Dave Snowden

Chaos & disorder

RSS Feed

I often get asked about the two transitionary domains in Cynefin and its not in frequent for people to assume they are the same sort of domain.   Now I can see how this happens but  its not the case and the reasons why are important.  At its most basic chaos is a legitimate state, engendered with control it has considerable utility as its the basis for distributed cognition, or Wisdom of the Crowds to use the more popular term.  If you can create unconstrained agents (remember I use the constraint based definition) then you have the pre-conditions to use those agents to help provide evidence under conditions of uncertainty.  Remember in Wisdom of Crowds all the agents need to guess the answer (or form a judgement) independently of each other, they must have tacit knowledge of the field and no significant personal stake in the outcome.  Chaos is also useful if entered with purpose as an innovation space, but entered accidentally it is a crisis, constraints and connectivity vanish there are no patterns.  I discussed this at some length with the chaos domain model some time ago and its a model I will be refining soon.

So Chaos fits with the the ideal of multi-ontology sense-making. Its a useful domain and we now how to behave through constraint management.  Disorder is very different.  The whole point of disorder is that we don’t know which domain we are in.  The situation could be complex, simple, complicated or chaotic.  So we don’t know what type of action we should take and fall back to that with which we feel most comfortable.  Over the years I have thought about renaming it.   Two options are still attractive:

  • Confusion would be an accurate name and I am sorely tempted by this one.  If you are in this domain your are ontologically confused which means you make inappropriate epistemic choices.  Most pre-complexity practice falls into this domain as the assumption of a single ontological approach (there is one solution) is of its nature confused and disordered.  You may accidentally get it right but accidental correctness is not a recipe for sustainability of resilience!  So I might still make that shift, but I have resisted it because the other alternative is also attractive, and that is …
  • Inauthentic which has a long history in various disciplines.  Slightly more academic, but more accurate as you are lack an authentic response.   Again I almost made this change, but then I realised that disorder is a legitimate transitionary domain.   When we talk about innovation for example this involves a transition through disorder.   So most recently I kept the Disorder name, but split it into authentic and transitionary.  Inauthentic means you should not be there, transitionary its a legitimate vector.  Someone (I keep forgetting who) suggested the metaphor of an open cast mine with lots of routes around the edges, but the lorries using those routes are in constant danger of falling over the edge.   I like that so its now part of the standard model.

That said I am still not sure its right and maybe confusion and transition would be the way forward.  Maybe draw the boundaries there in a different way; that has potential but I need to think about it a bit.  There is an older word for confusion that escapes me for the moment that might work better.

All ideas, questions and comments welcome

Top