Presented at the International Society for Ecological Economics Conference:
Designing Sustainability, August 4-7 1996

Beth Dempster, School of Urban and Regional Planning, University of Waterloo, Canada

The importance of understanding complex self-organizing systems, for designing sustainability, is only realized by the few who recognize their existence. This recognition, however, seldom extends to the practitioners, from bankers to foresters, who make the decisions that are relevant for actually achieving sustainability.

My objective in this brief paper is to illustrate a parallel between three conceptual maps regarding scientific approaches, assessment techniques, and system types. The key to what I offer is the latter, and in particular, the characterization of two types of self-organizing systems; autopoietic and synpoietic systems. Their description provides a perspective for practitioners to identify the systems they deal with. In turn, this can encourage recognition of the significantly different behaviours of the systems and the subsequent need for choosing relevant approaches.

Table 1 compares the characteristics of the two system types. Autopoietic systems were originally characterized in the 1970s by Maturana and Varela (1980). The defining characteristics of these systems are self-production and self-defined boundaries. The concept emphasizes the autonomy and unity of these systems without suggesting that they are independent of their environment. Though Maturana and Varela used the cell as the primary example, others, including myself, argue that organisms are also autopoietic systems.

In contrast, I characterize synpoietic systems - using the Greek for 'together' 'producing' (Dempster 1995). These systems are also self-producing, but are unbounded. They do not have completely encircling, self-defined boundaries. The concept emphasizes the cooperative, synergistic behaviour of these systems, suggesting that recognition is possible despite their complex amorphous nature. Primary examples are ecosystems and cultural systems.

The self-organizing potential of the two system types is quite different. In autopoietic systems, it is passed down through generations, which I have termed 'transmitted' self-organization. In contrast, the self-organizing potential of synpoietic systems is creative. By this, I refer to the self-organizing process identified by many, from Prigogine's description of dissipative structures to the emergent behaviour of complex systems recognized through computer simulations.

The resultant behaviour of the two system types is also very different. Autopoietic systems are primarily driven by negative feedback and are fundamentally equilibrium centered (in the homeostatic sense). They have finite trajectories that are development oriented. The trajectories are relatively predictable, based on observation of ancestral autopoietic systems.

Synpoietic systems are characterized by an interactive balance between positive and negative feedback. Rather than pushing systems beyond their ability to adapt, the self-reinforcing nature of positive feedback is a constitutive element of these systems. Synpoietic systems have potentially infinite trajectories, with the potential for dramatic and surprising changes. They are essentially evolutionary and are inherently unpredictable.

By plotting these system types against system domains, I create the diagram illustrated in Figure 1a. Note that the x-axis has been extended to include externally organized and other self-organized systems to cover the full range of possibilities. By domain, I refer to the space in which a system's components and relations are defined such as physical, biological, social and cultural domains. Note that each axis represents a gradation - there are not clear lines of demarcation between any of these categories. It is possible for practitioners to locate any system of concern within the region outlined by the diagram. Consideration of scale is critical. Since the diagram is offered as a heuristic, the system types must be understood as perspectives. Some systems may usefully be considered in different positions.

As an example, I illustrate the systems of concern for the management of Point Pelee National Park on the north shore of Lake Erie (Figure 1b). These include geomorphic processes, and hydrologic cycles in the physical, external organizing zone; terrestrial and aquatic ecosystems in the biological autopoietic-synpoietic zone; and social concerns such as the economic health of the region and the value systems held by the various actors in the social - cultural, synpoietic zone.

It is critical, however, to also recognize the links between domains. As humans, we link the biophysical and our cultural domains. Beliefs and values, formed in the cultural domain, become manifest, through our actions, in the biophysical domain. This clarification makes it essential to re-evaluate the placement of these systems on the diagram, viewing them instead as a single integrated synpoietic system (Figure 1c). Our socio-cultural systems have an influence on all other systems.

I suggest, that for achieving sustainability, all systems of concern must be perceived as located in the upper right region (or at least right hand side) of the diagram. Our tendency, however, has been to treat them as if they belonged in the lower left.

I now draw a parallel between the conceptual map of system type and other two conceptual maps. The three diagrams share the characteristic of increasing complexity, uncertainty and unpredictability moving from the core outwards.

The diagram of scientific approaches (Funtowicz and Ravetz 1993), illustrated in Figure 2, places increasing decision stakes to parallel domains on the y-axis. On the x-axis, system uncertainties parallel system type. Three zones are identified; applied science, consultancy and post-normal science.

The third diagram, Figure 3, represents different types of assessments (Nelson and Serafin 1993). On the y-axis, paralleling domains and decision stakes, is level of involvement. On the x-axis, paralleling system type and system uncertainties, is scope of interest. The three zones in this diagram include: the traditional analytical technique which emphasizes quantitative information, a reductionist approach and assumes predictable outcomes; the interpretive technique, which recognizes a plurality of views and involves public consultation; and the adaptive technique which encourages learning, understanding change, and shared decision making.

Due to the parallels between these concepts, I suggest that it is essential to use scientific and assessment approaches from the zones that correspond to the system type. By typing a system of concern, a practitioner may use the diagram as a guide for choosing relevant approaches.

I return to the example of Point Pelee to illustrate this potential. Figure 4 is an overlay of the three conceptual maps, including the systems of concern for Point Pelee National Park. I argue that for adequate treatment of the systems identified, it is necessary to use post-normal science and adaptive assessments.

It is interesting to note that changes in the management plans for Point Pelee National Park over the past few decades do reflect a movement outwards.

I stated earlier that I believe all systems of concern regarding sustainability must be perceived as synpoietic, particularly synpoietic-cultural systems, located in the right-hand region of the diagram. I conclude that for designing sustainability it is essential to use post-normal science and adaptive assessments. The traditional scientific approaches and analytical assessment techniques, though useful for systems located in the lower, left-hand regions, are inadequate for dealing with the complex, uncertain, and unpredictable systems we must deal with to achieve sustainability.

I gratefully acknowledge support from the Social Sciences and Humanities Research Council of Canada and Parks Canada, Ontario Region.