Emergent Behavior

One of the highlights of the last three months was reading Complexity, by Michael Waldrop. I highly recommend it. Reading about scientists grappling with the same things I’m fascinated by (and for the same reasons) felt like coming home. These scientists are intrigued by “complex, adaptive systems” and the dynamic way by which they learn, evolve, adapt. This science gave me words to crystallize concepts which had allured me but which had been too ungainly to manipulate verbally. One such concept was “emergent behavior”. But as I did in Shifting, let me start with two stories before I wax philosophical about Emergent Behavior.

The campfire circle at a national park I worked at was a series of benches. Beneath each bench was a ridge of loose soil. All the benches had the same uniform shaped ridge in precisely the same location. Someone had to have been playing in the dirt? But who? And why? I didn’t know the answer until the night I sat in the audience to watch another program. I stretched my legs out in front of me; my feet pushed across the ground and came to rest at the ridge ahead of me. Later, I pulled my feet back underneath my bench – and they came to rest next to the ridge beneath me. Back and forth moved my legs (and the legs of all others in the audience). Any soil in the back and forth area got pushed back and forth. But beneath the bench was a “safe zone”, out of reach of both one’s own drawn back feet and the stretched forward feet from the bench behind. Here is where all the dirt unconsciously pushed and pulled by feet gradually accumulated. These ridges are not built at the level of individual consciousness. Their formation emerges from the collective behavior of individuals.

My friend Slevin tells great stories. Several involve his summer work as an electric meter reader in LA during his college days. Meter readers were evaluated and ranked by two criteria. The first was mathematical accuracy. (This was before hand-held computers.) If a meter reader subtracted incorrectly, the customer would be upset. The second criteria was “lock-outs”. Some meters are, for a variety of reasons, behind locked doors. The company tried to get a key from the property owner but not all owners consented. So a meter reader was occasionally locked out from reading a meter. This posed two problems. First, the utility company had to pay someone additional to drive all over the city trying to read one meter here, another meter over there. Very inefficient. The second problem was that a meter reader who was tempted to finish up early could simply write “Lock Out” for some of the more time-consuming meters. To guard against that, the company explicitly and publicly ranked the meter readers by number of “lock outs” they accrued during the week.

At this level, the reasoning is logical. It all makes sense. Nothing sinister. However, from this set of initial conditions emerged the reality that the meter readers taught each other how to break into houses to avoid “lock outs.” Though breaking and entering wasn’t official company policy, it was an unofficial part of your training. Many of Slevin’s adventures as a meter reader began, as you can imagine, as he broke into a house.

An example from Shifting is how the behavior of runoff emerges from the condition of the slopes within a watershed. A healthy slope absorbs the vast majority of precipitation and leads it onto a route of nourishing plant growth and soil development that allows the slopes, over time, to absorb even more rain. But an eroding watershed will allow most of the precipitation to run off, depriving the slopes of the moisture they need to sustain strong plant growth and develop resilient, absorbent soil. Streams in such a watershed tend to be “flashy” and erosive.

Emergence is basically cause and effect – in the real world. Cause and effect does not stop at effect. The effect causes other effects. Not only does the sequence of cause and effect go on and on, it also spreads. A certain cause can precipitate many effects and many causes can converge to create a distinctive effect. This rippling and spreading of cause and effects can loop back on itself. A causes B which causes C which causes A. In Shifting, I poetically called these loops “spirals of cause and effect” (such as the spiral between the soil on a slope, the vegetation growing in that soil, the behavior of water falling upon that slope, and the behavior of the stream downslope). Science usually calls them “feedback loops”. An important discovery by the sciences of chaos and complexity is that feedback loops contains the power to make the future unpredictable. Feedback can create consequences as strong, sudden and unexpected as the screech of a microphone. Feedback can conversely minimize fluctuations. Feedback is part of the nature of the universe.

The science of complexity finds the source of complexity residing not so much in complex structures or complex behaviors but in complex patterns of interactions. With that introduction, here are two quotes about emergence from the book, Complexity.

“Every topic of interest had at its heart a system composed of many, many “agents”. These agents might be molecules or neurons or species or consumers or even corporations. But whatever their nature, the agents were constantly organizing and reorganizing themselves into larger structures through the clash of mutual accommodation and mutual rivalry. Thus, molecules would form cells, neurons would form brains, species would form ecosystems, consumers and corporations would form economies, and so on. At each level, new emergent structures would form and engage in new emergent behaviors. Complexity, in other words, was really a science of emergence. And the challenge was to find the fundamental laws of emergence.”

