The dynamics of self-renewal: A systems-thinking to understanding organizational challenges in dynamic environments

Abstract

In this paper I examine the conditions for the process of self-renewal via two different systems theories. I begin with a general overview of the development of systems thinking, providing a backdrop to the discussion of the two theories in focus. I then proceed to a more in-depth treatment of Ilya Prigogine’s theory of self-organizing systems. This opens the door to understand radical reform and renewal, particularly the innovative development process and the function of collective intelligence. Next, I move on to the autopoiesis theory of Humberto Maturana and Francisco Varela and to its social science application by Niklas Luhmann. This theory is particularly useful for understanding organizations as learning and evolving systems.

Full text

1
Chapter 7
The dynamics of self-renewal:
A systems-thinking to understanding organizational challenges
in dynamic environments
Pirjo Ståhle
1
1
The author is Professor at Finland Futures Research Centre, Turku School of Economics, pirjo.stahle@tse.fi

2
The dynamics of self-renewal: A systems-thinking to understanding
organizational challenges in dynamic environments
1. Introduction
2. The paradigms of systems thinking
3. Self-organizing systems according to Prigogine
3.1 Adaptations of self-organization in further research
4. Autopoietic systems by Maturana and Varela
4.1 Applications of autopoiesis theory
5. Luhmann‟s self-referential systems
5.1 The system‟s capacity for self-reference
6. The dynamics of self-renewal in organizations
1. Introduction
Organizations today operate in a dynamic, highly unpredictable global competitive environment.
The challenge is the same for both businesses and public organizations: how to constantly increase
speed and efficiency, to improve quality and innovation? In order to succeed in the competition at
both the company and national level, systems must show a capacity for continuous development
and even radical change. Increasingly, competitiveness now boils down to a capacity of self-
renewal in and by organizations, networks and nations. Continuous innovation and renewal
capability in organizations has indeed attracted growing research interest in recent years (e.g.
Nonaka & Takeuchi, 1995; Leonard-Barton, 1995; Weick and Sutcliffe, 2002; Brown & Eisenhardt,
1998; Ståhle et al., 2003; Pöyhönen, 2004).
To achieve the capacity for self-renewal, it is necessary to amalgamate and integrate different kinds
of expertise, interests, people and organizations. The management of these complexities presents a
huge challenge for every organization, and cannot be adequately met without an internal capacity
for self-organization. It is necessary therefore to understand the dynamics of self-renewal, which
unfolds as a result of a process of change involving multiple agents and driven from within the
system.
In this paper I examine the conditions for the process of self-renewal via two different systems
theories. I begin with a general overview of the development of systems thinking, providing a
backdrop to the discussion of the two theories in focus. I then proceed to a more in-depth treatment
of Ilya Prigogine‟s theory of self-organizing systems. This opens the door to understand radical
reform and renewal, particularly the innovative development process and the function of collective
intelligence. Next, I move on to the autopoiesis theory of Humberto Maturana and Francisco Varela
and to its social science application by Niklas Luhmann. This theory is particularly useful for
understanding organizations as learning and evolving systems.

3
2. The paradigms of systems thinking
Systems theories were developed in the twentieth century on both sides of the Atlantic, although
they have received greater emphasis in Europe than in the US (Checkland, 1988, 13). In the late
1940s there were two main schools of systems thinking: general systems theory and cybernetics.
These two approaches have provided the foundation for the development of systemic thinking and
systems theory up to the present day. Cybernetics was originally very much dominated by the
Newtonian paradigm, which means that systems where viewed mainly as ingenious machines.
Systems where dominated by general laws and as such they where predictable and controllable
(Dooley 1995, 1999). In perspective this view is still an important part of the cybernetic systems
thinking.
The Austrian biologist Ludwig von Bertalanffy founded general systems theory, and was the first
scientist to develop systems research outside the field of physics. In the 1920s and 1930s, von
Bertalanffy‟s theory focused on open systems and was initially grounded in organic biology, but it
was subsequently elaborated into a general systems theory (e.g. Bertalanffy, 1967, 1972a, 1975). In
this theory systems are looked upon as open and living organisms that communicate with their
environment. The processes taking place within the open system serve as continuous feedback
cycles, which are described as chains of inputs, throughputs and outputs. The system never rests and
the only force that maintains it is this perpetual motion. Feedback cycles generate a lot of
information that allow the system to choose different paths of development. In spite of its perpetual
motion, the system strives to achieve equilibrium and always remain in a steady state.
The other school of systems thinking was cybernetics, which was pioneered by American
mathematician Norbert Wiener. Cybernetics, according to Wiener, referred to disciplines that were
concerned with controlling machines and organisms by means of communication and feedback, i.e.
with the dissemination, manipulation and use of information for purposes of controlling biological,
physical and chemical systems (Wiener, 1948 and 1950; Porter, 1969, vii). Cybernetics is focused
on machine-like systems, whose operation and outcome are predetermined, or at least predictable. A
cybernetic system is a closed system in the sense that it has no exchange of energy or matter with its
environment. An open system, on the other hand, has several options with respect to its aims and
operation, and it is furthermore dependent on interaction with its environment.
From the 1960s onwards, systems thinking began to change. It was still mainly founded on the
theory of open systems, but the main focus of attention began to shift to the complexity of systems
and their innate capacity for change. This led to the emergence of new concepts and patterns of
systems thinking, including Forrester‟s system dynamics, Checkland‟s soft systems methodology
and Senge‟s learning organization. In 1956, Jay Forrester started the System Dynamics Group at
MIT, leaning largely on cybernetic thinking. However the group‟s main interest was in open
systems that communicated with their environment (e.g. Forrester, 1961 and 1968). Although they
focused on systemic change and problem-solving, Forrester (1991, 1) maintains that the approach
has universal application because system dynamics provides the foundation both for understanding
any processes of change and the tools to steer and influence them. Peter Checkland introduced his
soft system methodology as a critique against what he regarded as oversimplification of reality
(Checkland, 1981 and 1991, 1). His aim was to understand large social systems through feedback
cycles. Checkland emphasized that people create their own reality and are always active and
organic parts of the system. This is why systems formed by humans cannot be studied or
manipulated from the outside. Checkland was chiefly interested in identifying systemic changes

4
rather than regulating or manipulating them. From the early 1990s onward Peter Senge‟s concept of
learning organization gained wide currency. Organizational learning had previously been
successfully addressed by Argyris and Schön (1978), so in this sense the notion of systemic
learning of organizations was not entirely new. However, Senge (1990) was more clearly connected
to the tradition of systems thinking, especially to the idea of continuous renewal. He was interested
not only in the changes required by the environment or adaptive learning, but in learning processes
and organizational change that pave the way to generative learning (ibid., 14).
These three branches of systems thinking (systems dynamics, soft systems and learning
organization) highlighted a new research interest: the attempt to understand change and its
manifestations from a systemic point of view. Forrester, Checkland and Senge represented a new
way of thinking, but initially they were still quite firmly anchored either to the discourse of open
systems or cybernetics. However at the same time (in fact starting from the 1960s) a whole new
systems theory discourse began to evolve, and to gain ever-increasing recognition.
The evolving new systems paradigm was not based on open systems theory or cybernetics, but it
marked a complete departure from old ways of systems outlining and thinking. The new paradigm
focused on the chaotic and unpredictable behavior of systems (rather than on their stability) and on
the internal dynamics of systems (rather than on feedback cycles). The new perspective grew out of
three main sources:
1) Complexity and chaos research, as represented by Lorenz (1993 and 2005), Feigenbaum (1982
and 1993), Mandelbrot (1977a and 2004), the Santa Fe group (since 1984);
2) Prigogine‟s self-organizing systems (1980 and 1984);
3) Maturana and Varela‟s autopoietic systems (1981 and 1987).
Chaos and complexity research represent distinct traditions of their own, yet from a systems theory
point of view they also cover a lot of common ground, i.e. intra-systemic dynamics and changes
originating from within. Chaos theories emphasize the perspective of unpredictability and
permanent, uncontrollable laws, whereas complexity research places more weight on emergent
intra-systemic characteristics.
The most prominent instigator of this new line of thinking was American meteorologist Edward
Lorenz, who brought along a whole new perspective on dynamic and chaotic systems in the area of
meteorology. Whereas previously it was thought that chaos and discontinuity were instances of
system malfunction, Lorenz (1963) argued that they were in fact the normal state for many systems:
some systems, such as climate conditions, were in a constant state of chaos, however in an orderly
fashion. Chaotic systems are particularly sensitive to change because they are often composed of an
infinite number of interactions and are therefore in perpetual motion. Even the slightest change in
the original state of the system may have dramatic effects throughout the whole system. Another
noteworthy chaos researcher is Benoit B. Mandelbrot, whose studies on fractals formed by chaotic
systems have attracted much attention. Fractal theory means that the same structures and patterns
can be found within the system at different levels, i.e. that the system repeats itself at both the micro
and macro level (e.g. Mandelbrot, 1977a). A major influence in this field is the Santa Fe Institute:
founded in 1984, it is perhaps the world‟s leading research center on complex systems.
The chaos and complexity perspective implied three fundamental changes to the earlier systems
views of open and closed systems. These changes concerned the conception of a system,
possibilities to influencing the system and the focus of research interest.

5
1) The conception of the dynamics of the system. The focus shifted from equilibrium, stability
and continuity to imbalance, change and discontinuity. In contrast to earlier beliefs, the
continued existence of the system was not dependent on the maintenance of equilibrium.
Chaos was not a disruption or aberration in the system, but on the contrary often a
prerequisite for existence and development.
2) Conceptions of how the system could be steered and influenced. The interest was no longer
on manipulation or control of the system. Instead the system could be understood and it
could be steered and influenced from within, through involvement and participation in the
system, i.e. interaction. In order to glean information about the system, people had to be
actively involved in the system. Objective, external observation was merely a delusion.
3) The focus of research interests. Whereas previously researchers were interested in searching
for general laws, principles, symmetry and harmony, their interest now turned to
understanding the nature of change, the unfolding of changes and processes of radical
renewal.
We can distinguish between three different paradigms in the development of systems thinking. The
first paradigm refers to systems that are controlled by universal laws, regularities and stability.
Research under this paradigm aims to explain and define laws and principles and to predict events
on a theoretical basis. According to the underlying theories, systems are machine-like and obey
predetermined laws. Their foundation is provided by classical Newtonian physics, which is the
paradigmatic basis of Western science.
The second paradigm is based on general systems theory as developed by von Bertalanffy.
According to this theory systems are not regarded as closed or mechanical machineries, but on the
contrary as constantly evolving, open organisms that communicate and change with their
environments and their changes. The paradigm emphasizes both the system‟s interaction with its
environment and its alternative, open paths of development. Open systems are in a constant state of
controlled change, yet all the time striving for a new equilibrium, and permanent disequilibrium
would lead to system breakdown. The intra-system process is supported and maintained by input-
throughput-output feedback cycles, which are regulated by the system from within.
The third paradigm focuses on the system‟s own internal, autonomous dynamics. Here, the system
is looked upon as a highly complex entity that is in a state of inherent disequilibrium and chaos. The
paradigm emphasizes a) the capacity of the system for self-organization and renewal; b) the
system‟s discontinuity and non-determinism; and c) the non-locality of the system. The main
interests of the third paradigm lie in the system’s self-renewal, self-organization and its capacity for
radical change.
Paradigm
Origin
Characteristic
Research
interest
Objective
I
Closed
systems
Newton
static,
deterministic,
mechanical
principles,
rules,
laws
prediction,
control
II
Open
systems
von
Bertalanffy
balanced, near
equilibrium
equifinal,
living
feedback
processes,
changes,
adaptation
control,
maintenance,
development

6
III
Dynamic
systems
Lorenz
Prigogine
Maturana
Varela
imbalance, far-
from-
equilibrium,
uncontrollable,
complexity,
chaos
self-
organization,
self-renewal,
intra-systemic
dynamics
understanding/exploiting
system dynamics,
radical change,
innovation
Table 1. The paradigms of systems thinking (adapted from Ståhle, 1998, 43)
The three paradigms can be seen as complementary perspectives on systems thinking. None of them
is right or wrong as such, but can instead be seen as a (partly chronological) continuum of
understanding systems. The paradigms also refer to the existence of different kinds of systems with
different characteristics. Each paradigm still offers a valid point of departure depending on the
situation and the type of the system under scrutiny. However, it is crucial to understand the
paradigms and the different – even contradictory – prerequisites behind the respective systems.
There is no such scientific point of departure as “systems theory”, since every analysis always
involves certain tradition or perspective on systems, i.e. a system approach can refer to various
theories on systems. To refer generally to “systems thinking” or “systems theory” (as is more often
the case in the research literature) is meaningless unless one‟s point of departure is explicitly
anchored to a certain systems paradigm or at least a systems tradition.
From a practical point of view understanding of the system paradigms sheds useful light on how
organizations have been managed and continue to be managed today. These paradigms describe
comprehensive beliefs and mental models that are employed in the design and implementation of
change processes as well as in the management and leadership of organizations. They also help us
to understand the sometimes hard-to-resolve conflicts that arise between decision-makers and the
people responsible for implementation. If a creative development project is grounded in the first
(mechanical) system paradigm; it is easy to predict how the approach and results will differ
compared to the situation where the third (dynamic) paradigm of self-direction is adopted. An
awareness of these differences may in itself allow for the effective treatment of emerging undesired
conflicts or help in choosing an approach that is best suited to the situation.
2
In this paper I focus on the third paradigm which is the most informative with respect to the topic in
hand: understanding organizational challenges to self-renewal in dynamic environments. The
chapters below concentrate on the two major theories in this area: Prigogine‟s theory of self-
organizing systems and Maturana and Varela‟s theory of autopoiesis.
3. Self-organizing systems according to Prigogine
The results of chaos research only began to receive wider attention in the 1980s, even though many
key studies were initially published much earlier. These studies had shown that some systems are
capable of self-organization and self-development under the force of their own inherent (chaotic)
dynamics. Ilya Prigogine published his study on dissipative or self-organizing systems in 1967: this
provided the foundation for his analyses of the process of becoming as well as the evolution of
order out of chaos (Prigogine, 1967a, 1967b, 1967c). It was a revolutionary argument to suggest
2
For more details on organizational applications of the mechanical, organic and dynamic paradigm see Ståhle &
Grönroos, 2000.

7
that systems were capable of self-organization, without any external control (e.g. Nicolis &
Prigogine, 1977, Prigogine & Stengers, 1984); this marked a radical departure indeed from general
systems theory. Prigogine showed that self-organization was not in fact an exception, but on the
contrary quite a common systemic characteristic. Examples of self-organization include the
operation of markets, human biology or the movement of flocks of birds. An economic system is
created out of the countless decisions that are constantly made by people, consciously and
unconsciously, to purchase and to sell. The system is neither designed nor controlled by anyone; the
market simply creates and re-creates itself. In the same way, genes organize themselves in a certain
way as they form a liver cell and in another way to form a muscle cell, and flocks of birds are
organized without any external control. A modern example of self-organization is provided by the
Internet (see also the research on neuron networks by Kohonen et al., 2004).
Prigogine describes the phenomenon of self-organization from various perspectives in various
contexts. He points out that the phenomenon of self-organization is quite normal for different
systems, yet not all systems are capable of self-organizations. However, Prigogine does not offer a
clear universal description of the preconditions for self-organization, but deals with the issue in
several of his works. Based on the analyses of Prigogine‟s descriptions we can identify five core
concepts in self organization: 1) far from equilibrium, 2) entropy, 3) iteration, 4) bifurcation, and 5)
time. These core concepts have been drawn primarily from four works in which Prigogine describes
self-organization from different perspectives: Order out of Chaos (Prigogine & Stengers, 1984),
From Being to Becoming (Prigogine, 1980), Thermodynamic Theory of Structure, Stability and
Fluctuations (Glandsdorff & Prigogine, 1971) and Exploring Complexity (Prigogine & Nicolis,
1989). The concepts have their origins in chemical and physical phenomena, but Prigogine
frequently points out that they are also applicable more generally to social and human systems (e.g.
Prigogine, 1976, 120-126 and Prigogine & Nicolis, 1989, 238-242).
Far from equilibrium
According to Prigogine, most systems appearing in the world are capable of self-organization, but
only on certain conditions. Self-organization can only occur in systems that are capable of
remaining far from equilibrium, i.e. at the edge of chaos. Prigogine says that in all forms of life,
chaos or disequilibrium is the source of new order. In the state of disequilibrium, external change
and pressure constantly act on the structures and boundaries of the system: the system is being
pushed, as it were, towards disorder and chaos and is therefore under constant threat of collapse.
Instead of collapsing, however, the system is driven into a state of dynamic equilibrium, i.e. the
system possesses a dissipative structure: continuously disintegrating, destroying old structures the
system subsequently re-organizes new structures again. The self-organizing system is in a constant
state of chaos and order, i.e. it alternates between consecutive overlapping cycles of chaos and
order and order and chaos: after organizing itself and being driven into chaos, it re-organizes and
subsequently comes under threat and is driven into disorder, etc. It is noteworthy, however, that not
all systems are capable of self-organization: when a stable or balanced mechanical system comes
under pressure, it simply disintegrates and is unable to re-organize. (Prigogine & Stengers, 1984,
178, 278, 292; Prigogine, 1980,100, 123.)
It is impossible to understand the process of self-organization without an understanding of
disequilibrium, or what Prigogine refers to as “far from equilibrium”. Disequilibrium refers to a
state of intra-systemic conflict at the edge of chaos: for instance, in a thermodynamic system to the
simultaneous presence of hot and cold, or in a social system to the co-existence of conflicting
interests. These extremes create an inherent tension in the system and active interaction within the
system. Disequilibrium also results from the system being exposed to pressures from the outside, or

