March 06, 2003

A Pattern Language: Independent Regions

There's been a fair amount of talk on the web lately about the book A Pattern Language, by Christopher Alexander et al. Although it's a book on architecture and urban planning, I first heard of it in the context of software engineering. Alexander uses the term "pattern" to mean an archetypal solution to a certain kind of problem--a solution you can apply over and over again in different situations which nevertheless match the pattern. For a while, patterns were the next silver bullet in software engineering; that initial glow has faded, but both the idea of patterns and a number of specific patterns have taken firm root in the software community.

But I'm not intending to talk about patterns in software; instead, I want to talk about Alexander's book. It's been recommended to me by several people, some on line, and some I've actually met; the idea of patterns is appealing to me; and the recent discussions piqued my curiousity to the extent that I actually bought a copy. But the thing is, I find I can't review it the way I ordinarily would. It's a big, thick book, chock-full of ideas, all of them cross-referenced to other ideas. It's not the sort of thing you read cover to cover; it's the sort of thing you browse. And any short review I might right will utterly fail to do it justice. So I've decided to embark on a rather more ambitious plan.

I'm going to browse in it, and read it, and browse some more, and every so often I'm going to write about one single pattern. That's an idea that's completely incompatible with Alexander's goals, by the way; the patterns aren't intended to stand alone. But it's the only way I'll be able to present my thoughts.

The book is divided into three sections, called "Towns", "Buildings", and "Construction". The first section contains patterns on how to lay out towns and the things in them so as to make them delightful places to live. And the first of those is called Independent Regions (1). (That "1" is the pattern number; by convention that's always included in the name, to help people look them up.)

In Alexander's view, the world should be divided into a thousand or more independent regions, each with its own local government, and each part of a world federation. (What this has to do with architecture, I'm not sure.) Surprisingly, he doesn't suggest this as a means of centralizing urban planning within each region; Alexander doesn't believe in central planning. Within the region, each city and town is responsible for its own land; and within each city and town, each group or individual is similarly responsible for its own territory. If all of them conscientiously rely on Alexander's patterns as they do their planning, the world will be a beautiful place.

Instead, he lists several other reasons: government becomes unwieldly at any other size; the current trend toward globalization is homogenizing cultures the world over, whereas his notion would preserve them; those times and places where the basic political unit has been the city-state have been seen an outpouring of art and architecture.

What he seems to forget is that those same times and places also saw frequent outpourings of blood. The Italian Renaissance produced Leonardo, Michelangelo, Raphael, Donatello, and scads of other noteworthy turtles--ah, artists--but it was also the golden age of the European Mercenary Company, and the city-states of Italy were constantly at war. The great ruling families were patrons of the arts, it's true, but there's a reason why the names Borgia and Medici have an ominous sound in our ears.

I suppose he thinks this "world federation" will somehow manage to keep the peace among the thousand regions. But just because fifty states can be stable, it doesn't follow that 1000 states will also be stable. With fifty states, every state is important; even little New Hampshire is the star of every presidential campaign. With 1000 states, no individual state is important enough to make its voice heard in the assembly. I can't see that such a body can keep order without instituting such tight control over the member regions as to destroy the indepence that's their reason for being.

And then, I think that there's another reason the United States has gotten along so well (mostly) for over two centuries--they began with a relatively homogeneous political, legal, and moral culture rooted in the rights of Englishmen and the English common law. It's true that America is a melting pot, and I've no wish to disparage the contributions of any of the other groups involved. But there's a distinct different between the American Bill of Rights and the French Declaration of the Rights of Man--the Bill of Rights is a record of rights the founding fathers felt that they already had, even if they sometimes had them only in the breach. They wrote the Bill of Rights to protect those rights, and to make sure that no one could take them away in the future. They'd gone to war with England because England hadn't respected their rights as Englishmen. The Declaration of the Rights of Man, by contrast, is a record of rights most Frenchmen had never enjoyed up until that time, the rights the leaders of the Revolution felt that they should have--and given that the French have since had two emperors, a number of kings, and (if I recall correctly) five or six republics, not to mention two German invasions, it's not clear how often they've enjoyed them since.

Thus, I think the diversity of cultures that Alexander wishes to protect with his thousand independent regions would instead prevent his "world federation" from keeping the peace--and his plan would degenerate into warfare and bloodshed until the regions would be forced to join into larger countries for their own safety. Just as they did historically.

So OK, Alexander's a utopian dreamer. I suppose that shouldn't surprise me. But it does bug me a little that this is one of the patterns he and his co-authors marked with two asterisks, "**", indicating that it presents a solution that they are absolutely, positively, 100% sure of despite being architects rather than political theorists or historians. But I gather that no one has ever accused Alexander of modesty.

Posted by Will Duquette at March 6, 2003 07:43 PM