complex kinds of structures emerge over time, with important capacities that are not otherwise possible.
- Cities are cognitive and symbolic systems. The people of a city must navigate, make sense, and find meaning within the structure of the city, through all the scales of their experience. They must also find that the spaces they inhabit are comfortable, attractive, and conducive to their experience of well-being, if they and their city are to perform at their best.
Putting these six attributes together, we can draw three extremely important conclusions about the way cities work – and the ways we as planners can help them to work better, from a human point of view.
- The economic vitality of a city depends, to a remarkable degree, on the extent to which its physical structure provides – or does not provide – the diversified capacity for exchanges that support economic interaction and “spillovers.”We are beginning to see cities as, in effect, “socio-economic reactors,” able to promote the exchange and creative interaction between people – but only to the extent that their spatial structure supports that capacity. Research is bearing out that it is the public space created by infrastructure networks – the streets, sidewalks, parks and other spaces – that plays a fundamental role in social and economic interaction rates and economic “spillovers”. While electronic communications networks are increasingly important, research (by Putnam and others) is demonstrating that there is no replacement for the physical encounter between human beings within public spaces. It seems that Jacobs’ “sidewalk encounters” are indeed the “small change” of a city’s wealth.
One might wonder about the great cities of the US Sunbelt, which have little public space. Do they not have strong economic growth? They do indeed, but the lost public spaces for interaction and spillovers are replicated in private offices, schools and clubs, and in the infrastructure networks needed to sustain them: the automobiles, buildings and electronic systems on which this system depends.
There are two problems with this state of affairs. One, these systems are extremely resource-intensive. Two, they carry a number of other drawbacks, including limitations in the formation of “social capital,” as work by the sociologist Robert Putnam and others has indicated. Putnam suggests that while such networks (including electronic networks) can be complementary, they do not replace the physical encounters of public space, and the best cities offer “multi-stranded” connections.
A related finding of this work is that the socio-economic potential of a city is maximized when all of its citizens are socially connected and included, as West’s colleague Luis Bettencourt has noted. By contrast, to the degree that cities exclude some populations, or limit their transportation mobility and safety, then the entire city will under-perform socially and economically, creating a drag on overall economic performance. When it comes to cities, it seems clear that social inclusion is good for the bottom line.
For Bettencourt and others, this implies that effective urban planning should above all seek to promote the capacity for uninterrupted social connectivity, by promoting spatial land use and designing infrastructure that is commensurate with connectivity over the city as a whole. Barriers to this connectivity – railroad tracks, rivers, freeways, even large parks and campuses – must not be allowed to interrupt the continuous fabric of the city.
- The resource efficiency of a city and its inhabitants depends, to a surprising degree, on the extent to which its spatial structure provides – or does not provide – the capacity for connectivity at the human and pedestrian scale.Recent research into the characteristics of resource-efficient cities around the world shows consistent patterns of self-organizing networks of people, activities, structures and spaces, operating at remarkably distinct scales and configurations. To the extent the top-down actions of planners complement the formation of these self-organizing network patterns, the cities are much more resource-efficient (as we can observe empirically in many such cities). By contrast, to the extent the top-down actions of planners displace these patterns and restrict their formation, the cities are much less resource efficient by comparison. The cities of the US , with their mechanically ordered (and resource-intensive) freeways, suburbs and building distributions, demonstrate well the latter condition.Importantly, these self-organizing networks are centered around the scale of pedestrians, and as the evidence is showing, many benefits flow from this pedestrian-scaled organization: compactness, multi-modal travel, social interaction and health benefits, and behavioral and demand impacts. (This is also an active and promising area of research at present.) By contrast, many newer cities (including rapidly-urbanizing cities of the developing world) are organized largely at the scale of vehicles, and the functionally segregated systems on which they operate. This “tree-like” segregation of large-scale parts turns out to have profound consequences for the organization of neighborhoods, buildings and urban networks (an issue documented brilliantly by the mathematician and architect Christopher Alexander, in his 1965 essay “A city is not a tree.”)
- The vitality of a city depends, to a surprising degree, on its capacity to empower its residents to selectively modify the degree of connectivity of the spaces around them, so as to control and modulate their levels of interaction – and in time, to make beneficial transformations to the city as a whole. Of course not all connectivity is beneficial: there are negative factors that can come with connectivity, such as loss of privacy, stress, noise, traffic dangers, pollution and crime. Total connectivity would be no more pleasant or useful than living on a busy street corner. Therefore, the capacity to modulate these connections, and to regulate them as a function within the urban network, is equally critical.Some of this regulation can come top-down, through formal enforcements like policing. But importantly, the agents of a city, its residents, are able to perform a range of modifications to urban connectivity from the “bottom up”, based upon specific local information as well as their own goals and aspirations. In so doing they serve as an important force of urban self-organization.