“What captivated him wasn’t that science allowed you to reduce everything in the universe to a few simple laws. It was just the opposite; that science showed you how a few simple laws could produce the enormously rich behavior of the world. Science and math are the ultimate in reduction in one sense. But if you turn them on their heads, and look at the synthetic aspects, the possibilities for surprise are just unending. It’s a way of making the universe comprehensible at one end and forever incomprehensible at the other end.”

A classic image for emergent behavior is ant colony behavior. The colony exhibits adaptive behavior and yet the locus of this behavior does not reside in any ant. The adaptive behavior emerges from behaviors between ants that are fundamentally simple but which are occurring millions of times in richly interacting ways. One of the reasons I am grateful for this concept of emergence is that many of the problems we are grappling with as a culture are emergent behaviors that operate on a collective level higher than the individual. I think of these problems as “ant colony problems” and we are ” the ants”. Emergent behavior can’t be controlled the way we try to control individual behavior. It is more diffuse, more accumulative, and more unpredictable than that. How do we ants change either the simple behaviors or the complex pattern through which these simple behaviors interact in order to alter behaviors that jeopardize the long-term health of the colony?

Grappling with this challenging question is often frustrating; it can lead one into dark thoughts. For example, if I did not understand what created those ridges under the benches, I could easily be led to the conclusion that someone consciously created those ridges. The fact that they are found uniformly under all the benches makes me think the person is devoting an inordinate amount of their life energy to create something a bit weird which leads my imagination into areas far removed from people simply shuffling their feet back and forth. In exactly the same way, people can look at developments in our world and be easily led to conspiracy theories. (Weird people constructing weird structures to influence normal people visiting America’s National Parks. Utility companies conspiring to invade our homes.)

The reason conspiracy theories are so prevalent is because emergent behaviors are very life-like; they exhibit the dynamics of consciousness (just as an ant colony does). And yet, this behavior occurs too consistently or on too large of scale to be either chance or the work of one person. It must be a conspiracy. But it isn’t. It is consequences emerging from a level of behavior that might have no conscious awareness of its role in creating these consequences. So one reason to teach about emergent behavior is to reduce the amount of psychic energy flowing into conspiracy theories and paranoia. The concept of emergence helps us soar above the level of the individual and have a larger point of view.

So how do we “ants” alter emergent behaviors that threaten the long-term viability of our “colony”? One of the quotes from Complexity that has wormed its way deepest into my current thinking describes the way scientists of complexity have learned to computer simulate emergent behaviors. I view it also as a prescription for how to alter it.

“The way to achieve lifelike behavior is to simulate populations of simple units instead of one big complex unit. Use local control instead of global control. Let the behavior emerge from the bottom up, instead of being specified from the top down. And while you’re at it, focus on ongoing behavior instead of the final result. As Holland loved to point out, living systems never really settle down.”

I internalized this book right away because the scientists described in the book are grappling with the same issues I’m interested in. “Bottom up” means for them the same thing that “head of the watershed” came to mean to me. Nourishing the invisible power of individual grass blades contains the same ideas as “simple units” and “local control” and concentrating on the behaviors of these things upslope rather than on the resulting erosion emerging finally downstream.

One of my current projects that is helping me deepen my understanding of this quote is working on the charter for Chrysalis. (Chrysalis is a charter school we are trying to create which will have nature study as its curricular focus. Chrysalis will hopefully be the main topic in one of the next two or three Cairns.) Trying to define and establish our educational vision through a charter has sensitized us to the prevalence of emergent behaviors, unexpected consequences deriving from simple but richly interacting causes.

Public education, for example, can richly benefit from the prescription, “focus on ongoing behavior instead of the final result”. Public education is filled with people spending time and money creating documents that focus attention on “final results”. “Outcomes”. “Frameworks.” “National Standards”. Educational reform is dominated by such debates. And yet very little attention is directed to analyzing the structural barriers that channel “ongoing behavior” into unintended emergent behavior. For example, we have discovered that a significant barrier to educational reform is the salary schedule. The schedule is very simple; seems reasonable. It seems innocuous – and yet these simple rules interact with the rich complexity of educational administration to produce debilitating restrictions in unexpected locations throughout the system. In our charter work, we are trying to shift the attention from final results (which always sounds fine until you encounter the individuality of each student) to ongoing behavior. In fact, the ideas in Complexity have deepened our charter so much that we now refer to Chrysalis as an Emergent School.