1. The Distinctive Features Of Slavery

The traditional, and still conventional, approach is to define slavery in legal–economic terms, typically as ‘the status or condition of a person over whom any or all the powers attaching to the right of ownership are exercised’ (League of Nations 1938, Vol. 6). In this view, the slave is, quintessentially, a human chattel. This definition is problematic because it describes adequately mainly Western and modern, capitalistic systems of slavery. In many non-Western parts of the world, several categories of persons who were clearly not slaves, such as unmarried women, concubines, debt bondsmen, indentured servants, sometimes serfs, and occasionally children, were bought and sold. Conversely, in many slave-holding societies certain categories of slaves, such as those born in the household, were not treated as chattels.

Slavery is a relation of domination that is distinctive in three respects. First, the power of the master was usually total, if not in law, almost always in practice. Violence was the basis of this power. Even where laws forbade the gratuitous killing of slaves, it was rare for masters to be prosecuted for murdering them, due to the universally recognized right of masters to punish their slaves, and to severe constraints placed on slaves in giving evidence in courts of law against their masters, or free persons generally.

The totality of the master’s claims and powers in them meant that slaves could have no claims or powers in other persons or things, except with the master’s permission. A major consequence of this was that slaves had no custodial claims in their children; they were genealogical isolates, lacking all recognized rights of ancestry and descent. From this flowed the hereditary nature of their condition. Another distinctive consequence of the master’s total power is the fact that slaves were often treated as their surrogates, and hence could perform functions for them as if they were legally present, a valuable trait in premodern societies with advanced commodity production and longdistance trading, such as ancient Rome, where laws of agency, though badly needed, were nonexistent or poorly developed.

Second, slaves were universally considered as outsiders, this being the major difference between them and serfs. They were natally alienated persons, deracinated in the act of their, or their ancestors’, enslavement, who were held not to belong to the societies in which they lived, even if they were born there. They lacked all legal or recognized status as independent members of a community. In kin-based societies, this was expressed in their definition as kinless persons; in more advanced, state-based societies, they lacked all claims and rights of citizenship. Because they belonged only to their master, they could not belong to the community; because they were bonded only to their master’s household, they could share no recognized bond of loyalty and love with the community at large. The most ancient words for slaves in the Indo-European and several other families of languages translate to mean, ‘those who do not belong,’ or ‘not among the beloved,’ in contrast with free members of the community, who were ‘among the beloved,’ and ‘those who belonged.’

Third, slaves were everywhere considered to be dishonored persons. They had no honor that a nonslave person need respect. Masters could violate all aspects of their slaves’ lives with impunity, including raping them. In most slave-holding societies, injuries against slaves by third parties were prosecuted, if at all, as injuries against the person and honor of the master. Where an honor-price or wergild existed, as in Anglo-Saxon Britain and other Germanic lands, its payment usually went to the master rather than to the injured slave. Universally, slavery was considered the most extreme form of degradation, so much so that the slave’s very humanity was often in question.

For all these reasons, there was a general tendency to conceive of slaves symbolically as socially dead persons. Their social death was often represented symbolically in ritual signs and acts of debasement, death and mourning: In clothing, hairstyles, naming practices, and other rituals of obeisance and nonbeing.

2. The Modes Of Enslavement

Free persons became slaves in one of eight ways: capture in warfare; kidnapping; through tribute and taxation; indebtedness; punishment for crimes; abandonment and sale of children; self-enslavement; and birth. Capture in warfare is generally considered to have been the most important means of acquiring slaves, but this was true mainly of simpler, small-scale societies, and of certain volatile periods among politically centralized groups. Among even moderately advanced premodern societies, the logistics of warfare often made captivity a cumbersome and costly means of enslaving free persons.

Kidnapping differed from captivity in warfare mainly in the fact that enslavement was its main or sole objective, and that it was usually a private act rather than the by-product of communal conflict. Other than birth, kidnapping in the forms of piracy and abduction was perhaps the main form of enslavement in the ancient Near East and the Mediterranean during Greek and Roman times; and this was true also of free persons who were enslaved in the trans-Saharan and transatlantic slave trades.

