March 22, 2013
What Does “Resilience” Have to Do With Architecture?
Resilient systems may help architecture
The word “resilience” is bandied about these days among environmental designers. In some quarters, it’s threatening to displace another popular word, “sustainability.” This is partly a reflection of newsworthy events like Hurricane Sandy, adding to a growing list of other disruptive events like tsunamis, droughts, and heat waves. We know that we can’t design for all such unpredictable events, but we could make sure our buildings and cities are better able to weather these disruptions and bounce back afterwards. At a larger scale, we need to be able to weather the shocks of climate change, resource destruction and depletion, and a host of other growing challenges to human wellbeing. We need more resilient design, not as a fashionable buzzword, but out of necessity for our long-term survival.
Aside from a nice idea, what is resilience really, structurally speaking? What lessons can we as designers apply towards achieving it? In particular, what can we learn from the evident resilience of natural systems? Quite a lot, it turns out.
Resilient and non-resilient systems
Let’s start by recognizing that we have incredibly complex and sophisticated technologies today, from power plants, to building systems, to jet aircraft. These technologies are, generally speaking, marvelously stable within their design parameters. This is the kind of stability that C. H. Holling, the pioneer of resilience theory in ecology, called “engineered resilience.” But they are often not resilient outside of their designed operating systems. Trouble comes with the unintended consequences that occur as “externalities,” often with disastrous results.
More from Metropolis
On the left, an over-concentration of large-sale components; on the right, a more resilient distributed network of nodes.
Drawing by Nikos A. Salingaros
A good example is the Fukushima nuclear reactor group in Japan. For years it functioned smoothly, producing reliable power for its region, and was a shining example of “engineered resilience.” But it did not have what Holling called “ecological resilience,” that is, the resilience to the often-chaotic disruptions that ecological systems have to endure. One of those chaotic disruptions was the earthquake and tsunami that engulfed the plant in 2010, causing a catastrophic meltdown. The Fukushima reactors are based on an antiquated U.S. design from the 1960s, dependent upon an electrical emergency cooling system. When the electricity failed, including the backup generators, the emergency control system became inoperative and the reactor cores melted. It was also a mistake (in retrospect) to centralize power production by placing six large nuclear reactors next to each other. The trouble with chaotic disruptions is that they are inherently unpredictable. Actually we can predict (though poorly) the likelihood of an earthquake and tsunami relatively better compared to other natural phenomena. Think of how difficult it would be to predict the time and location of an asteroid collision, or more difficult yet, to prepare for the consequences. Physicists refer to this kind of chaos as a “far from equilibrium condition.” This is a problem that designers are beginning to take much more seriously, as we deal with more freakish events like Hurricane Sandy — actually a chaotic combination of three separate weather systems that devastated the Caribbean and the eastern coast of the U.S., in 2012.
Hurricane Sandy on October 28, 2012.
Courtesy LANCE MODIS Rapid Response Team at NASA GSFC
As if these unforeseen dangers were not enough, we humans are contributing to the instability. An added complication is that we ourselves are now responsible for much of the chaos, in the form of our increasingly complex technology and its unpredictable interactions and disruptions. Climate change is one consequence of such disruptions, along with the complex and unstable infrastructures we have placed in vulnerable coastal locations. (In fact, Japan’s technological infrastructure has been heavily damaged over a much wider area by the chaotic “domino” effects of the Fukushima disaster.) Our technological intrusion into the biosphere has pushed natural systems into conditions that are far from equilibrium — and as a result, catastrophic disruptions are closer than ever.
So what can we learn from biological systems? They are incredibly complex. Take, for instance, the rich complexity of a rainforest. It too generates complicated interactions among many billions of components. Yet many rainforests manage to remain stable over many thousands of years, in spite of countless disruptions and “shocks to the system.” Can we understand and apply the lessons of their structural characteristics? It seems we can. Here are four such lessons extracted from distributed (non-centralized) biological systems that we will discuss in more detail:
- These systems have an inter-connected network structure.
- They feature diversity and redundancy (a totally distinct notion of “efficiency”).
- They display a wide distribution of structures across scales, including fine-grained scales.
- They have the capacity to self-adapt and “self-organize.” This generally (though not always) is achieved through the use of genetic information.
Map of the Internet: a paradigmatic resilient network in part because it is scale-free and redundant.
