Friday, August 20, 2010

Cynefin Framework versus Stacey Matrix versus network perspectives


Lots of people seem to like the Cynefin Framework. Jess Dart and Patricia Rogers are some of my friends and colleagues of mine who have expressed a liking for it. It was one of the subjects of discussion in the recent Evaluation Revisited conference in Utrecht in May. Why don’t I like it? There are three reasons...

Usually matrix classifications of possible states are based on the intersection of two dimensions. They can provide good value because combining two dimensions to generate four (or more) possible states is a compact and efficient way of describing things. Matrix classifications have parsimony.

But whenever I look at descriptions of the Cynefin Framework I can never see, or identify, what the two dimensions are which give the framework its 2 x 2 structure, and from which the four states are generated. If they were more evident I might be able to use them to identify which of the four states best described the particular conditions I was facing at a given time. But up to now I just have to make a best guess, based on the description of each state. PS: I have been told by someone recently that Dave Snowden says this is not a 2x2 matrix, but if so, why is presented like one?

My second concern is the nature of the connection between this fourfold classification and other research on complexity, beyond the field of management studies and consultancy work. IMHO, I don’t think there is much in the way of a theoretical or empirical basis for it, especially when Dave’s fifth state of “disorder”, is placed in the centre. This may be the reason why the two axes of the matrix I mentioned above have not been specified, ...because they have not yet been found.

My third concern is that I don’t think the fourfold classification has much discriminatory power. Most the situations I face, as an evaluator, could probably be described as complex. I don’t see many really chaotic ones, like gyrating stock markets or changeable weather patterns, nor do I see many that could be described as simple, or just complicated. Except perhaps when dealing with single person’s task, not involving interactions with others. Given the prevalence of complex situations, I would prefer to see a matrix that helped me discriminate between different forms of complexity, and their possible consequences.


This brings me to Stacey's matrix, which does have two identifiable dimensions shown above: certainty (i.e. the predictability of events) and the degree of agreement over those events. Years before I had heard of "Stacey's matrix"" I had found the same kind of 2 x 2 matrix a useful means of describing four different kinds of possible development outcomes which had different  implications for what sort of M&E tools would be most relevant. For example, by definition you cannot use predefined indicators to monitor unpredictable outcomes (regardless of whether we agree or disagree on their significance). However methods like MSC can be used to monitor these kinds of change. And a good case could be made for more attention to the use of historian's skills, especially to respond to unexpected events that are of dispute meaning. More recently I argued that weighted checklists are probably the most suitable for tracking outcomes that are predictable but where there is not necessary any agreement about their significance. A quote from Patton could be hijacked and used here "These distinctions help with situation recognition  so that an evaluation approach can be selected that is appropriate to a particular situation and intervention, thereby increasing the likely utility -and actual use- of the evaluation" (page 85, Developmental Evaluation)

Post script: Here is an example of how I have used it for this kind of purpose, in  a posting on MandE NEWS about weighted checklists

From what I have read I think Ralph Stacey also produced the following more detailed version of his matrix:

This has then been simplified by Brenda Zimmerman, as follows

In this version simple, complicated complex and anarchy (chaos) are in effect part of a continuum, involving different mixes of agreement and certainty. Interestingly, from my point of view, the category taking up the most space in the matrix is that of complexity, echoing my gut level feeling expressed above. This feeling was supported when I read Patton's three examples of simple, complicated and complex (page92, ibid), based on Zimmerman. The simple and complicated examples were both about making materials do what you wanted (cake mix and rocket components), whereas the complex example was about child rearing i.e. getting people to do what you wanted. More interesting still, the complex example was raising a couple of children in  family, in other words a small group of people.So anything involving more people is probably going to be a whole lot more complex. PS: And interestingly along the same lines, the difference between simple and complicated was a physical task involving one person (following a recipe) and one involving large numbers of people (sending a rocket into space)