Debt bondage, which was common in ancient Greece up to the end of the seventh century BC, the ancient Near East, and in South East Asia down to the twentieth century, could sometimes descend into real slavery, although nearly all societies in which it was practiced distinguished between the two institutions in at least three respects: Debt-bondage was nonhereditary; bondsmen remained members of their community, however diminished; and they maintained certain basic rights, both in relation to the bondholder and to their spouses and children.

Punishment for crimes was a major means of enslavement in small, kin-based societies; China, and to a lesser extent, Korea, were the only advanced societies in which it remained the primary way of becoming a slave. Nonetheless, it persisted as a minor means of enslavement in all slaveholding societies, and became important historically in Europe as the antecedent of imprisonment for the punishment of crimes.

The enslavement of foundlings was common in all politically centralized premodern societies, though rarely found in small-scale slaveholding communities. It was the humane alternative to infanticide and was especially important in China, India, European antiquity, and medieval Europe. It has been argued that it ranked second to birth as a source of slaves in ancient Rome from as early as the first century CE until the end of the Western empire.

Self-enslavement was rare and was often the consequence of extreme penury or catastrophic loss.

In nearly all slave-holding societies where the institution was of any significance, birth rapidly became the most important means by which persons became slaves, and by which slaves were acquired. Contrary to a common misconception, this was true even of slave societies in which the slave population did not reproduce itself naturally. The fact that births failed to compensate for deaths, or to meet the increased demand for slaves—which was true of most of the slave societies of the New World up to the last decades of the eighteenth century—does not mean that birth did not remain the main source of slaves.

However, the role of birth as a source of slaves was strongly mediated by the rules of status inheritance, which took account of the complications caused by mixed unions between slaves and free persons. There were four main rules.

(a) The child’s status was determined by the mother’s status only, regardless of the father’s status. This was true of most modern and premodern Western slave societies, and of nearly all premodern non-Western groups with matrilineal rules of descent.

(b) Status was determined by the father only, regardless of the mother’s status. This unusual pattern was found mainly among certain rigidly patrilineal groups, especially in Africa, where it was the practice among groups such as the Migiurtini Somali, the Margi of northern Nigeria, and among certain Ibo tribes. The practice, however, was not unknown in the West. It was the custom in Homeric Greece and was the norm during the seventeenth century in a few of the North American colonies such as Maryland and Virginia, and in South Africa and the French Antilles up to the 1680s.

(c) Status was determined by the principle of deterior condicio, that is, by the mother or father, whoever had the lower status. This was the harshest inheritance rule and was the practice in China from the period of the Han dynasties up to the reforms of the thirteenth and fourteenth centuries. It found its most extreme application in Korea, where it was the dominant mode down to 1731. The rule also applied in Visigothic Spain, and medieval and early modern Tuscany. The only known case in the New World was that of South Carolina in the early eighteenth century.

(d) The fourth principle of slave inheritance, that of melior condicio, is in direct contrast with the last mentioned, in that the child inherited the status of the free parent, whatever the gender of the parent, as long as the father acknowledged his paternity. This is the earliest known rule of slave inheritance and may have been the most widely distributed. It was the norm in the ancient Near East and, with the exception of the Tuareg, it became the practice among nearly all Islamic societies. The rule was supported among Muslims by another Koranic prescription and practice: The injunction that a slave woman was to be freed, along with her children, as soon as she bore a son for her master.

The only known cases in Europe of the melior condicio rule both emerged during the thirteenth century. In Sweden, it was codified in the laws of Ostergotland and Svealand as part of a general pattern of reforms. In Spain, religion was the decisive factor in the appearance of a modified version of the rule. Baptized children of a Saracen or Jewish owned slave and a Christian were immediately freed. Throughout Latin America, although the legal rule was of the first type—the children of slave women were to become slaves—the widespread custom of concubinage with slave women, coupled with the tendency to recognize and manumit the offspring of such unions, meant that, in practice, a modified version of the melior condicio rule prevailed where the free person was the father, which was usually the case with mixed unions.

