This book reprints almost all articles on slavery published by the author.

Rather than summarize each chapter of the publication, we prefer to give a presentation of each of the major topics dealt with.

Definition of Slavery

So as to be useful from a sociological perspective, the term “slavery” should be reserved to describe a specific form of dependence, and should be carefully distinguished from numerous other types of subjection known in history (the Helots of Sparta, mediaeval serfdom, etc…) or in ethnography (individuals pledged against debts and all people known in English as bondmen). The material condition of slaves can be extremely variable: together with slaves exploited for labour, Islamic polities, the Roman Empire and African kingdoms have employed slaves as police corps, warriors or members of the administration. One of the most ancient uses of slaves in all societies since the Iliad was for pleasure. Another lesser known yet just as well attested function of the slave was to follow his master into the grave: this is particularly well-known in Africa or amongst Indians of the north-western coast of North America. One can safely emphasize that neither occupation nor function, nor in fact way of life constitute useful elements in defining slavery. Juridical status is the only valid criterion allowing a description of what slaves of a given society have in common, beyond differences in their material condition. No sociological approach to slavery can dismiss a study of the juridical status of slaves. Yet this status can vary from one society to another. Once juridical aspects have been acknowledged, one should go beyond them so as to establish a more general characterization from a sociological viewpoint.

The slave in antiquity was excluded from the polity (according to scholars working on the ancient world); the slave of African societies was excluded from kinship, he possessed neither a name nor a lineage. The private slave of ancient kingdoms was answerable to his master only, and did not owe anything to the king, neither paid taxes nor had any military obligations; he was not the ruler’s subject and therefore did not relate to the latter; etc… In one way or another the slave is a dependent outside one of the most fundamental social dimensions of society: excluded from the polity and the city’s institutions in the Classical World, outside kinship in lineage-based societies where such an aspect plays an essential part, excluded from any relationship with the king in monarchies, etc… This criterion of exclusion, added to the fact that the master could make a profit by selling the slave or physically putting him to work is enough to distinguish him from another type of dependent (for example the serf or the individual pledged in case of unpaid debts).

"Integration" and the Slave’s Fate

It has often been said that the slave in primitive societies (particularly in lineage-based communities of Africa and elsewhere) was entirely different from his counterpart in the ancient world in that he was part of the family. His status as adopted member or parent has also frequently been mentioned. These two ideas are the cause of considerable confusion and themselves derive from two fundamental misconceptions.

First of all, there is no distinction between a kinship group (such as a lineage, a clan, a Roman gens, etc…) and a domestic one (family unit), and the latter can include both kin relations and others. If this distinction is not clear the notion of “integration” means nothing at all. Two points have to made concerning antiquity :

1)- The slave in antiquity was integrated in the family, if one accepts the definition of a familia, whose initial meaning was “set of slaves” but which then came to designate a group of men, possessions and domestic animals living in the same domus under the authority of a paterfamilias. The slave was of course not part of a kin-related group (i.e. part of the set of agnates) and was not admitted into the gens.

2) - The slave was therefore simultaneously included and excluded; yet all other men (free men or citizens) were also included within a certain group (the city/polity, the family, the gens, etc…). Hence the slave was distinguished by his exclusion from everything apart from the family.

The same can be said for slaves in primitive societies, whose main characteristic was to be “without kin”, without identity and social protection (this protection was ensured in stateless societies by the threat of a blood feud, which could be started only by the victim’s immediate parents, collaterals and siblings). This did not prevent the slave from participating in domestic tasks, from being called “child” or even “son”, but these names were pronounced in a metaphorical sense since, literally, he was the “son of nobody” but accepted the authority of a master called “father” in this relationship of reciprocity. The second idea, as mistaken as the first one, has taken root among anthropologists as a result of a well-known and general fact observed in lineage-based societies: sooner or later the slave will be adopted by his master, will become a son and will have all the rights and prerogatives of the latter. Basing themselves on this established fact scholars have built false theoretical constructs by linking slavery and adoption, and by making one condition similar to the other. Once adopted, the slave looses his servile status, yet he remains a slave as long as he has not undergone this process of adoption. In the interim (this aspect has been underestimated by some scholars), he remains submitted to his master’s will and can be killed by the latter. Only good and faithful slaves will be adopted. The integration period can be quite long and the process can only happen during the second or third generation. Some primitive communities, moreover, never adopt their slaves: such is the case among societies of the north-western American coast. In any case, the fact that slaves can one day undergo adoption does not mean absence of slavery, and this conclusion should be considered completely absurd and groundless.

Role and Function of Slavery in primitive Societies (1) : Power

One could assume that in these societies there is no mode of production based on slavery. Yet to only speak of “domestic” or “patriarchal” slavery (these are, in our opinion, empty formulae) is to undermine both the reality and function of slavery.

