Dave,

You may have gathered that I'm relatively new to this game (KM and the like)...I'm having difficulty with the meaning of the words ontology and taxonomy and thought I'd found an answer in this paper >>
http://itc.scix.net/data/works/att/w78-2003-432.content.pdf

What you seem to be say in your post is that Van Rees has it wrong when he talks about ontology being "a set of well-defined concepts describing a specific domain. The concepts are defined using a subclass hierarchy, by assigning and defining properties and by defining relationships between the concepts et cetera". It sounds to me that he is talking about a very compicated taxonomy...

or is it me?

we have ITIL - based systems and processes implemented in my company, and the reality of the IT ticketing systems (should I say, with Husserl: Lebenswelt ;-) is that e.g. the difference between incident and problem is completely messed up and only the high priests understand that difference and wonder why the stupid users don't get it.

@MarkHarbor: I think that definition you reference is not wrong or different, just a more specialized information science definition/application of the more general Greek usage.

We could also distinguish between the word being a noun to represent a set of concepts (info science usage), or as a focus of inquiry (greek usage). I would think of it more in terms of a graph or hypergraph or set theory, rather than a hierarchical taxonomy.

The opening paragraph of the wikipedia page for "ontology" I think does a good job highlighting the major usages (point being that we cant talk about being, in terms of existence, without talking about what specifically exists and in what manner it exists):

Ontology (from the Greek ὄν, genitive ὄντος: of being (neuter participle of εἶναι: to be) and -λογία, -logia: science, study, theory) is the philosophical study of the nature of being, existence or reality in general, as well as the basic categories of being and their relations. Traditionally listed as a part of the major branch of philosophy known as metaphysics, ontology deals with questions concerning what entities exist or can be said to exist, and how such entities can be grouped, related within a hierarchy, and subdivided according to similarities and differences.

I must admit I'm not quite sure what you're getting at, Dave.

Cynefin is a decision making framework that assists people to take constructive actions in the presence of ambiguity. But you can't prefer some actions over others unless you are implicitly predicting systems behaviour. Even a prediction of unpredictability is a prediction!

So I don't understand the reason for picking on systems dynamics as a theory. Do you believe it is flawed in some way? Are you saying that all attempts to explain the behaviour of complex systems is worthless or futile?

I think you just made my point for me Stephen. It is a characteristic of the assumptions of systems dynamics that they focus on outcome/prediction rather than the NOW.

Perhaps 'prediction' is the word that causes trouble here. A prediction of system's behavior is not the same thing as predicting future states at particular times.

For example, I can predict that the behavior of a repeatedly rolled die will (over time) approximate an evenly distributed set of results. But I haven't said that the outcome of the next 5 rolls will be 6,4,5,5,3.

Similarly with complex systems, there are things that may be appropriate to predict, eg that evolutionary forces will move a system over time to an area of local maximum fitness. I say "may" because this isn't a ironclad rule for all complex systems -- but the general idea that this kind of predictability is possible still applies.

Stephen,

I'm curious why you suggest that, "you can't prefer some actions over others unless you are implicitly predicting systems behaviour."

The point is that some circumstances require actions that you cannot predict the outcome of.

Knowing that you can't predict the outcome, yet acting anyway, informs a very different kind of approach. This is the realm of experimentation, design, intuition and creativity (or 'abduction' as Dave and Max like to call it). This makes sense to act this way in complex environments, but not in simple or complicated ones.

Cynefin helps you diagnose the kind of strategic environment you're operating in and therefore which kind of knowledge is most appropriate. Prediction isn't necessary (or even possible) in the complex and chaotic realm.

Hope this helps clarify a bit.

Very good post... offers some great clarity.

I keep running into people battling symbolic meaning (words, classification, organization, identity, etc...) but who are still hesitant to look in any curious manner at semiotics and symbolic interactionism broadly (in terms of social complexity)... and im not quite sure why. Not in any definitive manner, I just find them to be enlightening in terms of what the founders observed and codified.

People seem to be more interested in battling the meanings of symbols and the symbols of meanings as if there were any true objective essence intrinsic in each, rather than look at the notion that symbols can have multiple meanings and meanings can have multiple symbols, which I feel you have touched on a lot in regards to signification, and the SenseMaker Suite.

I feel that it is this unpredictable (chaotic) generative capability in the present which enables order, complication, complexity to emerge, and as well allows us to contemplate the permanence or impermanence of past, present, and future meanings of symbols and symbols of meanings, giving rise to an illusion of order, complication, and complexity.

