Portland’s got good bones. No doubt about that. You’ll often hear me diss Portland’s planning and zoning policy at every level from the neighborhoods to the MPO. What you won’t hear me diss are the guys who built the damn place 100 years ago. Frankly, I kinda think of the planners as interlopers, taking a lot of credit for shit that was mostly there before they showed up. Human scale! Street trees!
So it’s interesting to read Fanis Grammenos and Douglas Pollard’s takedown of Portland’s “hippodamian grid.” It feels kinda like a straw man argument, because they’ve got this nice picture comparing Portland to a pre-grid city and a post-grid city:
Except, Portland doesn’t look like that. Portland looks like this:
But hey, let’s give these guys the benefit of the doubt for a moment and assume they’re talking about Downtown, which is pretty regular. That nice graphic shows a regular grid fifteen blocks on each side. But if you start walking west from the riverfront, you can scarcely to eight blocks before you run into the Park Blocks. The Park Blocks not only break up the regularity of the grid, they also provide a visual block which reduces the depth of field; the trees come out over the street, so if you’re looking down any east-west street, rather than staring off into endless space, you’re looking at a street that ends in a wall of greenery.
North-South travel is likewise broken up by Ankeny, which is the boundary between the original Overton/Lovejoy plat and Couch’s various additions. Lots of cool in that caddywhompus part of the grid. Ankeny was where I once ate vagina-shaped donuts with crunked daredevil cyclists at three in the morning. The Church of Elvis was on Ankeny. The Ohm was there. And I’m pretty sure I saw a couple burlesque shows on Ankeny, although I can’t find the club on Streetview. These sorts of little narrow out-of-the-way streets add interest, provided you don’t try to construct an entire city out of them.
Where Downtown has a few features that break up what could theoretically be a boring and monotonous grid, the rest of Portland has many, many more. The residential sections of Portland are cut by a thousand different Ankenies, which help define neighborhoods and break up traffic. For instance, here’s a chunk of the Hawthorne District, just east of Ladd’s Addition:
At a five second glance, that’s a grid. But look closer, and what you’ll soon see is almost no street goes through. Everything is misaligned. Try to use almost any street here as a “rat run”, and you’ll be forced into a bunch of street jogs, successive left-right and/or right-left turns. To a pedestrian or a cyclist, this isn’t a huge deal. To a motorist, it’s a deal breaker; he’ll probably just stick to the arterials.
To illustrate:
Note how almost all the streets don’t go through. Most of those that do end at the adjacent arterials. Only two north-south streets in this whole section of town connect to anything. The lone east-west through street was cobbled together from several other streets long after the original plat; it doesn’t even keep the same name, switching from Harrison to Lincoln as it zigs across 30th.
What I find more interesting is that this fine-grained, built-in resistance to through traffic has almost always been universally frowned upon by planners. Planning orthodoxy from 100 years ago, when this stuff was being laid out, held that everything should be regular, grids should match up with each other, like the Eixample. The only force favoring disaggregated, irregular grids like this was those damned developers’ profit motive – maximizing lot count within a given tract, and providing lots of various depths and sizes based on their perception of demand.
Later, after the 30′s had brought federally-subsidized mortgage insurance, and with it the stipulation that all neighborhoods should be laid out with Greenbelt-style curvilinear streets, the Traffic Engineers gave us the idea of minimum intersection spacing, which is still with us, enthroned in countless statutes and codes, including Houston’s.
Trying to reconcile these conflicting goals (“connect all streets perfectly” and “discourage through traffic”) results in the Calgary subdivision layout. But while the the Calgary layout can be made transit-friendly with sufficient midblock ped cut-throughs, it can never be as multi-modal friendly as the porous, easily traversable grids of Portland (and the Montrose, for that matter).
Street jogs, however, solve all these problems simultaneously. They discourage through automobile traffic while encouraging through bicycle and pedestrian traffic. They reduce the visual depth of field without turning the neighborhood into a maze of cul-de-sacs. They are an elegant way to design neighborhoods. And because they were born not out of any sort of planning ideal, but as an outgrowth of unplanned land speculation, they are now banned almost everywhere.
