HOW TO USE DIALOGUE AS A PRACTICAL WAY TO
START APPLYING THE PRINCIPLES OF
COMPLEXITY THEORY TO A TRADITIONAL
ORGANISATION
Elisabet Wreme
Australian Quality Council
Level 2, 700 High Street
Kew East, VIC 3102
Ph (61-3) 9810 8717 Fax (61-3) 9819 7165
E-mail e.wreme@aqc.org.au
There have been many articles written about chaos and complexity theories and how
they apply to oraganisations, but there is little written about how a traditional
organisation can actually start to apply these theories.
Stacey (1997) argues that you can view the organisation as two interlinked systems.
The legitimate system is described by the organisation chart, rules and procedures,
while the shadow system is a web of personal networks. He believes that it is the
shadow system that has creativity, the capacity to evolve and go through transitions
and that the shadow system impacts the legitimate system over time. However, he does
not give much assistance in relation to using these insights. The thesis of this paper is
that the use of dialogue, as described by Bohm (1989) and Isaacs (1993) is a possible
methodology for giving the shadow system a “head start”. A place for different minds
to meet and evolve together.
Introduction
There have been many articles written about chaos and complexity theories and how
they apply to organisations, but there have been very little written in relation to how
an organisation can actually start to apply these theories. The thesis of this paper is to
describe a workable way for an organisation to start applying chaos and complexity
theories through the use of the discipline of dialogue. However, dialogue has to be
seen as something bigger than a tool to facilitate Argyris’s “double loop learning” to
ensure it uses the principles of chaos and complexity theory. The paper starts with an
overview of chaos and complexity theory and how it applies to organisations. It
continues with a description of dialogue and how the application of dialogue can help
an organisation actively apply learnings from chaos and complexity theory. It
concludes by highlighting some implications for an organisation heading down the
dialogue path as well as future research questions.
Today’s Dominating Paradigms
Many authors like Stacey (1997) and Tetenbaum (1998) argue that today's dominant
view of how an organisation function is based on thoughts going back to Newton’s
mechanistic model of the universe and to Darwin’s model of evolution by competitive
selection.
The core of Newton’s thinking is the law of motion, which suggest the world is a well
behaved machine and therefore the relationships between cause and effect are simple
and linear. The belief is that if you would just know enough about the e& ng
conditions, you can accurately predict what is going to happen next. This mechanistic
worldview can also be found within organisations. It has lead to a focus on how to
predict and control ones surroundings. Organisational success depends upon creating
a stable system that can generate predictable result. If there are any deviations, it is up
to management to correct the situation and reestablish a new equilibrium. As a result
we have seen organisations focus on imposing order from above through command-
and-control leadership, as well as the use of visioning and strategic planning. For
instance, the purpose of the once highly popular scientific management approach,
(Taylor, 1911) was consistent with ensuring efficiency, regularity and predictability.
Stacey (1997) argues that the neo-Darwinian philosophy believes in life evolving
through a process of random mutations combined with competition. The change in
one species will impact the overall system and competition will select those systems
with the greatest chance of survival. This belief relies heavily upon chance. The
systems that happen to have the characteristics required by a change in the
environment is the one that survives. When this view gets applied to organisations, it
is usually modified so that managers can choose successful mutations in advance and
be ready to implement these mutations when the time is right. Hence there is also a
belief in predictability, which brings this view close to the Newtonian way of
thinking.
Chaos Theory
Today’s chaos and complexity theories have emerged from studies in a wide range of
disciplines like mathematics, fluid dynamics, physics and meteorology. The definition
of chaos theory, as originally formulated in mathematical physics, is “The qualitative
study of unstable aperiodic behavior in deterministic nonlinear dynamical system”
(Kellert, 1993). This expression sounds more difficult than it really is. Aperiodic
simply means that the system never settles into a regular sequence of values, which
means the system can never reach true equilibrium. A dynamical system is a system
that changes over time (Devaney, 1992). A dynamical system can change either in a
linear or nonlinear way. If it is linear, it changes in a “straight line”. An increase in x
will always cause a proportional increase in y. A nonlinear change does not follow a
straight line. A change in x will cause different changes in y depending on previous
events. Feedback is therefore a vital aspect of chaos theory. A tiny change in the
system can sometimes create dramatically changes in the future. The most referenced
example is the “butterfly effect” were a flapping of the wings of a butterfly in the
Amazons could cause a hurricane some weeks later in the US (Lorenz, 1963).
