Complexity science and regional development: Toward a computational regional science.

摘要

Regional Analysis is the quintessential spatial field. It is also quintessentially inter-disciplinary as regional dynamics are shaped by every dimension along the spectrum of human activity. As the density and complexity of human settlements has increased over time, the tacit linear mean field assumption undergirding the primary technique of regional analysis, linear regression, becomes increasingly suspect. It is not even clear that all the fields in effect across a human community are linearly additive. Significant progress in regional analysis will require a deepening of our understanding of social spaces, their topologies, and the dynamics of interaction among associated fields. It may be necessary to develop and promulgate a new mathematics to understand and speak to the combined effects of economics, social conditions, and political activity over a given space.;The current direction of the evolution of Big Data is not helpful in this regard because the current practice is to aggregate data from a variety of sources and to explore this data statistically for what it might reveal. Regions are highly complex systems; analysis of their complexity can only be served by use of data generated by the systems themselves. And the information sought is generally not the summary data provided by statistics, but pictures of the evolution of the state of the system over time in response to different stimuli or different initial conditions. Public policy has for centuries been proceeding as though we live in the land of the probable. Policy makers have been content to model human systems as though they are driven by the behavior of the average. This average behavior has been relentlessly mapped and poked and prodded in an effort to reveal the essential nature of the system so that intervention strategies could be designed that would enable us to correct anomalies or move the system in a desired direction. It is becoming increasingly clear that with the increasing density, number, and diversity of interactions in an environment of increasingly relaxed constraints, that our world is much more the land of the possible than the probable, and that our quest to understand and manage that world has much more to do with mapping the number, nature, and interactions of its infinite possibilities.;Techniques associated with the Analysis of Complexity are powerful tools to use in support of this quest. Another important tool in the quest to deepen our understanding of movement among social space is spatial econometrics. The physical sciences know well how to isolate and combine the forces due to diverse fields (gravitational, mechanical, electrical, . . . ). Social sciences (political, economic, cultural, . . . ) do not. In this age of computational analysis, the defining feature of the overlap between and among fields is not so much the subject matter covered as it is the mathematics employed. The path forward for regional economic development, then, is not so much through physics or biology as it is through spatial econometrics. That is not to say that the experiences of physics and biology are irrelevant to the emergence of regionalism as a computational field. The problem with using them as models is that they have learned to combine spaces in ways that regionalists have not. Spatial econometrics offers a mathematical platform that will enable us to visualize and work with combined spaces naturally without having to obliterate them through the tyranny of averaging. The developmental direction of regional analysis and the cutting edge of the evolution of computational regional science lies in increasing the sophistication of that field's analytic treatment of space. Complexity theory is explained from a perspective especially useful to the social science with special emphasis on definitions and the precise application of those definitions to social phenomena. Theories and modeling techniques offered by complexity theory are used to shape an alternative framework for regional analysis.;The purpose of this research is to contribute to the understanding of regions as dynamic, interactive, complex, adaptive systems, and to the precise migration of the tools of complexity analysis to the field of regional development. The expectation is that such migration will permit more precise problem identification, more creative scenario planning, and the development and implementation of more effective and innovative intervention strategies.