Anda di halaman 1dari 14


Complexity and Transition

Jan Rotmans and Derk Loorbach

complex adaptive systems
emergence This article presents a framework, transition management, for
governance managing complex societal systems. The principal contribu-
industrial ecology tion of this article is to articulate the relationship between
sustainable development transition management and complex systems theory. A better
understanding of the dynamics of complex, adaptive systems
provides insight into the opportunities, limitations, and con-
ditions under which it is possible to influence such systems.
Transition management is based on key notions of complex
systems theory, such as variation and selection, emergence,
coevolution, and self-organization. It involves a cyclical process
of phases at various scale levels: stimulating niche develop-
ment at the micro level, finding new attractors at the macro
level by developing a sustainability vision, creating diversity by
setting out experiments, and selecting successful experiments
that can be scaled up.

Address correspondence to:

Jan Rotmans
DRIFT: Dutch Research Institute For
Erasmus University Rotterdam
P.O. Box 1738
3000 DR Rotterdam

c 2009 by Yale University
DOI: 10.1111/j.1530-9290.2009.00116.x

Volume 13, Number 2

184 Journal of Industrial Ecology


The idea is that a better insight into the func-

tioning of societal systems provides insight into
Our society faces a number of persistent prob- the possibilities for directing these systems. We
lems whose symptoms are becoming more and use complex systems theory to study the dynamics
more apparent. Persistent problems are complex of societal systems to derive a collection of basic
because they are deeply embedded in our societal guidelines that can be used to direct those sys-
structures; uncertain due to the hardly reducible tems. Obviously, societal systems, because of their
structural uncertainty they include; difficult to complexity, cannot be directed in command and
manage, with a variety of actors with diverse in- control terms. We do, however, hypothesize that
terests involved; and hard to grasp in the sense it is possible to use the understanding of tran-
that they are difficult to interpret and ill struc- sition dynamics to influence the direction and
tured (Dirven et al. 2002). Persistent problems pace of a transition of a societal system into a
are the superlative form of what Rittel and Web- more sustainable direction. The explicit norma-
ber (1973, 160) refer to as “wicked problems.” tive orientation of sustainability is important, be-
An example of a persistent problem is the energy cause historical transitions often have not led to
problem, with anthropogenic climate change as a a more sustainable society (Rotmans 2005). Fos-
manifestation (Energy Council 2004). Persistent tering sustainability transitions is what we call
problems cannot be solved through only current transition management (Rotmans et al. 2001).
policies (Ministry of Housing, Spatial Planning In this article, we first treat basic principles of
and Environment 2002; Social and Economic complex systems theory and of managing com-
Council of the Netherlands 2001). Persistent plex adaptive systems. That results in the formu-
problems are related to the system failures that lation of core theoretical principles for transition
crept into our societal systems and that, contrary management, on the basis of which we present a
to market failures, cannot be corrected by the framework that contains guidelines for applying
market or current policies. System failures are transition management in practice.
locked-in flaws in our societal structures, such as
technological bias, weak or dominant networks,
Complex Systems Theory
institutional barriers, and path dependencies.
Combating system failures requires a restruc- Complexity theory, otherwise known as com-
turing of societal systems—that is, a transition. A plex systems theory, has its roots in the general
transition is a radical, structural change of a so- systems theory that Von Bertalanffy (1968) pub-
cietal (sub)system that is the result of a coevolu- lished in the 1930s. Systems theory is an interdis-
tion of economic, cultural, technological, ecolog- ciplinary field of science that studies the nature of
ical, and institutional developments at different complex systems in society, nature, science, and
scale levels (Rotmans et al. 2001; Rotmans 2006). technology. It provides a framework by which a
In “transition language,” we call the deep struc- group of interrelated components that influence
ture the incumbent regime: a conglomerate of each other can be analyzed. That group can be
structure (institutional and physical setting), cul- a sector, branch, city, organism, or even a soci-
ture (prevailing perspective), and practices (rules, ety. Systems theory evolved over the last century
routines, and habits). And we denote an emer- from deterministic to probabilistic, from a control
gent structure as a niche: a structure formed by engineering to a soft systems approach, and from
a small group of agents that deviate from the partial to integrated. In the 1970s and 1980s, in-
regime and that might build up a new regime tegral systems theory became an important field,
that is able to break down and replace the in- focusing on the integration of social, economic,
cumbent regime. This differs somewhat from the and ecological processes (Holling 1978; Hordijk
common definition of a niche as individual tech- 1985; Rotmans 1990). During this time, soft sys-
nologies, practices, and actors outside or periph- tems theory emerged; it takes a qualitative ap-
eral to the regime, as loci for radical innovation proach rather than a quantitative approach and
(Geels 2005). is mostly applied to companies and organizations

