What We Didn't Learn From Katrina

Cities are immensely complex self-organizing systems, not mere top-down designs—but they do need top-down interventions in strategic places. Unfortunately, we still have inadequate models and tools.

15 minute read

August 28, 2015, 5:00 AM PDT

By Michael Mehaffy


Tad Denson / Shutterstock

"Treat these people gently," I was told, "as if they've just lost a family member."

As indeed they had, in some cases. In many others they had lost beloved friends and members of the community. In almost all cases, they had lost their homes, their neighborhoods, and their quality of life. They lost the fabric of social connections that grows up over time in a neighborhood, a structure that is so easy to take for granted—but one that proves itself so vital in a time of crisis.

I had just arrived in a devastated coastal Mississippi when I was given this advice, and soon I would be in New Orleans, as part of work on three different recovery planning teams in 2005 and 2006. For a planner it was a deeply humbling experience, and at times a maddening one. Yes, government malfunctioned colossally at virtually every level, both before and after the storm, demonstrating an unnerving fecklessness at managing complex Katrina-level events.

An aerial view of Coastal Mississippi shows buildings wiped out along the coast, and severe damage to homes beyond visible under blue FEMA tarps. (Image: Michael Mehaffy)

But there was a deeper lesson, beyond the dysfunction of bureaucracies or the incompetence of individuals. Indeed there was a host of lessons, if we take the time to tease them out. But today, ten years later, I wonder how many of them we learned. I wonder if we are poised to relive the nightmare of the next inevitable disaster—or indeed, if we are already doing so, in a larger kind of slow-motion disaster.

By the time I got to New Orleans, the first lesson that struck home was how ill-prepared we were to accomplish what the residents most urgently asked us to do, over and over again. "Please help us to rebuild our homes." A smaller number also asked for help rebuilding their small businesses, which were often within or near their homes. But these were not the wealthy and powerful we were dealing with.

We dealt with renters too, but in New Orleans, a remarkable number of people, including the poor, owned their own homes. Of course they needed financial help—but more than that, they needed help navigating through a blizzard of bureaucracy, much of it private. They often found themselves in an absurdist nightmare, required to produce mortgages and insurance papers and other documents for homes that had washed away, along with all their records.

Nor were they prepared to navigate the bewildering complexities of contracting, financing, design, permitting—in effect, overnight to become residential homebuilders, a field in which they typically had zero experience (and zero time to perform, let alone learn). Nor did they have experience with the design of homes, or the correct restoration of historic properties, or the best way to respond to the neighborhood character and quality, or any of innumerable other demands. For almost all that I spoke to, it was an overwhelming ordeal.

In this context, the promise of endlessly delayed relief funding was doubly cruel. It had the effect of putting lives in limbo, shutting off alternative local solutions, and making the rebuilding, when it did finally come, all the more difficult. Worse, the money was often mismatched to the actual local need. If an insurance adjuster offered, say, $10,000 for a lost roof, but the lowest available contractor price was $20,000, the fed up homeowner—who was very unlikely to have another $10,000 lying around in the bank—might as well decide to pocket the $10,000 and move to Baton Rouge. Such a shortfall was all too common.

And this was no way to rebuild a city.

Our response was to propose immediate, locally based "neighborhood rebuilding centers," offering knowledge, expertise, access to resources—and above all, peer-to-peer sharing of information. These would be, in effect, small neighborhood research libraries, where residents could come and get solutions to their many detailed problems. Above all, they could help each other to learn the issues and solutions for their own neighborhood and situation.

Who were the best contractors, and when were they available? How do you talk to an insurance adjuster, and present your case? How can a plan be prepared for permitting and construction? How can we figure out any of the ten thousand things we have to know that are specific to this neighborhood—its base flood elevation, the history of its houses, the zoning requirements, and so on?

In the world of utilities, there is a well-known challenge called "the last mile problem." It turns out that, rather than the many miles that lead up to it, this last gap is the most difficult to bridge. In this last-mile gap is where many small and detailed connections and extensions have to be made, and it can't be done wholesale, top-down. It has to be done bottom-up, one house at a time.

Something similar was going on with the redevelopment after Katrina. We were focusing on everything except the last-mile problem—lots of talk about big solutions, big pots of money, massive rebuilding projects by movie stars and big national charities. But that wasn't where the real challenge was. It was in the last mile, in the humble case-by-case self-organization at the level of the house and the neighborhood.

New Orleans houses with the telltale cross-shaped code of damage spray-painted on the fronts. (Image: Michael Mehaffy)

Our solution was to propose resources targeted to the last mile—not only top-down ones, like expert technical knowledge and funding, but bottom up ones, like peer-to-peer knowledge sharing. We proposed a neighborhood wiki system that would allow sharing of tips, ideas, and how-to knowledge. It would be a tool of local empowerment and self-organization.