3. The Acquisition Of Slaves

Slaves not born to, or inherited by, an owner were acquired mainly through external and internal trading systems; as part of bride and dowry payments; as money; and as gifts. There were five major slave trading systems in world history. The Indian Ocean trade was the oldest, with records dating back to 1580 BC, and persisted down to the twentieth century AD. Slaves from sub-Saharan Africa were transported to the Middle and Near East as well as Southern Europe. It has been estimated that between the years 800 AD and 1800 approximately 3 million slaves were traded on this route, and over two million were traded during the nineteenth century.

The Black Sea and Mediterranean slave trade supplied slaves to the ancient European empires and flourished from the seventh century BC through the end of the Middle Ages. Over a quarter of a million slaves may have been traded in this system during the first century of our era. The European slave trade prospered from the early ninth century AD to the middle of the twelfth, and was dominated by the Vikings. One of the two main trading routes ran westward across the North Sea; the other ran eastward. Celtic peoples, especially the Irish, were raided and sold in Scandinavia. Most slaves traded on the eastern routes were of Slavic ancestry. It was the Viking raiding, and wide distribution, of Slavic slaves throughout Europe that accounts for the common linguistic root of the term ‘slave’ in all major European languages.

The transSaharan trade has persisted from the mid-seventh century AD down to the twentieth and involved the trading of subSaharan Africans throughout North Africa and parts of Europe. It has been estimated that some 6.85 million persons were traded in this system up to the end of the nineteenth century. Although it declined substantially during the twentieth century, largely under European pressure, significant numbers of Africans are still being traded in Sudan and Mauritania.

The transatlantic slave trade was the largest in size and certainly the most extensive of all these systems. The most recent evidence suggests that between the years 1500 and 1870, some 11 million Africans were taken to the Americas. Of the 10.5 million who were forced from Africa between 1650 and 1870, 8.9 million survived the crossing. Although all the maritime West European nations engaged in this trade, the main traders were the British, Portuguese, and French. Four regions account for 80 percent of all slaves going to the New World: the Gold Coast (Ghana), the Bights of Benin and Biafra, and West-Central Africa. Forty percent of all slaves landed in Brazil; and 47 percent in the Caribbean. Although only 7 percent of all slaves who left Africa landed in North America, by 1810 the United States had, nonetheless, one of the largest slave populations due to its unusual success in the reproduction of its slave populations. For the entire three and a half centuries of the Atlantic slave trade, approximately 15 percent of those forced from Africa died on the Atlantic crossing (some 1.5 million) with losses averaging between 6000 and 8000 per year during the peak period between the years 1760 and 1810.

4. The Treatment Of Slaves

It is difficult to generalize about the treatment of slaves, since this not only varied considerably between societies but also within them. There was no simple correlation of favorable factors. Furthermore, the same factor may operate in favor of slaves in one situation, but against them in the next. Thus, in many small, kin-based societies, slaves were relatively well treated and regarded as junior members of the master’s family, but could nonetheless be sacrificed brutally on special occasions. In advanced premodern societies such as Greece and Rome, as well as modern slave societies such as Brazil, slaves in the mines or latifundia suffered horribly short lives, while skilled urban slaves were often virtually free and sometimes even pampered. In the Caribbean, the provision ground system, by which slaves supported themselves, led to high levels of malnutrition compared to the US South, where masters provided nearly all the slaves’ provision. Nonetheless, Caribbean slaves cherished the provision ground system for the periods of self-determination and escape from the master’s direct control that it offered.