Rotmans and Loorbach, Complexity and Transition Management 185


(Senge 1990). In the 1990s, complex systems the- components. Complex systems have multiple at-
ory was introduced; it focuses on the coevolu- tractors. An attractor is a preferred steady system’s
tionary development of systems (Holland 1995; state set, to which a complex system evolves after
Kauffman 1993, 1995). Although the theory is a long enough time.
far from mature, it has attracted a great deal of Complex adaptive systems are special cases
attention and has many applications in diverse of complex systems. They are adaptive in the
research fields: in biology (Kauffman 1995), eco- sense that they have the capacity to change and
nomics (Arthur et al. 1997), ecology (Kay et al. learn from experience. Expressed differently, they
1999; Gunderson and Holling 2002), public ad- are able to respond to and adjust themselves to
ministration (Kickert 1991; Teisman 1992), and changes in their environment. What makes a
policy analysis (Geldof 2002; Rotmans 2003). A complex adaptive system special is the set of con-
single complex systems theory does not exist: stantly adapting nonlinear relationships. Com-
There are multiple manifestations of it. There plex adaptive systems contain special objects—
are (1) formalized and computational modeling agents that interact with each other and adapt
approaches, (2) a set of “understandings” of the themselves to other agents and changing condi-
behavior of complex systems, (3) metaphorical tions. This is why complex adaptive systems have
use to describe social phenomena, and (4) philo- unique features, such as coevolution, emergence,
sophical considerations about the ontology and and self-organization.
epistemology of complex systems. We take the In the biological or economic context, coevo-
second and, to a lesser extent, the first man- lution refers to mutual selection of two or more
ifestation as a starting point for our transition evolving populations (van den Bergh and Stagl
research. Within this context, complex systems 2004). In the complex systems context, however,
theory attempts to better understand the behav- coevolution is used to indicate the interaction
ior of complex systems that run through cycles of between different systems that influences the dy-
relatively long periods of equilibrium, order, and namics of the individual systems, leading to ir-
stability interspersed with relatively short periods reversible patterns of change within each of the
of instability and chaos. The primary focus is on systems (Kemp et al. 2007). The irreversibility
complex systems, which have the following char- aspect distinguishes coevolution from coproduc-
acteristics, as drawn from the work by Prigogine tion, which indicates mere interaction. Coevolu-
and Stengers (1984), Holling (1987), Holland tion means that a complex system coevolves with
(1995), and Kauffman (1995). its environment—that is, there are interdepen-
Complex systems are open systems that in- dencies and positive feedbacks between the com-
teract with their environment and constantly plex system and its environment (Mitleton-Kelly
evolve and unfold over time. Complex systems 2003). In such a coevolutionary process, both
contain many diverse components and interac- competition and cooperation have a role to play.
tions between components. These interactions Emergence can be defined as the arising of
are nonlinear: A small stimulus may cause a large novel and coherent structures, patterns, and
effect or no effect at all. Conversely, a big stim- properties during the process of self-organization
ulus may cause a small effect. Complex systems in complex systems (Goldstein 1999). Behind
contain feedback loops. Both negative (damping) the notion of emergence is the basic idea that
and positive (amplifying) feedbacks are key in- there may be autonomous properties at a higher
gredients of complex systems. Complex systems (macro) level that cannot be understood by re-
have a history; prior states have an influence on duction to lower (micro) levels (Sawyer 2005).
present states, which have an influence on future Here we speak of emergent properties if a group
states. This creates path dependence, whereby cur- of components has varying properties showing
rent and future states depend on the path of pre- deviant behavior at a higher scale level than the
vious states. Complex systems are nested and en- individual components at a lower scale level. De
compass various organizational levels. They have Haan (2006) distinguishes among three differ-
emergent properties—that is, higher level struc- ent types of emergence: discovery, mechanistic
tures arise from interaction between lower level emergence, and reflective emergence. In systems

186 Journal of Industrial Ecology


exhibiting the latter type of emergence, the ob- sity of components, of relations, of systems be-
servers are among the objects of the system and havior, and so forth.
have some reflective capacity, which enables
them to observe the emergence they produce.
Complex Systems and
Self-organization is a process in which the in-
Industrial Ecology
ternal organization of a complex system increases
in complexity without being guided or managed Without explicit reference to complex systems
by an outside source. This self-organization refers theory, industrial ecology (IE) can be consid-
to the ability to develop a new system struc- ered as a systems approach to societal, predomi-
ture as a result of the system’s internal constitu- nantly production−consumption, systems (Ayres
tion and not as a result of external management and Ayres 1996; Ehrenfeld 1997). A modest IE
(Prigogine and Stengers 1984). The notion of literature explicitly discusses complexity (e.g.,
organization is related to an increase in the struc- Allenby 1999; Kay 2002; Spiegelman 2003). IE,
ture or order of the system behavior. The new loosely based on the analogy with ecosystems,
structures are called dissipative because they fall views industrial systems in terms of material and
apart unless energy is fed from outside to main- energy flows and offers a comprehensive perspec-
tain them (Prigogine and Stengers 1984). Emer- tive, along with concepts and methods, for in-
gence and self-organization are related to each depth analysis. It has drawn attention to the need
other, but they are different. Self-organizing sys- to minimize energy and material flows and of-
tems usually display emergence, but not always. fers models to design ideal−typical “closed-loop
Self-organization exists without emergence, and systems” (Ehrenfeld and Gertler 1997). In its
emergence exists without self-organization. But systemic view, IE tends to be somewhat techno-
in complex, adaptive systems, emergence and cratic in that it fixates on measurable and physi-
self-organization occur together. cal streams and much less or not at all on culture,
Complex adaptive systems continuously adapt governance, agency, and power. It certainly offers
to their changing environment. Any kind of a fruitful basis for debate about (un)sustainable
adaptation and all self-organization (see below) production (e.g., De Vries and Te Riele 2006),
involves variation and selection that is internal to but it does not shed light on the institutional and
the system but may well be external to compo- societal embeddedness of these industrial systems.
nents of that system. Complex adaptive systems Although IE thus offers an analytical frame and a
constantly create variety, in terms of creating future vision, it is much less concerned with the
new components and relations, which provides process of change in-between and how to orga-
a source of novelty in these systems. Selection nize that (Green and Randles 2006).
then maintains the system in a dynamic equi- The complex adaptive system and transition
librium by preventing variation or by pushing it perspectives would consider production and con-
into a certain direction (Green 1994). The selec- sumption rather as subsystems of a societal system
tion process means that the system preferentially (Van der Brugge and Van Raak 2007). Produc-
retains or discards variations that enhance or de- tion of agricultural goods is, for example, largely
crease its fitness (the internalized system’s mea- determined by financial and institutional regu-
sure for success and failure). Most of the time, latory schemes, whereas production of mobility
complex adaptive systems are in a period of dy- technologies might be much more embedded in
namic equilibrium, with ongoing variation and a liberalized, consumer-driven market. In terms
selection but with selection as the predominating of sustainable development, it is clear that sus-
mechanism. External stimuli can force the system tainable production and industrial ecology are
to shift (across the chaotic edge) to a relatively concepts that push an increased eco-efficiency in
short phase of instability and chaos (punctuated production (Korhonen 2004). Herein also lies a
equilibriums), where variation predominates. We danger of optimizing the “wrong” systems by not
can express system variation in terms of diversity fundamentally questioning the need for certain
and heterogeneity. Diversity and heterogeneity are industrial production or the levels of consump-
key features of complex adaptive systems: diver- tion associated with these systems (Braungart and

Rotmans and Loorbach, Complexity and Transition Management 187


McDonough 2002). A complex, adaptive view implies that content and process are insep-
also needs to include at least the possibility for arable. Insight into how the system works is
structural change, along with the influence of out- an essential precondition for effective man-
side forces that could evoke such a transition. agement.
• Objectives should be flexible and adjustable at
the system level. The complexity of the sys-
Managing Complex, Adaptive
tem is at odds with the formulation of fixed
objectives. With flexible, evolving objec-
What does complexity, as described above, tives, one is in a better position to react to
mean in terms of management? Management—in changes from inside and outside the system.
the context of complexity theory—means influ- While being directed, the structure and or-
encing the process of change of a complex, adap- der of the system are also changing, so the
tive system from one state to another. Greater objectives set should change, too.
insight into the dynamics of a complex, adap- • Managing a complex, adaptive system means
tive system leads to improved insight into the using disequilibria rather than equilibriums. In
feasibility of directing it. In other words, appli- the long term, equilibrium will lead to stag-
cation of complexity theory can result in a col- nation and will, in fact, hinder innovation.
lection of basic principles or guidelines that can Nonequilibrium (the period in-between
be used to direct complex, adaptive systems. Re- multiple equilibriums) means instability
flexivity (i.e., reflection on the starting principles and chaos, which form an important impe-
defined) is inbuilt with respect to the assump- tus for fundamental change. The relatively
tions presumed as well as the possible effects of short periods of nonequilibrium therefore
such a form of direction. This results in an under- offer opportunities to direct the system
standing of the limitations of and scope for the in a desirable direction (toward a new
management of complex, adaptive systems and, attractor).
at the same time, provides insight into the oppor- • Creating space for agents to build up alterna-
tunities and conditions under which it is possible tive regimes is crucial for innovation. Stimulat-
to direct such systems. On the basis of theoretical ing emergence and divergence is crucial for
knowledge and practical experience with com- innovation. A diversity of emerging niche
plexity theory, we present a number of guide- agents at a certain distance from the regime
lines for management below. These guidelines can effectively create a new regime in a pro-
are partly descriptive, in the sense of basic princi- tected environment. For this to happen, a
ples, and partly prescriptive, in terms of rules for certain degree of protection is needed to
management. permit agents time, energy, and resources.

• Management at the system level is impor-

Managing Societal Systems
tant. Unintended side effects and adverse
boomerang effects can only be recognized The management principles underlying tran-
at the system level. A system-level perspec- sition management are built around the paradox
tive helps one to get a better insight into that societal change is too complex to handle
spillovers of the complex problem. This in terms of managing, but still we have formu-
implies management at various scale lev- lated a set of relatively simple rules regarding
els: Emergent properties might be hidden how to influence societal change. The rationale
at a higher (or lower) scale level but are for handling this management paradox is that
already beginning to emerge at other scale gaining insight into societal complexity by tak-
levels. ing a complex systems approach can help one
• The status (in terms of its performance) of the to fathom the possibilities for influencing soci-
system determines the way it is managed. The etal complexity. This logically connects content
dynamics of the system create feasible and and process, which are explicitly linked in tran-
nonfeasible means for management: This sition management: The complexity analysis of a

