Homework answers / question archive / How important was coerced labour for the development of preindustrial economies? The minimum number of citation is 8

How important was coerced labour for the development of preindustrial economies? The minimum number of citation is 8

Sociology

Share With

How important was coerced labour for the development of preindustrial economies?

The minimum number of citation is 8. The format is Chicago, and if the reference is from a book, you should edit the accurate page number.

Your plan should detail what you would include in:

• Your introduction (key debate/historiography, summary of your main argument)
• Your main paragraphs (key points, evidence, how it links to your overall argument)
• Your conclusion (summary of key points, overarching argument)
• The word limit is 1500 words per question
• The word limit excludes citations and the bibliography but these must be included
• You can also include tables and figures – these also don’t count to the wordcount (but don’t go overboard!)

THE CAMBRIDGE WORLD HISTORY O F SL AV E RY Volume 3: a d 1420–a d 1804 Most societies in the past have had slaves, and almost all peoples have at some time in their pasts been both slaves and owners of slaves. Recent decades have seen a significant increase in our understanding of the historical role played by slavery and wide interest across a range of academic disciplines in the evolution of the institution. Exciting and innovative research methodologies have been developed, and numerous fruitful debates generated. Further, the study of slavery has come to provide strong connections between academic research and the wider public interest at a time when such links have in general been weak. The Cambridge World History of Slavery responds to these trends by providing for the first time, in four volumes, a comprehensive global history of this widespread phenomenon from the ancient world to the present day. Volume 3 of The Cambridge World History of Slavery is a collection of essays exploring the various manifestations of coerced labor in Africa, Asia, and the Americas between the opening up of the Atlantic world and the formal creation of the new nation of Haiti. The authors, well-known authorities in their respective fields, place slavery in the foreground of the collection but also examine other types of coerced labor. Essays are organized both nationally and thematically and cover the major empires, coerced migration, slave resistance, gender, demography, law, and the economic significance of coerced labor. Nonscholars will also find this volume accessible. David Eltis is Robert W. Woodruff Professor of History at Emory University and research associate of the W. E. B. Du Bois Institute, Harvard University. He has also held visiting appointments at Harvard, Yale, and Oxford universities. Eltis received his PhD from the University of Rochester in 1979. He is author of The Rise of African Slavery in the Americas, co-author (with David Richardson) of Atlas of the Transatlantic Slave Trade, and co-compiler of Slave Voyages at www.slavevoyages.org. He co-edited and contributed to Extending the Frontiers: Essays on the New Transatlantic Slave Trade Database (with David Richardson) and Slavery in the Development of the Americas (with Frank D. Lewis and Kenneth L. Sokoloff ) and edited Coerced and Free Migrations: Global Perspectives. Stanley L. Engerman is John H. Munro Professor of Economics and Professor of History at the University of Rochester. He has also previously taught at Harvard, Yale, Oxford, and Cambridge universities. Engerman received his PhD in economics from Johns Hopkins University in 1962. He is the author of Slavery, Emancipation, and Freedom: Comparative Perspectives and the co-author of Time on the Cross: The Economics of American Negro Slavery (with Robert Fogel) and Naval Blockades in Peace and War: An Economic History Since 1750 (with Lance E. Davis). He is also co-editor of A Historical Guide to World Slavery (with Seymour Drescher); Finance, Intermediaries, and Economic Development (with Philip T. Hoffman, Jean-Laurent Rosenthal, and Kenneth L. Sokoloff ); and The Cambridge Economic History of the United States (with Robert E. Gallman). Cambridge Histories Online © Cambridge University Press, 2011 T H E C A M B R I D G E WO R L D H I S TO RY O F S L AV E RY General editors David Eltis, Emory University Stanley L. Engerman, University of Rochester Volume I: The Ancient Mediterranean World Edited by Keith Bradley and Paul Cartledge Volume II : ad 500–ad 1420 Edited by David Eltis and Stanley L. Engerman Volume III : ad 1420–ad 1804 Edited by David Eltis and Stanley L. Engerman Volume IV : ad 1804–ad 2000 Edited by Seymour Drescher, David Eltis, and Stanley L. Engerman Cambridge Histories Online © Cambridge University Press, 2011 TH E CA M BRIDG E WO R L D H I S TO RY O F SL AVERY VO LUME 3 ad 1420–ad 1804 Edited by DAVID ELTIS Emory University STANLEY L. ENGERMAN University of Rochester Cambridge Histories Online © Cambridge University Press, 2011 cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sa?o Paulo, Delhi, Tokyo, Mexico City Cambridge University Press 32 Avenue of the Americas, New York, ny 10013-2473, u s a www.cambridge.org Information on this title: www.cambridge.org/9780521840682 C Cambridge University Press 2011 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2011 Printed in the United States of America A catalog record for this publication is available from the British Library. Library of Congress Cataloging in Publication data Revised for volume 3 The Cambridge world history of slavery / edited by David Eltis and Stanley L. Engerman. p. cm. Includes bibliographical references and index. isbn 978-0-521-84066-8 (hardback) 1. Slavery – History. I. Eltis, David II. Engerman, Stanley L. III. Title. ht861.c34 2009 2009036356 306.3 62–dc22 isbn 978-0-521-84068-2 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate. Cambridge Histories Online © Cambridge University Press, 2011 CONTENTS page ix xi xiii List of Maps, Figures, and Tables Contributors Series Editors’ Introduction 1 Dependence, Servility, and Coerced Labor in Time and Space david eltis and stanley l. engerman 1 part i: slavery in africa and asia minor 2 Enslavement in the Ottoman Empire in the Early Modern Period ehud r. toled ano 25 3 Slavery in Islamic Africa, 1400–1800 rudolph t. ware iii 47 4 Slavery in Non-Islamic West Africa, 1420–1820 g . ug o nwokeji 81 5 Slaving and Resistance to Slaving in West Central Africa roquin ald o f erreira 111 6 White Servitude w i l l i a m g . c l a r e n c e - s m i t h a n d da v i d e l t i s 132 part ii: slavery in asia 7 Slavery in Southeast Asia, 1420–1804 kerry ward 163 8 Slavery in Early Modern China pamela kyle crossley 186 v Cambridge Histories Online © Cambridge University Press, 2011 vi contents part iii: slavery among the indigenous americans 9 Slavery in Indigenous North America leland donald 217 10 Indigenous Slavery in South America, 1492–1820 neil l. whitehead 248 part iv : slavery and serf d om in eastern europe 11 Russian Slavery and Serfdom, 1450–1804 richard hellie 275 12 Manorialism and Rural Subjection in East Central Europe, 1500–1800 edgar m elton 297 part v: slavery in the americas 13 Slavery in the Atlantic Islands and the Early Modern Spanish Atlantic World william d. phillips, jr. 325 14 Slavery and Politics in Colonial Portuguese America: The Sixteenth to the Eighteenth Centuries j o a?o fragoso and ana rios 350 15 Slavery in the British Caribbean philip d. morgan 378 16 Slavery in the North American Mainland Colonies lorena s. walsh 407 17 Slavery in the French Caribbean, 1635–1804 laurent dubois 431 18 Slavery and the Slave Trade of the Minor Atlantic Powers pieter emmer 450 Cambridge Histories Online © Cambridge University Press, 2011 contents vii part vi: cultural and demographic patterns in the americas 19 Demography and Family Structures b. w. higman 479 20 The Concept of Creolization richard price 513 21 538 Black Women in the Early Americas betty wood part vii: legal structures, economics, and the movement of coerced peoples in the atlantic world 22 Involuntary Migration in the Early Modern World, 1500–1800 dav id richardson 563 23 Slavery, Freedom, and the Law in the Atlantic World, 1420–1807 sue peabody 594 24 European Forced Labor in the Early Modern Era timothy coates 631 25 Transatlantic Slavery and Economic Development in the Atlantic World: West Africa, 1450–1850 joseph e. inikori 650 part viii: slavery and resistance 26 Slave Worker Rebellions and Revolution in the Americas to 1804 mary turner 677 27 Runaways and Quilombolas in the Americas m a n o l o fl o r e n t i n o a n d m a? r c i a a m a n t i n o 708 741 Index Cambridge Histories Online © Cambridge University Press, 2011 Cambridge Histories Online © Cambridge University Press, 2011 LIST OF MAPS, FIGURES, AND TABLES maps 9.1 12.1 14.1 14.2 Culture Areas of Indigenous North America East Central Europe, ca. 1500 Brazil, Eighteenth Century Portuguese Empire in America, Eighteenth Century page 216 296 348 349 figures 27.1 27.2 27.3 27.4 27.5 Sambabaia Quilombo River of Perdition Quilombo Quilombo on a Tributary of the Perdition River Ambrozio Quilombo Sam Gonc?alo Quilombo 730 731 732 733 734 tables 10.1 Debts to be collected by the postmaster of Cuyuni 14.1 Distribution of registered slave baptisms: Sa?o Gonc?alo, 1651–1668 15.1 Slave populations of the British Caribbean, 1650–1830 15.2 Annual percentage decline (and increase) in the slave populations of the British Caribbean, 1627–1825 18.1 Imports of slaves in Dutch Brazil, 1630–1653, by African region of origin 18.2 Suriname’s trade balance/balance of payments, 1766–1776, average per year 18.3 The Dutch slave trade, 1600–1800 18.4 Distribution of slave departures from Africa on Danish vessels 20.1 The African origins of Suriname slaves 22.1 Involuntary migration in the Old World, 1500–1800, estimates and projections ix Cambridge Histories Online © Cambridge University Press, 2011 262 363 383 384 456 461 465 468 518 570 x list of maps, figures, and tables 22.2 Africans and whites taken to the Americas, 1500–1800, by subperiods 22.3 National participation in transatlantic slave trade, 1500–1800 22.4 Numbers of slaves shipped by African region of departure, all carriers, 1500–1800 25.1 Merchandise carried to the African Coast by the Mary in 1684 25.2 Cowries carried to the Gold Coast from Britain, 1827–1850 (three-year averages in tons) 25.3 Distribution of commodities carried to the Bight of Benin from Britain, select years, 1681–1724 25.4 Distribution of commodities carried to the Bight of Biafra from Britain, select years, 1661–1791 25.5 Distribution of commodities carried to the Bights of Benin and Biafra from Britain, select years, 1828–1850 27.1 Demographic profile of slaves in Taubate? (1730–1830) and Rio de Janeiro (1789–1835) 27.2 Demographic profiles of escaped slaves advertised in newspapers in the Caribbean and the southern United States (1730–1805) 27.3 Population estimates of some Minas Gerais quilombos (1766–1770) Cambridge Histories Online © Cambridge University Press, 2011 574 582 586 665 666 666 667 668 709 714 726 CONTRIBUTORS Ma?rcia Amantino, Department of History, Universidade Federal do Rio de Janeiro, Brazil William G. Clarence-Smith, Department of History, School of Oriental and African Studies, UK Timothy Coates, Department of History, College of Charleston, USA Pamela Kyle Crossley, Department of History, Dartmouth College, USA Leland Donald, University of Victoria, British Columbia, Canada Laurent Dubois, Department of History, Duke University, USA David Eltis, Department of History, Emory University, USA Pieter Emmer, Department of History, Leiden University, Netherlands Stanley L. Engerman, Departments of Economics and History, University of Rochester, USA Roquinaldo Ferreira, Department of History, University of Virginia, USA Manolo Florentino, Department of History, Universidade Federal do Rio de Janeiro, Brazil Joa?o Fragoso, Department of History, Universidade Federal do Rio de Janeiro, Brazil Richard Hellie, University of Chicago, USA (deceased) B. W. Higman, Department of History, Australian National University, Australia Joseph E. Inikori, Department of History, University of Rochester, USA Edgar Melton, Department of History, Wright