The issue is not so simple. For example, sprawl is often blamed for urban poverty, on the grounds that peripheral growth drains jobs from the inner city. Yet Anthony Downs, a Brookings Institution researcher and longtime critic of sprawl, has found no significant relationship between sprawl and urban decline. “This was very surprising to me,” he wrote, “and went against my belief that sprawl had contributed to concentrated poverty and therefore to urban decline.”

 

What about sprawl using up land? Most people would tell you that sprawl threatens farmland, but there is no evidence that a shortage of agricultural land is a serious national problem; in fact, during the last three decades of rampant suburbanization, food prices have dropped, not risen. Environmentalists make sprawl sound like a voracious monster. Yet America is not running out of land. One researcher has calculated that to house the entire population of the United States at a low suburban density of one family per acre would require an area smaller than the state of Oregon. Only about five percent of the United States’ landmass is currently urbanized, that is, occupied by buildings, roads, and parking lots, compared with 20 percent devoted to farming, and more than 30 percent covered by forest. The balance—almost half—is wilderness. Indeed, as unproductive farms have been abandoned and people have moved from rural to urban areas, wilderness has actually increased. “If preserving large ecosystems and wildlife habitat is your priority,” wrote John Tierney in The New York Times, “better to concentrate people in suburbs and exurbs rather than scatter them in the remote countryside.”

 

Perhaps one reason for the confusion about sprawl is that there is no widely agreed-upon definition. Some describe sprawl as a particular type of low-density growth, and others as a symptom of runaway development. And for some it is merely a temporary stage in the urbanization process. Late-19th-century photographs of upper Manhattan show brownstones and apartment houses surrounded by open space—which looks like the sort of scattered development commonly associated with sprawl, yet in relatively short order the empty spaces were filled in, and sprawl turned into city.

 

Most people think they know sprawl when they see it. But do they? Los Angeles is popularly considered an example of sprawl, yet the population density of its built-up metropolitan area is actually greater than that of metropolitan New York. Likewise contrary to popular belief, Los Angeles is not a city of freeways; it has the fewest miles of freeway per capita of any American urbanized area (which is why its freeways are so congested). The least dense metropolitan areas in the United States are not around the new cities of the South and the West but around older cities such as Detroit, Philadelphia, and Boston. Between 1982 and 1997, the urbanized areas of all three increased more than five times as quickly as their populations. This reduction in population density is chiefly the result of home rule. All three cities are bounded by small independent municipalities whose zoning restricts growth by requiring large lots, or by creating other obstacles to development. This, in turn, reduces density and pushes new construction farther and farther into previously rural areas.

 

The media commonly fuel misperceptions about sprawl. A 1995 cover story in Newsweek titled “Bye-Bye, Suburban Dream” described the growth of Phoenix in alarming terms: Between 1950 and 1994, the area within city limits increased 26-fold although the population grew only 10-fold. Obviously a case of sprawl—or is it? When a city expands by annexation it acquires empty land, as well as unbuildable areas such as wetlands and mountain slopes. If one counts only the parts of metropolitan Phoenix that were actually urbanized in the 15 years leading up to 1997, the area of metro Phoenix increased only half as quickly as its population; that is, metro Phoenix grew denser. Moreover, in 1997 the population density per urbanized square mile was greater than the metropolitan densities of Chicago, Boston, and Philadelphia.

 

Sprawl is often contrasted with dense downtowns, as if the choice were between living in a suburban rancher and an urban high-rise. However, according to the 1990 Census, the densities of American suburbs and cities are not vastly different: The average gross population density of suburbs then was 2,149 persons per square mile, and that of cities was 2,813. The explanation for this similarity is the nature of American housing stock. As one might expect, the majority of suburban dwellings—almost three-quarters—are one- and two-story buildings. However, considerably more than half of city dwellings are also one- and two-story buildings. In fact, only five percent of city dwellings nationwide are in buildings of seven stories or more.

 

If American suburbs and cities are more similar than different, why does the specter of sprawl loom so large in the public’s imagination? One reason is that sprawl is often equated with suburbanization. Virtually all postwar metropolitan growth in the United States has been suburban, but not all suburban growth, as Los Angeles and Phoenix demonstrate, is sprawl. As Downs points out, “Sprawl is not any form of suburban growth, but a particular form of it.” (He lists low densities, leapfrog development, and extreme political decentralization as some of the traits.)

