Go back in history and take a look at the street pattern from subdivisions that sprouted across the purlieus of major metros in the 1950s and early 1960s—about the same time that the Federal Aid Highway Act of 1956 galvanized construction of limited-access expressways through the hearts of medium and large American cities. By the time President Eisenhower authorized the Federal Interstate Highway System, this segment of Levittown, PA (outside Philadelphia) had already enjoyed several consecutive years of astronomical growth:
By today’s standards, Levittown’s homes are modestly sized, the lots small enough to force the homes close together, and the garages/driveways barely accommodate two cars.
But they still represent a remarkable achievement in enabling homeownership to an emergent middle-class that had never enjoyed such a luxury in the past, particularly when the crippling Great Depression brought virtually all housing construction (and ensuing growth in independent households) to a screeching halt. Average people could afford these homes.
Now let’s do the time warp again thirty years more, by veering about 10 miles closer to central Philly. Witness the change in the layout of the streets:
While everything contained in this Google Map rests within the municipal boundaries of Philadelphia (unlike anything from the previous Levittown map), the most appropriate description of the neighborhoods outlined here is “transitional”. The street configurations to the west of the large, bisecting Pennypack Creek Park still mostly abide by the conventional street grid that Philadelphia and most American cities used as the basis for neighborhood design until around World War II. However, the east side of Pennypack begins to display a mix of conventional grids along with the same carefully ordered curvilinear streets that dominate Levittown further to the north. Housing developers after World War II, at the onset of the baby boom, began experimenting with street designs that boldly defied the four-way stop and quadrilateral. These new subdivisions meandered and undulated, continuing uninterrupted without any intersections for much longer intervals than any grid would allow. The William Levitts of the era gambled by speculating that Middle America would buy into sinuous streets that, though potentially more confusing and less suitable for navigation, helped break the visual monotony of a street that stretched to the horizon line. Curvilinear streets seemed to work well in the prestigious pioneering streetcar suburbs designed by Frederick Law Olmsted, such as Riverside, Illinois (outside Chicago) or Shaker Heights, Ohio (outside Cleveland). Why not export this typology to the middle classes?
The gamble paid off in spades, and first-time homeowners embraced the street design that all three of the major Levittowns employed. Before long, virtually every major metro witnessed the development of communities that followed the Levittown model, with most proliferating at about the same time that the central cities endured the cataclysm of interstate highway construction right through the central neighborhoods. Thus, it should come as no surprise that these 1950s suburbs nearly always stretch in close proximity to either their city’s major interstate highway, the circular beltways (with three-digit numbers instead of two) or both.
But this elegantly winding braid of streets still generally lacks something we associate with today’s suburbia: complete automobile dependency. It might not be easy or particularly desirable, but usually Levittowns are still tightly organized enough that it is possible to walk to a few destinations. Purely hierarchical street patterns—in which one “trunk” road provides access to “branches” and then still smaller sprouts—did not catch on until later, meaning we have to go back to the future (and much, much further out from the city center) to see the sort of development patterns that emerged in the late 1960s and 70s. The map of upper Bucks County (north of Levittown) proves this.
By the time the overwhelming majority of households had at least one car, it became de rigueur for developers to build according to these expansive street configurations, resulting in subdivisions that emphasized homeowners’ privacy at the expense of any real walkability. With homes spaced much further apart, yards were larger, and no amenities stood conveniently within walking distance, even back on the collector or arterial that provided access to the subdivisions. Nobody who bought into these developments gave any consideration that they would get anywhere except by car. Quite a few developments built during the 60s and 70s didn’t have sidewalks.
Housing from this time period isn’t as abundant as are the examples from the 1950s, because fewer households were organizing into families. A moderate baby bust followed the boom. Interestingly, home sizes grew even as the birth rate plunged; thus, individual households had more space to their own than any time in history—both inside and outside the house. Demand for shared green space reached a nadir. Public parks in these regions are particularly scarce, since private yards typically sufficed. The few parks in Upper Bucks County must devote a significant amount of space to parking lots, because they are inevitably unreachable by foot.
It was this time period that the cul-de-sac evolved from a suburban novelty to a sine qua non in subdivision development; families soon specifically sought them out. The early 1960s developments often didn’t even bother with the circular patch of pavement that allowed vehicles to turn around; the roads just terminated in dead ends, particularly when these new suburbs were pushing into unincorporated areas well beyond the inner-ring suburbs. Most municipalities eventually required roads to end in cul-de-sacs. While the curvilinear configuration of the 1950s remained popular, by this point many of the streets in a subdivision terminated in a court. The lack of a thruway discouraged all cars except for those belonging to people who lived on that cul-de-sac, dramatically lowering traffic volume, increasing privacy and giving the suburban subdivision a quiet, low-density settlement pattern that almost resembled rural living.
