SLAVERY AND INVOLUNTARY SERVITUDE

Many observers view slavery and freedom as polar opposites, but both slave and free wage labor systems rely on compulsion. Slave systems depend ultimately on physical coercion to force slaves to work for masters, although cultural, ideological, and economic pressures typically augment physical force. Wage labor systems, by contrast, depend on workers being free "in the double sense" (Marx [1867] 1967, pp. 168–169): Not only must workers be free to seek employment and choose among potential employers, they also must be free of all other means of subsistence that would allow voluntary withdrawal from the labor market. In the absence of subsistence alternatives, economic necessity compels "free" workers to exchange labor services for wages. Although wage labor systems depend primarily on labor-market processes to supply employers with workers, physical coercion often supplements those processes, especially during periods of economic decline. Cultural expectations and ideological appeals also reinforce market mechanisms. Nevertheless, large-scale labor systems are maintained primarily by a mixture of physical and economic coercion that varies with the availability of subsistence alternatives.

The way in which the constellation of physical and economic coercion and subsistence alternatives is determined by the power of contending groups as well as historically specific cultural and ideological factors has been of great interest to social scientists. Perhaps the simplest and most durable statement of the causes of slavery is a conjecture known as the Nieboer–Domar hypothesis (Nieboer 1900; Domar 1970; Engerman 1986a; see Patterson 1977b for a critique), which links slavery to an abundance of arable land combined with a shortage of labor. The way in which slavery differs from other forms of involuntary servitude is explained in the next section. The Nieboer–Domar hypothesis is then amended to provide a provisional explanation for the worldwide trend away from slavery and toward freedom in large-scale labor systems over the last several hundred years. Finally, the Nieboer–Domar hypothesis is reevaluated in light of current patterns of slavery and involuntary servitude around the world.

SLAVERY AND OTHER FORMS OF INVOLUNTARY SERVITUDE

Patterson (1982, p. 13) argues that slavery is defined by three conditions. First, slaves suffer perpetual domination that ultimately is enforced by violence. The permanent subjugation of slaves is predicated on the capacity of masters to coerce them physically. Second, slaves suffer natal alienation, or the severance of all family ties and the nullification of all claims of birth. They inherit no protection or privilege from their ancestors, and they cannot convey protection or privilege to their descendants. Third, slaves are denied honor, whereas masters are socially exalted. This condition appears to be derivative rather than definitive of slavery because all hierarchical social systems develop legitimating ideologies that elevate elites and denigrate those at lower levels. The first two conditions, which distinguish slavery from other forms of involuntary servitude, constitute the working definition used in this article.

In chattel slave systems, slaves are movable property owned by masters and exchanged through market processes. Because some societies constructed elaborate slave systems without well-developed notions of property and property rights, property relationships cannot be an essential defining element of slavery (Patterson 1982; 1977a). Nevertheless, property relations and economic processes had important effects on slavery and other forms of unfree labor in the Americas, Europe, and Africa in the period after the fifteenth century, which is the major focus of this analysis.

An unfree laborer cannot voluntarily terminate service to a master once the servile relationship has been established. Slavery maximizes the subordination of servant to master. Other servile workers, such as indentured and contract laborers, debt servants, peons, and pawns, are less dominated than slaves are and do not suffer natal alienation. Pawns, for example, were offered by their families in return for loans. Pawns maintained kinship ties to their original families, a situation which gave them some protection, and were freed once the loans were repaid. Indentured servants agreed to be bound to a master for a specific term, such as seven years, in exchange for a benefit such as passage to America or release from prison (Morris 1946; Smith 1947; Morgan 1975). Contract laborers also were bound for specified terms but could not be sold against their will to other masters, as was the case with indentured servants. Debt servitude consists of labor service obligations that are not reduced by the amount of work performed (Morris 1946; Sawyer 1986). Peons are tied to land as debt servants and owe labor services to a landlord. Serfs are not debt servants, but they are tied to land and perform labor services on their lords' estates. The right to labor services enjoyed by European feudal lords was vested in their political authority rather than in land ownership, although serfs were reduced to slaves in all but name in some instances (e.g., Russia in the nineteenth century) (Kolchin 1987).

Indentured servants and contract laborers may agree to the initial terms of their servitude, but they cannot willingly end it during its term once it begins. Usually some form of coercion, such as poverty, debt, or impending imprisonment, was necessary to force people to agree to terms of contractual servitude or pawnship. By contrast, the status of the slave, serf, peon, and debt servant typically was inherited or imposed on workers against their will.

SLAVERY, THE LAND–LABOR RATIO, AND THE STATE

In its simplest form, the Nieboer–Domar hypothesis states that abundant free land makes it impossible for free workers and nonworking landowners to coexist. If free land is available and laborers can desert landowners whenever they choose, landowners will be unable to keep enough workers to maintain their status as nonworkers. If landlords can compel workers to perform labor services despite the availability of free land, landlords become labor lords and workers are not free. By contrast, scarce land combined with an abundant labor supply drives wages down, making wage laborers less expensive than slaves and other servile workers. When they are denied access to land, hunger forces workers to labor for wages and wage labor systems displace slave labor systems.

