Convergence of description

One of the thing that was an undercurrent of thinking and discussion in Paris this week was the perennial debate between an extreme form of post modernism and those of those (to nail my colours to the mast) who take a more scientific view.   Now I should say that given the people there this was a serious discussion, unlike the level of trivia you often see in the blogosphere.  One recent example took the position that as observers influence the system we have to assume that everything is complex.  An extreme view triggered by reading a book that talked about reflexivity without bothering to check the wider context of that word.  Widely read I find does not always mean deeply read, the former may satisfy ego needs, the latter however has more value.  

Now the fact that you can find examples of situations where people believed things that turned out not to be true does not mean that we have to abandon the idea of evidence or for that matter of the use of order.  Newtonian mathematics works well if you understand the constraints under which it operates and the use of said mathematics is a form of analysis in the sense-analyse-respond instruction for the complicated domain of Cynefin.  In an operating theatre counting the number of swabs in/out and used is an essential aspect of not leaving some of them in the body  of the person you have just operated on.  Yes mistakes still happen, but it should be a simple matter of process control where best practice applies.

Treating everything as complex is as ineffective as assuming its all ordered. One problem (of many) with fad followers is that they want one universal and simplistic way of approaching the world, before moving on to the next bright new shinny idea.  We also get the old wine in new wineskins phenomena where the language of complexity is used as an overlay to the consultants pet practices.  You see a lot of that where people are ideologically driven to highly facilitated techniques but more on that another day.

Now no one, even the extreme post modernists, at the Paris event was that pathetic, neither were any sensitive to disagreement or trying to promote books or training programmes so the discussions were long, serious and very useful.   I spent a lot of time with Tim Allen Timothy Allen, Professor Emeritus of Department of Botany (and Environmental Studies) at the University of Wisconsin.  He talks about narratives within science and they way they blind people to evidence that they are wrong, or more frequently that there are different ways of looking at the problem that might produce more interesting results.  But at the same time he is passionate about evidence and how we get there.  Between him and Thagard yesterday (who takes a harder view of the role of science, I am somewhere between the two) we got to the key issue of how do you achieve convergence of description before you move to evaluation.  

Now this links strongly to work I am doing on monitoring and evaluation as well as avoiding premature convergence on plans and actions within organisations.  I got some really interesting ideas from Tim on how this is done, and how robust theory emerges within ecology and I am going to be using those in both some current SenseMaker® projects and in my own writing.   In essence (sic) the primary issue is how do we enable convergence of description without moving a synchrony (or group think).  It is one of the key aspects of the complex domain model.  There are however some key principles that apply here and I am currently at five and counting:

• Keeping parallel independent observers and comparing results is key, which is why approaches to complexity that just ask questions in facilitated conversation or techniques like open space are dangerous in this context (valid in others).
• There is a significant different between developing a theory and validating a theory, two steps not one that case based and purely inductive techniques miss out on.  Objectification of abductive plausibilities is going to be key here.  I'll explain that more in a future post.
• Context is everything, it enables you to understand constraints (another reason for using that definition of complexity) avoid interpretative conflict and generally move forwards
• Its never about the view of one observer, its about the independently formed views of multiple observers in many contexts
• Weak signal detection needs to focus at the liminal, at boundary crossings and boundary approaches to understand when context is shifting and limitations may be being reached.

A lot more on this in the next month or so, but I have a lot of new reading to do first.  I also need to read it in depth not simply skim!  In the mean time thanks to Mike Lissak who set it up and who first introduced me to Alicia Juarrero who was also there.  Good time in good intellectually stimulating company is food for the soul and real food not the junk food of trivialisation of big issues.