I’m not sure when the next subtle generation in subdivision design truly emerged, but it would probably correlate to the late 1970s and early 1980s, when the now-adult Baby Boomers began starting their own families—the echo boom. Household formation picked up pace, and these young families generally sought new construction just as their parents had. While some of the new developments took on an expansive form—particularly the luxury ones or those where land was cheap and abundant—quite a few homebuilders tightened the design. In many cases, average lot sizes retreated slightly, even while square footage to the homes continued a steady growth. Perhaps recognizing that they had renounced too many neighborhood essentials (or perhaps because more municipalities began mandating them), the developers of suburbs from the 1980s and onward regularly featured amenities such as storm sewers, curbs, and sidewalks (at least on one side of the street). As a compensation for the slightly smaller yards, these subdivisions (particularly the larger ones) would often include some shared open space in the form of greenery around a decorative retention pond, a community clubhouse, or a soccer field.
Why did this happen? Why did subdivisions return to a slightly higher, more urban density than the decades prior? My own suspicion is that some of it was prescriptive: through ordinances, municipalities started requiring culs-de-sac instead of dead ends, or storm sewers instead of drainage ditches, which drove up development costs. Also, formerly unincorporated lands began incorporating and immediately raising the minimum standards for subdivision design. Developers, in turn, couldn’t necessarily pass these imposed costs directly to the consumer, so instead they found a trade-off with smaller lot sizes, squeezing more homes into a newly platted piece of land. The result looks something like the quintessential 1980s/90s suburb below:
If this photo looks unusually flat for something in Pennsylvania, that’s because it isn’t Pennsylvania. The subdivision pictured above is in the outskirts of Baton Rouge, Louisiana. While such a shift in location may seem eccentric or even unexplainable, Baton Rouge offers much more abundant examples of the 1980s subdivisions, having grown considerably at a time when the more mature metro of Philadelphia was fairly stagnant.
This subdivision also demonstrates another, subtler evolution in street design: the 90-degree cul-de-sac. Notice the shift in the street from the photo below, as well as the parallel sidewalk.
The buffer strip between the sidewalk and the street seems a bit larger than normal, but other than that, nothing is likely to catch the eye as out of the ordinary. But then, upon entering the bend in the road, the sidewalk takes a more generous cut into people’s front yards, veering strangely close to their front porches.
Why would they have done this? It almost appears to me that authorities approved the site plan without thoroughly vetting it. While there’s nothing wrong per say about having a sidewalk so close to a house, it’s also hard to see why the average buyer would prefer it to a parcel where the sidewalk is much closer to the street.
Elsewhere, in a similar neighborhood down the road, one can witness another insertion of the cul-de-sac into a 90-degree-turn. Though the homes are more modest, the design of the sidewalk seems a bit more conventional, and, as a result, more effective.
But look at the grassy island in the middle. The lawn is poorly maintained, and the absence of any other landscaping makes it seem like an afterthought—like some sort of padding.
In neither of the two above examples is the execution as effective as it should be, which makes me suspect that the developers really didn’t know what they were doing. Or they didn’t care. It’s not a tough concept, though if these worm’s-eye photographs don’t convey it, this bird’s-eye Google Map should, taken from a subdivision of similar design in the same part of town.
Notice how the 90-degree bends stretch into curvy culs-de-sac? This road design is a shrewd method to make a little more money by cramming an additional parcel or two into the same space. If the road were to employ a conventional l-shaped bend, the lots that directly front the bend would be even more strictly wedge shaped, to the point that the street frontage wouldn’t be wide enough to allow individual driveways. Or, one house would claim an enormous side yard from the land that is completely unreachable by a driveway. With a gentle arc in its place, each house gets more or less the same linear footage as access to the street, and the surveyors who put together the original plat could fit in another parcel or two. And a more conscientious developer could transform grassy patch in the center of the cul-de-sac into an attractive, verdant amenity.
There’s nothing subversive or unethical about this design; it simply demonstrates that, particularly for moderate or middle-income households, developers have learned they can slice away at some of the yard size. The homes featured in the photos and map above all come from Gardere, a lower and moderate-income area on the otherwise affluent south side of Baton Rouge. My suspicion is that these subdivisions exist through some form of Low-Income Housing Tax Credits: they’re newer, in better condition, and bear some of the “neo-traditional” design features that federal housing programs love to employ. Because the average buyer for these properties doesn’t have the money to be choosy, developers can take huge liberties, and they don’t necessarily worry about details like attractively landscaped common area or well-designed sidewalks. And, in the grand evolutionary arc (pun fully intended) of American settlements, this represents one more design strategy pushing us further from the look and feel of the original Levittown. Suburban developments are neither uniform nor unchanging. I’m not confident that the conventional grid will ever become the standard again, even though New Urbanist advocates have successfully implemented it in specialized niche developments throughout the country. (After all, does anyone openly claim to be a New Urbanist anymore?) But I absolutely trust that culs-de-sac and street curves have a long way to grow before the design calcifies. And it probably never will. The evolution continues.