In general, the most important factors influencing the condition of slaves were the uses to which they were put, their mode of acquisition, their location— whether urban or rural—absenteeism of owners, proximity to the master, and the personal characteristics of the slaves. Slaves were acquired for purely economic, prestige, political, administrative, sexual, and ritual purposes. Slaves who worked in mines or in gangs on highly organized farming systems were often far worse off than those who were employed in some skilled craft in urban regions, especially where the latter were allowed to hire themselves out independently. Slaves acquired for military or administrative purposes, as was often the case in the Islamic world— the Janissaries and Mameluks being the classic examples—were clearly at an advantage when compared with lowly field hands or concubines. Newly bought slaves, especially those who grew up as free persons and were new to their masters’ society, usually led more wretched lives than those born to their owners. High levels of absenteeism among owners—which was true of the owners of slave latifundia in ancient Rome as well as the Caribbean slave societies and some parts of Latin America—often meant ill-usage by managers and overseers paid on a commission basis.

Proximity to the master cut both ways with respect to the treatment of slaves. Slaves in the household were usually materially better off, and in some cases close ties developed between masters and these slaves, such as those between masters and their former nannies, or with a favored concubine. However, proximity meant more sustained and direct supervision, which might easily become brutal.

Ethnic and somatic or perceived racial differences between masters and slaves operated in complex ways. Intra-ethnic slavery was uncommon in world history, although by no means nonexistent, while that between peoples of different perceived ‘racial’ groups was frequent. The common view that New World slavery was distinctive in that masters and slaves belonged to different perceived races is incorrect. Where there were somatic differences, the treatment of the slave depended on how these differences were perceived and, independently of this, how attractive the slave was in the eyes of the master. Scandinavian women were prized in the slave harems of many Islamic Sultans, but so were attractive Ethiopian and sub-Saharan women. Furthermore, in Muslim India and eighteenth-century England and France, dark-skinned slaves were the most favored, especially as young pages. In the New World, on the other hand, mulatto and other light-skinned female slaves were often better treated than their more African-looking counterparts.

Two other factors should be mentioned in considering the treatment of slaves: Laws and religion. Slightly more than a half of all slave societies on which data exist had some kind of slave laws, in some cases elaborate servile codes, the oldest known being those of ancient Mesopotamia. Slightly less than a half had none. Laws did make a difference; a much higher proportion of societies without any slave codes tended to treat slaves harshly. Nonetheless, the effectiveness of laws were mediated by other factors such as the relative size of the slave population, and religion.

The degree to which religion, especially Islam and Christianity, influenced the treatment of slaves is a controversial subject. Islam had explicit injunctions for the treatment of slaves, and these were sometimes influential, especially those relating to manumission. Although racism and strong preference for light complexion were found throughout the Islamic lands, it is nonetheless the case that Islam as a creed has been more assimilative and universalist than any other of the world religions, and has rarely been implicated in egregiously racist movements similar to those that have tarnished the history of Christianity such as apartheid, the Ku Klux Klan, Southern Christianity during the era of segregation, and the ethnic cleansing of Eastern Europe. However, Islam never developed any movement for abolition, and, in general, strongly supported the institution of slavery, especially as a means of winning converts.

For most of its history up to the end of the eighteenth century, Christianity simply took slavery for granted, had little to say about the treatment of slaves, and generally urged slaves to obey their masters. This changed radically with the rise of evangelical Christianity in the late eighteenth and early nineteenth centuries, during which it played a critical role in the abolition of both the slave trade and of slavery in Europe and the Americas. Throughout the world, Christianity appealed strongly to slaves and ex-slave populations, and in the Americas their descendants are among the most devout Christians. While Christianity may have had conservative influences on converted slaves, it is also the case that nearly all the major slave revolts from the late eighteenth century were influenced strongly by rebel leaders, such as Daddy Sharp in Jamaica and Nat Turner in America, who interpreted the faith in radical terms, or by leaders of syncretic Afro-Christian religions.

5. Manumission As An Integral Element Of Slavery

With a few notable exceptions, manumission, the release from slavery, was an integral and necessary element of the institution wherever it became important. The reason for this is that it solved the incentive problem implicit in slavery. The promise of redemption proved to be the most important way of motivating the slaves to work assiduously on their masters’ behalf.