Courtesy the Opte Project/Wikimedia
The Internet is a familiar human example of an inter-connected network structure. It was invented by the U.S. military as a way of providing resilient data communications in the event of attack. Biological systems also have inter-connected network structures, as we can see for example in the body’s separate blood and hormone circulation systems, or the brain’s connected pattern of neurons. Tissue damaged up to a point is usually able to regenerate, and damaged brains are often able to re-learn lost knowledge and skills by building up new alternative neural pathways. The inter-connected, overlapping, and adaptable patterns of relationships of ecosystems and metabolisms seem to be key to their functioning. Focusing upon redundancy, diversity, and plasticity, biological examples contradict the extremely limited notion of “efficiency” used in mechanistic thinking. Our bodies have two kidneys, two lungs, and two hemispheres of the brain, one of which can still function when the other is damaged or destroyed. An ecosystem typically has many diverse species, any one of which can be lost without destroying the entire ecosystem. By contrast, an agricultural monoculture is highly vulnerable to just a single pest or other threat. Monocultures are terribly fragile. They are efficient only as long as conditions are perfect, but liable to catastrophic failure in the long term. (That may be a pretty good description of our current general state!) Why is the distribution of structures across scales so important? For one thing, it’s a form of diversity. By contrast, a concentration at just a few scales (especially large scales) is more vulnerable to shocks. For another thing, the smaller scales that make up and support the larger scales facilitate regeneration and adaptation. When the small cells of a larger organ are damaged, it’s easy for that damaged tissue to grow back — rather like repairing the small bricks of a damaged wall.
Distribution of inter-connected elements across several scales.
Drawing by Nikos A. Salingaros
Self-organization and self-adaptation are also central attributes of living systems, and of their evolution. Indeed, this astonishing self-structuring capacity is one of the most important of biological processes. How does it work? We know that it requires networks, diversity, and distribution of structures across scales. But it also requires the ability to retain and build upon existing patterns, so that those gradually build up into more complex patterns. Often this is done through the use of genetic memory. Structures that code earlier patterns are re-used and re-incorporated later. The most familiar example of this is, of course, DNA. The evolutionary transformation of organisms using DNA gradually built up a world that transitioned from viruses and bacteria, to vastly more complex organisms.
Applying the lessons to resilient human designs
How can we apply these structural lessons to create resilient cities, and to improve smaller vulnerable parts of cities by making them resilient? Developing the ideas from our previous list, resilient cities have the following characteristics:
- They have inter-connected networks of pathways and relationships. They are not segregated into neat categories of use, type, or pathway, which would make them vulnerable to failure.
- They have diversity and redundancy of activities, types, objectives, and populations. There are many different kinds of people doing many different kinds of things, any one of which might provide the key to surviving a shock to the system (precisely which can never be known in advance).
- They have a wide distribution of scales of structure, from the largest regional planning patterns to the most fine-grained details. Combining with (1) and (2) above, these structures are diverse, inter-connected, and can be changed relatively easily and locally (in response to changing needs). They are like the small bricks of a building, easily repaired when damaged. (The opposite would be large expensive pre-formed panels that have to be replaced in whole.)
- Following from (3), they (and their parts) can adapt and organize in response to changing needs on different spatial and temporal scales, and in response to each other. That is, they can “self-organize.” This process can accelerate through the evolutionary exchange and transformation of traditional knowledge and concepts about what works to meet the needs of humans, and the natural environments on which they depend.
Resilient cities evolve in a very specific manner. They retain and build upon older patterns or information, at the same time that they respond to change by adding novel adaptations. They almost never create total novelty, and almost always create only very selective novelty as needed. Any change is tested via selection, just as changes in an evolving organism are selected by how well the organism performs in its environment. This mostly rules out drastic, discontinuous changes. Resilient cities are thus “structure-preserving” even as they make deep structural transformations. How do these elements contribute to resilient cities in practice, in an age of resource depletion and climate change? It’s easy to see that a city with networked streets and sidewalks is going to be more walkable and less car-dependent than a city with a rigid top-down hierarchy of street types, funneling all traffic into a limited number of “collectors” and “arterials.” Similarly, a city designed to work with a mix of uses is going to be more diverse and be able to better adapt to change than a city with rigidly separated monocultures.
A complex resilient system coordinates its multi-scale response to a disturbance on any single scale.
Drawing by Nikos A. Salingaros
A city with a rich and balanced diversity of scales, especially including and encouraging the most fine-grained scales, is going to be more easily repairable and adaptable to new uses. It can withstand disruptions better because its responses can occur on any and all different levels of scale. The city uses the disruption to define a “pivot” on a particular scale, around which to structure a complex multi-scale response. And it’s more likely to be able to self-organize around new economic activities and new resources, if and when the old resources come to be in short supply.
The evolution of non-resilient cities
So where are we today? Many of our cities were (and still are) shaped by a model of city planning that evolved in an era of cheap fossil-fuel energy and a zeal for the mechanistic segregation of parts. The result is that in many respects we have a rigid non-resilient kind of city; one that, at best, has some “engineered resilience” towards a single objective, but certainly no “ecological resilience.” Response is both limited and expensive. Consider how the pervasive model of 20th century city planning was defined by these non-resilient criteria:
- Cities are “rational” tree-like (top-down “dendritic”) structures, not only in roads and pathways, but also in the distribution of functions.