We also proposed a "pattern language" format—a locally customizable system of design elements to assist with homes and neighborhoods. (It is also related to Wiki—both are modifiable, peer-to-peer, and hyperlinked.) We brought in our friend Ward Cunningham, the inventor of Wiki, to help to develop the software. We brought in Christopher Alexander, the father of pattern languages, to consult.

To their credit, the Louisiana Recovery Authority (LRA) strongly supported the proposal, even offering to fund a pilot project—until their lawyers got their hands on it. "There is a data protection issue," officials were told. "The government could be held liable if information were used by one party to damage another." Curiously, that kind of liability seems to have been no problem for the likes of Facebook, Twitter, and other peer-to-peer social media—and, for that matter, Wikipedia itself.

But the government was risk-averse, and in this there was a kind of paralysis. A lack of certain action meant no action. More than that, the LRA seemed to be unsure even whether this kind of rebuilding effort was really the right priority after all. In frank discussions with Andy Kopplin, the director of the LRA, he candidly told me that "perhaps the best strategy is just to rebuild in Baton Rouge." I found myself wondering whether this was a recovery program, or a program of retreat.

To be sure, the rebuilding of a great city like New Orleans is a daunting challenge. It simply cannot be done top-down, and I understand if Kopplin felt as though he'd hit a brick wall. But I think the lessons do show that a city like New Orleanscanbe rebuilt bottom-up, with top-down support. The challenge is daunting but not insurmountable. Indeed, that's how a great city like New Orleans grew up in the first place.

New Orleans is a marvel of informal order. Its older neighborhoods are loose jazzy improvisations of buildings and details and quirky outdoor spaces, all exquisitely human scale and aimed at pedestrian delight. A walk down one of its streets reveals the layers of human activity and change that have grown up there, re-organizing, and transforming neighborhoods bit by bit. It is a marvel of durable livability.

But this self-organizing growth is hardly random. New Orleans displays a complex cultural language of patterns and elements that were borrowed, adapted and applied—some of them French, some Spanish, some purely local—and in many cases, a kind of creole fusion of them all, into something new and remarkably coherent.

There were also top-down patterns that supported bottom-up growth. For example, the historic French farm tracts had property lines that extended perpendicular to the river, in a complex pattern that fans out and back as the river wends its way. Along these property lines, major streets would later be built, giving the city's street pattern its distinctive radial form.

So it goes: bottom-up, top-down, bottom-up again, forming a fugue, a weave, an urban fabric. People interacting, creating businesses, building houses, shaping gardens and details, layering on top. Applying carefully refined patterns, with careful evaluation and skill. It is neither all top-down nor all bottom-up, but a catalytic weave of the two.

And how the two weave together and support one another is everything.

The gardening analogy is apt. You cannot command plants to grow in the form you specify. But you can take steps to assure that they do grow, bottom-up, with your supportive top-down actions. You can plant the right seeds—good genetic material. You can build trellises that will support the growth of the plants. You can fertilize, and water, and prune where necessary. You can shape the growth in the way you need and want.

Just so with cities: you can plant genetic seeds as design patterns and types, you can build "trellises" in the infrastructure and streets, and you can fertilize with incentives, and prune with regulations and disincentives. You can empower local people to grow and diversify into something infinitely richer and more complex than you could have ever specified as an act of top-down design.

Two New Orleans houses that started out identical have transformed slowly over time into very different buildings.  Stewart Brand, in his bookHow Buildings Learn, used similar examples from New Orleans. (Image: Michael Mehaffy)

You can develop genetic material, the "DNA of place," through a rich repository of design ideas—and where better to do so than from the several thousand years of human experience building successful settlements? So Greece passes its genetic material on to Rome, Rome on to Spain, Spain on to the West Indes, and finally it arrives in New Orleans in a delightful, well-adapted mix (along with English, French, and other sources) creating an exquisitely local, high-performing architecture.

Contrast that with another strategy, which is to empower individual heroic top-down designers, the modern architects. They come to New Orleans with the best intentions, intent on creating something "fresh," "innovative," "exciting and new." It is a kind of novelty act—and the lessons of history show that this top-down approach is doomed to fail. But no one learns these lessons.

No one wants to believe them, because everyone has the best of intentions. Take, for example, the Make It Right Foundation, Brad Pitt's project to build exciting new "designer" homes in the Lower9th Ward. Celebrities have come to raise money. Star architects have come to offer their designs. In ten years, over 90 homes have been built.

But over 300,000 homes were destroyed by Katrina, almost 100,000 of them in New Orleans. At the current rate, Make It Right would fully rebuild New Orleans in a little over 1,000 years.

More troubling is the precedent the Make It Right homes have set. Let us not learn from the successes and delights of New Orleans itself, they suggest. Let us not empower local people with local solutions. Instead, let us bring international architects to craft novelty inventions, and bestow them upon these lucky denizens. If these novelties happen to perform poorly—if they rot quickly, or have other problems—well, who knew?

New Orleans knew, if we wanted to look at the lessons of time and climate and locality. (Indeed, the Make It Right Foundation's Wikipedia entry notes that "the foundation has been under fire in recent months after over two dozen of the green homes built started exhibiting severe signs of rotting.") New Orleans has the embodied knowledge of how to make an exquisite street, a delightful house, a durable and enduring piece of the city. But we are in the bad habit of ignoring it.

The heroic top-down model is failing us, whether employed by architects, movie stars, or government agencies, because it cannot solve the last-mile problem. It cannot tap into the genetic repository of solutions and small adaptive actions that form the self-organization of a great city.

But just as wise top-down actions can support the bottom-up evolution of a great city, so unwise top-down actions can thwart it, and even destroy it. New Orleans, like many great American cities, was already in deep trouble well before Hurricane Katrina. It was in trouble because it failed to recognize that an equitable urbanism—an equitable geography—Is not only more just, it is good for the bottom line.

During the first half of the 20th century, New Orleans offered a growing employment base around shipping, and thousands of people—many of them African-Americans—migrated to the city to work on the docks. At its peak, the city population was over 600,000. But by the 1960s, containerization had put an end to many of those jobs, and the population shrank for the first time.

Around that same time, suburban "white flight"—aided and abetted by government policy—accelerated the population decline, so that by the time of Hurricane Katrina, the city population was well below 500,000. Many of the remaining residents were poor and African-American. To the legacy of racial discrimination they could now add crumbling infrastructure and lack of access to viable urban resources and opportunities.

This is what, in New Orleans, we came to call "the hurricane before the hurricane"—the slow disaster that is 20th century urban planning, as it existed in New Orleans. This is what has happened to so many U.S. cities like Detroit, Baltimore, Cleveland—and so many other suburbs. This phenomenon is sometimes termed "shrinking cities"—but that is a misnomer. The New Orleans metro area is over 1.1 million, almost double the city's population at its peak. The city has not shrunk, but rather, has turned into a kind of demographic donut—emptying at the core, and thickening at the edges.

The same pattern exists in Detroit, the poster child of "shrinking cities"—and so many other American cities that have devastated cores. These conditions were not overtly planned, but the actions that brought them about certainly were: GI mortgages, the Interstate Freeway System, mortgage redlining, functional zoning, bad transportation planning, and a host of other policies, incentives, and disincentives.

The fact is, in focusing simple-mindedly on the physical plan, we have taken our eye off the real ball: the larger system of incentives, disincentives, laws, rules, codes, standards, and models that constitute our urban "operating system for growth." And we have failed to recognize that the system that produced New Orleans' urban disaster—the hurricane before the hurricane—is still with us.

That system is built on an old, mechanical, top-down idea of the city. It sees the city as a collection of parts that need to be rationally segregated and then recombined to work properly—just as the parts of an engine are rationally segregated by function. If there is something wrong in one of these segregated compartments, the way to fix it is to target that spot with some additional top-down resources. If poor people lack housing, build some new housing for them. If a street is too congested with traffic, build a wider road or even a freeway.

One of the fiercest and most effective critics of this idea of cities was the great urbanist Jane Jacobs. In place of rationally segregated compartments, sorted out by use and type and population, she was a champion of diversity and connectivity right across the city. She noted that cities are indeed self-organizing systems, and they thrive on the kind of pervasive connectivity that can be seen in a city like Manhattan, with its highly interconnected grid of streets and sidewalks.

There was an important economic dimension to Jacobs's ideas too. It was not a coincidence that a city like New York could take penniless immigrants from Russia and Italy and Ireland, and convert them in due course into middle-class shop owners and executives and creative people of all types. The city functioned as a connective human-development system, allowing them to connect to ideas, knowledge, and opportunity and benefit from a kind of metabolic process.

The best way to get a city working well, she said, is to get all its parts connected and interacting in balance, neither letting some areas get over-concentrated with growth and wealth, nor others depleted of activity and left in poverty. For the poorest areas, the best thing that can happen is "unslumming"—allowing residents to contribute more to the overall economy, instead of creating a burden on it. Increasing wealth, if it increases diversity, is a good thing. A monoculture of poverty is not a good thing.

Similarly, for wealthier areas, gentrification past a certain point also becomes unhealthy, if it means the loss of diversity and a monoculture of wealth. It then becomes important to decant some growth to other areas of the city where it can be more beneficial. It is a bad thing for the entire city to allow these wealthier areas to continue hypertrophic growth—just as it is to allow poorer areas to remain mired in poverty.

In time Jacobs's ideas have come to be taken up by many others, including prominent economists and complexity scientists. She was right, they agree, that cities are complex adaptive systems that self-organize, and our challenge is to understand this "kind of problem a city is" as she put it, and to manage its "organized complexity." If we fail to understand or to manage cities in this way, we are likely to make all kinds of mistakes that are devastating to the people who live in them.

And so we have. We have continued to mistreat cities, and to allow them to under-perform. We get hyper-performance not from the natural dynamics of cities but from artificial (and unsustainable) injections of resources, into artificial urban environments. Since World War II, we have built an enormous new world of sprawl, requiring massive (and massively profitable) consumption of fossil fuels and other finite resources. We have used this resource Ponzi Scheme to prop ourselves up—and now it is a global process. The institutions that are "too big to fail" grow ever bigger, until now they are entire nations: China, the United States...

Sprawl and an emptying city core produced a 'hurricane before the hurricane' in New Orleans and the Gulf Coast. Blue FEMA tarps show major damage to relatively recent projects along a strip highway in coastal Mississippi. (Image: Michael Mehaffy)

So there is much more at stake than whether New Orleans rebuilds—or whether the other devastated areas of the Gulf Coast rebuild. There is more at stake than any one disaster, be it Sandy, Rita, Fukushima, or any of a thousand other calamities. What is at stake is whether we can understand and support the vitality of cities, their capacity to promote resilience, ecological performance, and sustainable prosperity. New Orleans is not just one place, but an exemplar of what is happening to us as a species and the question of whether we will develop the models and tools to change our "operating system for growth," as we surely must to avert disaster.

In a real sense, we are all stranded in New Orleans.

Michael Mehaffy is Executive Director of Sustasis Foundation, an NGO created in New Orleans to take forward recovery planning work in 2007. He is also Chair of the College of Chapters of the International Network for Traditional Building, Architecture and Urbanism, based in London. He served with three different recovery planning teams organized by the Congress for the New Urbanism, Duany Plater-Zyberk and Company, and the Unified New Orleans Plan. 

Aerial view of homes on green hillsides in Daly City, California.

Depopulation Patterns Get Weird

A recent ranking of “declining” cities heavily features some of the most expensive cities in the country — including New York City and a half-dozen in the San Francisco Bay Area.

April 10, 2024 - California Planning & Development Report

Aerial view of Oakland, California with bay in background

California Exodus: Population Drops Below 39 Million

Never mind the 40 million that demographers predicted the Golden State would reach by 2018. The state's population dipped below 39 million to 38.965 million last July, according to Census data released in March, the lowest since 2015.

April 11, 2024 - Los Angeles Times

A view straight down LaSalle Street, lined by high-rise buildings with an El line running horizontally over the street.

Chicago to Turn High-Rise Offices into Housing

Four commercial buildings in the Chicago Loop have been approved for redevelopment into housing in a bid to revitalize the city’s downtown post-pandemic.

April 10, 2024 - Chicago Construction News

Woman with long hair wearing Covid mask sitting on underground train station bench looking at her watch as subway train approaches in background at Hollywood/Western station in Los Angeles, California.

How California Transit Agencies are Addressing Rider Harassment

Safety and harassment are commonly cited reasons passengers, particularly women and girls, avoid public transit.

April 17 - The American Prospect

Nighttime view of wildfire in Los Angeles hills.

Significant Investments Needed to Protect LA County Residents From Climate Hazards

A new study estimates that LA County must invest billions of dollars before 2040 to protect residents from extreme heat, increasing precipitation, worsening wildfires, rising sea levels, and climate-induced public health threats.

April 17 - Los Angeles Times

Bird's eye view of oil field in New Mexico desert.

Federal Rule Raises Cost for Oil and Gas Extraction on Public Lands

An update to federal regulations raises minimum bonding to limit orphaned wells and ensure cleanup costs are covered — but it still may not be enough to mitigate the damages caused by oil and gas drilling.

April 17 - High Country News

News from HUD User

HUD's Office of Policy Development and Research

Call for Speakers

Mpact Transit + Community

New Updates on PD&R Edge

HUD's Office of Policy Development and Research

Write for Planetizen

Urban Design for Planners 1: Software Tools

This six-course series explores essential urban design concepts using open source software and equips planners with the tools they need to participate fully in the urban design process.

Planning for Universal Design

Learn the tools for implementing Universal Design in planning regulations.