Homogeneity worries

click on images for full-size:

New York City at twilight

Different periods and styles on an Oxford street

Edinburgh's New Town

Edinburgh's Old Town

Door on a house in Lunenburg, Nova Scotia

Increasing the complexity of suburbia might bring so many links and allusions and differences that everything is present everywhere, so everywhere looks the same. This turns out to be a problem of competing scales.

It is not necessary for a place to possess a single unified architectural character, and one area may possess multiple and overlapping place grammars. But that is not to say that every place should resemble every other in a homogeneous mixture of allusions and links to everywhere else. In discussing the "mosaic of subcultures" in a city, Christopher Alexander worries about the way in which too much intermixing could reduce a city's built environment to a bland homogeneity. (Alexander 1977, 42-50) He urges that intermediate sized regions in the city need to feature different building styles, and different types of retail and restaurants. His model seems to be the old ethnic neighborhoods, though he is thinking more of life style than ethnic or class differences. He argues that such diversity benefits the city population as a whole, but it can be maintained only by a degree of residential specialization in local base populations supporting the mosaic's atmospheres and commercial areas (and keeping such areas from becoming only spectacles and themed places for visitors). Without some such specialization everyone's locations might look the same. (Kenneth Frampton's "critical regionalism" responds to similar worries. (See Frampton 1983, and 1995)

If large cities sometimes provide an interplay between specialized districts, suburbia tends toward a repetitious and large-scale homogeneous mixture of the same types (residential enclaves, scattered housing, malls, retail strips, scattered offices and factories, and so on).

Alexander is right that a city or a suburban region would benefit from areas of differing character. But he is also right that if we want variety at larger scales in the city or the sprawl, then we need some local homogeneity at smaller scales.

However, today's multiple populations will not be confined to "their own" areas in city or sprawl, and today's places overlap in non-centered and non-hierarchical ways. The desire for some architectural and spatial distinctiveness conflicts with today's less neatly compartmentalized populations and place grammars. (See Kolb 2007.)

Complexity and linking can help reduce this conflict. First, architecturally, there is no minimum size for links. An area of the city or suburbia could maintain an overall architectural character while being porous at smaller scales through a variety of links and signs of insertion into varied networks and different larger contexts. An overall architectural character need not correspond to one single overarching place grammar, or if it does, that grammar can be inflected by the small-scale presence of links and niches where other sets of norms and expectations show through. Second, there may be situations where architectural homogeneity is treasured precisely because of the differences among the people living there.