The Future Of Architecture Must Be “Agile”

Technology is triggering a cascading series of unintended consequences that degrade quality of life on Earth.

As humanity progresses into an increasingly technological 21st century, we are confronted with a historic and alarming paradox.

Over the last two and a half millennia, our species has made historic progress in achieving (partially but substantially) ancient ideals of democracy, human rights, justice, and equality. Our institutions of science and technology have made brilliant advances, while the global economy has created unprecedented wealth. Billions of people around the world are healthier, better educated, and more empowered to shape their own lives and futures.

And yet, as many are well aware, we are entering an era of growing existential threat—caused, ironically, by our very technological successes. We are depleting our resources at unsustainable levels, and creating unprecedented damage to the critical Earth systems on which prosperity and even life itself depend. Our own technology—including our economic technology—is triggering an interacting, cascading series of unintended consequences that degrade quality of life, and now threaten to become catastrophic. The most notable example (though by no means the only one) is anthropogenic climate change.


More from Metropolis


Closely related to the malfunction of our technology is the malfunction of our institutions that are critical to learning, governance, regulation, and reform: politics, economics, journalism, law, and others. Worrisome evidence is growing that those institutions are unable to address the real problems we face—dangerously degraded by unintended consequences and perverse incentives, fragmenting and confusing the essential processes of an intelligent culture. This is a formula for sure disaster in the years ahead. (Hence the clear warning of the great urbanist Jane Jacobs’ final book, Dark Age Ahead).

A case in point is in the professions of architecture, planning, and development. Research in environmental psychology reveals a huge gulf between what most people judge to be good quality development, and what architects, planners, and developers actually build (and celebrate through relentless and effective group indoctrination). The chasm is so large that it’s common to hear ordinary people, un-bewitched by marketing, remarking on the ugliness, strangeness, or inferior quality of most new development [see “The Architect Has No Clothes” in On the Commons magazine].

Those perceptions are also backed up by research into the actual performance of these places—even highly touted “green” ones built by world-famous architects. As we have written previously, many of the claims to sustainability and resilience are disproven by remarkably damning post-occupancy evaluations.

[O]ur current conception of design is bound up within a pathological form of growth.

These lessons remind us that the problem is not simply that we need to be a bit more efficient with our resources, or to recycle more. That will only buy a small amount of extra time. To survive and prosper, we will need to change our fundamental relationship to the planet’s resources, and the ways we go about extracting, structuring, and transforming them.

Among other things, this means a fundamental re-conception of what it is to design—that is, how we transform resources into the structures of our world. We must recognize that our current conception of design is bound up within a pathological form of growth, which relies upon unsustainable levels of waste and debt. A fundamentally more sustainable, more resilient kind of growth—more like the evolutionary, cyclical growth that occurs within biological systems—could save us. This in turn will require a different, intrinsically resilient kind of institutional system.

Crucially, this new kind of growth must occur within our systems of settling and inhabiting the Earth: the architecture of our cities, towns, and countrysides. This “ecologically resilient” architecture, in the words of resilience pioneer C. S. Holling, must be able to withstand chaotic, non-linear events, beyond the narrow parameters of “engineered resilience.” More than that, our technological growth needs to become, as political economist Nassim Nicholas Taleb has termed it, “antifragile”—able to learn, and even to gain from disorder.

What sort of profound institutional changes will be required? A key insight comes from software design, and the methodology known as “Agile.”

(L) Hundreds of thousands of gated communities and privatized pseudo-public spaces have arisen around the world, such as this isolated, car-dependent community in Argentina. (R) Contrast the continuous open, walkable network of great cities like Rome, shown in the famous Nolli plan. This network structure has profound economic implications.

Photo by Alex Steffler, Wikimedia

Generate, don’t specify 

Some years ago in the world of computer software, programmers recognized problems with increasingly “cluttered” computer code. Its unpredicted interactions produced unacceptable malfunctions—much as we are experiencing in other forms of technology today. One of the most effective responses came to be called “Agile.” As software pioneer Ward Cunningham described it, specifying the desired behaviors always required elaborate definitions and standards, while, paradoxically, generating them often only required the identification (through a process of adaptive iteration) of a much simpler set of generative rules.

