Thinking Systems

February 06th, 2010 | Category: Essays, Reviews

In his epic, futurist tome The Third Wave, Alvin Toffler (1980) wrote that we need to “move from a Second Wave culture that [has] emphasized the study of things in isolation from one another to a Third Wave culture that emphasizes contexts, relationships, and wholes” (p. 300-301), what Herman Witkin calls “field dependence.” Taking the long view, considering the context, and how one thing influences another — these are all things we would do well to do at all times. General system theory as conceived by Ludwig von Bertalanffy provides a rich framework for just this type of thinking.

“A system,” writes Bertalanffy’s biographer Mark Davidson, “like a work of art, is a pattern rather than a pile. Like a piece of music, it’s an arrangement rather than an aggregate” (1983, p. 27). In other words, a system is an assemblage that is arranged to serve a purpose. Whereas Camus insisted that there were no ends, only means, Bertalanffy saw them as one and the same. The system is its own means and its own end.

Framing things as systems inherently simplifies them. Sometimes this is done by leaving certain aspects out, sometimes by artificially drawing boundaries around a “whole.” As Manuel De Landa puts it, this

point of view allows for the emergence of wholes that are more than the sum of their parts, but only if specific historical processes — specific interactions between ‘lower scale entities’ — can be shown to have produced such wholes. Thus, in my view, institutional organizations like bureaucracies, banks, and stock markets acquire a life of their own from the interactions of individuals. From the interactions of those institutions, cities emerge, and from the interactions between cities, nation states emerge. Yet, in these bottom-up approaches, all the heterogeneity of real nation states can be pockets of minorities, the dialect differences, the local transience — unlike when history is modeled on totalities (concepts like ‘society’ or ‘culture’ or ‘the system’). In this latter situation, homogeneity has to be artificially injected into the model (quoted in Miller, 2007, p. 71-72).

The main criticism of systems theory is its quasi-functionalist embrace of the needs of the system over those of the humans involved. Where one view seems to favor the system over all else (cf. Mary Douglas and Aaron Wildavasky’s Risk and Culture; 1982), Donella Meadows’ Thinking in Systems (2008) shows how framing risks and problems in their context can help us understand and even control them better. Acknowledging the artificial nature of “closing off” systems for study, Meadows (2008) wrote, “The right boundary for thinking about a problem rarely coincides with the boundary of an academic discipline, or with a political boundary” (p. 98).

De Landa (1997) wraps it up eloquently, writing, “…[M]uch as sedimentary rocks, biological species, and social hierarchies are all stratified systems, so igneous rocks, ecosystems, and markets are self-consistent aggregates, the result of the coming together and interlocking of heterogeneous elements” (p. 66). We are — and we live in — a system of interacting systems. The better we understand them as such, the better off we will be.


Bertalanffy, L. v. (1968). General system theory. New York: George Braziller, Inc.

Davidson, M. (1983). Uncommon sense: The life and thought of Ludwig von Bertalanffy. Los Angeles: Tarcher.

De Landa, M. (1997). A thousand years of nonlinear history. New York: Zone Books.

Douglas, M. & Wildavasky, A. (1982). Risk and culture. Los Angeles: University of Southern California.

Meadows, D. H. (2008). Thinking in systems: A primer. White River Junction, VT: Chelsea Green.

Miller, P. D. (2007). ILLogical Progression. In R. Christopher (Ed.), Follow for now: Interviews with friends and heroes (pp.). Seattle, WA: Well-Red Bear.

Toffler, A. (1980). The third wave. New York: Bantam.

Witkin, H. A. & Goodenough, D. R. (1981). Cognitive styles: Essence and origins. New York: International Universities Press.

Further Posting: