Smart growth supporters tend to prefer grid systems to cul-de-sacs, for excellent reasons. A proliferation of cul-de-sacs artificially lengthens walking distances: if streets don't connect to each other, you might have to walk a mile to go just a few hundred feet. In addition, cul-de-sacs increase traffic congestion by dumping most vehicular traffic on a few major streets. And because biking is less safe on busy, high-traffic streets, bikers benefit from a grid system as well.
But banning cul-de-sacs on residential streets might go too far: many homeowners understandably prefer cul-de-sacs because of the absence of "cut through traffic" on those streets (that is, traffic cutting through from one major street to another). Is there a middle ground between current subdivision ordinances (which sometimes require new subdivisions to be dominated by cul-de-sacs) and wiping out cul-de-sacs altogether?
One option might be for government to neither encourage nor discourage cul-de-sacs. It seems to me that this result would clearly be preferable to the status quo, as it maximizes consumer choice while reducing at least some of the social harms caused by cul-de-sacs.
On the other hand, one common justification for government regulation is to prevent situations where it makes sense for individuals to do X, but if lots of individuals do X, we all lose something. It could legitimately be argued that the prevalence of cul-de-sacs may be such a situation: if I live on the only cul-de-sac in the neighborhood, I have less traffic on my street (presumably a good thing) but still live in a basically walkable and uncongested neighborhood. But if everyone else lives on a cul-de-sacs, I have to suffer through all the disadvantages of cul-de-sacs as well: more traffic congestion because everyone has to drive on a couple of main streets, and reduced walkability as distances between houses multiply.
A second "middle ground" alternative is to allow cul-de-sacs in new subdivisions, but to create a quota limiting their number- for example, to provide that there be no more than one cul-de-sac for every intersection.This rule might accommodate consumer demand for cul-de-sacs, but would ensure that there were enough interconnected streets to accommodate driving, walking and biking.
A third compromise is the "fused grid." Under a fused grid street system, there is a grid of main streets and a set of cul-de-sacs branching off from those streets. But the difference between the fused grid and a cul-de-sac system is as follows: in the latter situation, there is nothing to connect one cul-de-sac to another, so walkers and bikers have to travel out of their way to reach other cul-de-sacs. By contrast, a fused grid "fuses" the cul-de-sacs with miniature parks or pathways designed for bicycles and pedestrians, thus allowing nondrivers to go from one house to another. When streets in an existing grid network are closed off to cars, something similar to a fused grid is created: cars are limited as in a cul-de-sac system, while pedestrians are as mobile as in a grid system.
Comments
Furlong Grid
I think another reasonable compromise is a grid with edges of an eighth-mile (a furlong)—not as tight as traditional grids, though rather walkable—and within each block the street layout could be some short cul-de-sacs (about 200 feet, longer if open-ended), a circle, a through street or two, anything reasonable as long as the furlong grid remains.
Fused Grid
Fused grid clearly sounds best: it has the main advantage of cul-de-sacs for residents (safety from through car traffic) and the advantage of the grid for pedestrians and bicyclists. Some of the fused cul-de-sacs should have traffic lights at the grid streets, so they make a good set of bike routes.
Incidentally, we have something like this in the Berkeley flatlands. We had an old-fashioned grid. But in the 1970s, the city decided to protect local streets from traffic, so it installed diverters almost all streets except the through streets. Now the local streets do not work for through car traffic but do work for through bicycle and pedestrian traffic.
Charles Siegel
One concern I have about fused grids
Suppose that a neighborhood has a fused grid system with small, walkable parks or pathways in back of the cul-de-sacs. These pathways will of course have no vehicular traffic- and I would guess that they would have a lot less pedestrian traffic than the grid streets leading to commercial streets. In short, these parks will have far fewer "eyes on the street" than the neighborhood's grid streets.
Given that likelihood, will these parks/pathways be as safe from crime as ordinary streets?
Pieces parts. Fused.
I'm not against a few culs-de-sac myself, as long as there is a realization that cut-thru traffic means more options for driving. I can't say I know the magic formula for a ratio, but there has to be added consideration for safety in a fused grid, having fewer auto options and resultant additional traffic on other streets (until gasoline gets high enough to negate this).
Best,
D
(the title is from an old, old Wendy's TV commercial - parts is parts. Only the older on this thread may get it).
Use The Berkeley Design For Fused Grids
I think that concern disappears with the design we have in Berkeley, where you have an ordinary grid of streets that are made into functional cul-de-sacs by diverters (that are just made of bollards).
A little park or pathway at the end of a cul-de-sac does sound unsafe. Instead, we have ordinary streets that function as cul-de-sacs, and that are as safe as any ordinary street with little traffic.
Charles Siegel
Paths connecting cul-de-sacs can be safe...
...if they're straight (so you can see from one end to the other), don't contain any landscaping large enough for someone to hide behind, and have some level of light from surrounding streetlights. If there's a bend in the path, it will be very unappealing to pedestrians and cyclists, particularly if the path is hemmed in by six-foot fences.
Greg Redeker