It turns out that many biological systems work in just this way. The complex pattern of bird flocks, to take just one example, is not created out of a kind of rigid blueprint, specifying the complex shape at any given instant. That would be an overwhelmingly vast set of instructions. Rather, each bird has only a few simple rules for maintaining its position relative to its leader and neighbors. From the interaction of these simple instructions, the beautiful complex geometric patterns of the flock are generated.

As we have noted previously, the beauty of such patterns is closely related to their capacity for resolving problems (such as the complex challenge of flock migration). We humans add other layers of structure into our designs, including symbolic, artistic, and abstract components. But it is mistaken to think of these as fundamentally different. Each aspect of structure, in its own way, helps the complex function of a living process.

An “Agile” approach helps us to resolve the challenges created by our own technologies. Instead of adding more “bolt-on” gadgets to address each of the malfunctions we encounter, we make an “agile” transformation of the system we are dealing with, which allows it to adapt better to the living function. Often this is a surprisingly simple change, in a surprising part of the system.

An “Agile” approach helps us to resolve the challenges created by our own technologies.

Simple “Agile” principles can be applied to the older, more entrenched systems of designing our buildings and cities. The needed reform is not simply to make further bolt-on additions to the current “operating system,” in the form of additional regulations, laws, and restrictions. Those are what one immediately thinks of in such a discussion, but they have proven ineffective at best. Such additions are likely to increase the problem of unintended consequences and perverse outcomes, and not actually remedy the problem.

A central principle, as with Agile Methodology in software design, is that the operating system should be re-written, not to specify the behavior desired, but rather, create the conditions in which that behavior is most likely to be generated. This “generative design approach”—employing complex adaptive transformations, engaging economic processes, and exploiting Agile self-organizing capacities—is emerging as the key to resilient design for the future.

Two hospitals in Portland, Oregon. (L) Providence St. Vincent Medical Center is an isolated “campus” style that disrupts the urban fabric around it. (R) Legacy Good Samaritan Medical Center is fully integrated into the walkable mixed-use street grid of Northwest Portland.

Photos courtesy Bing

Transform, don’t replace

Another principle of Agile Design is to invest more time understanding the existing structure, and seeking to find “agile” ways to transform it. This, too, is often a simpler approach than starting from scratch, requiring fewer but more effective rules.

Design up to now is widely conceived as directing the complex assembly and composition of elements into consumable “products” (including buildings). This process also typically bestows a consciously created veneer of aesthetic novelty onto these products, as a way of promoting their (temporary) desirability. (Though often characterized as great art, it is only rarely regarded as such by later generations.) This linear process continues with the rapid obsolescence and disposal of the products, and the creation of new and improved products (with fashionably new artistic veneers) to replace them.

This is a fundamentally unsustainable process.

Adaptive design is a continuous (and continuously beneficial and profitable) process of transformation, in which novel aspects are typically combined with enduring and recurrent ones. Artistic aspects have to work in service to this evolutionary pattern, and not be allowed to dictate it. Design, according to this definition, creates a transformation “from existing states to preferred ones,” as the great polymath Herbert Simon put it.

Of course Simon’s wonderful definition raises more questions than it answers—but they are the right questions. For example, who is doing the preferring? It has to be a larger democratic process, and not solely specialists—architects, developers, art-connoisseurs, or city-boosters. But how is this process going to proceed in an intelligent way, reconciling the preferences of multiple agents into an effective optimum? Beyond mere marketing to consumers, sustainable design must address this profound civic question too.

Furthermore, a “preferred state” itself is not fixed, but by its nature, represents an optimum balance between a number of factors. Those factors are in a dynamic interaction, which requires that we experiment within an iterative, adaptive process to achieve what we prefer.

It is also challenging to know what is the existing state, and how to transform it. Perhaps, as we learn more about the existing state, our sense of what we prefer will also transform. We may find that there are aspects of the existing state that we will choose to retain. In this sense, design is necessarily an evolutionary process of discovery and redefinition, which can’t be predicted in advance.

As these evolutionary alternatives proliferate, some of them will frequently re-incorporate features that occurred previously (simply because those are still the best solutions). In nature, one notable example is the dorsal fin of a porpoise, which re-incorporates the much older shape of a shark fin.

