Available via license: CC BY 4.0
Content may be subject to copyright.
Edited by Damian A. Pargas · Juliane Schiel
The Palgrave Handbook
of Global Slavery
throughout History
The Palgrave Handbook of Global Slaver y throughout
History
Damian A. Pargas · Juliane Schiel
Editors
The Palgrave
Handbook of Global
Slavery throughout
History
Editors
Damian A. Pargas
Institute for History
Leiden University
Leiden, The Netherlands
Juliane Schiel
Institute for Economic
and Social History
University of Vienna
Vienna, Austria
ISBN 978-3-031-13259-9 ISBN 978-3-031-13260-5 (eBook)
https://doi.org/10.1007/978-3-031-13260-5
© The Editor(s) (if applicable) and The Author(s) 2023. This book is an open access publication.
Open Access This book is licensed under the terms of the Creative Commons Attribution
4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use,
sharing, adaptation, distribution and r eproduction in any medium or format, as long as you
give appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license and indicate if changes were made.
The images or other third party material in this book are included in the book’s Creative
Commons license, unless indicated otherwise in a credit line to the material. If material is not
included in the book’s Creative Commons license and your intended use is not permitted by
statutory regulation or exceeds the permitted use, you will need to obtain permission directly
from the copyright holder.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher
nor the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regar d to jurisdictional claims in published maps and institutional affiliations.
Cover illustration: narvikk/Getty Images
This Palgrave Macmillan imprint is published by the registered company Springer Nature
Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Acknowledgments
This publication is based upon work from COST Action CA18205—Worlds
of Related Coercions in Work (WORCK), supported by COST (European
Cooperation in Science and Technology).
COST (European Cooperation in Science and Technology) is a funding
agency for research and innovation networks. Our Actions help connect
research initiatives across Europe and enable scientists to grow their ideas
by sharing them with their peers. This boosts their research, career and
innovation. https://www.cost.eu.
The publication is co-funded by the Faculty of Historical and Cultural
Studies of the University of Vienna.
v
Contents
1Introduction: Historicizing and Spatializing Global Slavery 1
Damian A. Pargas
Part I Ancient Societies (to 500 C.E.)
2Mesopotamian Slavery 17
Seth Richardson
3Ancient Egyptian Slavery 41
Ella Karev
4Slavery in Ancient Greece 67
Kostas Vlassopoulos
5Slavery in the Roman Empire 87
Noel Lenski
6Injection: An Archaeological Approach to Slavery 109
Catherine M. Cameron
Part II Medieval Societies (500–1500 C.E.)
7Slavery in the Byzantine Empire 123
Youval Rotman
8Slavery in Medieval Arabia 139
Magdalena Moorthy Kloss
9Slavery in the Black Sea Region 159
Hannah Barker
10 Slavery in the Western Mediterranean 179
Juliane Schiel
vii
viii CONTENTS
11 The Question of Slavery in the Inca State 195
Karoline Noack and Kerstin Nowack
12 Injection: A Gender Perspective on Domestic Slavery 215
Ruth Mazo Karras
Part III Early Modern Societies (1500–1800 C.E.)
13 Slavery in the Mediterranean 227
Giulia Bonazza
14 Slavery in the Ottoman Empire 243
Hayri Gök¸sin Özkoray
15 Slavery in the Holy Roman Empire 259
Josef Köstlbauer
16 Slavery and Serfdom in Muscovy and the Russian Empire 279
Hans-Heinrich Nolte and Elena Smolarz
17 Slavery in Late Ming China 297
Claude Chevaleyre
18 Slavery in Chos˘on Korea 319
Sun Joo Kim
19 Slavery in the Indian Ocean World 339
Titas Chakraborty
20 Maritime Passages in the Indian Ocean Slave Trade 359
Pedro Machado
21 The Rise of Atlantic Slavery in the Americas 379
Michael Zeuske
22 Plantation Slavery in the British Caribbean 395
Trevor Burnard
23 Injection: Atlantic Slavery and Commodity Chains 413
Klaus Weber
Part IV Modern Societies (1800–1900 C.E.)
