1497977733-featured.jpeg
Dave Snowden

Liminal Cynefin

RSS Feed

The Cynefin framework tends to be used in the main to categories situations within the four main domains (and for more advanced use all five). I’ve never had a problem with that but the dynamics are as important as the domains and they are too often neglected. A concerned effort over the last year in presentations and teaching has resulted in a higher utilisation so it’s time to move on the develop the boundary zones. I should make it clear that this is an additional layer to Cynefin so it does not need to be used per se. This is not complete and I may switch the representation as i develop it.

I’ve called it the liminal version and thought for some time before committing to that. Various disciplines use the idea of the liminal on various ways. So to be clear I am using the word in the sense of its original latin root, namely a threshold state; a sense of being in a state of tension during transition. Cynefin assumes a phase shift between the ontological states of order, complexity and chaos, unlike some people in the field I don’t see it as a spectrum or gradient, but a clear boundary.

The way I am currently drawing it is with a narrow (red) and a broad (green) liminal zone around the boundary. The narrow boundary implies a definitive shift, a commitment which made will be difficult to undo. The broad boundary implies an ability, albeit at a cost, to maintain a threshold state before commitment, more than holding your options open, the idea of liminality here is more fluid in nature. Indeed the broad boundaries could be considered an extension of transitionary disorder, a state of change in which the ontological shift has not yet been made; options implies the change has been done.

So we have a broad zone as things move from complex to complicated. The domain as things emerge from the complex space and start to exhibit linear causality is important. When you make the transition successfully then constraints will provide predictable outcomes, you have linear causality. That is valuable as scalability can be efficient, ordered and robust in nature. But assuming that transition too early is deeply problematic hence holding the threshold state until you are certain. So in the complex domain you do parallel safe-to-fail experiments and as you move over the boundary (a choice) you start linear iteration to check you have it right. But you need to be ready to reverse with either minor variation.

The same applies in reverse moving out of complicated into obvious. That shift is a major commitment to robustness and an assumption that failure is implausible, you can afford the energy expenditure required for rigid order and predictability. So you hold for a long period of time before you make the final abrupt transition. If you go too soon then the switch to chaotic failure, induced by complacency and an inability to detect the weak signals of legitimate deviancy results in catastrophic failure.

Finally relaxing constraints for innovation (complex to chaotic) is risky and needs containment and more energy that the more experimental emergence possible within complexity. You may need to do it, but once you make the decision you are committed.

Linking this more with disorder is key so expect new versions over the next week or so.

  • janicefingler

    I really like this addition of liminality – it feels like a good representation for evolving systems. At first glance it reminded me of those windspinners that catch the wind from which ever direction it blows !

    As a visual person, I also envision the domains like flat paint trays with colours that tip dynamically in different directions (maybe yellow – complex, red – complicated, blue – obvious) and maybe black for chaos – that as we tip them with the boundaries transitions becoming red – yellow or yellow – red “orangish” depending on the direction of vector. Is there a very narrow 50:50 state – in an evolving system ? Moving from blue to black (order to chaos) “cliff” is still very dark but would the yellow to black transition be lighter ? Or the chaos paint tray is deeper near obvious transtion (like an actual paint tray). Using colours, I can map project tasks, experiments or projects in a portfolio as being collections of drops of paint – each of which can be combined into a larger paint tray mosaic – fractals !

    From a complexity and cynefin standpoint – what might be the limitations to my visual ?

    • janicefingler

      Now that think about safety practices – if people experiment with pouring from their paint can (as in complexity), they will be more used to the shifting weight of the can and other factors. ie adaptive. If people don’t handle the paint can or others do it for them (as in obvious), they risk spilling the can too quickly – and catastrophically. Minor spills might not be minor after all. Drift to failure ?

      • Dave Snowden

        I’d go more with old dried out paint in obvious, congealed paint in complicated and fluid paint in complex with chaos either the absence of paint or a spray can running loose without control

        • janicefingler

          I think i understand better with that metaphor as it preserves the fundamental nature of constraints for each domain. A colour scheme would only convey vector and dynamics. Thanks !

  • Bhav Patel

    You wrote – “So we have a broad zone as things move from complex to complicated.” That makes me think that the blue zone should be on the other side of the red to allow for the broad zone and then a clear decision to shift into linear?

    • Dave Snowden

      The switch from parallel experiments to a pilot or small number of pilots is a big decision – hence the way it is shown but will think about it

  • Andrew Wagner

    ” The broad boundary implies an ability, albeit at a cost, to maintain a
    threshold state before commitment, more than holding your options open,
    the idea of liminality here is more fluid in nature.”
    This reminds me of the Set-Based Concurrent Engineering approach used by Toyota in their product development. This method emphasizes continued exploration of new design choices until they prove unfeasible or impractical, rather than performing the concept down-select early and committing fully in the presence of uncertainty.
    http://sloanreview.mit.edu/article/toyotas-principles-of-setbased-concurrent-engineering/

  • janicefingler

    One thing I’m not sure about is if transitions between complex and obvious are possible, without moving to chaos or complicated. And if they are, what they look like? I think about typical mineral exploration activities (complex) in cultures/countries that are very hierarchal (obvious). It’s possible to get work done, but holding those together in the project space requires a lot more time and money. We’re usually aware of the disconnect, though.

    • Dave Snowden

      Not desirable and very high risk. But a lot of BPR and Sick Stigma exercises do it

  • Mark Anderson

    Throughout the blog you have written the word “you” frequently. Who is (or are) the “you” in this context? Are you possibly committing the act that Ralph Stacy rails against of assuming there is a “you” (singular or collective) who somehow is outside of the world represented by the model and in control? I have always thought the model was not just about dynamics but also deeply social. Each social use of the model to make sense of the dynamics constructs the model anew.

    • Dave Snowden

      As in the normal use of the world in English, namely any actor in the system. Ralph tends to rail against many things including anyone who doesn’t adopt his somewhat absolutists positions so I wouldn’t worry about it (and don’t as one is in good company if a target)

Top