An anthropologist emails:
Steve Sailer has recently posted on isteve on trying to come up with a definition of class. Here are a few thoughts.
I don't think we can get too far from the standard sociological notion that "class" has to do with inequalities in power, wealth and status in stratified societies, without completely changing the meaning of the word. But we can add the idea that class is not only a matter of social stratification, but involves assortative mating based on Power, Wealth, and Status. I take it this is what Steve is getting at. This would mean that a rich powerful celibate priesthood would not be a class.
Why bother? On reason is that over time classes may differentiate genetically if different genes help people get into different classes. This is part of the argument of The Bell Curve. The genetic consequences of a pure class society will be different from those of a caste or ethnically stratified society. In the former situation, only genes relating to class (or linked genes) will differ between classes, in the latter, where descent not assortative mating is driving things, all sorts of other genes may differ between strata.
Even without genetics, marriage practices can make a difference to class. The anthropologist Jack Goody has spent a lot of time looking at broad differences between African and Eurasian societies. He says that by and large, African societies, even when stratified, don't form Eurasian style classes, because African polygyny means that high status groups incorporate low status females in large numbers. So you don't, Goody claims, get the distinctions between "high cuisine" and "low cuisine," and other high/low culture distinctions in traditional Africa as much as in traditional
I don't have any well-worked definition to offer, but the basic idea seems to be that we have to take into account that people are more than just isolated monads floating around (as in a lot of classical economics) but have families and kin and (most of us hope) descendants, and our definitions of social aggregates ought to reflect this.
I had never thought about class (or its relative absence) in traditional African societies before. It's one of those dog-that-didn't-bark phenomena that are so hard to notice, but are often very illuminating when you finally realizing they are missing.
I have a book by Goody sitting around, but the prose style is awfully academic so I haven't gotten very far.