"The World's Oldest Trade": Dutch Slavery and Slave Trade in the Indian Ocean in the Seventeenth Century^*

Markus Vink
SUNY–Fredonia

"Compared with the Atlantic trade, none of this Indian Ocean flow of captive labor, legal or illegal, has been well researched, and there are no conclusive quantitative studies of its volume ..." P. Finkelman and J. C. Miller eds., MacMillan Encyclopedia of World Slavery (New York, 1998), p. 851.   "In comparison with the literature on the trans-Atlantic slave trade, a number of topics in the slave trade in Asia and Oceania remain under-researched ..." S. Drescher and S. L. Engerman, eds., A Historical Guide to World Slavery (New York, 1998), p. 364.   "The evidence of the slave trade in the Indian Ocean is scanty and periodic, and could reflect the nature of the trade. There are huge gaps in the evidence, which might reflect the spasmodic and periodic nature of the slave trade but also the sheer lack of information for long stretches of time." S. Arasaratnam, "Slave Trade in the Indian Ocean in the Seventeenth Century," in K. S. Mathew ed., Mariners, Merchants and Oceans: Studies in Maritime History (New Delhi, 1995), p. 195.
Slavery, far from being a "peculiar institution," has deep and far-reaching roots, stretching back at least to the beginnings of historical times in many parts of the world. In his five-volume magnum opus on the Dutch East Indies, Oud en Nieuw Oost-Indiën (1724–26), Calvinist minister François Valentijn appropriately called the enslavement of human beings "the world's oldest trade" (den oudsten handel in de wereld).1 For most of the seventeenth and eighteenth centuries the Dutch were active participants in the Atlantic and Indian Ocean slave trades. For brief spells during the seventeenth century they even dominated the Atlantic slave trade, while for nearly two centuries they were "the nexus of an enormous slave trade, the most expansive of its kind in the history of Southeast Asia."2 Whereas the Atlantic slave trade has been mapped out in relatively great detail in numerous studies, its Indian Ocean counterpart has remained largely uncharted territory and overlooked in Asian colonial historiography. Indeed, the sufferings of the slaves in Asia occurred mainly in silence, largely ignored by both contemporaries and modern historians. Moreover, if we are to believe one Dutch colonial historian, Indian Ocean slavery "as a topic will never play such an important role as it does in the Caribbean."3	1
Two notable exceptions exist to the "history of silence"4 surrounding the Indian Ocean slave trade: the east coast of Africa (though mostly centered on the period after 1770) and the Dutch Cape Colony (1652–1796/1805). The Afrocentric focus of Indian Ocean historiography is a derivative of the Atlantic slave trade in general, and reflects the takeoff of plantation slavery on the Swahili coast and the Mascarene Islands (Mauritius and Réunion) in the late eighteenth century5 along with its obvious connections with the modern biracial system of apartheid in South Africa (1948–94) in particular.6	2
The underdevelopment of an Asian-centric historiography on colonial slavery and slave trade can be attributed to a number of factors. First, a fragmented administrative superstructure belied the underlying unity of the Indian Ocean world system, making archival research a formidable task to any serious scholar. The archival materials of the major European powers of the early modern period (Dutch, British, French, and Portuguese) are deposited in the national archives at The Hague, London, Paris, and Lisbon, along with the numerous archives overseas in Indonesia, India, South Africa, and elsewhere. Despite the presence of a central Asian headquarters in Batavia (modern Jakarta), even the numerous settlements of the Dutch East India Company or VOC (1602–1799) had separate administrations and record keeping. Second, unlike the Atlantic slave complex, European and preexisting indigenous forms of bondage seemingly shared many forms of similarities. Except for South Africa, European colonial powers took over and interacted with existing Indian Ocean systems of slavery, rather than imposing their own system in a relative vacuum as in the New World. Third, though recently challenged by revisionist historians, Indian Ocean slavery, focused on the household, has been traditionally portrayed by apartheid apologists and other "settler historians" as paternalist, patriarchal, and relatively "benign" compared to its Atlantic plantation counterpart.7 Fourth, the focus of Asian colonial historiography has been on indentured workers and coerced labor systems of a later period or "new systems of slavery," such as the Indian, Chinese, and Javanese "coolie" and the Cultivation System and Liberal Period in nineteenth-century Java and Southeast Sumatra.8 Finally, studies of the Dutch East India Company period (1602–1799) concentrate on trade, political economy, and, more recently, urban history.9 Because slave trade was in general insignificant in monetary terms, most surveys and regional studies on the Dutch East India Company mention slavery and the slave trade only in passing. In the eighteenth century, for instance, when the volume of the Dutch slave trade was substantially larger than in the seventeenth century, slaves only accounted for 0.5% of the total value of company trade.10 These works on the company history ignore the extensive slave trade of company servants as private individuals, along with European settlers or free burghers and Asian subjects of areas under company jurisdiction. Moreover, they obviously gloss over the crucial economic, social, and cultural significance of slave labor in Dutch colonial settlements, which, as we will see, were in fact true "slave societies." As Hubert Gerbeau acutely observed more than twenty years ago, "The specialist in the slave trade is a historian of men and not of merchandise, and he cannot accept the silence of those transported."11	3
This article is a first step to "unsilence" the history of the world's oldest trade and to correct or "re-Orient" the historiographical imbalance by looking at the organization and numerical aspects of Dutch slavery and slave trade in the Indian Ocean in the seventeenth century. The term "Dutch" consists of three components: (1) the Dutch East India Company or VOC as a semi-official institution or chartered company with delegated government rights; (2) Dutch East India Company officials as private individuals; and (3) European settlers or free burghers and Asians in areas under Dutch jurisdiction. Despite the problematic nature of the term "slave" in an Indian Ocean context,12 its special characteristics included property or chattel status and the ensuing potential of re-isolation, institutionalized coercion and systemic exploitation, outsider status or essential kinlessness defined as "social death," and lack of control over physical reproduction and sexuality.13	4
Two basic systems of Indian Ocean slavery can be distinguished. The "open system" of slavery could be found in the commercialized, cosmopolitan cities of Southeast Asia and elsewhere where the boundary between slavery and other forms of bondage was porous and indistinct and upward mobility was possible. In the "closed systems" of South (and East) Asia, with some notable exceptions, it was inconceivable for a slave to be accepted into the kinship systems of their owners as long as they remained slaves because of the stigma of slavery; instead they were maintained as separate ethnic groups. The term "slave" here includes so-called "true slaves," those recently captured and sold in open systems, and those in closed systems of slavery, along with all other forms of bondage and ties of vertical obligation.14	5
