Regularities & reduction 2/2

Following yesterday’s post, I am continuing the pictorial theme of an ordered system in the banner picture and an emergent one in the opening picture.  Emergence is the phenomenon by which a coherent entity emerges with properties that can not be determined from the properties of its ‘parts’ but arise from the interaction of those ‘parts’ over time.  I would replace parts with actants for two reasons: firstly, actant is a richer term which allows distinctions between actors, constraints and constructors; secondly, it shifts us away from talking about parts, which implies an aggregative way of thinking about things – assemblies rather than assemblages.  If we go back to the French, assemblage is better understood as agencement, which is about playing things out, piecing things together, a form of Bricolage.  Dynamic words are essential in understanding complexity, although they have less flow, more flux, less river, and more estuary.  It’s one reason why I have never been comfortable with the comparison of cynefin to the German heimat (a sense of belonging of not being alien) or the Maori tūrangawaewae (domicile, standing, place where one has the right to stand – place where one has rights of residence and belonging through kinship and whakapapa).  Both are important words with long histories and contexts that I am sure I don’t fully understand.  But cynefin has a sense of being entangled in many threads of meaning, which are evolving; its origins are in the footprints left by animals as they explore the landscape; it is a nomadic word with all that implies, more kairos than chronos but that distinction is for another time.

Now, there are two fundamental errors that people make that irritate me.  There are probably more, but let’s keep it to two for now.  Both are variants of an excuse for not doing analysis or actively managing the system.  The first is the common statement in systems thinking that we need to take a whole system or holistic approach.  Now, in a complex system, you cannot understand the system as a whole, and if you attempt to do so, your perspective on the system will be distorted by the most influential perspective in the room.  This may be why it is popular in government labs and the like.  It also tends to have many idealistic statements about how things should be, which tend to be platitudinous, and we get a whole buzzword bingo of phrases appropriated from Indigenous thinking and nature without any fundamental understanding.  This always reminds me of my favourite quote from T S Eliot:

“Nothing pleases people more than to go on thinking what they have always thought, and at the same time imagine that they are thinking something new and daring: it combines the advantage of security and the delight of adventure.” 

The second was very common in parts of the Agile Community in which the phrase its emergent was a simple excuse for the rather infantile we don’t want to be managed.   That has rightly been called out by many people in the software development community. Still, unfortunately, they have often thrown out the baby with the bathwater and rejected the real insights possible through an understanding of complexity science.  I say ‘was ‘ although you still hear it from time to time from the Dedicated followers of fashion, the Agile Magpies, who pick up bright, shiny things they don’t understand to feather their nests.   Be deeply suspicious of those who keep creating something new but highly derivative, with only training, certification and speaking as their revenue sources.  This morning, I picked up a particularly shameless example on social media, but there are a lot around.

 The reality is that there is a lot we can manage, and I wrote a long post on this back in my 2020 Twelvetide series, which needs some updating and will take place in the Cynefin Field Guide).   Understanding that you can change the actants, you can change the interactions and need to monitor for what emerges.  That means weak signal detection and a broad and diverse monitoring system (a human sensor network as described in the EU Field Guide).  You can also create order, which has always distinguished the Cynefin Framework from the Stacy Matrix and others. 

Reduction

As I indicated in my last post, there is nothing reductionist about breaking things down, in anthro-complexity we do that with actant and interaction analysis, which also leads to mapping affordances (what we can change) and assemblages (the patterns of belief and attitudes, which act as enablers and blockers) as well as making changes to the agency within the system. A complex system scales by decomposition to the lowest level of granularity and then recombination. The reductionist error in a complex system is assuming that you can predict the qualities of those re-combinations from the actants and interactions which have generated them. Hence, there is a need for monitors and constraining aspects of the system so they provide some scaffolding or certainty going forward.

Regularity

Every snowflake is different, but they have regularities; they exhibit a six-fold radial symmetry arising from the ice’s hexagonal structure.  We see familiar political patterns, such as the current trend toward populism.  We have high levels of path dependency in human systems – much of the individualism that dominates thinking in North America and Northern Europe can be traced back, for good or bad, to the Reformation.  All organisations have histories, and you can never create a change programme based on a greenfield site.  Even when you create a company from scratch, you inherit much from your past and those of your key employees, not to mention the environment and affordances in which you operate.  Much of the time, those regularities persist and can be relied on.  If you are a radiologist, the probability of a gorilla appearing in an X-Ray is remote, so you don’t look for it.  Constantly scanning a system for unexpected weak signals consumes too much energy for it to be done regularly.   We depend on regularities to survive in the world, but those regularities, if disrupted, can trigger phase shifts in the system and significant failure – remember what happened to Kodak.  Just extolling people not to repeat such mistakes is remarkably stupid, although it seems to be a commonplace response.  We need to think differently and, as importantly, act differently to distinguish when we can make assumptions and when we should not.   That will be the subject of my next post, but living with emergent regularities, understanding the limits of emergence and, above all, managing it remains key.  

Emergence is an opportunity, not an excuse.

The banner picture is cropped from an original image of a server cabinet licensed under the terms of Unsplash (which I have paid for, so don’t try to reuse it). The opening picture of snowflakes by Elisabeth Lee was obtained from Unsplash.