Wednesday, January 9, 2013

Stephen Shennan, Genes, Memes, and Human History

Archaeology is by its nature an inductive, data-rich, detail-oriented science. What most archaeologists do is extract objects from the ground using extraordinary care, making meticulous records of the process, and then analyze them using their minute knowledge of similar sites left by the same or similar cultures. You can read thousands of pages of archaeology without encountering anything the deserves the name of "theory," or any big idea about the past.

Yet there are big ideas in archaeology. Especially as it is taught in universities, archaeology is all mixed up with notions of cultural evolution, ethnic identity, gender studies, and so on. I find, though, that these sorts of ideas have almost no impact on how I do my work or what I hear about when I go to archaeological conferences. This brings me to Genes, Memes, and Human History: Darwinian Archaeology and Cultural Evolution (2002), by British archaeologist Stephen Shennan. This is an excellent book, and in fact I would say it is the best theoretical, big-picture work by an archaeologist I have read in a decade. It is full of fascinating information about humanity's past, and some of the ways that past might have been shaped by natural selection. As an attempt to re-invigorate archaeological practice with an infusion of evolutionary science, though, it largely fails, because the issues it deals with intersect only seldom with archaeology as it is normally done.

Shennan, like the rest of the world, is fascinated by humanity's evolutionary history and the impact our genes have on who we are and how we live. Like almost anyone with any scientific interests, he is dismayed by the casual way assertions about genes for this and that are thrown around in the popular discourse. He suggests that there is one discipline that can show us how the intersection of our genetic heritage and our cultural creativity has worked out in practice, and that is archaeology. Archaeology shows us how people actually lived in the past, providing a million years of real data instead of mere "men are from Mars" speculation. In the other direction, evolutionary theory can inform archaeologists' attempts to figure out what their data means, and questions about genetic vs. cultural evolution provide an agenda for archaeological research.

There are a few places in which Shennan makes this work wonderfully. My favorite concerns the apparently discontinuous nature of the archaeological record. Based on how much they have found from various cultures, archaeologists get the impression of major fluctuations in the populations of the areas they study. In the Chesapeake region, for example, you can hardly put a shovel in the ground without finding artifacts from what we call the Savannah River culture of around 2500 to 2200 BCE. Artifacts for the following 500 years are, by comparison, rare. Shennan presents data from intensely studied parts of Europe, such as the Swiss lakes, showing what appear to be waves of intense occupation, separated by empty centuries. But are these population fluctuations real, or are they an artifact of how archaeologists work? In the Chesapeake region there are archaeologists who believe the fluctuations are illusory. They think there were plenty of people around in the 2200 to 1500 BCE period, but we don't recognize them because their tools were not distinctive, there settlements were in places where they were less likely to survive, or we are simply mis-identifying their remains.

Shennan presents data from human evolution that may have a real bearing on these questions. He notes that humans are capable of breeding very fast. We can, when we want to, produce children three times as fast as chimpanzees or bonobos. In fact our breeding potential is so great that if you look at any traditional culture, whether peasants or hunter-gatherers, you find that people are taking active steps to limit fertility. (Abortion, infanticide, requiring young couples to accumulate property before they can marry, and so on.) Why would we have evolved this ability to breed rapidly, if in practice we rarely use it? One explanation would be a history punctuated by population crashes and rebounds. Those who could produce children quickly to occupy the depopulated landscape would have a huge evolutionary advantage, leading over the millennia to our remarkable fertility. The convergence of these two lines of evidence, archaeological and physiological, makes a very strong case that human populations have in fact been highly unstable, and therefore that the discontinuities in the archaeological record do represent population booms and crashes.

Another fascinating section of the book concerns what Shennan calls life history strategies. You cannot get a very clear idea of an organism's evolutionary fitness by observing it an any one time in its life. If you consider just that one moment, you might think the organism was sabotaging its own reproductive chances and that Darwin had things all wrong. For example, sometimes wolves old enough to breed stay with their parents for another year, helping to raise a new litter of siblings instead of looking for a mate and setting up their own families. Long-term studies of many wolf packs have shown, though, that wolves who do this acquire skills that help them when it comes to raising their own offspring, so forgoing a year of breeding potential may be worth it in the evolutionary long haul. This is especially true in lean years, when most pups of inexperienced parents die anyway. Putting off mating is a strategy that may actually lead to more surviving offspring over the whole life of the animal.

Humans, of course, use many different life history strategies to maximize their evolutionary success. Shennan cites two amazing, long-term studies of well-documented human groups to show this. Farmers in the German district of Friesland left most of their property to a single son, and other sons had to leave the district or look for other kinds of work. Daughters were married off to other farming families. Historian Eckart Voland has shown that these families invested  much more in their daughters, through dowries, than in their non-inheriting sons. Why? First, Voland showed that property owners left many more descendants than those without property. Providing large dowries made it possible for daughters to marry other property-owning farmers and leave many descendants, whereas non-inheriting sons would likely not leave many descendants no matter what was done for them. Similar effects led lower-ranking nobles of Renaissance Portugal to provide large dowries for daughters while sending younger sons overseas, into the army, or into the church. Historian James Boone showed that the daughters of minor families had more descendants than the sons even when they lived long lives, which partly explains why those families treated younger sons as expendable. Younger sons of minor families were far more likely to die in war than either elder sons or the younger sons of the very top clans.

There is much, much more in Shennan's book, from history, ethnography, and archaeology, and I highly recommend it.

As an archaeologist, though, I found it in many ways discouraging. The data Shennan uses to advance evolutionary arguments comes mostly from ethnography or history, not archaeology. Where archaeology does appear the argument depends on remarkable sites and remarkable circumstances, for example an entire German valley that was minutely explored by archaeologists before the whole place was dug up by coal companies. The preservation of animal bone seems to be quite extraordinary at coastal sites in northern California, allowing us to make lots of inferences about hunting and fishing strategies that could never be made from the typical site, or from any site in my part of the world. The neolithic villages around Swiss lakes have been exhaustively explored by archaeologists for a century, and the preservation of large amounts of waterlogged wood allows us to date them very precisely using tree rings. It is only in these special cases, it seems, that archaeology is exact enough to support the kind of theorizing Shennan wants to do. Does that make the rest of it a waste of time?

I would like to make two criticisms of Shennan's approach. One concerns disease, about which Shennan says next to nothing. Over the past 10,000 years infectious disease has been the biggest killer of humans. How has this impacted calculations of advantage, life history strategies, and the like?

My other serious comment involves Shennan's approach to culture, which is very mechanistic. He spends a great deal of time on food acquisition and the attendant technologies, with a whole chapter on "optimal foraging." He also devotes much attention to levels of political organization, that is the rise of chieftainships and early states, and ways to distinguish between chiefly tribes and those with more egalitarian organizations. All fine. But what about the other side of culture, the things that fall under headings like art and religion? Humans put great effort into these things, and in many cultures investment in artistic expertise or religious knowledge is a very successful path to worldly success. Why?

Shennan's approach to culture -- and I have the impression that this is generally true in British anthropology -- is missing the dimension of psychology. A huge amount of what we do is designed to make us feel good. It is designed to entertain us, to move us, to give us confidence, to make us feel that we are part of something larger than ourselves. Some of the liveliest questions in evolutionary theory concern exactly these matters: what is the evolutionary purpose of religious ritual or fertility magic? On these matters the best book I know is still Weston La Barre's The Ghost Dance, and I wish  Shennan would read it and incorporate some of its insights into his view of past cultures. Unless you understand that people in the past felt all the psychic pains and glories that we feel, you will never understand their lives or the record they left behind.

No comments: