The Complex Adaptive Systems Approach. a Sociocybernetic Reading

Since the foundation of the Santa Fe Institute in the United States in 1984, and especially during the 90s, was formed a new wave in the science of complexity: the Complex Adaptive Systems (CAS) approach. CAS is based on modeling and simulation of complex systems using advanced computational techniques such as cellular automata or agent-based modeling. With an interdisciplinary and universal claim, its mainly methodological approach has dominated the scientific discourse around complexity. Numerous interdisciplinary complex adaptive systems centers have spread throughout the world, from United States to Europe, from Latin America to East Asia.

We need a more reflexive and critical approach to CAS, taking into account its epistemological limits. Sociocybernetics, through authors such as von Foerster, Varela, Maturana, Luhmann or Morin, is very useful here, because of its emphasis on the question of the observer/subject and its construction of the observed/object. Something that CAS models ignore, seeking to determine objectively and algorithmically probabilistic patterns in complex systems, without considering that system and complexity are primarily attributes of an observer, an operation of “punctuation” in his knowledge of the world. The simulated systems are doubly result of the intervention of the observer: the real system is an "invented reality" (von Foerster) and the system that simulates the real system would be doubly so.

CAS epistemic limits become more evident in its sociological models called "Artificial Societies". The intervention of the modeler/observer here is even more evident in self-reflexive systems that are able to observe themselves, "self-referential" (Luhmann) systems of meaning in which both the constructed nature of the subjects themselves (Foucault) and the structures or social spaces of objective positions where the subjects are located (Bourdieu) cannot be ignored in favor of a mere behaviorist study of patterns of interaction among agents and their resulting and irreducible emergences.