From time to time in social media I resist the temptation to respond when people talk about the Stacy Matrix; if they conflate the two I generally put up something mild along the lines of They are very different you know and Stacy himself deprecates its use. More recently I reacted with more vigour to a mash up of the two in which my disorder became a more extreme zone in the upper right hand area of the stacy matrix. That was wrong on so many counts and it, in part, stimulated this post. The other reason was the two days last week when Mary Boone and I, helped by other Cognitive Edge staff completed the planning for the Green and Red Mini-books of Cynefin which will hopefully be published this year. I have several blog posts to come over the next week to ‘fill in the gaps’ for those and this is therefore the first of a series. It is also worth reading the 2019 St David’s Day series of five posts which provided an update on Cynefin. Keep an eye out here and on social media for an invitation to review the outline and content in our new Haunt for members of the Cynefin Network.
To be clear I am not going to attempt a detailed description of the Stacy Matrix, or the Zimmerman variation. There is a lot of material on the web and in books; my purpose here is to establish similarities but more importantly differences. For the sake of clarity I have shown all three below. The basic starting point tends to be Simple-Complex-Chaotic but while the words are the same, the meaning and context of use are different. I read Stacy, but also drew on Paul Cilliers, Max Boisot and Alicia Juarrero. Stacy was the initial stimulus but I wasn’t happy with his use of the terms which didn’t conform with my understanding of some of the science. I had a lot of resonance with Paul and many discussions over the years drew me to the conclusion that in human systems we would have to use the chaos word iwith a different meaning from that in Physics; Alicia’s work on constraints made that possible. I mostly talk about using the constraint based approach to complexity and that remains a key component of what I have termed anthro-complexity as a field.
Often when I see people (particularity in the Agile world) use the Stacy Matrix they actually use the simplification adopted by Brenda Zimmerman. I first met her several years before she published Getting to Maybe at a seminar at New York University where I was running a workshop on the Cynefin framework and she was aware of Ralph’s (Stacy) work. She also introduced the commonly quoted (but rarely attributed) examples of Simple as baking a cake using a recipe, Complicated as getting a rocket to the moon and Complex as raising a child. While I find that useful and ‘not-wrong’ it has become problematic as people start their understanding at the wrong level of granularity. All three examples actually encompass at different stages or in different contexts all five domains (or eight if you add liminality) of Cynefin.
The commonality between Ralph and Brenda is that their matrices have two axes with a gradation between states based on human perception. Cynefin in its first manifestation also had axes but I (sorry about this but I can’t resist) took an axe to that approach fairly early on in the development. Cynefin has domains and critically the boundaries between Order (Obvious and Complicated), Complexity and Chaos are all phase shift boundaries. The best way I have found to explain this is to think of latent heat. If I boil water to 100º it doesn’t shift from liquid to gas for some time, as energy is required to achieve the shift. Equally it always warms up a little before snow as heat is thrown off to achieve the change from liquid to solid. In an human system complexity has the lowest energy gradient (as human systems are open) and it requires considerable energy to shift the system to a complicated state, or to temporarily remove constrains for a deliberate shift into the liminal zone between complex and chaotic. The shift from Obvious to Chaos is a catastrophic shift – every thing seems fine as you paddle down a broad, swift flowing river and you get complacent; by the time you see the waterfall its too late.
So that is the first key distinction; in Cynefin the movements between the three primary domains ( Order, Complexity and Chaos) are phase shifts not gradients, and the phase shifts involve the use or release of energy. Cynefin defines order into two based on human perception and consent, so that boundary is a gradient. It also adds Dis-order the state of not being aware of which type of system you are in which can be authentic (you deliberately confuse things to allow a shift but at risk, the liminal aspect of disorder) or you can be inauthentic which means you are complacently unaware of any distinction and carry on managing complex situations as if they were ordered.