- “Efficiency” demands the elimination of redundancy. Diversity is conceptually messy. Modernism wants visually clean and orderly divisions and unified groupings, which privilege the largest scale.
- The machine age dictates our structural and tectonic limitations. According to the most influential theorists of the modernist city, mechanization takes command (Giedion); ornament is a crime (Loos); and the most important buildings are large-scale sculptural expressions of fine art (Le Corbusier, Gropius, et al.).
- Any use of “genetic material” from the past is a violation of the machine-age zeitgeist, and therefore can only be an expression of reactionary politics; it cannot be tolerated. Novelty and neophilia are to be elevated and privileged above all design considerations. Structural “evolution” can only be allowed to occur within the abstracted discourse of visual culture, as it evaluates and judges human need by its own (specialized, ideological, aestheticizing) standards.
From the perspective of resilience theory, this can be seen as an effective formula for generating non-resilient cities. It is not an accident that the pioneers of such cities were, in fact, evangelists for a high-resource dependent form of industrialization, at a time when the understanding of such matters was far more primitive than now. Here, for example, is the architect Le Corbusier, one of the most influential thinkers in all of modern planning, writing in 1935, and providing a blueprint for modern sprawl: “The cities will be part of the country; I shall live 30 miles from my office in one direction, under a pine tree; my secretary will live 30 miles away from it too, in the other direction, under another pine tree. We shall both have our own car. We shall use up tires, wear out road surfaces and gears, consume oil and gasoline. All of which will necessitate a great deal of work … enough for all.” Sadly, there is no longer enough for all! This relatively brief age of abundant fossil fuels — and the non-resilient urban architecture that it has spawned all over the globe — is rapidly drawing to a close. We must be prepared for what has to come next. From the perspective of resilience theory, the solutions are not going to be simple techno-fixes, as so many naively believe. What is required is a deeper analysis and restructuring of the system structure: admittedly not an easy thing to achieve since it doesn’t make money short-term.
Postscript: a lesson from our own evolution
People tend to be carried along by the present, and put both past and future out of their mind. Even in our information-glutted age, the past is remote and abstract–just another set of images like any movie. And so we ignore where we have come from, and the path that brought us here to our marvelous technological culture. We are ill prepared to see where we must go next. For our techno-consumerist culture, tomorrow will bring no surprises. But new research in anthropology, anthropogeny, and genetics suggests that we humans are, quite literally, creatures of climate change. Thanks to ingenious detective work, we now know that 195,000 years ago, our species very nearly became extinct — down to hardly more than 1,000 survivors clinging to the southern African coast, as a mega-drought swept that continent. Our evident response was to diversify, and to develop many new sources of food as well as new technologies for acquiring them: fishhooks, barbs, baskets, urns, and other innovations. More complex language probably followed, allowing us to coordinate more sophisticated strategies for hunting and gathering. 10,000 years ago, it now appears, we adapted once again to a mini-ice age, prompting us to innovate with new agricultural technologies, and new forms of settlement around them. These innovations arose more or less simultaneously in many parts of the then-disconnected world, suggesting that the trigger was very likely the climate. Now we are facing the third great adaptation of our history to climate change. But this time it is we, ourselves, who have triggered it with our own technologies. If we are going to adapt successfully, we will need to understand the opportunities to innovate yet again, in the way we design and operate our technology. Our comfortable lifestyle (in the wealthy West, and among those socioeconomic classes that can afford to copy us) is significantly less resilient than most people would care to admit, or even dare think about. If we are going to continue our so-far remarkably successful run as a technological civilization, we had better take the lessons of resilience theory to heart.
Michael Mehaffy is an urbanist and critical thinker in complexity and the built environment. He is a practicing planner and builder, and is known for his many projects as well as his writings. He has been a close associate of the architect and software pioneer Christopher Alexander. Currently he is a Sir David Anderson Fellow at the University of Strathclyde in Glasgow, a Visiting Faculty Associate at Arizona State University; a Research Associate with the Center for Environmental Structure, Chris Alexander’s research center founded in 1967; and a strategic consultant on international projects, currently in Europe, North America and South America.
Nikos A. Salingaros is a mathematician and polymath known for his work on urban theory, architectural theory, complexity theory, and design philosophy. He has been a close collaborator of the architect and computer software pioneer Christopher Alexander. Salingaros published substantive research on Algebras, Mathematical Physics, Electromagnetic Fields, and Thermonuclear Fusion before turning his attention to Architecture and Urbanism. He still is Professor of Mathematics at the University of Texas at San Antonio and is also on the Architecture faculties of universities in Italy, Mexico, and The Netherlands.
Read more posts from Michael and Nikos here.