In the first of these articles, I looked at Borton’s W³ question from the 1970s and more recent adaptations (with varying degrees of acknowledgment of their source) by Driscoll, Eoyang and Liberating Structures. I’ve found all useful in various ways but I have concerns based on their heavy reliance of workshops (a general concern by the way to which I will return as it makes scaling very difficult and is too subject to facilitator bias), and what seems to be a linear approach – the question assumes a common context in most of the uses that I have seen or read reports of. In effect, the facilitator of the process starts with perceptions and attempts to move to knowledge. This is an incomplete reversal of the Cynefin perspective where ontology (the nature of what is) is meant to proceed epistemology (how we know things) and radically reduce the dependency on perception.
The nature of What, remembering the meaning of Wherefore is an ontological question, not a matter of shifting perception to agreed knowledge. However, most consultancy approaches start with perception and attempt to gain knowledge of the way forward. Now, this can work, but to my mind is lacks energy efficiency and over privileges the role of the facilitator. As in Rugby matches (sorry this is a sore point on the day I am authoring this) if you have an experienced referee all goes well; if you get an inexperienced one then angling from the side at a scrum goes unnoticed and they lack the confidence to give a penalty try for a deliberate knock-on. A good facilitator can usually get things to work well however and a lot of these methods sell based on the work of the originator – then things get industrialised and we start to get failure. To give two examples: David Cooperrider’s work on Appreciative Inquiry as and is excellent but the industrialisation into always look on the bright side of live platitudes is not; Clayton Christianson’s work on disruptive innovation was brilliant but the consultancy practice that arises less so.
So I decided to take a look at the W³ question through the eyes of the Cynefin framework and came up with the illustration to the right. I’m not sure it is right yet, so consider this a work in progress.
As I have already implied, to simply ask the W³ question without first getting the context is in effect sitting in disorder (or confused which is a current idea for renaming that I am toying with). This is shown as a red arrow. The suggestion is an alternative staged approach in which the What question related to the nature of the system before any move to the right is essayed.
- That means the first question is if the system is chaotic, which means there are no discernible pattern or direction and we need to act decisively to create constraints and see if we can get any stable pattern, If we can the question we have to ask is we maintain several patterns (the shift to complex) or focus on a single one (the authoritarian shift to Clear (which used to be Obvious, another experiment. Just to note this is not the same as entering chaos with deliberate purpose into its liminal state.
- If it isn’t chaotic then we test for complexity – are there multiple competing hypotheses about what is going on, all of which are supported by evidence, but the question of which is right or wrong cannot be determined in the time frame for a decision? If it is then we shift to the right, first mapping constraints and propensities then creating parallel safe-to-fail experiments to see if a clear pattern can emerge. We have a sense of direction in doing this (which may be defined by what we don’t want more than what we do) and as patterns stabilise toad the position becomes clear we ask if we can shift this, via the liminal aspect of the Complex-Complicated boundary onto something with high stability.
- If we can resolve that the situation has cause and effect which we can discover then its complicated and we do the appropriate analysis of what is happening, agree on clear goals and then create action plans to execute. Most consultancy processes start here and forget about the rest. Of course, when we come out of that we may find we cot it badly wrong in which case we shift back to Complex (my two-way arrow should probably return to the far left). If it turns out we have it right and change is unlikely to happen in the near to middle-term future then we shift it to Clear as best practice.
- Finally, if the situation and the solution are Clear then we just get on and do it.
Now note as you go down the scale things may fall out for the right shift, others may carry on down for diagnosis. At any stage, you may shift – something that requires more arrows! One of the functions of Cynefin is to deal at the right level of granularity so a sifting takes place as you go through the chaotic-complex-complicated-clear sequence and you may end up with multiple horizontal paths.
As such we reduce the chance of error (Disorder, or confused) and reduce the energy cost of change. We can also potentially automate the selection and allocation of tasks in a distributed model.
As I said early – this is not complete, I am going to experiment a little. It is part of the project I am engaged to map Cynefin to other methods and tools (my own version of ‘So What”) and I need to try it out a few times so it will change, Anyone else is welcome to experiment and reports would be appreciated.
The Letter W in the text is by Leo Reynolds discovered in Flickr as is the banner picture, both are used under the terms a creative commons license I’ve used the same pictures today as yesterday, but reversed their order