188 Journal of Industrial Ecology


societal system under observation also determines velopment of these niches and the emergence of
the opportunities for managing such a system niche regimes.
(Loorbach 2007, 86). Analytical lenses such as The focus on frontrunners is a key aspect
the multistage, multilevel (Rotmans et al. 2001), of transition management. In complex system
and multipattern concepts (de Haan and Rot- terms, frontrunners are agents with the capac-
mans 2008) provide us with opportunities for ity to generate emergent structures and operate
identifying patterns and mechanisms of transi- within these deviant structures. They can only do
tional change. Once we have identified transi- that without being (directly) dependent on the
tional patterns and mechanisms, we can deter- structure, culture, and practices of the regime. In
mine process steps and instruments to influence the context of transition management, we mean
these patterns and mechanisms. Our approach by frontrunners agents with peculiar competen-
differs from earlier attempts to use a complex cies and qualities: creative minds, strategists, and
systems approach for management of policy is- visionaries. If a new regime is to be created ef-
sues (e.g., Kickert 1991; Kooiman 1993; Stacey fectively, agents are needed at a certain distance
1996) in that it is more oriented toward reflexive from that regime.
planning—not deterministic but reflexive rules. Another principle of transition management
We have formulated rules for managing societal is guided variation and selection. This is rooted in
change, but we realize that once we apply these the notions of diversity and coherence within
rules in a process context, they need to be ad- complexity theory. Diversity helps avoid rigidity
justed because the conditions and dynamics (con- within the system; without it, the system could
tent) will change as a result of the application of respond flexibly to changes in its environment.
these rules. Therefore, learning, searching, and Coherence refers to the level of interrelatedness
experimenting are crucial in transition manage- among the entities of a complex system. In the
ment. In that sense, it has similarities with strate- equilibrium phase, there is continuous variation
gic niche management—that is, experimenting and selection, but when a regime settles, it be-
with new technologies in an experimental space comes the dominant selection environment and
(Kemp, Schot and Hoogma 1998). thus decreases the diversity. But a certain amount
of diversity is required for us to explore a variety of
innovative options instead of looking for the op-
timal solution. Rather than selecting innovations
Principles of Transition
in a too early stage, we keep options open to learn
about the pros and cons of available alternatives
Here we briefly describe the theoretical prin- before making a selection. Through experiment-
ciples of transition management that arise from ing, we can reduce some aspects of the high level
complexity theory. The first principle is that of of uncertainty, which leads to better informed
creating space for niches in so-called transition are- decisions.
nas. The notion of arena originates from that part The principle of radical change in incremental
of complexity theory that indicates that a small steps is a paradox that is derived from complex-
initial change in the system may have a great im- ity theory. Radical, structural change is needed
pact on the system in the long run. In systems to erode the existing deep structure (incumbent
terms, this is called an emergent structure: an en- regime) of a system and ultimately dismantle
vironment that offers some protection for a small it. Immediate radical change, however, would
group of agents. The self-organizing capacity of lead to maximal resistance from the deep struc-
the system generates new, dissipative structures ture, which cannot adjust to a too fast, radical
in the form of niches. A niche is a new structure, change. Abrupt forcing of the system would dis-
a small core of agents, that emerges within the rupt the system and would create a backlash in
system and that aligns itself with a new configu- the system because of its resilience. Incremental
ration. The new alignment is often the emergent change allows the system to adjust to the new cir-
property of the system. An emergent structure cumstances and to build up new structures that
forms around niches, stimulating the further de- align to the new configuration. Radical change in

Rotmans and Loorbach, Complexity and Transition Management 189


incremental steps implies that the system heads emergent structures arise from the interactions
in a new direction toward new attractors, but in between components at the micro level. Every
small steps. transition domain has its own dynamics: Cul-
Empowering niches is an important principle of tures only change slowly, but economic changes
transition management. By empowering, we mean take place in the short term, whereas institutional
providing with resources, such as knowledge, fi- and technological changes are somewhere in be-
nances, competences, lobby mechanisms, exemp- tween. The various domains shift over each other
tions of rules and laws, and space for experiment- and constantly influence each other through in-
ing (Avelino 2007). An empowered niche may teractions and feedbacks. The resulting dynamics
cluster with other empowered niches and emerge are a hybrid picture of alternating fast and slow
into a niche regime. Multiple regimes coevolve change. Analyzing the interactions and feedbacks
with each other—a dominant regime and one or across levels and domains is of importance for
more niche regimes. Crucial is the coevolution of identifying patterns and mechanisms of transi-
a regime within the existing power structure and tional change and for determining instruments to
a niche regime outside the power realm. Coevolv- influence these patterns and mechanisms.
ing regimes influence each other in an irreversible Through experimental implementation of the
manner, with an unknown outcome. The niche complex adaptive systems approach to transitions
regime may take over the incumbent regime but in societal systems, we have translated the theo-
may also be absorbed and encapsulated by the retical principles underlying transition manage-
incumbent regime. ment into so-called systemic instruments. Table 1
Anticipation of future trends and develop- summarizes the main insights from complexity
ments, with account taken of weak signals and theory and their translation into theoretical prin-
seeds of change that act as the harbingers of the ciples of transition management as well as these
future, is a key element of a proactive, long-term system instruments. The next section describes
strategy of transition management. This future a framework for doing transition management in
orientation is accompanied by a strategy of adap- practice, using theoretical principles of complex
tation, which means adjusting while the struc- systems theory.
ture of the system is changing. This requires ad-
equate insight into the dynamics of a complex
system. Although in general, complex system dy-
Transition Management: The
namics are highly nonlinear and unpredictable,
there are periods when the system behaves in a The challenge with transition management
relatively orderly manner and, to a limited ex- is to translate the above, relatively abstract
tent, is predictable. But there are also periods in management rules into a practical management
which chaos rules and the behavior of the system framework without losing too much of the com-
is quite unpredictable. So although the degree of plexity involved and without becoming too pre-
predictability is rather small, transitions do imply scriptive (Rotmans and Kemp 2008). We have
generic patterns that indicate the future pathway. attempted this by delineating transition manage-
Path dependency is an example of such a pattern. ment as a cyclical process of development phases
A transition is the result of a coevolution at various scale levels. In complex system terms,
of economic, cultural, technological, ecological, transition management can be described as con-
and institutional developments at different scale sisting of the following steps (Loorbach 2007;
levels. So transitions, by definition, cross multiple Loorbach and Rotmans 2006):
domains and scales (Rotmans et al. 2001). Com-
plex systems also involve multiple domains and 1. Stimulate niche development (emergence,
scales. They are nested and encompass various variation) at the micro level and try to
organization levels, where higher level structures interconnect niches with the same di-
arise from interaction between lower level com- rection. In the transition management
ponents. The transition literature often makes framework, one does this by establish-
clear that there is a macro level at which novel ing and organizing a transition arena, a