Another take on this is given by Chris Rodgers comments on Stacey’s views:
Although the framework, which Stacey had developed in the mid-1990s, regularly crops up in blogs, on websites and during presentations, he no longer sees it as valid and useful.  His comment explains why this is the case, and the implications that this has for his current view of complexity and organizational dynamics.  In essence, he argues that
  • life is complex all the time, not just on those occasions which can be characterized as being “far from certainty” and “far from agreement” …
  • this is because change and stability are inextricably intertwined in the everyday conversational life of the organization …
  • which means that, even in the most ordinary of situations, something unexpected might happen that generates far-reaching and unexpected outcomes …
  • and so, from this perspective, there are no “levels of complexity” …
  • nor levels in human action that might usefully be thought of as a “system”.
Well maybe,… but this is beginning to sound a bit too much like the utterances of a Zen master to me :-) Like Rodgers, I hope we can still make some kind of useful distinctions re complexity.

Back to Snowden

Which brings me back to a more recent statement by Dave Snowden, which to me seems more useful than his earlier Cynefin Framework. In his presentation at the Gurteen Knowledge Cafe, in early 2009, as reported by Conrad Taylor, "Dave presented three system models: ordered, chaotic and complex. By ‘system’ he means networks that have coherence, though that need not imply sharp boundaries. ‘Agents’ are defined as anything which acts within a system. An agent could be an individual person,or a grouping; an idea can also be an agent, for example the myth-structures which largely determine how we make decisions within the communities and societies within which we live."
  • "Ordered systems are ones in which the actions of agents are constrained by the system, making the behavior of the agents predictable. Most management theory is predicated on this view of the organisation."
  • Chaotic systems are ones in which the agents are unconstrained and independent of each other. This is the domain of statistical analysis and probability. We have tended to assume that markets are chaotic; but this has been a simplistic view."
  • "Complex systems are ones in which the agents are lightly constrained by the system, and through their mutual interactions with each other and with the system environment, the agents also modify the system. As a result, the system and its agents ‘co-evolve’. This, in fact, is a better model for understanding markets, and organisations.”

This conceptualization is simpler (i.e. has more economy) and seems more connected with prior research on complexity. My favorite relevant quote here is Stuart  Kauffman’s book: At home in the Universe: The search for the laws of complexity (p86-92) where he describes the behavior of electronic models of networks of actors (with on/off behavior states for each actor) moving from simple to complex to chaotic patterns, depending on the number of connections between them. As I read it, few connections generate ordered (stable) network behavior, many connections generate chaotic (apparently unrepeating) behavior, and medium numbers (where N actors = N connections) generate complex cyclical behavior. (See more on Boolean networks).

This relates back to conversation that I had with Dave Snowden in 2009 about the value of a network perspective on complexity, in which he said (as I remember) that relationships within networks can be seen as constraints. So, as I see it, in order to differentiate forms of complexity we should be looking at the nature of the specific networks in which actors are involved: Their number, the structure of relationships, and perhaps the extent to which the actors have own individual autonomy i.e. responses which are not specific to particular relationships (an attribute not granted to “actors” in the electronic model described).

My feeling is that with this approach it might even be possible to link this kind of analysis back to Stacey’s 2x2 matrix. Predictability might be primarily a function of connectedness, and therefore more problematic in larger networks where the number of possible connections is much higher. The possibility of agreement, Stacey’s second dimension, might be further dependent the extent to which actors’ have some individual autonomy within a given network structure.