8
to stabilization being prevented by the system‟s internal entropy. (Glandsdorff & Prigogine, 1971,
278.)
Entropy
Entropy, Prigogine says, has a function of paramount importance in the process of self-organization.
Entropy refers to energy or information that the system produces but that it cannot use. In this sense
it may be described as a surplus residue. A high degree of entropy is also indicative of disorder,
wasted resources, untapped information, or insecurity within the system. Entropy is created when
the system exchanges or produces information and energy beyond its needs, or when information is
disorganized, unclassified or devalued. Established thinking and the second law of thermodynamics
had it that entropy was superfluous and a threat to the system, and only increased the system‟s
destructive instability. Prigogine, however, showed that in systems capable of self-organization,
entropy was in fact necessary and indispensable. Entropy introduces uncertainty, imbalance and
confusion into the system – and it is this very instability that gives the system its capacity for self-
organization. In other words in the process of self-organizing excess entropy is both used and
absorbed. (Glandsdorff & Prigogine, 1971.)
Iteration
The foundation of all self-organizing systems lies in abundant information exchange, abundant
interaction. Intra-systemic interaction, when it is as its most sensitive and most abundant, refers to
the second precondition for self-organization, i.e. iteration. Iteration means a continuous, highly
sensitive feedback process or activity via which the information and models produced by the system
are rapidly disseminated throughout the system. Iteration gives the system its capacity of self-
renewal, its ability to copy internal models from the micro to the macro level and vice versa. In a
sense, it is the system‟s engine room. For iteration to work properly in the system, intra-systemic
interaction must meet two criteria: first of all it must be non-linear, and secondly it must be based
on feedback. The basis for feedback refers to the basic condition of iterative dynamics, i.e. sensitive
dependence on the original circumstances. (Prigogine & Nicolis, 1989, 219; Prigogine, 1976, 95.)
Iteration as positive and negative feedback (and feed forward) functions makes the system
spontaneous and utterly sensitive to change, i.e. the system dynamics is nonlinear. This often finds
expression in what is known as the butterfly effect: initially the effect is seen in only a small part of
the system, but it then advances and gradually gathers momentum so that “the flap of a butterfly‟s
wing in Brazil sets off a tornado in Texas” (Lorenz, 1993, 14). This would not be possible without a
sensitive and continuous reciprocal feedback process between different components of the system.
Iteration is the driving force of self-organization, because it constantly generates new information
and new structures and carries the effect throughout the system. Iteration guarantees that whatever
happens within the system, it spreads and multiplies. (Prigogine & Stengers, 1984, 154; Prigogine
& Nicolis, 1989, 72.)
Bifurcation
Bifurcation is a zone in-between determinism and free choice. It means that a) there are certain
periods in the life of a system when it can make genuine choices, b) these choices cannot be
predicted and c) the choices are irreversible. The system has a choice between two or more
alternatives when it is driven ever further from its state of equilibrium. Bifurcation, therefore, is
always” a manifestation of a new solution” (Prigogine, 1980, 105), and it produces a solution that is

9
not a logical or necessary extension of the previous structure. “The event of bifurcation, therefore, is
also always a source of innovation.” (Prigogine & Nicolis, 1989, 74).
The change of a system to a new state of equilibrium happens suddenly, as if in a single (quantum)
leap. At the point of bifurcation, the system rejects huge amounts of information, reducing the
amount of entropy and paving the way to the creation of a new order, a new dynamic structure.
Between the old and the new system structure there is a moment of discontinuity and non-location,
i.e. neither old nor new system structure exist. The point of bifurcation is a key concept with respect
to irreversible changes in self-organizing systems. When the system drifts ever further from its
original state of equilibrium, it can only choose between existing new possibilities; there is no
returning to the old. Bifurcation does not necessarily require chaos; a state of equilibrium will
suffice, together with a genuinely open and non-deterministic situation. The choice made by the
system cannot be predicted, i.e. the choice is never made by necessity and in this sense it is a
genuine free choice. (Prigogine & Stengers, 1984, 169; Prigogine 1980, 106; Prigogine & Nicolis,
1989, 72.)
The historical path that the system has followed in its development includes a series of stable stages
dominated by deterministic laws, and a series of unstable stage or points of bifurcation where the
system can make a free choice between several alternatives. This mixture of necessities and
possibilities constitutes the history of the system. (Prigogine & Stengers, 1984, 169.)
Time
From the system‟s point of view, time is both subjective and objective. Subjective time means that
the system creates its own history via its choices. Bifurcations not only create a new order, but also
at the same time equip the system with new unique characteristics and structures. The constant
production of entropy forces the system constantly to move forward, to continuously develop and
find new forms. This kind of evolution requires time, and is built into the system: it is the system‟s
way of being. Over time, all parts of the system and their subsystems contribute to driving forward
the process of evolution. (Prigogine & Stengers, 1984, 106; Prigogine, 1980, 127.)
According to Prigogine everything has its own forward-looking dynamics; all development is
geared forward. In self-organization the main role is played by entropy, because entropy is also key
from the point of view of time and evolution. In nature and in human life, entropy constantly
produces development and forward movement, which has both an innovative and on the other hand
a deterministic side. We live our lives on the interfaces of both necessity and creativity, of being
and becoming. (Prigogine & Lefever, 1973, 132.)
In this sense the perspective of timing is crucial to the process of creation, to the changeover from
one point of bifurcation to another. Bifurcations appear unexpectedly, and the possibility of choice
and change is only opened up with the occurrence of the point of bifurcation. All systems have their
own history, an irreversible series of events that go together to form a path of a unique life. It can be
argued that each process, with time, produces its own unique pattern out of the alternation between
chaos and new order. For self-organizing systems, this means that it is essential to be able to master
and cooperate with time. The accumulation of entropy takes time, the exchange of information takes
time, the iterative feedback processes takes time – and points of bifurcation have their own timing
that needs to be recognized. (Prigogine & Nicolis, 1989, 242; Prigogine, 1976, 124.)

10
Main concepts of self-
organization
Far from
equilibrium
Entropy
Iteration
Bifurcation
Characteristic
continuous or
periodic
excess residue,
not directly
applicable
non-linear,
based on
feedback
between necessity
and freedom
Manifestation
fluctuations,
intra-system
conflicts,
pressures
from outside
system
abundant
information
exchange,
tolerance of
insecurity and
confusion
sensitive
reaction and
feedback,
positive and
negative
feedback
proper timing
Significance to self-renewal
precondition
for radical
change
creation of new
order
cumulative
power
innovation and
new solutions
Table 1. Self-organizing systems according to Prigogine (based on Ståhle, 1998, 72)
3.1 Adaptations of self-organization in further research
Prigogine‟s work on self-organizing process might have universal applicability, even though his
findings were made in the context of chemistry and physics. He himself is convinced on the
universal nature of the principles of self-organization (Prigogine & Stengers, 1984, 298). There are
no detailed scientific analyses of the self-organizing process that would provide conclusive
evidence in either direction. However, self-organization as such has raised increasing interest in
various branches of research, and the angles and results of this work are closely connected to
Prigogine‟s findings.
Prigogine‟s key concepts “order out of chaos”, “self-organization through bifurcation” and
“dissipative structures” have been applied in a variety of fields, ranging from quantum physics (e.g.
Wheatley, 1999) to the mental process of knowledge creation and neurology (e.g. Piaget, 1975;
Collier, 2005). Applications in the study of social systems in general (e.g. Mileton-Kelley, 2003),
economics (e.g. Arthur 1994 and Chen, 2000), organizational systems (e.g. Marion, 1999; Griffin,
2002) and developmental and innovation processes (e.g. Fischer, 2001; Kuschu, 2001; Nonaka,
2006) are particularly interesting.
Self-organization has also attracted increasing research interest in social and organizational
sciences, where it has been studied in the context of collective intelligence among others by
Hakkarainen (2006) and Engeström and colleagues (Engström et al., 1999). Collective intelligence
refers to processes of intelligent activity that find expression at the collective rather than individual
level. Many animals are capable of coordinating and self-organizing mutual activities at such a high
level of sophistication that they can be considered to possess a kind of swarm intelligence. The
swarm intelligence of ants, for instance, is a form of self-organizing activity. Humans engage in
various processes of collective intelligence that resemble swarm intelligence, both metaphorically
and literally. Many of the manifestations of human collective intelligence are outcomes of self-
organizing activity rather than representing coordinated, organized or directed individual processes.