The second key distinction is the treatment of chaos – described in the two matrices as Anarchy or Disintegration; two negative words. In Cynefin the accidental collapse into chaos is catastrophic and requires dramatic action to recover, but there is also a positive aspect namely the deliberate removal of effective constraints for innovation and distributed decision support. That involves a recognition that a highly uncertain situation in which there is no agreement is something that will happen; it represents the unknowable unknowns and is an domain of threat but also opportunity. It is going to happen, and often should happen so we should not dismiss it or seek to avoid it in all circumstances. Cynefin is a realist framework which leads us int the third area of difference.
Anyone familiar with Ralph’s work (and I suspect I am more familiar with his that he is with mine) will know that it has two distinct phases. He was a prize winning economist who became a corporate strategist. In that capacity he took on board complexity science to explain uncertainty. He was not the first economist to take that path; both Arthur and Beinhocker did to good effect. It was during this phase that he created the Stacy Matrix. The second phase is very different. After qualifying as a group psychotherapist he supervised the PhD programmes of Patricia Shaw and Doug Griffiths and they became a team with a succession of useful publications. Personally I think Stacy is at his best when writing with Griffiths. In this second phase he argued that natural science could not be applied to human systems except as an analogy and decried the use of the Stacy Matrix as allowing the continuation of the dominant discourse of management. This is when we get the idea of complex responsive processes which draw on the ideas of Mead, Elias and some aspects of the American pragmatists. Cynefin in contrast is a framework established within the naturalising school of sense-making and sees natural sciences as a constraint on understanding of human behaviour in organisations and societies. Complexity science on its own is not enough as humans also possess intelligence, identity and intention (the three ‘I’s of anthro-complexity) so we also draw on cognitive neuroscience, aspects of anthropology, constructor theory and others frameworks.
So our third key distinction is that Cynefin retains an assumption that adopting natural sciences as a constraint gives us greater confidence and anticipatory capability under conditions of uncertainty. Stacy in effect also takes a phenomenological stance (the axes labels alone create this) while Cynefin is an ontological framework that allows radically different conversations mediated by epistemology and pheunominaology. To make that simpler: how things are, how we know things and how we perceive things are distinct and increasing the alignment between the three is key to coherent action.
The fourth key distinction is the assumption in Cynefin of what I have called bounded applicability. Ralph constantly derogates systems thinking and lumps anyone he disagrees with into that heretical cast. Within Cynefin (and the phase shifts help here) there is nothing wrong with techniques such as BPR if you are dealing with a highly constrained system, but everything wrong if you only have the enabling constraints of complexity. This means we are not asking people to reject everything they have done in the past, but rather to realise the boundaries of its applicability and do things differently on both sides of the boundary. One of the key themes of the introduction to the Green Book of Cynefin is the need to break the fad cycle. Complexity allows ontologically incongruent systems to co-exist and that is a major breakthrough. It allows and even encourages coherent heterogeneity resisting the homogenisation of common values and the like that characterises much of systems thinking, and ironically Ralph’s work. Ralph and I are agreed that Systems Thinking and Complexity are significantly different, but I see both as having value.
Finally we get to the danger of axes labels that imply human agreement on position. If we are all in agreement and think we are close to certainty we may well be subject to pattern entrainment or inattentional bias. It may not be simple (or obvious) it may mean we are teetering on the edge of chaos. The value of disorder as the fifth domain of Cynefin is that it recognised both the legitimacy and the illegitimacy of being confused as to the nature of the system. The four points method for contextualising Cynefin emphasises that this is often the default position and shifting from it to complex, complicated or obvious is a resilient approach to making decisions. Cynefin is not drawn to identify decision points positioned between two perceptual linear scales, it is open and by its representation allows more multiple movements. The dynamics within Cynefin also suggest that in many events a constant movement or cadence between domains is more resilient that being in one specific area. So our fifth key distinction is the representation itself.
Opening picture is a part of an outward-facing wall on the side of the building in Spijkenisse by by Echo Grid; banner picture is the Cataratas do Iguaçu in Brazil by Guilherme Madaleno both on Unsplash