This article discusses various aspects of Dutch slavery and slave trade in the Indian Ocean: the markets of supply and demand or geographic origins and destinations of slaves; the routes to slavery or the diverse means of recruitment of forced labor; the miscellaneous occupations performed by company and private slaves; the size of Dutch slavery and the volume of the accompanying annual slave trade; and the various forms of slave resistance and slave revolt. The findings presented here are tentative, illustrating broad contours in bold, sweeping strokes. Further research will be necessary to fill in the details and shed new light on the world's oldest trade in the Indian Ocean basin, but the protracted history of silence has finally ended.	6
The discussion that follows tries to transcend the ahistorical, incomplete, descriptive, static, one-dimensional picture and conventional generalized abstractions of slavery that characterize much of traditional scholarship. Instead, an alternative historicized, holistic, analytical, dynamic, and multi-dimensional, open model is suggested, sensitive to chronological and geographic variations, socioeconomic and political contexts, and cross-cultural interactions. To the extent that the effort has been (un)successful, it is to some extent a reflection of the current state of the field.15	7
Markets of Supply: Origins of Slaves
Inspired by the spatial and temporal geo-historical conceptualizations of Fernand Braudel and the "Annalistes," various studies have looked at the Indian Ocean basin as a meaningful unit of analysis.16 Sometime before 300 B.C.E., sailors of various nationalities began to knit together the shores of the Indian Ocean into a world-system by riding the seasonal monsoon winds blowing off and onto the Asian continent. The resulting multifaceted diffusionist process of "southernization" integrated the societies and cultures of what the Arabs styled al-bahr al-Hindi ("Indian Ocean"), the Persians darya'i akhzar ("Green Sea"), and the Chinese the Nan-yang ("Southern Ocean").	8
Southernization is characterized by both multidimensionality of relationships and nonsimultaneity of geographic boundaries. Ross Dunn considers the concept of southernization as a useful "post-civilizational narrative" and possible new paradigm: "It is not enough simply to say that there was significant trade among the societies clustered around the South Asian center. The products and exchanges created broad regions of similarities which distinguished one distinctive 'world' from another."17 Thus a "world of slaves" emerges as a meaningful unit with interacting elements (e.g., "slave societies") rather than a series of separate "worlds" that have relationships with each other. (In many ways, this post-civilizational narrative is commodity history, a variation on cross-cultural trade, taking a particular commodity as the starting point for the analysis of one or more societies, economic systems, groups involved in production, transportation, marketing, and consumption, and so forth.)	9
In his classic two-volume study on the Indian Ocean world, K. N. Chaudhuri convincingly demonstrates that capitalist, long-distance trade in luxury goods and bulky commodities provided an underlying unity of the region, cutting across geographical and cultural watersheds. Chaudhuri distinguishes "four significant expansionary forces" from the rise of Islam in the seventh century to the onset of European colonialism in the mid-eighteenth century: the emergence and spread of Islam from the Middle East, the massive presence of Chinese civilization and political power, the periodic migration of nomadic peoples from Central Asia, and (after 1500) European maritime expansion. During this life cycle, Chaudhuri argues, "[m]eans of travel, movements of peoples, economic exchange, climate, and historical forces created elements of cohesion." Elaborating on Chaudhuri's work, Janet Abu-Lughod styles the Indian Ocean as "that great 'highway' for the migration of peoples, for cultural diffusion, and for economic exchange."18	10
The Dutch Indian Ocean slave trade precisely involved, among others, the movement of peoples, cultural diffusion, and economic exchange during the later stages of the Indian Ocean ancien régime.19 The Dutch further integrated the Indian Ocean basin by creating direct and more extensive linkages across the region and incorporating previously isolated cultures and societies on the southwestern and eastern peripheries. In the seventeenth century, Batavia (Jakarta), the company's central Asian headquarters and seat of the governor-general and Council of the Indies, became the hub of a flourishing intra-Asiatic or country trade. As the company directors intimated to the high government at Batavia in 1648, "The country trade and the profit from it are the soul of the Company which must be looked after carefully because if the soul decays, the entire body would be destroyed."20	11
The Dutch Indian Ocean slave system drew captive labor from three interlocking and overlapping circuits of subregions: the westernmost, African circuit of East Africa, Madagascar, and the Mascarene Islands (Mauritius and Réunion); the middle, South Asian circuit of the Indian subcontinent (Malabar, Coromandel, and the Bengal/Arakan coast); and the easternmost, Southeast Asian circuit of Malaysia, Indonesia, New Guinea (Irian Jaya), and the southern Philippines.21	12
In general, the Dutch slave trade took people from segmented microstates and stateless societies in the East outside the "House of Islam" to the company's Asian headquarters, the "Chinese colonial city" of Batavia (Jakarta),22 and its regional center in the "western districts" of the Indian Ocean, coastal Ceylon (Sri Lanka). Other destinations included strategic emporia such as Malacca (Melaka) and Makassar (Ujungpandang), along with the plantation economies of eastern Indonesia (Maluku, Ambon, and Banda Islands), and the agricultural estates of the southwestern Cape Colony (South Africa). Though each company settlement had some specific catchment area, certain common patterns prevailed.	13
The first circuit or subregion, the Indian subcontinent (Arakan/Bengal, Malabar, and Coromandel), remained the most important source of forced labor until the 1660s (see Table 1 ). Between 1626 and 1662, the Dutch exported with reasonable regularity 150–400 slaves annually from the Arakan-Bengal coast. During the first thirty years of Batavia's existence, Indian and Arakanese slaves provided the main labor force of the company's Asian headquarters. For instance, of the 211 manumitted slaves in Batavia between 1646 and 1649, 126 (59.71%) came from South Asia, including 86 (40.76%) from Bengal. 23 Slave raids into the Bengal estuaries were conducted by Magh pirates using armed vessels (galias), joining hands with unscrupulous Portuguese traders (chatins) operating from Chittagong outside the jurisdiction and patronage of the Estado da India. These raids occurred with the active connivance of the Taung-ngu (Toungoo) rulers of Arakan. The eastward expansion of the Mughal Empire, however, completed with the conquest of Chittagong (renamed Islamabad) in 1666, cut off the traditional supplies from Arakan and Bengal. 24 Until the Dutch seizure of the Portuguese settlements on the Malabar coast (1658–63), large numbers of slaves were also captured and sent from India's west coast to Batavia, Ceylon, and elsewhere. After 1663, however, the stream of forced labor from Cochin dried up to a trickle of about 50–100 and 80–120 slaves per year to Batavia and Ceylon, respectively. 26	14
Table 1. Company slave exports from the Indian subcontinent (Arakan/Bengal and the Coromandel Coast) in the seventeenth century * This list only represents Coromandel exports in so-called "boom" years. † Exported by private traders to Ceylon. Source: see footnote 25.