State University, USA Philip D. Morgan, Department of History, Johns Hopkins University, USA xi Cambridge Histories Online © Cambridge University Press, 2011 xii contributors G. Ugo Nwokeji, Department of African American Studies, University of California, Berkeley, USA Sue Peabody, Department of History, Washington State University, USA William D. Phillips, Jr., Department of History, University of Minnesota, USA Richard Price, Department of Anthropology, College of William and Mary, USA David Richardson, Wilberforce Institute for the Study of Slavery and Emancipation, University of Hull, UK Ana Rios, Department of History, Universidade Federal do Rio de Janeiro, Brazil Ehud R. Toledano, Department of History, Tel Aviv University, Israel Mary Turner, Institute of Commonwealth Studies, University of London, UK Lorena S. Walsh, Colonial Williamsburg Foundation, USA (retired) Kerry Ward, Department of History, University of Michigan, USA Rudolph T. Ware III, Department of History, University of Michigan, USA Neil L. Whitehead, Department of Anthropology, University of Wisconsin, USA Betty Wood, Department of History, Cambridge University, UK Cambridge Histories Online © Cambridge University Press, 2011 SERIES EDITORS’ INTRODUCTION This is the third volume of The Cambridge World History of Slavery, exploring the various manifestations of coerced labor in Africa, Asia, and the Americas between the opening up of the Atlantic world and the formal creation of the new nation of Haiti. Slavery has been among the most ubiquitous of all human institutions, across time and place, from earliest history until, some would argue, the present day. Yet its durability and ubiquity are not widely recognised and, where they are, they seem poorly understood by the general public and scholars alike. A central aim of these volumes, which cover many different times and places, is to help to place the existence and nature of slavery against the backdrop of the broader human social condition. Slavery has appeared in many different forms and is not always easy to separate from other forms of coerced labor. Nevertheless, there are basic similarities that emerge from the contributions that follow. Most critical of these is the ownership of one human by another, and the ability to buy and sell the human chattel such ownership creates. A second common characteristic is the fact that chattel status is a heritable condition passed down through the mother. Such characteristics are not to be found in the more general category of ‘coerced labor’, as normally practiced. The latter typically involves a general loss of citizenship rights, but not necessarily ownership of one person by another and inherited status. Some scholars regard slavery as part of a spectrum of coerced labor and dependency, but the institution has maintained a distinctive legal existence in almost all societies. xiii Cambridge Histories Online © Cambridge University Press, 2011 Cambridge Histories Online © Cambridge University Press, 2011 CHAPTER 1 DEPENDENCE, SERVILITY, AND COERCED LABOR IN TIME AND SPACE david eltis and stanley l. engerman Slavery is generally regarded as the most extreme form of dependency and exploitation. This project attempts to cover types of dependency in addition to slavery, although it is clear from both the overall title and the program for the project’s third volume that slavery gets considerably more attention than do other types of dependency. This reflects in part the modern preoccupation with individual freedom and equality before the law accorded by citizenship now acknowledged, at least as an ideal, just about everywhere in the modern world. Slavery may not be completely eradicated today, but it had lost irrevocably the ideological struggle perhaps as early as the first half of the nineteenth century, with only minor rearguard actions (in ideological terms, that is) in the antebellum South and less certainly in Hitler’s Germany and the Soviet gulags. Such a circumstance – amazing in its rapidity and completeness from a worldwide historical perspective of human behavior and beliefs – is taken for granted today. The more complete the victory of the view that slavery should not exist nor should have ever existed, the more remote slavery itself appears, but at the same time the greater the modern fascination with the institution becomes. And the more remote it appears, the easier it is to treat slavery simply as an evil practiced by evil men, and the harder it is to understand it in human terms. At the very least, modern preoccupations with freedom and individual rights drive the fascination with slavery. This phenomenon, an outcome of the Enlightenment, shapes the form of the modern assault on slavery. General explanations of the rise and fall of slavery have not fared well in recent years, as the great resources thrown into the study of slavery from primary sources have revealed the richness and complexity of the institution. As this suggests, such explanations tend to date from an era predating our present age of extensive empirical research, and for the most part focus on slavery – or rather separate slavery from other forms of dependency – counter to what we wish to do. Such explanations are quite good at describing how slavery functions but are weakest at accounting for first, its rise, second, its fall, and third, why at times nonslave dependency (for instance, serfdom) emerges as more important than chattel slavery. 1 Cambridge Histories Online © Cambridge University Press, 2011 2 the cambridge world history of slavery Most important of all, perhaps, they fail to explain the eligibility issue – in other words, why certain peoples are seen as qualifying for slavery (whereas others are not), and why this changes over time. This last issue has become of much greater interest in the last decade or so, as the realization spreads that all peoples in the world have been at some time in their history both slaves and owners of slaves, often at one and the same time. Having dismissed general theories, we will nevertheless mention three of them here as sometimes helpful. There is the general Marxist position, implicit in the work of those who followed Marx, if not Marx himself, who had little to say on the subject, which in broad terms takes the position that any ruling class would wish to impose slavelike conditions on the rest of society and is prevented from doing so only by resistance on the part of the potential slaves. This position is tempered by an argument – quite incorrect, in our view – that chattel slavery is not compatible with industrialization because, in crude terms, advanced capitalism needs consumers and skilled workers who respond to incentives. Thus, it is argued, slavery exists when conditions hobble the ability of people to resist enslavement and tends to disappear with the onset of industrialization. A second general position is that of Jack Goody, who accepts the overwhelming power element of the previous argument but interprets it in terms of states rather than classes. This has the advantage of recognizing that most peoples in history have not enslaved full members of their own society and have sought slaves from elsewhere. It also projects to the level of the state the explanation Adam Smith offered for slavery at the personal level, which was “man’s love to domineer.” Such an impulse would probably hold for both states and individuals even if using free rather than slave labor might lead to more profits. Based mainly on his study of African societies, Goody offers the general proposition that any time a state was significantly more powerful than its neighbors, one could expect the powerful state to use the weaker as a source of slaves. A third general explanation is the now well-known Nieboer-Domar hypothesis that focuses on the environment. It is a land-labor argument that elegantly lays out the social consequences of land abundance. In short, it holds that slavery will tend to emerge in such an environment because one cannot have free land (in other words, a frontier open for settlement), free workers willing to work for wages, and a nonworking land-owning class at the same time. Only two of these three elements can exist at once. Hence serfdom emerged in early modern Eastern Europe, and slavery emerged in the Americas. We find this persuasive, but there is nothing to account for why serfdom emerged and not slavery (and vice versa), why slavery never appeared in many land-abundant environments – especially hunter-gatherer societies – and why slavery disappeared in the Americas at least several generations before the closing of the land frontier on the two continents. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 3 Instead of dwelling further on these general theories, we would like, at least at this stage of the project, to note the different forms of forced dependency that have existed, as well as some common patterns in the institution of slavery and how these have changed over time. If we are to gain any insight into slavery, however, it must be assessed as part of a continuum of dependency typically seen as occupying the opposite pole from free labor and separated from it by such institutions as indentured servitude, convict labor, debt peonage, and serfdom, to mention just a few of the intervening categories. Institutionalized dependency and servitude had been accepted without question in Western and non-Western cultures alike, from the dawn of recorded history until the modern historical era, and they have formed one of the basic institutions that have appeared in almost every culture. Earlier discussions of dependency, and more specifically slavery, where they occurred, were couched in terms of how individual slaves should be treated, who should be a slave, and how one could fall into or lose slave status, but not whether the institution itself should exist. Moreover, however firmly the modern mind sees free labor as the antithesis to slavery, free labor arguably did not exist at all until the nineteenth century in the sense of the master-servant contract being enshrined in civil rather than criminal law. For example, free labor emerged first in the United States. As late as 1875 in England, a worker who refused to comply with the terms of his contract was viewed as stealing from the employer. Indeed, when the post-emancipation British West Indies colonial authorities introduced what the Colonial Office in London regarded as a harsh labor code, it was pointed out that the new code was basically adapted from the British Master and Servant Act. More recently, Kevin Bales has estimated that 27 million slaves lived in the late-twentieth-century world. It is possible to question the definition he uses – it appears to cover a range of dependency relations rather than chattel slavery per se – but even accepting it for the moment, 27 million constitutes far less than 1 percent of today’s global population. Two and a half centuries ago (as Arthur Young, among others, pointed out), a definition of “unfree status” similar to that employed by Bales would have encompassed a majority share of the mid-eighteenthcentury’s working population, whereas a definition of free labor in the modern sense would have covered few, if any, waged workers in 1750 or in any preceding era. Broadly, then, institutionalized coercive relationships, whether for profit or for some more overtly social purpose, were normal before the nineteenth century and have diminished rather dramatically since. Perhaps the first step is to recognize changes in the way societies have defined the various forms of dependency. Thus, as already hinted, even the nature of free labor has changed substantially within the confines of the period to which volume three of the present project is devoted – waged Cambridge Histories Online © Cambridge University Press, 2011 4 the cambridge world history of slavery labor in seventeenth-century England (and even in mid-nineteenth-century America) being taken as a sign that an individual could not possibly be a full citizen. Among the more overt forms of dependency and coerced labor, convict labor (in the sense of those guilty of offenses being required to labor) by the state has increased dramatically since the early modern period. Prior to this, and in many non-Western