Dave Snowden’s Cynefin framework
Copyright Graham Berrisford. One of several hundred papers at http://avancier.website. Last updated 09/04/2019 19:40
A
system theorist may say our behaviour is shaped by
our internal state, which is in turn shaped by external events.
A
social systems thinker may say our behaviour is
influenced by our place, our experience, culture etc..
Cynefin is a Welsh word covering habitat, place, experience,
culture etc.
http://cognitive-edge.com/uploads/articles/Origins_of_Cynefin.pdf
http://mandenews.blogspot.co.uk/2010/08/test3.html#comment-form
The Cynefin framework is commonly
shown as below with Disorder in the middle.
Complex Probe - Sense – Respond emergent
practice. --------------------------------Dis Chaotic Act - Sense – Respond novel
practice. |
Complicated Sense - Analyze – Respond good
practice. order------------------------------- Simple Sense - Categorise – Respond best
practice. |
Dave Snowden says the framework not a categorisation of systems (see below for that).
It
lies at the problem analysis end of “systems thinking”, rather than the system
theory end.
It is a “sense-making framework”.
It is a self-organising tool, a
guide to choosing between processes or methods for systems analysis.
Snowden says when first placed in
a situation, start in the central Disorder domain and then decide which kind of
behaviour to adopt.
You may move from any domain to an
adjacent domain, but not from simple to chaotic
Mistakenly behaving in the simple
way is dangerous; it can lead you over the cliff from simple into chaos and
failure.
The framework is a tool for working out how to solve a human problem, which approach or method to use
Given a task, a problem to solve,
or transformation to make, we can use the framework to decide how we or other
people best behave.
The problem domain is |
The path from cause to effect (event to result, input
to output) |
Our behaviour pattern should be |
We should |
Simple (ordered) |
Obvious |
Sense - Categorise - Respond |
apply established best practice. |
Complicated (ordered) |
Requires analysis, investigation
or the application of expert knowledge |
Sense - Analyze - Respond |
apply a good/expert
practice (any of several) |
Complex |
Can be recognised after testing,
but not determined in advance, |
Probe - Sense - Respond |
experiment and test what practice
emerges |
Chaotic |
None, there appears to be no
relationship |
Act - Sense - Respond |
act fast using a novel
practice to reach order |
Disorder |
Unknown, we don’t know which of
the above 4 domains we are in |
|
work out what domain we are in! |
The framework could be used to justify the choice of design/development method to managers.
If Conrad Taylor has it right, Dave Snowden speaks of three kinds of system: ordered, complex and chaotic.
He distinguishes them as shown in the table below; I have added notes in [square brackets].
Cynefin terms |
Agents [their roles and rules] |
Activities |
Snowden says |
Ordered systems |
Constrained |
are predictable [meaning deterministic] |
Most management theory is predicated on this view. |
Complex systems |
Lightly constrained |
Agents interact with each other and external entities, and co-evolve [their roles and rules] |
A better model for understanding human societies, organisations and markets. |
Chaotic systems |
Unconstrained |
Agents act independently of each other |
Some assume markets are chaotic; but agent behavior is analysable using statistical analysis and probability. |
The difficulty with this is that
it confuses social entities and social systems.
The actors in a human social
entity are somewhat constrained not by one set of roles and rules, but by many.
They have to choose not only how
to act in one system, but also which system’s roles and rules take precedence.
As Ashby would have said, one
social entity may realise infinite social
systems.
The Cynefin framework is a tool for situation thinking rather than systems thinking.
It is the
kind of mental model loved by management consultants.
It is a sense-making framework for those wanting to make a “management intervention”.
It is a tool for problem domain analysis – for choosing how to approach a business problem.
The four-way classification of
situations or problem domains may be useful.
But in an
exposition of system theory, there some difficulties to address.
To call every problem, situation or social network “a system” is unhelpful.
The big issue in social systems thinking is the confusion between:
· A social network in which people communicate
· A social system in which people realise role and rules.
Ackoff was inconsistent; he started off distinguishing abstract
systems from concrete systems.
Later, he spoke of a human organisation as being a system regardless of any abstract system description.
This confusion of a human network with a system runs through much
systems thinking discussion.
This table is an attempt to separate three concepts that have
become entangled.
Abstract social system |
A set of roles
and rules (the logic or laws
actors follow) |
Concrete social system |
Actors
playing the roles and acting
according to the rules |
Social network |
Actors who inter-communicate
and act as they choose |
With respect to a system, social
networks might be classified into three kinds.
·
In
a formal social network, actors
are wholly constrained by the roles and rules of the system.
·
In
an informal social network, actors
are somewhat constrained by the roles and rules.
·
In
a chaotic social network, actors
completely ignore the roles and rules.
Read Social networks versus social systems for further discussion.
Dave Snowden says The Cynefin
framework is not a categorisation of systems.
However it does tend to be
regarded that way.
For just for fun, let us try to
map it to a system classification.
One might try to map the Cynefin framework to Ackoff’s scheme as below.
Cynefin
problem domain |
Ackoff’s system class |
Simple |
Mechanistic |
Complicated |
Animate |
Complex |
Social |
Chaotic |
Ecological |
But actually, the two schemes are unrelated
The Cynefin framework is looser, more generic, and used differently.
Avancier suggests classifying an
operational system by scoring it on nine dichotomies or scales on which systems
might compared.
This table selects four of Avancier’s nine scales that can be related the Cynefin framework.
Bear in mind that deterministic means predictable in theory, predictable means predictable in practice.
Facile |
Easy transformations / services |
Difficult transformations / services |
Complex |
Predictable |
Inputs have predictable effects |
Inputs have unpredictable effects |
Unpredictable |
Rigid |
Fixed components & processes |
Refactorable
components & processes |
Malleable |
Static |
Constrained, rule bound |
Adaptable, self-directing, evolving |
Changing |
I propose a rough mapping between the two schemes as follows.
Cynefin problem
domain |
Avancier’s
system scales, applied to a human problem domain |
|||
Simple |
Predictable |
Facile |
Rigid |
Static |
Complicated |
Predictable |
Complex |
Malleable |
Static |
Complex |
Unpredictable |
Complex |
Malleable |
Changing |
Chaotic |
Unpredictable |
Complex |
Malleable |
Changing |
The tools are different. So the mapping here is done for fun.
Ackoff’s system
class |
Avancier’s
system scales |
|||
Mechanistic |
Predictable |
Facile |
Rigid |
Static |
Animate |
Unpredictable |
Complex |
Malleable |
Static |
Social |
Unpredictable |
Complex |
Malleable |
Changing |
Ecological |
Unpredictable |
Complex |
Malleable |
Changing |