Home Thumbnail Summary Introduction Table of Contents and Excerpts Excerpts from Reviews About the Author Buy the Book
















What Is Social Complexity?


...Cultural evolutionists usually speak of human societies evolving toward greater "complexity" or higher degrees of "organization." And precisely defining complexity or organization is such a notoriously frustrating task that many people give up and fall back on an intuitive definition, like Supreme Court Justice Potter Stewart's famous definition of pornography: "I know it when I see it."... 


...One scholar who took on the stiff challenge of [defining complexity was the anthropologist] Robert Carneiro. In the early 1960s, Carneiro actually tried to quantify social complexity and then document its growth. He started by taking 100 diverse societies—hunter-gatherer and agricultural, literate and pre-literate, that were well described in the anthropological or historical literature. He listed some of their conspicuous traits. Did they have special religious practitioners? Temples? Craft specialization? Craft specialization for trade?

It wasn't enough to just invent such traits out of thin air. Their significance had to be validated by showing, through a technique called Guttman scale analysis, that they were logically interrelated. In particular: some traits had to imply the existence of others with high probability. Sometimes these implications were almost obvious. Thus all societies with temples had religious specialists. Sometimes the implications were less obvious. All societies with formal codes of law had towns with over 2,000 people. The reverse, of course, was often not true: not nearly all societies with religious specialists had temples, and not all societies with towns of over 2,000 had legal codes.

Carneiro ranked the traits in order. The most common—such as trade between communities—were at the bottom, and the less common calendars, cities of over 100,000were at the top. In general, societies with any given trait would tend also to have the traits below it. I stress the word tend. The "higher" traits don't imply the lower with 100 percent confidence. And that's especially true of traits at nearly the same level. Thus, communities with clearly marked social status (trait number 7) don't always have formal political leadership (trait number 3), but societies with markets (trait number 26) do.

All told, when you stack up Carneiro's fifty traits in order, you find considerable predictive power. If a society has a given trait, then the chances are over 90 percent that it will have any randomly selected "lower" trait.

Of course, it would be nice if the figure were 100 percent. And, in general, it would be nice if laws in the social sciences were airtight, like laws in the "hard" sciences, rather than just statistically predictive. But, alas, human society is the most complex phenomenon in the known universe, so identifying the key to its dynamics is a messy task...


...What did these results mean? Carneiro postulated that this list of traits was a kind of ladder, a general evolutionary sequence. If you could have watched these different societies evolve, rather than just look at them statically—if you had movies instead of snapshots— you'd see them all acquire these traits in roughly the same order, from the bottom of the list to the top. Cultures in general would tend to have trade between communities and special religious practitioners and craft specialization long before they had temples and laws and big cities.

Years later, to test his model, Carneiro looked at the evolution of Anglo-Saxon England from the fifth through the eleventh centuries. In historical sources he found 300 cultural traits whose order of appearance he could document, and 33 of those traits had been among the 100 on his ladder. He looked at every pair of traits among those 33—all 528 combinations—to see, in each case, if the two had indeed appeared in the order predicted by his ladder. The answer was yes in 86.5 percent of the cases. Moreover, almost all of the exceptions—the 13.5 percent—came when comparing traits that had been very near one another on the ladder to begin with, such as "towns of 2,000 or more" (trait number 27) and "state or church employs artisans (28)"; or "taxation in kind" (24) and "military conscription" (25). In other words: they were pairs of traits whose relative order you would have guessed would vary from one case of cultural evolution to the next.

All of this gave Carneiro confidence that he had indeed found a meaningful list of cumulative cultural traits: traits that tend to appear in a particular order and, having appeared, tend to persist. Hence, he suggested, a rough gauge of cultural complexity: add up the number of these traits a society possesses, and thereby assign it a rank. As it happened, another scholar, Raoul Naroll, had earlier developed a quite different measure of social complexity, involving, for example, the number of craft specialties and the number of "team types." And these two measures of complexity ranked societies in almost exactly the same order. This finding, Carneiro wrote in 1969, "strengthened my conviction that cultural complexity is something real, objective, and measurable."

Notwithstanding Carneiro's confidence, there is today no consensus on how to measure social complexity. Some analysts stress the number of "levels of hierarchical control" in social organization. And many, like Naroll, consider the degree of division of labor to be relevant (rather as some biologists use the number of "cell types" to gauge an organism's complexity). But the Potter Stewart litmus test— "I know it when I see it"—retains a certain appeal.

In defense of cultural evolutionism, it should be noted that the "hard" sciences aren't doing much, if any, better when it comes to defining complexity. Physicists and chemists have rigorous definitions for "order." (A pure substance with its many identical molecules neatly arrayed is the ultimate in order.) And they have rigorous definitions for "entropy." (The ultimate in entropy is a heterogeneous substance with its many different kinds of molecules randomly distributed.) And they know that complexity is something between the two—something that has pockets of order yet isn't pure order. But there is no consensus on what exactly is the essence of complexity, or on how to quantify complexity.

An excerpt from Nonzero: The Logic of Human Destiny, By Robert Wright, published by Pantheon Books. Copyright 2000 by Robert Wright. Other excerpts are available at