Emergent Urbanism

Rediscovering Urban Complexity

The Meaning of Emergent Urbanism, after A New Kind of Science

Stephen Wolfram is celebrating the tenth anniversary of the publication of A New Kind of Science, a milestone in the development of complexity science that is more significant than any other for me, as it was reading through that book in 2007 that gave me the motivation and the sense of purpose to begin writing about urbanism and complexity science.

Although I had taken a personal interest in urbanism as an undergraduate student of economics and computer science, mostly out of a sense of outrage at its condition as a practice whether technical, artistic or political, I was hard pressed to explain to my entourage my decision to go to graduate school for an urban planning degree. It seemed utterly disconnected from anything I had been involved in previously, and offered dire career prospects as well (in my defense, this time period is described by famous technologists as the ”nuclear winter” of the industry, so my planned career had already been derailed). I was following an intuition: that the texts I was reading by thinkers such as Christopher Alexander, Nikos Salingaros and Jane Jacobs, combined with the economic theories of outcasts like F.A. Hayek, were deeply meaningful and not just the amusements of intellectuals engaging in discourse for its own sake. I thought that they were revolutionary, but in what sense I couldn’t understand, much less articulate.

Wolfram allowed me to understand, yet it has taken me another five years to articulate why, and reading his series of blogs on the legacy of his book crystallized my ideas about the purpose of this website.

The new kind of urban science

Back in the 1950's, when Jane Jacobs was writing Death and Life of Great American Cities, she arrived at a conclusion that was a precursor for Wolfram’s scientific treatise. She claimed that city planning was a problem of organized complexity, and that it should be done by starting with the particulars and observing their behavior, instead of trying to control for a single variable useful in mathematical formulas. Wolfram discovered the same idea by observing not cities but extremely simple computer programs. He found that some of these elementary cellular automata showed emergent patterns so complex in their behavior that mathematics were useless in analyzing them. (Wolfram was the creator of the mathematics software suite Mathematica, so he speaks on greater authority than anyone in the world.) Despite that, he found that by using computers to synthesize, instead of analyzing, all sorts of simple programs, he could find programs whose rules generated complex patterns similar to the natural world.

Wolfram thus arrived at his methodology for complexity science: instead of analyzing the natural world to arrive at knowledge, one could search through the ”computational universe” for patterns found in the natural world, and know their rules immediately without necessarily understanding their behavior, which for complex systems defies analysis.

Wolfram proposes ”mining” the computational universe for scientific facts and technological ideas, and this is precisely what I’ve tried to do with every posting on Emergent Urbanism.

Recovering past urban science

Wolfram’s thesis is not only meaningful for future research, but also essential to reviving traditions that were cast aside generations ago for not meeting scientific rigor. Urban planning, beginning in the 19th century, was overwhelmingly statistical. Cities were reduced to variables that prescribed the average apartment block or suburban lot for the average family, and the force of scientific authority was used to coerce everyone to accept this plan as standing for the common good. What I found when I started searching for patterns of traditional urbanism that had common features and properties with advanced computations was that traditional urbanisms were in many aspects more sophisticated than modern ones, only they had been adopted by trial and error and, not being understood, could not be adapted to modernity’s scale. My first postings therefore sought to legitimize traditional urbanism using my scientific intuitions.

Designing a new kind of urbanism

Understanding the nature of our past success certainly is good for our existential confusion, but it is pointless unless this understanding allows us to tackle modern challenges. As such, the second mission of Emergent Urbanism is to use computational patterns to tackle contemporary problems, for example with the complex grid model. It does not prescribe any specific plan of action for any particular place, but it provides the inspiration for any city on what to do next to improve itself.

One of the conclusions reached by Wolfram is that human behavior, despite all of its complexity, is no different than any other simple computation that shows complex emergence, and as such if we choose to act within the same rules as the computer models we find, we will produce the same complexity. Searching for models of complexity and applying them is how we turn the new kind of science into a new kind of urbanism.

Sources of patterns

Complexity science offers us a plethora of new patterns to study. I’ve used fractal geometry to show the principles of chaotic beauty in urban architecture, or scale-free movement in urban grids. Fractal analysis is also meaningful to identify when we are witnessing a complex or simple system, but it does not help us generate complexity.

Digital computers are not the only sources of computations. All of nature is a computation, according to Wolfram, so the actions of humans in the past are just as interesting to us if we know by what rules they were acting. This makes the study of urban code of laws, and the morphology that resulted from them, another useful inspiration.

And lastly, the web itself, where I now have a busy career, provides a source of knowledge by merging hard computational rules with human interaction. Fascinating systems like Wikipedia offer us an experimental window into new human social creations.

A new kind of urbanism, to fix the other kind

Most of the urban countercurrent of the past generation has been focused on regenerating traditional towns and cities from the devastation of modern planning. That made sense for the reason that the old processes were easy to recover, and the places still generated a strong enough attachment to be worth fighting for. While I am endlessly grateful to all of those who labored to save traditional cities, this is not my focus. Having grown up and escaped from the suburban fringe, possibly the deadest place suburbia has to offer, I know that there are many generations who, for social and economic reasons, will continue to suffer from the choices of modern urban planning. I want Emergent Urbanism to be a guidebook on how to gradually recover these places and make them as complex as the traditional cities we love, by changing one tiny detail at a time. Doing so requires first understanding the aspects that made traditional cities wonderful (a description of their features is not enough). When traditional planning was attempted in the modern system, the result was uninspired and blamed on all sorts of factors.

The malady of modern planning is actually simple to diagnose: everything is frozen at enormous scale. The other legacy, the chaotic spread of shantytowns, is a side-effect of the loss of control bizarrely produced by the attempt to freeze growth at such enormous scale. Sprawl and shantytowns must be understood as the same problem. 

The cure for these places will come by improving one simple part at a time, whether it is the pattern of space or the pattern of building processes, to make it more simple in its definition but more emergent and more complex in its behavior. We will adopt the complexity of nature and computation as our model.

How to contribute

There are many people who have written to me asking in what way they could contribute to this cause. Many seem ensnared in the world of academia, with its conventions and expectations that root back to conventional science, which are not particularly useful in the case of doing new science. The best way to contribute, I advise you, is to participate in a change of processes in your country or community. This is what the BIMBY project in France has endeavored to do, and I have enthusiastically contributed my time and advice to help them succeed. If you are involved in such a program in your area, I would gladly give you support.

If you find yourself isolated, you can become a contributor to this website by producing a posting that utilizes the methodology I’ve described before. If it was first written for a different mode of publication, I will contribute the edits that ensure its success on the web.

I hope that this website will continue to empower those who feel something wrong with their world, but find themselves without the means to act about it. Believe that there are irrefutable facts in your defense, not merely opinions.

Comments

First of all, I really like the site! Good stuff all round.

On the topic of Jane Jacobs, you should look into the work of Geoffrey West (et al.) in understanding complexity in cities if you haven't already (eg/ http://www.youtube.com/watch?v=DFFVSvAr7Wc and http://blog.ted.com/2011/07/26/qa-with-geoffrey-west/). The presence of underlying laws that emerge as a result of the interactions occurring within a city is incredibly interesting.

On stage at TEDGlobal 2011, Geoffrey West talked about the universal mathematics that govern cities and corporations. Knowing only the population of a city, he can predict the number of patents, the crime rate, the average walking speed and many other features of a city.

This is from the TED interview, and I want to emphasize what kind of predictions he is trying to make. Traditional physics and science in general is analytical, it can take a large body of facts and reduce some essence or average from it, for example by taking all the mass of a planet moving in space and producing an orbital path for all of it. Then, using this essence, it can analyze the path to make a prediction of where the average will be later in time (computing the orbit of the planet). However, it can know nothing about any of the specific mass, only the average mass.

This may be interesting to a physicist, but it is not very useful complexity science. Wolfram points this out in his theory of complexity. If you start out with analytic tools (such as mathematical regressions and equations), you can only go looking for facts that are subject to analysis. By doing that, you have cast aside all of the interesting complexity of nature.

All of the results of analysis, however, cannot tell us what to do or how to build cities. To arrive at these facts, we need to do science the other way around, by doing synthesis and computation. This is what Jane Jacobs, Christopher Alexander and Stephen Wolfram do, and what some past economists such as Hayek did. It produces knowledge that looks weird and unscientific to traditional scientists, but it gives us much more clarity as to how things work and how we go about solving our present problems.

Now I have no problem with a complexity scientist predicting the average walking speed of a city, but my question for you is what am I supposed to do with this information? If I know that on average people will walk one block a minute, should I make blocks a bigger or smaller size? What about people who walk much slower, such as children and the elderly?

Enormous mistakes were made in 20th century urban planning by relying on average this or that to set baselines. Jane Jacobs warned us all about it.

Yeah, agreed although I'm not so sure that you cast aside all of the interesting complexity.
And yes, definitely a lot of it isn't useful in design and planning. A lot is simply interesting for it's own sake.

However, some is useful. For example, as density increases, so does the productivity. This would lend itself to planning in thinking about creating places that promote interaction, rather than separating things out (single home dwellings etc of urban sprawl). Also while these types of analytical equations can't explain why certain cities are different, it can identify which ones persistently perform better or worse than others, which can lead to insights into their differences on a planning level.

I personally just see it as interesting, and it certainly can't hurt in viewing things from many different perspectives (even if it isn't used directly in any way.)

mathieu-

i couldn't agree more with your thoughts on both west and batty. it is fine and neat that you can predict crime rates and walking speed from population. similarly if you know the structure of hydrogen and oxygen you can predict H20, but you will know nothing about the water. population will not tell you anything about how satisfying it is to be in a place, but knowing block size and shape might; knowing how a park is bounded and knitted to it's surrounding structures might; knowing the distribution of distances to a bar might.

batty suffers from almost an opposite problem. he's abstracted away everything that distinguishes a city system. yes, diffusion aggregated growth of crystals creates a structure reminiscent of a city. but this tells us desperately little about the internal behavior or external relation of an urban system.

i love this site and look forward to its continued development.

Do you have an RSS feed for this website? If so, I'd follow it. Great site.

A feed is available at http://emergenturbanism.com/rss.xml

There are two reasons I have faith in humanity and our country. #1 Mothers It is a constant that through the ages has saved most of us from catching pneumonia and breaking our necks. #2. People like you. Instead of bashing the same 5 ideas together over and over you take a couple of steps back and look toward the resources all around.
Pioneers each and every one. What beautiful minds.
Karla

I've been keeping your blog as a personal reference in the last two years and I'd like to congratulate and to thank you. It seems we have close interests... I was wondering if you know "The Self-Organizing Universe" by Eric Jantsch. I was able to find just a pdf version of it 'cause it is a really expensive book. Thanks again, I will go through your posts again.

Haven't read it, but it's going on my list.

Further comment

Please send your comments by email at mthl@mthl.info, or find me on Twitter @mathieuhelie. The commenting system is closed at the moment as no measures can hold back blog spamming bots.

Loading