190 Journal of Industrial Ecology


Table 1 Linking of complexity characteristics, theoretical principles of transition management, and systemic
instruments for transition management
Complexity Theoretical principles of Systemic instruments for
characteristics transition management transition management
Emergence Creating space for niches Transition arena
Dissipative structures Focus on frontrunners Transition arena and competence
Diversity and coherence Guided variation and selection Transition experiments and transition
New attractors, punctuated Radical change in incremental Envisioning for sustainable futures
equilibriums steps
Coevolution Empowering niches Competence development
Variation and selection Learning by doing and doing by Deepening, broadening, scaling up
learning experiments
Interactions, feedbacks Multilevel approach, Complex systems analysis
multidomain approach
Patterns, mechanisms Anticipation and adaptation Multipattern and multilevel analysis

quasi-protected area for frontrunners importance in each cycle. In the real world, the
(niche players and change-inclined regime transition management activities are carried out
players). partially and completely in sequence, in parallel,
2. Try to find new attractors for the system and in a random sequence.
by developing a sustainability vision and In effect, transition management comes down
derived pathways at the macro level that to creating space for frontrunners (niche play-
can act as guidance for niche development. ers and change-inclined regime players in tran-
3. Try to stimulate the formation of niche sition arenas), forming new coalitions around
regimes by creating coalitions and new net- these arenas, driving the activities in a shared
works around the transition agenda and and desired direction, and developing coalitions
the different pathways. and networks into a movement that puts soci-
4. Create diversity by setting out transition etal pressure on regular policy. In the transition
experiments that are related to specific management framework, activities related to the
pathways onto the vision. content (integrated systems analysis, envision-
5. Select the most promising ones that can ing, agenda building, and experiments) are linked
be scaled up to a higher level as you learn to activities related to the process (network and
from these experiments and develop an up- coalition building, execution of experiments, and
scaling strategy. process structuring). The preferred actors to be
6. Try to further modulation between the mi- involved (based on the necessary competencies)
cro and macro levels (coevolution) by ad- and instruments (e.g., scenarios, transition agen-
justing the vision, agenda, and coalitions, das, monitoring instruments) are derived from
if necessary, by monitoring and evaluat- this framework. The four activity clusters de-
ing (analyzing patterns and mechanisms) picted in figure 1 are described in more detail
the transition management process, after below.
which the cycle starts again.
Integrated Systems Analysis and Actor
For the sake of simplicity, we present the cy-
cle of transition management as a sequence of
steps, as presented in figure 1. In practice, how- An integrated systems analysis forms the basis
ever, there is no fixed sequence of steps in tran- of every transition management process, provid-
sition management, and the steps can differ in ing a common ground for a variety of actors and

Rotmans and Loorbach, Complexity and Transition Management 191


Figure 1 The transition management cycle.