To be continued…

PS1:Michael Quinn Patton's book on Developmental Evaluation has a whole chapter on "Distinguishing Simple, Complicated, and Complex". However, I was surprised to find that despite the book's focus on complexity, there was not a single reference in the Index to "networks". There was one example of a network model (Exhibit 5.3) , contrasted with a Linear Program Logic Model..." (Exhibit 5.2) in the chapter on Systems Thinking and Complexity Concepts. [I will elaborate further here]

Regarding the simple, complicated and complex, on page 95 Michael describes these as "sensitising concepts, not operational measurements" This worried me a bit, but it is an idea with a history (Look here for other views on this idea). But he then says "The purpose of making such distinctions is driven by the utility of situation recognition and responsiveness. For evaluation this means matching the evaluation to the nature of the situation" That makes sense to me, and is how have I tried to use the simple version of the Stacey Matrix (using dimensions only). However, Michael then goes on to provide, perhaps unintentionally, evidence of how useless these distinctions are in this respect, at least in their current form. He describes working with a group of 20 experienced teachers to design an evaluation of an innovative reading program. "They disagreed intensely about the state of knowledge concerning how children learn to read..Different preferences for evaluation flowed from different definitions of the situation. We ultimately agreed on a mixed methods design that incorporated aspects of both sets of preferences". Further on in the same chapter, Bob Williams is quoted reporting the same kind of result (i.e conflicting interpretations), in a discussion with health sector workers. PS 25/8/2010 - Perhaps I need to clarify here - in both cases participants could not agree on whether the situation under discussion was simple, complicated or complex, and thus these distinctions could not inform their choices of what to do. As I read it, in the first case the mixed method choice was a compromise, not an informed choice.

PS2: I have also just pulled Melanie Mitchell's "Complexity: A Guided Tour" off the shelf, and re-scanned her Chapter 7 on "Defining and  Measuring Complexity". She notes that about 40 different measures of complexity have been proposed by different people. Her conclusion, 17 pages later, is that "The diversity of measures that have been proposed indicates that the notions of complexity that we're trying to get at have many different interacting dimensions and probably cant be captured by a single measurement scale" This is not a very helpfull conclusion. But I noticed that she does cite earlier what seem to be three categories of measures that cover many of the 40 or so measures: These are: 1. how hard the object or process is to describe?, 2. How had it is to create?, and 3. What is its degree of organisation?

PS3: I have followed up John Caddell's advice to read a blog post by Cynthia Kurtz (a co-author of the IBM Systems Journal paper on Cynefin) recalling some of the early work around the framework. In that post was the following version of the Cynefin Framework included in the oft-mentioned "The new dynamics of strategy: Sense-making in a complex and complicated world" published in the IBM SYSTEMS JOURNAL, VOL 42, NO 3, 2003.
In her explanation of the origins of this version she says it had two axes: "the degree of imposed order" and "the degree of self-organization." This I found interesting because these dimension have the potential to be measurable. If they are measureable, then the actual behavior of four identified systems could be compared. And we could then ask "Does their behavior differ in ways that have consequences for managers or evaluators?" I have previously speculated that there might be network measures that could describe these two measures: network density and network centrality. Network centrality could be the x axis, being low on the left and high on the right, and network density could be the y axis, low on the bottom and high on the top. How well the differences in these four types of network structures might capture our day-to-day notion of complexity is not yet clear to me. As mentioned way above, density does seem to be linked to differences between simple, complex and chaotic behavior. Maybe differences in centrality moderate/magnify the consequences of different levels of network density?

PS4 (April 2015: For more reading on this subject that may be of interest, see Diversity and Complexity by Scott E Page, Princeton, 2011

PLEASE NOTE: To make a Comment, or to read the Comments already made on this post, click on the Leave a Comment  link below, or directly here


  1. This is a duplicate post of another made two days ago. I have not been able to pen or edit that post, so I have had to re-create the post here. I will try to copy the 4 Comments across to this version as well.

  2. Well for a start RIck Cynefin is not a matrix, its an emergent framework. If you read this you will see how it is created so it can't start with the intersection of two dimensions. Its a sense-making framework which is socially constructed from peoples experience of their past and also their anticipated futures.

    It is of course derived from the three fold classification of systems that are defined in terms of constraints (the bit you like) You might (if you merged the simple and complicated and call them order, prefer this representation. The prime model is the three systems, Cynefin simply divides order into simple and complicated and adds disorder to recognize that human perception is also important.

    I don't like 2by2 matrices because they create a categorisation approach in which the model precedes the data so people make things fit. As you can see from the first link the framework emerges from the data so its better for sense-making and is more likely to recognise a changed or changing context.

    I know Stacy's model well, ad his latter rejection given that he now things everything is complex. I also think the danger with gradient models is that don;t enable people to think differently. They just settle where they are most comfortable (which is what Stacy has done). - a framework with boundaries (such as Cynefin) allows people to see that they need to behave differently in different contexts.

  3. Oh, and a PS. I am currently working on a constrain based model to allow discrimination within the complex domain - my previous comment handled your first two concerns, hopefully this will help with the third

  4. And thanks to Dave Snowden for such a prompt response, along with some useful links on the origins of the Cynefin framework.

    I will be especially interested in the "constrain based model to allow discrimination within the complex domain" that he mentions he is developing

  5. In my attemtp to copy over Dave Snowden's comments I lost two embedded links.

    Here they are:

    "Well for a start Rick Cynefin is not a matrix, its an emergent framework. If you read this you will see how it is created so it can't start with the intersection of two dimensions. Its a sense-making framework which is socially constructed from peoples experience of their past and also their anticipated futures.

    It is of course derived from the three fold classification of systems that are defined in terms of constraints (the bit you like) You might (if you merged the simple and complicated and call them order, prefer this representation. The prime model is the three systems, Cynefin simply divides order into simple and complicated and adds disorder to recognize that human perception is also important."

  6. Rick, in a recent couple of posts Cynthia Kurtz (a co-author of the IBM Systems Journal paper on Cynefin) recalled some of the early work around the framework and some other ways of thinking about it that may help address some of your questions. The diagrams are very thought-provoking and may be of use.

    Here's the first one (the comments are also very interesting):

    regards, John

  7. Rick, in respect of your PS. Cynthia's post says when she encountered the Cynefin framework (which was pretty much formed at that time) she went away to think about the subject in general and came up with a model based on two axes which generated the idea of different network structures in the different domains. Those tetrahedrons from that (which she is further developing) were added to the Cynefin framework which added greatly to it. If you check here you will see the story of that. It was the penultimate leg of a long development. The final stage was to conform the language to avoid confusion between ontology and epistemology.

    As I identified in my earlier response the Cynefin framework is a sense-making one and is normally created as an emergent property of social interaction. One of the reasons for this is the need to root any sense-making model in peoples' own understanding of their past and possible futures. A system may in practice be complicated (ordered) but the nature of human knowledge means that for them it is complex or chaotic, and their interaction with meaning needs to reflect their understanding of the system. This is alignment between what is, what we know and how we perceived that I described in the final History of Cynefin post here.

    If we are looking for measurement systems that allow us to determine where something sits within the framework then I think we have to create persuasive power with the "facts". In operational practice we do that with a wisdom of crowds variant namely the capture of large volumes of self-signified narrative. The combination of statistics with anecdotes in a single system seems to provide that. You'll see more of that on the site shortly and it relates to our various conversations about linking SenseMaker® with Most Significant Change and (I think) to complexity models.

    1. @Dave Snowden: Unfortunately links within your post from are broken. Seems that there was a relaunch meanwhile. Is it possible to provide permalink or redirects? The whole story is of major interest to me and others too. Thanks :-).

  8. I need to catch up with Dave’s initial comment on this blog….

    1. The origins of the Cynefin Framework versus its current form and use.

    • Dave’s first link provides a useful background to how the Framework was developed. Its development is described as a process of of social construction. So perhaps any critique needs to be framed in terms of what can be expected from such a social process. That is not what I have tried to do in this blog, but would be interesting to explore. Caveat: Under “Social construction of the Cynefin framework” Dave seems to describe a social process of its application, but as far as I can see the basic typology is introduced to the participants. See Workshop Para 1.

