I’ve been finishing the Cynefin Field Guide, which will be ready for pre-orders in early December. All the primary text is now there, either original writing or cutting, pasting and then updating from blog posts. It will be in Field notebook size (memories of keeping notes on geology field trips here), which is 21cm by 11cm closed. When open, the left page will contain a picture, three/five key points, and space for notes; the right side is text-constrained to around 1800 characters. If it needs more than that, then it’s a new section. Writing to that constraint has been an interesting and valuable experience. I will be crowdsourcing the bullet points with a review panel shortly.
The intention is to create a single source of ‘truth’, part of an ecosystem containing the open source wiki and Hexi; like Hexi, there will be QR codes to connect to the wiki pages. The Field Guide is designed as a reference for consultants and a handout for workshops, employees, etc. The focus is Cynefin and anthro-complexity, but it also includes other material and sections on the other sense-making frameworks. The following field guide will be about using narrative, where we have a mature body and some new methods and tools. I’ve already started to map out its structure. After that, I will write the Strategy Field Guide, including Estuarine Mapping and some new material that will begin to be explored on this blog over the holiday period. I hope to complete Narrative and Strategy by the end of January, with Cynefin going to print in December; after that, I will move on to jointly authored ones in application-specific areas: coaching, design, change and others.
People familiar with the Field Notes format will know you can buy leather folders, which can contain many such guides and blank notebooks – we may produce a limited edition collector’s item next year to celebrate twenty-five years as an independent organisation – April 2004 is when I left IBM to make all of this real.
Following an internal review of the Field Guide and feedback from training courses, we realised that the guide needed to elaborate more on the nature of Cynefin’s chaotic domain and provide more advice on how to act in chaos, anticipate it, or avoid it. I thought the information was clear, but I may have been overconfident, hence this blog post. An abbreviated version will be included in the field guide.
Over the years, there have been a few debates about whether Chaos is the right name for the domain, and those debates are ongoing and unlikely ever to be fully resolved. To a degree, in Cynefin, it represents a form of randomness, or to use the long-standing definition, a state with no practical constraints. In the three plus one version of Cynefin, using latent heat as an underlying metaphor, it is gas, as complexity is to water and order is to solid.
One of the reasons for a possible change in the Cynefin label is that I have never been happy with the use of Chaos in the so-called Forest Cycle, which appeals to many and is a critical element of Chaordic systems, which assume chaos is necessary for novelty. I believe it may be sufficient, but it isn’t essential, and to insist on it is dangerous. In everyday English, it means confusion, but it is used differently in mathematics and physics.
That makes it the responsibility of anyone using the language to define what they mean, and people reading and responding to respect or at least recognise the intent behind the definition used.
A good starting point is to look at the various definitions given in the Oxford English Dictionary. The following uses come from paraphrasing and directly quoting that august tome.
And, of course, in the modern day, it is used to depict the forces of evil or disorder in some fantasy fiction and games. However, the likes of Roger Zelazny in Princes of Amber have a more interesting take, seeing the two as a type of balance. It’s a brilliant series; not the best written, but it has some of the best ideas.
The etymology is worth quoting in full:
In 1750, it occurred twice in every million words, steadily rising to eleven in 2010, which may tell us something.
So, at a base level, in the three plus one version, we have three states of order, complexity and chaos with the aporetic domain standing athwart all three, a state of liminality from which aspects can be allocated to the other three. Order is highly constrained to the point where it is predictable. Complex has a myriad of connections or enabling constraints that make things unpredictable, but we can discern and influence patterns. Chaos has no effective constraints, so it’s a type of randomness. Now, in this, we have departed from much common usage. People say that traffic in Mumbai is chaotic, but it isn’t. It is tightly connected, so the traffic will adjust around you if you walk across the stream in a straight line. It just appears chaotic to a stranger.
All transitions between the main domains are phase shifts (think latent heat in the solid/liquid/gas metaphor), but that between order and chaos is catastrophic. We assume order and that assumption may lead to complacency and collapse or we may not be ready to accept a catastrophic event – the twister in the opening picture. So here we are using chaos in the common English language use as a state without form and, assuming resignation is not an option, the exit route is to seek to create some form of order.
Here, Cynefin departs from the Hobbsian Hobbesian view that absolute authority is the only alternative to chaos and argues that the role of a leader is to create constraints that allow solutions to emerge, not to solve the problem. In the EU Field Guide, this is known as the aporetic turn. Creating sufficient structure for issues to be assigned to complex or ordered domains for appropriate solutions.
When we add liminality to Cynefin, a closed subdomain is created, approximating what is known as deterministic chaos. None of the actants has any constraint, and therefore, a distribution pattern will follow a Gaussian distribution, which enables us to apply the wisdom of crowds. SenseMaker® MassSense involves presenting an issue to a network (for example, the whole workforce) and asking them to interpret it without any conversation or consultation with colleagues. Any conversations or interaction would make it a Pareto (the fat tail), not a Gaussian (the bell curve) Distribution, and we could not trust the result. Remember, those guessing the weight or counting the jelly beans are unaware of what other people have guessed, which is critical to using the quantitative results of multiple perspectives to create a degree of objectivity in managing uncertainty.
So, in Cynefin, chaos can mean an accidental loss of constraints, the sudden collapse into randomness (think Thom and Catastoprhy Theory), which will always be temporary. It can also mean (and in SenseMaker®, this is a critical function) the deliberate creation of an environment in which no actant can engage in discussion with any other actant. In the latter case, we have a powerful decision-support technique, but the energy required to keep people from talking with each other is high. Similarly, in accidental chaos, someone will impose order quickly. And you want to be the one doing that. In many organisations, the trigger to chaos may not be apparent to those in senior decision roles until it is too late, which can compound the problem.
One consequence is that, unlike physical systems, human systems have higher energy costs for chaos than for complexity, which is a more natural state with a low energy gradient. We don’t like randomness, and we seek to create order.
So, in Cynefin, there are several different but coherent uses of the word chaos, and critically, we break the order or chaos dichotomy by introducing complexity.
Farmer’s excellent book Making Sense of Chaos is important in establishing that you can’t model a complex system but can simulate it. In SenseMaker®, we would add stimulation to simulation in dealing with human systems. He also makes the same point, namely that “the word chaos can be misleading because chaos in the mathematical sense does not imply chaos in the same sense that we use the word in everyday English”. Further, he states, “Although complex systems can be complicated, they don’t need to be complicated. The words complicated and complex are not synonyms! Complicated means that something has many moving parts, whereas complex means it exhibits emergent behaviour – that is, it does things that are not easily predicted a priori”. In his work on simulation and the mathematics of understanding impact in economics, he distinguishes between simple chaos and complicated chaos, which lies at two ends of the spectrum between the Lorenx equations and weather prediction. People have asked if this can be mapped onto Cynefin, and my provisional answer is yes, but the manner of its mapping needs some work. I will include that in my reworking of the old three-by-three domain models. So there is some more work to do on Cynefin.
The banner picture is cropped from an original by Steve Johnson, and the opening picture is by Nikolas Noonan, both sourced from Unsplash.
Cognitive Edge Ltd. & Cognitive Edge Pte. trading as The Cynefin Company and The Cynefin Centre.
© COPYRIGHT 2025
October 12th is, if you believe it, World Migratory Bird Day. If you’ve been reading ...
This month, 21 years ago, I got caught up in a nasty bit of IBM ...