Slaves were manumitted in a wide variety of ways, the most important being: Self-purchase or purchase by others, usually free relatives; through the postmortem or testamentary mode; through cohabitation or sexual relations; through adoption; through political means or by the state; and by various ritual or sacral means. Self-purchase or purchase by relatives or friends of the slave into freedom was by far the most important means in the advanced slave economies of Greece and Rome, and in the modern capitalistic slave regimes. However, it was not the most widespread in other kinds of slave systems, and in parts of the world such as Africa it was uncommon. Post-mortem manumission by wills and other means was common in Islamic lands and in many parts of Africa. This form of manumission was usually intimately linked to religious practices and with expectations of religious returns for what was often defined as an act of piety.

As indicated earlier, in many slave societies slave concubines sometimes gained their freedom for themselves and their children from the master. Manumission by the state for acts of heroism or for military action was an important, though episodic, form of manumission not only in the ancient world, but in many New World slave societies. Thousands of slaves gained their freedom in this way, not only in Latin America, but also in North America during the American war of independence and in wars against Spain in the southern USA. Manumission by adoption was unusual, but in certain societies, such as ancient Rome, it constituted the most complete form of release from slavery.

Slaves were sometimes manumitted for ritual or religious reasons, or on special celebratory occasions. Although thousands of slaves were manumitted at Delphi, ostensibly by being sold to Apollo, such manumissions had become merely legal formalism by the second century BC, although the practice may have harked back to an earlier era when they were genuinely religious in character. In other societies, religious or ritual manumissions were often substitutes for earlier practices in which slaves were sacrificed. Since manumission meant the negation of slavery, for many peoples, freeing slaves was symbolically identical to killing them. Such practices were common in some parts of Africa and the Pacific islands, and among some indigenous tribes of the Northwest Coast of America slaves were either killed or given away in potlatch ceremonies. There is an extension of this primitive symbolic logic in Christianity, where Christ’s sacrificial death is interpreted as a substitute for the redemption of mankind from enslavement to sin and eternal death, ‘redemption’ (from Latin redemptio) literally meaning ‘to purchase someone out of slavery.’

In all slave societies, certain categories of slave were more likely to be manumitted than others. The most important factors explaining the incidence of manumission were: Gender, the status of parents, age, skill, residence and location, the means of acquisition and, where relevant, skin color. These factors are similar to those influencing the treatment of slaves and will not be discussed further.

They were also important in explaining varying rates between societies. Thus societies with relatively higher proportions of skilled slaves, a greater location of slaves in urban areas, higher ratios of female to male slaves, and higher rates of concubinage between masters and slaves were more likely to have higher rates of manumission than those with lower levels of these attributes. Added to this is another critical variable: The availability of slaves, either internally or externally, to replace manumitted slaves. As long as such sources existed and the replacement value of the manumitted slave was less than the price of manumission, it suited slave-owners to manumit slaves, especially when the manumitted slaves were nearing, or had already reached, the end of their useful life.

However, on rare occasions the supply of slaves was cut off when demand remained high or was on the increase. This always resulted in very low rates of manumission. The most striking such case in the history of slavery was the US South, where the rise of cotton-based, capitalistic slavery in the early nineteenth century came within a few years of the termination of the Atlantic slave trade to America. Planters responded by reducing the manumission rate to near zero. A similar situation, though not as extreme, developed in the Spanish islands of the Caribbean during the nineteenth century, when the plantation system developed in the context of British enforcement of the abolition of the slave trade in the region. The result was that previously high rates of manumission plunged to very low rates.