 

Another reason that sprawl appears pervasive is that the effects of growth can be so visible. Since moving to Philadelphia, my wife and I sometimes drive through Bucks County to a large flea market near Lambertville, New Jersey. It’s as much a chance to get out in the country as to look at cracked teacups. Bucks County, roughly halfway between New York City and Philadelphia, used to be strictly a rural area; then it was a place for weekend retreats, and now city people are moving there permanently, drawn by good schools and relatively inexpensive housing. Over the last 10 years, the quiet country roads we take have become congested thoroughfares, and the picturesque fields have filled up with housing developments and discount malls. In fact, development in the county is generally concentrated and large parts of the countryside remain open, but that is not the view we have from the road.

 

A lot of the new houses in Bucks County are the work of K. Hovnanian Homes, a company that has built more than 150,000 homes across the United States since it was founded in 1959. According to president and CEO Ara K. Hovnanian, “The challenge for home builders is to try and figure out the type of housing that will be demanded by buyers, and where the demand will occur geographically. The good news is that, over the long term, the size of the actual demand for new homes is entirely predictable.” The predictability he describes is the result of three conditions. The first is population growth. Thanks largely to immigration, the U.S. population has been increasing every year by more than two million persons. These people need somewhere to live. The second is steadily increasing prosperity. As people become better off, they want newer, better-equipped, and larger homes. The third is mobility. New jobs don’t necessarily coincide with existing housing, and as people move—from cities to suburbs, from suburbs to rural areas, from one coast to the other—they, too, need places to live. As a result, every year, year in and year out, the American home-building industry produces between one and two million new homes, four out of five of which are single-family houses. Add to these new workplaces, new shopping places, new entertainment places, new schools, new hospitals, and new roads tying them all together, and you have a Monopoly game in full play.

 

It’s unsettling to live in a state of perpetual upheaval. That’s probably why sprawl has become a whipping boy for so many of the things we don’t like about modern life: traffic jams, overcrowding, instability, change itself. George Galster, an urban economist at Wayne State University, in Detroit, described sprawl as “the metaphor of choice for the shortcomings of the suburbs and the frustration of central cities . . . a conflation of ideology, experience, and effects.” I have a friend who has lived in Chester County, Pennsylvania, west of Philadelphia, for the last 50 years. He originally had an old house on a piece of land large enough so that he could shoot rabbits without disturbing his neighbors. Over the years, he has seen the surrounding horse farms gradually replaced by residential subdivisions. Naturally, he grumbles about the influx of newcomers, the increased traffic, the noise, the slow disappearance of his bucolic surroundings. More than a decade ago, he subdivided his 15 acres into three lots, sold two, and built himself a new house on the third. In other words, in a small way, he became a real estate developer. But if I were to call him that, he would be outraged—sprawl is always somebody else’s fault.

 

According to the Oxford English Dictionary, “sprawl” first appeared in print in 1955, in an article in the London Times that contained a disapproving reference to “great sprawl” at the city’s periphery. Lewis Mumford referred to “sprawling suburbia” in his 1961 classic The City in History. A 1965 article in Land Economics defined sprawl as “areas of essentially urban character at the urban fringe but which are scattered or strung-out, or surrounded by . . . underdeveloped sites or agricultural uses.” At that time, a more neutral term, “scatteration,” was also used to describe this phenomenon. Thanks to a famous 1974 study titled The Costs of Sprawl, which computed the direct costs and adverse environmental effects of low-density development, “sprawl” entered the planning lexicon. The methodology of the study was later called into question, but the term stuck. There is no better way to occupy the high ground in a debate than to define its language.

 

The Costs of Sprawl study was prompted by the fact that in 1970, for the first time, more Americans lived in suburbs than in rural areas or cities. The authors of the study predicted that suburbanization between 1970 and 2000 would be almost as great as in the previous 20 years, which had been “the period of greatest suburban growth in the nation’s history.” They underestimated on two counts. Suburban growth was not 70 percent, as expected, but 80 percent, and the overall population grew not by 46 million but by 76 million. As a result, the increase in the number of people living in the suburbs turned out to be almost twice as great as predicted. The United States had become, in the words of one commentator, a “nation of suburbs.”

 

When railroads and streetcars opened up the urban periphery in the 19th century, only the well-off could afford to commute, whether it was from Chestnut Hill to Center City Philadelphia, from Brookline to downtown Boston, from Lake Forest to Chicago’s Loop, or from Tuxedo Park to Manhattan. That might have remained the pattern—a select number of wealthy garden suburbs on the distant fringes of dense, blue-collar, industrial cities—but for Henry Ford. Inexpensive automobiles gave mobility to everyone.

 

John Nolen, who was a student of Frederick Law Olmsted Jr. and one of the most prolific American planners of the early 20th century, predicted the revolutionary impact that cars would have on urbanization. In 1927, in New Towns for Old, he wrote, “If the movement away from the cities assumes the formidable aspect of a hegira (and the magnitude of recent modern developments like the automobile and the radio makes this appear quite likely), then it is immensely important that it be organized and directed accordingly.”

 

Nolen’s solution to suburban growth was to channel it into planned garden suburbs, among them his exquisitely planned model town of Mariemont, outside Cincinnati. He believed in design, but unlike most city planners today, he was not wedded to high-density development. He agreed with his friend Raymond Unwin, who once wrote a pamphlet called Nothing Gained by Overcrowding! Nolen and Unwin decried the congested tenements and walkups of the old industrial cities. They wanted everyone—not just the rich—to have their own homes, their own gardens, and access to nearby parks and playgrounds. Garden suburbs delivered on that promise. Nolen and Unwin’s suburban strategy still appears sound. As Gregg Easterbrook wrote in The New Republic in 1999, “If suburbs are where Americans choose to live—and that verdict is in, the suburban class now constituting the majority of Americans—then brainpower should be applied to making ’burbs as livable as possible.”

 

One of the planning ideas advanced as an antidote to scattered development is so-called smart growth, which originated in the 1990s. Smart growth, like sprawl, is a slippery concept, not least because it is espoused by anti-growth environmentalists as well as pro-growth developers. In a 2001 article in Planning magazine, Anthony Downs wrote that advocates of smart growth do have some things in common. They are for walkable communities and mixed-use town centers, and generally favor preserving open space and redeveloping inner cities. However, depending on who is speaking, smart growth can also include such controversial ideas as subsidizing mass transit to reduce car dependency, creating regional governments, and establishing urban growth boundaries to restrict growth into rural areas. While environmentalists see smart growth as a way of placing limits on growth, developers would like to change zoning to permit higher densities, and land conservationists would like to restrict development to selected areas. Downs concluded that, as a national strategy, smart growth is simply too contradictory to be effective, and he argued for elements of smart growth to be applied selectively at the regional level. As he succinctly put it, “What is ‘smart’ in New York City may be ‘dumb’ in Phoenix.”

 

The battle over sprawl and smart growth usually comes to the fore when a community is faced with new development. For the last five years I’ve been following the creation of a small subdivision called New Daleville, in Chester County. It is an example of New Urbanism, also loosely referred to as neotraditional planning, an idea that has gained currency among some developers and planners. In brief, this approach aims to build walkable, compact communities, with smaller lots and higher densities than conventional subdivisions. There is more emphasis on common areas, such as parks and playgrounds, and because the houses are bunched close together, these communities sometimes resemble old-fashioned villages—hence, the neotraditional label. New Urbanism hardly dominates construction on the suburban fringe, but it’s yet another factor that confounds the stereotypes. Neotraditional development appeared at New Daleville because the community had been resisting a conventional half-acre-lot proposal, and decided it wanted to try something different. The new plan involves more houses on smaller lots, and sets aside half of the 90-acre site for a township park.

 

How smart is New Daleville? If sprawl is measured in consumption of land, the fact that there are more lots on less space appears to limit sprawl. Compared to the 86 houses that were originally planned for this site, New Daleville will have 125, an increase of almost 50 percent. However, since the lots at New Daleville will be smaller, it is likely that the houses will appeal to smaller families and empty-nesters. If the average family size in New Daleville is three rather than four, the total population will be 375 persons versus 344. Still an increase, but nowhere near as dramatic.

 

On the other hand, if sprawl is defined as building over farmland, then New Daleville will contribute to sprawl. Since the township has no real master plan, merely a collection of zoning districts, any development, however well designed, will remain an isolated residential island. Although the New Daleville planner has designed a walkable community, there will not really be anywhere to walk to, since the place will be too small to support a village center. Since the density of the township will always be too low for mass transit, the future inhabitants of New Daleville will be heavily dependent on their cars. Their comings and goings will add to the traffic and congestion of the back roads of Chester County. Thus, for hard-core, transit-first, rebuild-the-center-city, regional planning advocates of smart growth, New Daleville is merely more of the same, what they don’t want.

 

Yet New Daleville’s compact layout will likely foster a greater sense of community than if the houses were spread out. Children will play in the parks—and probably in the back lanes. People will more easily meet their neighbors. They may even organize public events on the common green. With its compact plan, New Daleville will be a nice place to walk—for exercise and for pleasure. The narrower streets and denser layout will reduce the amount of asphalt. Hence, there will be less polluted runoff; more rainfall will be absorbed into the ground naturally. Half of the site will be left unbuilt in perpetuity—no small accomplishment. Kids will be able to walk or bicycle to the playing fields. Above all, New Daleville, unlike other subdivisions in the area, will include shared public spaces: sidewalks, walking trails, play lots, village greens, parks. These will be small reminders to the people living there that they are not only private homeowners but also members of a community. That will be smarter growth indeed.