The nature of how organization, pattern, identity, unity, etc... is recognized, entrained, amplified, dampened, stabilized, symbolized, communicated, etc... I think is a very strong base for any thought.

The underlying "problem" which persists is the very fact that we all self-signify in the permanent present in regards to the past, present, and future... and I feel many people who are talking about order, complication, complexity, chaos are completely ignoring this simple notion.

So when people individually (which you could replace "people" or "individual" with something more academically general if we want, such as "unity", "organization", etc...) signify the same future, in terms of thinking, believing, and acting as so... the illusion of order in the present arises. and vice versa for the illusion of chaos, complication, or complexity and such in the present.

The more we all signify a word as having a particular meaning in the present, the more it appears that the word has an intrinsic meaning, and intrinsic order regarding its usage. But we have to recognize that this is a "feed-forward" notion. We are constantly feeding forward to the future from the present. And in the present we retain the capability of what will be fed forward, lending the concept of inherent chaos or unpredictability.

You call a child by a certain name enough times and they will grow up until one day they begin declaring "I am Mark Spivey". It is one of the most permanent and persistent things, leading to the illusion of me thinking I am now it.

The ability to contemplate impermanence allows us to meditate on the inherent freedom and chaos and unpredictability in the present, presenting us with infinite opportunities. Change initiatives, as well as organizational initiatives in general, fail when too many different people are holding on to too many different pasts, while still contemplating the permanence of a common unity in the present moving forward into the future.

Companies fail when the stakeholders begin to believe that the company is greater than them, causing them to believe they dont have to persist its signification in the present, mistakenly believing they can walk away and it still carries on.

I think it has a lot to do with expectation of future, or the illusion of order arising in the present because of an assumed certainty in the present against the future, while the unordered side would accept uncertainty leading to less expectation of anything particular.

re: "They see everything in terms of a model, and the success of a model is defined by its predictive capability. We really need to move on from this restrictive thinking. . ."

In April I delivered an invited presentation about the Cynefin framework (with rich attribution to Dave's work) that discussed the utter failings of models in two important areas: U.S. financial system and 'man-made global warming.' In both of these areas the complete failure of the predictive capability of the models should convince decision-makers to abandon the model approach for CAS predictions. Unfortunately, models continue to be a sophisticated and 'smart-looking' surrogate for real thought for those comfortable with their position at the top of the command-and-control hierarchy. Acknowledging the inherent messiness of economic, ecological, or human states is uncomfortable as it means sharing decision-making power. Misplaced 'ordered' approaches in which models are used in a vain attempt to understand or predict a CAS results in ecological damage and human suffering. Conversely, understanding the dangers of entrained thinking and embracing the value of a CAS with weak constraints can improve systems - as indicated by events surrounding the Longitude Prize, Netflix Prize and traders vs. Marines.

Concerning recent posts: I see no need to rearrange the Cynefin framework so it fits into an old 'wine skin' of thinking but rather to understand it fully and spend my time applying it in many different areas. Enjoy the presentation and podcast. It was well received.

Abstract and bio

Scroll down to: "Cognitive Kanban: Improving Decisions in a Complex World," by Mike Sivertsen, at this page for a richly annotated PDF (PowerPoint notes view) and the MP3 podcast.

Is this not a dichotomy like the Heisenberg Principle in quantum mechanics?

In one of the more recent posts on the history of Cynefin is a diagram showing Unordered on the left hand and Ordered on the right. In the complex realm we can find stability but there is not a discernable pattern or history. In the complicated realm there may be a clear pattern, but inadequate evidence to assess its stability.

To sow further confusion, doesn't each realm have an ontology? In the complex we can talk about emergence and history and attractors or local optimization. In the complicated realm we use theory and knowledge and metaphors of structure of all kinds.

I find it hard to keep it straight ... is it a particle or a wave?

Hi Noah,

I'd invite you to consider what "retrospective coherence" means, and Dave's exhortation to "amplify positive patterns and dampen negative patterns".

Once you diagnose a system as complex, you *are* making predictions about the behaviour of the system, like it or not. To wit: (1) that actions will make sense in hindsight, and (2) that acting in a similar way *after positive effects have been observed* will tend to continue to produce further positive effects.

Tony - Heisenberg is more of a paradox/dialectic than a dichotomy but otherwise I'm not sure of the implications here. The essence of the complicated domain in Cynefin is that there is a right answer and it is knowable through use of experts or research, the circumstances you describe would be complex not complicated. Boundaries and attractors are attributes of a compex system, not sure that classes as an ontology per se or why its confusing

RichardIf you identify a system as complex then you are making statements about its nature, to call those predictions is to distort the word, easier to admit you were wrong surely?