Source: Loorbach (2007)

enough information for informed debates and dis- understanding that leads to a new perspective on
cussions. Informed insight into the complexity a transition issue. Such a transition arena has to
of the system, its major defining subsystems, the be supported by political actors or regime powers
dominant causal relations, feedback loops, the but not dictated by them—for example, through
roots, and the nature of structural problems estab- the support of a minister or a director. In gen-
lish a baseline as well as conditions for discussing eral, around 15 to 20 frontrunners (i.e., pioneer-
visions, strategies, and actions in the future. In ing individuals) are involved in the beginning of
addition, such a preliminary assessment yields the transition arena, although, over time, only
knowledge about the main actors influencing the around 5 will become the core group.
system in both a conservative and an innovative Within the transition arena, multiple in-
way and helps to guide the selection of partici- depth discussions take place, structured accord-
pants for the transition arena. Such a selection ing to the system approach. Facilitators synthesize
is of vital importance. Participants need to have discussions and work toward convergence of per-
some basic competencies at their disposal: They spectives, assumptions, and ambitions. The tran-
need to be visionaries and frontrunners, and they sition arena develops a shared understanding of
must have the ability to look beyond their own the persistence of a problem at the level of a soci-
domain or working area and be open-minded. etal system, the necessity of a transition or radical
change, and the definition of the challenge this
poses. Key outcomes are a new, shared perspec-
Problem Structuring and Envisioning:
tive; language to discuss the transition; and the
Establishment of a Transition Arena
definition of a set of guiding principles for the
The transition arena is best viewed as a vir- envisaged transition. This relates to the earlier
tual network, which is a legitimate experimen- mentioned notion of emergence (De Haan 2006):
tal space in which the actors involved use social The awareness of and insight into the complexity
learning processes to acquire new knowledge and of their environment helps individuals to better

192 Journal of Industrial Ecology


understand the complexity and realize that they are either related to or combined with existing
can, on a small scale, exert influence. activities. Transition experiments are high-risk
experiments with a social learning objective that
are supposed to contribute to the sustainability
Development of Sustainability Images,
goals at the systems level and should fit within
Pathways, and a Transition Agenda
the transition pathways. It is important to for-
Transition images are the “translation” of the mulate sound criteria for the selection of exper-
generic guiding principles or “sustainability cri- iments and to make the experiments mutually
teria” to specific concrete settings, subsectors, coherent. The crucial point is to measure to what
or themes. These images must be appealing and extent the experiments and projects contribute
imaginative so as to be supported by a broad range to the overall system sustainability goals and to
of actors and inspire and guide short-term action. measure in what way a particular experiment re-
Inspiring images are useful for mobilizing social inforces another experiment. Are there specific
actors and represent a consensus among different niches for experiments that can be identified?
actors on what sustainability means for a specific What is the attitude of the current regime toward
transition theme, which could evolve over time as these niche experiments? The aim is to create a
new insights emerge. Transition images embrace portfolio of transition experiments that reinforce
multiple transition pathways to represent a vari- each other and contribute to the sustainability
ety of possible options. They include transition objectives in significant and measurable ways.
goals, which are qualitative rather than quan- Around and between these experiments, all sorts
titative and are multidimensional, representing of actors can be involved that will not engage
the three dimensions of sustainability: economic, regularly in debates about long-term issues: small
ecological, and sociocultural. businesses, consumers, citizens, local groups, and
Various transition pathways lead to a partic- so on. Here, as well, the emphasis is on involving
ular transition image (i.e., a sustainability vision frontrunners.
comprises various transition images), and from
various transition images a particular transition
Monitoring and Evaluating the Transition
pathway may be derived. The transition images
can be adjusted as a result of what has been
learned by the players in transition experiments. Continuous monitoring is a vital part of the
The transition process is thus a goal-seeking pro- search and learning process of transitions. We
cess, in which the transition visions and images, distinguish between monitoring the transition
as well as the underlying goals, change over time. process itself and monitoring transition manage-
During the course of the transition process, the ment. Monitoring the transition process involves
actors will choose the visions and images that attending to physical changes in the system in
appear to them as the most innovative, promis- question, slowly changing macro-developments,
ing, and feasible. The transition agenda contains fast niche developments, seeds of change, and
content objectives, process objectives, and learn- movements of individual and collective actors at
ing objectives. Although the transition visions, the regime level. Monitoring of transition man-
images, and objectives form the guidelines for agement involves different aspects. First, the ac-
the transition agenda, the transition agenda it- tors within the transition arena must be moni-
self is the compass for the frontrunners, to which tored with regard to their behavior, networking
they can refer during their search and learning activities, alliance forming, and responsibilities
process. and also with regard to their activities, projects,
and instruments. Next, the transition agenda
must be monitored with regard to the actions,
Initiation and Execution of Transition
goals, projects, and instruments that have been
Experiments and Mobilization of Actors
agreed on. Transition experiments need to be
From the transition vision, images, and path- monitored with regard to specific new knowl-
ways, transition experiments can be derived that edge and insight and how these are transferred

