Redistribution

What Carneiro calls ``the shibboleth of chiefly redistribution'' ([Carneiro 1988]: 500) was an important component of some early formulations of the chiefdom concept. Sahlins and Service, particularly, in influential statements, characterized chiefship as a functional institution which served to redistribute goods and to pool subsistence risk across a heterogeneous territory ([Service 1962,Sahlins 1958,Sahlins 1963,Fried 1967]). Redistribution was such a central concept for Service that he defined chiefdoms as ``redistributional societies with a permanent central agency of coordination'' ([Service 1962]: 144).

The concept of chiefly redistribution suffered the reverses of functionalism generally in the '70s and '80s. Various researchers pointed out that in fact many of the goods collected by chiefs from their subject population were never redistributed at all, or were distributed only to a restricted circle of other elites in a politically opportune manner (summarized in [Earle 1987a]). That is, that chiefly redistribution was not in fact functional but rather was oriented to the aggrandizement of the chiefly personage him- or herself.

Whether chiefly redistribution was, in fact, an adaptive mechanism or simply tribute, the fact remains that chiefs - or more generically, leaders - in small scale societies world-wide collected and the redistributed (``dissipated'', as Spencer would have it [[Spencer 1998]]) a portion of the agricultural and other products of the populace at large. It is important to remember, however, that the surplus agricultural product collected by leaders was eventually eaten. It was not, as a matter of course, destroyed. What this means is that the chief or leader supports other people, to a greater or lesser extent. The distribution of this surplus could take many forms. The chief could take many wives and support a large household, as was commonly the case. Or the chief could support retainers or artisans, and participate in a system of wealth finance ([Earle 1997]). Alternatively, the chief or leader could distribute at least a portion of the surplus in massive acts of generosity ([Hayden 1996a,Dietler 1996]). Most empirical cases will involve many different strategies employed in varying combinations. In all of these cases, however, the chief or leader acquires dependents.

Conceptualized in this way, a chief or leader is a point resource on the landscape. S/he is a location from which goods emanate, and proximity to this location could prove advantageous to other persons. It is in this way that the location of certain chiefs or leaders can become a settlement determinant. And it is in this way that the differential success of leaders can lead to the kind of unequal population growth rates that we observe in Taraco Peninsula prehistory. To return to the example, if the leaders of one village are more successful than are those of surrounding villages, then access to these leaders will be an inducement for some people to establish their residence in the first village. I emphasize that this effect does not have to be a strong one, nor does the difference in chiefly success between villages have to be pronounced. The slightest difference might be enough to produce a systematic bias in residential decisions which over the course of centuries could produce the observed population growth differences between villages.