Hi Stephen,

Thanks for your response. I think I see where you're coming from but still disagree.

Recognition of the complex mode implies a continual re-evaluation of our actions and their effects.

Just because something worked in one place doesn't mean it will work again elsewhere. And just because something worked in one place in the *past* doesn't mean it will work again the same place at another time.

I wrote about this a while back in a post called "Why the use of best practice can be dangerous and misleading".

I quote;

"Unfortunately our social and organisational lives take place in domains that are complex, dynamic and interactive. Simply reproducing something that works in one place and re-applying it somewhere else can be a naive and dangerous thing to do. Even in the same organisation, if one were to try to re-apply an intervention that worked once at at a different time, it would produce different results because the component parts and the relationships which constitute it have changed. Furthermore, the relationship between the organisation and its environment will have changed as well. So the imposition of the intervention is not only likely to fail, but likely to produce entirely unpredictable and surprising results."

Amplifying or dampening patterns always happens in the present and does not require either, a) an understanding of the reason for their amplification or diminishment, or b) a belief that continued efforts to do so will necessarily produce the desired outcome.

Think of dancing with a new partner. You don't know how they'll respond to more forceful guidance or this move or that. So you experiment, try a little something here, a little something there, until you both adjust to a common rhythm.

None of this requires predictive knowledge; it just requires astute observation, heightened awareness of change and fast reaction times.

"Just because something worked in one place doesn't mean it will work again elsewhere. And just because something worked in one place in the *past* doesn't mean it will work again the same place at another time."

The key thing here I think is bringing in time, in regards to our thoughts, beliefs, and actions about the nature of systems.

To adhere to some key language of Dave Snowden, but with my interpretation added:

To be "robust": Be more hermeneutically assuming of certainty in the present, contemplating the permanence of the past, meditating on the security of the past so that current thought, belief, and action may have more predictable value and meaning against/into the future. Feedback loops enable more robustness.

To be "resilient": Be more hermeneutically accepting of uncertainty in the present, contemplating the impermanence of the past, meditating on the freedom from the past so that current thought, belief, and action will fuel emergence against/into the future. Feedforward loops enable more resiliency.

Also, I would have to say there is little resilient relevance or value in recognizing the nature of a system being anything but chaotic in the present. This simply enlightens me to the fact that the person is holding on to some "assumption of certainty" they aren't yet willing to let go of. What I would advocate is that people come to awareness and terms with their assumed certainties and accepted uncertainties, so they can at least be understanding of their activity.

We have to escape talking about systems as if we cannot affect their very nature in the present, giving rise to a completely different nature.

Ordered, complicated, and complex domains ALL do the exact same things, just on different levels and scales, in regards to assumptions and acceptances of permanence... complexity just takes it farther... while chaotic domains is the true nature of the present social condition. Any extremely simple experiment using linguistics can show us what I mean.

As you move from chaotic, to complex, to complicated, to ordered... assumed certainties increase, and accepted uncertainties decrease... to the point where ordered domains almost become self-fulfilling, until a revolutionary breakdown taking all participants to the edge of chaos, requiring them to order their way out once more.

Is an "organization" chaotic, complex, complicated, or ordered? It depends. It depends on how people think, believe, and act in regards to it and other things, in the present most likely based on how we understand the past and desire the future. We won't know until we look back on today from tomorrow.

Predictive markets exist because assumed certainties exist.

Contemplate impermanence and we can escape predictive markets... but would likely give rise to creative markets or something of the sort.

Give people security, and they dream of freedom, give people freedom, and they dream of security.

I see an ontology as a prescriptive domain in which agents and operators can be derived.
It is also the name of the house of being -- the initial state in which languages -- with syntaxes, taxonomies, and vocabularies take place.
I think you can build a purposeful ontology, such a for these types of active (and passive) concerns -- such as learning, and patterning in social domains.
The most interesting open question is this: what are the system and design thinking requirements for capturing and patterning the "in the moment?" That's what I've been working on.

Noah: Thanks for your thoughtful response. I don't disagree with most of what you've written but apparently my use of the word 'predict' (which I was using in your sense of the word 'theorise') is causing problems.

But regardless, to say "None of this requires predictive knowledge; it just requires astute observation, heightened awareness of change and fast reaction times" is a little intellectually cute.

