Complexity Theory, Nonlinear Dynamic Spatial Systems

Abstract

This article provides an overview of the origins of complexity science, a summary of its ontological and epistemological implications, and a discussion of those implications for human geography. It is argued that the currently dominant North American perspectives on complexity science have their origins in post-World War II optimism concerning the applicability of computational approaches to systems of organized complexity. Systems of organized complexity are considered to be intermediate between simple systems with few interactions between elements, and large systems with many millions of elements whose behavior is adequately described in statistical terms. Such systems exhibit a number of expected structural and behavioral characteristics. Aspects of structure include distinctive hierarchical or networked patterns of organization. Behavioral characteristics of interest may include emergence, self-organization, chaotic dynamics, positive feedback, path dependence, and tipping points, and each of these is considered in turn. The important role of computational modeling in the study of complex systems is discussed with particular attention paid to the importance of the unsolved problem of validating the representational adequacy of any computational model. Finally, the relatively limited impact to date of complexity science in human geography is discussed, and prospects for future engagement are considered.