This model appears to be deficient in at least four ways. First, as Domar recognized, political factors determine the degree of freedom enjoyed by workers. Chief among those factors is the extent to which the state protects the interests of landowners when they conflict with those of laborers. Large-scale slave labor systems cannot exist without states that defend the power of slave masters to control and utilize the labor of slaves. A powerful state is essential for protecting slave masters against slave rebellions, capturing runaways, and enforcing slave discipline. State power is required for the enslavement of new supplies of slaves. If the state is responsive to the demands of workers or if workers can voluntarily withdraw their labor services, unfree labor systems cannot be maintained.

Second, the model presumes that slave masters exploit slaves in response to economic incentives, but slaves and other unfree laborers often provided military, administrative, domestic, and sexual services largely unrelated to economic activities (Roberts and Miers 1988; Patterson 1982). The Nieboer–Domar hypothesis therefore does not apply to societies that employ slaves and other servile workers in noncommercial or minor economic roles (Lovejoy 1983; Finley 1968). It also does not apply to states that use race, religion, gender, or other status criteria to restrict the freedom of workers for noneconomic reasons ( James 1988).

Third, the key issue from an employer's perspective is not simply the ratio of land to labor but the relative costs and benefits of different forms of labor that can be profitably employed using existing capital (including land). A more general version of the Nieboer–Domar model compares the stock of available capital to the availability of different forms of labor at prevailing prices. Thus, labor scarcity means the scarcity of labor at prices that allow it to be employed profitably.

Fourth, the simple version of the Nieboer–Domar hypothesis ignores the organizational capacities of workers and capitalists' ability to adopt labor-saving innovations. If workers demand concessions that threaten profits or engage in strikes and other production disruptions, capitalists experience "labor shortages" that stem not from insufficient numbers but from the organized resistance of the workers who are present (Miles 1987). Faced with such disruptions, capitalists with sufficient capital may adopt labor-saving innovations if they are available. When capitalists are unable to adopt those innovations, they may resort to coercive strategies to curb workers' market-based demands (Paige 1975). This case contradicts the Nieboer–Domar hypothesis, which assumes that high ratios of labor to capital (or land) make coercive labor control strategies unnecessary.

UNFREE LABOR IN THE AMERICAS

From the fifteenth through the nineteenth centuries, Europe, Africa, and the Americas were closely linked by flows of people and commodities (Lovejoy 1983; Eltis 1987). The colonization of the Americas by strong European states provided vast, lightly populated lands for commercial exploitation. Expanding markets in Europe for sugar, cotton, tobacco, coffee, and other commodities stimulated the demand for greater supplies of servile labor to work the plantations and mines of the Americas. Weak states in large areas of sub-Saharan Africa left large populations vulnerable to armed predation by stronger states that supplied the expanding markets for slaves.

Estimates of the numbers of bondsmen and slaves transported to the Americas are subject to sizable errors because of the paucity and unreliability of existing records, but relative magnitudes are thought to be reasonable (see Table 1). Differences in the sources of servile labor produced different racial compositions across American regions. Slaves from Africa outnumbered arrivals from Europe nearly four to one before 1820, and most were bound for sugarcane plantations in Brazil and the West Indies. British North America was atypical because its early immigrants were predominantly white indentured servants from Britain, Ireland, and Germany; perhaps two-thirds of the white immigrants who arrived before the American Revolution were bonded servants (Smith 1947, p. 336). Before being displaced by African slaves, white bondsmen were the principal source of labor in the plantation regions of all British colonies, including those in the Caribbean (Engerman 1986a; Galenson 1981).

Indentured servitude was the principal method of defraying the costs of supplying the colonies with workers. British laws and customs regulating master–servant relationships were modified significantly to fit American circumstances (Galenson 1981). Because of the high costs of transatlantic passage, longer periods of service were required, typically four to seven years rather than one year or less in England. English servants could not be sold against their will to another master, but that practice was sanctioned in colonial laws and customs because European servants could not negotiate terms with perspective masters before immigrating to America. Finally, opportunities for escape were much greater in America. Consequently, elaborate state enforcement mechanisms were implemented to discourage runaways and to catch, punish, and return those who did. Most indentured servants were transported to plantation regions

because plantation labor produced greater returns than did any other economic activity in the Americas (Galenson 1981). Employers in areas such as New England could afford few or no servants because they specialized in trades with lower labor productivity and lower profit margins.

White servile labor was replaced by black slavery throughout the Americas between 1600 and 1800. Racial prejudice encouraged the shift but probably was not decisive (Morgan 1975). First, the limited supply of indentured servants could not satisfy the demand for servile labor, whereas the supply of African slaves was almost completely elastic. Improving economic conditions in Britain and state restrictions on the emigration of British servants reduced the numbers seeking passage to America, causing the price of servants to increase. As the price of servants exceeded the price of slaves, first for unskilled and later for skilled workers, slaves came to be preferred to bonded servants (Galenson 1981). Second, Africans were more resistant to the diseases of the tropics, where the most important export crops were grown (Eltis 1983).