Your metaphor of a dance is actually quite useful (although note I have zero talent at dancing). There is an expectation when you are dancing with someone that they won't spontaneously explode. Similarly it's unlikely that kicking them in the shins as a "new dance move" will lead to positive outcomes.

There is an expectation of coherence in complex systems. But taken literally, your language abrogates our responsibility as agents to identify coherent outcomes in an intelligent way and filter our possible actions on that basis.

Note that this is different from choosing an action in the expectation of achieving an outcome within a defined timeframe.

Confusing an "ought" with an "is" Stephen?

Mark, Stephen,

Thanks for your responses. An interesting thread struck me between both of your comments.

Mark wrote:

"We have to escape talking about systems as if we cannot affect their very nature in the present, giving rise to a completely different nature."

And then Stephen wrote:

There is an expectation when you are dancing with someone that they won't spontaneously explode. Similarly it's unlikely that kicking them in the shins as a "new dance move" will lead to positive outcomes.

I see more of what you mean here Stephen and think that this is actually a useful contradiction to Mark's earlier point. To wit, the world isn't totally chaotic and certain regularities and laws do exist in the natural world (physical and chemical laws) and to some extent in the social world as well (slowly changing norms and behaviours).

So you are right in that there is much that we cannot effect and is also rather reliable and predictable (per Stephen's point and contrary to Mark's) - just as there are also many things that we can neither effect or predict (such as the daily weather).

Within these bounds of physical and social regularities, however there is still plenty of meaningful room to make things like social interaction sufficiently complex and uncertain as to render them functionally unpredictable.

While we both agree (at least Stephen and I) that certain things are predictable, I would suggest that for all intents and purposes these things belong to the Simple realm and that with the exception of these regularities and rules, most important things in life are variable enough to require a high degree of attention and fast reaction (contrary to Mark's assertion that "Ordered, complicated, and complex domains ALL do the exact same thing", but maybe I'm misunderstanding you here).

I wasn't meaning to be cute or pedantic. I was just trying to say that for most practical purposes prediction isn't necessary or even possible. That doesn't imply powerlessness or inactivity of course (per one of Mark's other point). Simply that the we can take these regularities for granted in most cases and therefore need to focus on the things which we perceive to be regular but actually aren't.

A bit of reference to the literature, just for good measure:

Weick and others observed that the continuous change we see in complex environments is emergent in nature (i.e., complex) and that this means organisations need to excel at “improvisation, translation, and learning." They go on to define a successful strategy as, "one that possesses a change repertoire which enables a response proactively or reactively to strengthen, enlarge, change, and unlearn skills and knowledge, as necessary" (Weick & Quinn, 1999).

Thus my dancing metaphor and one of the core values of sensemaking, in my opinion.

Let me try and clarify some things, because what I feel yall are referring to as contradictions are really just me taking the same principles to application at many more levels, both macro and micro.

I will use time as a metaphor:

The past conditions and informs (feed-forward) the present, but does not govern it...

Similarly, the present, includes the past, but offers opportunity for "elaboration"... as well as the future includes the present.

We only "live" in the present though...

What I feel may be claims of present condition (order, complication, complexity), need to recognized as observation of past condition being contemplated as persistent or permanent into the present, but merely on a superficial level, and certainly not on an actual level.

Notice my use of language in my comments referring to "expectation", "assumption of certainty", "acceptance of uncertainty", etc...

So Stephen's comment is not a useful contradiction to mine, but actually a useful example of my comment. "There is an expectation..." When I see a statement like that, it affords me the understanding of there being a certain expectation of permanence which was observed and or concluded of condition in the past, which is being contemplated into the present, but has not yet been observed empirically in the present.

The problem with scientific reasoning is that empirical evidence must always precede conclusion, but empirical evidence never stops, therefore conclusion must never stop... The present is constant, meaning that in the present there is new empirical evidence interacting with old, so to expect similar output is a fallacy... which is why I would advise the use of language such as "assumption of certainty" or "contemplation of permanence".

No science would ever dare conclude that ANY thing is predictable, but when we disconnect ourselves from such petty observation, we assume so...

To say that some thing is predictable would be to take a walk in the future looking back back at the present. No degree of statistical probability can account for present opportunity. What many people mistakenly believe to be "laws" of nature are simply assumed judgements made of past empirical evidence, which are incapable of including the present as justification.

A machine is not predictable... it is merely more evidence of underlying patterns remaining in tact...

Many people in the complexity community have taken "abduction" to heart, but have failed to apply it in the constant present.

