An Introduction to Complex Adaptive System Theory

1
An Introduction to Complex Adaptive System Theory
Key Concepts of Complexity Science

• Dr Carol Webb
• Manufacturing Dept, Bldg 50, RmF9b,
• School of Applied Sciences

2
Why Complexity Science?

• Problem with legacy of scientific management
• Traces of scientific management in much
  management theory and discourse
• Dominant metaphor mechanical, reductionist,
  linear
• OK for target driven activities
• But, something else needed for
• What emerges between people
• Non-linearity
• Uncertainty unpredictability

3
Why Complexity Science?

• Complexity refers to the condition of the
  universe which is integrated and yet too rich and
  varied for us to understand in simple,
  mechanistic or linear ways.
• We can understand many parts of the universe in
  these ways but the larger and more intricately
  related phenomena can only be understood by
  principles and patterns not in detail.
• Complexity deals with the nature of emergence,
  innovation, learning and adaptation
• Lissack, M. (1997). Mind your Metaphors
  Lessons from Complexity Science in Long Range
  Planning, Vol. 30/2 pp294

4
Complexity Science changing the way we think

• Complexity theory deals with systems which show
  complex structures in time or space, often hiding
  simple deterministic rules. Complexity theory
  research has allowed for new insights into many
  phenomena and for the development of a new
  language. The use of complexity theory metaphors
  can change the way managers think about the
  problems they face. Instead of competing in a
  game or a war, they are trying to find their way
  on an ever changing, ever turbulent landscape
• Lissack, M. (1997). Mind your Metaphors
  Lessons from Complexity Science in Long Range
  Planning, Vol. 30/2 pp294
• Weicks concept of sensemaking can be
  summarized as an organisations need to interpret
  and make sense of the environment around it if it
  is to survive
• K. E. Weick and K. H. Roberts, Collective Mind
  in Organisations Heedful Interrelating on Decks,
  Administrative Science Quarterly, September
  (1993), And K. E. Weick, Sensemaking in
  Organisations, Sage Press, Thousand Oaks, CA
  (1995).

5
Complexity Science Changing what we do
"Complexity science offers a way of going beyond
the limits of reductionism, because it
understands that much of the world is not
machine-like and comprehensible through a
cataloguing of its parts but consists instead
mostly of organic and holistic systems that are
difficult to comprehend by traditional scientific
analysis. it remains very much a science -
that is, a body of observation and analysis of
natural phenomena - rather than being deep
theory" (Lewin, R., 1999)
However, let us consider some of the theory
generated by this body of observation
6
Complex Adaptive Systems (CAS)?

• Ever wondered how to describe

7
Complex Adaptive Systems

• A flock of birds might be thought of as a
  complex adaptive system. It consists of many
  agents, perhaps thousands, who might be following
  simple rules to do with adapting to the behaviour
  of neighbours so as to fly in formation without
  crashing into each other.
• A human being might be seen as a network of
  100,000 genes interacting with each other. An
  ecology could be thought of as a network of vast
  numbers of species relating to each other. A
  brain could be considered as a system of ten
  billion neurones interacting with each other.
• In much the same way, an organisation might be
  thought of in terms of a network of people
  relating to each other. Complexity science seeks
  to identify common features of the dynamics of
  such systems or networks in general
• (Stacey 2003a238).

8
Complex Adaptive Systems
9
Complex Adaptive Systems

• A Complex Adaptive System (CAS) consists of a
  large number of agents, each of which behaves
  according to some set of rules
• These rules require the agents to adjust their
  behaviour to that of other agents
• In other words, agents interact with, and adapt
  to, each other
• Out of these interactions, novelty, spontaneity
  and creativity emerge sometimes in
  unpredictable ways

10
Think of a flock of birds as a complex adaptive
system

• Complexity science seeks to
• identify common features of the dynamics of such
  systems or networks in general
• The emergent outcome in the case of the
  self-organisation of the birds is the order
  present in the formation of the flock.

11
Innovation as an emergent outcome of system-wide
self-organisation how?

• Key questions
• How do such complex non-linear systems with their
  vast numbers of interacting agents function to
  produce orderly patterns of behaviour (or
  innovation)?
• How do such living systems evolve to produce new
  orderly patterns of behaviour (or innovation)?

