Despite the historical importance of slavery in many parts of the world anthropologists study, the subject remains somewhat marginal within the discipline. Much more has been written by historians, most of this dealing with the transatlantic trade during which more than 9 million Africans were transported to the New World, principally to work on European and American-owned plantations. Much of this historical work has dealt with the political and economic dimensions of the supply-side of the trade and its effects on coastal and interior populations of Africa (see also Wolf 1982: 195-231), whilst more recent studies have added to our knowledge with some particularly fine social histories of slavery in the Caribbean and Southern United States. The subject was, however, largely ignored in modern anthropology until it received both ethnographic and theoretical attention in the 1970s from French Africanists working within a Marxist framework, and British and American scholars drawing on ethnohisto-rical data (see e.g. Meillassoux 1975; Miers and Kopytoff 1977).
Given the historical significance of the Atlantic trade it is not surprising that the dominant stereotype of slavery is that of the New World Afro-American plantation system, a stereotype in which ‘slavery is monolithic, invariant, servile, chattellike, focused on compulsory labour, maintained by violence, and suffused with brute sexuality’ (Kopytoff 1982: 214). Yet examples from different times and places of what is usually taken to be slavery reveal a great variation in both the type of servitude slaves experienced (a common difference often being noted between domestic and chattel slaves), and the political and economic systems in which the institution existed. There are cases of slaves owning property (even other slaves), occupying important offices, and selling their product. And as an institution slavery was to be found among peoples with polities and economies as diverse as those of the American Northwest Coast, Borneo, and sub-Saharan Africa.
Theories seeking an economic explanation of slavery have frequently dwelled on the land-labour-capital triad, attempting to account for the incidence or type of slavery in terms of the relative availability of each in any given case. However, a central problem for recent anthropological discussions of slavery has been one of definition. Most authors abjure legalistic and economistic definitions which they see as rooted in capitalist notions of property, although there is disagreement as to who has been successful in this regard. Legal definitions of slaves as property come up against the predictable problem of what is to be defined as property? The common distinction made between slavery and freedom also raises problems, given differing notions of social belonging in the societies that anthropologists study. Finally, the varieties of work performed by slaves in different societies also makes an economistic definition of slavery based on labour (compulsory or otherwise) similarly problematic.
Another common theme in recent discussions has been the relationship between slavery and kinship. In the introduction to a collection on African slavery, Miers and Kopytoff (1977) for example discuss slavery in terms of transfers in ‘rights-in-persons’ within kinship systems open to absorbing outsiders, thereby making them quasi-kin. Thus whilst starting off as outsiders, slaves are subsequently absorbed by a kinship group, an argument which proposes that slavery must be looked at as a process rather than as a state of being. In a comparison of African and Asian systems of slavery, Watson (1980) contrasts the ‘open’ system described by Miers and Kopytoff with ‘closed’ systems where kin groups tend to be exclusive, and where slaves remain outsiders rather than being incorporated. This difference Watson links to land, arguing that in ‘closed’ systems land is a scarce valuable, whereas in ‘open’ systems land is plentiful and wealth resides in people. The status of slaves as outsiders is similarly central to Meillassoux’s (1991 [1986]) Marxist analysis of African slavery, although he is at pains to point out the radical differences between his own model and that of Miers and Kopytoff. For ^Meillassoux slaves are and remain ‘aliens’ par excellence, and he suggests that a fundamental characteristic of slavery is ‘the social incapacity of the slave to reproduce socially -that is, the slave’s juridical inability to become kin’ (1991 [1986]: 35). Thus for Meillassoux there is no ‘slavery-kinship continuum’ as Miers and Kopytoff suggest, for slavery is the antithesis of kinship.
Despite their various differences, the issue of identity is implicit in the theories of both Kopytoff and Meillassoux. The incorporation of outsiders as kin and the incapacity to reproduce socially are both arguments about personhood and identity, issues which may prove central to further developments in the anthropology of slavery as it begins to draw on recent anthropological discussions of the person and gift and commodity exchange. Kopytoff himself suggests that ‘The sociological issue in slavery is … not the dehumanization of the person but rather [their] rehumanization in a new setting’ (1982: 222). However whilst one may agree that slaves are not dehumanized (i.e. reduced from persons to things), neither are they fully rehumanized in their new setting, and we find that systems of slavery exhibit great variation in the ways that the status of slaves is marked out in terms of their personhood. On the New World plantations we perhaps find the system in which the reduction of persons to things was most marked; elsewhere, however, we find the personhood of slaves marked differently — sometimes slaves were sacrificed, their tombs were symbolically devalued through their positioning and construction, and they were prevented from having descendants or achieving ancestorhood. In short their status as slaves was marked out by their inability to fully realize their personhood in various ways, an issue touched on in a recent paper by Guyer (1993). Thus whilst slaves are not wholly reduced to the status of things, they are always to be found as persons of a different order.
It has been the case up till now that anthropological studies of slavery have relied on archives and oral history for their data. But as anthropological interest in slavery grows, so too does its incidence in the contemporary world, with cases reported for places as different as Eastern Europe and Eastern Indonesia. Although cultural analyses of slavery may extend ‘our understanding [of slavery] … in the very process of dismantling the concept’ (Kopytoff 1982: 227), some people now find themselves faced with economic conditions under which the institution thrives.