Chaos theory models are also bounded and deterministic, which means an underlying
pattern of order can be found. A few equations can describe seemingly random events.
Waldrop (1992) describes how Reynold was able to simulate bird-flocking behavior
through having bird-like objects called “boids” flying across the computer screen. Just
like real birds, the boids formed into flocks, flowed smoothly around objects and
merged together again in a constantly changing formation. These behaviours were
created using only three rules: (1) fly in the direction of other objects, (2) try to match
the velocity with neighboring boids and (3) avoid bumping into things. This is the
beauty of chaos theory; simple deterministic equations can generate complex and
random outcomes. All in all, “chaos” is in many ways a misnomer and may be better
described as in Prigogine and Stenger’s definition (1984) as the study of complex and
dynamic systems that reveal pattern of order out of seemingly chaotic behaviours.
Take for instance the creation of snowflakes. They are always recognizable through
its six-sided shape, but each snowflake is at the same time unique. Similarly, each
human is different, but we recognize one when we meet one. Stacey (1992) calls this
“bounded equilibrium”.
Complexity Theory
Complexity theory is according to Johnson and Burton often used interchangeably
with chaos theory. However, in reality it is a collection of theories that use similar
concepts to chaos theory, but they loosen some of the strict restrictions like being
deterministic. This makes complexity theory much more applicable to organisational
theory. A system within complexity theory can “learn” or change the rules as times
moves on. In this theory, determinism and free will can coexist. (Begun, 1994). The
systems are deterministic in that the driving forces of systems can be specified, but at
the same time they have free will in that their futures depend on unpredictable shifts.
This shifts often happen due to self-learning.
The use of complexity theory, instead of chaos theory, addresses one of the major
criticisms of applying this type of thinking to social systems, which is that social
systems are not deterministic because they have a free choice and can learn over time.
However, it does not address another major concern that some authors, like Johnson
and Burton (1994), have in relation to social systems having much more complex
rules and behaviours than other adaptive systems. More research is required in this
area to see if the complexity of the rules will impact how the self-organising systems
work and evolve.
Self-Organising Systems
Complexity theory takes us into the territory of self-organising systems. The most
common approach is for researcher in this area to conceptualize all of nature and its
many subsystems as complex systems (Stacey, 1997). These systems consist of many
autonomous agents that interact with each other according to their own principles and
rules. For instance, the brain can be seen as a complex adaptive system with the
neurons as the agents. Other examples include flocks of birds, colonies of termites, in
fact the whole ecology of species on earth including our own organisations. Even the
human mind can be viewed as a complex adaptive system according to Stacey (1997).
In this case the agents can be seen as the imaginal symbols we use. The world is build
up be nesting systems. What is a system at one level will be an agent on a higher
level.
Researching these types of systems have revealed some major insights. At low levels
of energy/information flow, and when each agent is connected to and interacting with
only a few other agents, the system displays the dynamics of stability. It behaves very
much like the Newtonian way of thinking, a small disturbance will soon be corrected
and the system returns to a predictable behavior. However, if the level of
energy/information flow is very high and each agent is interacting with a very large
number of other agents, the system becomes explosive and there is a real possibility it
will disintegrate when it comes up against a constraint. This system amplifies any
deviation instead of damping it.
The real insight is, however, that at some critical level of energy/information flow and
connectedness between the agents, the system reaches a transition phase where new
forms can emerge. For instance, when low levels of heat are applied to a particular
gas, the molecules moves in random ways, but when the energy levels are increased to
a critical point, the molecules spontaneously self organize so that they point in the
same direction. A laser beam is born as a result. A new form of stable behavior has
emerged out of the disorder. Wheatley (1992) explains the concept in simple terms:
“Life seeks order in a disorderly way... mess upon mess until something workable
emerges.”
We can apply the same way of thinking to living and adaptable systems. These
systems are still complex and consisting of many agents, but with one major
difference, the rules and principles that are used by the agents evolve overtime as the
system learns.
Experiments and computer simulations have demonstrated that these adaptive systems
also behave as the previous described laser beam. They have the same three states of
dynamic stability, dynamic of disintegration and between them a phase transition. The
phase transition is just at the edge of disintegration, therefore the expression “the edge
of chaos”. In this phase, the system is capable of evolving into new forms, and it uses
self-organisations as the way to achieve it. However, there are researchers like
Johnson and Burton (1994) who questions if these findings will hold true for human
systems which have very complex rules of interactions and conscious, purposive
actions compared to most performed experiments and simulations. There have been
successful simulations of some human systems like the stockmarket by the Santa Fe
Institute (Caulkin, 1995). However, as previously mentioned, this is an area that will
require further research.
According to the worldview of self-organising systems, there are many systems on
many different levels, and they are all interacting with each other. If what is a system
on one level is an agent on another level (eg the brain nests in a mind who belongs to
a group which belongs to a society) as one system learns, it will have an impact on the
other systems. The new behavior will trigger a response from the other systems as
they learn and respond to the new behavior. The result is that the agents co-evolve
with the system and in doing so they effect the system they are a part of. This view of
the world does not take away the freedom of choice, but it does imply that the price
we pay for this freedom is an inability to foresee the long-term outcome of the choices
we make (Stacey, 1997).
Self Organising Organisations
There are very few examples of organisations that are truly built on the principle of
self-organisation. The most frequent talked about example is Visa. Visa is described
as being built on a biological instead of a mechanical metaphor. Visa has grown more
than 10.000% since 1970 (Tetenbaum, 1998) and is a trillion dollar business operating
in more than 200 countries. Despite its size and growth, you don't know where it is
located, who owns it or how it is operated. Dee W. Hock, founder and former
president of Visa International, calls Visa and the Internet “chaords”, which he claims
are seamless blends of the principles of chaos and order, competition and cooperation
(Caulkin, 1995).
However, the great majority of the world’s organisations are build according to
traditional models. How can these organisations start to use the insights generated
through the application of complexity theory? How can they apply the concept of
agents and self-organisation within their rigid organisational structures?
Stacey (1997) has suggested one model that tries to provide insights into those
questions. He conceptualizes the organisation as two parts, the legitimate system and
the shadow system. The legitimate system is the hierarchy, bureaucracy and officially
approved and shared ideology at the time. The shadow system is the network of
personal, social and emotional relationships that underlies and intertwines with the
legitimate system. The shadow system is according to Stacey a self-organising
system. No one can tell us who to network with and you can not control the network
since it requires cooperation from those you network with.
The shadow system can serve as a learning community and the location of the
organisation’s narrative and tacit knowledge. Nobody fully understands their
organisation’s shadow system, nobody is in control of it and everyone contributes to it
through their interaction in the local network. A change in this system will over time
bring about changes in the legitimate system.
If the shadow system is a self-organising adaptive system, how can an organisation
create the right conditions for this system to be “at the edge of chaos” and become a
creative drive for the organisation? I propose that the formal use of dialogue is a
workable approach for an organisation to move in this direction. I will describe the
discipline of dialogue and how it applies to self-organising adaptive systems.
The Need for Dialogue
If we believe the human mind is a complex and self-organising system, with imaginal
symbols as it agents, we need to create an environment where these agents can
interact in a creative way. As described earlier in this article, the optimal conditions
require the right level of energy/information flow and the right amount of
connectidness and diversity, which will bring the system “to the edge of chaos”.
However, this is clearly very stressful for the individuals and will produce a lot of
anxieties that are not contained through the organisation’s normal procedures and
structure. Something else has to be introduced or the system will either disintegrate
into what Bion (1961) calls basic assumption behavior in which the group escapes the
work tasks through acting out a shared fantasy like fight/flight or pairing, or the
system will be blocked by the triggered organisational defenses. These types of
defenses are described in detail by Argyris (1990).
The individuals who “lives on the edge of chaos” have to have reached a certain
psychological maturity to be able to hold the discovered ambiguities of our existence.
According to Klein (1975) we need to be able to operate in the depressive position. In
this position we can love and hate the same object and see good and bad in the same
situation. If we cannot stay in the depressive position, we will tip over into the
schizoid-paranoid position. This position is a form of mental disintegration and will
trigger neurotic defenses that will make any form of genuine inquiry impossible.
For the organisation to start “living on the edge of chaos” we need individuals who
can live with ambiguity and an environment that can contain the triggered anxieties.