12
CAS Methodological considerations

• No search for an overall blueprint for the whole
  system
• model agent interaction
• each agent behaving according to their own
  principles of local interaction
• No individual agent, or group, determines the
  patterns of behaviour
• bottom-up emergence

13
Ants as an analogy to convey the meaning
potential of self-organisation to solve business
problems

• To understand the power of self-organisation,
  consider how certain species of ants are able to
  find the shortest path to a food source merely by
  laying and following chemical trails. Individual
  ants emit a chemical substance a pheromone
  which then attracts other ants. In a simple case,
  two ants leave the nest at the same time and take
  different paths to a food source, marking their
  trails with pheromone.
• The ant that took the shorter path will return
  first, and this trail will now be marked with
  twice as much pheromone (from the nest to the
  food and back) as the path taken by the second
  ant, which has yet to return.
• Their nest mates will be attracted to the shorter
  path because of its higher concentration of
  pheromone. As more and more ants take that route,
  they too lay pheromone, further amplifying the
  attractiveness of the shorter trail.
• The colonys efficient behaviour emerges from the
  collective activity of individuals following two
  very basic rules lay pheromone and follow the
  trails of others (Bonabeau and Meyer 2001108).

14
Computer programmes to study CAS

• Genetic algorithms
• developed by John Holland of the Santa Fe
  Institute (Holland, 1992)
• The Boids simulation
• developed by Reynolds (1987) to simulate the
  flocking behaviour of birds
• The Vants simulation
• developed by Langton (1996) to simulate the
  trail-laying behaviour of ants
• The Tierra simulation
• developed by Ray (1992) using the analogy of
  biological evolution to evolve computer
  programmes.

15
Conversation in complexity science method
Analogies from the complexity sciences provide
insight into stabilising features of
communicative interaction.
Narrative and propositional themes that Stacey
describes as organising themselves into
conversation can take various forms (Stacey
2003a362) fantasies myths rituals ideology
culture gossip rumour discourses and speech
genres dialogues discussions debates and,
presentations.
These are responsible for organising the
experience of relating in different ways, by
e.g. selecting what is to be attended to
shaping how what is attended to is to be
described selecting who might describe it
accounting by one to another for their actions
articulating purpose in the form of themes
expressing intentions (Stacey 2003a 363)
Importance of acknowledging feelings,
reflection-in-action, and abstract thinking
(Stacey, 2001)
16
(No Transcript)
17
Self-Organisation

• No single person absolutely in command or control
  of the situation
• No-one really planning and managing the situation
  even though they might think they are
• Obvious hierarchy in complex systems are not
  immediately noticeable
• Agents continuously organising themselves without
  a leader
• Agents interacting with each other in simple ways
• Complex systems structure themselves out of
  themselves
• Interacting elements act according to simple
  rules
• Order is created out of chaos

18
Emergence

• You cant easily predict what is going to happen
  next
• The way people are interacting appears to be
  random
• You see new things emerging from interactions
• If you were to look on a wide scale there might
  be some patterns emerging
• Patterns emerge from interactions
• Patterns inform the behaviour of a system
• New qualities arise through particular types of
  networks
• Higher complexity is produced out of many simple
  components
• Each individual component outgrows usual
  capabilities e.g. people outgrow their
  competencies.

19
(No Transcript)
20
The edge of chaos

• Not a fixed state a transitional phase!
• Lots of creative activity going on
• Lots of transitions and changes from one state to
  another
• Living networks reside in a critical phase
  between chaos and order where networks find
  creativity and stability in an optimal balance
• Living systems are most creative, with the
  greatest potential for discovering order that
  expresses an emergent property for the whole
  system, when they are living near the edge of
  chaos
• Living systems naturally undergo transitions from
  current order to chaos, from which emerges new
  order.

21
(No Transcript)
22
Diversity

• If differences are not flattened out or levelled
  change happens easily
• Interaction and change appears flexible
• The system seems strong in these cases
• Networks combine the most different variants,
  characters, functions
• High diversity creates more possibilities to
  react flexibly, on environmental changes
• The greater the variety within the system the
  stronger it is
• Ambiguity and paradox abound
• Contradiction is used to create new possibilities
  to co-evolve with their environment.