environments long afterward, those guilty of crimes against the community might be physically chastised or expelled. Punishment had few implications for labor. In Western societies, physical chastisement came to be supplemented by, or in some instances replaced with, incarceration, and expulsion became systematized into transportation. In both cases, however, convicts were frequently expected to labor as well. The Siberian case is well known. Exile was stipulated as early as 1582, but the forced labor of exiles is an eighteenth-century phenomenon, with, in the British case, a rapid switch from colonial North America to the antipodes as the place of exile. The most striking example is perhaps Australia, where shortly before the ending of transportation in the 1850s, convicts brought halfway around the world formed a similar proportion of the total population as had slaves in South Carolina less than a century earlier, and a far greater proportion than was ever the case in Siberia. They were also responsible for much of the infrastructure that accelerated the economic development of Australia. Despite this, the exaction of labor was never the major reason for the creation of convicts in the first place, or even, after conviction, for the existence of schemes that used the labor of those convicted, such as workhouses, prison gangs, galleys, soviet gulags, and transportation to distant colonies. Indeed, the history of coerced labor in the context of the history of the community’s or state’s need to punish transgressors seems a story of lost economic opportunity. One possible reason for this is that few schemes to harness the labor of convicts appeared to have warranted the expenditures they incurred – at least within the norms that most societies regarded as acceptable for the treatment of convicts. If convicts had been treated like African slaves, then there might have been different economic consequences. In classical times, prisoners of war were probably the major source of slaves, especially in the early expansionary days of the Roman Empire, as was also the case more recently in Africa and the indigenous Americas. Historically, capture in war has always been a justification of slavery. If a victor has the power to end a person’s life, then presumably the victor also has the power to inflict social death, or slavery, as opposed to biological death. A typical pattern at the conclusion of a battle was to inflict the latter on adult males and the former (slavery) on women and children. Such behavior is observed in the struggles between core states in Western Europe and the peoples that spearheaded the great migration prior to the fall of the Roman Empire and on down to the early Middle Ages. It was Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 5 also prevalent in struggles between most premodern polities everywhere in the world. The first effect of the emergence of large states and empires – whether in China, Mesoamerica, or the aforementioned case of Rome, where state structures allowed the control of men as well as of women and children – was that men, too, became slaves. Yet in the European world, treatment of prisoners of war changed rather decisively around the twelfth century, as relative equality of power between European states (and also between Islamic and Christian powers) and the attendant fear that the defeated power might be the victor in the next conflict meant that gradually more and more prisoners of war came to be exchanged or ransomed. Yet when Western European nations ended enslavement of one another, they still carried on extensive warfare resulting in large-scale deaths, rape, and pillaging. Whatever the reason, there is almost no evidence of prisoners of war being enslaved in the European Atlantic world during the era of American slavery, and indeed, no indication of servitude of any length being exacted by the victors in the many intra-European wars of the era (except, perhaps, for Dutch prisoners being put to work draining the English fens in the seventeenth century for the duration of hostilities). The major exception was prisoners of civil wars and those on the Celtic fringe that resisted the expansionary impulses of the core states of Western Europe, they were sent in large numbers to American plantations, at least in the seventeenth century, but always as servants with fixed terms rather than as chattel slaves, and with offspring who were free. Debt bondage was a form of servitude based upon an initial agreement to borrow funds and continued until the time, if ever, the debt was repaid. The debt was payable by the family of the borrower if the latter was unable to repay while alive. Lenders were accused of extending too much credit or charging an excessively high interest rate so that repayment was never possible. The borrower would therefore become bound for very long periods, perhaps for life. Debt bondage was a system of coercion sometimes associated with the post-chattel-slavery era, as manifested in nineteenthcentury India, but it was practiced widely and in some cases earlier in other parts of Southeast Asia, as well as in Latin America, Africa, and China. Serfdom has a history going back to at least ancient Greece and formed the basis of agricultural production and rural social structure alike in Western European medieval countries. The classic explanations of its rise, in what might be called its first resurgence in the aftermath of the fall of the Roman Empire, allow for some peasant agency. The feudal contract provided some protection from marauding invaders for those working the land in return for feudal obligations to the lord, who provided the security. From the late fifteenth to seventeenth centuries, serfdom went through a second renaissance in Eastern Europe and, on a much smaller scale, in Cambridge Histories Online © Cambridge University Press, 2011 6 the cambridge world history of slavery Scotland after seemingly heading toward insignificance in the West. Both the scale and the intensity (that is, the restrictions applied to the peasant) increased in the east through to the eighteenth century, as the Russian and Prussian states extended the area under their control eastward. By late in that century, there were probably more serfs in Europe, including Russia, than ever before. Expansion also meant that the term “serf ” came to cover a much wider range of servile relationships than earlier. Serfdom may have disappeared in Scandinavia, England, and the Netherlands, but in most parts of Western Europe, including Germany and France, peasants still owed residual obligations to landholders. Indeed, in Germany, such obligations acted as a major restraint on German migration to both east and west, as German peasants had to compensate their lords before they could legally migrate. Peasant support for the early stages of the French Revolution is testimony enough to the significance of similar obligations west of the Rhine. The new “full” serfdom that developed in Eastern Europe from the sixteenth century varied somewhat from its Western predecessor. Although primarily a means of ensuring that landholders would have a supply of labor, and the state a pool of potential soldiers, a new form of serfdom also showed up, stripped of its military aspects, in mines in Scotland, Germany, and even in the lead mines of Elizabethan England. In the Scottish case, valuations of the collieries reflected the number, age, and sex of the serf workforce in a way familiar to those who have studied probate records or deeds in plantation regions in the Americas. In addition, the second serfdom showed much less evidence of the contractual (implicit or otherwise) basis for serf status that historians have seen in its Dark Ages predecessor. The new lands acquired by an expanding Russian state were taken from indigenous, mainly Turkic, peoples and remained highly insecure. Hundreds of thousands of Russians and other Slavic peoples fell victim to slave raids and died in servitude in Islamic and Christian Middle Eastern regions, as indeed the origin of the term “slave” suggests. Nevertheless, there is little sense of a contractual relationship between the peasant on the one hand and the state, or the local pomeschiki class in Russian history, on the other. The expansion of serfdom occurred overwhelmingly at the initiative of an expanding militaristic state. Equally important, some Eastern serfs came to have fewer ties with the land in law, in the sense that both state and seigneurial peasants in Russia could be forcibly moved to new lands in a way that would not have been imaginable in medieval Western Europe, and which was redolent of chattel-slave status. Under such circumstances – given the heritability of serf status – drawing a legal or behavioral line between serf and slave status becomes difficult. If the resurgence of serfdom in the east changed the nature of serfdom, completely new forms of coercive relationships appeared in northwestern Europe. The aforementioned master-servant contract, as it evolved in Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 7 the aftermath of the Great Plague, recognized the right of the master to physically chastise the servant and charge the servant with theft in the event that the latter did not meet the terms of the contract. From the broad global perspective, what is extraordinary about such a relationship is the voluntary nature of the initial contract and the fact that it could be renewed at least once a year. Rural fairs in northwestern Europe became not just markets for surplus produce but, late in each year, nascent markets for labor as well. In the global history of dependency and coercive labor, this was a watershed in the evolution of agency on the part of those without property or without kin. The evolution of the master-servant relationship has received very little attention, at least from the comparative perspective. Equally unique in global terms was the system it spawned for facilitating large-scale transoceanic travel. As it evolved in England, the master-servant contract provided the initial basis for the repeopling of the Americas, and much later, the first large-scale movement of Asian peoples to the semitropical Americas. In its first manifestation, it came to be called indentured servitude; in its second, contract labor. In both cases, there was a largely voluntary contract in which individual workers gave up several years of their working lives in return for the cost of passage. During the period of the contract, there were clear analogies with slavery in that the contract could be sold and severe restrictions placed on the rights of the worker to move or to avoid the obligations incurred. Once more, the full weight of the criminal law was applied against the servant for noncompliance, but not against the master. The length of the term of labor required appears to have varied closely with key variables such as the age and skill level of the laborer and the distance (and thus the cost) of the migrant’s passage. Major change occurred within the slavery category over the centuries preceding its abolition. There are, arguably, three aspects of slave societies that at a preliminary view are to be found across cultures, although the incidence and distribution of these forms do seem to vary in a systematic fashion. As with attempts at definition, these may seem vague and indefinite, but they help provide some analytical grounding for an important issue. First, and perhaps most common from a transglobal perspective, was slavery as a system of augmenting and sustaining the survival of the group as a social entity, whether based on some conception of kinship or set of religious beliefs. Such slavery is more likely to be “open,” that is, to provide for eventual entry into full membership of society through a process of “a gradual reduction in marginality” of either the slave or, more likely, the descendants of the slave (though the stigma of slave origins could survive for many generations). Slavery of this type could be associated with large state structures, as in many Islamic polities, or in smaller societies on either side of the shift to settled agriculture, as in the indigenous Americas and pre-nineteenth-century Africa. Cambridge Histories Online © Cambridge University Press, 2011 8 the cambridge world history of slavery A second type of slavery was, as a system, directly organized by the state to achieve communal goals – perhaps the maintenance of public works, as in irrigation systems, fortifications, or the clearing of salt deposits to permit agriculture, or to provide soldiers for offensive or defensive purposes. Examples could be found in most phases of Chinese history (referred to sometimes as “Oriental despotism”), in fourteenth- to sixteenth-century Korea, and in Ancient Egypt. Both the Janissaries of the Ottoman Empire and the genizaros of Spanish New Mexico would also qualify.1 A third type is as a system for extracting high levels of output from labor for profit of private individuals. Although the state was not directly involved as an owner (though in the early modern period, Atlantic European navies did ship some slaves across the Atlantic, and European armies bought African slaves for military purposes – galley oarsmen as well as the regular army), the state normally had to provide the legal structure for the enforcement of ownership rights of slaveholders and, ultimately, the armed force that sustained the private use of slaves. There are probably no occupations that have been performed by nonslaves that have not also been performed by slaves, yet historically, some activities have clearly had a larger slave component than others. Concentration of slaves in particular tasks may be attributed broadly to the ability of nonslaves to avoid activities that were particularly unpleasant. For two centuries after the mid-seventeenth century, field labor on plantations in the Americas was evidently one such activity. In some societies in the classical era, the focus on production did not preclude the eventual entry of some slaves into mainstream society. We can probably all think of cases that fit none of these three categories – the tribute slaves that came into the Aztec Empire from the north, many of whom ended up as sacrificial victims, to provide one example.2 Yet some broad categorization is useful to get an analytical grasp on an institution as ubiquitous as slavery – few peoples on the globe have not at some point in their history been slaves and owners of slaves, often at the same time. Given these changing conceptions of dependency, it is somewhat tricky to evaluate the relative importance of the different forms of dependency and coercion over time. Even without such a consideration, the different types do on occasion occur together. Thus, the bulk of European convicts sent overseas before 1800 were in fact sold in the same manner as indentured servants to private owners, with only a longer term of service separating them from their nonconvict counterparts. But as social observers from 1 James F. Brooks, Captives and Cousins: Slavery, Kinship, and Community in the Southwest Borderlands (Chapel Hill, NC, 2002), pp. 121–42. 2 The historiography on slavery in the Aztec Empire is extremely thin, but see Robert D. Shadow and Maria J. Rodriguez, “Historical Panorama of Anthropological Perspectives on Aztec Slavery,” in Barbro Dahlgren and Ma De Los Dolores Soto de Arechavaleta (eds.), Arqueologia del Nort y del Occidente de Mexico: Homenaje al Doctor J. Charles Kelley (Mexico City, 1995), pp. 299–323. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 9 Aristotle to Marx and Foucault have noted, there can be no doubt that in addition to changes within a given form, major shifts have taken place in the relative importance of different forms. As already suggested, recent interpretations stress that free labor as we understand it today did not exist prior to the nineteenth century. But even understood in seventeenthcentury terms, it had neither a long history nor a very wide currency outside relatively small enclaves in Western Europe. For convicts and perhaps prisoners of war, significant numbers could not be expected before the creation of a state system and bureaucracy to maintain them and administer their activities. Galleys in the Mediterranean drew on this form of labor (as well as on nonconvict slaves) from antiquity to the eighteenth century, but it is unlikely that convicts ever formed more than a tiny share of either the labor force or, more broadly, the unfree, even in societies with sophisticated state structures. The same is true of indentured servitude and contract labor, which did not appear at all until the seventeenth century and thereafter never accounted for anything approaching majority status in any society. Serfdom, by contrast, was usually widespread if it existed at all, especially if we define it in the broadest possible way to include all relationships where individuals gained access to land to produce their own commodities in exchange for varying circumscriptions of personal actions and the acknowledgment of obligations to others. The chronology of the initial appearance of the three systems discussed in this chapter broadly follows the order in which they were described. Slavery dedicated to augmenting the numbers and sustaining the identity of societies or religions is usually associated with Islam, sub-Saharan Africa, or the indigenous Americas, but it now seems to have application for many parts of the premodern world. As that world is also largely preorthographic, historical evidence of it tends to come from oral tradition or from those post-orthographic societies with which the premodern society interacted. This means essentially that evidence of such slavery is scarce in the years before Chinese and European expansion, but there seems little reason to doubt that it existed and, indeed, may well have been universal in post-neolithic societies. More broadly, an argument might be made that the basic social structure in such environments was not class but kinship, and that slavery was a normal component of kinship structures. This is not to suggest that slavery then was widespread. Too many slaves would be likely to overwhelm the absorptive function of the institution and threaten collective identities – as indeed happened in several indigenous American societies in the aftermath of the demographic calamity triggered by Old World contact. A slave in the two later types of slave systems described earlier was usually without any rights in law and passed on his or her status to any offspring. In kin-based societies, by contrast, slaves or their descendants might gradually receive back certain rights as they Cambridge Histories Online © Cambridge University Press, 2011 10 the cambridge world history of slavery demonstrated acceptance of kinship identity through their behavior. As there is no clear dividing line between slave and nonslave, assessments of the extent of such slavery must necessarily be fragile. Nevertheless, in the absence of severe demographic stress, people without rights at any given point in time must have formed a very small proportion of the populations of kin-based societies. From another perspective, however, one that counts as servile all those who were not full members of the kin group and were therefore in part dependents of those who were full members, then we might say that the servile would often, perhaps normally, account for the majority of the population. Systems of slavery dedicated to the extraction of labor, whether for public projects or for the production of export crops organized for the benefit of private individuals, are normally associated with stratified societies that have moved some distance beyond the agricultural revolution. When these appear, it is possible to think in terms of “slave societies” instead of “societies with slaves,” to use Moses Finley’s well-known designations. It is also probable that slavery of this type was what the major social science modelers of slavery, both Marx and Engels, Nieboer, and Domar, had in mind. Indeed, this form of slavery is what most people have in mind when they think of the subject at all, especially those who have used the term “slavery” to draw attention to abusive or exploitative labor situations from early times to the present day. Many Caribbean islands had more than three quarters of their populations as chattel slaves with no prospect of change of status prior to the abolitionist era. Brazil probably approached a point where half of its population was enslaved at several points prior to the early nineteenth century. Yet because of the absolute nature of the definition of slavery in these societies, and the rarity of any intermediary category between slavery and freedom, the proportion of the population that had full rights was actually quite high from the global historical perspective adopted here, and high, too, compared to the share of free people – using here modern definitions of freedom – that existed in the countries of Western Europe that owned these islands. Though the share of slaves in Rome, Greece, and the slave Americas was much higher than was ever the case in kinbased societies that used slavery as a way of augmenting their numbers and sustaining their identities, there have been relatively few “slave societies” in history. They appeared relatively late in human social evolution, and though they have had a very high profile in recorded history – being associated usually with imperial systems and “human progress” to borrow David Brion Davis’s ironic association – they probably never accounted for anything like the majority of slaves on the globe at any point in history. Thus, most slaves in history have experienced their servitude in what are today termed premodern social environments. It also seems highly probable that the number of slaves in the Americas has always lagged behind the number of serfs in the Old World. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 11 The advent of large-scale slave societies did not mean that the original kin-based form of slavery disappeared. The two, sometimes all three, forms of slavery existed at the same time. In the Atlantic world, some scholars argue that the kin-based system of slavery at the periphery of capitalist development in both Africa and the indigenous Americas was transformed by a burgeoning Atlantic-based market system into something more akin to slavery in the plantation Americas. Thus, by the nineteenth century, the Cherokee in the United States owned cotton plantations worked by African slaves, and slaves owned by Africans in different parts of subSaharan Africa grew peanuts and cloves for sale into the Atlantic economy. Yet the total value of such activities is so small when compared to the value of any major crop in the white-dominated plantation Americas that such a slippage into a new form of slavery cannot have been extensive. A much stronger consequence of contact between different systems was that plantation societies drew on their kin-based counterparts for slaves, first in the Americas, then on the African coast, and finally in Dutch Asia. Slaves traded between the two systems were individuals without any rights whatsoever in either sphere, but the trade ensured that they shifted from an environment where a reduction of their social marginality was possible to one in which the gradual reclaiming of rights was an unlikely eventuality. Returning to the overview of dependency and coerced labor over the very long run, we can observe three major patterns. First, though slavery was ubiquitous, the share of slaves in kin-based slave systems was not likely to have been very great. However, if we define freedom as emanating from full membership of a given society so that, first, one has the right to participate in the decision making of the kin or community in which one lives, and second, one is in possession of most of the bundle of rights that make up possessive individualism, then the share of free individuals in kin-based societies was also small. Thus, the vast majority of people in most societies in history have been neither slave nor free, once we consider the limited rights to political participation that existed, and not just freedom from labor coercion. A second pattern is the polarization process that appears to have been associated with the rise of more complex economies and imperial systems. The share of both slave and free in such societies appears to have risen sharply, and the intervening categories of dependence have almost disappeared. This observation is another way of approaching the paradox that has drawn the attention of Orlando Patterson, who has argued that our understanding, indeed awareness, of freedom was dependent on slavery.3 The lines between slave and free (defined in terms of citizenship) were clearly delineated in Greek, Roman, and, with a religious orientation, Islamic societies, too. The slave-free dichotomy was perhaps at its starkest in the Americas. 