Winnicott (1971) describes this environment as the “transitional space” where
“transitional objects” can be played with and manipulated. He believes that
adequately contained spaces enable teacher and learners, analyst and patient to play
together (French 1997). If the teacher, facilitator or analyst can not fulfil the role of
containment, the mutual play of learning can turn into coercion. I believe the same is
true for dialogue
I believe the process of dialogue is one way of “living on the edge of chaos” while
containing the triggered anxieties. The process allows inquiry and the play of
imaginary symbols, as well as providing a safe container for triggered anxieties. The
next part of the paper will describe dialogue in more detail and how it links with
“living on the edge of chaos”.
Dialogue
The purpose of dialogue (Isaacs, 1993) is to establish an environment which facilitate
genuine inquiry. It is a setting in which people can allow a free flow of meaning and
vigorous exploration of the collective background of their thoughts, their personal
predispositions, the nature of their shared attentions and the rigid features of their
individual and collective assumptions. The belief is that something new can emerge if
we can start to see how we create our own rigid structures in our minds and learn to
suspend our judgment, while collectively exploring our thoughts and assumptions
through different individuals eyes.
Isaacs (93) describes dialogue as a discipline of collective thinking and inquiry, a
process for transforming the quality of conversation and, in particular, the thinking
that lies beneath it. A slightly easier and alternative definition he uses is “dialogue can
initially be defined as a sustained collective inquiry into everyday experience and
what we take for granted” (Senge et al 1994). It is important to highlight that the word
“thought” has a very wide meaning. Bohm et al (1991) writes: “We are using the
word “thought” here to signify not only the products of our conscious intellect but
also our feelings, emotions, intentions, and desires.”
These definitions highlight the key aspects of the dialogue process. First of all it is a
collective inquiry process. Bohm (1989), who is one of the major thinkers behind the
process of dialogue, believes a “larger pool of meaning” is only accessible to a group
It is therefore recommended that dialogue has a relatively large number of participants
20-40 (Isaacs, 1992). Schein (1994) claims he has conducted successful dialogue
sessions with 60 people and that he has heard about sessions containing 100
individuals. The process works better if the participants come from different
backgrounds and operate from different roles. Ideally, you want your whole system
you are inquiring into participating in the process. These requirements are perfectly
aligned to the complexity theory, which suggest you need diversity and high level of
connectedness for you to enter the phase of transition or reach “the edge of chaos”.
Secondly, the focus is on everyday experience and the inquiry into a whole system. It
is not a problem-solving tool with a very strict focus. Again we find similarities with
complexity theory, where the focus is on the whole, not the individual parts.
Thirdly, dialogue is based on the assumption that we have mental models in our minds
that we take for granted. Isaacs (1993) states that “human beings operate most often
within shared, living fields of assumptions and constructed embodied meaning, and
that these fields tend to be unstable, fragmented, and incoherent”. Bohm et al (1991)
believe that the pervasive incoherence in the process of human thought is the essential
cause of the endless crises affecting mankind. The poor use of our capacity for
abstraction is blamed for this fragmentation of our mind. We need to use our capacity
for abstraction to deal with complex issues and deal with everyday life. However, we
often forget our thoughts are abstractions and treat them as facts, forgetting how these
opinions and facts were created in the first instance. Senge (1990) calls this tendency
to generalize with a lightning speed the “leaps of abstraction”. If our mental models
would change, we would see things differently. Dialogue is a way to collectively help
each other to see issues from different perspectives and see if something new can
emerge. The mental models fit well with the concept of the human mind as a self-
organising system with imaginary symbols as its agents. Dialogue will help the
imaginary symbols reach the phase of transition!
Dialogue requires a sustained effort over a long time period, sometimes years, to
achieve any transformation. First of all it will take time to build the individuals skills
and awareness levels required for true dialogue. Secondly, you can not put a timeline
on dialogue. You need to spend time to reach the phase of transition and you can not
guarantee that you will get a workable output. Things will take its own course, just
like a truly self-organising system.
Dialogue requires genuine cooperation to be effective (Isaacs, 1993). This is
according to Stacey (1997) also true of an adaptive and self-organising system. If the
mind is as an adaptive self-organising system with imaginary symbols, dialogue
provides the setting for these symbols to combine in new ways and create a new
system (in this case a view of the world). For this to happen, the members will need
inquiry skills and be able to deal with the created anxieties. The aim of dialogue is to
get the participants “to know the thought when they have it”.