Rotmans and Loorbach, Complexity and Transition Management 193


but also with regard to the aspects of social and dynamics, which will change when these princi-
institutional learning. Finally, the transition pro- ples are applied. On the basis of this approach,
cess itself must be monitored with regard to the the management framework itself has been the
rate of progress, the barriers, and the points to be result of experiences within testing grounds and
improved, for example. Evaluating these moni- has evolved in the past several years. The concept
toring aspects within each phase may stimulate a of transition management and the derived frame-
process of social learning that arises from the in- work is promising but still needs to largely prove
teraction and cooperation between different ac- itself empirically. It is a great challenge to empir-
tors involved. ically validate the partly descriptive and partly
In each of the above activity clusters, coali- prescriptive parts of transition management in
tion and network formation are of vital impor- such a manner that the framework can be further
tance, combined with the systemic structuring developed and used in a broad international so-
and synthesizing of discussions. The transition cietal context. One of its potential contributions
arena is meant to stimulate the formation of new lies in application to nonenvironmental domains,
coalitions, partnerships, and networks that to- such as health care and city restructuring. In
gether create a new way of thinking. Mostly, this sense, transition management can be consid-
coalitions emerge around transition pathways or ered as an extension of and a step beyond indus-
experiments or around specific subthemes, where trial ecology into broad societal (socioeconomic)
subarenas arise. The very idea behind transition systems.
management is to create a societal movement
through new coalitions, partnerships, and net- References
works around arenas that allow for building up
Allenby, B. 1999. Industrial ecology: Policy framework
continuous pressure on the political and market
and implementation. Englewood Cliffs, NJ: Pren-
arena to safeguard the long-term orientation and tice Hall.
goals of the transition process. Arthur, W. B., S. N. Durlauf, and D. A. Lane. 1997. The
economy as an evolving complex system. Reading,
MA: Addison-Wesley.
Avelino, F. and J. Rotmans. 2007. Power in transition.
In this article, we have presented a transition Submitted for publication to European Journal of
management framework for addressing persistent Social Theory.
societal problems that is grounded in complex Ayres, R. U. and L. Ayres. 1996. Industrial ecology—
systems theory. Variation and selection, emer- towards closing the materials cycle. Cheltenham,
UK: Edward Elgar.
gence, coevolution, attractors, diversity and co-
Braungart, M. and W. McDonough. 2002. Cradle to
herence, and interactions and feedbacks are key
cradle: Remaking the way we make things. New York:
elements of transition management. The under- North Point Press.
lying premise is that a better understanding of the De Haan J. 2006. How emergence arises. Ecological
dynamics of complex, adaptive systems provides Complexity 3(4): 293–301
insight into the opportunities, limitations, and De Haan, J. and J. Rotmans. 2008. Patterns in tran-
conditions under which it is possible to influence sitions. Submitted for publication to Research
such systems. This implies a strong linkage of con- Policy.
tent and process: The combination of analytic in- De Vries, J. and H. Te Riele. 2006. Playing with hyenas:
sights into systems complexity and understanding Renovating environmental product policy strat-
of the process of governance complexity is new egy. Journal of Industrial Ecology 10(3): 111–127.
Dirven, J., J. Rotmans, and A. P. Verkaik. 2002. Samen-
and has resulted in a set of management principles
leving in transitie: Een vernieuwend gezichtspunt [So-
that forms the basis for the management frame-
ciety in transition: A new viewpoint]. The Hague,
work. The management principles are reflexive the Netherlands: Innovatienetwerk Agrocluster
rather than deterministic, reflecting a belief that en Groene Ruimte.
transitions toward sustainability can be directed Ehrenfeld, J. 1997. Industrial ecology: A framework
to a limited degree. Applying these principles im- for product and process design. Journal of Cleaner
plies adjusting them to the new conditions and Production 5(1–2): 87–96.