11
According to Hakkarainen the highly complex problems that people have to resolve in knowledge
work or in high technology require ever greater reliance on socially distributed intelligence and
competence. Collective intelligence is based upon the self-organization of the social collectivity‟s
intelligent systems into a collective intelligent system. The self-organization of intelligent activity
within the social collectivity is crucial to overcoming and exceeding the individual‟s intellectual
resources. (Hakkarainen, 2006.)
Engeström et al (1999) have argued for the emergence of an historically new type of work, which
they call knotworking. Knotworking is characterized by the absence of an organizing structure or
centre; instead the participants collaborate to self-organize their work, its objectives and modi
operandi. The approach developed by Engeström and colleagues (1999, 1987) on the basis of the
cultural-historical theory of activity offers a conceptually advanced way of understanding collective
intelligence. That approach is now emerging as an international metatheory of collective
intelligence that provides a unified foundation for the analysis of human collective activity (Minnis
& John-Steiner, 2001). Engeström‟s theoretical frame of reference does not draw directly upon
Prigogine‟s work, but his research certainly stands as an excellent concretization of the
phenomenon of self-organization as it is discussed in this article.
Social collectivities spontaneously produce an accurate understanding of the distribution of
knowledge and know-how within an organization, which refers to transactive memory (Wegner,
1986; Moreland 1999). This is true particularly in situations where people are working with highly
complex information and knowledge for extended periods of time. A team that works closely
together for long periods, such as an elite anti-terrorist police group, a football team or an
emergency room team may develop a collective mind (Weick & Roberts, 1993). Intensive
interaction makes it possible to transcend the boundaries of the individual‟s skills and competencies
and to form a socio-cultural system with hybrid expertise that cuts across those boundaries
(Howells, 1997; Spinardi, 1998). It has also been suggested that the current era of information
networks is changing our conceptions of how human intelligence works, indeed that it calls for a
new understanding of humans as networking cultural creatures whose intelligence in socially and
physically divided (Salomon, 1993).
Theories of collective intelligence owe their origin to early pragmatists such as John Dewey and
George Herbert Mead, but they have now been sidelined from mainstream psychology. Psychology,
however, was dominated by a natural science ideal that did not provide solid enough premises for
an investigation of social intelligence, and researchers in this field were unable to offer
methodologically reliable tools. The situation has now changed, for three reasons. First of all,
modern audio, video and network technology means that complex collective phenomena can be
recorded. Secondly, the analysis of social networks provides the methodological tools that are
needed to analyse relations between individual agents. And thirdly, the theory of self-organizing
dynamic systems helps us to conceptualize complex phenomena of interaction. (Hakkarainen, 2006,
398.)
4. Autopoietic systems by Maturana and Varela
Prigogine described a process of renewal that does not necessarily lead to incremental development
through small steps, but on the contrary that is geared to producing whole new solutions and
structures. These may also be described as innovations, since the new solution always introduces
genuinely new information to the system (Prigogine & Stengers, 1984, 307; Prigogine & Nicolis,

12
1989,132, 140). Chilean biologists Humberto Maturana and Francisco Varela, on the other hand,
approached the process of renewal from a different perspective. They focused in their research on
living systems as self-copying, self-reproducing organizations, thus addressing the principles of a
system‟s self-renewal from a completely new angle as compared to the perspectives adopted in
earlier complexity studies. In the discussion below I first consider autopoietic systems by reference
to Maturana and Varela, and then proceed to a social scientific application of their theory by
German sociologist Niklas Luhmann.
Maturana and Varela published their study on autopoietic systems in the early 1970s. The concept
of autopoiesis was originally coined in the field of biology to describe the capacity of cells for self-
reproduction. The theory belongs to the category of new emerging paradigms dealing with
spontaneous phenomena and the self-organization of physical, biological and social systems
(Zelenyn, 1981a, xv).
Autopoiesis means self-production, self-maintenance, sameness and harmony (autos = self, poiein =
to do, to produce, to maintain existence, to do again, to conceptualize). In autopoietic systems
relations and interaction constitute both the system itself and its boundaries, not just the system
components, i.e. relations and interactions are the main components of a system. The constituent
parts influence the whole and the whole influences the constituent parts, i.e. the relations within the
system are organized in such a way that they are constantly reproduced. Autopoiesis refers to “the
process of self-production and self-renewal in living systems” (Dobuzinskis, 1987, 214). The
coherence of the autopoietic system is always the outcome of the close contacts and interaction
between constituent factors (Maturana, 1981, 23).
According to Maturana (1981, 23) the autopoietic system is defined as follows: “The unity of an
autopoietic system is the result of the neighborhood relations and interactions (interplay and
properties) of its components.” Thus autopoietic systems are thus entities a) where the components
create the network and the network creates the components – i.e. interaction between the constituent
parts of the system maintains and constantly reproduces the network, but on the other hand the
network also produces and maintains the constituent parts; and b) whose boundaries are formed by
the parts of the network that are involved in building the network (ibid., 21, 22).
All social systems are dependent on communication between their members. If there is not enough
communication, the system cannot function properly. In the words of Varela: “In defining a system,
when conceiving something about it, one is already part of it.” (Varela & Johnson, 1976, 31).
According to this theory, then, passive membership of an autopoietic system is impossible;
membership has to be based on active involvement and interaction. Each individual in the network,
for instance, influences the system and contributes to its reproduction, but at the same time the
network also constantly changes the individual and the individual‟s relations of interaction.
The theory of autopoiesis emphasizes being as something. Being, however, is not seen as a static
condition, but above all as a process in which the system continuously produces and reproduces
itself. The aim of autopoietic organization, then, is the system itself and its existence – not “doing”
or “representing”. Autopoiesis is a property of a system, reproducing itself (internally) in a way so
as to preserve its organization, which is to say its identity. The way the system is organized is, in
fact, the system‟s identity: it is on this basis that the system can be identified and distinguished from
other systems (cf. Rapoport, 1986, 114).
The autopoietic system has a special relationship to its environment. von Bertalanffy‟s open
systems and Prigogine‟s self-organizing systems are both dependent on the environment – or at

13
least the environment heavily influences them. The autopoietic system, by contrast, is essentially
autonomous. Maturana and Varela say that the environment is a mirror - or point of reference - for
the autopoietic system, i.e. the system lives in relation to the environment but is not dependent on it
(Maturana & Varela, 1987, 75). Seen from the point of view of its organization and maintenance,
then, the autopoietic system is closed. This means that the system only realizes its own autopoiesis,
i.e. its own existence. However, autopoietic systems are closed only as far as their essence is
concerned; this does not apply to any other of their functions. In order to ensure that their other
functions remain effective, autopoietic systems must engage in exchange with their environment. A
cell, for example, communicates with its environment, unlike the genetic code that controls the cell.
Whereas structure and function in self-organizing systems can sometimes change quite radically, in
autopoietic systems they usually remain constant (Jantsch, 1981, 65).
Varela described the autopoietic phenomenon in the context of a social system as early as 1976. He
defined a system as a being that always has clear boundaries, although those boundaries vary
depending on the observer. Varela argued that in reality, persons who define the boundaries of the
social system are themselves an integral part of the system, and personal needs and perspectives
always influence their view on the system. This means that all social systems are self-referential in
that they always define themselves (Varela & Johnson, 1976, 26-31). The logic of self-reference
can be summarized as follows: what we see is always a reflection of what we are. According to
Varela (Varela & Johnson, 1976, 29), every characteristic that we identify in an object is always
dependent on ourselves as observers. In other words, objects never appear to us objectively, as
assemblies of their own inherent characteristics, but every individual perceives that object through
the lens of their own characteristics, and partly as a result of the interaction that they themselves
have created. All system characteristics are thus filtered and expressed through the observer‟s own
characteristics.
Renewal is not a fundamental characteristic of autopoietic systems; instead the key lies in the
“constitution of the unity to be reproduced” (Maturana, 1981, 23). Autopoietic renewal does not
primarily mean regeneration, but rather maintenance of the core of the autopoietic system. As
Kickert (1991, p.198) has shown, renewal requires a constant and ongoing struggle. Even though
autopoiesis refers primarily to maintenance, it also requires constant renewal of the system. As
systems everywhere are in a constant process of natural degradation (according to the second law of
thermodynamics), maintenance itself requires constant renewal. However even maintenance does
not simply mean the reproduction of the same models in similar conditions, but the system also
works constantly to renew its elements and their mutual relationships.
To sum up, an autopoietic system has two distinctive characteristics:
1) A core that finds expression through interaction. The essence of a system cannot be
understood without studying the interaction taking place within that system. The main
purpose of an autopoietic system is its existence, which is characterized by the reproduction
of its own core, i.e. the continuity of its own identity.
2) An overall view of the system cannot be gained from the outside. When an individual
describes or defines a system, he or she is already part of that system (Varela & Johnson,
1976, 29). The process of defining is itself active involvement and participation, a process in
which the individual‟s view of the system is formed only in interaction. The essence of the
system cannot be defined from the outside; it can only be properly understood by someone
actively involved in the system.
A human being, for instance, is an autopoietic system. In accordance with the first of these two
characteristics, the sole purpose of the human individual is “existence” and “becoming a self”. The

14
true nature of the individual is always expressed in his or her way of interacting with the
environment and other people. In accordance with the second characteristic, it is impossible to
define or characterize someone else without the person‟s own characteristics impacting that
assessment. Whatever statements or arguments the person makes about another person (s)he will
always simultaneously reveal something about him- or herself. For example, from the comment that
“He is an extremely dominating person”, it will not be clear to the listener whether that person is
bossier than usual, or whether the comment reflects more on the person making the statement, say
that it is hard for that person to hold his own or that he takes the view that people are supposed to be
modest and humble.
Like Prigogine, Maturana and Varela also deal with the process of change and renewal. Their
perspectives, however, are quite different. Prigogine emphasizes dramatic changes that affect
structures and basic functions, i.e. vacillation between chaos and order. Maturana and Varela, by
contrast, emphasize continuity and maintenance as a core a systemic function, which implies
ongoing, incremental change for system maintenance. For instance, almost all cells in the human
body are replaced over a period of two years, yet people can still be identified throughout their life.
Thus both incremental change and stability are simultaneously present in autopoietic systems.
is demonstrated in
the system´s
being
is not possible without
always influences
how one perceives
interaction
self-reference
Figure 1. The autopoietic nature of systems (Ståhle, 1998, 81)
4.1 Applications of autopoiesis theory
The concept of autopoiesis has attracted widespread attention and applications have been put
forward in virtually all fields of systems research. The most significant applications have come in
the fields of biology and medicine (e.g. Boden, 2000; Naohide, 2005) and in the fields of human
networking (e.g. Plass et al., 2002), knowledge management (e.g. Okada, 2004; Jackson, 2007) and
knowledge creation (e.g. Ratcheva, 2007; Thompson, 2004), physics (e.g. Tsytovich et al., 2007)
and social sciences (Nomura 2002). For the most part, however, the references are recognitions of
the phenomena rather than rigorous theoretical analyses of autopoiesis. Consequently the
applications suffer from some grave weaknesses. First, the interpretations offered tend to over
emphasize the concept by treating autopoiesis synonymously with autonomy. Second, no clear
conceptual distinction is elaborated between autopoietic, complex, chaotic and self-organizing

15
systems, i.e. it is assumed that complex systems are autopoietic and that self-organizing systems are
autopoietic.
One noteworthy exception is Maula‟s (1999) more analytic treatment of autopoiesis. ShHEr
application of autopoiesis theory is embedded in the context of multinational companies‟ learning
and evolving in complex environments. Maula points out that an autopoietic analyses allows for the
identification of new principles that can explain the evolution of firms. At the same time this
analysis sheds light on multinational companies‟ underlying structures and processes, especially on
the knowledge flows and the consistency of their strategic composition. The findings indicate that
autopoiesis theory can be extended to cover the production of various non-physical components. In
particular, it provides a new tool for the analyses of strategic composition, i.e. a selection of
strategic components and their relationships. Furthermore, it allows for the redefinition of such
concepts as identity, knowledge and strategy can be redefined in a larger interconnected self-
producing system. The research suggests that the less-structured „informal‟ and „chaotic‟
communication can have far-reaching implications for the evolution of firms and is therefore a
relevant topic for further research. (ibid., 346, 347, 350.)
5. Luhmann’s self-referential systems
Maturana and Varela were biologists and developed their theory of autopoiesis primarily in the
context of natural sciences. German sociologist Niklas Luhmann has expanded this theory and
applied it to social systems in a noteworthy manner. According to Fuchs (1988, 21), “at present,
Luhmann‟s theory of social systems is the only general theory that can claim to introduce a new
paradigm … Luhmann‟s proposal will radically change the conventional ways of doing social
theory”. Luhmann is convinced that social systems are autopoietic, and it is a recurring theme in
several of his works (1982b, 1984a, 1984b, 1986, 1990, 1995a, 1995b). He goes so far as to argue
that the theory of autopoietic social systems requires a conceptual revolution in sociology.
According to Luhmann the foundation of the system lies in communication. Social systems use
communication as a means of autopoietic renewal: it is only by means of communication that the
system is capable of maintaining and duplicating itself. By communication, Luhmann refers to
activity or to an event rather than the subject of communication (Luhmann, 1986a, 174). In the
theory of autopoietic systems, communication is the basic unit of self-referential processes
(Luhmann, 1986a, 177). Communication is based on contacts that are constantly created and
renewed by the network of interaction and that cannot exist outside of the network. In this sense
autopoiesis means that continuity requires communication (Luhmann, 1990, 3, 14).
Maturana, too, repeatedly points out that the autopoietic (social) system is composed of
communication, not components (e.g. people). An autopoietic system can be defined as an entity
that consists of the relationships in which its components are reproduced (Maturana, 1981, 29).
Luhmann agrees that the autopoietic system constantly creates itself, i.e. its essence. This is a
process in which the system constantly reproduces its basic components in a network formed by
those components. The outcome may be some form of biological life, consciousness or (in the case
of a social system) communication. Whatever the outcome, the system that is created in the
autopoietic process is always distinctive and clearly identifiable in relation to its environment.
Autopoiesis is these systems‟ way of being and self-reproducing (Luhmann 1989, 143).

16
Communication as the basic unit of systemic processes refers to activity, an event and
understanding. Understanding does not mean one has to approve of the content communicated, but
that communication always leads in open situations either to approval or rejection of the content. In
other words the function of communication is not to achieve mutual understanding. By contrast
communication may force situations to change because it leads to choices without which interaction
would never happen. Only communication itself can create situations that open up new possibilities
to achieve a point of bifurcation, which in turn pave the way to different future scenarios.
(Luhmann, 1986c, 176.)
According to Luhmann (1995a, 37), the most important factor in the system‟s self-renewal is
controlling its complexity. This is not, however, a matter of manipulation from outside the system,
but rather of controlling complexity from within. This perspective is also reflected in the way that
Luhmann defines the autopoietic system. In addition to open, closed and self-organizing systems,
Luhmann introduces a new category of systems, namely self-referential systems. Self-referential
systems can regulate their own boundaries, i.e. they open and close autonomously and are thus at
once both closed and open.
Below, I proceed to discuss Luhmann‟s concept of self-referential systems in somewhat more
detail. I examine his main ideas via three of his main concepts: 1) self-referential closure, 2) double
contingency and 3) processing meaning.
Self-referential closure: the foundation of autonomy
Luhmann (1995a) says that the autopoietic system is fundamentally autonomous and independent of
its environment, and in that sense closed. Self-referential closure means that the system can choose
either to open up or to remain closed and use the information gleaned from the environment in its
own processes of renewal. In this way the system remains autonomous and independent, but at the
same time communicates with the environment and is open to the environment on its own terms
The self-referential autopoietic process is dependent on the ability to make a distinction between
oneself and the environment. Luhmann says that if the autopoietic system did not have an
environment, it would need to create one in which to reflect itself (Luhmann, 1986a, 175). However
Luhmann‟s notion of self-reference does not mean that the system would directly create an image of
itself on the basis of what it sees in the mirror. Rather, it would be looking into what may be
described as a “negative mirror”, which means that the system creates an image of itself on the basis
of the image in the mirror, but it does not draw information about itself directly from the image;
rather it uses the image to create a perception of itself as distinct from its environment. This process
may be described as one of negative mirroring in which the system learns to recognize what it is not
like, i.e. how it differs from the other (systems).
According to Luhmann self-referential systems are characterized by the ability of self-referential
closure (Luhmann, 1990). Without this ability the system would be unable to set itself apart from
the environment as an autonomous being and become interwoven as part of the environment. The
system reflects its autonomy via self-defined and self-regulated boundaries. Because system
renewal takes place via the system‟s internal dynamics, the role of the environment in transactions
is seen in a different way when compared to the theory of open systems. Despite these differences
the views are not fundamentally at odds with each other, because even self-referential systems
exchange information with their environment – it is only that they regulate this interaction
autonomously by opening and closing their boundaries depending on the situation. Luhmann
(1995a, 29) emphasizes that the role of system boundaries is highly significant in new systems

17
thinking. Boundaries represent the evolutionary peak of the system and reflect the operation of the
most advanced systems.
Double contingency: trust and equality
According to Luhmann (1995a, 118) the basic explanation for social action lies in the relationship
between two persons. Systemic change is not primarily reduced to individuals, but to their
relationship. The core of a self-referential system is manifested in double contingency: all
individuals in the system live in a network of reciprocal dependencies. Without these dependencies,
the system lacks the necessary connectivity. It is precisely by virtue of its internal relationships that
a system can form a coherent entity – without those relationships there would be no system.
From a social point of view the key issue with regard to autopoietic systems does not have to do
with self-reproduction, but with systemic development: how the system moves from one point of
departure to the next. According to Luhmann (1995a, 36) the answer lies in the system‟s
relationships of double contingency: these contingencies determine the possibilities of change and
learning.
Double contingency relations are always symmetrical and voluntary. Symmetry means that both
parties are aware of their contingent relationship. Voluntariness, then, means that both parties
accept this relationship of reciprocal dependence (Luhmann, 1995a, 108, 125). Communication in
this type of relationship always involves risks. If the individual is unable to take risks, or to
overcome the fear of the unknown, “the system is undetermined and thereby blocked.” (ibid., 131).
One of the key preconditions for double contingency is the development of trust or distrust. A
person who shares a lot of trust also enhances his or her scope of action. However trust inherently
includes the possibility of distrust and is therefore highly sensitive. Breaking trust will necessarily
bring changes to the relationship (ibid., 128). Trust is always freely handed out according to the
situation; it cannot be forced or manipulated. Trust makes it possible for the system to develop and
on the other hand provides it with the power for ever riskier self-renewal. Trust is not based on
reported factual information, but information serves as an indicator of trust. Trust is a universal
precondition for action (ibid., 129). Luhmann emphasizes that every system first puts trust to the
test and only then proceeds to process meanings – and specifically and only in this order (ibid.,
112).
Processing meaning: information as an experience
The processing of meaning takes places in double-contingency relationships (Luhmann, 1995a,
113). When information is considered in a systemic context, it refers more to an “event” than to a
“fact”. Information, in a systemic context, refers to the kind of facts, information or knowledge that
has some impact on the system. In other words, information is defined not through its form or other
characteristics, but only through its impact. Information that is repeated in a system no longer
serves in that system as actual information, because it no longer changes the state of the system.
When information is repeated in identical format, it does retain its meanings, but it can no longer
impact the systems – therefore it does not function as information. Information changes the state of
the system. Information is more of an experience than a fact. Information is the basic unit of an
event in a system: this is not just data referring to facts, but information that affect people
personally. Only if information causes reactions (i.e. changes the state of the system) will it become
a process element. (ibid.,67, 69.)
According to Luhmann (1995a) meanings are core structural elements of the system. Psychic and
social systems have developed in and through a process of evolution in which both complexity and
self-reference are necessary. This achievement of evolution can be described as “meaning” (ibid, p.