From an article title "An Abductive Theory of Scientific Method" (Brian Haig):

"The theory of method outlined in this article is a broader account of scientific method than either the inductive or hypothetico-deductive theories of method. This more comprehensive
theory of method endeavors to describe systematically how one can first discover empirical facts and then construct theories to explain those facts. Although scientific
inquiry is often portrayed in hypothetico-deductive fashion as an undertaking in which theories are first constructed and facts are then gathered in order to test those theories, this should not be thought of as its natural order. In fact, scientific research frequently proceeds the other way around. The theory of method described here adopts this alternative, facts-before-theory sequence, claiming that it is a search for the understanding of empirical phenomena that gives explanatory theory construction its point. With this theory of method, phenomena exist to be explained rather than serve as the objects of prediction in theory testing."

To say that anything is "ordered" in the present would required us to conclude that any and all related phenomena has concluded... but this is not so if time progresses allowing new phenomena to both emerge and interact. Notice that I am not saying that we do not have the ability to act as though things were not ordered, which certainly can contribute to an observation of self-fulfilling order.

What you may conclude to be predictable and ordered, even complicated or complex in the present, I would challenge anyone to dig deep down and expose the contemplations of permanence, the assumed certainties, etc... that it is based on, even if they are believed as truths, and then think about them yourself, but also search for people who are currently challenging that dogma.

I can illustrate using this line of thinking, how a company, a country, a marriage, a machine, etc..., may be considered "ordered", "complicated", "complex", in the present... as well as expose what the person who made that comment must be contemplating to be permanent in order to arrive at such conclusion... as well as illustrate how to bring them to the edge of chaos to see that which they thought to permanent is in fact conditioned, and therefore impermanent... so they can look at assumption dead on, and at the very least recognize it as such.

And to say that this illustrates "powerlessness" or "inactivity" is not my point either, but actually the opposite. The point is to recognize how much power we have over the present, rather than how much power conclusion of the past have over us in the present (predictability, ordered domains, etc...).

@NOAH: "I see more of what you mean here Stephen and think that this is actually a useful contradiction to Mark's earlier point. To wit, the world isn't totally chaotic and certain regularities and laws do exist in the natural world (physical and chemical laws) and to some extent in the social world as well (slowly changing norms and behaviours).

So you are right in that there is much that we cannot effect and is also rather reliable and predictable (per Stephen's point and contrary to Mark's) - just as there are also many things that we can neither effect or predict (such as the daily weather)."

You see Noah, I believe this is my precise point, and I will use this as an example (not to be argumentative, because I am trying my best to illustrate that we really all agree, just differ in terms of levels of application). I am saying more so that different hermeneutics exist, and each hermeneutic is made up of contemplations of permanences and impermanences in temporal terms, which lead that hermeneutic to certain judgements and expectations, in terms of objectives, and subjectives (drawing heavily on my own work and expansions related to Semiotics and Symbolic Interactionism). And it is precisely this which gives rise (emergence) to the appearance of systemic complications, complexities, and orders as truly existing at any point. The very fact that in your comment concluded that certain things are predictable, and I may not share that belief, is the very root of complexity, moving up to complicated, to order (shared beliefs)... and the very fact that it is POSSIBLE for us to have or not to have these differences or sameness in the present is evident of chaotic conditions in the present.

I do note that I am not addressing specifically differences between conscious unities (a human), and unconscious unities (a rock)... but moreso in terms of a unity capable of hermeneutic thought, belief, and action.

If a group of hermeneutic unities (people) think, believe, and act in the present in regards to the same contemplations of permanence, then it will appear that order pervades that system... but at each point in the ever constant present, each hermeneutic unity reserves the capability of contemplative adjustment.

If you and I use a word for the first time, it is highly likely that you and I will share an intimate experience with it and realize the meaning we LENT the word as a symbol... but over time we may use it so much, as well as entrain others to its particular usage, that we start to believe that the word actually means something intrinsically, and will begin to debate when we see new unthoughtof uses... because we contemplate it's permanence from the past into the future... but certainly not because of any inherent order or predictability, etc... In the present you and I reserve the chaos (unpredictability).

Organizational Change initiatives fail when they do not address and affect hermeneutic unity contemplations of permanence and impermanence.

User Interface Design plays off the same thing. Writing a book in a particular language, the same. Me writing this blog comment in the way I choose, rather than in some new random language (which I am certainly capable of right now), the same...

Order pokes its nasty head in when assumptions are high, which is mistakenly observed as an intrinsic nature of the system itself.

A system will collapse when those that caused it to emerge begin to think, believe, and act as though it was greater than them.