System theoretic limits on the cybernetic paradigm

Abstract

The cornerstone of modern cybernetic theory is still the first-order, negative feedback scheme as when it began. Though there have been attempts to move toward a n-order formulation (Forrester 1968) and some attempts to define the possibilities of positive feedpack systems (Maruyama 1963), the original formulation is still the most widely referenced and employed. It is so widely employed that its effects are felt in fields as diverse as automata theory, computer engineering, organization theory, and behaviorist psychology, and, of course, it has become the axis of much work in the management sciences, particularly that associated with system control. Indeed, the cybernetic scheme has now attained the status of a proper paradigm, which means that its first premises are often simply accepted rather than investigated. Indeed, some enthusiasts feel the cybernetic paradigm largely exhausts the substance of system theory. However, when we analyze the fundamental attributes of the cybernetic paradigm from a system theoretic perspective, we find some limits on both its scientific and operational utility, limits which have received only the most casual attention. It is these limits which this paper will explore.