18
59). “Not all systems process complexity and self-reference in the form of meanings but for those
that do, this is the only possibility… Systems bound to meaning can therefore never experience or
act in a manner that is free from meaning (ibid., 61-62). According to Luhmann the core of
interaction lies in meaning, because meaning is created and it materializes in the event of
interaction. Meanings are created in an evolutionary process as a result of human interaction. In the
search for meaning the system realizes its potential: contradictory experiences and views engender
activity, which in turn evolves into goal-oriented action. The social structure in self-referential
systems is always created through the processing of meanings (ibid., 61-65, 113).
The speed of systemic renewal is proportional to the speed at which meanings develop. This refers
to the ability of the system to make rapid choices to develop and create information. Systemic
coherence is important, because without it there can be no double contingencies in the system; and
without double contingencies, the system would not be able to produce or test meanings. The
function and purpose of the system are based on meanings (Luhmann, 1995a, 119), and double
contingency serves as a kind of internal accelerator in the system (ibid., 131).
5.1 The system’s capacity for self-reference
Self-reference is the starting-point for all communication within the system. Self-reference is the
core that is fundamentally autonomous, but it evolves in a reflective relationship with the
environment. Being and becoming thus lie at the heart of all renewal.
Luhmann says that self-renewal can be seen as an event that is based on three decisive criteria. The
first is double contingency. The quality of social relations is essential to the system‟s capacity for
self-renewal, i.e. the participants must encounter one another at the same level. The mutual
dependence must be recognized and admitted, the risk involved in developing relations of trust must
be taken into account and the participants must act accordingly. Double contingency does not
require mutually shared values, symbols or consensus (Luhmann, 1995a, 172-173, 126). Interaction
does, on the other hand, necessarily require mutual trust and recognition of a mutual relationship of
dependence.
The second criterion concerns the quality of information. Exchange of information, i.e.
communication is a necessary condition for the system to function, because no action is produced
without communication. Whether or not the system is capable of autonomously renewing itself
depends on the quality of information that is exchanged within the system. Luhmann emphasizes
the importance of information that becomes a driving force and process element of the system. By
this, Luhmann refers to information that is shared in the discourse of experience – i.e. information is
related to the speaker‟s experience and at once engenders experiences in the listener. In practice this
means that the information exchanged influences the people that constitute the system and in this
way changes the state of the system. Information that does not change the state of the system is
meaningless. A systemic message is never superficial, but on the contrary it always has some
impact or another. (Luhmann, 1986c, 174.)
The third criterion is related to meanings. Luhmann shows that meanings are created collectively
within the system, as a consequence of collectively produced events. Meanings are never fully
ready, nor can they be directly transferred to others. The creation of meaning always requires
double-contingent relations, which in turn engender action. Meanings are thus basic structural
elements of a system and all operations are based and steered by meanings.

19
From Luhmann‟s presentation we can extract the following criteria for self-renewal that serve as the
basic operational preconditions for a self-referential system:
(1) connection with other systems
 use as a point of reference
(2) double contingency
 symmetric dependence (balance of power)
 voluntary provision of trust
(3) experiential information
 information as an event
 information produced in an experiential discourse that has the power to
change the state of the system
(4) creation of collectively produced meanings
These criteria are demonstrated in the way that the system refers to itself. At the same time they
demonstrate how self-reference is the way in which the system controls its own internal complexity.
MAIN
CONCEPTS
Self-reference
Double
contingency
Information
Meaning
Characteristic
Identifiable
core, self -
reproduction
Mutual,
recognized
positive inter-
dependency
Experience –
not
representation,
Event – not fact
Basic element of
system operation
and structure
Manifestation
Definition of
self, self-
referential
closure,
contacts with
other systems
Equality,
trust,
risk-taking
Information
exchange,
reactions and
responses,
dialogue
Collective
processing (in
double-
contingent
relations)
Contribution to
self-renewal
Internal control
of complexity
Internal system
accelerator
Power of
renewal
Actualizes
system‟s
potential
Table 2. Description of self-referential systems (adapted from Ståhle, 1998, 90)
Autopoiesis as set forth by Luhmann as a social systems theory has had an immense impact on
social systems thinking and its recent development. The concept of autopoiesis has been developed
in intrinsic detail in fields as diverse as gender research (Misheva, 2001) and history (Gregory,
2006). It has been deployed in a wide range of studies from organizational studies to small
companies (Koivisto, 2005; Christensen, 2003) to research on global politics and law (Albert, 2002;
Albert & Hilkemeier, 2004; D‟Amato, 2003) and the competitiveness of multinational corporations
(Hessling & Pahl, 2006). At the same time it has been the focus of conceptual criticism and
development (Jalava, 2003; Gumbrecht ,2001).

20
6. The dynamics of self-renewal in organizations
The third paradigm of systemic thinking highlights the fact that each system has its own built-in
spontaneous dynamics and potential that can be exploited in the right circumstances. In a
mechanical way of thinking the components of the system, say the members of an organization can
be harnessed to pursue predetermined goals and objects with the support of management and
control systems. But for example Prigogine‟s main idea is that in certain circumstances and under
certain conditions, systems are capable of organizing themselves, i.e. producing completely new
physical, social and mental structures that are not just an incremental step forward, but an
innovation-like shift. How this happens in real organizations is a huge challenge that may generate
significant competitive advantage for companies in knowledge economies where competitiveness is
mainly based on brands (company “self” and identity) and innovations.
The secret of innovative development and by the same token of organizational competitiveness lies
in whether or not people can learn to make good use of the capacity of self-organization, or whether
that potential is constrained by excessive control. As Prigogine points out, in some circumstances
chaos produces nothing but confusion; in others it may produce radical innovations. The systemic
preconditions for innovation are concentrated, for example, in the system‟s ability to cope
constructively with conflicts and threats to its own power structures and ways of thinking. The
questioning of the status quo and openness to new possibilities presents a huge challenge at both the
individual and organizational level.
3
The processing of information, i.e. an entropy-producing communication process also runs counter
to the order on the strength of which organizations have learned over time to operate. The process
of self-organization requires a great deal of the participants involved. First of all it is necessary to
have a high tolerance of the uncertainty that grows out of the initial confusion. If a solution is
forced before there is a sufficient amount of entropy, self-organization – and with it any new
solutions or innovations – will not happen. On the other hand self-organization also requires an
ability to make good use of points of bifurcation, i.e. to reject even good templates or ideas and
most of the work that has gone into them, to make the right decision and proceed accordingly. In
this process solutions do not come about by vote, but by communication: as a rule the material will
speak for itself and begin to self-organize so that the next steps are clearly evident. In the
individual‟s creative work process this stage often follows close on the heels of the moment of
insight, with the solution effectively surfacing out of its own accord. It appears to come out of
nowhere, but it has in fact been preceded by extensive, both conscious and unconscious collection
and processing of information.
For Luhmann, self-renewal is a rather different concept than it is for Prigogine. Whereas Luhmann
emphasizes the established identity of the system, i.e. the capacity of the system to constantly
reproduce itself as an identifiable self, Prigogine is interested to study the system‟s visible self-
organization, its spontaneous transformation that eventually produces a new structure. Luhmann
emphasizes continuity, process-like development without crises, whereas Prigogine emphasizes
more sudden and dramatic change.
Assessments of organizational renewal and competitiveness tend usually to focus exclusively on
action, on what it has done to achieve certain goals. In today‟s high-paced and insecure competitive
3
For a detailed synthesis of the preconditions for a self-renewing system see Ståhle, 1998, 227.

