Attractors
Fragments on the shape of living systems
Continued existence of complex interactive systems depend upon preventing the maximization of any variable.
(Gregory Bateson quoting W. Ross Ashby)
Living systems are perhaps best characterized as systems that dynamically avoid attractors. (Chris Langton)
Attractors are states a system wants to fall into. “Attractor”, because the system is attracted toward certain outcomes, like a marble rolling around a in landscape of basins and channels.
If the marble rolls into a deep basin, it gets stuck. The system stops generating new states. It is in static equilibrium.
If it is in equilibrium, it must be dead. (John Holland)
The marvelous thing about nature is that it doesn’t have a fixed goal toward which it is moving. It moves in every direction at once, generating an ever-expanding diversity of ways to be alive.
…nature is a stepping stone collector, accumulating steps towards ever-more complicated novelties, marching onward onto the mist-cloaked lake of possible-life-forms, heading eternally both everywhere and nowhere in particular.
(Stanley, Lehman, 2015. Why Greatness Cannot Be Planned)
Evolution is never “finished”. It is open-ended. The landscape of nature keeps changing, raising up new hills, carving out new valleys, and life, a marble rolling along this dancing landscape.
This open-ended evolution turns out to be a challenging behavior to reproduce, but one key factor is the number of interacting variables.
One variable, and you get a possibility landscape with a single attractor. A basin.
Two or more variables, and you start to get a rugged landscape of hills and valleys.
Something interesting begins to happen when variables interact. As one variable influences the value of the other, the landscape begins to dance. The more interacting variables, the more the landscape dances.
Interacting how? Through coevolution! In nature, everybody interacts with everybody else, all evolving at once. Species compete and cooperate and create new niches, and change their goals, constantly altering the competitive landscape.
So, fixed landscapes generate fixed attractors. Dancing landscapes don’t.
I’m fascinated by this quality of open-endedness and how it is that some systems seem to avoid getting stuck in attractors.
Social media has some of this living quality, but I don’t think it is fully open-ended. The explicit objective of social media is to maximize a variable called "engagement”. An algorithmic feed will amplify whatever works, whatever is most virally contagious. This generates a lot of novelty, yet often seems to converge around fundamental attractors in our psyche—brainstem stuff like angry, cute, lol, wow, horny.
But it’s not just engagement. Maxing any single variable creates an attractor. Maximization, attraction are two ways of talking about the same thing. Maxing is asking the system to continue producing more of the same state. For open-endedness to emerge, you gotta want more than one thing.
There is no logic that can be superimposed on the city. People make it, and it is to them, not buildings, that we must fit our plans. (Jane Jacobs)
Yet we do create systems that are open-ended. Cities, markets, the internet, and the web all have some of this open-ended quality, at least within certain operating ranges.
These open-ended systems are where innovation comes from. In fact, innovation and open-endedness are synonymous. Open-endedness is the ability of a system to continue adapting, continue generating novelty, continue knocking itself into new states… to continue.
Only that which can change can continue. (James Carse)
None of these open-ended systems are “for” just one thing. There are many dreams lived out in a city, many different ways to make a buck, many different things you can build on the internet, many ways to be.
Ecosystems can't possibly exist for a particular purpose. (Hayao Miyazaki)