But in “modern” human design, we have allowed the dictates of artistic novelty to usurp this evolutionary process. As Jane Jacobs pointed out, this confusion of roles is bad for art, and even worse for cities. The art-clad novelty regime dominating today’s consumer-oriented architecture and design is fundamentally non-resilient and unsustainable. 

Art’s profoundly important role in design has become corrupted, and turned into novelty packaging, which has created a dangerous form of cultural clutter. The essential contribution of Art to communication, to legibility, to elucidating meanings, is now exploited as a kind of “Trojan Horse” for those who would profitably industrialize the built environment, without regard to the long-term consequences.

Correcting this damaging state of affairs requires that we recapitulate good solutions from any evolutionary source, from any era. If we want to be truly sustainable, we need to freely utilize, as nature utilizes, the recurrence of evolutionary geometric patterns. (This includes the best solution-patterns evolved from centuries of human tradition, but foolishly rejected by naïve “modern” designers.) Within this resilient framework, art can take its fully creative place.

Two shopping centers in the Portland, Oregon area. (L) Washington Square in Tigard, Oregon is a large car-dependent campus. (R) Pioneer Square in central Portland extends over several blocks, engaging and preserving the walkable, transit-served street grid.

Photos courtesy Bing

Simplify and adapt the “operating system for growth”

Closely related to the way we design is the way we relate to others in the production process. At present, design is largely conceived as occurring within a specialist’s intricately prescribed context. This has to change.

Instead, design must be expanded to engage the required tactical changes in what we refer to here as the “operating system for growth.” Working collaboratively with others, we can transform the interacting collection of incentives, rewards, penalties, regulations, standards, laws, and models — constituting a kind of “operating system” that generates the growth of environmental structures (and other artifacts and systems) around the world.

Like a computer operating system, or the rules of a game, this “operating system for growth” allows some activities but not others. If you want some processes to function in ways that are not currently permitted, you will have to reform the operating system — in this case, the models and other standards of planning, architecture, and development.

Many of the results of the current operating system are “perverse” — that is, they are not what people would have wanted originally, but they result from the distorted way that the incentives and other requirements interact. Those incentives (few of which seem objectionable in isolation) encourage some people to make excuses for these perverse outcomes — to pretend (or to believe) that they were desired all along. Perhaps, some are misled to think, these outcomes are even something wonderful. This is what is known as a “cognitive illusion,” and it helps to explain a number of current professional dysfunctions in architecture and planning.

But again, the goal is not to clutter up the operating system with rules and procedures that are fragile because of their intricacy and “patched-on” quality, but to identify (through an evolutionary cultural process of adaptation) relatively simple, agile sets of incentives and processes that will create the conditions under which the desired behavior is most likely to be generated. Sometimes, the solution is to remove over-complicated and malfunctioning elements; sometimes, however, it is a matter of making transformations with slight additions. (This is not at all a simple-minded libertarian prescription.)

Two universities. (L) The Evergreen State College in Olympia, Washington is an isolated, car-dependent campus. (R) Portland State University is fully integrated into the walkable street network of Portland, Oregon.

Photos courtesy Bing

Value externalities

The most powerful incentive fueling any “operating system for growth” lies simply in the way the underlying economic processes work, and how certain kinds of “economies” (strategies to achieve economic efficiency and benefit) are rewarded, while others are neglected. We wrote in “The Geometry of Resilience” about the powerful economies of scale and standardization displacing two other essential economies needed for resilience: the economies of place and differentiation. Correcting this imbalance will bring in more local economic functionality, as well as more diverse kinds of economic activities. (A familiar example of both might be, say, a local farmers’ market that offers a wide variety of heirloom vegetables by different growers. The opposite might be a standardized kind of corn that dominates international markets and drives local varieties to extinction.)

At present there is an unequal contest between the economics of scale and standardization on the one hand, and those of place and differentiation on the other. This comes from the failure of markets to value what are known as “externalities” — negative (and sometimes positive) external factors that are not taken into account in the economic calculations for a project. Such a project could end up damaging these external resources, or it may fail to create the benefits that would otherwise be possible.

