Monday, April 4, 2011

A (Defense!) of the Grid: Portland and 19th Century American City Grids

This blog has not been kind to the grid in the past, but when it is criticized unfairly, it deserves a defense.  In a lengthy critique of Portland's street grid, planner Fanis Grammenos argues that the grid makes inefficient use of valuable urban land:
"The Portland grid uses 42% of land in right of ways for streets and has the highest length of road infrastructure of any alternatives. Simply put, nearly half of the land is used up in accessing the other half. ... Evidently, Portland’s founders either understood little about infrastructure costs or judged them irrelevant; a judgment that no planner, developer or municipality today would take at face value. When economic efficiency matters, Portland’s grid fails the grade."
Writing in response, Jarrett Walker of Human Transit perceptively notes that the amount of land devoted to streets is a function of the width of the streets rather than the grid plan itself.  Portland, like most other American cities founded in the 19th century, was laid out on a regular grid with streets of generally uniform width.  In Portland's case, this meant 200-square foot blocks separated on all sides by a 60 foot right-of way.  Although sixty feet may not even seem especially wide in comparison to some other cities, its constant repetition results in 41 percent of all downtown land devoted to streets and sidewalks (27,600 sq. ft./67,600 sq. ft.), virtually the same as Grammenos' figure above. 

If the founders of Portland could dodge the infrastructural expenses imposed by roadways of such width by simply leaving the streets filthy, unpaved and sometimes dangerous,* their successors in the early automobile age were not so fortunate.  In a 1919 article "City Planning For Portland," it is evident that the costs by that time have become too great to ignore:
"Whereas city engineers used to think that every street should be laid out as a through street, it is now found that we can only afford to provide the wider and heavier pavement on about every fifth or sixth street, called a main thoroughfare or major street.  ... Considered from this new point of view, some municipalities have begun to realize that a lot of money put into paving which is now unnecessary could have been used to splendid advantage in other ways. ... [I]t becomes evident that 200-foot blocks, particularly in outlying districts, are an extravagance ...."
Among other things, the planners of 1919 recommended much narrower streets and longer blocks for future residential areas, the identification and selection of a network of through streets, while also mentioning a "necessity for diagonal routes ... in many parts of the city."  A narrowing of certain existing rights-of-way, which would have gone a long way to addressing the planners' primary concerns, was not advised, although it apparently was within the realm of imagination.

Portland's case is not unique, but rather is representative of hundreds of American cities and towns of the 19th century.  The seemingly irrational decision to give all city streets an extravagant scale was replicated in plan after plan, even in rural hamlets.  Even if an emphasis on facilitating transportation ("wide enough for an ox-cart to make a U-turn") may have played a role, any such benefits were more than canceled out by the city's inability to pave the wide streets, leaving them a muddy quagmire in many cases well into the 20th century.  The vast amount of land set aside for streets also diminished urban density, leading to, on average, greater travel distances than in a city set along narrower ways.

It would be a mistake to fault the grid plan itself for this, however, as a grid may with only slightly more thought given to the initial plan, and less of an obsessive focus on perfect regularity, incorporate a range of street widths and block dimensions (Manhattan's grid being probably the best known example).  This will not cure it of all of its deficiencies, but it certainly will address Grammenos' criticism by helping the grid achieve a more favorable ratio of buildable to non-buildable land.

*From Wikipedia: ""[Portland in the mid-to-late 19th century] was a place where “stumps from fallen firs lay scattered dangerously about Front and First Streets … humans and animals, carts and wagons slogged through a sludge of mud and water … sidewalks often disappeared during spring floods.” ...  In 1889, The Oregonian called Portland "the most filthy city in the Northern States", due to the unsanitary sewers and gutters. The West Shore reported "The new sidewalks put down this year are a disgrace to a Russian village.""


  1. This is fantastic. Wonderful historical background on the basic design disasters of the 19th century.

    Yes, a grid can work when the streets are pedestrian-sized, around 10-15 feet from building to building. However, it is not at all necessary to use a grid, and it doesn't seem to confer many advantages. In fact, many suburban developments of the last thirty years have a lot of swoops and curves.

    We have spent so much time repllicating the grid that it is a good time for other experiments. However, whether a grid or some more biomorphic layout, in the end, it doesn't matter nearly as much as the width of the street itself.

    Link to pictures of the city of Seijo, west of Tokyo. This is basically a grid layout, but the streets are about 12 feet wide.

    Nathan Lewis

  2. Wider streets and lower density are features, not bugs. They make the heart of Manhattan more expansive, private, well-aired, gracious - and in summer, livable - than the old sections of London or Amsterdam.

    Modern folk will order by phone/internet, hail a cab, or use public transport for tasks and distances that our grandparents covered on foot. And they don't want their windows nose-to-nose with their neighbors.

  3. If you read McColl on early Portland you find that the fellows who laid out the streets had an interest in maximizing the number of expensive corner lots they could sell.

  4. The corner lot theory sounds plausible. And in fact there's nothing wrong with short blocks -- Jane Jacobs favored them for their friendliness to pedestrians. When paired with very wide streets, however, that advantage is largely negated. Nathan's link shows basically the same size block with streets 1/5 as wide as those in Portland, which makes for an 89 percent overall building coverage as compared to 59 in Portland.

  5. "Wider streets and lower density are features, not bugs. They make the heart of Manhattan more expansive, private, well-aired, gracious - and in summer, livable"

    Are you joking? You can say many good things about Manhattan, but "expansive, private, well-aired, gracious, and in the summer, livable" wouldn't be high on the list. Manhattan, I would say, is the exact opposite of "gracious." Either this is a person not from NYC who is just making stuff up, or someone in NYC who is delusional.

  6. First off, love the blog. Your thoughts bring a wonderful sense of humanity back to the often over-technical discipline of urban planning.

    However, wide blocks or not, it is what we have. It would be great if our streets were a bit smaller but they aren't and there is realistically no way to start shrinking block size now.

    To add a bit of reality here, perhaps you can look into examples of innovative ways of creating more pedestrian scaled streets and spaces. One example would be the promotion of alleys as commercial streets. Melbourne has done some great work here. Seattle's Post Alley is also wonderful.

    There are also possiblities of reinventing wide streets as front yards for gardening, bike boulevards, or more spacious outdoor seating.

    The United States simple isn't Europe and our cities will never look like theirs. While its great to compare and contrast the differences it would be helpful to also look to a hopeful future as well.