23
(No Transcript)
24
History Time

• History and time irreversible you cant go back
  in time and change things
• Some specific decisions brought you to where you
  ended some you were aware of, others you were
  not (what might have been???)
• In a social context, the series of decisions
  which an individual makes from a number of
  alternatives partly determine the subsequent path
  of the individual
• Before a decision is made there are a number of
  alternatives after, it becomes part of history
  and influences the subsequent options open to the
  individual
• Unique histories mean every decision the
  organisation makes is context specific (therefore
  questions the idea of best practice and one
  size fits all treatments)
• Also, think about path dependency e.g.
  technological path dependency systems are
  locked into using dominant tools and processes
  because of historical factors
• Think about our present day road systems these
  often date back to Roman times!

25
(No Transcript)
26
Unpredictability

• Detail and order of outcomes not determined by an
  elite group
• Not really possible to forecast or control
  behaviour in details
• No actions isolated
• Interlinked groups or networks with lots of
  people acting and reacting among each other
• Things happening in one place create consequences
  elsewhere and vice versa
• Due to complicated interrelations, its very
  difficult to foresee or to control behaviour of
  the nodes of the network, when reacting to
  impulses (from outside or inside the network).
• Emergent order is holistic a consequence of
  interactions between elements of the system
• All systems exist within their own environment
  and they are also part of that environment
• As their environment changes they need to ensure
  best fit
• When they change, they change their environment
  too

27
(No Transcript)
28
Pattern Recognition

• You cant always see direct and proportional
  links of cause and effect
• People and groups dont really link in random
  ways
• Small numbers of people are loosely coupled to
  others
• Small changes are amplified - You can see big
  effects coming from small changes
• You see patterns of activity being repeated over
  and over again
• The ways agents in a system connect or relate to
  each other is critical to the survival of the
  system
• From these connections patterns are formed and
  feedback disseminated
• Relationships between agents are more important
  than agents themselves
• Self-organised, living networks always show
  similar patterns.
• Feedback is the systems way of staying constantly
  tuned to its environment and landscape and
  enables the system to re-adjust its behaviour.
• In far from equilibrium conditions change is
  non-linear, so small changes can be amplified,
  and produce exponential change
• Novel, emergent order arises through cycles of
  iteration in which a pattern of activity, defined
  by rules or regularities, is repeated over and
  over again, giving rise in coherent order.

29
6 Properties of Complex Adaptive Systems (CAS)

• Self-Organisation Emergence
• Diversity
• The Edge of Chaos
• History Time
• Unpredictability
• Pattern Recognition
• there are more (!) these are just some basic
  principles
• Dont forget interconnectivity and the importance
  of networks!
• Networks are the assumed context of CAS
• (also see references in the bibliography for how
  CAS theory is applied to different contexts)

30
Linking theory and method

• Systems practice as a way of managing in
  situations of complexity
• Systems thinking shows there is no right answer
  when dealing with complexity
• We avoid terms like manage and managed with
  deterministic overtones in favour of managing
  which is an active process associated with daily
  living
• Need to see the parts in the context of the whole
• Engaging with complexity entails
• Engaging in situations of complexity
• Using systems or complexity thinking to learn
• Learning our way towards purposeful action that
  is situation improving

31
Conversation in complexity science method

• Analogies from the complexity sciences provide
  insight into stabilising features of
  communicative interaction.
• Narrative and propositional themes that Stacey
  describes as organising themselves into
  conversation can take various forms (Stacey
  2003a362)
• fantasies myths rituals ideology culture
  gossip rumour discourses and speech genres
  dialogues discussions debates and,
  presentations.
• These are responsible for organising the
  experience of relating in different ways, by
  e.g.
• selecting what is to be attended to shaping how
  what is attended to is to be described selecting
  who might describe it accounting by one to
  another for their actions articulating purpose
  in the form of themes expressing intentions and,
  justifying actions in the form of themes that
  express ideology (Stacey 2003a 363).
• Importance of acknowledging feelings,
  reflection-in-action, and abstract thinking
  (Stacey, 2001)

32
What Enables Self-Organising Behaviour in
Businesses?

• Self-organising behaviour will naturally occur
  without addressing what causes it
• Behaviour is self-organising when people (agents)
  are free to network with others and pursue their
  objectives
• Even if this means crossing organisational
  boundaries created by formal structures
• Self-organisation as the natural default
  behaviour
• Organisation studies recognise barriers to such
  freedom in bureaucratic structure
• Understand self-organising behaviour in
  adaptation to change by applying concepts of
  organisation theory and organisation behaviour

Coleman, H. J. (1999)
33
What Enables Self-Organising Behaviour in
Businesses?

