1481442392-featured.jpeg
Dave Snowden

A return to constraints

RSS Feed

I’m getting increasingly excited about the work I am doing on constraints. I first posted on this with an initial taxonomy back in September. Since then I have been working the ideas into a series of keynote speeches and more recently workshop exercises. One of the reasons this is important is that we can manage constraints and a large part of doing something within a complex systems is understanding what you can manage and how. Given that you can’t reduce complexity by simply setting goals (the main error of popular forms of systems thinking) this is one of the most important questions. Complexity is all about making changes in the present with a sense of direction but not with a specific goal, other than in very limited circumstances.

So mapping the present is key in order to know what to change. This links to my long standing three simple and eliminative questions to get started in any complex intervention:

  1. What can we change?
  2. Out of the things we can change, where can we monitor the impact of change?
  3. Where we can monitor the impact, can we rapidly amplify success or recover from failure?

The first two have always seemed to be self-evident; just because you want to change something doesn’t mean that you will be able to; if you can’t see what is happening then you might be on a pathway of catastrophic failure. The third contains an important insight from complexity: when an opportunity presents itself you will have limited opportunities to take action.

Now all of that has worked to get people talking about the issue, but the next stage is to make it more tangible as a method, or series of methods that can be put into operation and scaled. Given that constraints are things that can be mapped, and where change is possible, a typology as much as a taxonomy. Now the way I work is the explore an idea through conversations and reading, then I start to play with the ideas in front of an audience, post here then repeat several times until I can teach it. So this is all still a work in progress. However I have been modifying my September thoughts somewhat and I’ve increasingly made a distinction between robust and resilient types of constraint. Then helps with the order-complex distinction in Cynefin and also (I think) helps people get the idea. More on what we do with these over the next week with some interruptions on other themes. Oh and I am keeping with the idea of Containment and Coupling as a master distinction. Each of the constraint types below apply to both.

Robust constraints can be …

  1. Rigid or fixed which means they are clearly defined, visible and can be enforced. If you want a metaphor think of a sea wall, or dykes. They have all the advantages of certainty, all the disadvantages of sudden catastrophic failure when their limits are reached.
  2. Elastic and thus have the benefit of adapting to a degree of change which is good, but they can still break and possibly give a false level of confidence.  An elastic waist band may give you the illusion of maintaining a healthy weight but only for a time.  The failure is more extreme.
  3. Tethers which provide a backstop, like a tow rope for example which has the slack taken up before it applies.  These can be fail safes, backups only coming into play in extremis.  The danger is damage when they snap into effect, both for the object being tethered and for the tether itself

Resilient constraints can be …

  1. Permeable (for containers) or conditional (for coupling), in both cases the constraint is contingent.  Some things can get through others can’t.  Think of a salt marsh to contrast with the sea wall if you want an example.  Or a system where rules can be broken in specific cases subject to heuristic control.
  2. Mutating which means they change over time, they don’t switch on and off, you can see the evolutionary pattern.  Case based legal systems on contrast with those based on the Napoleonic code would be good example of these.  In companies they are hardly even used, but could be valuable.
  3. Dark are like dark matter or dark energy, we can see their impact but we don’t know what they are.  Aspects of organisational culture fall into this category as do taboos, rituals and the like.  Far more prevalent in modern organisations than people realise they are almost a complex system in their own right, only knowable or changeable by interaction.
Top