The Mystery of Cities

There is so much bad news about American cities that it’s surprising to discover that they are not shrinking but growing. According to the last US census, the twenty-three largest cities, which have populations over one half million, grew by 6 percent in the period between 1980 and 1990; the forty medium-sized cities with populations between one quarter and one half million increased by 20 percent in the same period; as for the 131 cities with between 100,000 and 250,000 people, they have grown by more than 15 percent, and today, for the first time in the twentieth century, they are home to more people than the great metropolises.

If this urban expansion has not had much attention, it is probably because the form of these cities is both brash and ambivalent. “Impressive in scale but limited in vision, creating new opportunities but also providing massive new problems,” wrote Asa Briggs.¹ He was describing Victorian cities in Britain, but he may as well have been writing about the American city of today: spread out, socially fragmented, recklessly entrepreneurial, relying almost completely on the automobile, often lacking a clearly defined center, and without the trappings of urbanity that have characterized cities in the past. In the eyes of many, a distinctly un-urban place.

But what is a city? Lewis Mumford asked the question more than thirty years ago in the opening of his great work, The City in History. Although Mumford never provided a single, simple definition of a city, he did point out that neither size, nor population density, nor physical extent, nor function provided reliable yardsticks. The cities of ancient Greece, for example, with only a few thousand inhabitants, were the size of today’s small towns; indeed, Aristotle maintained that the ideal city should contain about 5,000 people, and that a settlement with 100,000 inhabitants could no longer be considered a city. In medieval Germany, towns of 3,000 inhabitants were granted the status of cities, and in medieval France, there were numerous examples of walled cities with as few as 200 or 300 households, that is, probably no more than 2,000 or 3,000 people. In the American Old West, the grandly named Dodge City, which was laid out in 1872, was described by a contemporary visitor as consisting of about a dozen frame houses and two dozen tents, beside a few sod houses. Even in its heyday, when Dodge City had the reputation of being the Western frontier’s wildest town, rivaled only by Deadwood and Tombstone, the permanent population was probably well under one thousand people, by today’s standards a small town.

Most people take it for granted that there is an important difference between a city and a town, a difference of kind as well as degree. It comes as a surprise, then, to discover that in most of the European, languages, the same word can be used to refer to a “city” and a “town.” Some languages, like German, French, or Polish, do not distinguish between them at all; a city is a ville and a town is a ville. This suggests a view of cities and towns that recognizes continuity rather than differences, as if cities were larger versions of towns, or vice versa. On the other hand, in English, and especially in American English, the perceived difference between cities and towns, especially between big cities and small towns, is considered paramount.

If size doesn’t define a city, perhaps population density does. Medieval European cities were closely packed behind defensive walls, and the citizens were privileged to live inside the walls—the poor lived outside, in the suburbs or faubourgs. But the vast Chinese cities from the Yuan dynasty onward were spread out and included affluent suburbs, and many modern American cities are exceedingly expansive—the city limits of Houston encompass more than five hundred square miles. On the other hand, there are many large cities in souther Africa where the presence of farm animals and agricultural land, for example, suggests a vastly expanded rural village rather than a town. But then, in the early nineteenth century, foraging pigs were a common sight in New York City streets.

The functions of cities can vary widely. Cities have been vividly described by Jane Jacobs as places of commercial exchange and industrial innovation, but in many Greek cities all commerce was explicitly forbidden, and Mumford makes a convincing case that the three cities that stand as the best example of Greek urban culture are Delphi, Cos, and Olympia. Delphi was an oracular religious center, Cos was a combination sanatorium, research hospital, and spa, and Olympia was the site of quadrennial athletic games. Noncommercial cities are equally prominent. Few cities are as influential as the holy cities of Rome, Jerusalem, and Mecca; one of the most charming cities in Europe, Venice, has been a sort of amusement park, as Mary McCarthy put it, for almost three hundred years; and Las Vegas, a city that stands as an almost mythic symbol in the imaginations of millions of Americans, is purely a play-town. There is no longer a single city associated with sports, but it is indisputable that for many Americans what distinguishes a “big” city in the United States today is not its size, or how tall its skyscrapers are, or the influence of its newspaper, but its ability to attract a major-league baseball team.