The problem may be understood as the unequal competition between distinct temporal scales (processes defined on very different time intervals). Natural and human systems eventually adjust to forces that are destroying them, but with a relatively long reaction time. By contrast, global finance can move very quickly to intervene significantly in human and natural ecosystems. This is a historically novel phenomenon, which our technological systems are not equipped to handle. If such moves damage long-term processes—and they often do, because they are oriented strictly towards the extraction of short-term profits — then the system cannot possibly respond in time.

True modernity lies…in a different way of thinking about what it is to design for the full participation of all human beings.

One evident example is depletion of natural resources, which is typically not valued until the resource in question is so close to total depletion that shortages put upward pressures on prices. Another related one includes the externalities of conventional “sprawl” development. Current developers are not required to pay for the external costs (externalities) that include future infrastructure repairs, damage to ecosystems, impacts on human health, degradation of air quality, and many more.

Economies of scale and standardization by themselves create significant externalities, which can only be re-balanced by the finer-grained economies of place and differentiation. But if the operating system provides no information feedback (such as financial reward) for employing these other economies, they will continue to be neglected, and the balance between the economies will remain unequal (as they are today).

One thing we can do is to put a price on externalities. That means, for example, paying for carbon when it is emitted into the air, or for ecosystems when they are destroyed, or for other resources when they are depleted. Systems need to do this pricing in a largely consensual way—exposed to the checks and balances of a larger participatory process, and not based on the decisions of one self-interested party or another.

Two industrial districts. (L) The Intel Ronler Acres campus in Hillsboro, Oregon, planned for an area with limited industrial uses. (R) The new North American headquarters of Vestas Wind Systems, in the walkable street grid of Northwest Portland, an area popular with high tech employees.

Implementation requires a functional participatory process—a democratic process—and, as most people recognize, in almost all countries we need reforms here too. The current political process that would develop those reforms is “broken” in the USA and other countries; specifically, it is unduly influenced by narrow financial interests.

There is another major piece of economic reform that is related. Right now, human creativity is taxed in a way that is largely indistinguishable from the way depletion of resources is taxed. That makes no sense long-term. We need to transition to a form of public valuation and pricing (that is, of taxation) that prices the consumption and depletion of resources in a very different way — and usually at a significantly higher rate — than the creative activities of people. We suggest that this “Georgist” approach to tax policy (so named for 19th century political economist Henry George) is likely to be a key economic ingredient in the resilient transition ahead.

Two “big box format” buildings exhibiting very different urban models. (L) An IKEA in suburban Portland, Oregon, characteristic of many thousands of similar retail facilities around the world. (R) A Target Store in a 200,000 sq ft building in downtown Portland, occupying two floors (about 90,000 sq ft). This historic building was one of the first department stores west of the Mississippi River in the USA, and now includes an adjoining parking garage. The format on the left vastly outnumbers that on the right.

Toward a “new modernity”

Most of those reading this post are, like us, at the higher end of the global economic pyramid.  It is sobering to recognize that the wellbeing of many billions (as well as our own descendants) depends disproportionately on our own actions in coming years. We face the difficult task of balancing enjoyment in our own daily lives with the larger responsibilities of stewardship over the culture and the planet.

It is understandably fun to engage in the edgy, attention-getting art-novelties of our consumer-based design culture. But it is silly to suppose that this approach is in any genuine sense progressive, sustainable, or “modern.” In fact it is only reactionary orthodoxy, clinging to a nearly century-old, outmoded conception of industrial modernity. True modernity lies in the embrace of new models of global growth, embodying evolutionary pattern, organized complexity, and adaptive morphogenesis. It lies in a different way of thinking about what it is to design for the full participation of all human beings, for living systems, and for a living planet.

Comparison of “object buildings” on superblocks with “fabric buildings” within a walkable, transit-served, multi-modal mixed-use community. The structure on the left is closer to a sculpture gallery than to a functioning city.


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.  He is a Research Associate with the Center for Environmental Structure, Alexander’s research center founded in 1967, and Executive Director of the Sustasis Foundation, a Portland, OR-based NGO dedicated to developing and applying neighborhood-scale tools for resilient and sustainable development.

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.

Recent Projects