    • However, the product of that process has taken on a life of its own (a meme no less), and its dissemination has been facilitated by Dave and many others (e.g. see Shawn Callaghan’s explanation video). It is this freestanding aspect of the Framework which I am addressing, the one which is most widely known and responded to. This product is important, because it is about a process of generalisation and wider application of knowledge, from those who originally constructed it, to many others.

    2. Categories versus continuums used in Frameworks and Matrices: I hear Dave saying two things:
    • "I don't like 2by2 matrices because they create a categorisation approach in which the model precedes the data so people make things fit"
    • "I also think the danger with gradient models is that don’t enable people to think differently"

    While I can sympathise with both views as cautions, if they are taken as prohibitions then we are left with no way of constructing a modeI that is usable by others.

    PS: In one of his posts Dave mentions the need to avoid “reductionism”. Like the word “linear” this word is taking on the automatic connotation of being “A BAD THING”, at least in some of the evaluation literature tying to address complexity issues. As this Google define: reductionism search result shows, there are many ways of defining reductionism. In my view the search for the simplest workable explanation of apparently complex phenomena is an intrinsic part of science.

  9. I'm pretty careful on the dissemination Rick, but I agree that others can interpret it too easily as a two by two (Shawn's video does that). I do my best to correct it when I find it. Oh and the typology is not given away other than in narrative descriptions of the extreme states.

    Cautions should never be prescriptions, but they should still be cautionary. Its linked back to the criticality of constraints which I introduced into the framework after leaving IBM to bring it closer to the literature, in particular the work of Brian Goodwin of Schumacher and others. For me categorisation models are highly appropriate in ordered systems (simple and complicated) and have utility there, but are dangerous elsewhere. Gradient models don't create boundaries and humans need boundaries to think differently. With a gradient people settle where they feel comfortable.

    Ditto reductionism (I am using it in the sense of its use within the complexity literature) which is OK for order where the whole is the sum of its parts and you can reduce a system to components. In a complex system this is impossible, and if attempted creates a reduced perspective.

    This is all part of a key Cynefin concept, namely that of bounded applicability. Different (and contradictory) approaches work in different domains.

  10. Rick, nice to meet you and thanks for reading my blog post. When you say "these dimension[s] have the potential to be measurable" - that's somewhat true. But they also have the potential to be misunderstood and misrepresented. I do use stories and questions to map the space, but I consider this as measuring perception and experience rather than measuring reality per se.

    As to the category issue, in my view there isn't a huge difference between an explicitly categorical model and a model that just uses the same names over and over across many contexts. You don't have to SAY things are categories in order for people to use them as categories; in fact you can categorically deny it. But if you keep using the same names for well-defined states across many contexts, people will pigeonhole things into them whether you like it or not. Indeed, I've found that permanently named bounded areas lead many (but not all) people into simplistic thinking: the situation is "in" this area or "in" that area. I think it's more fruitful to think the other way round: areas are in situations, not situations in areas. Almost all of the situations I help people make sense of are spread across the landscape of hierarchical and meshworked connection, with aspects in many locations intermingling and interacting. That's why they require sensemaking.

    What I like to do is help people use the confluence framework to develop an emergent sense of the landscape of their context. This may include various features such as boundaries and bounded areas. Then I help them come up with their own names to describe the relevant features of their landscape, wherever they are located. Such context-specific names are more useful to context-specific goals than generic terms can ever be, and they don't carry the danger of categorization out of context. This is why I describe the entire confluence framework as negotiated space: because negotiation of the landscape produces something unique and uniquely useful to each group and goal and problem. The only generic terms I use are the terms that define the space itself, not bounded areas within it. I don't find comparing those across broad contexts to be very useful in practice.