24 The Second Slavery in the Americas 429
Michael Zeuske
25 Slavery in the US South 441
Damian A. Pargas
26 Slavery in the Middle East and North Africa 459
Ismael M. Montana
CONTENTS ix
27 Slavery in Islamic West Africa 479
Jennifer Lofkrantz
28 Urban East African Slavery 497
Michelle Liebst
29 Slavery in South Asia 517
Emma Kalb
30 Slavery in Southeastern Europe 535
Viorel Achim
31 Injection: The Global Spread of Abolitionism 553
William Mulligan
Part V Contemporary Societies (1900–Present)
32 American Slaveries Since Emancipation 567
Catherine Armstrong
33 Slavery in Francophone West Africa 583
Benedetta Rossi
34 Slave Labor in Nazi Germany 605
Marc Buggeln
35 State-Introduced Slavery in Soviet Forced Labor Camps 625
Felicitas Fischer von Weikersthal
36 North Korean Slavery and Forced Labor in Present-Day
Europe 643
Remco E. Breuker
37 Modern Slavery in the Global Economy 661
Bruno Lamas
38 Injection: Modern Slavery and Political Strategy 681
Joel Quirk
39 Conclusion: Situating Slavery Studies in the Field
of Global History 701
Juliane Schiel
Index 709
NotesonContributors
Viorel Achim is a Senior Researcher at the Nicolae Iorga Institute of History,
Romanian Academy, Bucharest. His research fields include the history of the
Gypsies (Roma) in Romania, slavery in the Romanian lands, ethnic minorities
in Romania between 1918 and 1948, population policies in Romania during
World War II, and the Holocaust. He published five author books (one of
which was also published in three translations, and another in one translation)
and over 100 book chapters and journal articles. His most cited book is The
Roma in Romanian History (Budapest, New York: Central European Univer-
sity Press, 2004). He also edited three volumes of archives documents and
edited or co-edited four collective volumes. In 2007, he was a postdoctoral
fellow at the Gilder Lehrman Center for the Study of Slavery, Resistance, and
Abolition (Yale University). In recent years, together with a small team from
the Nicolae Iorga Institute of History, he has been studying slavery, aboli-
tionism, and emancipation from slavery in the Romanian principalities between
ca. 1830 and ca. 1860.
Catherine Armstrong is Reader in Modern History at Loughborough
University and Director of People and Culture for the School of Social
Sciences and Humanities there. She has recently published a monograph with
Cambridge University Press entitled American Slavery, American Imperialism:
US Perceptions of Global Servitude 1870–1914. She is currently working on
a book length project investigating the forgetting and misremembering of
chattel slavery in the aftermath of the American Civil War.
Hannah Barker is an Associate Professor of medieval history at Arizona State
University. She studies connections between the Black Sea and the Mediter-
ranean during the thirteenth through fifteenth centuries, especially the slave
trade which flourished during this period and the transmission of plague
leading to the Black Death. Her book, That Most Precious Merchandise: The
Mediterranean Trade in Black Sea Slaves, 1260–1500 (2019), received the
xi
xii NOTES ON CONTRIBUTORS
Paul E. Lovejoy Prize, the ASU Institute for Humanities Research book prize,
and honorable mentions for the Mediterranean Seminar’s Wadjih F. al-Hamwi
Prize and the Middle East Medievalists book prize. Other recent publications
include “The Risk of Birth: Life Insurance for Enslaved Pregnant Women in
Fifteenth-Century Genoa,” Journal of Global Slavery 6 (2021) and “Laying
the Corpses to Rest: Grain Embargoes and the Early Transmission of the Black
Death in the Black Sea, 1346–1347,” Speculum 96 (2021).
