A New Systems Thinking: Implications of the Sciences of Complexity for Public Policy and Administration

Article excerpt


This paper discusses the implications of the sciences of complexity for public policy and administration. It is argued that the sciences of complexity have implications for our thinking in mainly three areas. First, they revise our conceptions of systems, causal relations, and determinism and depict a picture of mostly indeterministic reality composed of open systems. Second, they offer an "endophysical" and phenomenological view of system - observer relations. Third, although they are heavily quantitative, they illustrate the importance of qualitative interpretations in quantitative analyses and thus bridge the chasm between quantitative and qualitative methodologies. The insights of the sciences of complexity can help us improve our understanding of the complexities of public policy and administrative processes.


The implications and significance of the sciences of complexity (1) have been matters of controversy. There are those who think that complexity theory offers only an improvement in scientific understanding, but does not revolutionize science or offer a new worldview (Eve, 1997; Wilson, 1998). Others (e.g., Kiel, 1992; Dent, 1999) see complexity theory as a new paradigm, an alternative to the old Newtonian paradigm. In Kiel's view, the new "nonlinear paradigm" has important implications for public policy. He suggests that the new paradigm both shows the limits of the mathematics of certainty (e.g., in forecasting) and enhances the power of analyses by revealing the patterns and degrees of stability in systems' temporal behaviour. It also shows the inadequacies of the prevailing theoretical approaches in public policy studies-the social engineering approach, which is based on a Newtonian notion of prediction and control, and the laissez faire model, which is based on the notion of an equilibrating invisible hand.

I agree with Kiel's assessments. In this paper I will give an overview of the epistemological and methodological alternatives complexity theory offers. I want to show that the sciences of complexity have implications for our thinking mainly in three areas: (1) characteristics of systems, particularly causal relations and determinism; (2) system observer relations, and (3) quantitative and qualitative forms of inquiry. As I will illustrate below, some of the concepts of complexity theory have equivalents in the social science theories and they resonate well with the epistemological concerns of some social scientists and public policy theorists.

The sciences of complexity are grounded in systems thinking, which is a theoretical framework used in both the natural and social sciences. Public administration and policy theorists use systems concepts too. We think of organizations and public policies as closed or open systems, for example. Complexity theory changes the traditional understanding of systems in important ways. Most significantly, it blurs the distinction between "simple" and "complex systems" (Cilliers, 1998: 2). First, it suggests that systems are dynamic and that during the phase transitions that systems undergo simple turns into complex and vice-versa. The deterministic nonlinear and indeterministic relations between the elements of systems generate these dynamic transformations. Second, complexity theorists argue that the simplicity and complexity of a system and what constitutes a system depend partly on an observer's definitions. Thus they pose a major challenge to empiricist epistemologies and proffer a phenomenological perspective for scientific knowledge.

In the next section I will discuss the systemic transformations described in different versions of complexity theory. In the following section I will address the phenomenological perspective it offers. I will discuss its implications for quantitative and qualitative forms of inquiry in a separate section.


Determinism and Indeterminism

Newtonian science is based on the belief that the universe is completely deterministic. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.