In contrast with other areas of the Indian subcontinent, Coromandel remained the center of a spasmodic slave trade throughout the seventeenth century. In various short-lived booms accompanying natural and human-induced calamities, the Dutch exported thousands of slaves from the east coast of India. A prolonged period of drought followed by famine conditions in 1618–20 saw the first large-scale export of slaves from the Coromandel coast in the seventeenth century. Between 1622 and 1623, 1,900 slaves were shipped from central Coromandel ports, such as Pulicat and Devanampatnam. Company officials on the coast declared that 2,000 more could have been bought if only they had the money.	15
The second short-lived boom in the export of Coromandel slaves occurred during a famine in the wake of the revolt of the Nayaka Hindu rulers of South India (Tanjavur, Senji, and Madurai) against Vijayanagara overlordship (1645) and the subsequent devastation of the Tanjavur countryside by the Bijapur army. According to indigenous informants, more than 150,000 people were taken by the invading Deccani Muslim armies to Bijapur and Golconda. In 1646, 2,118 slaves were exported to Batavia, the overwhelming majority from southern Coromandel. Some slaves were also acquired further south at Tondi, Adirampatnam, and Kayalpatnam.	16
A third short-lived boom in slaving took place between 1659 and 1661 due to the devastation of Tanjavur resulting from another series of successive Bijapuri raids, creating the usual "famine-slave cycle." At Nagapatnam, Pulicat, and elsewhere, the company purchased 8,000–10,000 slaves, the bulk of whom were sent to Ceylon while a small portion were shipped to Batavia and Malacca. A fourth boom (1673–77) was initiated by a long drought in Madurai and southern Coromandel starting in 1673, exacerbated by the prolonged Madurai-Maratha struggle over Tanjavur and resulting oppressive fiscal practices. Between 1673 and 1677, the VOC exported 1,839 slaves from the Madurai coast alone. A fifth boom occurred in 1688, caused by a combination of poor harvests and the Mughal advance into the Karnatak. Reportedly thousands of people from Tanjavur, mostly girls and little boys, were sold into slavery and exported by Asian traders from Nagapattinam to Aceh, Johor, and other slave markets. In September 1687, 665 slaves were exported by the English from Fort St. George, Madras. The Dutch decision to participate was belated for the boom ended as abruptly as it had started as a result of the abundant rice harvest in early 1689. Finally, in 1694–96, when warfare once more ravaged South India, a total of 3,859 slaves were imported from Coromandel by private individuals into Ceylon.27	17
After 1660 relatively more slaves came from the second circuit or subregion, Southeast Asia. Warfare and endemic raiding expeditions provided a steady supply of slaves from the region's stateless societies and microstates, especially after the collapse of the powerful sultanate of Makassar (Goa) in Southwest Sulawesi (1667/1669).28 The slave trade network in the archipelago revolved around the dual axis of Makassar and Bali. Makassar was the main transit port for slaves from Borneo (Kalimantan), Sulawesi, Buton (Butung), and the northeastern islands, as well as the eastern Tenggara islands (Lombok, Sumbawa, Bima, Manggarai, and Solor).29 The kingdoms of Bali were not only independent slave exporters, but also reexported slaves from eastern Indonesia as far as New Guinea (Irian Jaya). Of almost 10,000 Indonesian slaves brought to Batavia by Asian vessels between 1653 and 1682, 41.66% (4,086) came from South Sulawesi, 23.98% (2,352) from Bali, 12.07% (1,184) from Buton, 6.92% (679) from the Tenggara islands, and 6.79% (646) from Maluku (Ambon and Banda).30 Similar proportions could be found among Indonesian slaves in the other Dutch posts in the archipelago and at the Cape of Good Hope, where South Sulawesi contributed 46% and Bali 6.6% of the 166 Indonesian slaves imported in the last three decades of the seventeenth century. Between 1680 and 1731, 30.2% (201) of the 666 company slaves imported to the Cape came from Indonesia, 24.8% (165) from India, and 22.1% (147) from Madagascar and other parts of Africa.31 Between 1620 and 1830, Hindu Bali, internally divided among various rival states after the collapse of the kingdom of Gelgel, exported at least 100,000 members of its own population and neighboring Lombok, Sumbawa, Sumba, and elsewhere as slaves.32	18
The third circuit or subregion, the African mainland, Madagascar, and the Mascarene Islands (with a few notable exceptions) remained a relatively insignificant basis of supply for slaves throughout the seventeenth century. As long as supplies from the Indian subcontinent and Southeast Asia could meet demand, slaving voyages to the southwest Indian Ocean were avoided because of the region's peripheral location and marginal commercial importance in the VOC's overall intra-Asiatic trade network. East Africa and the islands off the continent's coast, however, were occasionally tapped into by the Cape Colony and by other company settlements for special projects. Several hundred Bantu-speaking slaves from West and Central Africa were imported in the Cape Colony after 1652 in the first decades of the colony's existence. In addition, 502 slaves were exported from Madagascar between 1641 and 1647. Between 1652 and 1699, 12 company-sponsored voyages were undertaken to Madagascar returning with 1,069 Malagasy slaves and one voyage to Dahomey with 226 slaves, a total of 13 voyages with 1,290 slaves. Malagasy slaves made up 24.1% of all slaves imported into the Cape during this period.33 In comparison, Madagascar supplied about 70% (31,076) of all 44,394 slaves reaching French Mauritius and Réunion between 1670 and 1769. The balance of imports came from Mozambique and the Swahili Coast (19% or 8,435), India (9% or 3,995), and West Africa (2% or 888).34	19
Dutch exports of Malagasy slaves pale in comparison with the French, Arab, Portuguese, and English slave trades from Madagascar in the seventeenth century. During the so-called "age of troubles," the French carried approximately 1,000 slaves from Madagascar to the Mascarene Islands between 1670 and 1714. Arab traders exported as many as 3,000 slaves per year from northwestern Madagascar, including 800–1,000 slaves to Oman, while at least 40 English voyages (mostly interlopers evading the monopoly of the Royal African Company on the west coast of Africa) left Madagascar to the New World between 1675 and 1700 with slaves as cargo.35 Apart from Batavia and the Cape Colony (including Mauritius), a significant portion of Dutch slave exports from Madagascar were intended for the company-operated gold mines along the west coast of Sumatra, where between 1670 and 1696 some 200–500 slaves toiled at Sillida and a few other mines nearby.36	20
Markets of Demand: Destinations of Slaves
Influenced by "central place theory" or "location theory" of the German geographers Walter Cristaller and August Lösch, Fernand Braudel asserts that preindustrial world trade was urban-centered, consisting of an "archipelago of towns." A complex hierarchy of central places, from multifunctional "world-cities" to less specialized regional and local centers, serviced and drew on their respective hinterlands. Applying these concepts to the Indian Ocean, Chaudhuri points to the dominant role of the region's great trading emporia from the Swahili and Red Sea coasts to Southeast China littoral. This thalassic network of Indian Ocean port cities was linked by the sailing ship.37 In the case of the Northern European chartered companies, these central places in the East became, in the words of Philip Curtin, an integral part of "trading-post empires" and tightly controlled "militarized trade diasporas."38	21
During its first centuries, European expansion had a markedly, almost exclusively urban character. Port cities were the nexus of the maritime power of the Iberian state-run enterprises and the Northern European chartered companies. The Dutch Indian Ocean slave system was therefore urban-centered, concentrated around a small number of central places in areas where they exercised political authority "out of conquest," usually designated as "governments." Like the markets of supply, these markets of demand can be divided in three groups: Southeast Asia (including Malacca, westcoast of Sumatra, Jambi, Palembang, Bantam, Batavia and Java's north coast, Makassar, Ambon, Banda, and the Moluccas) South Asia (Surat, Malabar, Ceylon, Coromandel, and Bengal), and "greater South Africa" (Cape of Good Hope and various "outer factories").	22
