Escaping the complexity dilemma

Abstract

The characteristic behaviour of a social-ecological system (SES) emerges from feedback interactions between its parts. This means that its response to human activities cannot be predicted on the basis of studies of its individual parts taken separately. As expressed by Ackoff (1986), “A system is more than the sum of its parts; it is the product of their interactions. If taken apart, it simply disappears”. But anyone who tries to look at an SES as a whole will be overwhelmed by its complexity – such a system has too many parts, interacting in too many different ways, at too many different scales, for it to be understood as a whole. This is the ‘complexity dilemma’. On the one hand, practical approaches to policy development require the identification of relatively simple, understandable sub-systems. On the other hand, the reductive process of isolating a sub-system draws a boundary around a limited set of state variables and breaks causal links with key variables in other sub-systems. This process runs the risk of leading to unsustainable policies that sooner or later are besieged by unexpected outcomes – often outcomes that the new policies themselves have triggered.