Cynthia Kurtz, Story Colored Glasses, 9/3/10
This is an interesting and insightful blog entry (although somewhat long). Cynthia Kurtz describes herself as a “researcher, writer, and programmer who works on the ‘listening side’ of organizational and community narrative.”
In this essay, she looks at the subject of complexity through the lens of narrative, and apologizes to her readers. “Since I posted some thoughts about complexity a while back, I've been surprised both by the number of people who have been interested in what I said (and encouraged me to write more), and by the degree to which I find myself wanting to write more. I still have much to write about narrative, but I also seem to want to write more about complexity as it relates to sensemaking and decision support. Since this blog is supposed to be about narrative, I keep feeling a need to apologize whenever I write about complexity.
But really there is nothing to be sorry for. My work on the "listening side" of narrative centers on the place where stories and patterns come together. Narrative incorporates complexity because stories self-organize into emergent patterns, and complexity incorporates narrative because complex systems are historical systems. I remember when I first realized this, and what a rush it was to discover synergy between two fields I had come to love. Thus, dear reader, henceforth I resolve to stop apologizing for combining these topics!”
The metaphor (story) about the butterfly in Brazil and the tornado in Texas came from Lorenz. “Lorenz's first talk on the topic in 1972 was titled ‘Predictability: Does the flap of a butterfly's wings in Brazil set off a tornado in Texas?’ His answer was not ‘yes’ but an emphatic ‘impossible to say’.” Kurtz shows how this correct interpretation of the impact of minor perturbations in the weather morphed into predictability, linking cause and effect.
She gives a selection of excerpts and then points out, “These excerpts, and almost all popular and business interpretations of the butterfly effect, transform a statement about uncertainty to one about certainty. The second story changes the butterfly effect from tiny actions piling up in unpredictable ways to tiny actions having predictable and controllable impacts. I have taken to calling this second story the "underbutterfly effect," because the butterfly in these versions is a powerful underdog who changes the world, not one of millions of other butterflies (not to mention innumerable more powerful creatures) whose feeble flaps are lost in the sea of uncertainty in which we live.”
Kurtz examines two other narratives that have morphed into stories we want to believe in. We changed the story of the butterfly because we can’t seem to accept that a large part of our world exists without cause and effect in play. Causality is deeply ingrained in our world view.
“In 1969 Robert T. Paine introduced the idea of a keystone species to ecology”, she writes. “Paine discovered this phenomenon in an experiment during which he removed a single species of sea star from a small area of shoreline and found that it had far-reaching effects on species diversity. Most importantly, the effect produced by its removal was out of proportion to its relative abundance in the community.” Like the butterfly narrative, this story morphed driven by our desire for certainty. The keystone became what she calls a topstone. “As with the underbutterfly story, the topstone story began to surface soon after Paine started publishing papers about his concept. What seems to have happened is that people involved in wildlife conservation started trying to identify keystone species in order to wisely use limited conservation budgets. Political, cultural and special-interest complexities joined the mix, and the keystone species concept widened and weakened as a result. For environmental study, retrospective discovery might suffice, but for environmental action, people wanted certainty.”
Her last narrative is about the adaptive landscape. “I first encountered Sewall Wright's adaptive landscape metaphor in college, and as a visual thinker I found it useful right away. Even though the metaphor is severely limited, many evolutionary theorists use it as a visual shorthand for thinking about genetic change. The way the metaphor works is this. Populations are located in X and Y dimensions to describe their genetic makeup (in a radically simplified way). They exist on a landscape where the height of each point describes the fitness of the population with that particular XY combination of genetic variables.” By describing fitness as a peak, the metaphor conveyed a lot of uncertainty. A small nudge could send the species down in any direction into a valley. Physicists tend to look at this type of metaphor in the opposite direction where potential energy increase in the peaks and valleys are lower potential energy.
I have to admit that with my physics background I had trouble with section of her discussion because when I first saw the metaphor, I thought to my self that it was upside down. The trend would be for the species to always roll down the hill into the “genetic valley of death.”
Kurtz writes, “One more and then I'm really done, I promise. This is from The Quark and the Jaguar by Murray Gell-Mann:
Biologists conventionally represent fitness as increasing with increasing height, so that maxima of fitness correspond to the tops of hills and minima to the bottom of pits; however, I shall use the reverse convention, which is customary in many other fields, and turn the whole picture upside-down.
Gell-Mann then gets into the difficulties of this flip, seemingly without realizing it:
If the effect of evolution were always to move steadily downhill -- always to improve fitness -- then the genotype would be likely to get stuck at the bottom of a shallow depression and have no opportunity to reach the deep holes nearby that correspond to much greater fitness. At the very least, the genotype must be moving in a more complicated manner than just sliding downhill.
