After the new year, I will again be teaching the course I call Place and Power in the City, and one of the things I am thinking about as I start to get ready for it is the way that built environment creates, in the most literal way, “facts on the ground.” That is, whatever we think of specific elements of that environment— individual buildings, the locations of roads or transit stations or crosswalks, and so on— or the processes by which they were designed and constructed, once they are in place we are more or less forced to work within them, or around them. The routes we take, the time we dedicate to commuting, where we live and where we work will be determined, or at least influenced, by the affordances of a particular constellation of architecture, infrastructure, and regulation.
Once built, too, all of this is relatively hard to change. A building or a road that turns out to have been inconveniently placed can usually be relocated only with difficulty and expense— so much so that we often simply find ways to live with such inconveniences.
The costs of change can be relatively easy to estimate, but the costs of not changing are often less visible, since they necessarily rely on guesses about what would happen if things were different. Even when the status quo is clearly bad, the concrete costs of unbuilding can seem overwhelming. “The World Needs to Quit Coal. Why Is It So Hard?” by Somini Sengupta showcases this problem. The problems with coal are clear: it is the most polluting fossil fuel source, and its use is pretty obviously a major driver of climate change. But it’s also cheap and abundant, including in developing countries where resources are scarce but demand for energy is high. At the same time, the infrastructure— mines, power plants, even railroad cars designed to carry it— is already in place.
“The main reason why coal sticks around is, we built it already,” said Rohit Chandra, who earned a doctoral degree in energy policy at Harvard, specializing in coal in India.
This path dependence— a situation where the longer one way of doing things continues, the more expensive changing it becomes— characterizes a lot of what happens in cities. The most obvious example of this is in transportation, where cities built (or extensively rebuilt in the 1950s and 60s) around cars find the costs of building mass transit prohibitive, or at least hard to support politically. The roads, in other words, are already there, so it seems easier and cheaper to stick with them rather than tear everything up to put in train tracks, even if this is actually a much costlier option in the long term, and even if the roads are in poor repair or inadequate for the traffic they now have to carry.
More generally, change is a problem for complex systems. If they can’t change, they can’t adapt to new conditions, and they eventually collapse. At the same time, complexity makes change hard, since any change in one part causes a cascade of knock-on effects throughout the system.
“The Upgrade”, by Atul Gawande, illustrates this clearly. He’s writing about the “computerization” of medical records, which seems like a good and even necessary step in the abstract, but creates many new problems in practice. First and foremost, it increases the amount of time that doctors spend on documentation, and that means less time talking with patients— and higher rates of burnout among physicians. Gawande frames the problem as one of balancing the need for innovation with the need for ways of assessing the value of those innovations:
Medicine is a complex adaptive system: it is made up of many interconnected, multilayered parts, and it is meant to evolve with time and changing conditions. Software is not. It is complex, but it does not adapt. That is the heart of the problem for its users, us humans.
Adaptation requires two things: mutation and selection. Mutation produces variety and deviation; selection kills off the least functional mutations. Our old, craft-based, pre-computer system of professional practice—in medicine and in other fields—was all mutation and no selection. There was plenty of room for individuals to do things differently from the norm; everyone could be an innovator. But there was no real mechanism for weeding out bad ideas or practices.
Computerization, by contrast, is all selection and no mutation. Leaders install a monolith, and the smallest changes require a committee decision, plus weeks of testing and debugging to make sure that fixing the daylight-saving-time problem, say, doesn’t wreck some other, distant part of the system.
One problem is that the new medical record systems must try to be universal, to be usable by all kinds of doctors meeting all kinds of patients, ordering all kinds of tests and procedures and recording all kinds of results. The possibly inevitable result of that is a compromise that’s ideal for nobody; every user is forced to alter, in one way or another, the way they do things to fit into the system’s design. Changing anything to suit a particular group of users is difficult, both because of the complexity of the system and because a change that improves things for one might make things worse for another.
A consequence of that is that such systems are always breaking down, in ways large and small, and require constant support to keep them going. “Maintenance and Care”, by Shannon Mattern, argues that the way we maintain and repair such systems deserves as much attention as the systems themselves.
What we really need to study is how the world gets put back together. I’m not talking about the election of new officials or the release of new technologies, but rather the everyday work of maintenance, caretaking, and repair.
We pay absurdly little attention to maintenance in general— a problem discussed in an older Atul Gawande piece, “The Heroism of Incremental Care”. Building something new is more exciting, more visible, and easier to get credit for, but keeping things working once they are built is essential to their function.
Mattern identifies a number of kinds of work and effort that could be seen as kinds of maintenance in one sense or another— not only the obvious repair of roads or bridges, but childcare, cleaning, and content moderation on social media. This kind of labor, as much as invention, produces the world we live in; it is one thing for an object or system or institution to exist, but how well it works will also determine how we use it, and indeed whether we use it or avoid it. Maintenance is often the difference between a resource and an obstacle. We are dependent not only what there is, but how well it is maintained.
Interestingly, like Gawande, Mattern also relates this to adaptation, suggesting that repair is also a source of innovation:
…Stephen Graham and Nigel Thrift identify breakdown and failure as “the means by which societies learn to reproduce,” because the repair of broken systems always involves elements of “adaptation and improvisation.”
That is, systems adapt not only because people try things that are new altogether, but because people find ways of fixing, retrofitting, or stabilizing what has stopped working. Problems in maintenance and repair can also influence the design of new things. From this perspective, design and repair should be seen as steps or phases of the same, continuous process.
This also provides a new angle from which to look at the idea of path dependence, as I was talking about it earlier. One aspect of the built environment is the problems of, and opportunities for, repair and maintenance that it presents. To say that we are stuck with things once they are built is also to say that we are presented with a particular set of maintenance tasks, and thus possibilities for creative repair and reconstruction. Even hard facts like infrastructure are more or less open to revision or reconstruction. An obvious example would be the so-called “gatos” or illegal utility connections in the favelas of Rio de Janeiro; admittedly, those are not exactly about repairing systems, but if we take seriously the idea of repair as a mechanism for innovation, then the line between repairing systems and “hacking” them becomes fuzzy.
I also can’t help but think here of Geoff Manaugh’s description of burglary as just “another way of using a building.” Burglars “use” buildings and structure in ways that were not intended by their designers, but the possibilities for doing so are still determined by the designs— dimensions, materials, entrances and and exits, and so on. Everybody is working with the same materials, the same facts on the ground, but those materials may enable a wide range of possible responses.
So I find myself again thinking of the concept of affordances, the “action possibilities provided to the actor by the environment.” These include not only those possibilities envisioned or intended by designers and planners, but others they did not foresee, and might even find undesirable. The built environment imposes limitations, but also generates its own distinct possibilities; the way a city works (or doesn’t) will be a result of how people respond to both.