Indeed, if we look at the way people use urban space, we can see them operating a remarkably complex network of user-controlled connections between room-like enclosures that extend from the most private realms (the literal rooms of bedrooms and baths, for example) to the most public realms (the “urban rooms” of streets, sidewalks and parks, for example) and the many transitional spaces in between. These places have a network relationship that is generally hierarchical, but that also has many overlapping inter-connections. Because we are not only considering abstract spatial configurations but also human-experienced places, with all their psychological and phenomenological attributes, my colleagues and I refer to this emerging unified conception as “place network theory.”
Moreover, these varied places afford, to some degree, control of the connectivity between them. We can open or close (and even lock) doors and windows, draw blinds, close gates. We can retreat to the quietest and most sheltered places, or advance into the busiest and most interactive places, and once we are in any given place, we can modulate our degree of connection (by opening a window, say).
At larger scales, we can make more permanent changes by building structures – doors, windows, walls, fences whole buildings – or make choices that prompt others to build structures (like shopping at some stores and not others). Through an immense number of such small actions over varying time spans, the city self-organizes from the bottom-up, into a highly complex pattern of evolving structures.
Moreover, the connections modulated by our structures are not just connections of human movement, but also of sight, sound and smell. We can install a window that allows sight but not sound (or movement), or a hedge that allows sound but not sight. We can then open the window and allow both sight and sound, or close it and draw a blind, shutting off both sight and sound.
When we step back and look at this inter-connected network at a city scale, we see that it is a vast and very complex structure, and moreover one that is growing and evolving in time. The individual residents are, of course, constituents of many local and informal social structures that mediate all these outcomes, good and bad – the groups of citizens who informally regulate public spaces, the owners and proprietors who look after their spaces and the people within them, and other informal groups and activities that serve to informally regulate public space. They are also agents acting within markets and legal systems to meet their their own needs and interests within collective and cooperative structures, and within the ultimate political entity of the city itself.
The growth of great cities like Rome and Venice can be seen as an immense collection of just such transformations, consisting of many small-scale and progressively fewer larger-scale adjustments, all interacting and compounding over time, and creating beautifully intricate, human-adapted structures. We now see that these structures were not just aesthetically beautiful, but (for related reasons) they offered to their residents remarkable capabilities for opportunity, efficiency and vitality. On many metrics of sustainability they perform remarkably well, in many cases putting our “modern” structures to shame.
By contrast, cities that limit the extent of these spatial networks, or the degree of user-controlled connectivity within them (including many “modern” cities) also exhibit a cascading series of problems, of just the sort Jacobs described in her book. At best their neighborhoods are “dull and inert” places, lacking vitality. At worst, as Jacobs pointed out, they profoundly damage the lives, squander the resources, and limit the opportunities of their residents.
A timely and urgent agenda
We are entering an age of critical challenges to humanity, among them depletion and degradation of resources, climate instability, and disruptions to sustainable economic activity, often caused by the unintended consequences of our own actions. The new research offers to us something that all good science should do, particularly in light of our current challenges. In this case, it gives us a useful picture of how cities work, affording us the basis to learn to make them work better. Specifically, we now see that cities (and smaller towns too, to some extent) have an inherent structural capacity to connect us to opportunities, resources and enriching experiences, while providing remarkable ecological benefits. But this capacity must be understood, supported and expanded (and certainly not damaged) by our conscious acts of planning and design.
Unfortunately, at just the historic moment that we most need this inherent capacity of cities to work in our favor, it seems the older, failing models of city-making decried by Jacobs are still dominating the professions, and even accelerating – perversely so, in light of our new knowledge. In planning, vast new areas of rapidly-growing cities around the world are following the sprawling models of Post-War America, with disturbing consequences for long-term global resource depletion. In architecture, the closely related 1930s CIAM model of object-buildings on superblocks survives and grows into ever larger scales, costumed only in extravagant new art packaging. In view of our challenges, these practices would seem to qualify as Jacobs’ “pseudo-science” at best, and professional malpractice at worst.
But the lessons of the new sciences of cities do not suggest a marginalized role for planners and urban designers. They do not suggest that we are hopeless before the forces of economics, capital or bureaucracy – far from it. They certainly do not support any kind of simplistic libertarian prescription, reflecting the charming but quite wrong view that great cities arose in a vacuum.
On the contrary, these insights offer a different strategy for planning – more directly utilizing the dynamics by which a city spontaneously self-organizes, so that those dynamics can be influenced to operate in much more beneficial ways for their human occupants. Among other exciting revelations, these insights open the door for us to develop and apply effective new tools for a more “generative” planning approach, working in networks across many scales of democratic and participatory action. Like doctors who understand and support the body’s own immune system function, we can exploit the age-old vernacular processes that are already occurring, to produce much better cities – better able to meet both old and new human challenges.