The application of Social-Ecological Systems (SES) thinking to evaluation


Aaron Zazueta    

Among the many approaches to complex systems thinking, the Social-Ecological Systems (SES) proponents have developed a set of concepts to understand and model the interlinked dynamics of social and environmental change. When addressing the transformation of large complex systems, I find particularly useful the SES  concepts of the Social-Ecological Systems, boundaries, domains, scales, agents, adaptive behavior, and emergence and system development trajectory.  I have used these concepts to construct theories of change (TOCs), to help me understand how and what extent projects, programs, or policies interact with social-ecological systems to steer development processes in the direction of a given trajectory or policy goals. The notion that all systems are composed of subsystems that are interconnected helps focus the attention on the phenomena relevant to the desired policy goals. The concepts of domains help to identify further the critical conditions that can enable or hamper change in the direction of a given development trajectory (such conditions can include the presence of sound governance, the availability of knowledge and technology to solve a problem or the necessary institutional capacities). As domains cut across the whole SES, the concept of a domain is helpful to trace system interactions across different scales of space and time.

The agents and their adaptive behavior underlie the phenomena that encompass the system and its components. SES scholars assume that systems operate through the actions and reactions of the agents (the agents’ adaptive behavior). While agents command different resources and are influenced differently by the conditions in the various domains, they are linked, either directly or through other agents). Even relatively minor agents under the right conditions can generate reverberating behaviors across the system. The aggregated adaptive behavior of the agents responding to other agents and other factors external to the system result in the emergence of system-level shapes that can be quite different from the behaviors of the agents. Based on these concepts, it is possible to develop a model of the conditions that are likely to steer agent behavior direction and the extent to which an intervention is has contributed or is likely enable adaptive behavior consistent with the desired long term policy goals.

SES also assumes that the adaptive behavior of the agents also contributes to various degrees of unpredictability and non-linearity. It is thus important not to expect that in complex systems, outputs or results will correspond to inputs. Therefore, when dealing with complex systems, effective development interventions are those which mimic other agents in the system and adopt adaptive management as the approach to steer the development trajectory of the system. Adaptive management entails clear long-term goals (or trajectory direction), identification of alternative management objectives,  the development of a set of the hypotheses of causation and procedures for data collection to adjust hypotheses (ongoing evaluation).