That is precisely the problem with flipping the landscape: that when you do so, things that are in reality very complicated seem simple, and certain.”
This desire to correct uncertainty runs strong in us. She writes, “If you think about it, this is not at all surprising. We have been conditioned since an early age to believe in this equation:
uncertainty + science = certainty
When we meet an equation like this:
uncertainty + science = more uncertainty
We react, and a second story arises. That can't be right. There must be another explanation. That's what Edward Lorenz said when his computer generated a new weather system based on what he thought were the same inputs. He called in the hardware engineers to find the broken vacuum tube.”
She summarizes her essay in a section titled “A Generation, or Two Or Three, to Sink In. “If people tell second stories about complexity because we aren't ready for the first stories, I think our children are ready. In many of my narrative projects I ask people to rate the predictability of events in stories -- it's useful to map perceptions of stability and instability across conceptual space. I've noticed a pattern across several projects that older people are more likely to associate instability with negative outcomes in stories. Younger people are more likely to mark stories as both unstable and positive.
I was thinking about all this the other day while playing with my son, and two things happened that gave me food for thought. The first was that we watched the movie Clifford's Really Big Movie. It's a great movie, and it's in our pantheon now and will probably be watched many more times. One of my favorite parts of the movie is this lovely song, which my six-year-old understood and liked immediately. It goes, in part:
You've gotta get lost if you wanna get found
Gotta wind up to get unwound
Things only look up from down below
And I can't come home until I go
It only gets better after it gets worse
Happy ever after needs a scary part first
You've gotta fall down to get back on
And I can't come home until I'm gone.”
Later, she writes, “According to my surely-biased reading, there are three ways the authors of business books about complexity and chaos provide reassurance to those in charge. One is to drain the power out of the major discoveries by highlighting the second stories, which do exist in science and so can be called scientific: the underbutterfly, the topstone, and the easy roller. The first stories can be waved away as "internal disputes" that don't matter.
The second method of making complexity palatable to those in charge is to make it seem magical. That is why the phrase "order for free" is so wildly attractive in these books, and why people love to throw around terms like "strange attractor" and "fitness landscape" and "coevolution" -- because they are magic words of power that seem to promise something for nothing. Even "nonlinear" effects are spoken of mainly for the idea of something small going in and something big (and beneficial to the reader) coming out. I see this rose-colored view of complexity in most (but not all) business writings about complexity. Consider for example the way people talk about coevolution: nearly every treatment of business coevolution I have read has talked about it like nothing can possibly go wrong. But in real coevolution things can and do go horribly wrong at times. It is these sorts of distortions that not only give complexity concepts a bad name (because rarely can such wild claims of magical power be justified) but also spread confusion about the true utility of complexity and chaos based approaches.
And finally, the third and most used tool in the business complexity writer's toolkit is that the sky is falling. You need this, say the business books, because the world has changed in such dramatic ways that you can't possibly survive without it. People who use this tool ramp up the fear quotient by making claims such as that "an organization is a complex adaptive system" -- the implication being, and you had better find out what that means, and quick. But organizations are not complex adaptive systems! More precisely, they are not only complex adaptive systems. An organization is a lot of people. Those people interact with each other in many ways, some of which are complex and emergent, and some of which are not. Organization and self-organization, hierarchy and meshwork are inextricably bound up together in organizations, and saying an organization "is" one without the other is sheer nonsense and is probably meant to entice rather than inform. There are no only-complex social groupings in human life. Every gathering of ten huts has a path through it. Every lunch meeting has a leader. Every subway car has a social structure, if even only for the two minutes the same people are in it. That's what we do. There may be such things as only-complex systems in the lives of social insects, but even there some hierarchy (in the form of central pheromonal control) is usually mixed in.”
She concludes with, “My advice, if anyone wanted it, would be this. First, stop saying everything is complex, and start talking about how complexity and hierarchy can work to mutual benefit.”
“When you don't fear complexity, when you see it as a part of reality but not a ‘whole new world’ dominated by a falling sky, you don't have to muzzle it. The underbutterfly and the topstone and the easy roller can go plague other people. You can take the butterfly and the keystone and the steadfast climber as they come to you, in stride. You can learn to recognize them, deal with them, work with them and even in time welcome them as old friends. You'll just know better than to hand over your car keys to them.”
As it turns out, this is the conclusion I had recently reached intuitively. She has done all the work that suggests that this is correct. I was drawing from my experience with physics and squaring Newtonian, Quantum and Relativistic physics. All are present in everything. It’s just in certain conditions one view works best. But you have to know when you are in what type of physics. To complicate matters, there are at least three different types of “messy” complexity and one type more tractable. For example in the behavior of markets there are trends, cycles and complexity. Complexity is in the fine structure of the behavior of the market.
Read Her Blog