Giulia Bonazza is currently Marie Skłodowska-Curie Global fellow at
Columbia University (NY) and Ca’ Foscari University (Venice) with a project
titled The Darker Shades of Black. The Value of Skin Colour in the Mediter-
ranean and Atlantic Slave and Labour Markets, 1750–1886. She is a former
fellow at the German Historical Institute in Rome (2018) and a former Max
Weber postdoctoral fellow at the European University Institute (Florence)
(2016–2017).
Remco E. Breuker is Professor of Korean Studies at Leiden University, the
Netherlands. A historian, he has published extensively on Korean and North-
east Asian medieval history. Next to his work on historical Korea, he also works
on contemporary North Korea, in particular on exile narratives, human rights
issues, and DPRK overseas economic activities. A present research focus is
the presence of North Korean forced labor and modern slavery in interna-
tional supply chains. He is (co-) author of among other titles: Forging the
Truth: Creative Deception and National Identity in Medieval Korea (ANU,
2009), Establishing a Pluralist Society in Medieval Korea (918–1170): History,
Ideology and Identity in the Kory˘oDynasty (Brill, 2010), People for Profit:
North Korean Forced Labour on a Global Scale (edited with I.B.L.H. van
Gardingen, LeidenAsiaCentre, 2018), Invented Traditions in North and South
Korea (edited with Andrew Jackson, Codrut
,a Sîntionean, and CedarBough
Saeji, Hawai’i University Press, 2022), and Samguk yusa: Forgotten Legacies
of the Three Kingdoms (with Boudewijn Walraven, Grace Koh, and Frits Vos,
Hawai’i University Press, forthcoming 2023).
Marc Buggeln is Assistant Professor at the Friedrich-Meinecke-Institute at
the Free University Berlin and Privatdozent at the History Department at
Humboldt-Universität zu Berlin. His main research interests are: German and
European history in the nineteenth and twentieth centuries, concentration
camps, (forced) labor, National Socialism, social inequality, tax and budget
policy, governmentality. His publications include: Das Versprechen der Gleich-
heit. Steuern und sozialeUngleichheit in Deutschland seit 1871 [The Promise
of Equality.Progressive Taxation and the Reduction of Social Inequality since
1871] (Berlin: Suhrkamp, 2022); and Slave Labor in Nazi Concentration
Camps (Oxford: Oxford University Press, 2014).
Trevor Burnard is the Wilberforce Professor of Slavery and Emancipation
at the University of Hull and the Director of the Wilberforce Institute. He
is the author of Mastery, Tyranny, and Desire: Thomas Thistlewood and His
NOTES ON CONTRIBUTORS xiii
Slaves in the Anglo-Jamaican World; Planters, Merchants, and Slaves: Planta-
tion Societies in British America, 1650–1820; The Plantation Machine: Atlantic
Capitalism in French Saint-Domingue and British Jamaica; Jamaica in the
Age of Revolution; and Writing Early America: From British Empire to the
American Revolution (Charlottesville: University of Virginia Press).
Catherine M. Cameron is professor emerita in the Department of Anthro-
pology at the University of Colorado. She has worked in the northern part
of the American Southwest, focusing especially on the Chaco and post-Chaco
eras (AD 900–1300). Her research interests include prehistoric demography,
the evolution of complex societies, and processes of cultural transmission. She
has worked in southeastern Utah at the Bluff Great House, a Chacoan site,
and nearby Comb Wash, publishing a monograph on this research in 2009
(Chaco and After in the Northern San Juan, University of Arizona Press). She
also studies captives in prehistory, especially their role in cultural transmission.
She edited a volume on this topic in 2008 (Invisible Citizens: Captives and
Their Consequences , University of Utah Press). Her single-authored volume
on captives was published in 2016 (Captives: How Stolen People Changed the
World, University of Nebraska Press).