• Diversity seen as important in context of
  interconnected people translating ideas into
  innovation
• Agents co-evolve with the environment of fitness
  landscapes through a process of self-organisation
  intended for both survival and growth from
  innovation
• Impetus for creativity comes from shadow system
  of learning communities with enough diversity to
  provoke learning but not enough to overwhelm
  legitimate system and cause anarchy
• Degree of connectivity between agents in a
  system necessary variety in behaviour depends on
  strength and number of ties
• Few and strong ties producing stable behaviour
  too little for effective learning
• Many and weak ties producing unstable behaviour
  too much variety for effective learning

Coleman, H. J. (1999)
34
What Enables Self-Organising Behaviour in
Businesses?

• To operate at the edge of chaos, agents and
  systems balance canalisation and redundancy
• Need for creative tension and experimentation
• Space for creativity in an organisation
• Tension between over-control (in legitimate
  system) and chaos (in shadow system)
• Confident employees risk-takers and
  experimenters
• Some organisational stability required and some
  order necessary for employees to recognise
  novelty
• Organisations learn when there is new information
  combined with knowledge and applied to new
  opportunities provided by changes in the external
  environment
• People in learning communities seize such
  opportunities to be innovative
• If structure is flexible enough the firm can
  adapt and form new project teams or even new
  business units, or found new companies

Coleman, H. J. (1999), Eden and Ackermann, (1998)
35
What Enables Self-Organising Behaviour in
Businesses?

• Organisational open systems assumed
• Open to flows of data and information
  facilitating learning and construction of new
  knowledge
• Goal is to encourage experimentation (planned or
  naturally occurring)
• Some failure needs to be tolerated (e.g. Post-It
  notes developed from the failure of a search for
  an adhesive substance)
• Judicious ignoring of local constraints helps
  avoid being trapped on poor local optima
• Entrepreneurial behaviour is spontaneous in
  response to perceived opportunities to create an
  organisation

Coleman, H. J. (1999)
36
What Enables Self-Organising Behaviour in
Businesses?

• Organisational theory and organisational
  behaviour
• Need for innovation leads to particular emphasis
  on knowledge management
• Adaptation in turbulent environments necessary
• Small teams (or cells) pursue entrepreneurial
  opportunities and knowledge sharing among
  themselves (leads to a potent organisation)
• Operating logic based on flexibility with
  knowledge sharing in place of hierarchical
  controls
• Stability created for confident risk-taking and
  experimentation
• New knowledge constructed in communities of
  practice (COPs)

37
What Enables Self-Organising Behaviour in
Businesses?

• Organisation Design
• Organisation design/structure can facilitate
  change by being flexible
• Design org for purpose of evolution with the
  changing environment
• Design for emergence by avoiding rigidities of
  bureaucratic hierarchy
• Create org environments not inhibiting
  evolutionary change and accept discontinuous
  change
• Leadership may be anywhere, and everyone is a
  champion of change
• No need to bust bureaucracy because there is none
• When an organisation is operating on the edge of
  chaos, not even its leaders can know its future
  direction
• Becomes relevant to operate in a mode of inquiry,
  surfacing and questioning assumptions

Coleman, H. J. (1999)
38
What Enables Self-Organising Behaviour in
Businesses?

• Loose-tight controls
• Freedom of activity
• Relative autonomy within boundaries
• Management confidence and trust in employees to
  act according to shared values
• Tension between empowerment and control reached
  through accountability
• Satisfying human needs for interaction to obtain
  other needs
• Computers and telecommunications increase
  interconnectedness of people and speed of sharing
  knowledge and information
• Empowerment
• Staff taking initiative - Intrinsic motivation in
  staff to contribute
• Enabling feelings of meaning in work, autonomy,
  choice, and having an impact on outcomes
• Releasing self-motivation of employees to take
  responsibility by trusting them to think,
  experiment and improve

Coleman, H. J. (1999)