This is a post in my series on organizing ”between and beyond.” Other posts are here. The purpose of this post is to reflect on subjects occupying my mind. I make no claim to fully believe what I write. Neither do I pretend that others have not already thought or written about the same subject. More often than not, I take up, combine, and add to already existing thoughts and ideas.
What is on my mind?
In today’s reflection I’m looking into the Cynefin complexity framework. Here is an interview where Dave Snowden, the creator of Cynefin, shares the philosophy underpinning his work. He talks about how people can apply his insights to leading and managing organizations.
Dave Snowden says among other things that (my emphasis in bold):1
We should manage the evolutionary potential of the present, rather than aiming for some idealized future state.
If you have a highly constrained environment, you can manage it through rules and objectives, because you’ve got predictability. … In a complex system, you have to manage in a different way. …
The great liberation of complexity science is that it gives you a base in science to say you’ve got a non-causal system. The minute you realize that systems can be non-causal, everything becomes simple … If you believe causality is a necessary condition, life becomes very, very, complicated.
There’s a basic difference between … an enabling constraint and a governing constraint. A governing constraint is context free … and an enabling constraint is context sensitive. … A governing constraint is a container. … Within this boundary you can do whatever you want. A fixed constraint says, this is the way you do it. No variation is permissible.
Excessive constraints actually produces deviant behavior. … Human beings will accept constraint. … One of the great things about humans is that we actually have constraints … like laws, and also things like acceptable forms of behavior, and rituals. … We like order. We are really good at it. There’s nothing wrong with it.
But there is a big difference in Cynefin between order which is self-evident, which everybody buys into, and order which could only be understood by experts. Obvious vs. complicated, best practice vs. good practice, fixed constraints vs. governing constraints.
Cynefin is a typology, not a taxonomy. Taxonomy puts things into rigid chategories. Typology says this is different perspectives, different ways of looking at it. Actually, cynefin is a mixture of both. … The primary division of ordinary, complex and chaotic is a taxonomy. … Within that there are different gradations and that’s typology.
The difference between the obvious and the complicated is basically a gradient, it’s not a rigid boundary. … The point is that there are right answers. … The boundary between obvious and chaotic is a catastrophic cliff … If you become complacent you restrain a system which shouldn’t be constrained because it will break catastrophically. …
Complex to complicated is when you stop doing your multiple safe-to-fail experiments. … You’ve come out of the mist, you know roughtly what to do, but you’ve not settled yet. … You kind of know where you’re going, then it becomes complicated.
The liminal domain to chaos is drawn as a closed space. It’s open on the other one, because that’s where you dip into chaos for innovation. Or, you dip into chaos for mass sensing. No agent is connected with anyother agent. … The issue is, if you enter into chaos accidently, it leads to disaster. If you enter into it deliberately, … it’s a good thing to do. …
If people are arguing about the details, that’s liminality. … We know this is probably right, but we don’t know how to do it yet. That’s liminal. … Liminality is a good concept, because it’s a state of transition. And the longer you hold it in a liminal state, the more reliable is what comes out of it. … You’ve got a tradeoff between speed and reliability.
You move technically from deductive to abductive logic. … Deductive, if A then B. Inductive, all the cases of A have B, therefore the likely association. Abductive is a logic of hunches, plausable connections between apparently unconnected things. …
Human beings have evolved to think abductively. … Human beings have evolved to make decisions collectively, not individually. … That’s our strength, we can cooperate. … If you can increase the number of people in the collective decision-cycle, you can make it more objective.
One of the dangers we got with the engineering approaches which came in the 80s is people try to get rid of human judgment. … One of the big things over the next two decades is human judgment. … Artificial intelligence … is the second existential threat to humanity after nuclear war. … Part of the problem is that we’re reducing human beings to following rigid processes …
Vector measures says am I going in the right direction, at the right speed, for the right effort. It doesn’t have a specific outcome. … It basically says I need to move in this direction, I need to shift in this direction at this pace. Am I doing it? … You still measure, but you measure appropriately.
Are you riding a wave of uncertainty, which means you have to have a sense of direction, and keep moving to maintain balance? Or are you in a highly stable position where you can say what you should achieve? Context is everything. … Always start from where people are, unless you can kill them and start from fresh, but that’s rare.1
Generative organizing is appropriate for riding waves of uncertainty. It relies on collective decision-making, abductive logic, and human judgment. Generative organizing is impossible if constraints are fixed.
1 #12 Managing in Complexity—Dave Snowden | Being Human, 2018-06-15 (accessed 2018-08-10).
Organizing in between and beyond posts