Nevertheless, whatever else they are, cities are also physical artifacts. They are not the biggest man-made objects in the world—they are not as big as works of pure engineering like the Great Wall of China or the Panama Canal, or the continental telephone system—but what they lack in extent they make up for in visceral impact. The US telephone system is huge but largely invisible, and only a small part of the Great Wall can be seen at a time—the immensity of these creations makes itself felt only in the imagination. But a city is experienced all of a piece. That is why great city views, whether of Paris spread out below the heights of Sacré Coeur, or the crowded island of Hong Kong seen from Kowloon, or lower Manhattan looming over the Staten Island ferry, are so moving. Such views are a potent reminder that the construction of cities represents a great human achievement. “It was divine nature which gave us the country,” wrote the Roman scholar Marcus Terentius Varro, “and man’s skill that built the cities.”

The city as artifact is the theme of Spiro Kostof’s The City Shaped and its companion volume, The City Assembled, which together comprise the most ambitious review of urban form since The City in History. Like Mumford, Kostof takes as his theme the city in a historical setting, although he broadens the scope and includes cities in China, Japan, India, and Latin America. Unlike Mumford, however, he does not organize his material chronologically but thematically, for his intention is not to study cities as examples of a sort of evolution (or devolution, as Mumford would have it) but to examine the art of making cities in different historical periods.

Economists see them as engines for commercial development, sociologists as settings for human behavior, and art historians as settings for works of architecture. Kostof was a professor of architectural history at the University of California (he died in 1991), and his subject is urban form, that is, the way in which buildings are arranged in a city, and the patterns of the public spaces: the streets, squares, and parks. His aim is not so much to judge the success or failure of these arrangements, as to understand their cultural meanings and to define the processes that produced them. In this, he succeeds admirably.

The City Shaped describes five approaches to urban form that have been used throughout history in a variety of cultures: cities that have evolved according to organic patterns, cities planned as grids, symbolic and diagrammatic layouts, the Baroque plan, the urban skyline. This scheme is roughly historical since the organic plan corresponds to the medieval town, which eventually gave way to the grand devices of Baroque town planning—including long axial views—which in turn were supplanted by the vertical modern city. On the other hand, grids have existed since the time of the ancient Greeks and are a constant theme in town planning, whether in eighth-century Heijokyo, Japan, thirteenth-century Suzhou, China, eighteenth-century Philadelphia, nineteenth-century Barcelona, or in twentieth-century English new towns. Similarly, towns whose layouts incorporate political or religious symbolism appear at different periods: as exquisite starbursts in the fortified towns of the European Renaissance, as religious mandalas in Hindu India, as imperial diagrams in British New Delhi, or as a great political symbol of democracy in Washington, DC.

Washington, DC, was planned mainly by Pierre Charles L’Enfant, and New Delhi was the work of Edwin Lutyens and Herbert Baker, but most cities, as Kostof underlines, “came about without benefit of designers, or once designed, set about instantly to adapt themselves to the rituals of everyday life and the vagaries of history.” The strength of The City Shaped is that, unlike many books on city form, it avoids the idealized theories of urban design—and the proposals for idealized but unbuilt cities—and deals, instead, with the messy, haphazard, and often makeshift processes according to which real cities are fashioned.

Several years ago, Kostof was the host for a public-television series titled America by Design. This experience obviously colored the organization of The City Shaped, which is handsomely illustrated and is as much a book to be looked at as read (there are more than 350 illustrations, many in color, all beautifully printed). There are not only diagrams and plans (which urbanists and architects will appreciate) but also reproductions of paintings, engravings, and drawings, as well as many interesting photographs. The analogy to a television documentary does not end there. The five long chapters resemble one-hour episodes: in each, a general theme is introduced and the reader is presented with a kaleidoscope of visual images and written descriptions. Kostof is particularly skillful at weaving together unexpected material.