21
environment, business organizations must constantly work to identify and define their own
competitive assets. Much attention has been paid to products and services, but there is also a
growing recognition that the creation of attraction, an image or brand is in fact often more important
that the development of a specific product. People are intrigued by the aura and identity of a
business, and their decisions are largely driven by their desire of identification. Persona or identity,
the system‟s fundamental essence, Luhmann says, is reflected in interaction. In other words the true
essence of a business company is reflected not in what the company says it is, but in how it is
reflected in all its activity. It is not enough that the company declares its mission and its values
verbally or in writing, but the core of the system lies in its genuine action. The more strongly that
core is transmitted to others, the greater its appeal and attraction – both from internal and external
perspective.
The quality of the information or knowledge processed in organizations is of paramount importance
to the achievement of results. This means that the micro-level communication processes are
quintessential from the point of view of the capacity for organizational self-renewal. Luhmann does
not subscribe to the importance of the distinction between explicit and tacit information (elaborated
by Polanyi, 1958, and made famous by Nonaka, 1995). Instead he underlines the impact of
knowledge as communication: whatever the form of knowledge, it should act as a force that can
change the system, i.e. knowledge is more an event than a fact. If product development people
exchange information either verbally or by technical means but are not really interested in each
other‟s arguments, the information exchanged will remain meaningless and will not contribute to
development. Indeed there is good reason to ask whether development projects or change
management today pay enough attention to making sure the information exchanged becomes a
shared experience, or whether it merely remains a dead letter. Very often it is much easier to
intervene in structures, processes and forms than it is to genuinely analyze what is really happening
in an organization.
Both Prigogine and Luhmann encourage us to focus on communication processes at the micro level.
They argue that the possibility of self-renewal is reduced precisely to communication. The system’s
capability of interaction will at once determine its changes of renewal, radical change, innovation
and influence. Both Prigogine and Luhmann also draw attention to power structures and the
manifestations of power that steer the processes of communication. In Prigogine‟s view the
production of entropy requires equal exchange of information without power concentrations, which
is a key precondition for self-organization. Luhmann, on the other hand, emphasizes double
contingency and the equality and mutual trust it requires, without which meanings cannot develop
in a system. These are interesting preconditions for development in organizations and other social
systems. When the aim is to put the system‟s development potential to full use, it is necessary above
all to focus on the power structures demonstrated in communication. Recently studies of social
capital in particular have paid much attention to the role of trust in economic productivity and in the
success of partnerships (Blomqvist, 2002). In this sense, too, the pioneering work by Prigogine and
Luhmann opens up important insights that can pave the way to building up competitive advantage
in dynamic business environments.

22
References
Albert, M. (2002). Governance and democracy in European systems: on systems theory and European
integration. Review of International Studies 28: 293-309.
Albert, M. & Hilkermeier, L. (2004). Observing international relations: Niklas Luhmann and world
politics. Routledge.
Argyris, C., & Schön, D. A. (1978). Organizational learning: A theory of action perspective. Reading,
MA: Addison-Wesley.
Arthur, W. B. (1994). Increasing returns and path dependence in the economy. University of Michigan
Press.
von Bertalanffy, L. (1967). General System Theory. In N. J. Demerath & R. A. Peterson (Eds.), System,
change, and conflict (pp. 115-129). New York: The Free Press.
von Bertalanffy, L. (1972a). The history and status of General Systems Theory. In G. J. Klir (Ed.), Trends
in General Systems Theory (pp. 21-41). New York: Wiley-Interscience.
von Bertalanffy, L. (1975). Perspectives on General Systems Theory: Scientific-philosophical studies.
New York: Braziller.
Blomqvist, K. (2002). Partnering in the dynamic environment: The role of trust in asymmetrical
partnership formation. Acta Universitatis Lappeenrantaensis 122.
Boden, M. A. (2000). Autopoiesis and life. Cognitive Science Quarterly, 1, 117-145.
Brown, S. & Eisenhardt, K. (1998). Competing on the edge. Strategy as structured chaos. Boston,
Harvard Business School Press.
Checkland, P. B. (1981). Systems thinking, systems practice. Chichester: Wiley.
Checkland, P. B. (1988). Images of systems and the systems image. In W. J. Reckmayer (Ed.), General
systems (Vol. XXXI, pp. 13-20). New York: ISSS.
Chen, P. (1996). A random walk or color chaos on the stock market? Time-Frequency Analysis of S&P
Indexes. Studies in Nonlinear Dynamics and Econometrics Quarterly Journal July 1996, Volume 1,
Number 2. Available OnLine at: http://www.e-m-h.org/Chen96.pdf
Chen, P. (2000). The frisch model of business cycles - A spurious doctrine, but a mysterious success.
CCER Working Paper 1999007. Available OnLine at:
http://old.ccer.edu.cn/workingpaper/paper/e1999007.pdf
Christensen, K. (2003). Knowledge management in a project-oriented organization: three perspectives.
Journal of Knowledge Management, Vol. 7 no. 3 2003, pp. 116-128.
Collier, J. (2005). Anticipatory systems: Dealing with the unexpected. Available OnLine at:
http://www.kli.ac.at/theorylab/jdc/papers/CASYS2000final.pdf
D‟Amato, A. (2003). International law as an autopoietic system. Max Planck Institute for Comparative
Public Law and International Law. Available OnLine at:
http://anthonydamato.law.northwestern.edu/auto/PLANCK-1.pdf
Dobutsinskis, L. (1987). The self-organizing policy: An epistemological analysis of political life.
Boulder, CO: Westview.
Dooley, K. (1999). Evolution and revolution in the history of the discipline. In Advances in the
Management of Organizational Quality, Vol. 5, JAI Press: 1-28.
Dooley, K. & Bush, D. (1995), TQM, chaos, and complexity. Human Systems Management, 14(4) 1995:
1-16
Engeström, Y., Engeström,R. & Vähäaho, T. (1999). When the center does not hold: The importance of
knotworking. In S.Chaiklin, M. Hedegaard & U.Jensen (Eds.), Activity theory and social practice:
Cultural-historical approaches (p. 345-374). Aarhus: Aarhus University Press.
Faulkner, W. & Fleck, J. (Eds.), Exploring expertise: Issues and perspectives (s. 265-285), London:
McMillan.
Feigenbaum, M. (1982). Universal behavior in nonlinear systems. In Physica D: Nonlinear Phenomena,
Volume 7, Issue 1-3, p. 16-39.

23
Feigenbaum, M. (1993). The transition to chaos. In Chaos: The new science (Nobel Conference XXVI)
University Press of America , 1993.
Fischer, M. M. (2001). Knowledge, complexity and innovation systems. Springer
Forrester, J. W. (1961). Industrial dynamics. Cambridge, MA: MIT Press.
Forrester, J. W. (1968). Principles of systems. Cambridge, MA: Wright-Allen.
Forrester, J.W. (1991). System dynamics and the lessons of 35 years 1. Available OnLine at
http:/www.rtpnet.org/~gotwals/stella/sdg/sdlessons.txt
Fuchs, S. (1988). The constitution of emergent interaction orders: A comment on Rawls. Sociological
Theory, 6, 122-124.
Fuchs, S. (1988). Translation and introduction. In N.Luhmann, Tautology and paradox in the self-
descriptions of modern society. Sociological Theory 6, 21-37.
Glandsdorff, P., & Prigogine, I. (1971). Thermodynamic study of structure, stability and fluctuations.
New York: Wiley.
Gregory, A. (2006). Systemic reflections on our past, present and future. Wiley, Systems Research and
Behavioral Science, Volume 23, Issue 5, 685 – 694.
Griffin, D. (2002). The emergence of leadership: Linking self-organization and ethics. Routledge.
Gumbrecht, H. (2001). How is our future contingent? Reading Luhmann against Luhmann. Theory,
Culture & Society, Vol. 18, No. 1, 49-58.
Hakkarainen, K. 2006. Kollektiivinen älykkyys. Psykologia 06/03, 384-401.
Hakkarainen, K., Palonen, T., Paavola, S. & Lehtinen, E. (2006). Communities of networked expertise:
Professional and educational perspectives. London: Elsevier/Pergamon.
Hessling, A. & Pahl, H. (2006). The global system of finance. Scanning Talcott Parsons and Niklas
Luhmann for theoretical keystones. American Journal of Economics and Sociology 65 (1), 189–
218.
Howells, J. (1997). Management and the hybridization of expertise: EFTPOS in retrospect. Journal of
Information Technology, Volume 12, Number 1, 83-95.
Jackson, T. W. (2007). Applying autopoiesis to knowledge management in organisations. Journal of
Knowledge Management, Volume: 11, Issue: 3, 78 – 91.
Jalava, J. (2003). From norms to trust: The Luhmannian connections between trust and system. European
Journal of Social Theory, Vol. 6, No. 2, 173-190.
Jantsch, E. & Waddinton, H. C. (Eds.) (1976). Evolution and consciousness: Human systems in transition.
Addison Wesley.
Kickert W. J. M. (1991). Applicability of autopoiesis to administration science. In R. J. in‟t Veld, L.
Schaap, C. J. A. M. Teermeer & M. J.W. van Twist (Eds.), Autopoiesis and configuration theory:
New approaches to societal steering (pp. 193-205). AA Dordrecht, Holland: Kluwer Academic
Publishers.
Kohonen, T, Kaski, S, Lagus, K, (2004). Mining massive document collections by the WEBSOM
method. In Information Sciences, Vol 163/1-3, pp. 135-156
Koivisto, T. (2005). Developing strategic innovation capability of enterprises. VTT Publications, 586,
Espoo. Available OnLine at: http://www.vtt.fi/inf/pdf/publications/2005/P586.pdf
Kuscu, I. K. (2001). An adaptive approach to organizational Knowledge Management. Knowledge and
Innovation: Journal of the KMCI, Vol. 2.
Leonard-Barton, D. (1995). Wellsprings of knowledge. Building and sustaining the sources of innovation.
Boston, MA, Harvard Business School Press.
Lorenz, E. (1963). Deterministic non periodic flow. Journal of Atmospheric Science, 20, 130-141.
Lorenz, E. (1993). The essence of chaos. London: UCL Press.
Lorenz, E. (2005). Designing chaotic models. Journal of the Atmospheric Sciences: Vol. 62, No. 5, pp.
1574–1587.
Luhmann, N. (1982). The differentiation of society. New York: Columbia University Press.
Luhmann, N. (1984a). Soziale systeme: Grundrib einer allgemeinen theorie. Frankfurt: Suhrkamp Verlag.