Implications
The dialogue discipline, as discussed, has very strong similarities with adaptive self-
organising systems. It can therefore be used as a vehicle for letting the shadow system
experiment with reinventing the systems the organisation belongs to. It ensures a
“safe” environment where we can hold the anxieties in relation to creative play. This
is extremely important since this type of learning can not take place without triggering
different types of anxieties. Isaacs (93) describes a series of “crises” that a group
participating in dialogue experiences. The first is “the initiatory crises” which happens
when the participants realize they cannot force dialogue to take place. This is
followed by “the crises of suspension’. This happens when points of view which used
to make sense, no longer does. The group feels that they do not know where they are
headed. If the group can get through these crises, they start inquiring at a level where
the conversation takes new forms and the energy that has been trapped in rigid and
habitual patterns of thoughts and interactions are freed up. New ways of thinking can
emerge. However, even at this level, there are crises. Isaacs (1993) calls it “the crises
of collective pain”. It happens when the group sees the self-created limits of human
experience.
Anxiety is impossible to avoid if learning is to take place on these fundamental levels.
It is therefore critical to manage anxiety to ensure it does not block learning, or to use
the words of Schein (as quoted in Vince, 1996): “Paradoxically, anxiety prevents
learning, but anxiety is necessary to start learning as well. Managing learning or a
change process means managing these two kinds of anxieties”. Long and Newton
(1997) further highlights the difficulties related to inquire into and challenge long held
assumptions and emotions: “The starting point for significant change is the system
internal to the thinker and learner, but first it must be brought into awareness. The
history of psychoanalysis is about the difficulty of achieving a state of being that will
allow a denied system of emotional experience to impress itself on the conscious
mind”. These difficulties have led to dialogue being viewed as a powerful but slow
process. According to Isaacs, the dialogue process often starts with spending time
trying to understand what dialogue really means. It is a "safer" topic to use when
learning the process as well as learning to suspend judgement and holding
ambiguities, conflicting thoughts and emotions. Participating in a well functioning
dialogue process will over time lead to personal growth, but it also means
experiencing a lot of pain when the different "crises" as described by Isaacs (1993)
are worked through.
Another benefit of dialogue is that it allows the self-organising system to evolve at its
own pace partly protected from the harsh reality of competition. The ideas born, as
part of the dialogue process will have had time to evolve, mature and gain support,
before it has to be subjected to the normal competition within the organisation.
However, there will still be enormous difficulties for any insights achieved in the
dialogue process to be shared with an adopted by the formal process.
Dialogue, the way it has been discussed in this article, takes an organisation beyond
“double loop learning” to generate “triple loop learning” (Isaacs, 1993). Double loop
learning was coined by Argyris and focuses on inquiry into the context of an issue.
For instance, single loop learning refers to trying to get the thermostat to a certain
temperature, while double loop learning would question why we chose a certain
temperature. Triple loop learning would question why we even worry about the
temperature. It is looking for whys and questioning of the context, not just improving
the effectiveness of the existing system.
If the dialogue process inquires into the holistic systems we belong to, we could find
that we want to substantially change the way we operate. It is not just a problem
solving tool (Cayer, 1997). Dialogue has the potential to recreate systems, not just
improve them within a defined context. It is a “relative safe” way to start to apply
complexity theory to a traditional organisation, but it will trigger deep seated anxieties
that have to be contained to stop the process from exploding or being blocked. The
organisation needs to understand this and the need for long term time frames before
they head down this path of encouraging the shadow system to reinvent itself.
Research Implications
MIT’s program for dialogue research has so far focused on understanding the process
of dialogue. However, the proposed thesis of using dialogue as an active way for the
shadow system to reinvent itself based on the philosophies of complexity theory
opens up new directions for research. There needs to be a focus on how the process of
dialogue impacts the shadow system over time and how any generated insights are
shared with an adopted by the legitamate system. Some of the key questions to study
are:
e How are the insights gained in dialogue transmitted and taken up in the shadow
system?
e How are changes in the shadow system impacting the legitimate systems of the
organisation?
e What will facilitate the transmition and adoption of the insights gained in dialogue
to the legitimate system?
Research, based on the assumptions used in complexity theory, raises two added
complications: What is our unit of analysis and the need for longitudinal studies.
According to Begun (1994), the use of complexity theory forces organisation
scientists to dwell on the question; "What is our unit of analysis?" As a result we have
to explore:
e¢ In what way is our unit of analysis similar or disimilar to the unit of analysis in
other sciences?
What is the nature of interdependence between the different systems?
How do we classify our unit of analysis?