Third, slaves could be compelled to comply with the labor-intensive plantation work regime that developed (Fogel 1989). Slaves were more efficient and profitable than free or indentured workers in sugar, cotton, coffee, rice, and tobacco agriculture because the work required by those crops could be performed efficiently by slave work gangs. Work gangs were organized according to specialized tasks, and slaves were assigned to particular gangs according to their skills and capacities. The work was performed under close supervision to maintain work intensity and quality. Slave masters often used brutal violence to enforce discipline, but naked force might have been used less than once was thought. Slave masters experimented with different mixtures of positive and negative incentives, to encourage slaves to maximize their output (Fogel 1989). Thus, slave plantations anticipated the discipline of workers in the great factories of industrial capitalism, where assembly lines regulate the rhythms and intensity of work.

Forced migration from Africa greatly exceeded all migration from Europe as sugar production became the greatest consumer of servile labor in the Americas. High death rates and a preference for male slaves in the sugar-producing regions led to net population declines among blacks and mulattoes (compare immigration numbers to population sizes in Table 1), but the proportion of blacks in the British West Indies increased from 25 to 91 percent between 1650 and 1770 (Fogel 1989, p.
30). By the 1820s, the proportion of blacks and mulattoes in Brazil, the Guyanas, and the West Indies reached 75 percent (Table 1).

British North America was an exception to this pattern as both black and white populations had high rates of natural increase. Almost all major slave societies were unable to maintain the size of slave populations without continuous replenishment from outside sources. By contrast, the slave population in the United States multiplied because of unusually high fertility rates and low mortality rates (see Table 1 and Fogel 1989).

Political factors also encouraged the transition from white servitude to black slavery (Engerman 1986b; Galenson 1981). As British citizens, indentured servants retained state-protected natal rights that their masters were obliged to respect. For example, masters could beat servants and slaves to enforce work discipline, but colonial courts protected servants against unfair punishment (Smith 1947). Importantly, Europeans could choose the place of their servitude, and most refused transportation to the plantation regions from the eighteenth century on. African slaves could not avoid the plantation regions and were citizens of no state in Africa or America that could or would defend their interests.

Because Spain conquered the continental regions with the largest Native American populations (Table 1), it had less need of African slaves. Instead, Spanish colonists installed a coercive labor system patterned on Spanish feudalism that forced natives to work part-time on colonial estates although slavery was still preferred in the mines (Slicher Van Bath 1986; Kloosterboer 1960). Unfree labor markets and compulsory labor endured for 400 years, eventually evolving into debt servitude in the nineteenth century. Native Americans and mestizos accounted for nearly 80 percent of the population of continental Spanish America by 1825 but were almost annihilated in the West Indies (Table 1).

Nowhere in the Americas was slavery in danger of withering away economically at the time when it was abolished (Eltis 1987). Furthermore, with the principal exception of Haiti in 1804, slave rebellions were not successful in conquering slave masters and transforming a slave system into a wage labor system. Paradoxically, Britain played the dominant role in abolishing slavery and the transatlantic slave trade even though it controlled half the transatlantic commerce in slaves and half of the world's exports in sugar and coffee, which were produced primarily on slave plantations (Eltis 1987). Britain outlawed the slave trade in 1808 and freed the slaves in its West Indian colonies in 1833 over the strenuous objections of slave owners. The United States prohibited the importation of slaves after 1808, and civil war led to abolition in 1865. By the 1870s, all the major European and American maritime and commercial powers had acquiesced to British pressure and outlawed the slave trade. Brazil, the last state in the Americas to abolish slavery, did so in 1888.

The land–labor ratio strongly affected planters' responses to abolition. In places where exslaves could find no alternative to plantation work, such as Barbados and Antigua, the transition to free labor was rapid, and plantation production did not decline appreciably (Boogaart and Emmer 1986). In places where land or alternative employment was available, such as Jamaica and Trinidad, the ex-slaves abandoned the plantations, and plantation productivity declined (Engerman 1985). In response, planters implemented a variety of servile labor systems with mixed results. A second wave of indentured servants was imported chiefly from Asia, especially China and India, which more than compensated for the labor shortages induced by abolition in some cases, such as Mauritius and British Guiana (Engerman 1985, 1986b). China and colonial India eventually banned the recruitment of servants because of objections to employers' poor treatment of servants, and Brazil was never able to gain access to Asian indentured laborers (Boogaart and Emmer 1986).

In areas where planters retained a degree of political power, such as the West Indies and Brazil, vagrancy statutes and other compulsory labor schemes forced workers to accept wages below free market levels (Kloosterboer 1960; Huggins 1985). Indentured labor and other forms of involuntary servitude were banned in the United States in 1865 by the Thirteenth Amendment to the U.S. Constitution, but planters regained substantial influence over black workers through their control of racially discriminatory state and local governmental institutions ( James 1988). Blacks were disfranchised by 1900, making them vulnerable to racial segregation, physical coercion, and economic discrimination. The extent to which racial discrimination interfered with free labor markets in the South is controversial (Wright 1986). Nevertheless, the most determined resistance to the civil rights movement of the 1960s occurred in the plantation regions ( James 1988). The success of that movement led to increased protection of the citizenship rights of blacks and doomed widespread coercive labor control practices. The transition to capital-intensive agricultural practices was rapid during that period.

UNFREE LABOR IN AFRICA AND ASIA

Slavery was an indigenous institution in Africa and Arabia for centuries before Europeans entered the African slave trade (Thornton 1998). While approximately 9.9 million Africans were transported to the Americas before the Atlantic slave trade was suppressed (Fogel 1989), an additional
5.2 million African slaves were transported across the Sahara, the Red Sea, and the Indian Ocean into the Islamic world between 1500 and 1900. Moreover, perhaps 6.4 million more were exported to Islamic societies between a.d. 650 and 1500 ("a rough approximation," Lovejoy 1983, p. 24). Many thousands more were enslaved in African societies in that period (Thornton 1998).

Whereas chattel slavery in the Americas was predicated on profit making, African slavery typically did not have a narrowly economic basis. African slaves were menial servants and field workers, but they also were concubines, surrogate kin, soldiers, commercial agents, and candidates for human sacrifice (Roberts and Miers 1988, p. 5). Female slaves were especially valued because women performed most agricultural and domestic work. African societies were based on kinship relations in which all individuals were linked in a complex network of dependency. Because power in kinship systems depends on the size of social groups, slave masters could increase their power by obtaining more slaves. Furthermore, slaves were immune to the appeals of their masters' rivals within kin groups because they had no kinship ties that mediated their subordination to their masters. Large numbers of persons were enslaved as a result of military victories in wars between African kingdoms and societies.

African slave masters also responded to economic incentives. An increasing number of slaves were provided to the Atlantic slave trade as the demand for slaves in the Americas increased. Thornton (1998, p. 125) concludes that African participation in the slave trade was voluntary because European slavers did not have the economic or political power to force African leaders to sell slaves. The established African practices of holding and trading slaves made it possible for African states to respond to the increasing European demand so long as the prices paid were attractive.

Islamic slavery also differed from chattel slavery in important ways. Islamic law prohibited the enslavement of Muslims but permitted the enslavement of people born to slave parents or captured for the purpose of conversion to Islam (Gordon 1989). Concubines could not be sold if they bore a child to a master, and the child could not be enslaved. Allowing slaves to purchase their freedom brought honor to former masters. Manumitting slaves was also meritorious and could atone for certain sins and offenses.

Islamic slaves typically were employed as household servants, domestic workers, concubines, and to a lesser extent soldiers. Female slaves typically brought higher prices than did males because the heads of patriarchal Muslim families prized female slaves for assignment to sexual and domestic roles in their households. Slave eunuchs performed special tasks in large households and usually brought higher prices than did female slaves. Consequently, pre-twentieth-century slave traders castrated large numbers of African slave boys in crude operations that killed up to 90 percent of them (Gordon 1989, pp. 91–97). However, Islamic slave masters also responded to economic incentives as did their American counterparts when market opportunities arose. During the nineteenth century, over 750,000 slaves were transported to the clove plantations on Zanzibar and other locations on the east coast of Africa, for example (Cooper 1977; Lovejoy 1983, p. 151).

British diplomatic and military pressure finally led to the suppression of the Islamic and African slave trades as it did with the transatlantic traffic. In 1890, all the European powers agreed to suppress slave trading and slave raiding and to assist ex-slaves, a commitment that legitimated the conquest of Africa in the eyes of European citizens. However, European colonial administrators were reluctant abolitionists (Roberts and Miers 1988). Inadequate military and administrative power, fear of economic and political disruptions, and unfamiliarity with African customs delayed the process.

Colonial governments outlawed slavery almost everywhere in sub-Saharan Africa by the 1930s, but involuntary servitude persisted. Roberts and Miers (1989, pp. 42–47) identify three factors that retarded the emergence of free labor markets in Africa. The first two were responses to abundant land and scarce labor. First, colonial states conscripted natives, imposed labor levies that local chiefs had to fill, and implemented other compulsory labor mechanisms to maintain a supply of cheap labor for European employers and administrators. Second, many Africans had access to land or livestock and were unwilling to work for wages. Colonial states tried to reduce the attractiveness of nonwage occupations by, for example, raising taxes above what peasant agriculturalists and pastoralists could pay and prohibiting Africans from growing lucrative cash crops. In settler colonies such as South Africa, native Africans were pushed off the land and confined to strictly regulated labor markets by pass laws. Third, Africans resorted widely to pawnship after abolition.

The reluctance of colonial administrators and the power of postcolonial states allowed slavery to survive in some nations in north Africa and the Arabian peninsula well into the twentieth century. In 1926, the League of Nations codified its opposition to slavery by adopting the Convention to Suppress the Slave Trade and Slavery, which defined slavery as the ownership of another person. Gradually, the remaining slave states abolished slavery officially: Ethiopia in 1942, Saudi Arabia in 1962, Muscat and Oman in 1970, and Mauritania for the third time in 1980. Nevertheless, reports of slavery persisted. Saudi Arabia allegedly failed to free some 250,000 slaves in the late 1960s; an estimated 100,000 chattel slaves existed in Saharan regions of Mauritania in 1980, although many were freed by 1984; and nomadic tribesmen allegedly held 250,000 slaves in the Sahelian districts of Mali in 1984 (Gordon 1989, pp. 232–234; United Nations 1984, pp. 18–19; United Nations 1988, p. 197; Sawyer 1986, p. 14).

In 1956, the United Nations increased the international attack on slavery and involuntary servitude by adopting the Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery. In addition to outlawing slavery, the Supplementary Convention pledged signatory nations to suppress debt bondage, serfdom, the pawning of children, and servile marriage (forcing women to marry in exchange for payments to their family members or assigning wives, after the death of their husbands, to others as an inheritance). Progress has been slow. For example, India outlawed bondage in 1976, but a survey found more than 2.5 million bonded workers in 1978; only 163,000 had been freed by 1985 (Sawyer 1986, pp. 124–134). Debt servitude has been reported since 1970 among landless peasants in India and Nepal and among Native American rubber collectors in the Peruvian Amazon. As late as 1986, the Dominican Republic used its army to round up Haitian immigrants for forced work on sugar plantations during the harvest season (Plant 1987).

PATTERNS OF SLAVERY AND UNFREE LABOR SINCE 1990

Large-scale systems of slavery and involuntary servitude can be maintained only if slave owners and labor lords can use physical coercion to maintain labor discipline. Hence, large-scale systems of unfree labor depend on state institutions that deny citizenship rights to unfree workers and augment the power of dominant classes to coerce their workers physically. Today no nation officially protects the rights of employers to reduce their workers to slavery or involuntary servitude. Virtually all members of the United Nations have ratified the Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery (see United Nations [1957] 1999 for the current list of ratifying nations). Ratification of the Supplementary Convention officially commits a nation to the elimination of slavery and involuntary servitude within its borders and obligates it to cooperate with other nations in suppressing those practices. Although it took 200 years, the international antislavery campaigns led by politically powerful nations with wage labor markets were successful. Large-scale systems of slavery and involuntary servitude supported and protected by complementary state institutions no longer exist.

The expansion of capitalism and increasing world population displaced large numbers of people from subsistence agriculture and other means of support in many regions. Great disparities between rich and poor nations drive people across state boundaries in search of jobs and improved living conditions. State power plays a crucial role in shaping migration and molding the relationship between capital and labor, but states with expanding economies now prevent the entrance of many willing workers rather than compelling the entrance of the unwilling. The whip of unemployment and poverty replaces the slave master's lash as free labor replaces slave labor.

However, slavery and forced labor persist and are widespread in some areas. Anti-Slavery International, the world's oldest human rights organization, estimates that over 200 million people, about 3 percent of the world's population, labor in some form of bondage. Table 2 provides examples of some existing systems of unfree labor. Because reliable information is difficult or impossible to obtain in some cases, the examples in Table 2 should not be considered exhaustive or the most egregious. The best available information suggests that slavery and involuntary servitude occur with the greatest frequency in nations that are ravaged by civil war or have weak states that are unwilling or unable to suppress coercive labor practices.

Somalia, Sudan, and Uganda provide examples of how civil war places defenseless people at the mercy of powerful military groups. A United Nations special rapporteur confirmed that armed militia groups abducted people in southern Sudan for use as forced laborers or for sale as slaves. Prisoners were subjected to beatings, electric shock, exposure to the sun for long periods, pouring of cold water on the naked body, rape and the threat of rape, sleep deprivation, and the refusal of food and medical treatment. Sudanese government security forces and allied militias as well as insurgent groups were guilty of conscripting children and forcing them to fight as soldiers (United Nations 1997). The civil war in Sudan has disrupted agricultural production to the extent that some impoverished parents give or sell their children to others to prevent their starvation (Finnigan 1999).

A large and active chattel slavery market is in operation in Sudan. Its magnitude is not known,

but international human rights groups estimate that chattel slaves number in the tens of thousands and that the market may extend as far as Saudi Arabia and the Gulf states (Finnigan 1999, p. 71). Anti-Slavery International (1999a) reports that more than 2,700 slaves were freed during the first four months of 1999 in return for over $100,000 in payments. Finnigan (1999) photographed a Sudanese Arab slaver who hoped to sell over 130 individuals to a Christian antislavery organization for $50 each. If that plan failed, he decided that he might return the former slaves to their families for a much lower price. Many others were not so fortunate. Slaves in Sudan are subject to severe punishment; are stripped of their cultural, religious, and personal identities; and can become the property of another person for life, traded and inherited, branded and bred.

