"How Buildings Learn: what happens after they're built"
by Stewart Brand
Link: http://www.amazon.co.uk/How-Buildings-Learn-Happens-Theyre/dp/0140139966/
Reading history and reviews
Finished on 15th June 2009
This is a book packed with fascinating observations. The core idea is that the life of a building doesn't end when it's first built, instead it's only beginning. While this idea might initially feel counterintuative - we think of buildings as being the very essence of unchanging permanance - Brand asserts that in fact the "best" buildings evolve, sometimes radically, in response to the changing needs of their inhabitants. "Best" in this context means those that are most able to provide a suitable environment for the activities that take place within them - whether that's a home, office or manufacturing or retail environment. Buildings that can't be adapted can be expensive to maintain and modify, and are likely to fall into disrepair, or else be demolished.
Following on from this idea, the book contains an exploration of the various ways that buildings evolve (or don't). It's clear that various factors militate against this evolution - most buildings don't have adaptability or maintainability "designed in" at the start, but there are also financial,legal and planning constraints that prevent inhabitants from making changes. Brand berates "magazine architecture" (the tendency for architects to focus on the look of a building at the expense of how it might be used or adapted in future) as a factor in producing unworkable designs, but also highlights how a lack of accurate documentation (particularly during the building phase) makes changes more difficult and costly.
Once we recognise that change is a vital part of a building's lifecycle, Brand asks how can adaptibility be designed into new buildings, and what can be learned from what's gone before. "Successful" buildings use tried and tested materials, and designs which can accommodate a range of possible usage scenarios. Extensive documentation of both the design decisions and of what is actually built (which can differ from the plans) is a significant factor in enabling changes to a building in the future. Architects should return to buildings that they've designed to see if there are lessons that they can take forward in future.
By the end the book had changed my perspective on architecture and how I look at buildings around me. A number of the themes are visited again in Brand's The Clock of the Long Now, and on a personal note with both books I felt that there were also some striking parallels with software development and maintenance (in particular, could a scenario-planning approach could be applied to improve software adaptability?). A thoroughly thought-provoking book and highly recommended.