First of all, in these societies the slave’s condition is synonymous with hardship. The master possesses an arbitrary power of life and death over his slave. Moreover slavery plays an essential role that is part and parcel of the overall strategies typical of these societies. Ethnology has demonstrated that slaves in lineage-based societies primarily strengthened the lineage chief’s power, first by increasing the global number of dependants (parents, slaves, clients, etc…), and then by multiplying the number of kin relations (this would happen either when the offspring of a concubine slave would be admitted inside the group or when the faithful slave would be adopted by the lineage). This does not imply that in primitive societies slaves did not accomplish real tasks; the point is to re-emphasize the importance of the latter: lineage or household chiefs and all powerful people in these primitive communities would vie for power and direct influence over men. This was a primary aim, as important, if not more, as the acquisition of wealth. The slave is the best dependent able to serve a man with the will to dominate, because he possesses no rights at all, and is totally and literally in his master’s hands. He is devoid of any social protection because he is without blood relations, he is more submissive, servile and reliable than a son whose rights are at least partly acknowledged by tradition. The slave can be used as an assistant in raids or warfare.

The Indians of the north-western coast never adopted their slaves. Why ? The reason is that these societies are not based on lineage: a (free) man could move to his maternal uncle or his father’s household, or be admitted in any other family. To free a slave or to accept him as a family relation was to risk loosing a dependent.

Role and Function of Slavery in Primitive Societies (2) : Wealth

In primitive societies the division of labour is weakly developed and production is hardly specialized; consequently wealth is aimed at acquiring material goods through trade. It is meant as a way to fill one’s obligations: reimbursing debts owed by blood relations, paying fines resulting from both deliberate or unintentional and religious or secular offences harming the honour or the goods of a community member, and above all satisfying kin-based duties. These include paying for indispensable marriage transactions (above all for the price of the bride; by paying this price, the husband would acquire (partial) rights over his spouse (over her offspring, work, etc…). The similarities between these (partial) rights over certain people (wives, children, etc…) acquired through payment and slavery have been efficiently demonstrated by Kopytoff and Myers. In African traditional law or according to the customs of many South-East Asian societies, the sale of rights over people is a widespread phenomenon: slavery is but an extreme form of this traditional practice.

There is in addition a significant correlation between societies practicing slavery and others where a price is to be paid for securing a bride. The sociological consequence is the reciprocal convertibility of wealth and dependence. Wealth enables the acquisition of dependents and dependents render possible the provision of wealth. Slavery only exists in societies where one also finds a certain type of social wealth. It presupposes wealth but also prevents its excessive growth by always allowing its conversion into a number of dependents whose acquisition is a major objective of these communities.

The convertibility of wealth and dependence is complete, however, only when slavery as a consequence of debt insolvency exists. The poor and debt-ridden indigent is at risk of being cast into slavery. This is frequent in most West and Central African societies, as well as in tribes of South-East Asia.

The first paradox, if one is to consider the Soviet theory of “general slavery”, is the low number of slaves, their unimportant function, the proven absence of any slave-based production. An Egyptologist like B. Menu can successfully maintain that there were no slaves in ancient Egypt. In China the measures adopted by the Emperor Wang Mang (the prohibition of the sale of slaves in 9 AD) would have led to the abolition pure and simple of slavery, had it not been for the opposition and uproar caused by such a decision. These instances are nevertheless proof of the will to abolish slavery on the part of a political structure traditionally described as despotic. Some nine centuries later Japan officially abolished slavery after adopting most Chinese institutions, only to restore it when central power declined, during the period known as “feudal”. Yet the intentions of the Imperial regime were obvious, and one must admit that this abolition was a decision taken by an administration that was anything but liberal or democratic. The Emperors of Japan are the offspring of the gods. Neither Rome nor Athens, and for that sake none of their philosophers and thinkers, ever thought of abolishing slavery.

The second paradox is that the slave is generally protected against the master’s abuses, and the latter does not possess the power of life and death over his servile dependent. A slave’s status was far better in a despotic regime than in primitive societies or ancient democracies.

In fact the above-mentioned paradoxes are non-existent. An all-powerful ruler does not need any slaves. The ordinary power that he exerts enables him to extort from his subjects huge taxes and lengthy military duties. Yet much more can be added. Slavery is not only useless to the despotic ruler, it also undermines his power. We are speaking here of course of private slaves owned by an empire’s elites and nobility (high officials, princes, rich merchants, court favourites, etc…).The slave has only one master, depends totally on him and will not serve another. He is not a subject of the king, does not pay any taxes, nor does he carry out any military duties. Hence a powerful man’s private slave can only serve his master against the king. Wilbur has very convincingly demonstrated these points. This is why it is in the ruler’s interest to abolish slavery, to at least to limit this institution or to set himself up as the protector of slaves against their masters, thus weakening the link between them. It is once more slavery’s political function that explains its inessential role in despotic states.