11th FIS Discussion Session:


Joseph A. Tainter

Global Institute of Sustainability and School of Human Evolution and Social Change,
Arizona State University, Tempe, Arizona, USA

& & & & & & & & &

Several years ago I began a fruitful collaboration with Timothy Allen, an ecologist at the University of Wisconsin and a leading thinker on hierarchy theory. Many stimulating discussions and several worthwhile products emerged from our work together, as well as a few surprises. One surprise was to find, after writing together for five years, that we meant different things by the term complexity. I suspect this kind of discovery is not unusual in interdisciplinary collaborations.

How many kinds of complexity are there? Perhaps I should phrase that as: How many concepts of complexity are there? There are, to begin, social complexity and ecological complexity. There are algorithmic complexity and computational complexity. A simple foray into Wikipedia shows that computational complexity can be linear, logarithmic, or exponential. Complexity classes can be P or NP. There are complex systems, complex mechanisms, and complex behavior. Complexity can be specified, irreducible, or unruly. To some ecologists, complexity is equivalent to diversity, while to others complexity emerges through hierarchy (Allen and Starr 1982; Allen and Hoekstra 1992; Ahl and Allen 1996). Complexity can be hierarchical or heterarchical. Hierarchies can be simple and short, or complex and elaborate. Complexity can occur within a system, or by embedding types of systems. Organization in complex systems emerges through constraints, which can be asymmetrical (Allen, Tainter, and Hoekstra 1999; Allen et al. 2001) or symmetrical. Amidst all this conceptual fluorescence, the irony of complexity is that it simplifies. Elaboration of structure and organization simplifies behavior, making it possible for assemblages of entities to function as systems.

I will cut through the conceptual diversity by describing how many anthropologists, including myself, came to think of complexity. Two of history’s truisms are that most societies of today are more complex than those of our ancestors, and that at least since the end of the Pleistocene many societies have shown a seemingly inexorable tendency to increase in complexity. But of what does this complexity consist?

One essential feature of the least complex societies is that they show little differentiation in structure. There are few social roles except those arising immediately from gender, age, and personal abilities. Social and cultural complexity has meant, in part, the emergence of differentiated roles and institutions. Hunter-gatherer societies may contain no more than a few dozen distinct social personalities, while modern European censuses recognize 10,000 to 20,000 unique occupations, and industrial societies may contain overall more than 1,000,000 different types of personalities (McGuire 1983: 115).

Is this differentiation of structure all there is to social complexity? To conclude so would be like concluding that ecological complexity can be reduced to species diversity. Differentiation in structure is a useful starting point for understanding cultural complexity, but it is only the beginning. In western North America, anthropologists in the early 20th century compiled lists of “culture elements” among the remaining native peoples. It was not a very sophisticated approach. A culture element could be anything from the practice of a ritual to a kinship category, as if all such things were commensurate “elements.” But Julian Steward pointed out the quantitative contrast between the 3000 to 6000 culture elements documented among the native people of western North America, and the 500,000 artifact types that U.S. military forces landed at Casablanca in 1942 (1955: 81). The figures give an indication of great difference in the complexity of these respective societies.

But such enormous differentiation tells only part of the story. Complexity consists of organization as well as structural differentiation. The challenge of the logistical train that headed for the coast of northwest Africa was not just the great diversity in artifact types, but also how they were packed aboard ship. In proper combat loading, material should be stowed in reverse order from the sequence needed upon landing. No doubt the U.S. military understood this in principle, but failed to apply it in practice. Material cascaded into the docks chaotically, and was loaded onto ships haphazardly. Soldiers broke the windshields of stowed vehicles that they had to climb over to find items. Explosives wound up in passageways, staterooms, and troop holds. Needless to say, unloading under fire was chaotic. Guns arrived on the beach without gunsights, without ammunition, without gunners. Important radio equipment had been stored as ballast because it was heavy. Medical supplies remained on the ships for 36 critical hours (Atkinson 2002: 34-35, 138-139). To find any specific thing it was necessary to unload nearly everything.

The problem with the 500,000 artifact types shipped to Casablanca is that they did not, in fact, comprise a complex system. The system lacked organization. Differentiation in structure without corresponding organization makes a system complicated (Allen, Tainter, and Hoekstra 1999, 2003), not complex. In a complex system, certain elements constrain others, making the behavior of the constrained elements simple and predictable. This is the essence of the evolution of social and cultural complexity–differentiation in structure combined with organization that increases to constrain the structure (Tainter 1988). As a human system becomes elaborate and organized, the behavior of its individual elements–institutions, social roles, or just individual people–is channeled and simplified. Where once in human societies most people could perform most tasks, today specialization is the norm. Behavior is constrained and predictable.

Of what does human organization consist? Historians, working with the records of kings and governments, have tended to emphasize the evolution of one facet of organization, the social and political hierarchy. The essence of hierarchy is asymmetric relations, in which some elements constrain the behavior of others (Allen and Starr 1982). Certainly hierarchy is a salient kind of human organization, but it is not the only kind. In a human system, symmetric relations also constrain. Heterarchy is a kind of organization in which elements are unranked relative to each other, or can be ranked in a number of ways (Crumley 1979). The states comprising the European Union are a heterarchy, for they are in principle unranked, or ranked only temporarily. They constrain each other through symmetric relations. The emphasis and guiding principle is common policy. Hierarchy and heterarchy differ in the ways in which they integrate elements, the speed at which they can operate and reach decisions, and their effectiveness in various circumstances. The advantage of hierarchy is that it can act quickly, uniformly, and over a large area. Heterarchy is better suited to incorporating varieties of experience, but is slower to act. The two organizing principles each perform best in different circumstances (McIntosh, Tainter, and McIntosh 2000).

A new area of investigation is how structure and organization emerge from the interactions of individuals. This is the basis of much of the work done in the area of agent-based modeling. The advent of telecommunications and electronic databases now allows social scientists to identify small social networks, and see the structures that emerge from them. Applied to the conflict in Iraq, researchers have employed regularities in the formation of networks to conclude that the number of contending factions has grown from 15 to 35 in 2003 to 100 to 130 today (Bohannon 2006).

So in a human system, structure and organization emerge from top-down constraints, from the horizontal constraints of peers, and from the atomistic interactions of gregarious individuals. The intersection of these makes for the complicated, complex, confusing life that we all know, and through which daily we try to navigate.

It was once thought that complex society–what we conventionally call “civilization”–emerged solely through human creativity, or from the availability of surplus energy provided by agriculture and fossil fuels. We now understand that this was naive: Complexity imposes costs, and humans are frequently averse to it. Human systems often increase in complexity through the mundane process of solving problems (Tainter 2000, 2006; Allen, Tainter and Hoekstra 2003). As problems increase in scale and complexity (a phenomenon that is perhaps inevitable), problem solving efforts must complexify correspondingly. Often the solution to a problem is seen as differentiating structure (perhaps through technology or formal institutions) or as increasing organization (to further channel behavior). The best recent example is the response of the US government to the attacks of September 11, 2001. The governmental response was to increase structural differentiation, by forming new bureaus and rearranging existing ones, and to increase organization, in order to constrain behavior still further. The official commission convened to investigate the incident issued a thick report that recommended still more complexity in order to forestall possible future attacks.

In the world of complex adaptive systems, there is (to use a colloquial expression) no free lunch. In any living system, complexity costs. Once structure and organization increase, a way must be found to pay for the increase. Complexity that emerges through problem solving typically increases prior to the energy (or other currency) that is required to fund it. Thus governments never have sufficient funds to pay for all necessary programs, and firms, families, and individuals must always make choices about which problems they will solve, and which they cannot afford to address. Complexity that emerges through problem solving is a benefit/cost function. It can enhance sustainability, but also undermine it (Tainter 1988, 2000, 2006).

Are there other concepts of complexity that can fruitfully be applied to human systems? Deliberately I refrain from undertaking to answer this question. Instead I pose it for our forum, in the hope that interesting and worthwhile discussions will emerge.


Ahl, V. and T. F. H. Allen. 1996. Hierarchy Theory: A Vision, Vocabulary, and Epistemology. Columbia University Press, New York.

Allen, T. F. H. and T. W. Hoekstra. 1992. Toward a Unified Ecology. Columbia University Press, New York.

Allen, T. F. H. and T. B. Starr. 1982. Hierarchy: Perspectives for Ecological Complexity. University of Chicago Press, Chicago.

Allen, T. F. H., Joseph A. Tainter, and T. W. Hoekstra. 1999. Supply-side sustainability. Systems Research and Behavioral Science 16:403-427.

Allen, T. F. H., Joseph A. Tainter, and T. W. Hoekstra. 2003. Supply-Side Sustainability. Columbia University Press, New York.

Allen, T. F. H., Joseph A. Tainter, J. Chris Pires, and Thomas W. Hoekstra. 2001. Dragnet ecology­“Just the facts, ma’am”: the privilege of science in a postmodern world. BioScience 51:475-485.

Atkinson, Rick. 2002. An Army at Dawn: The War in North Africa, 1942-1943. Holt, New York.

Bohannon, John. 2006. Tracking people’s electronic footprints. Science 341:914-916.

Crumley, Carole L. 1979. Three locational models: an epistemological assessment of Anthropology and Archaeology. In Advances in Archaeological Method and Theory, Volume 2, edited by Michael B. Schiffer. 141-173. Academic Press, New York.

McGuire, Randall H. 1983. Breaking down cultural complexity: inequality and heterogeneity. In Advances in Archaeological Method and Theory, Volume 6, edited by Michael B. Schiffer, pp. 91-142. Academic Press, New York.

McIntosh, Roderick J., Joseph A.Tainter, and Susan Keech McIntosh. 2000. Climate, history, and human action. In The Way the Wind Blows: Climate, History, and Human Action, edited by Roderick J. McIntosh, Joseph A. Tainter, and Susan Keech McIntosh, pp. 1-42. Columbia University Press, New York.

Steward, Julian H. 1955. Theory of Culture Change. University of Illinois Press, Urbana.

Tainter, Joseph A. 1988. The Collapse of Complex Societies. Cambridge University Press, Cambridge.

Tainter, Joseph A. 2000. Problem solving: complexity, history, sustainability. Population and Environment 22: 3-41.

Tainter, Joseph A. 2006. Social complexity and sustainability. Ecological Complexity 3: 91-103.

fis mailing list

Reply via email to