The chapter on “The Urban Skyline,” for example, includes not only a predictable discussion of American twentieth-century skyscraper cities, but also descriptions of medieval and Renaissance towns. In addition to such well-known examples as New York and Chicago, Kostof introduces the skylines of Moscow, Jerusalem, and Edinburgh. He provides some arresting urban images: post–World War II Moscow, with its pompous ring of high-rise buildings, a mid-nineteenth century panorama of Istanbul, punctuated by the domes and minarets of several mosques, and contrasting views of London, one—painted in 1746 by Canaletto—a beautiful skyline of church spires (the majority built by Christopher Wren), and a 1990 photograph in which the dominant monument is the grossly overdone tower of Canary Wharf, a lonely symbol of financial miscalculation and architectural hubris.

The contrasting images are stimulating; still, watching television is one thing, and reading books another. Contemplating the rich mixture of cityscapes that Kostof has put together the reader is occasionally left wondering exactly where the author is headed. Like many public-television-documentary hosts, Kostof is knowledgeable, genial, and evenhanded—he has no polemical point to make, except a general affection for the traditional, pedestrian city center. While this allows for a more balanced appraisal of the urban past than was contained in The City in History, say, one misses Mumford’s occasional bursts of indignation and righteous outrage. Sometimes, Kostof’s detachment is downright odd, as when it produces a dispassionate description of the urban qualities of Neuengamme, a Nazi concentration camp, for example. Nor does his uncritical description of the new capital of Brazil ring true. Robert Hughes once called Brasília “an expensive and ugly testimony to the fact that, when men think in terms of abstract space rather than real place, of single rather than multiple meanings, and of political aspirations instead of human needs, they tend to produce miles of jerry-built nowhere.”² To Kostof, the Brazilian city is merely “iconoclastic as a product of the Sixties.”

Kostof spent the three months before his untimely death completing the second part of his urban study, The City Assembled (his research assistant, Greg Castillo, is credited as a collaborator). Although the book resembles The City Shaped in general format and in its visual richness, its organization is different. The City Assembled examines four physical constituents of cities: streets, public places, districts, and city edges. This deceptively simple scheme provides a solid underpinning for what Kostof does best, synthesizing large amounts of diverse information. The City Assembled provides the most comprehensive discussion of the relationship between human activity and urban form that I know of in a single volume.

The experience of a city is, first of all, the experience of its streets. Indeed, as Kostof writes, without streets there can be no city, not simply for practical reasons, but because the public nature of streets provides the required setting where the heterogeneous population of the city can mingle, politically as well as socially. This mingling occurs in ritualistic fashion in the form of parades and Sunday promenades, spontaneously in demonstrations and protest marches, and daily in activities such as strolling, shopping, and going to work. The design of streets shows an astonishing variety: straight or curved, with or without trees, flanking a waterway, arcaded, lined with buildings, or walls, or gardens.

Ever since Roman times, the vehicular street has accommodated pedestrians on separate, raised pavements. Sidewalks went out of fashion during the Middle Ages, but reappeared in seventeenth-century London and have been a fixture ever since. Streets reserved entirely for pedestrians were introduced in the 1960s, and have been very popular in European countries such as Germany and Denmark; Americans seem less drawn to carless streets, however, and many pedestrian malls have been returned to mixed use. The most prominent twentieth-century additions to the street lexicon are the elevated and underground pedestrian walkway systems, adopted by Minneapolis, Houston, and Montreal. Kostof is probably right in describing these skyways and tunnels as pseudo-streets, but his claim that “the obdurate pedestrian…can be seen shunning these carefully engineered environments and seeking the ground level when given the opportunity” is not borne out in Montreal, whose residents have adopted the underground city with great enthusiasm.

All cities, no matter the culture or the epoch, provide their citizens with places to congregate, not only squares and plazas, but also public gardens and parks. These vary enormously in scale, from Siena’s intimate Campo to Beijing’s totalitarian Tiananmen Square, from a neighborhood playground to Central Park. Urban public places also differ in the type of activities they contain. Palace Square, in St. Petersburg, was designed specifically for military parades, the Plaza Mayor, in Madrid, was built to hold bullfights, Covent Garden was a market square, and although the royal squares of Paris, Nancy, and Bordeaux (Place Dauphine, Place Vendôme, Place des Vosges, Place Stanislas) had groundfloor shops, their chief function was to celebrate royal power, and incidentally to provide an opportunity for aristocratic real-estate speculation. In the modern city, the enclosed shopping mall has become a ubiquitous public gathering space, not only in the United States but also in Europe, Latin America, and Asia.

Looking at how different cultures at different times have undertaken to create streets and public places points up an interesting feature of cities—their transcendence of history. Cities, unlike most artifacts, do not really become obsolete or, at least, rarely completely so. The center of Paris can be experienced as a historical setting, but it is also a place for contemporary life. Old weapons—or old clothes—are put away in museums, but with rare exceptions (Williamsburg is one) old cities continue to be lived in, even if, as in the case of Venice or Charleston, their commercial functions have drastically altered.

Cities, unlike villages and towns, are characterized by a rich mixture of specialized functions which are usually accommodated in distinct districts. According to Kostof, cities traditionally contained four types of district: administrative, religious, commercial, and residential. Perhaps because his chief concern is aesthetic, Kostof omits industrial areas, despite the fact that manufacturing plays an increasingly important role in the life of cities after the industrial revolution. The architectural impact of workshops, breweries, loft buildings, and small plants is minor, but their economic role is crucial. Indeed, Jason Epstein has argued that the decline of New York City since the 1950s is due in large part precisely to its loss of almost 800,000 manufacturing jobs.³

But cities were not only divided according to different functional requirements, there were also spatial divisions of class (often vertical, as in Philadelphia’s Society Hill, Boston’s Beacon Hill, San Francisco’s Nob Hill, or Los Angeles’s Beverly Hills) and ethnicity. The latter was sometimes formalized in the plan of a city, which was the case of many Jewish ghettos (originally a Venetian term that referred to an island reserved for Jews), the native quarters of French colonial towns, and the black townships of South Africa. In American cities, ethnic neighborhoods were—and are—the rule rather than the exception, as is shown by the many Germantowns, Little Italys, Little Tokyos, Chinatowns, Koreatowns, and Little Havanas, as well as the hundreds of black de facto ghettos.

Cities have to have edges, whether these are fortifications, industrial slums, greenbelts, or suburbs. Kostof writes engagingly of these various conditions, and also includes a discussion of “natural” edges: river fronts, corniches, and harbors. His international survey underlines the contrast between European cities, which usually treated rivers and lakes as civic amenities, and American cities, for whom waterfronts were merely access to cheap transportation. Chicago’s lakefront parks and boulevards are a notable exception.

Kostof closes his chapter on the city edge with a discussion of suburbs. Interestingly, he reminds us that this is not an exclusively American phenomenon. Canada may have a national health plan and many social-democratic programs, but the suburban character of its metropolitan areas differs little from that of the United States. Well-to-do suburbs appeared outside London in the eighteenth century, and are found surrounding many European cities today. Residential suburbs—for the poor as well as for the rich—are springing up outside Mexico City, Nairobi, Bombay, and Hong Kong.

Kostof is not optimistic about what he calls the redistribution of urban vitality from the traditional center to the edges. Of course, the development of some sort of residential satellites around a metropolitan center is inevitable in large cities; after all, 18 million people can hardly live on the island of Manhattan, any more than 10 million Frenchmen can crowd into the twenty arrondissements of the old center of Paris. But I think that what concerns the author, as it does many urban critics, is the nature of the relationship between the different parts of the modern metropolis. “There has always been some vestigial pretense that suburbia is not complete in itself,” Kostof writes, “that the umbilical link that ties it to the city center cannot be totally severed.” This pretense is no longer possible, for many satellite cities have become almost totally self-sufficient, providing their own recreation, shopping, and workplaces, and effectively isolated from the traditional downtown; in many cases, suburban communities are economically and racially separated as well. American suburbs tend to be middle class and predominantly white; French suburbs, on the other hand, tend to be working class, and often nonwhite.