3 Orlando Patterson, Freedom in the Making of Western Culture (New York, 1991). Cambridge Histories Online © Cambridge University Press, 2011 12 the cambridge world history of slavery A third major pattern has been the rise and fall of the incidence of coercive systems in the last five centuries, in a world in which kin-based systems of slavery continued to thrive. From the fifteenth to the early twentieth centuries, systems of serfdom, slavery, convict labor, and indentured servitude expanded dramatically and in close unison. Four out of five transatlantic migrants prior to 1820 arrived in the Americas owing service to another, most of them having been physically coerced into leaving their country of origin. Yet in little more than a century, coercive migrant systems had disappeared. The last slave ship crossed the Atlantic in 1867, the last transoceanic contract labor vessel (with terms of service enforced with penal-code sanctions) arrived in 1917, and the last convicts returned from Devil’s Island to France in 1952. A related and even more important development was the virtual disappearance of all ideological justifications of inequality and dependence. In the twentieth century, there have been intense debates on the meaning of freedom, but none at all on its desirability. The net result is that from the perspective of the early twenty-first century, while inequality is clearly rife in the modern world, there is no attempt to justify it in the terms employed in the earlier debates. The ideological shift has swept away not only the American slave plantation, but, almost as comprehensively, the kin-based systems of slavery in the indigenous Americas, in Africa, and in Asia. At no point in history has the share of the global population who see themselves as full members of society been as great as it is now. Although slavery today is seen as the epitome of evil, its stigma is not entirely a function of modern conceptions of freedom. However much slavery has historically formed part of a range of dependent relations, it has tended to be regarded across cultures at best as a particularly hard and unfortunate fate, and at worst as the ultimate degradation for any human being. In many social environments, it has been viewed not as an alternative to death, but as a fate worse than death, although most societies that had some form of human sacrifice also had slavery. Individuals who sold themselves into slavery did so only as a last resort, thus suggesting that avoidance of slavery was of paramount importance to them. The stigma of a slave-ancestor in most non-Western societies was (and in many, still is) widespread. Long before the abolition process was complete, Frederick Douglass made it clear to supporters of other social reforms that antislavery should have priority because there was nothing at all to compare with its malevolent impact.4 Scholars of the social history of the colonial Americas have equated the conditions of indentured servants, convict and contract 4 David Roediger, “Race, Labor, and Gender in the Languages of Antebellum Social Protest,” in Stanley L. Engerman (ed.), Terms of Labor: Slavery, Serfdom and Free Labor (Stanford, CA, 1999), pp. 175–83. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 13 laborers, and even free wage-workers with those endured by slaves.5 Nevertheless, there could have been few slaves at any point in the history of slavery in the Americas who would have spurned an opportunity to switch their status with that of any one of these nonslave groups, just as few in history have opted to enslave themselves. The distinctiveness of slavery in historical (as opposed to modern) terms seems to lie in the close to absolute one-sidedness of power in the master-slave relationship, at least in formal legal terms. Even where slavery might offer freedom from starvation and on occasion greater life expectancy, the disutility of the institution in the form of being in the power of another was overwhelming. Nonslaves always had more protection against the power of a social superior or an employer than did slaves. In the end, social norms offered far more protection for serfs, convicts, servants, prisoners of war, contract workers, debt peons, apprentices, and the myriad other forms of dependency (including children and wives) than they did for slaves. Put another way, these groups were less marginal to society than were slaves – a conclusion that appears to hold for all societies. Even in societies where the exaction of labor was not the central function of slavery, they were less likely than slaves to be sacrificed, sold off in times of social stress, or denied rights over offspring and spouses. What follows from the uniquely degrading nature of slavery observed here is a central set of questions for the present volume. What is it that determines who is to be a slave, and how does this shift over time and between societies? Given that the potential for abolition has always existed in the sense that, in every culture, there were large numbers of people – usually the vast majority – who were considered exempt from slavery. Is abolition, then, nothing more than the extension of this exemption to everyone in a given society and, eventually, the attribution of all the characteristics of full personhood to all aliens as well? If so, then just as important as the type and function of coercion is the question of which groups are viewed as eligible for coercion, and why direct coercion has come to play a very much smaller role in the way societies function than has hitherto been the case. It is striking that few of the major models of slavery have made much effort to address the issue of eligibility for enslavement. Whether land-labor ratios (Nieboer-Domar), or power imbalances between societies (Goody), or simply the love to domineer (Adam Smith), general explanations have focused very much on the conditions under which slavery might appear or intensify, and on the prerequisites of its abolition. For most of the history of slavery, such a focus was entirely appropriate. Major centers of slavery have often drawn slaves from one particular region so that the name for slave became synonymous with the name of the dominant peoples in the 5 See, for example, Hilary McD Beckles, White Servitude and Black Slavery in Barbados, 1627–1715 (Knoxville, TN, 1989). Cambridge Histories Online © Cambridge University Press, 2011 14 the cambridge world history of slavery region of provenance. In French Canada, “panis,” an ethnic designation, was the standard name for slaves whatever the ethnicity of the slave. In colonial Brazil a similar merging came about under the term “tupi,” and the origin of the term “slave,” as is well known, comes down to us from a time when the great bulk of slaves entering the Mediterranean area were drawn from slavic regions. Yet prior to the fifteenth century, it was rare to have eligibility for enslavement defined in terms of physical characteristics or even racial constructions. Cranial deformation, or its absence, among Northwest Pacific Coast peoples comes closest, but it was never an absolute marker for slavery.6 For the great modelers of slavery, it was enough to acknowledge that slavery was associated with extreme degradation, and then move on to the social, psychological, or environmental factors that shaped how extensive the institution of slavery would be, and what form it would take. And most of the historiography on slavery has followed suit by keying on rather narrow cost-benefit considerations and power relationships between groups when addressing historical shifts in the composition of people making up slave populations – as opposed to explaining why slavery per se has existed.7 It is impossible to address the question of eligibility without taking into account how any group responsible for enslavement perceived and defined itself in relation to others. In recent decades, this has come to be known as the question of identity. Societies have tended to reserve enslavement for those whom they have defined as not belonging, but this has not always meant that all aliens were enslaved, or that all slaves were aliens. There have been many instances in history of societies generating slaves from within their own ranks, but this has usually occurred only after the potential slave has violated, or is thought to have violated, the most profoundly held norms of society. In addition, exposure of infants (parents abandoning a child), typically practiced by all social ranks, was a source of internally generated slaves in many societies, including ancient Rome and China, which suggests that some acculturation or nurturing process was a prerequisite of “belonging,” or insider status. In early Rome, citizens could be reduced to slaves, and twins in many Igbo communities were sold into the Atlantic slave trade directly from Igboland.8 It was easier to become a slave from within some societies than from within others, just as the ease of reduction of marginality (and 6 Leland Donald, Aboriginal Slavery on the Northwest Coast of North America (Berkeley, CA, 1997), pp. 94–5. 7 H. Hoetink’s work on “somatic norms,” not often cited recently, is an exception to this comment. 8 Almost all the twins in a sample of 57,000 Africans taken out of slave ships by British cruisers and landed in Sierra Leone between 1819 and 1845 were on vessels that left Bonny, New Calabar, and Old Calabar [“Nutritional Trends in Africa and the Americas: Heights of Africans, 1819–1839,” Journal of Interdisciplinary History, 12 (1982): 453–75], as were the vast majority of the small number of recaptives in the Liberated African Registers with disabilities. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 15 thus protection against the worst consequences of enslavement) varied. In parts of Europe in the Middle Ages, the problem of “excess children” was solved by infanticide and abandonment, whereas in some societies in Africa and Asia, the same issue was resolved by involuntary enslavement of such children. In the slave Americas, manumission – a clear example of reduction of marginality – was more possible in Iberian than in Englishspeaking areas, but on the other hand, the Iberian Americas were the very last to abolish slavery on the two continents – an unexpected negative correlation. The most well-known survey of slavery and kinship in an African context focuses almost entirely on the movement from outsider to insider status.9 The reverse process – in effect how an insider becomes a slave – has received little attention for any society. Generally, however, there was some formal process whereby the erstwhile insider was redefined as an outsider, or else, as in the case of Russia, owners believed that their human chattels were physically different from themselves when the reality pointed in quite the opposite direction. Nevertheless, the vast majority of slaves in history have originated from outside the group that was responsible for their enslavement. The conception of not belonging appears to form the core element of eligibility for slavery across cultures whether or not the institution functioned primarily to extract labor or to add and integrate newcomers to the slave owners’ social group. In addition, however, gender and age were major considerations at different times. Where the main aim of slavery was to augment one’s social or religious group, then women and children would likely be preferred to adult males, who, as already suggested, might be put to death immediately or, as in Tupinamba societies in Brazil, held for sacrifice at a point in the future decided by the captor. The trade in slaves across the Sahara Desert to the Islamic Mediterranean, which grew from a trickle of people in the early days of Islam to a stream ultimately rivaling in numbers the better-known transatlantic trade, was overwhelmingly female, and some of the few male slaves involved were destined to be eunuchs. As the previous discussion suggests, societies seeking to augment their numbers, and ultimately their cultures and/or religions, were extremely eclectic in their selection of potential slaves. The whole point of acquiring such slaves was not just to inflict “social death,” but also to facilitate social rebirth. The basic aim was to create a new social identity – to produce more people who in the end behaved and thought like the host group, and might fight alongside them. Children from any culture presumably have the potential for assuming new identities, and the chief purpose of preserving women after capture was to ensure a broader base for society 9 Suzanne Miers and Igor Kopytoff (eds.), Slavery in Africa: Historical and Anthropologival Perspectives (Madison, WI, 1977). Cambridge Histories Online © Cambridge University Press, 2011 16 the cambridge world history of slavery to reproduce itself in its own image. Slavery supported by major state structures could more easily cope with adult male slaves. The ultimate aim, however, was nevertheless to refashion newcomers according to the needs of the host society. Most countries in the modern world have had the same attitude toward today’s voluntary immigrants. In the United States and Europe, there is a set of core values, loosely described as Western, that all newcomers are expected to believe in and accept. Assimilation was intended to be the main outcome of the “melting pot.” The major difference between this and enslavement as traditionally enforced is that the decision to migrate today is voluntary. Entry into the new society is no longer preceded by social death – or in the case of the Mintz and Price formularization of creolization in the Americas, by the traumatic relocation inflicted by the Middle Passage. But in some cases, women and children were simply not available. From the point at which Christendom and Islamic societies reached military stalemate in early Middle Ages, slaves acquired by one from the other tended to be drawn from the mainly male world of ships and the military, and despite the fact there may have been more English slaves held in North Africa than black slaves in the English Caribbean in the second half of the seventeenth century, neither were at that time very numerous. In recent years, historians have begun to draw explicit comparisons between kinship structures and bondage in widely separated parts of the world, especially sub-Saharan Africa and the temperate areas of the indigenous Americas. Whereas Europeans carried off 12.5 million Africans to the Americas in just more than three centuries, Africans absorbed few if any Europeans into their own societies.10 In the celebrated case of Bullfinche Lambe, an Englishman was held captive by the king of Dahomey for several years and was eventually released. But the basic reason for the imbalance, apart from the fact that few Europeans could survive in sub-Saharan Africa, was the almost total absence of European women and children on the African coast, and the essentially nonconfrontational nature of the relationship between African polities and European slave traders. Not only were there few European captives, but Europeans could usually pay what was necessary to gain the release of captives before the “reduction of marginality” had proceeded very far.11 In the temperate North Americas, by contrast, there was large-scale settlement by Europeans in the aftermath of the demographic disaster that overtook the Indian population. These factors ensured that some French, English, and Spanish (and many more of Euro-aboriginal 10 David Eltis and David Richardson, “A New Assessment of the Transatlantic Slave Trade,” in idem (eds.), Extending the Frontiers: Essays on the New Transatlantic Slave Trade Database (New Haven, CT, 2008), pp. 1–60. 11 See, for example, the process described by Suzanne Schwarz, Slave Captain: The Career of James Irvine in the Liverpool Slave Trade (Wrexham, 1995). Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 17 parentage) women and children, and even some men, became absorbed into indigenous societies via the enslavement mechanism, though obviously the numbers never approached those of Indians enslaved by Europeans. Though initially Europeans sought labor, they later strove to bring about the complete assimilation of whole Indian populations in what was one of the most blatant of European attempts to recast social identity. Even African slaves were expected to take a role in white society – a partial assimilation presaged in the Iberian areas with conversion to Christianity, learning a new language, and conforming to European behavioral norms. From this perspective, the differences between slavery designed to extract labor and slavery designed to augment and perpetuate the host social group fades somewhat. If societies seeking slaves for any purpose happened to draw disproportionately on a particular region or ethnicity, this was rarely because the people of the target group were seen as having the best potential for enslavement. “Slavish personalities” attributed to slaves and serfs, or the peoples from whom slaves were drawn, have formed part of the ideology of slavery since written records first appeared. Nevertheless, any examination of the composition of slave-labor forces and conquered peoples in the major empires from ancient China down to the very early modern Atlantic empires indicates that the enslaved comprised a bewildering mix of peoples. The same was true of systems such as those in the indigenous Americas, where the Iroquois, for example, brought in so many newcomers from so many ethnic backgrounds that social identity could never become uniform in spite of the fact that enslavement and absorption were designed to achieve the opposite effect.12 Broadly, for all systems, shifting power relationships between societies, as well as environmental and demographic considerations, usually ensured a heterogeneous mix of slaves in the long run. In the late Middle Ages, this situation began to change. Partly because of the aforementioned stalemate between Islam and Christendom, and more particularly because of rising slavic, especially Russian, power in the east, the flow of slaves into Islamic areas from the north and west gradually declined. As a consequence, the relative importance of southern regions – specifically sub-Saharan Africa – as a source of slaves for Islam increased. Growing distinctions between white and black slaves are quite apparent in Islam from the ninth century if not even earlier. Different terms evolved – “abd” for all slaves, but “mamluk” came to be used for white slaves – and higher valuations for whites appear in the record, presumably reflecting the greater likelihood of whites being ransomed. The harder 12 Richard White, The Middle Ground: Indians, Empires and Republics in the Great Lakes Region, 1650–1815 (Cambridge, 1991). Cambridge Histories Online © Cambridge University Press, 2011 18 the cambridge world history of slavery tasks were assigned to blacks.13 Thus, the linking of race and slavery in the context of the Americas had little precedent in the Old World. By the nineteenth century, the overwhelming share of slaves in the Arab world was from sub-Sahara Africa, or was of sub-Saharan African descent, and the association of black skin with slavery became ever stronger. European expansion into the Atlantic world from the early fifteenth century brought, first, cheap transoceanic transportation, second, a demographic calamity in the Americas, third, the prospect of exportable quantities of precious metals and high-value crops, and fourth, labor productivity that was much higher in the New World than in the Old. The resulting transatlantic slave trade, after relatively modest beginnings in the sixteenth and seventeenth centuries, built up to hitherto unimaginable levels of mass movement of peoples in the eighteenth and early nineteenth centuries. That the vast majority of the people who made this move were unwilling Africans was precisely because the peoples of Europe had some choice over the decision to migrate. When the options in the Americas narrowed to working on a sugar estate – the major reason for people crossing the Atlantic for nearly three centuries after the late 1500s – then voluntary migrants avoided the plantation labor force option and did their moving within Europe rather than between continents, but when alternative forms of agriculture developed, free-labor migration was renewed on a much larger scale. Once again, the fact that conceptions of freedom in Europe had shifted to permit most individuals control over the decision to migrate, plus, as described later, the invisible barrier that prevented Europeans from enslaving other Europeans, generated more coercion and more slavery for non-Europeans. In both the Atlantic and, less certainly, the Islamic world, for the first time eligibility for enslavement began to be defined not in terms of which group to exclude, but rather which groups to include. Muslims debated the issue extensively, and in addition, had a formal proscription against enslaving other Muslims (as opposed to automatically manumitting those enslaved chattels who converted to Islam). The Spanish debated the enslavement of indigenous Americans, but the striking feature of the establishment of African slavery in the Americas was the set of underlying assumptions about who could be enslaved. Indeed, the absence of a major debate – except for the Spanish case – is probably responsible for the failure of historians to explore the eligibility issue in the European context. There was nothing in European history to suggest either a turning away from coercion or restrictions on who should be subject to enslavement. Europe was a conglomeration of competing polities with extensive written records of military conflicts, civil wars, and the repression of minority systems of thought, especially religious thought. Most states evolved unequal 13 Bernard Lewis, Race and Color in Islam (New York, 1971), pp. 63–4. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 19 and rigid social structures early on that fostered some slavery and a great deal of serfdom, as well as centralized judicial systems that meted out punishments for wrongdoing and heresy that appear harsh compared to slavery itself. Slavery had been extensive in Roman times, and for nearly six hundred years after the fall of Rome, a slave trade from the less developed north, west, and east of Europe sent a stream of slaves drawn from various European peoples to the more prosperous areas of the south and the Mediterranean – increasingly Islamic after the seventh century.14 Relations between European polities and the fringe areas of Europe, especially the marauding leading edges of the Great Migrations, had a large enslavement component on both sides, and the system of serfdom thought to have developed as a response to these pressures was clearly related to slavery. In addition, in the late Middle Ages, plagues reduced western European populations by one-third and created a large shortage of labor. But despite all the precedents and pressures that appeared to point to more coercion and more slavery, between the eleventh and fifteenth centuries, the internal European slave trade dried up completely, slavery itself diminished significantly throughout Europe, and the institution disappeared altogether in the north and west of the subcontinent. More remarkably, when Europeans expanded into new lightly populated, land-abundant territories, the overseas component of that expansion in the West (but curiously, not its eastern counterpart) demonstrated that Europeans were prepared to enslave the peoples they found in those territories, and to relocate millions of others – Indians, Africans, and Asians alike. Demographic collapse, and in the Spanish case some ideological reservations, soon eliminated indigenous Americans as slaves,15 and though the Dutch carried Asians to South Africa as slaves in the seventeenth and eighteenth centuries, and all Europeans used enslaved Asians in Asia in mainly domestic contexts, only distance from the Americas ensured that no Asians worked in Caribbean cane fields prior to the nineteenth century. In early Brazil, the Portuguese drew extensively on Indian communities for slave labor to produce sugar. The Spanish used variations of corve?e labor as well as some Africans to exploit precious mineral deposits of New Spain and South America. In the Chesapeake, the English in the seventeenth century had some Indian slaves and many peoples of African descent who were not only not slaves, but full members of society as well. French Canadians were prepared to buy African slaves but could not afford them, and they 14 See William D. Phillips, Slavery from Roman Times to the Early Transatlantic Trade (Minneapolis, 1984) for the main slave trade routes across Europe down to the eleventh century. 