The other key issue is the need for longitudinal studies. According to Begun (1994),
cross-sectional studies do not make sense when you are studying dynamic systems
heavily based on feedback loops. The focus of the study has to be on how the system
changes over time, not detailed understanding of the relationship between different
variables at one point in time. Therefor research in the areas of dialogue and shadow
system require longitudinal studies. Longitudinal studies would ensure the real
assimilation process would be studied instead of a sanitized construction after the fact.
References
Argyris, C. (1977). Double Loop Learning in Organizations, Harvard Business
Review, September-October.
Argyris,C. (1990). Overcoming Organizational Defenses: Facilitating Organizational
Learning, Allyn and Bacon, Prentice-Hall, Boston.
Begun, J.W. (1994). Chaos and Complexity: Frontiers of Organiszation Science,
Journal of Management Inquiry, December.
Bion, W.R. (1961). Experiences in Groups and Other Papers, Tavistock Publications,
London.
Bohm, D. (1989). On Dialogue, Ojai, California.
Caulkin S. (1995). Chaos Inc. Across the Board, July/August.
Cayer, M. (1997). Bohm’s dialogue and Action Science: Two Different Approaches,
Journal of Humanisitc Psychology, Spring.
Devaney, R.L. (1992). A first course in chaotic dynamical systems: Theory and
experiment. Reading, Addison-Weasly, MA.
French, R,B. (1997). The Teacher as Container of Anxiety: Psychoanalysis and the
Role of the Teacher, Journal of Management Education, November.
Isaacs, W.N. (1992). Dialogue: The Power of Collective Thinking, Proceedings from
System Thinking in Action Conference: Creating a Learning Organiszation, Pegasus
Communication.
Isaacs, W.N. (1993). Taking Flight: Dialogue, Collective Thinking, and
Organizational Learning, Organizational Dynamics, Autumn93, Vol.22 Issue 2.
Johnson, J.L. & Burton, B.K., (1994). Chaos and Complexity Theory for Management
Journal of Management Inquiry, December.
Kellert, S.H. (1993). In the wake of chaos, University of Chicago Press, page 2,
Chicago.
Long, S. and Newton, J. (1997). Educating the Gut: Socio-Emotional Aspects of the
Learning Organization, The Journal of Management Development, Vol 16. No 4.
Lorenz, E. (1963). Deterministic and Periodic Flows, Journal of Atmospheric Science,
Vol 20.
Prigogine, I. & Stengers, I. (1984). Order out of chaos: Man’s new dialogue with
nature, Bantam Books, New York
Schein E.H. (1994). On Dialogue, Culture, and Organizational Learning,
Organizationals Dynamics.
Senge, P.M. (1990). The Fifth Discipline: The Art and Practice of the Learning
Organization, Century Business, London.
Senge, P.M., Roberts, C., Ross, R.B.,Smith, B.J., Kleiner, A., (1994). The Fifth
Discpline: The Art and Practice of the Learning Organization, Nicholas Brealey
Publishing Limited, London..
Stacey, R. (1992). Managing the Unknowable: Strategic Boundaries Between Order
and Chaos in Organizations, Jossey-Bass, San Francisco, CA.
Stacey, R. (1996). Management and the Science of Complexity, Research Technology
Management May-June.
Stacey, R. (1997). The Implications of Complexity Theory for Psychoanalytical
Thinking about Organizations, Symposium for the International Society for the
Psychoanalytical Study of Organization.
Stacey, R. (1997). Excitement and Tension at the edge of Chaos in Smith E. (ed),
Integrity and Change - Mental Health in the Market Place, Routledge, London and
NewYork.
Taylor, F.W. (1911). The Principles of Scientific Management, Harper and Row, New
York.
Tetenbaum, T.J. (1998). Shifting Paradigms: From Newton to Chaos, Organizational
Dynamics, Spring.
Vince, R. (1996). Managing Change: Reflections on Equality and Management
Learning, Policy Press Bristol, UK.
Waldrop, M.M. 1992. Complexity: The emerging science at the edge of chaos, Simon
& Schuster, New York.
Warren, K., Franklin, C. and Streeter, C. (1997). New Direction in Systems Theory:
Chaos and Complexity, Social Work, July.
Wheatley, M. (1992). Leadership and the New Science: Learning about
Organizations from an orderly Universe, Berrett-Koehler, San Fracisco, CA.
CC BY-NC-SA 4.0