24
Luhmann, N. (1984b). Religious dogmas and the evolution of societies. New York: Edwin Mellan.
Luhmann, N. (1986c). The autopoiesis of social systems. In F. Geyer & J. van der Zouwen (Eds.),
Sociocybernetic paradoxes: Observation, control and evolution of self-steering systems (pp. 172-
192). London: Sage.
Luhmann, N. (1989). Ecological communication. Cambridge: Polity Press.
Luhmann, N. (1990). Essays on self-reference. New York: Columbia University Press.
Luhmann, N. (1995a). Social systems. Stanford, CA: Stanford University Press.
Luhmann, N. (1995b). Why „systems theory‟? Cybernetics and human knowing 3, 3-10..
Mandelbrot, B. B. (2004) Fractals and chaos: The Mandelbrot set and beyond. New York: Springer.
Mandelbrot, B. B. (1977a). Fractals: Form, change, and dimension. San Francisco: Freeman.
Marion, R. (1999). The Edge of Organization: Chaos and complexity theories of formal social systems.
Sage Publications Inc
Maturana, H. R. (1981). Autopoiesis. In M. Zeleny (Ed.), Autopoiesis: A theory of living organization
(pp. 18-33). New York: Elsevier-North Holland.
Maturana, H. R. & Varela, F. J. (1987). The tree of knowledge: The biological roots of human
understanding. Boston: New Science Library.
Maula, M. (1999). Multinational companies as learning and evolving systems. A multiple-case study of
knowledge-intensive service companies. An application of autopoiesis theory. Helsinki School of
Economics and Business Administration. Acta Universitatis Oeconomicae Helsingiensis. A-154.
Helsinki: HeSE print.
Minnis, M. & John-Steiner, V.P. (2001). Are we ready for a single, integrated theory? Human
Development 44, 296-310.
Misheva, V. (2001). Systems theory from a gender perspective. Annual meeting of The Swedish
Sociological Association, Working Group on Gender Research, Uppsala. Available OnLine at:
http://www.soc.uu.se/publications/fulltext/vm_system-gender.doc
Mitleton-Kelly, E. (2003). Complex systems and evolutionary perspectives on organisations: The
application of complexity theory. Elsevier.
Moreland, R.L. (1999). Transactive memory: Learning who knows what in work groups and
organizations. In L.L.Thompson,, J.M.Levine & D.M. Messick (Eds.), Shared cognition in
organizations: The management of knowledge (s. 3-31) Mahwah, NJ: Erlbaum.
Naohide, T. (2005). Basis of bioengineering: Paradigm shift of bio function interpretation. New Diamond
Journal, Vol.21/3, 38-41.
Nicolis, G. & Prigogine, I. (1977). Self-organization in nonequilibrium systems: From dissipative
structures to order through fluctuations. New York: Wiley.
Nomura, T. (2002). Formal description of autopoiesis for analytic models of life and social systems.
Proceedings of the Eighth International Conference on artificial life. Artificial Life VIII, 15 –18.
Available OnLine at: http://alife.org/alife8/proceedings/sub358.pdf
Nonaka, I. 2006. Organizational knowledge creation Theory: Evolutionary paths and future advances.
Sage Publications.
Nonaka, I & Takeuchi, H. (1995). The knowledge-creating company. New York, Oxford University
Press.
Okada, A. (2004). The collective building of knowledge in collaborative learning environments.
Available OnLine at: http://www.projeto.org.br/alexandra/pdf/L1_%20roberts2004_okada.pdf
Piaget, J. (1975). The equilibration of cognitive structures: The central problem of intellectual
development. Chicago: University of Chicago.
Plass, J. L. & al (2002). A living-systems design model for web-based knowledge management systems.
Educational Technology Research and Development. Vol. 50/1, 35 – 56.
Polanyi, M. (1958): Personal Knowledge, Towards a Port-Critical Philosophy. Chicago: University of
Chicago Press
Porter, A. (1969). Cybernetics simplified. London: English University Press.

25
Prigogine, I. (1967a). Dissipative process, quantum states and field theory. XIVe Conseil de Physique
Solvay. Oct. 1967. Bruxelles.
Prigogine, I. (1967b). Introduction to thermodynamics of irreversible processes (3rd. ed.). New York:
Wiley.
Prigogine, I. (1967c). Quantum theory of dissipative systems and scattering processes. Nobel Symposium
V. Stockholm.
Prigogine, I. (1976). Order through fluctuations: Self-organization and social systems. In E. Jantsch & C.
H. Waddington (Eds.), Evolution and consciousness: Human systems in transition (pp. 93-133).
Reading, MA: Addison-Wesley.
Prigogine, I. (1980). From being to becoming: Time and complexity in the physical sciences. San
Francisco: Freeman.
Prigogine, I. & Lefever, R. (1973). Theory of dissipative structures. In Synergetics cooperative
phenomena in multi-component systems. Proceeding of the Symposium on Synergetics, from April
30 to May 6, 1972, Schloss Elmau. Ed. by H. Haken. Pub. B.G. Stuttgart.
Prigogine, I., & Stengers, I. (1984). Order out of chaos: Man‟s dialogue with nature. New York: Bantam.
Prigogine, I., & Nicolis, G. (1989). Exploring complexity: An introduction. New York: Freeman
Pöyhönen, A. (2004). Renewal ability of organizations: Synthesis of theoretical literature and suggestions
for measurement. In M.Hannula & P.Ståhle (Eds.), New directions of Intellectual Capital
Management. Lappeenranta University of Technology
Rangel, O. & Jose, A. (2005). The Systemic theory of living systems and relevance to CAM: The theory
evidence-based complementary and alternative medicine. Oxford University Press.
Rapoport, A. (1986). General Systems Theory: Essential concepts and applications. Tunbridge Wells:
Abacuss Press.
Ratcheva, V. (2007). Redefining multidisciplinary team boundaries in resolving heterogeneous
knowledge dilemmas. International Journal of Intelligent Enterprise. Vol .1/1, 81 – 97.
Reynolds, S. (2003). Higher education administration in a dynamic system. Available OnLine at:
http://www2.unil.ch/pavie/documentation/workingdocument.pdf
Santa Fe group and others (2007) OnLine at: http://www.santafe.com
Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization. London:
Century Business.
Spinardi, G. (1998). Nuclear weapons experts: Bomb building inside and outside the laboratory. In R.
Williams, W. Faulkner & J. Fleck (ed.), exploring expertise: Issues and perspectives (s. 245-264),
London: Mcmillan.
Ståhle, P. (1998). Supporting a System‟s Capacity for Self-renewal. Helsinki, Yliopistopaino.
Ståhle, P. & Grönroos, M.(2000). Dynamic Intellectual CapitalTM. Knowledge Management in Theory
and Practice. WSOY: Porvoo, Helsinki, Juva.
Ståhle, P., Ståhle, S. & Pöyhönen, A. (2003). Analyzing dynamic intellectual capital: System-based
theory and application. Acta Universitatis Lappeenrantaensis, 152. Lappeenranta University of
Technology.
Thompson, E. (2004). Life and mind: From autopoiesis to neurophenomenology. A tribute to Francisco
Varela. Phenomenology and the Cognitive Sciences, Vol. 3, Issue 4, 381-398
Tsytovich, VN. & al (2007). From plasma crystals and helical structures towards inorganic living matter.
New Journal of Physics.
Varela, F. J., & Johnson, D. (1976). On observing natural systems. Co-Evolution Quarterly, Summer
Issue, 26-31.
Zeleny, M. (Ed.). (1981c). Autopoiesis: A theory of living organization. New York: Elsevier-North
Holland.
Weick, K. & Sutcliffe, K. (2001). Managing the unexpected. Assuring high performance in an age of
complexity. Wiley, San Francisco.

26
Wegner, D.1986. Transactive memory: A contemporary analysis of the group mind. In B.Mullen & G.R.
Goethels (ed.), Theories of group behaviour (p. 185-208). New York: Springler-Verlag.
Weick, K. & Roberts, K.H. (1993). Collective minds in organizations: Heedful interrelating on flight
decks. Administrative Science Quarterly 38, 357-381.
Wheatley, M. J. (1999). Discovering order in a chaotic world. Berrett-Koehler Publishers.
Wiener, N. (1950) The Human Use of Human Beings, Houghton Mifflin, Boston, 1950.
Wiener, N. (1948). Cybernetics or Control and Communication in the Animal and the Machine, Hermann
et Cie, Paris, The MIT Press, Cambridge (Mass.), Wiley and Sons, New York, 1948. Second
edition, revised, with two more chapters, The MIT Press, Cambridge (Mass.), Wiley and Sons, New
York, 1961