15 Presumably if the Spanish had developed a large export sugar industry in the sixteenth century, like the Portuguese, their reservations on the use of Indian slaves would have been more in tune with those of the Portuguese. Cambridge Histories Online © Cambridge University Press, 2011 20 the cambridge world history of slavery ended up with a slave-labor force that was exclusively Indian.16 But from the mid-seventeenth to the early nineteenth centuries, the economic center of gravity of the Americas was in the Caribbean and subtropical South America – a centering that was made possible by a labor force of slaves who were exclusively African or of African origin. Slavery here came to have a racial exclusivity that was more pronounced than in the Islamic world. In both Islam and Europe, a core culture characterized a wide variety of ethnicities and languages as major elements of a common identity, and therefore, insider status. Both Islam and Europe traveled a road that began with the exemption of its own members from slavery, and by the end of the period, both cultures were conferring eligibility for enslavement on only one of the many groups of “others,” or outsiders – Africans. But was the next step in this process of redefining eligibility necessarily the elimination of the final category such that slavery was seen as inappropriate for any human being? The intensity and depth of abolitionism in parts of the West suggests that the final step was more likely to happen (or to happen first) in the Atlantic rather than in the Islamic world, and within regions where slavery played a marginal role. As Adam Smith described it in the case of the Pennsylvania Quakers, the demand curve for emancipation could be downward sloping, emancipation occurring first where slavery was less important. But interpreting the economic argument is itself often difficult, and the current literature has provided at least three arguments – the first, that slave owners found the institution unprofitable, the second, that while still profitable, slavery was less so than sectors of the economy drawing on free labor, and third, that slavery came to play an ever-diminishing role in the major slave-owning nations and could be abolished without serious implications. Each of these arguments has different implications for the nature of slavery. The debates on the causes of abolition have perhaps drawn too sharp a dichotomy between economic and other (moral, religious, cultural) factors in the process. The greater the costs of emancipation, whether because of the ongoing profitability of the slave system, or the costs of compensating slaveholders, the more likely emancipation will be delayed. Nevertheless, whatever the cost, unless there is a moral argument of some kind pointing to the need for slavery to end, the institution will continue. Even where the costs to ending slavery were low, this situation alone has never by itself led to abolition, nor apparently have high costs ever permanently prevented it from happening. Compensation, though not always paid, was always an issue. First, should it be paid (and to whom, and in what form)? Here the answers were generally clear – compensation to the slave owners in cash, bonds, or labor time. Second, should emancipation 16 Robin Winks, Blacks in Canada: A History (New Haven, CT, 1971), chapter 1. Cambridge Histories Online © Cambridge University Press, 2011 dependence, servility, and coerced labor 21 be immediate or gradual, and what should be the role of government in the process? An argument for gradual emancipation goes back to Jean Bodin and Condorcet – if slavery is as destructive to slave psychology and culture as the antislavery argument claims, then immediate freedom without some adaptive period or government control could only be disastrous. In no case were slaves or serfs ever provided with compensation by their owners or the state, a pattern that reflects the belief in the importance of property rights. There were a few discussions of compensation to freed people at the end of the U.S. Civil War, but this was not a source of major debate. Claims for compensation to the descendants of the enslaved (reparations) developed as an issue only in the twentieth century. Viewing abolition through the lens of social identity does offer some prospect of finessing these older debates, as well as coming to terms with the continuance of slavery in those parts of Asia and Africa that viewed slavery as an integral part of societies organized around kin groupings. Such an approach also reduces the distance between slave systems dedicated to the exaction of labor and those whose aim is to augment the social group. The former always attempted some assimilation, and the latter always had labor needs, the most unpleasant of which were invariably performed by slaves or those who were most marginal to society. Cambridge Histories Online © Cambridge University Press, 2011 Cambridge Histories Online © Cambridge University Press, 2011 PART I SLAVERY IN AFRICA AND ASIA MINOR Cambridge Histories Online © Cambridge University Press, 2011 Cambridge Histories Online © Cambridge University Press, 2011 CHAPTER 2 ENSLAVEMENT IN THE OTTOMAN EMPIRE IN THE EARLY MODERN PERIOD ehud r. toledano introduction From the middle of the fifteenth century until its demise after World War I, the Ottoman Empire was arguably the most important Islamic power on the face of the earth. At the height of its expansion, it ruled a vast territory from the western Mediterranean to the Persian Gulf, from southern Poland to southern Sudan. Many of the sultan’s subjects were not Muslim, did not speak Ottoman Turkish, and were illiterate, poor, and lived in villages, not in cities. Yet they were all governed by a Muslim, Turkish-speaking, urban, affluent, and predominantly male elite of officeholders. Perhaps the only phenomenon that cut across all these social barriers was enslavement, for despite the at times enormous differences in lifestyle, enslaved persons came from all walks of life: They were male and female, rich and poor, powerful and powerless, rural and urban, Muslim and non-Muslim, and speakers of all the dialects in the empire, with origins as far-flung as central Africa and the eastern Caucasus. What united them was a shared legal status of bondage, with the variety of social impediments it entailed in each predicament.1 Perhaps more than anything else, it was this me?lange of types that made Ottoman enslavement unique, complex to study and explain, and highly intriguing as a social phenomenon. For its significance lay mostly in its social and cultural aspects rather than its role in the Ottoman economy. Whereas practically all historically known societies – including Islamic ones – enslaved individuals either from within or from outside their boundaries, few had evolved such a stratified and highly diversified unfree population. If until the early seventeenth century most of the enslaved were prisoners of war, from that point on – but mainly in the eighteenth and nineteenth centuries – people were reduced to slavery through capture and trade. The main reason for this shift in recruitment was rather simple – territorial expansion as a result of military conquest ceased almost completely, 1 For the detailed arguments underlying this essay, see the following books by Ehud R. Toledano: As If Silent and Absent: Bonds of Enslavement in the Islamic Middle East (New Haven, CT, 2007); and Slavery and Abolition in the Ottoman Middle East (Seattle, 1998). 25 Cambridge Histories Online © Cambridge University Press, 2011 26 the cambridge world history of slavery and few prisoners of war were available, although some were still being enslaved even as late as the nineteenth century. Hence, during the period of contraction and diminishing military successes, demand for unfree labor had to be met through an evolving network of trafficking in humans. It was then too that the ethnic makeup of the enslaved population in the empire shifted according to the changing origins – from the Balkans and eastern Europe to central and eastern Africa and the Caucasus, largely Circassia and Georgia. Scattered data and reasonable extrapolations regarding the volume of the slave trade from Africa to the Ottoman Empire yield an estimated number of approximately 16,000 to 18,000 men and women who were being transported into the empire per annum during much of the nineteenth century.2 Estimates for the total volume of coerced migration from Africa into Ottoman territories are as follows: from Swahili coasts to the Ottoman Middle East and India – 313,000; across the Red Sea and the Gulf of Aden – 492,000; into Ottoman Egypt – 362,000; and into Ottoman North Africa (Algeria, Tunisia, and Libya) – 350,000. If we exclude the numbers going to India, a rough estimate of this mass population movement would amount to more than 1.3 million people. During the middle decades of the nineteenth century, the shrinking Atlantic traffic swelled the numbers of enslaved Africans coerced into domestic African markets, as well as into Ottoman ones. Although the numbers were possibly smaller during the eighteenth century, the patterns of trade remained fairly stable, with seasonal shifts occurring as a result of local factors. Such were the internal wars within Africa between rival Muslim states, as between them and non-Muslim ones, which resulted in the enslavement of large numbers of men and women. Changing economic conditions on the continent also affected the reduction of individuals to slavery as a result of debt or the inability of dependent entities to pay the tributes imposed on them in cash or kind. Brigandage on the overland routes and corsair activity on the high seas also affected the traffic, as did circumstances on the northern Black Sea shores and in the Caucasus. One thing remained fairly constant through most of the eighteenth and nineteenth centuries – Ottoman demand for unfree labor, 2 The most reliable work on this is by Ralph Austen: “The 19th Century Islamic Slave Trade from East Africa (Swahili and Red Sea Coasts): A Tentative Census,” in William Gervase Clarence-Smith (ed.), The Economics of the Indian Ocean Slave Trade in the Nineteenth Century, Special Issue of Slavery and Abolition, 9 (1988): 21–44; and “The Mediterranean Islamic Slave Trade out of Africa: A Tentative Census,” Slavery and Abolition, 13 (1992): 214–48. See also Thomas M. Rick’s thorough consideration in “Slaves and Slave Traders in the Persian Gulf, 18th and 19th Centuries: An Assessment,” in ibid, 60–70. For Lovejoy’s higher numbers and criticism of Austen’s figures, “Commercial Sectors in the Economy of the Nineteenth-Century Central Sudan: The Trans-Saharan Trade and the Desert-Side Salt Trade,” African Economic History, 13 (1984): 87–95; see also Paul E. Lovejoy, Transformations in Slavery: A History of Slavery in Africa, 2nd edn (Cambridge, 2000), chapter 7. Cambridge Histories Online © Cambridge University Press, 2011 enslavement in the ottoman empire 27 mainly of domestic and menial workers. Agricultural slavery, which had been widely practiced until the sixteenth century, was abandoned thenceforward until it reappeared in the second half of the nineteenth century. Even then, only in Egypt during the American Civil War was agricultural bondage used to supply global demand for cotton, and in the 1860s, the empire had also absorbed a large population of Circassian agricultural slaves who were pushed by the Russians with their landlords out of the Caucasus. Immigration and emigration in the Ottoman Middle East and North Africa have not been an unusual phenomenon. In a region still supporting large nomadic-pastoralist