Some slavery and slavery-like practices in India, Benin, Mauritania, Morocco, and Pakistan are supported by local customs and traditions that have long histories (Table 2). Mauritania and India have attempted to eliminate these practices, whereas other states are more reluctant or unable to act decisively. Debt bondage is widespread in the rural areas of India, but involuntary servitude has been adapted to the economic opportunities provided by the operation of global markets. Just as European demand for sugar and coffee drove the Atlantic slave trade in the seventeenth and eighteenth centuries, employers of servile labor in poor countries find opportunities to supply consumer demands for cheap goods in economically advanced countries. For example, India, Pakistan, and Nepal employed as many as one million servile child workers in the hand-knotted carpet industry in 1994 (United States Department of Labor 1995). The United States imported $329 million worth of hand-knotted carpets in 1996, a large proportion of which were produced in those countries (United States Department of Labor 1997). Children in India have been kidnapped and transported from their villages to face years of forced labor in the carpet industry. Forced child labor is characterized by long hours, threats of violence, dangerous conditions, little or no pay, and poor or nonexistent health care (United States Department of Labor 1994, 1998; Anti-Slavery International 1999b). Other commercial products produced by bonded and forced child workers for export are brassware, silk cloth and silk garments, and stone and glass products (United States Department of Labor 1997).

Thailand, Nepal, Indonesia, and Cambodia are examples of weak states or states crippled by corruption that facilitate the persistence of forced labor and forced sexual prostitution. The flourishing Asian sex trade involves the transport of women and children across borders to work in brothels for foreign and domestic customers. Victims are lured by false promises of decent employment, kidnapped or sold by family members, or reduced to debt bondage by poverty. In Thailand, debt bondage sometimes continues from generation to generation as a result of very low wages, high interest charges, and fraudulent debts that cannot be repaid (U.S. Department of Labor 1994, 1996, 1998). Because of the spread of AIDS, brothel customers increasingly prefer very young girls who are supposedly disease-free. Consequently, brothel owners purchase or kidnap young girls from the surrounding countries to supply the demand. Girls from Burma, Cambodia, China, Laos, and Vietnam can be found in the brothels of Thailand (United States Department of Labor 1995).

The Asian sex trade is huge. UNICEF estimates that 100,000 child prostitutes are employed in Thailand alone, but the number could be more than 200,000 (U.S. Department of Labor 1995). Patterns vary from country to country, but the brothels typically are patronized by local customers and to a lesser extent by foreigners, including tourists, businessmen, and military personnel from the United States and Europe. Travel agencies arrange sex tours that include accommodations and a choice of escorts (United States Department of Labor 1995, 1996). As is the case with many handmade and manufactured products, the Asian sex trade and the exploitation of servile workers are intimately connected to the global economy.

In contrast to Britain's use of the navy to suppress the Atlantic slave trade in the nineteenth century, current international efforts to suppress slavery and involuntary servitude are weak. One enforcement tactic is to expose governments that do not suppress servile labor practices to the condemnation of world opinion. For example, the United Nations Commission on Human Rights investigates patterns of human rights violations within countries, including slavery and servitude, and disseminates its reports widely. The International Labor Organization, a special agency of the United Nations, formulates international labor standards and monitors the compliance efforts of governments (United Nations 1991). A wide variety of nongovernmental organizations are involved in the publicity campaigns against human bondage. Perhaps the most famous is Anti-Slavery International, which promotes the eradication of slavery and slavery-like practices by supporting the victims of those practices and through the collection and dissemination of information on specific cases. The negative publicity created by the public exposure of servile labor practices can diminish servile labor practices, but governments often do not cooperate because negative publicity alone is a feeble enforcement tool.

Economic pressure may be more effective than the investigative reporting and publicity efforts of the United Nations and other governmental and nongovernmental organizations. For example, the annual reports of the U.S. Department of State on Human Rights Practices are provided to the U.S. Congress to assist it in formulating foreign aid policies (Table 2 is based on the 1999 report). Countries are encouraged to suppress slavery and unfree labor practices to continue receiving aid from the United States and international organizations. Pressure from consumer organizations is effective in some cases. For example, a number of product-labeling programs were created to reassure buyers that products imported into affluent countries were not made by children. A nongovernmental organization in India initiated the RUGMARK program in 1994, which encourages carpet manufactures to stop using child labor by providing a labeling service that certifies that products are not made by children (United States Department of Labor 1997). By mid-1997, RUGMARK inspections had found over 1,000 illegal child workers in the carpet industry. Other labeling programs cover leather footwear, soccer balls, and the tea industry (United States Department of Labor 1997).

The historical decline in the land–labor ratio did not produce the abolition of slavery and involuntary servitude. The simple version of the Nieboer–Domar hypothesis is inadequate. From the middle of the nineteenth century until the present, political factors have played a decisive role in breaking the link between the availability of land and unfree labor. Rather than defending slavery when land was plentiful, Britain used its political power to abolish slavery and suppress the slave trade even though it was not in its economic interest to do so.

By the late twentieth century, most nations had officially prohibited slavery and slavery-like practices, but both persist. As the modified Niebor–Domar hypothesis predicts, some employers will use physical coercion to drive down the cost of labor when opportunities arise. Weak or corrupt state institutions cannot or will not defend the rights of those who are most vulnerable to coercion. Civil wars destroy the ability of states to maintain order and subject citizens to the depredations of warring militias. In some areas of the world, cultural support for servile labor and grinding poverty combine to make it difficult to eliminate some forms of slavery and forced labor. In all cases, the typical victims of these practices are the weakest and most vulnerable groups: women, children, migrant workers, low-status class or caste groups, and racial, ethnic, and religious minorities (Table 2).

All nations regulate the passage of individuals across their borders and assign superior rights and privileges to citizens compared to noncitizens. In advanced capitalist democracies with ostensibly free labor markets, the state-enforced distinction between citizen and noncitizen is a key mechanism in maintaining dual labor markets that disproportionately relegate noncitizens to the lowest-paying jobs (e.g., Thomas 1985; Miles 1987; Cohen 1987). Typically, noncitizen "guest workers" are less likely to enjoy state protection and more vulnerable to discrimination. Because the demand for cheap labor often can be satisfied by choosing among citizens and noncitizens who have no other labor market alternatives, democratic states can regulate noncitizens' access to domestic labor markets rather than forcibly import unfree workers from foreign lands.

However, many states are not liberal democracies. Thousands were confined for political reasons in forced labor camps during the Stalin era in the Soviet Union. Nazi Germany forced Jews and other minorities into slavery where they were to be "worked to death" (Sawyer 1986). Blacks were disfranchised and rigidly segregated in the southern United States for much of the twentieth century, making them vulnerable to coercive labor practices. The Republic of South Africa's now abolished policy of apartheid denied citizenship status to indigenous blacks and exposed them to forced labor practices. Since 1988, the military government of Burma has engaged in systematic human rights abuses, including the imposition of forced labor on large segments of the population for military purposes and for the construction and development of infrastructure (Bureau of International Labor Affairs 1998).

The international condemnation of slavery and involuntary servitude represents a great victory for those who support and defend human rights. The use of forced and other forms of servile labor has not been eliminated, but it has been widely branded as criminal activity. As a consequence, those who would employ servile labor must risk prosecution or search for opportunities in countries with weak or corrupt political institutions that cannot or will not suppress slavery and involuntary servitude. Although the struggle to eradicate servile labor practices has not been won, nations and international human rights organizations appear to be more concerned about the loss of life and other human rights violations that accompany civil wars and international conflicts. For example, NATO recently intervened to stop the murder and forced displacement of thousands of ethnic Albanians in Kosovo. A similar intervention into the affairs of anther nation to eliminate slavery or involuntary servitude is unlikely.

references

Anti-Slavery International 1999a "Slavery in Sudan—Appeal for Releases." News Release available from the Anti-Slavery International website (http://www.charitynet.org/˜asi/).

—— 1999b. "What Is Modern Slavery?" Available from the Anti-Slavery International website (http://www.charitynet.org/˜asi/)

Boogaart, Ernst Van Den, and P. C. Emmer 1986 "Colonialism and Migration: An Overview." In P. C. Emmer, ed., Colonialism and Migration. Dordrecht, Netherlands: Martinus Nijhoff.

Bureau of International Labor Affairs 1998 "Report on Labor Practices in Burma." Washington, D.C.: U.S. Department of Labor.

Cohen, Robin. 1987. The New Helots. Brookfield, Vt.: Gower.

Cooper, Frederick 1977 Plantation Slavery on the East Coast of Africa. New Haven, Conn.: Yale University Press.

Domar, Evesy D. 1970 "The Causes of Slavery or Serfdom: A Hypothesis." Journal of Economic History 30:18–31.

Effah, Josephine 1996 Modernised Slavery: Child Trade inNigeria. Lagos, Nigeria: Constitutional Rights Project.

Eltis, David 1983 "Free and Coerced Transatlantic Migrations: Some Comparisons." American HistoricalReview 88:251–280.

—— 1987 Economic Growth and the Ending of theTransatlantic Slave Trade. New York: Oxford University Press.

Engerman, Stanley L. 1985 "Economic Change and Contract Labour in the British Caribbean." In D. Richardson, ed., Abolition and Its Aftermath. London: Frank Cass.

—— 1986a "Slavery and Emancipation in Comparative Perspective: A Look at Some Recent Debates." Journal of Economic History 46:317–339.

—— 1986b "Servants to Slaves to Servants: Contract Labour and European Expansion." In P. C. Emmer, ed., Colonialism and Migration. Dordrecht, Netherlands: Martinus Nijhoff.

Finley, M. I. 1968 "Slavery." In David Sills, ed., International Encyclopedia of the Social Sciences, vol. 14, pp. 307–313. New York: Macmillan.

Finnigan, William 1999 "The Invisible War." The NewYorker, January 25, pp. 50–73.

Fogel, Robert W. 1989 Without Consent or Contract. New York: Norton.

Galenson, David W. 1981 White Servitude in ColonialAmerica. New York: Cambridge University Press.

—— and Stanley L. Engerman 1974 Time on the Cross:The Economics of Negro Slavery, 2 vols. Boston: Little, Brown.

Gordon, Murray 1989 Slavery in the Arab World. New York: New Amsterdam Books.

Huggins, Martha K. 1985 From Slavery to Vagrancy inBrazil. New Brunswick, N.J.: Rutgers University Press.

James, David R. 1988 "The Transformation of the Southern Racial State: Class and Race Determinants of Local-State Structures." American Sociological Review 53:191–208.

Kloosterboer, W. 1960. Involuntary Labour since the Abolition of Slavery. Leiden, Netherlands: E. J. Brill.

Kolchin, Peter 1987 Unfree Labor: American Slavery andRussian Serfdom. Cambridge, Mass.: Harvard University Press.

Lovejoy, Paul E. 1983 Transformations in Slavery. Cambridge, UK: Cambridge University Press.

Marx, Karl. (1867) 1967 Capital, vol. 1. New York: International Publishers.

Miles, Robert 1987 Capitalism and Unfree Labour. London: Tavistock.

Morgan, Edmund S. 1975 American Slavery, AmericanFreedom. New York: Norton.

Morris, Richard B. 1946 Government and Labor in EarlyAmerica. New York: Columbia University Press.

Nieboer, Herman J. 1900 Slavery as an Industrial System. The Hague, Netherlands: Martinus Nijhoff.

Paige, Jeffery 1975 Agrarian Revolution. New York: Free Press.

Patterson, Orlando. 1977a "Slavery." Annual Review ofSociology 3:407–449.

—— 1977b "The Structural Origins of Slavery: A Critique of the Nieboer Domar Hypothesis." Annualsof the New York Academy of Sciences 292:12–34.

—— 1982 Slavery and Social Death. Cambridge, Mass.: Harvard University Press.

Plant, Roger 1987 Sugar and Modern Slavery. London: Zed.

Roberts, Richard, and Suzanne Miers 1988 "The End of Slavery in Africa." In S. Miers and R. Roberts, eds., The End of Slavery in Africa. Madison: University of Wisconsin Press.

Sawyer, Roger 1986 Slavery in the Twentieth Century. London: Routledge and Kegan Paul.

Slicher Van Bath, B. H. 1986 "The Absence of White Contract Labour in Spanish America during the Colonial Period." In P. C. Emmer, ed., Colonialismand Migration. Dordrecht, Netherlands: Martinus Nijhoff.

Smith, Abbot Emerson 1947 Colonists in Bondage. Chapel Hill: University of North Carolina Press.

Sutton, Alison 1994 Slavery in Brazil. London: Anti-Slavery International.

Thomas, Robert J. 1985 Citizenship, Gender, and Work. Berkeley: University of California Press.

Thornton, John 1998 Africa and Africans in the Making ofthe Atlantic World, 1400–1800, 2nd ed. Cambridge, UK: Cambridge University Press.

United Nations 1984 "Slavery: Report Prepared by B. Whitaker, Special Rapporteur of the Sub-Commission on Prevention of Discrimination and Protection of Minorities." New York.

—— 1988 "United Nations Action in the Field of Human Rights." New York.

—— 1991 "Contemporary Forms of Slavery." New York: Center for Human Rights.

—— 1997 "Situation of Human Rights in the Sudan." New York: Commission on Human Rights.

—— (1957) 1999 "The Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery," including a current list of ratifying nations as well as all the major documents and treaties concerning human rights. Available on the United Nations website, www.unhchr.ch/html/intlinst.htm.

United States Department of Labor 1994 By the Sweatand Toil of Children vol. 1: The Use of Child Labor inAmerican Imports. Washington, D.C.: Bureau of International Labor Affairs.

—— 1995 By the Sweat and Toil of Children, vol. II: TheUse of Child Labor in U.S. Agricultural Imports & Forcedand Bonded Labor. Washington, D.C.: Bureau of International Labor Affairs.

—— 1996 Forced Labor: The Prostitution of Children. Washington, D.C.: Bureau of International Labor Affairs.

—— 1997 By the Sweat and Toil of Children, vol. IV:Consumer Labels and Child Labor. Washington, D.C.: Bureau of International Labor Affairs.

—— 1998 By the Sweat and Toil of Children, vol. V:Efforts to Eliminate Child Labor. Washington, D.C.: Bureau of International Labor Affairs.

United States Department of State 1999 Country Reportson Human Rights Practices for 1998. Washington, D.C.: U.S. Government Printing Office. Available on the State Department website, www.state.gov/www/global/human_rights/1998_hrp_report.

Wright, Gavin 1986 Old South, New South. New York: Basic Books.

David R. James

Sara Heiliger