The Southeast Asian group included Neira Town and the government of Banda (Banda and the southern Maluku Islands), Kotah Ambon and the government of Ambon (Ambon and neighboring islands), Ternate and the government of the Moluccas (northern Maluku Islands), Vlaardingen and the government of Makassar (Makasar and southwestern Sulawesi), Batavia, seat of the governor-general and the Council of the Indies (west and north Java littoral), and the city and government of Malacca (Melaka and its immediate environs on the Malaysian peninsula).	23
The South Asian group included Nagapatnam and Pulicat and the government of Coromandel (east coast of India), Colombo and the government of Ceylon (Jaffnapatnam and the lowlands of southwestern Sri Lanka), and Cochin and the "commandement" of Malabar (Cochin and several forts and guard stations in coastal Kerala). Greater South Africa consisted of the single case of Cape Town and (after 1690) the government of the Cape of Good Hope (the southwestern Cape and increasing sections of the dry interior of South Africa).39	24
All of these urban centers and their surroundings were true slave societies, in which slaves played an important part in both luxury and productive capacities, empowering particular groups of elites, deeply influenced cultural developments, and formed a high proportion (over 20–40%) of the total population (see Table 2 ). 41 Slaves accounted for 51.97% (16,695) and 56.93% (12,505) of the urban population of Batavia in 1679 and 1699, respectively. In 1694, the cities of Colombo and Kotah Ambon consisted of 53.36% (1,333) and 52.31% (2,870) slaves. In 1687 and 1697, 35.18% (649) and 42 .33% (938) of the households in Cochin consisted of slaves. The only figures available for Cape Town fall somewhat outside our time period, but are nevertheless revealing. In 1731, 42.22% (1,333) of the urban population of Cape Town were slaves. Somewhat deviant from the median are the cases of Malacca and Vlaardingen (Makassar). On the low end, Malacca (and its immediate surroundings) had a slave population of "only" 25.28% (1,134) and 40.07% (1,853) in 1680 and 1682, respectively. At the other extreme, 66.55% (921) of the population of the "hamlet" Vlaardingen consisted of forced laborers.42 In comparison, in the typical sugar islands of the Caribbean, 75–95% of the population were slaves, and most of the free people were of African descent. In the American South, generally, most people were not slaves at all, but colonists of European descent. 43 No statistically significant differences in regional patterns can be discerned between the urban slave societies of South Asia (Cochin and Colombo), Southeast Asia (Batavia, Kotah Ambon, Malacca, and Vlaardingen), and South Africa (Cape Town). Slavery was a defining component of Dutch colonial settlements throughout the Indian Ocean, grafted on the preexisting open system of slavery in the commercialized, cosmopolitan cities in Southeast Asia and elsewhere in the Indian Ocean.	25
Table 2. Total population and slave population of various Dutch central places in the Indian Ocean in the late seventeenth century Source: See footnote 40
Routes to Slavery
Arriving in the Indian Ocean in the late sixteenth century, the Dutch took over and interacted with preexisting systems of slavery and dependency.44 Indigenous sources in the Indian Ocean prescribing the "peculiar institution" originated from three major traditions: Hindu, Muslim, and Southeast Asian. Throughout the Indian Ocean basin, however, there were significant geographical and chronological variations within each of these traditions.	26
The normative texts circumscribing routes to slavery in Hinduism were the law books (dharmasastras or smritis) of Manu and Narada. Manu "the Lawgiver" (third century B.C.E.), author of the prototypic dharma text, recognized seven types of slavery: persons (dasas) captured in battle; those enslaved in return for food; those born in the house of the master; those who are bought; those inherited as part of patrimony; those who are given away by their parents; and those enslaved for not paying a fine or in execution of a judicial decree. In one of the latest of the major dharmasastras, Narada (c. 300 C.E.) subdivided and added categories to bring the number up to fifteen, including a child whose adoption was defective, while another resulted from a lost wager. In addition, bondage for debt and nonpayment of taxes were introduced as separate categories.45	27
In Islam, the authoritative sources, the Qur'an, Shari'a, and (to Sunni Muslims only) the innumerable hadiths (traditions), recognized four main ways of recruitment of slaves (‘Abd): capture of infidels in holy struggle (jihad) against the "Abode of War" (dar al-Harb), tribute (bakt) required from vassal states beyond the Islamic frontiers, offspring from or birth to slave parents, and purchase by slave merchants known as "importers" or "cattle-dealers." Raids and petty warfare by Muslims could at best be interpreted as a genuine jihad or holy war to extend the faith. In 1669, for instance, the Pangeran Dipati of Jambi, a Muslim port city on the east coast of Sumatra, justified the equipping of armed vessels for a slave raid against Ujang Salangh on the Malaysian Peninsula arguing that the inhabitants "were heathens, and hence [the raiding] could not be considered an injustice." At worst, however, these expeditions represented ruthless raiding for helpless human prey and continued on all the boundaries of Islamic communities throughout the Indian Ocean basin.46	28
The Southeast Asian legal codes (Bugi, Malay, Javanese, and other texts, such as the Undang-undang Melaka, Agama Djawa, or Adat Aceh) were based on ancient Indian dharmasastras, Islamic law books, decrees issued by the ruler in power, and numerous local traditions (adat). Though widely differing, they commonly acknowledged five paths by which a person could enter a state of bondage: inheriting the bondage of one's parents; sale into bondage by parents, husband, or oneself; capture in war or raids; judicial punishment (or inability to pay fines); and, most importantly, failure to meet debts. Debts were acquired through trading, gambling, inability to pay for the ceremonies associated with rites of passage (naming ceremony, initiation, marriage, funeral, and so forth), and crop failure or other calamity.47	29
Grafting on these preexisting indigenous traditions, the Dutch brought with them an intellectual, theoretical mentalité steeped in Christian humanism mixed with a healthy doses of pragmatism.48 In the Dutch Republic, majority apologists (moderate Calvinist ministers, theologians, and jurists) used biblical references and sources of classical antiquity to produce a specific slaving discourse with a Protestant face. The so-called "Ham ideology" based on the curse of Canaan (Genesis 9:25–27) served as the most important biblical justification for the enslavement of Africans and Asians until the nineteenth century emancipation.49 The authority of the Old and New Testaments was supplemented by the writings of Greco-Roman authors, condoning slavery "within natural limits" or under certain safeguards stated in company ordinances. Dutch jurists, such as Hugo Grotius (1583–1645), recognized voluntary sale in order to escape famine and starvation, capture in a just war (bellum iustum), judicial slavery as an alternative to execution, and inheritance or birth from a female slave.50	30
Whereas ideological motivations on a theoretical plane dominated discourse in Europe, in Asia slavery found virtually universal acceptance on a practical level among self-righteous religious, military, and civil officials of the Dutch East India Company. Using reasons of state or pragmatic politics to defend the trade, these company servants rather opportunistically resorted to a variety of ad hoc arguments in an unsystematic manner. These arguments included Christian humanitarian compassion (saving the body and soul of the slave), the need to establish and populate settlement colonies ("peuplatie"), the right of war and conquest (bellum iustum), the "uncivilized" nature of the "servile" indigenous peoples, natural, contractual law (pacta sunct servanda), and financial-budgetary considerations ("mesnagie"). To God-fearing Calvinists in patria and overseas, the enslavements of African and Asian peoples did create a serious moral predicament for both opponents and apologists alike. Torn between gain and godliness, or the practical temptations of the "Dutch trading empire" and the principle requirements of the "Empire of Christ," the "peculiar institution" was carefully circumscribed, permissible only "for good and sufficient reasons" and "within natural limits" under certain safeguards punctiliously stated in company ordinances.51	31