A final point concerns the status of freed persons. This varied considerably across slave societies and bore little relation to the rate of manumission. Thus manumission rates were high in the Dutch colonies of Curacao and in nineteenth-century Louisiana, but the condition of freedmen was wretched. Conversely, in the British Caribbean, where manumission rates were low, the condition of freedmen was relatively good, some groups achieving full civil liberties before the end of slavery. The main factor explaining the difference in treatment was the availability of economic opportunities for freedmen, and the extent to which the dominant planter class needed them as allies against the slaves. In the Caribbean, the small proportion of slaveholders and Europeans and the existence of a vast and rebellious slave population gave much political leverage to the small but important freed population. No such conditions existed in the United States, where the free and white population greatly outnumbered the slaves, and conditions for rebellion were severely restricted. In the ancient world, ethnic barriers meant generally low status and few opportunities for the manumitted in the Greek states, in contrast with Rome, where cultural and economic factors favored the growth and prosperity of a large freedmen class, a class that eventually came to dominate Rome demographically and culturally, with major implications for Western civilization.

Bibliography:

1. Blackburn R 1997 The Making of New World Slavery. Verso, London
2. Cohen D W, Greene J P (eds.) 1972 Neither Slave Nor Free. Johns Hopkins University Press, Baltimore, MD
3. Davis D B 1966 The Problem of Slavery in Western Culture. Cornell University Press, Ithaca, NY
4. Drescher S, Engerman S L (eds.) 1998 A Historical Guide to World Slavery. Oxford University Press, New York
5. Engerman S 1973 Some considerations relating to property rights in man. Journal of Economic History 33: 43–65
6. Engerman S L (ed.) 1999 Terms of Labor: Slavery, Serfdom, and Free Labor. Stanford University Press, Stanford, CA
7. Engerman S, Genovese E (eds.) 1975 Race and Slavery in the Western Hemisphere. Princeton University Press, Princeton, NJ
8. Eltis D, Richardson D (eds.) 1997 Routes to Slavery. Frank Cass, London
9. Findlay R 1975 Slavery, incentives and manumission: A theoretical model. The Journal of Political Economy 83(5): 923–34
10. Finley M I (ed.) 1960 Slavery in Classical Antiquity: Views and Controversies. Heffer, Cambridge, UK
11. Fogel R W 1989 Without Consent or Contract: The Rise and Fall of American Slavery. Norton, New York
12. Garlan Y 1995 Les Esclaves en Greece Ancienne. Editions de la Decouvert, Paris
13. Kirschenbaum A 1987 Sons, Slaves and Freedmen in Roman Commerce. Catholic University Press, Washington, DC
14. Landers J (ed.) 1996 Against the odds. Slavery and Abolition, Special issue 17(1)
15. League of Nations 1938 Report to the League of Nations Advisory Committee of Experts on Slavery, Vol. 6. League of Nations, Geneva, Switzerland
16. Lovejoy P E 2000 Transformations in Slavery: A History of Slavery in Africa. Cambridge University Press, New York
17. Meillassoux C 1991 The Anthropology of Slavery: The Womb of Iron and Gold, trans. A Dasnois. Athlone, London
18. Miers S, Kopytoff I (eds.) 1977 Slavery in Africa. University of Wisconsin Press, Madison, WI
19. Miller J C 1993 Slavery and Slaving in World History: A Bibliography:, 1900–1991. Kraus International, Millwood, NY
20. Patterson O 1967 The Sociology of Slavery: Jamaica, 1655–1838. McGibbon & Kee, London
21. Patterson O 1982 Slavery and Social Death. Harvard University Press, Cambridge, MA
22. de Queiros Mattoso K M 1986 To Be a Slave in Brazil, 1550–1888, trans. A Goldhammer. Rutgers University Press, New Brunswick, NJ
23. Reid A (ed.) 1983 Slavery, Bondage and Dependency in Southeast Asia. University of Queensland Press, St. Lucia, Queensland
24. Rodriguez J P (ed.) 1977 The Historical Encyclopedia of World Slavery. ABC-CLIO, Santa Barbara, CA
25. Shepherd V, Beckles H (eds.) 2000 Caribbean Slavery in the Atlantic World. Ian Randle Publishers, Kingston, Jamaica
26. Watson J (ed.) 1980 Asian and African Systems of Slavery. Blackwell, Oxford, UK