In addition to my hissy fit about the “agreement x certainty” matrix I have also been in a bit of a lather about the typology in the Cynefin model that identifies four states of systems – simple, complicated, complex, and chaotic. I like this model and I think it is exceedingly useful for helping people understand the program and evaluation scenarios they are working in. But at the same time I always bristled at it, and I finally figured out why.

As I see it, the way Cynefin draws on complexity concepts only partially overlaps with the way complexity science deals with complexity. Those four domains only partially overlap with what CAS researchers and theoreticians think of as complex systems. As I read the complexity literature I find no discussion of “simple” and “complicated”. Those concepts are not part of the research and they are not part of the theory.

The Cynefin  tradition borrows from the CAS tradition in some ways but not others. And, the borrowing is not rigorous. For example the notion of “complex” in the Cynefin mode pulls on ideas such as “emergent” but the CAS folk never discuss “probe- sense – respond”, which is how Cynefin proposes that complex activity be treated. PSR are human activities that we use to understand and guide our actions. But they are not complexity science concepts, the understanding of which has nothing to do with how  human beings do planning, management, or evaluation. Also, I don’t know for sure but I’d bet that the PSR definition of “emergence” is not totally congruent with the definition in the CAS tradition. Ditto for ”chaos”.

All this makes for great confusion, not to mention mental health stress for me.  Or at least it has, because I could not get away from the notion that both traditions were using words in the same way. They are not, but the words always seemed similar enough that I could not escape believing they were the same.  What I need to do is to disentangle the conversations, and not assume that one is related to the other. Ah, I feel much better.

Much better but not completely cured. It all still bothers me because I think the there is widespread belief that the Cynefin view says something about complex systems as the rest of the world thinks of complex systems. (Prove me wrong, I would be very happy to be disabused of this notion.) The result is that the real value of understanding complex systems is lost to our field.

17 thoughts on “Simple, Complicated, Complex and Chaotic VS. Complexity Science. Jonny finally resolved his confusion

  1. Alas, the fields of complexity and its application to human systems are both in states of evolution. When Brenda Zimmerman first started talking about simple, complex, complicated, chaotic way back in the early 90s, it was a revolutionary path out of the control/command models that had us stuck at that time.

    Some people continue to use this taxonomy, but I don’t find any of the embedded intelligence in it that I see in the LD (as decribed above). I would much rather people find ways to engage with the system that is in front of them than have them step away to decide what label to attach to it before they can decide what to do.

    Yes, I agree with you entirely. A tool, model, or method can only serve well when it is understood, critiqued, and consciously fit to the task at hand. The concepts underlying complexity are hard to understand, they are shiny new objects, and they are sold as cure-alls. No wonder we find ourselves stuck with things that don’t work. It is especially sad because we are trying desperately to replace an old set of things that didn’t work either.

    Of course I don’t think any of these things is true about CDE, but I won’t bore you with why.

  2. Cynefin only borrows some aspects of complexity because it didn’t originate from complexity science. David Snowdon and Cynthia Kurtz were, for want of a label that probably David would challenge, information scientists working for IBM and appear to me to have based a lot of their early thinking on knowledge management concepts and network theory. The complexity stuff came later, and to a large extent I think has been more of a vehicle for carrying David’s ideas than a core set of principles. No criticism implied here – I think that’s how many of us come to and use ideas from the systems domain.

    In any case Cynthia wrote a piece a couple of years ago that – from memory – sought to distance Cynefin from complexity science … but I may be very wrong about that.

    Which isn’t to say that Cynefin’s conceptualization of simple, complicated and complex is wrong, indeed I find it extremely useful. But as I said in my Evaltalk post recently, we really need to do our homework about where each persons conceptualisation of simple, complicated and complex comes from. They don’t come from the same place and thus don’t necessarily have the same meanings, strengths, weaknesses and uses.

  3. I think you are still confused, more so if anything. Probe-Sense-Respond is our suggested response to engagement with a complex system not a description of it. The five domains (not four you keep getting that wrong) cannot overlap with complexity science because only one of them (complex as it happens) is intended to. So you will not find any discussion of simple or complicated because they are ordered system decisions not complex.

    I suspect you are still seeking a single ontological approach. The essence of Cynefin is that it recognises multiple ontologies, each of which has its own epistemology and of course as we are dealing with humans there is a phenomenological overlay.

    If you read the literature then you will find there is no consistent use of Chaos so everyone has to choose one. I work with the constraint based definition (per Juarrero and others).

    Bob – Cynefin emerged (sic) from a fusion of the original knowledge work and complexity work. The history of Cynefin article makes this clear.

    Oh and Glenda, Zimmerman’s use was derivative not original

  4. And an after thought, you will find the ‘understand through engagement’ (probe-sense-response) throughout a large section of the literature in the field. I’ve spoken on this at the European Complexity Science Conference, the complexity interest groups at the NSA in the USA and EPSRC in the UK (where I was Director of their Complexity Programme). I don’t remember anyone having a problem with it.

  5. I appreciate David Snowden’s comments. I’ve read many times in the Snowden/Kurtz paper the explanation of the ontological nuances of Cynefin. It’s the part of the paper I return to most often. I think it would benefit others in this blog if David could expand a bit more on his earlier comment.