Bring back the jog.
keephoustonhouston
Excellent diagram there. Up until the 1930s-1940s, there was widespread awareness among developers and planners about this. They knew from experience that jogs, bends, and T-intersections reduced traffic speed and volume. The pattern also provided a high degree of connectivity that allowed convenient intra-neighborhood travel and reduced the load on arterials.
Camillo Sitte promoted the concept and Le Corbusier disparaged it as “the pack-donkey way.” All that wisdom of experience was tossed aside when automobile-adoring planners and engineers decided that because cars were capable of traveling 100 mph, cities had to be redesigned to handle those speeds. Arrow-straight expressways to the horizon! At the same time, articles started appearing that complained about how boring and oppressive the grid pattern was. The Grammenos and Pollard article shows that complaint still exerts a powerful influence 80 years later.
Fascinating breakdown of my town. As a bicyclist, I would say that the broken street network does cause problems when we won’t put bike lanes on major through streets. Shunting cyclists off to side roads that don’t connect and don’t connect with shopping, for example, is Jim Crow planning–separate and unequal. While planners should get their share of the blame (mostly for promoting monocultures of uses that isolates housing from jobs/shopping), i reserve my ire for the engineers who see the streets as sewers that should be designed to move “stuff” through fast. Hence, pedestrians are “friction”, little streets should flow into ever bigger streets, speed is king and people are problems not the purpose of cities.
As someone who lived a ways south of the pictured area of SE Portland, I’ll disagree with “To a pedestrian or a cyclist, this isn’t a huge deal” about the lack of through streets. Since there are so few through north/south streets, there’s a _lot_ of car traffic concentrated on them, and they mostly don’t have dedicated bike lanes. So, for a cyclist, if you want to traverse that chunk of Portland north/south, there are very few choices between 16th and the very jog-heavy 42nd-43rd-41st bike route that don’t either put you in traffic (often without much passing room) or have you stopping every two blocks for a stop sign or a turn.
Don’t get me wrong, Portland’s the most bike-friendly metropolitan area I’ve ever been to and I love it for that, but the example you picked shouldn’t serve as a role model for a transportation network that’s serious about cycling.
I think Laurence is right that a lot of this stuff was once sort of intuitive knowledge of the building/construction industry that had been passed down for generations, but got cut off with the nationalization of planning and subdivision standards in the 30′s and 40′s.
As far as joggy bike routes go, when I lived in Portland and rode a bike I just blew those stop signs. And my 21-speed could cut a nicely arced S-curve through any of those intersections. I can see how this pattern would be annoying if you’re a cyclist who actually comes to a full stop at every red octagon. However if you’re really that masochistic, I would suggest finding other outlets for it (Portland has a pretty substantial BDSM scene), and shaving minutes off your commute by treating all stops as yields.
Incidentally, I’d consider 39th a fault in the original layout. If a street is to be the only continuous north-south street for two miles in either direction, it needs more than a 60′ right-of-way. I’d say 100′ is about the lower bound, really. But can you imagine the NIMBY outcry if Portland was to go in and propose a widening, even if it was entirely for bike lanes/wider sidewalks/medians/street trees/other wholesomeness? The lawsuits alone would tie up the project for ten years, easy.
I’m not sure why the “no through streets” thing is a deal-breaker to someone in a car but not to someone on a bike or on foot, and I rather think your whole argument hangs on that point.
Can you elaborate a little why you believe there is a difference? What negative effects will I suffer in a car that I’m spared from in another mode?
@Daniel
Basically it has to do with varying expectations of speed. A pedestrian walks 3-5mph, a bike commuter rides 15-20mph, and a car drives 30-40mph. Cars want to go straight and fast, and have limited maneuverability. Peds on the other hand have no turning radius.
For a car cutting through a residential neighborhood, a street jog forces a 30-0mph decel, a turn, another turn, and a 0-30mph accel. Lots of effort, gas, and annoyance. For a ped walking through the same neighborhood, the same jog marginally increases the distance traveled, but there’s no reduction in speed or comfort. For a cyclist it’s somewhat in between.
The point is to *discourage* through traffic (for neighborhood tranquility) while simultaneously *allowing* it (for neighborhood connectivity, transit accessibility, etc). The disjointed grid and its street jogs work on a sliding scale, providing progressively more discouragement for progressively faster modes.
Pingback: The Reddit Effect | Keep Houston Houston.