In a small but important book, Cities Without Suburbs, David Rusk proposes that this separation may be overcome not, as his title might suggest, by building different kinds of cities, but by organizing existing American cities in a different fashion. Rusk, the former mayor of Albuquerque, studied 522 central cities in 320 metropolitan areas in the United States between 1950 and 1990. He divides cities into two categories: “elastic” and “inelastic.” Elastic cities are cities that either have enough vacant land inside their boundaries to accommodate growth, or have expanded inside their city limits by annexing undeveloped land or surrounding communities. Inelastic cities are locked into existing boundaries and metropolitan growth occurs outside their city limits. Examples of elastic cities are Houston, whose area increased from 160 to 540 square miles between 1950 and 1990, Columbus, Ohio (an increase of 39 to 191 square miles in the same period), Nashville (22 to 473 square miles), Indianapolis (55 to 362 square miles), and Rusk’s own Albuquerque (48 to 132 square miles). Examples of inelastic cities are Detroit, Cleveland, and, of course, New York City, whose city limits have remained virtually unchanged since 1950.

Elastic cities are “cities without suburbs” because a very large portion of their “suburban” neighborhoods are within city limits. Rusk presents convincing evidence that this produces a series of positive effects on the city as a whole. Because the tax base is broadened, elastic cities provide better services. Because the city limits are so wide, middle-class flight from the city is vastly reduced, which also reduces the disparity of incomes that characterizes many older, inelastic cities. An elastic city like Houston has an average per capita income almost equal—95 percent—to the average of its metropolitan area, while the average per capita income in an inelastic city like Cleveland is only 61 percent of its metropolitan area. The data that the author presents suggests that vigorous, elastic cities are better able to cope with the deindustrialization that has adversely affected almost all American cities, perhaps because new industry prefers peripheral land, which is located within city limits in elastic cities. In any case, with a more heterogeneous population and more available land, elastic cities attract growth, they create more jobs, and they even have higher bond ratings (an average of AA1, compared to A1 for inelastic cities). Finally, and not least, Rusk demonstrates that racial segregation is reduced in elastic cities since independent suburbs enforce exclusionary zoning, something that city neighborhoods cannot do. For example, the metropolitan areas of inelastic Detroit and elastic Houston contain about the same proportion of black citizens (21 percent in Detroit, 19 percent in Houston), but the city of Detroit is 75 percent black, while Houston is only 27 percent black.

The evidence that Rusk has marshaled here makes a clear and cogent case that the survival of many American cities depends on making city and suburb one. I am afraid that this common-sense argument will find many opponents. First, the well-to-do suburbs, which have seceded from the city and all its problems, will obviously prefer to continue their insulated existence. Can they be convinced that a broader, public interest would be served by annexation, or will they prefer to build walls—legally and, increasingly, physically—between themselves and their less fortunate neighbors? Opposition will also come from black big-city mayors and their African-American constituents, who would cease to be a majority if cities were to be integrated with their (largely white) suburbs. Rusk points out that black and Hispanic communities have waited a long time to achieve real power at city hall, although in a period of deficit reduction and federal spending cuts it is likely that this will be a hollow victory. There will also be opposition from many middle-class Americans who fervently believe that small government is better, more effective, and fairer than big government. For these people, it goes without saying that small towns are superior to big cities, and the suggestion that the latter should become bigger still will be difficult to accept.

Many intellectuals will resist Rusk’s implicit acceptance of the living arrangements that most Americans apparently prefer if they can afford it—a house in the suburbs. They would much prefer to ignore the existence of suburbs, and turn their attention to restoring downtown districts and rehabilitating inner-city neighborhoods. This has been the credo of urban advocates for the last thirty years, but the mounting evidence of failed projects suggests that as long as cities are concentrations of unemployment, poverty, and dependency, such rebuilding, necessary as it is, will be doomed to failure. “Ghettos can only become bigger ghettos,” writes Rusk.

Turning cities from clusters of ghettos into authentic, heterogeneous reflections of American society in which productive activities can have a place remains the greatest challenge of the next decade. There is little cause for optimism except, perhaps, the mutable nature of cities, themselves. “Cities are never still,” writes Kostof in the concluding pages of The City Assembled, “they resist efforts to make neat sense of them. We need to respect their rhythms and to recognize that the life of city form must lie loosely somewhere between total control and total freedom of action.” We need to remain open to the possibility that the form of the future American city could be an unexpected amalgam: of both downtown and suburb, of both Main Street and shopping mall.