communities of various ethnicities – Turcoman and Bedouins immediately come to mind – inbound and outbound movements of people have been a common feature of history. People moved in and out both as groups and as individuals. They brought with them their languages, religions, and cultures. They interacted with the already diversified populations in the empire; they left their mark, contributed their share, and enriched and were enriched by the me?lange of traditions that permeated these lands, littorals, river basins, and mountains. Out of all this wealth of human experiences, our concern here is with the trade in enslaved Africans and Circassians transported into the Ottoman Empire during the eighteenth and nineteenth centuries. However, the traffic, which has been treated in a number of studies, will be examined here from a rather different perspective: not as an economic or political phenomenon, but as a question of cultural inflow, interaction, and fusion. Indeed, as has been publicly proclaimed recently, the forced movement of enslaved persons was “one of the largest migrations of history,” and also one of its “greatest crimes.” A number of important social and cultural insights concerning enslaved Africans and Circassians in the Ottoman Empire can be gained by examining their forced transportation as a type of migration. For our purposes here, such an approach to the Ottoman slave trade yields some interesting cultural insights, for instance, the view that migration tends to occur at certain junctions in the life cycle, thus becoming “a part of the rites de passage.” The evidence shows that the overwhelming majority of enslaved persons who were taken into the Ottoman Empire, whether African or Circassian, were very young, often in their early to midteens. Their coerced recruitment into the Ottoman unfree labor market occurred, in many cases, just as they were passing into puberty, entering the workforce, and – for the young females – also becoming sexually active either as concubines in urban households or as wives in the countryside. Although they would have also gone through these passages in their origin societies, their enslavement meant that all this took place amidst the heightened stresses of resocialization and reacculturation in unfamiliar surroundings, without the support of family and friends, and without the Cambridge Histories Online © Cambridge University Press, 2011 28 the cambridge world history of slavery comforting and soothing effects that a home culture would normally provide. The important elements from that perspective are the construction of individual and collective identity, the redefinition of notions such as origin/home and destination/host cultures and their interrelationship, and the reevaluation of the concepts of struggle, conflict, choice, and agency with regard to the enslaved. The main divide within the enslaved population in the Ottoman Empire was between elite and nonelite slaves, or rather between military-administrative slaves and their female consorts or wives (hereafter kul/harem slaves) on one hand, and the rest of the unfree laborers, that is, domestic, agricultural, and menial bondsmen and bondswomen, on the other. This has raised the question whether or not the kul/harem class should be considered in the same category with the other enslaved populations. Alternative terms have been suggested to describe the predicament of people in that group, including “the sultan’s servants” and “state servitors.”3 Others have rightly argued that “the privileges of elite slavery were temporary,” because they were not allowed to bequeath their wealth or status to their offspring, and their property reverted to the treasury upon their death. The sultan controlled his unfree servants’ religious and cultural identity, their material environment, and their right to life, which he could take if he believed they had betrayed his trust. As one writer put it, elite slavery was “a paradox at the heart of the Ottoman system”; that is, that ordinary subjects enjoyed immunity from the sultan’s direct power of life and death, which was denied to those who governed them, namely, kul/harem slaves.4 Although certain elements of kul/harem servitude were gradually removed in practice toward and during the nineteenth century, all legally bonded subjects of the sultan should – for the purpose of social analysis – be treated as enslaved persons. In fact, there was no difference of kind between kul/harem and other types of Ottoman slaves, although there certainly were differences of degree among them. By and large, the unfree can be classified according to four main criteria, which determined their position in society and affected their treatment and fortune. The first criterion was the tasks the enslaved performed – whether they served as domestic, agricultural, menial, or kul/harem. Second was the stratum of the slaveholders – whether they were employed by urban elite members, rural notability, small cultivators, artisans, or merchants. The third was location – whether they lived in core or peripheral areas. Finally, type of habitat – urban, village, or nomad – was of central importance. 3 See Metin Kunt, All the Sultan’s Servants: The Transformation of Ottoman Provincial Government 1550–1650 (New York, 1983); Suraiya Faroqhi, “The Ruling Elite between Politics and ‘the Economy,’” in Halil Inalcik and Donald Quataert (eds.), An Economic and Social History of the Ottoman Empire 1300–1914 (Cambridge, 1994), pp. 564 ff. 4 Leslie Peirce, Morality Tales: Law and Gender in the Ottoman Court of Aintab (Berkeley, CA, 2003). Cambridge Histories Online © Cambridge University Press, 2011 enslavement in the ottoman empire 29 In addition, two other parameters have to be factored in, namely, gender and ethnicity. Generally, women were more exposed to sexual exploitation, whereas men were more vulnerable to harsh physical treatment. And, more often than not, enslaved Africans ranked socially lower than enslaved Circassians and Georgians. Some preliminary generalizations emerge from this matrix. Enslaved domestic workers in urban elite households were better treated than enslaved people in other settings and predicaments. The greater the distance from the core, the lower the enslaved were in the social strata, and the less densely populated was the habitat, the greater the chances the enslaved would receive worse treatment. Finally, the lives of enslaved Africans and enslaved women were more often than not harder. Thus, for example, it follows that women in urban elite households – where arguably enslavement was the mildest – could be, and not infrequently were, exposed to uncomfortable, compromising situations, which we might call today sexual harassment. The single most important factor that sustained a fairly stable demand for unfree labor within the Ottoman Empire was the constant dwindling of the enslaved population and the absence of a capacity to replenish the supply of slaves internally. Both should be attributed to sociocultural practices that must be considered as mitigating circumstances of Ottoman – and more generally Islamic – enslavement. According to Islamic law as practiced in the Ottoman Empire during the period reviewed here, enslaved women could be absorbed into the slaveholding society through concubinage. There were no injunctions against cross-racial or cross-cultural unions. If an enslaved concubine became pregnant, it was illegal to resell her, and if she gave birth, her child was considered free and she was to be manumitted upon the death of the father. Although little choice on the woman’s part existed in such cases, the status that she came to possess in effect provided her with considerable protection. It also meant that such women and their offspring would regularly disappear from the enslaved population, reducing its size. In addition, an Islamic moral encouragement to manumit enslaved persons after long service – in Ottoman practice this meant, on average, seven to ten years – was largely observed, although not by all slaveholders. Again, this imperative constantly released individuals from legal bondage, even if they often chose to remain as free servants in the same or another household. At the same time, on the supply side there was in Ottoman societies an absence of slave-breeding practices. Enslaved persons were not married to each other in order to produce enslaved children for the household or the farm. When such marriages occurred it was usually between freed persons, socially and financially encouraged by their former slaveholders to marry within their ethnic group after manumission, as an act of benevolent Cambridge Histories Online © Cambridge University Press, 2011 30 the cambridge world history of slavery patronage. With such strongly entrenched social mechanisms, demand for slaves would persist on the same level as long as slavery remained legal, that is, until the demise of the empire and the establishment of the Turkish Republic in the 1920s. Attempts to abolish Ottoman slavery and suppress the slave trade into the empire were launched during the second half of the nineteenth century but do not form a major theme of this chapter. Suffice it here to note that British efforts to induce the Ottomans to suppress the slave trade from Africa and the Ottoman government’s own actions against the traffic in Circassians and Georgians produced a significant reduction in the volume of the traffic toward the turn of the twentieth century. But these measures also pitted two value systems – the European and the Ottoman – against each other, producing a whole set of dilemmas for Ottoman intellectuals and statesmen. Part of the reason why the Ottoman elite did not enthusiastically embrace British-style abolitionism lay in different perceptions, indeed, in a bifurcated view of the Ottoman version of enslavement itself. Compared to other modes of dependency in Ottoman society, slavery was not necessarily the worst, which is probably true for most other premodern societies. Significant social disabilities, reflected in law and practice, were part of everyday life – in varying degrees – for all women as well as for all non-Muslims. Poor people often suffered greater deprivation than many enslaved people, and marriage was also a form of male ownership over women. Military service in the sultan’s armies did not always have a clear end in sight for the common soldier. Peasants frequently worked for bare subsistence, and when pushed under that line, had to abandon the land and fend for themselves as brigands or nomadic beggars in the nearby desert for food and shelter. Consequently, it has been argued that many slaves were better off – that is, better cared for – than many of the sultan’s free subjects, especially in material terms. Many slaves, it was further asserted, would not have traded their position for the uncertainties and vulnerabilities of the free poor and other marginals in Ottoman societies. Still, slavery is rightly considered to be the most extreme form of domination. There were other, at times quite harsh, forms of coerced denial of freedom, such as incarceration or indentured labor. Even in its mild forms, slavery seems to remain such a stark instance of deprivation and coercion that it stands apart from the other phenomena of “unfreedom.” Hence, what is perhaps sometimes hard to grasp, or even simply to realize, is that even under enslavement, the capacity of slaveholders to extract labor was not unlimited, nor was the slave’s powerlessness absolute. A better understanding of slavery can be gained only if we conceive of it as an involuntary relationship of mutual dependence between two quite unequal partners. Within this broad definition, there were certainly cases in which slaves had little impact on their lives, as were other situations in which they had a Cambridge Histories Online © Cambridge University Press, 2011 enslavement in the ottoman empire 31 great deal of influence vis-a?-vis their masters. In all cases, the slave’s ability to stand her or his ground in the relationship depended on the extent to which she or he could withhold labor in order to attain their minimal requirements. In other words, their agency depended on denial of services, whether in the fields, the mines, or the household – the latter including sex, ...