194 Journal of Industrial Ecology


Ehrenfeld, J. and N. Gertler. 1997. Industrial ecology Kemp, R., D. Loorbach, and J. Rotmans. 2007. Tran-
in practice: The evolution of interdependence at sition management as a model for managing
Kalundborg. Journal of Industrial Ecology 1(1): 67– processes of co-evolution towards sustainable de-
80. velopment. International Journal of Sustainable De-
Geels, F. W. 2005. Technological transitions and system velopment and World Ecology (special issue on co-
innovations: A co-evolutionary and socio-technical evolution) 14: 1–15.
analysis. Cheltenham: Edward Elgar. Kickert, W. J. M. 1991. Complexiteit, zelfsturing en dy-
Geldof, G. 2002. Omgaan met complexiteit bij inte- namiek. Over management van complexe netwerken
graal waterbeheer [Dealing with complexity in bij de overheid [Complexity, self-steering and dy-
integrated water management]. Twente, the namics. On management of complex networks by
Netherlands: Universiteit Twente. government]. Rotterdam, the Netherlands: Eras-
Goldstein, J. 1999. Emergence as a construct: History mus Universiteit.
and issues. Emergence 1(1): 49–72. Kooiman, J. 1993. Modern governance: New government-
Green, D. G. 1994. Evolution in complex systems. In society interactions. London: Sage.
Complex systems: Mechanism of adaptation, edited Korhonen, J. 2004. Industrial ecology in the strategic
by R. J. Stonier and Ju Xing Huo. Oxford, UK: sustainable development model: Strategic appli-
IOS Press. cations of industrial ecology. Journal of Cleaner
Green, K. and S. Randles. 2006. Industrial ecology and Production 12(8–10): 809–823.
spaces of innovation. Cheltenham, UK: Edward Loorbach, D. 2007. Transition management: New mode
Elgar. of governance for sustainable development. Utrecht,
Gunderson, L. H. and C. S. Holling. 2002. Understand- the Netherlands: International Books.
ing transformations in human and natural systems. Loorbach, D. and J. Rotmans. 2006. Managing
Washington, DC: Island Press. transitions for sustainable development. In In-
Holland, J. H. 1995. Hidden order: How adaptation dustrial transformation—disciplinary approaches to-
builds complexity. Ulam Lectures Series. Cam- wards transformation research, edited by A. J.
bridge, MA: Helix Books/Perseus Books. Wieczorek and X. Olshoorn. Dordrecht, the
Holling, C. S., ed. 1978. Adaptive environmental assess- Netherlands: Kluwer Academic Publishers.
ment and management. London: Wiley. Ministry of Housing, Spatial Planning and Environ-
Holling, C. S. 1987. Simplifying the complex: The ment. 2002. A world and a will: Working on sus-
paradigms of ecological function and structure. tainability. Fourth national environmental plan. The
European Journal of Operational Research 30: 139– Hague: Ministry of Environmental Affairs.
146. Mitleton-Kelly, E. 2003. Ten principles of complex-
Hordijk, L. 1985. A model for evaluation of acid de- ity and enabling infrastructures. In Complex sys-
position. In Systems analysis and simulation 1985, tems and evolutionary perspectives of organizations:
volume 2, edited by A. Sydow, M. Thoma, and R. The application of complexity theory to organizations,
Vichnevetsky. Oxford, UK: Pergamon Press. edited by E. Mitleton-Kelly. London: Elsevier.
Kauffman, S. 1993. The origins of order. Oxford, UK: Prigogine, I. and I. Stengers. 1984. Order out of chaos:
Oxford University Press. Man’s new dialogue with nature. Boulder, CO: New
Kauffman, S. 1995. At home in the universe: The search Science Library.
for laws of complexity. Oxford, UK: Oxford Uni- Rittel, H. and M. Webber. 1973. Dilemmas in gen-
versity Press. eral theory of planning. Policy Sciences 4(2): 155–
Kay, J. 2002. On complexity theory, exergy and indus- 159.
trial ecology. In Construction ecology: Nature as the Rotmans, J. 1990. IMAGE: An integrated model to assess
basis for green buildings, edited by C. J. Kibert et al. the greenhouse effect. Dordrecht, the Netherlands:
New York: Spon Press. Kluwer Academic Publishers.
Kay, J., H. Regier, M. Boyle, and G. Francis. 1999. An Rotmans, J. 2003. Transitiemanagement: Sleutel voor een
ecosystem approach for sustainability: Addressing duurzame samenleving [Transition management:
the challenge of complexity. Futures 31(7): 721– Key for a sustainable society]. Assen, the Nether-
742. lands: Koninklijke Van Gorcum.
Kemp, R, J. Schot, and R. Hoogma. 1998. Regime shifts Rotmans, J. 2006. Societal innovation: Between dream
to sustainability through processes of niche for- and reality lies complexity. Rotterdam, the Nether-
mation: The approach of strategic niche manage- lands: RSM Erasmus University.
ment. Technology Analysis and Strategic Manage- Rotmans, J. and R. Kemp. 2008. Detour ahead: A re-
ment 10: 175–196. sponse to Shove and Walker about the perilous

Rotmans and Loorbach, Complexity and Transition Management 195


road of transition management. Environment and economic behaviour and institutions: Towards a
Planning A 40(4): 1006–1011. theory of institutional change. Journal of Evolu-
Rotmans, J., R. Kemp, and M. van Asselt. 2001. More tionary Economics 13: 289–317.
evolution than revolution: Transition manage- Van Der Brugge, R. and R. Van Raak. 2007. Facing the
ment in public policy. Foresight 3(1): 15–31. adaptive management challenge: Insights from
Rotmans, J., D. Loorbach, and R. Van Der Brugge. transition management. Ecology and Society 12(2):
2005. Transitiemanagement en duurzame ontwikke- 33.
ling: Co-evolutionaire sturing in het licht van com- Accessed 21 March 2009.
plexiteit. [Transition management and sustainable Von Bertalanffy, L. 1968. General system theory: Foun-
development: Co-evolutionary steering in light of dation, development and applications. New York:
complexity]. Beleidswetenschap Juni. The Hague: Braziller.
Lemma. VROM (Energy Council of the Ministry of Hous-
Sawyer, R. K. 2005. Social emergence: Societies as complex ing, Spatial Planning and Environment) and
systems. Cambridge, UK: Cambridge University VROM Council. 2004. Energy transition: Climate
Press. for new chances. The Hague, the Netherlands:
Senge, P. M. 1990. The fifth discipline: The art & prac- VROM.
tice of the learning organization. London: Random
Social Economic Council. 2001. Ontwerpadvies Na- About the Authors
tionaal Milieubeleidsplan 4 [Recommendations Na- Jan Rotmans is professor in transition man-
tional Environmentl Policy Plan 4]. The Hague,
agement at the Erasmus University Rotterdam,
the Netherlands: SER.
Rotterdam, the Netherlands, and scientific di-
Spiegelman, J. 2003. Beyond the food web: Connec-
tions to a deeper industrial ecology. Journal of In- rector of the Dutch Research Institute for Tran-
dustrial Ecology 7(1): 17–23. sitions (Drift) at the same university. Derk Loor-
Stacey, R. D. 1996. Strategic management and organisa- bach is a senior researcher at Drift and received
tional dynamics. London: Pitman. his doctorate in transition management in
Van Den Bergh, J. and S. Stagl. 2004. Coevolution of 2007.

196 Journal of Industrial Ecology