The Dutch acquired the majority of their slaves indirectly through purchase from indigenous suppliers, which, similar to the other universal religions of Buddhism and Islam, was rendered in religious humanitarian terms as a "work of Christian compassion" (christelijcke mededogentheyt) based on the alleged material and spiritual salvation of the individual slave's body and soul. Apart from saving people from physical starvation in instances of severe famine, enslavement also allegedly saved the soul of infidels ensnared in the trappings of the devil.52 Enslavement of company subjects via debt bondage, however, was prohibited (with little success) in numerous ordinances as "contrary to all divine and Christian laws and unacceptable to the high authorities."53	32
Numerous armed conflicts with indigenous societies also provided a major source of captive labor. In accordance with prevailing mercantilist thought and a paranoid siege mentality, these conflicts were invariably depicted by the righteous Dutch as just wars. In theory, the company only accepted legally acquired slaves and rejected those acquired via outright kidnapping and robbery. In practice, however, the Dutch distinction between legal acquisition and illegal kidnapping and robbery was as nebulous as the Muslim differentiation between "true slaves" (Arabic ‘abd) resulting from genuine jihad and ruthless raiding for helpless human prey.54	33
In addition, "rebellious" peoples, once subdued, were often forced at gunpoint to sign treaties with slave clauses, promising the delivery of a fixed number of slaves and other commodities as fine or tribute (boete ofte amende). The symbolic exchange of gifts reestablished relationships and reordered the proper hierarchy between individuals, groups, and the "honorable" company. Between 1650 and 1675 alone, the Dutch concluded numerous slave-clause agreements with indigenous chiefs and headmen (orangkayas, penghulus, sangajis, and so forth) on and near the islands of Sumatra, Timor, New Guinea, and Sulawesi.55	34
According to Dutch Roman law, slavery was hereditary through the female "on the principle that the fruit follows after the womb." Offspring from female slaves, however, was a relatively insignificant source of captive labor due to low levels of reproduction among slave populations. Despite several "moments of creolization" (when the slave population was more than 50% locally born), the gender imbalance among slaves (i.e., an overwhelming preponderance of male slaves especially among non-company slaves), high mortality rates, and poor living and working conditions limited female fertility and weighed heavily against natural reproduction.56	35
In 1682, for instance, the slave population of Malacca consisted of 1,589 adults (880 men and 709 women) and 264 children, an adult-to-child ratio of 6.0:1 (see Table 3 ). In 1687, the Cape slave population consisted of 758 adults (503 men and 253 women) and 145 children (ratio 5.2:1), while the Dutch settlements along the Malabar coast (Cochin, Quilon, Cannanur, Cranganur, and Pallipuram) had 716 adults (536 men and 180 women) and 137 children (ratio 5.2:1). In 1688, the slave population of Ambon consisted of 8,335 adults (4,446 men and 3,889 women) and 2,381 children (ratio 3.5:1), whereas Banda had 3,213 adults (1,631 men and 1,582 women) and 503 children (ratio 6.4:1). In 1689, the slave population of Batavia consisted of 22,570 adults (12,557 men and 10,013 women) and 3,501 children (ratio 6.4:1). 57	36
Table 3. Gender and age composition of slave populations in various company settlements in the late seventeenth century * The Zuidervoorstad (southern suburbs) excluded. Source: See footnote 58
Judicial punishment, in the form of political exiles and convicts (gecondemneerden or bandieten), represented a small, but distinct category.59 The ordinances issued by company authorities in Batavia, Ceylon, the Cape, and elsewhere contain numerous references in which transgressors, such as captured runaway slaves, are condemned to chain labor for specified terms at the public works (ad opus publicum).60 Batavia, Colombo, Robben Island off the coast of South Africa, and Rozengain, one of the Banda Islands, served as special penal colonies for several hundreds of convicts.61	37
Throughout the Indian Ocean, war captives came largely from animist, stateless societies of shifting cultivators or hunter-gatherers and microstates too weak to defend themselves against the stronger and wealthier societies of the cities and the rice-growing lowlands. Sometimes non-Muslim societies and hill peoples sold the victims of their own internal warring; more often they were simply raided for slaves by outsiders. In Southeast Asia, these so-called "victim societies" included, among others, the Papuas of New Guinea; the Alfurs of Maluku, northern Sulawesi, and Mindanao; the Torajas and Mandars of Southwest Sulawesi; the Dayaks of Borneo; the Minangkabau and Bataks of Sumatra; and the Orang Asli (lit., "original peoples") of Malaysia. These groups enslaved each other in low-scale, endemic intercommunal conflict or were enslaved by neighboring (often Muslim) states, such as the sultanates of Ternate and Tidore in eastern Indonesia; Magindanao on the island of Mindanao; Makassar (Goa) and Bone on Sulawesi; Aceh, Jambi, and Palembang on Sumatra; and Johore on the Malaysian peninsula.	38
Oftentimes, the slaving frontier coincided with the juxtaposition of upstream and downstream (ulu-ilir). Somewhere between 500 and 200 B.C.E., coastal communities in Sumatra, Borneo, Malaya, and elsewhere started acting as downstream, "gatekeeper" polities for the upstream interior in the larger drainage systems of the region.62 Upstream and downstream communities were distanced by geography, language, and customs, but nonetheless drawn together by politico-economic ties between the interior and coastal centers.63	39
South Asian slavery appears to have originated in the Aryan conquest (after 1500 B.C.E.) and assimilation into the caste structure of autochthonous peoples who were enslaved as dasas. Similar to the upstream-downstream division in Southeast Asia, the "inner frontier" in South Asia separated tribal peoples (adivasis) of hunter-gatherers, shifting cultivators, and pastoral nomads in the interior forest and hill tracts from the sedentary wet-rice farmers of the coastal and riverine floodplains. With the expansion of the agricultural frontier in the medieval and early modern period, these forest and pastoral peoples were reduced, along with war captives and other outsiders without community protection, to various degrees of dependency upon landed households of higher standing.64 Slaving also occurred intermittently along the Muslim-Hindu "frontier" in South Asia, starting with the numerous pillaging expeditions of Mahmud of Ghazni into the subcontinent between 1000 and 1025 C.E., and continued throughout the periods of the Delhi Sultanate (1206–1526) and the Mughal Empire (1526–1857). Mughal policy alternated between pragmatism and orthodoxy, represented most clearly in the reigns of Akbar (r. 1556–1605) and Aurangzeb (r. 1658–1707), respectively. The term "Muslim-Hindu frontier," however, should be used with caution. It implies a rather precise line where none existed. Muslims ruled over Muslims and non-Muslim subject populations all over South Asia. In fact, the Indo-Islamic ulama was deeply divided between pragmatism, personified by "Akbar's intellectual" Abu'l Fazl Allami (1551–1602), and orthodoxy, represented by the leader of the Muslim opposition to Akbar, Shaykh Ahmad Sirhindi (1564–1624), over how non-Muslims should be treated.65	40
Frequent warfare among the major confederations and kingdoms on Madagascar (Sakalava, Tsitambala, and Merina), compounded by the rise of militant Islamic sultanates such as Maselagache on the northwest coast of the island, led to a steady stream of captive humanity throughout the seventeenth and eighteenth centuries. The hereditary slave caste (andevo) of the highland Merina, for instance, was supplemented by war or slave-raiding, judicial punishment, and debt bondage.66 On the African mainland, non-Muslim, Bantu-speaking peoples of the interior of east and central Africa fell victim to slave raiding by rival Muslim and Afro-Portuguese coastal networks. The real takeoff in the southwest Indian Ocean, however, did not occur until the late eighteenth century with the rapid expansion of Omani and French agricultural plantations, the demand of Brazilian merchants, and a series of severe droughts and epidemics.67 No slavery existed in South Africa prior to the arrival of the Europeans among the local Khoikhoi and San populations. Slaves therefore had to be imported from outside. In the pastoral region of the Cape Colony, however, increasing numbers of Khoikhoi and San on the interior "highveld" were dispossessed and employed by Dutch colonists as indentured laborers in the eighteenth century. Though this system of "inbooking" (inboekstelsel) was not formalized until 1775, local company magistrates approved of the practice long before this date.68	41
Inheritance and judicial punishment were the most common sources of forced labor in the closed systems where a money economy was little developed. There existed, however, no institutional obstacles against the sale of slaves on a massive scale once a strong external demand made itself felt along with the spread of a money economy in the absence of a strong state. All major indigenous powers prohibited the export of slaves as an intolerable loss of the country's most precious resource and a violation of the collective moral code of society. On the Indian subcontinent, for instance, the Dutch encountered difficulties with the Mughals and Marathas in the north and the Nayaka rulers in the south opposing the slaving activities of the Europeans. In 1643, for instance, the ruler of Senji, Krishnappa Nayaka, lectured a visiting Dutch envoy—Calvinist minister "that selling human beings was not only disgraceful to the world, but was also considered one of the greatest sins by our gods." In Southeast Asia, with the completion of the process of Islamization of Java and expansion of the state of Mataram in the sixteenth century, the island ceased to export its people. No slaves were exported from South Sulawesi during the height of the sultanate of Makasar (1600–68).69	42
Sale and indebtedness were more important routes to slavery in the cities and other areas open to the money economy. Debts were, as we have seen, an important source of bondage, whether acquired through trading, gambling, inability to pay for the ceremonies associated with rites of passage, crop failure, or other calamity. The prominence of debt and judicial slavery seems to be distinctive in the Southeast Asian pattern. Sale into bondage in times of famine was more prevalent in South Asia and China.70	43
Slaves were transported to Dutch settlements by company ships or other European slavers, free burgher vessels, and Asian craft. Special slaving voyages were occasionally undertaken by the company in times of great demand or for special projects, though normally "pieces" of slaves were stowed as supplementary cargo on board Dutch East India ships along with other commodities. Company officials as private individuals also engaged in the legal and illegal slave trade. Foreign slavers would occasionally sell some of their forced labor when calling at one of the VOC ports en route to their ultimate destination. Free burgher and Asian merchants (rulers, nobility, and commoners) routinely carried slaves throughout the Indian Ocean as part of the so-called country or intra-Asian trade.	44
Slave Occupations
Slaves were general laborers and used in a wide variety of occupations in the Dutch slave societies across the Indian Ocean basin. Specialization among private and company slaves, however, occurred in accordance with the size of the individual slave household and the particular position the settlement occupied within the company's overall trade network. The majority of slaves acted as domestic servants in small or large slave households of company officials, free burghers, and Asian subjects in areas under Dutch jurisdiction. They served as cooks, lamplighters, houseboys, housemaids, concubines, seamstresses, bread bakers, tea makers, coachmen, musicians, masseuses, honor guards, valets, and so forth.71 They performed menial labor as coolies in the construction of fortifications, buildings, roads, canals, and trenches, and as porters and stevedores in the ports and warehouses.	45
In agriculture, slaves grew food crops (rice, wheat, potatoes, and vegetables), cash crops (pepper, nutmeg and mace, cloves, sugar, cotton, tobacco, and grapes), and herded cattle and sheep. In mining, slaves dug for gold, tin, and other minerals, and broke coral stone for the burning of lime. They served as fishermen, sailors, and country traders in the intra-Asiatic trade. In manufacturing, slaves labored in artisan workshops as carpenters, furniture makers, coopers, tailors, cobblers, gold-, silver-, and blacksmiths, and numerous other artisanal occupations. They worked in gunpowder mills, sulfur and saltpeter refineries, arak distilleries, sawmills, shipyards, and sugar mills. In the service sector, they were active in retail, (lower) administrative functions, medical professions (nurses, midwives, etc.), and so forth. Political exiles and criminals (gecondemneerden and bandieten), the result of judicial punishment, formed a small but separate category. Invariably, they were condemned to perform hard physical labor at the public works (ad opus publicum) at the company's fortifications or elsewhere often as part of a chain gang.	46
In Batavia, the administrative center, central rendezvous, and port of transshipment of the VOC in Asia, several hundred company slaves served on the island of Onrust's (Pulau Kapal) shipyards to repair and service the visiting Dutch East Indiamen, while others worked in the company hospitals. In the environs of Batavia, thousands of Asian and free burgher slaves cultivated the sugar, rice, and pepper gardens in the late seventeenth century.72 In eastern Indonesia, the center of spice production, thousands of free burgher and Asian slaves worked the clove gardens in Ambon and the nutmeg plantations in Banda. In 1694, for instance, 1,879 free burgher slaves labored on the some 70 nutmeg gardens or perken of Banda, though 2,500 were deemed necessary.73	47
At Malacca, a vital port of transshipment in the VOC's intra-Asiatic trade network, the majority of the 207 company slaves in 1662 were held in the "Slave Castle" (Slavenburgh) as a flexible pool of labor "for the loading and unloading of the ships and elsewhere where needed." In 1678, 53 out of 187 company slaves were directly employed "in the chain gang on the common works." It was recommended, however, that "[a]ll slaves, both men and women ... shall, if occasion arises for something to be done quickly, whether the unloading or loading of ships or anything else, be united with the common gang and employed where they are needed."74	48
In the 1660s and early 1670s, when the company tried to achieve self-sufficiency on Ceylon, over 2,000 company slaves worked the fields around Galle and Colombo in the southwest of the island, growing rice, nacheri (fine cereal), cotton, tobacco, potatoes, and other crops. In Colombo, seat of the Dutch governor and council of Ceylon, 533 adult slaves (including 348 females) and 196 slave children served as fortification workers in 1685, carrying bricks, lime, and earth to the city's and castle's ramparts.75	49
At the Cape of Good Hope, the vast majority of the 300 company slaves tilled the 40-acre urban vegetable garden in Cape Town, while hundreds of free burgher slaves worked on the intensive wheat and wine farms in the southwestern Cape. Smaller numbers herded the free burgher livestock in the pastoral regions to the north and east, though the preference here was to recruit the local Khoikhoi and San populations.76	50