Titas Chakraborty is Assistant Professor of History at Duke Kunshan Univer-
sity. Her research investigates slavery and unfree work and their relationship to
colonial state formation in seventeenth-, eighteenth- and nineteenth-century
South Asia and the Indian Ocean world. Her work has appeared in Slavery and
Abolition, International Review of Social History, and Journal of Social Histor y.
She is also the co-editor of Global History of Runaways: Workers, Mobility, and
Global Capitalism, 1650–1850 (University of California Press, 2019). She is
currently completing a book titled “Empire of Labor: Hired Workers, Mobility
and the Rise of the East India Company State 1651–1817.”
Claude Chevaleyre is CNRS researcher at the Institute of East Asian Studies
in Lyon. He is also affiliated to the Bonn Center of Dependency and Slavery
Studies. He has a Ph.D. in Chinese history from EHESS (Paris) and an MA
in Chinese studies from INALCO (Paris). He has published articles and book
chapters on bondage and trafficking in early modern China and is currently
heading a project entitled “China Human Trafficking and Slaving Database.”
Felicitas Fischer von Weikersthal is a Postdoctoral Researcher and Lecturer
in the History of Eastern Europe at the University of Heidelberg. Her
research interests include gulag studies, cultures of memory, and the history of
terrorism. She has published a monograph on camp newspapers and cultural
education in the early Soviet forced labor camps (Die “inhaftierte” Presse. Das
Pressewesen sowjetischer Zwangsarbeitslager 1923–1937 , 2011) and several arti-
cles dealing with reeducational tendencies of the camps system, the role of
sports, and Gulag economy as part of war preperation.
xiv NOTES ON CONTRIBUTORS
Emma Kalb completed her Ph.D. in South Asian Languages and Civilizations
at the University of Chicago in May 2020 and is currently a postdoctoral
researcher at the Bonn Center for Dependency and Slavery Studies at the
University of Bonn. Her current book project is entitled “Unequal Intimacies:
Gender, Slavery and Servitude in the Mughal World.”
Ella Karev is a Postdoctoral Teaching Fellow in the Department of Near
Eastern Languages and Civilizations at the University of Chicago, an Egyp-
tologist and papyrologist whose research focuses on the social and economic
aspects of enslavement and bound labor throughout Egyptian history with a
particular focus on the Late and Ptolemaic Periods. She additionally inves-
tigates body modification for both enslaved and unenslaved persons, and is
preparing a book on scars as an identifying feature in Graeco-Roman papyri.
Ruth Mazo Karras is Lecky Professor of History at Trinity College Dublin.
She is the author of five monographs and editor of three volumes as well
as author of dozens of articles and book chapters on a variety of topics
in medieval social history and the history of gender and sexuality. She
has discussed medieval slavery in Slavery and Society in Medieval Scandi-
navia (1988) and in Unmarriages: Women, Men, and Sexual Unions in the
Middle Ages (2012).
Sun Joo Kim is Harvard-Yenching Professor of Korean History in the Depart-
ment of East Asian Languages and Civilizations at Harvard University. She has
a broad range of research interests in social and cultural history of Chos ˘on
Korea (1392–1910) including regional history of the northern part of Korea,
popular movements, historical memory, micro history, history of emotions,
law and society, slavery, and art history.
Josef Köstlbauer is a historian based at the Bonn Center for Dependency and
Slavery Studies (University of Bonn). His publications cover a broad range
of topics, from historical semantics of dependency to slavery in early modern
Europe, colonial borderlands, the history of digital games, and baroque alle-
gorical painting. Together with Rebekka von Mallinckrodt and Sarah Lentz he
recently published a volume titled Beyond Exceptionalism: Traces of Slavery and
the Slave Trade in Early Modern Germany, 1650–1850 (DeGruyter 2021).
Bruno Lamas an architect by training, worked in the fields of urbanism and
urban and regional planning between 2004 and 2017. He is currently a fellow
of the FCT, the Portuguese national funding agency for science, research, and
technology, and a Ph.D. student in Economic and Organizational Sociology
in UL-ISEG, Lisbon School of Economics and Management, developing the
dissertation “The Metamorphoses of Modern Slavery: Labour, Self-ownership
and the Problem of Slavery in the History of Capitalism.”
Noel Lenski is the Dunham Professor of Classics and History at Yale Univer-
sity. His previous scholarship includes Failure of Empire: Valens and the
Roman State in the Fourth Century AD (Berkeley 2002), Constantine and
NOTES ON CONTRIBUTORS xv
the Cities: Imperial Authority and Civic Politics (Philadelphia 2016), and the
co-edited volume What is a Slave Society? The Practice of Slavery in Global
Perspective (Cambridge 2018).
Michelle Liebst is a Postdoctoral Research Fellow and historian of East Africa
specializing on the themes of religion, labor, humanitarianism, gender, and
citizenship. She is the author of Labour and Christianity in the Mission (James
Currey, 2021), and based at UCL, she works on the ERC-funded project,
“African Abolitionism: The Rise and Transformations of Anti-Slavery in
Africa” (AFRAB).
Jennifer Lofkrantz is an Associate Professor of History at Gulf University for
Science and Technology, Kuwait. She specializes in the history of precolonial
West Africa and the contemporary history of northern Nigeria, particularly on
Muslim West African debates on legal and illegal enslavement, the ransoming
of war prisoners, and historical and contemporary jihadism.
Pedro Machado is a global and Indian Ocean historian with interests in
commodity histories, labor and migratory movements, and the social, cultural,
environmental, and commercial trajectories of objects. He is based at Indiana
University, Bloomington, and is the author of several works, among which
are Ocean of Trade: South Asian Merchants, Africa and the Indian Ocean,
c. 1750–1850 (Cambridge University Press, 2014); “Views from Other Boats:
On Amitav Ghosh’s Indian Ocean ‘Worlds,” American Historical Review,Vol.
121, No. 5 (December 2016); Textile Trades, Consumer Cultures and the
Material Worlds of the Indian Ocean (Palgrave Macmillan, 2018); and Pearls,
People and Power: Pearling and Indian Ocean Worlds (Ohio University Press,
2020). He is currently at work on a global history of pearl shell collection and
exchange while also developing resear ch on eucalyptus and colonial forestry in
the Portuguese empire in the nineteenth and twentieth centuries.
Ismael M. Montana received his Ph.D. from York University in Toronto. He
is an Associate Professor of History at Northern Illinois University. Montana
is the author of The Abolition of Slavery in Ottoman Tunis (Gainesville, 2013)
and a co-editor with Paul E. Lovejoy and Behnaz M. Asl of Islam, Slavery,
and Diaspora (Trenton, 2009). His r esearch interests include the social and
economic history of slavery, culture, and citizenship in Northwest Africa and
the Western Mediterranean Basin from the eighteenth century to the present.
Magdalena Moorthy Kloss is a researcher at the Austrian Academy of
Sciences’ Institute for Social Anthropology specialized in Yemen, slavery,
marginalized groups, historical anthropology, and gender. She holds a Master’s
degree and Ph.D. in anthropology from the University of Vienna, and a
Master’s degree in anthropology and development from the London School
of Economics and Political Science. Her latest article is entitled “Eunuchs
at the Service of Yemen’s Ras¯ulid Dynasty (626–858/1229–1454)” and was
xvi NOTES ON CONTRIBUTORS
published in Der Islam (Vol. 98/1 (2021). She previously worked for the
United Nations and GIZ in Yemen and Egypt.
William Mulligan is a Professor in the School of History at UCD. He has held
visiting fellowships at the Institute for Advanced Study, Princeton, and the
Wissenschaftskolleg, Berlin. He has published widely on international history
in the nineteenth and twentieth centuries and is editor, with Maurice Bric, of A
Global History of Anti-Slavery Politics in the Nineteenth Century (Palgrave,
2013).
Karoline Noack holds a Ph.D. and habilitation in cultural anthropology (Freie
Universität Berlin). Since 2009, she has been Professor of Anthropology of
the Americas at the University of Bonn and Director of the BASA Museum
(Bonn Collection of the Americas). Her focal points include research on social
and cultural phenomena in the Andes in the longue durée, museum and
collection research, also with “source communities.” She conducts depen-
dency research in the Inca state and is currently researching on Cochabamba
in Bolivia, a hotspot of mobilization and production strategies in the late
Inca state, within the framework of the Bonn Center of Dependency and
Slavery Studies (BCDSS), of which she is one of the deputy speakers. Her
most recent book publications, together with other colleagues, include “From
‘Bronze Rooster’ to Ekeko. Impulses for Ethnological Provenance Research in
University Collections and Museums” (2020), “Global Turns, descolonización
y museos” (2020), “Digitization of Ethnological Collections. Perspectives
from Theory and Practice” (in German) (2021), and various articles on the
above-mentioned topics.
Hans-Heinrich Nolte is Emeritus Professor of Eastern European history at
the University of Hannover. His work has focused primarily on Eastern Euro-
pean history and the history of the USSR, but Nolte is also consider ed a
pioneer of global histor y. He is the author of numerous articles and nine
monographs, including Weltgeschichte des 20. Jahrhunderts (2009) and most
recently Kurze Geschichte der Imperien (2017).
Kerstin Nowack obtained her Ph.D. in Cultural Anthropology of the Amer-
icas at the University of Bonn. Nowack is an independent scholar, working
on the ethnohistory of the Inca and of early colonial Peru. Her main r esearch
interests are the politics and history of the Inca empire, war, religion, and
knowledge in the Inca imperial system, Inca elites before and after the Spanish
conquest, and sources on the Inca, on which she has published numerous arti-
cles. She is currently preparing the publication of a book about the Inca civil
war.
Hayri Gök¸sin Özkoray is Assistant Professor of Early Modern History at the
Aix-Marseille Université (France). He deals mainly with the history of slavery
and labor in the Ottoman Empire (fifteenth to seventeenth centuries).
NOTES ON CONTRIBUTORS xvii
DamianA.Pargas is Professor of North American History at Leiden Univer-
sity as well as Director of the Roosevelt Institute for American Studies.
Specialized in the history of slavery, he has published numerous articles and is
the author or editor of six books, including Slavery and Forced Migration in
the Antebellum South (Cambridge University Press, 2014) and most recently
Freedom Seekers: Fugitive Slaves in North America (Cambridge University
Press, 2021). Pargas is also the founder of the Journal of Global Slavery.
Joel Quirk is Professor of Politics at the University of the Witwatersrand. His
work focuses on enslavement and abolition, work and mobility, social move-
ments, gender and violence, historical repair, and the history and politics of
Africa. Joel is the author or co-editor of eight books, and has also co-edited
special issues of Social & Legal Studies, Review of International Studies, The
Anti-Trafficking Review,and Slavery & Abolition. He is a founding editor of
openDemocracy’s Beyond Trafficking and Slavery.
Seth Richardson is a historian of the ancient Near East who focuses on
Mesopotamia’s Old Babylonian period (ca. 2000–1600 BC) and the specific
problem of the Fall of Babylon in 1595 BC. His work addresses political-
economic questions of state sovereignty such as legitimacy, territory, rebellion,
membership, and community practices of violence. He is an Associate at the
Oriental Institute of the University of Chicago and the Managing Editor of
the Journal of Near Eastern Studies.
Benedetta Rossi is Professor of History at University College London.