    By the way, I have followed your work on Most Significant Change with interest. We have some things in common! Happy to connect off-line.


  11. This is a second part of my previous comment (which should have waited for it really). In response to what you said about lots of measures of complexity being proposed, I'm not surprised, because I've also found lots of remarkably similar sensemaking frameworks that draw on complexity, not least the Native American medicine wheel, which goes back several thousand years. I call these frameworks and systems of thought "siblings" because they represent the lively diversity of thought created by many people thinking in their own ways about similar things. I agree with those who say that many approaches work in different areas, and I also think that many approaches work in different contexts and for different thinkers in the same areas. It's healthy to have lots of different ways of thinking about things, and it's healthy for each of us to find what works best for each of us. May we all find what we need.

  12. Hi Rick - Some important insights here - thank you!

    I particularly note your comment that "Predictability might be primarily a function of connectedness, and therefore more problematic in larger networks where the number of possible connections is much higher." Agreed; and predictability is also not so much problematic as (by definition) impossible in 'Chaotic' contexts, or domains of inherent uniqueness. Complexity-science techniques can be very useful, but since every complex real-world context also contains at least some aspects of uniqueness, those same techniques can at times also be misleading.

    To practitioners, a real danger of complexity-science in all its forms is that by definition it's attempting to use a Simple concept of order to describe something that is inherently Complex and unordered. Hence it may be the limitations of the science itself that are the real cause of an apparent problem. In sensemaking it's essential to be able to switch between models so as to contrast and cross-compare, to allow new insights to arise from the cognitive dissonance between them, much as you've described above. Reliance on a single model such as Stacey or Cynefin or whatever could lead to unfortunate results, where the attempt to make things fit the model prevents us from seeing what's actually going on, and where the actions espoused by the model only make a wicked-problem worse, yet being unable to see how and why. As in your teachers' example above, assigning too much primacy to a single 'scientific' model creates a tendency to fall back to positions that owe more to religion than to science itself: "they disagreed intensely" etc. As you say, the only way out of that kind of impasse is a mixed-methods approach - which technically could not be described as 'scientific'.

    So from my own experience I would say that most of your doubts about Cynefin are spot-on. (Struggling with similar doubts led me to some alternate approaches that you may find useful for your own work - see my book 'Everyday Enterprise Architecture: sensemaking, strategy, structures and solutions', .)

    But in practice Cynefin is merely one framework amongst many: it has its usefulness, but is also problematic in its own ways. I've found the 'official' Cynefin framework valuable in the past, in my own field of enterprise-architectures, but - much like Taylorism or hard-systems theory - it now perhaps carries too much baggage and too many assumptions to be usable outside of a fairly narrow set of constraints. The core categorisation (Simple, Complicated, Complex, Chaotic, and order versus unorder) remains useful, especially when used in a layered, recursive manner where the model may explore and reflect upon itself. (Note, though, that Snowden himself describes such tactics as 'illegitimate' - see - although he's never explained why.)

    The key point, though, is that you're doing important work here, in this kind of cross-comparison of models: please don't allow yourself to be distracted from that overall aim!

  13. I would agree with Stacey's latest views. To help you appreciate where he is coming from, imagine areas 1-3 as domains of first-order cybernetics, area 5 of the second order, and area 4 as the third order. I would presume that Stacey operates comfortably within the 3rd order domain, as I do. To do so enables you to see the human narrative in its entirety from a first order perspective.


  14. Hi, I think misunderstandings in this area lie in a) miss-assigning one's own concept to somebody else's term and b) trying to show everything in a two dimensional graph.

    Surely we need to separate these distinct dichotomies?
    Deterministic v not
    Predictable v not
    Rigid v malleable
    Stable v changing
    Complicated v simple (internal view)
    Complex transformation v facile transformation(external view)
    Complex structure v complex behaviour
    Self-aware v unconscious
    (and some other dichotomies I've listed elsewhere.)