The division of slave labor roughly followed ethnic, gender, and age lines based on colonial taxonomy and preexisting indigenous beliefs and practices that characterized local slave systems. Comparative psycho-physiology decided the typical qualities and defects assigned to representatives of the various races and, in consequence, the functions for which they were considered best suited. Administrative ethnography and categorization, however, should be treated with caution since labels could be deceptive or misleading due to miscegenation, political expediency, and other reasons. Moreover, ethnic stereotyping or racial profiling differed in each company settlement and changed over time. Nevertheless, Indian and Southeast Asian slaves in general were deemed to be cleaner, more intelligent, and less suited to hard physical labor than African slaves.77 Indian and Southeast Asian slaves therefore frequently worked as artisans or domestic servants, while African slaves commonly served as field laborers. Despite the importance of "context-dependent particulars" and local variations, slave women were not used regularly in fieldwork, but were mostly involved in domestic labor as housemaids, specialized seamstresses and knitters, laundresses, or wet-nurses.78 Slave children could be employed in seasonal work, such as sheaf-binding or domestic occupations, serving as "companions" to their master's children, guarding younger white children or babies. At the Cape, older farm slaves often worked with the livestock either as shepherds or leading the oxen in plowing.79	51
Size of Dutch Slavery and Volume of the Slave Trade
The number of company and total Dutch slaves and the accompanying volume of the annual slave trades were subject to great volatility and varied greatly from year to year. Famine, wars, epidemics, and natural disasters could wreak havoc among local slave populations, already tending to melt away due to high mortality rates, low levels of self-reproduction or creolization, manumission, and widespread desertion. As has been noted by many scholars, there is no slavery without the slave trade because of the long-term demographic imbalance. Mortality rates in most of the company settlements in Southeast Asia were significantly higher compared to those at the Cape, and, to an even greater degree, in South Asia.80	52
In Ambon, ruthless wars (1618, 1625, 1636–37, 1641–46, 1650–56, 1658–61, 1680–81) and recurring malaria epidemics (1633–34, 1651, 1656–58, 1666, 1671–72, 1677–78, 1682–84, 1689–91) occasionally ravaged all groups of the population, including slaves. Continuous raiding by Alfurese "headhunters" in southwest Ceram, outward migration, and a massive earthquake followed by a tidal wave (1674) further exacerbated the existing demographic problems. Between 1643 and 1671, the population of Ambon (Hitu, Larike, Hitu Tenggara, and Leitimor), the Lease Islands (Haruku, Saparua, and Nusalaut), Western Islands (Ambelau, Buru, Boano, Manipa, and Kelang), and southwest Ceram declined with approximately 30,000, or 37%, recovering only moderately until 1691 (approximately 1% growth annually), and stabilizing in the period afterward (1692–1708).81	53
In Banda, the original population was deported, driven away, starved to death, or massacred by the company in 1621 and replaced by Dutch colonists or perkeniers using slave labor. The worst malaria epidemics generally followed volcanic eruptions. In 1638, 375 people died in Neira Town alone. In 1678, 376 people died in the Dutch government of Banda altogether. In 1693, 771 people died, including "very many slaves" in the nutmeg gardens or perken. In 1702, 351 free burgher slaves died and subsequently 356 had to be imported from eastern Indonesia in order to replace them.82	54
Southwestern Ceylon was plagued by intermittent warfare and hostilities with the interior kingdom of Kandy (1670–75) and recurrent droughts or floods (1659, 1661, 1664, 1669, and 1673), spreading famine and disease across the island. In 1661, for instance, 900 slaves died, including 400 "old and nearly worn out people." In 1669, large numbers of people were dying in the lands of Colombo. The 100 Dutchmen and 800 slaves reportedly being treated in the local company hospitals were hardly better off, since these facilities were often virtual death traps.83	55
Batavia and its environs were the scene of several wars involving the Javanese states of Mataram (1628–29, 1677–81) and Bantam (1619, 1633–39, 1656–59, 1680–83), and increasingly virulent diseases (especially malaria) due to natural and human-induced changes in the local environment. Between 1676 and 1677, for instance, during hostilities involving Mataram, the slave population inside Batavia declined from 17,279 to 15,776. Between 1688 and 1690, diseases reduced the number of slaves inside the city (excluding the Zuidervoorstad) from 12,125 to 11,172. In August and September 1688 alone, over 1,000 people "of all nations" died in Batavia due to "an evil form of measles and smallpox." High mortality rates in Batavia were compounded by cramped living conditions and the practice of housing domestic slaves in rooms without air circulation.84	56
Barring unexpected high mortality, the number of company slaves was overall relatively stable. At times, however, drastic changes could occur as a result of policy decisions. Following the completion of the local fortifications, the number of company slaves at Batavia between 1664 and 1671 was reduced by one-third from 1,519 to 1,008. The cuts apparently were too deep and the company was subsequently forced to hire 500 slaves from private persons. To reduce expenses, the high government in January 1678 decided to employ more company slaves.85 More permanent was the decision to decrease the number of company slaves in Ceylon as part of a series of budgetary measures. Between 1677 and 1679, the number of company slaves at Ceylon was slashed almost in half from 3,932 to approximately 2,000, respectively. Some old slaves were manumitted against payment, while the rest were sent to Batavia and Malacca. Even in Colombo, the company was subsequently forced to hire 500 Christian Paravas from southeast India as coolies to help strengthen Ceylon fortifications.86	57
Whereas the slave population of the company and its officials as private individuals was relatively stable, that of European and Asian subjects in areas under company jurisdiction displayed a distinct secular upward trend, growing throughout the seventeenth and the early eighteenth centuries. Only in the late eighteenth century did the number of slaves in Dutch "conquests" drop dramatically along with the declining fortunes of the VOC.	58
A sample from the late seventeenth century provides some valuable insights into the number of company and total Dutch slaves and the accompanying volume of the annual slave trades. The numbers presented here are, to borrow a phrase from Ralph Austen, in the form of a "tentative census." 87 Future archival research will refine the estimates for various VOC settlements. In 1688, there were about 4,000 company slaves and perhaps 66,000 total Dutch slaves in the various settlements spread out across the Indian Ocean basin (see Table 4 ). Not surprisingly, the two most important VOC settlements, Ceylon (1,500) and Batavia (1,400), accounted for the bulk of the 4,000 company slaves, with the Cape of Good Hope (382), Banda (166), Malacca (161), and Makassar (112) as important secondary centers of forced labor. Batavia (26,000) and Ambon (10,500) made up over half of the total of 66,000 Dutch slaves, with Ceylon (4,000), Banda (3,700), Malacca (1,800), and Makassar (1,500) as significant second-rank slaveholding societies. The number of slaves in South Asia was much smaller than in Southeast Asia and greater South Africa due to the ability to compel "free" populations to perform labor services as part of their caste obligations or to demand extraordinary services from resident foreign communities.	59
Table 4. Number of company slaves and total Dutch slaves along with estimates of the size of the accompanying annual slave trades, ca. 1688 a With the exception of the Cape of Good Hope, all figures on company slaves are derived from Gaastra, De geschiedenis van de VOC, pp. 84–85, 94. b According to Knaap, 8–9% of the total number of slaves in Ambon were imported each year. See Knaap, Kruidnagelen en christenen, pp. 107, 132. c In 1680, the company estimated that annually 150–200 slaves had to be imported in Banda. See Generale Missive IV, p. 431. According to Jacobs, in the eighteenth century more than 100 slaves had to be imported each year in Banda in order to compensate for the losses due to deaths and desertions. See Jacobs, Koopman in Azië, p. 26. d Prior to 1733, mortality rates in Batavia and Ceylon ranged between 5 and 10%. See Van der Brug, Malaria en Malaise, p. 59. According to Reid, about 1,000 slaves per year may have been imported on average into Batavia in the seventeenth century, though this figure seems to be on the low side for 1689. See Anthony Reid, "Introduction: Slavery and Bondage in Southeast Asian History," p. 29. In 1671, in the aftermath of the campaigns waged by the company and its allies against Makassar, 1,584 slaves were imported in Batavia by Asian ships. In 1720, 2,149 slaves were imported legally into Batavia. Raben, "Batavia and Colombo," p. 122; Generale Missiven VII, pp. 579, 581. e According to Armstrong and Worden, for most of the eighteenth century 100–200 slaves were imported into the Cape by the burghers and company officials (excluding those imported by the company for its own use). See Armstrong and Worden, "The Slaves, 1652–1834," p. 120. According to Shell, 63,000 slaves were imported into the Cape between 1652 and 1808, 5,400 of those by the company between 1652 and 1795. By 1717 2,579 slaves (40 per year) had been imported by the company and 3,997 slaves (60 per year) by free burghers and local officials (total of 6,576 or 100 per year). See Shell, Children of Bondage, pp. 4, 40–41, 68. f With the exclusion of Matura. g In 1694, the city of Colombo alone had a slave population of 1,761. See Knaap, "Europeans, Mestizos and Slaves," p. 88. In 1661, 10,000 slaves had been put to work by the company and by private individuals on the lands in southwestern Ceylon, including 2,000 company slaves. Due to poor conditions, manumission, and budgetary measures of the late 1670s the number of slaves decreased rapidly thereafter. VOC 1234, fls. 125r-v, Miss. Gouvr. Van Goens aan H. XVII, 5.4.1661. h In 1676, the city of Vlaardingen alone had a slave population of 921 out of 1,384 inhabitants. Vlaardingen had 1,125 and 2,929 inhabitants in 1679 and 1694, respectively, with 58% of the population (c. 2,500 based on interpolation) reportedly being slaves in 1688. i In 1687, Cochin, the seat of the Dutch commander and council and main Dutch settlement on the Malabar coast, had (excluding the Asian population) 1,845 inhabitants, including 649 slaves (plus 142 free women and 24 free household servants). The other company settlements (Quilon, Cannanur, Cranganur, and Pallipuram) had 704 inhabitants, including 204 slaves. VOC 1434, OBP 1688, fls. 263v–265v, Samentrekking huisgezinnen, 17.12.1687; s'Jacob, "De VOC en de Malabarkust in de 17de Eeuw," p. 88. j The mortality rates in Malabar in 1768–69 among VOC personnel was 5%. Van der Brug, Malaria en Malaise, Table B, p. 29. k There were 330 male slaves in the Moluccas in 1686 along with 385 (free and enslaved) women and children. Generale Missiven V, p. 33. l In 1687/88, the other company settlements (Amoy, Bantam, Bengal, Coromandel, Jambi, Japan, Japara, Palembang, Persia, Siam, Sumatra, Surat, Timor, and Tonkin) had 2,839 European VOC servants, 24.58% of the total of 11,551.
To replenish or increase these numbers, 200–400 company slaves and 3,730–6,430 total Dutch slaves had to be imported each year. Assuming average mortality rates en route of circa 20% on slaving voyages, 240–480 company and 4,476–7,716 total Dutch slaves were exported annually from their respective catchment area.88 To place these numbers in a comparative global perspective, 9,500 slaves were exported each year in the trans-Saharan slave trade in the seventeenth century; 3,000 slaves were shipped annually from the Swahili and Red Sea coasts during the same time period. In addition, 29,124 slaves were exported each year in the Atlantic slave trade during the last quarter of the seventeenth century, 2,888 of which transported by the Dutch West India Company. The exact volume of the Crimean Tatar trade in Polish and Russian slaves is impossible to gauge, though one (inflated) estimate suggests that in the seventeenth century Poland lost an average of 20,000 captives yearly. In the period 1607–17 the Tatars may have seized 100,000 Russians and in the next 30 years another 100,000.89 The volume of the total Dutch Indian Ocean slave trade was therefore 15–30% of the Atlantic slave trade, slightly smaller than the trans-Saharan slave trade, and one-and-a-half to three times the size of the Swahili and Red Sea coast and the Dutch West India Company slave trades.	60
Slave Resistance and Slave Revolt
With some notable exceptions, relationships of power in Dutch territories were tilted too heavily in favor of the owners for a large or even a medium-scale slave revolt to break out. The virtual absence of concerted mass slave rebellions in the face of slave-owner hegemony does not mean, of course, that the slaves meekly acquiesced in their subordination. Resorting to what James C. Scott has called the "weapons of the weak," they expressed their discontent in a variety of everyday forms of resistance, but above all by marooning or deserting.90	61
Using Eugene Genovese's list of preconditions favoring "massive revolts and guerilla warfare," numerous factors worked against any large-scale slave rebellion in Dutch-controlled territories throughout the Indian Ocean basin.91 First, the nature of the master-slave relationship was by and large immediate and personalized. Though the incidence of absenteeism was probably higher than most historians have assumed, there appears to have been no serious cultural estrangement between "white" and "black." In fact, "Low Portuguese" and Malay were the means of communication between master and slave in Batavia, Ceylon, the Cape, and elsewhere.92 Second, despite several booms and busts in the economies of Dutch colonial settlements, there were only occasional signs of serious distress and famine due to inadequate local food production and/or external sources of supply. In fact, the high government was very conscious in assuring plentiful rice supplies and low prices at Batavia and the "outer factories" in order to forestall bread riots and other forms of social unrest.93 Third, the small size of slaveholding units in Dutch conquests worked against any massive slave insurrection. Unlike the sugar colonies in the New World with an average size of 100 to 200 slaves per owner,94 few slave owners in the Indian Ocean had more than 100 slaves. In 1688, for instance, the 88 perkeniers of the nutmeg gardens on Banda owned 2,149 slaves, an average of less than 25. Some 25 small perkeniers only had 6, 8, or 10 slaves. In the same year, the average married household in Stellenbosch district at the Cape consisted of 6.3 individuals, including free burghers, VOC-knechts, and slaves. Most of the wine farms owned between 5 and 20 slaves, some farms having only 1 or 2 slaves.95	62
Fourth, despite politico-economic tensions between private settlers or burghers and VOC leadership, the basic fabric of colonial society was never seriously threatened. The exceptions to the rule at the Cape are the affair that led to the sacking of Governor Willem Adriaen van der Stel in 1707, the Barbier rebellion of 1739, and the Patriot movement of the 1780s.96 Though the Patriot movement had little impact on the citizenry of Batavia, during the early history of the Dutch Asian headquarters a burgher movement lobbied for a city government, a Council of Aldermen (Schepenbank) of their own, which could not be manipulated by presiding company officials. The letter, which they sent in 1649 directly to the States General of the Dutch Republic (effectively bypassing the high government in Batavia and the VOC directors in Amsterdam), in fact represented the culmination of a long-standing feud between the burghers and company authorities, which had been fermenting since the founding of Batavia in 1619. All requests from the Batavian citizenry, however, were brushed aside. They had to live under institutions, which were similar in name only to those back in Holland, presided over by company officials, mostly members from the Council of the Indies. Henceforth, the free burghers lived subject to the whims of the successive governors general.97	63
Fifth, although slaves generally outnumbered the whites, the slave/master ratio was never sufficiently tilted especially when including local Asian populations under Dutch