  6. I planned after reading this to write a blog post as a response, and sweep up some of Rick’s other comments on research and the like at the same time. May not be for a week or so though

  7. Dave, I look forward to that. I’m thinking the epistemological/ontological distinction is an important one as we think about how these “competing” models are true/useful. In the meantime, I’ll look into the Snowden/Kurtz article and see what I can make of the ontological nuances of Cynefin

    It looks to me like “probe, sense, respond” and Adaptive Action (what? so what? now what?) are essentially the same with one possible exception. Do you find probe, sense, respond equally useful in all parts of Cynefin? Adaptive Action works fine regardless of the dynamics, it is just more interesting when constraint decreases and degrees of freedom increase.

    Sorry to have rewritten history. Where did the simple, complex, complicated, chaotic distinctions show up first?

    Thanks for this and all.

    1. In Cynefin epistemology follows ontology, that is clear in Kurtz/Snowden as well as Snowden/Boone and the earlier articles I wrote before I met Cynthia and afterwards. So I am puzzled by your question on the applicability of “probe-sense-respond” as all the articles make it clear that is the strategy if the situation is, or is perceived to be, complex while in the other domains different strategies apply. In complicated we sense-analyse-respond, simple sense-categorise-respond and chaos act-sense respond (although this is only a temporary state)

      I don’t think that asking “what, so what and now what” can ever do any harm in any domain but it isn’t as you say differentiated by type of system. So in my language you have a single epistemology of action to deal with multiple ontologies. The action is not, as you say, unique to a complex situation. Having just skim read your recent book I think this is a significant difference as (and I reserve the right to modify this when I read in more detail before reviewing) you are using complexity theory to validate an ideology of mediated question asking and conversation.

      My view is that we need to take a different approach in each domain, including not wasting time in multiple conversations if the position is ordered, simple or complicated. In a complex system that approach does not include ideas of parallelism, obliquity or naiveté except by accident, and that is just for starters! You also don’t seem to distinguish that much between chaotic and complex systems (and you have others in that camp) I think the lack of that distinction limits the range of actions that can be taken.

      The complex-complicated distinction is an old one. I normally attribute it to Paul Cilliers, but when he was alive he used to say that he knew he got it from someone else.

      1. I’m with Dave on the ontology of this, and as well as his idea of responding according to the nature of the situation. For me this is one of the great contributions of the Cynefin framework – it’s a means of informing how to manage different aspects of a situation and thus use scarce resources efficiently, effectively and with optimal efficacy. If we managed every aspect of a situation as if it were complex, we’d hardly get anything done.

        In my wanderings around the systems field I’ve observed two areas where folks talk past each other. There is a strong ontological orientation in some parts of the systems field, especially in the USA, and those people find the more epistemological orientations quite unintelligible. It is especially true in the evaluation field, where I can count on the fingers of both hands those who don’t struggle with the epistemological orientation. The other area contains those who consider as entirely complex any situation that involves human beings. For them, ontologically there is no such thing as simple or complicated aspects of a situation, even if they agree that there may be some benefits in considering these categories in an epistemological sense. I wish there was a way we could actually discuss these two tensions, because it really is a brake on taking these important systems and complexity ideas further.

        Incidentally, according to Gerald Midgley, the earliest reference to “complexity” as a field of inquiry was Warren Weaver back in the late 1940s. Out of curiosity I’ve just looked him up in Wikipedia. I was delighted to read he was fascinated by Alice’s Adventures in Wonderland, and in particular how to translate the riddles and puns into different languages ….. a rather nice example don’t you think of the balance between ontology and epistemology as well as the simple, complicated and complex?

      2. Love the Alice and Wonderland example! I’m with Gerald, Alicia and others in Paris tomorrow so I will ask around to try and get more of the history

  8. Thanks to Dave Snowden for the link to this blog and post. I would disagree with the original post re the claim that notions of Simple and Complicated are not in literature on complexity and what constitutes a complex adaptive system. The whole notion of complex systems is how the interconnection of autonomous agents (we humans, e.g.) who are conscious and capable of both self-reflection and non-linear actions, results in systems different from simple or complicated examples. It’s been a while since I looked, but my middle-aged mind suggests a look at briggs and Peat’s Turbulent Mirror; Eoyang and Olson’s Facilitating Organization Change (citing Kevin Dooley on CAS definition, as I recall); Wheatley’s A Simpler Way; and so on. I’d also note that with regard to program and policy evaluation, Michael Quinn Patton’s book, Developmental Evluation, makes clear the distinctions in sense-making and response, between simple, complicated, and complex situations. This book also draws significantly on complexity sources including Cynefin, Eoyang, and Zimmerman among others.

  9. This is really interesting, You’re a very skilled blogger.
    I’ve joined your feed and look forward to seeking more of your great
    post. Also, I’ve shared your web site in my social networks!

  10. excellent points altogether, you simply received a new reader.
    What would you suggest about your publish that you just made a few days
    in the past? Any positive?

  11. Hi! Someone in my Myspace group shared this site with us so I came
    to give it a look. I’m definitely enjoying the information. I’m book-marking and will be tweeting this to my
    followers! Fantastic blog and great design and style.

Leave a Reply to snowded Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s