She is author of From Slavery to Aid (CUP, 2015) and editor of Reconfig-
uring Slavery: West African Trajectories (LUP, 2009, 2016); LesMondesde
l’Esclavage (Seuil 2021, with C. Vidal, dir. P. Ismard); “Developmentalism,
Labor, and the Slow Death of Slavery in Twentieth-Century Africa,” Special
Issue, ILWCH 92 (2017); and “Slavery and Marriage in African Societies,”
Special Issue, Slavery & Abolition 43 (2022, with J. Quirk). She holds the
ERC Advanced grant “African Abolitionism: The Rise and Transformations of
Anti-Slavery in Africa” (AFRAB, grant agreement no. 885418).
Youval Rotman is a social historian of the Byzantine Mediterranean world,
and teaches history in the department of Jewish history at Tel Aviv University.
His scholarship focuses on social and cultural processes in the Mediterranean
and Near East in the first Millennium. He is the author of Byzantine Slavery
and the Mediterranean World (Harvard University Pr ess, 2009), published
first in French (Les Belles Lettres, 2004), which contests the idea of the decline
of ancient slavery, ascribing a central role to slavery in the medieval transfor-
mation of the Byzantine world. His second book Insanity and Sanctity in
Byzantium: The Ambiguity of Religious Experience (Harvard University Press,
2016), analyzes religious forms of abnormality as a motor of social change.
His third monograph (ARC Humanities Press, 2021), is a study of compar-
ative history of slavery. It narrates the formation of the Middle Ages from
the point of view of slavery, and challenges the traditional dichotomy between
xviii NOTES ON CONTRIBUTORS
ancient and medieval slaveries, revealing the dynamic, versatile, and adaptable
character of slavery to changing circumstances. Prof. Rotman is the co-editor
of the Mediterranean Historical Review.
Juliane Schiel is Associate Professor for Economic and Social History in
Late Medieval and Early Modern Times at the University of Vienna. She has
published widely on the history of slavery in the Mediterranean as well as on
the history of labor and coercion. Her work is informed by a micro historic
approach, the methods of historical semantics and global history perspectives.
Since 2019, she is chairing the COST Action CA18205 “Worlds of Related
Coercions in Work” (WORCK), assembling about 150 social and labor histo-
rians, anthropologists, and sociologists from about 40 countries worldwide
(https://www.worck.eu).
Elena Smolarz is a postdoctoral researcher at the Bonn Center for Depen-
dency and Slavery Studies. As a research associate of competence network
“Crossroads Asia,” funded by the German Ministry for Education and
Research (BMBF), she conducted studies on coerced mobility and slave trade
networks in Central Asia in the eighteenth and nineteenth centuries. Her
main research interests include serfdom in the Russian Empire, representa-
tions of freedom and dependency in archival documents and narrated sources
as well as transformation of autochthonous societies in Southern borderland
regions under Russian imperial influence with special focus on spatial and social
mobility.
Kostas Vlassopoulos is Associate Professor of Ancient History at the Univer-
sity of Crete and collaborating member of the Institute for Mediterranean
Studies (IMS-FORTH). He is the author of Unthinking the Greek Polis:
Ancient Greek History beyond Eurocentrism (2007); Politics: Antiquity and its
Legacy (2009); Greeks and Barbarians (2013); ‘My whole life’: Stories from the
Daily Lives of Ancient Slaves (2020); Historicising Ancient Slavery (2021);
and Greek and Roman Slaveries (2022). He has co-edited Communities and
Networks in the Ancient Greek World (2015) and Violence and Community:
Law, Space and Identity in the Ancient Eastern Mediterranean World (2017).
Klaus Weber wrote his Ph.D. thesis on German merchants in eighteenth-
century Atlantic trade at Hamburg University. He was a research fellow at the
National University of Ireland (Galway), at The Rothschild Archive (London)
and at the Institut für die Geschichte der deutschen Juden (Hamburg). Since
2011, he has been Professor of European Economic and Social History at
the European University Viadrina in Frankfurt (Oder). Among his research
interests are labor regimes and welfare regimes in moder n Europe and the
Atlantic world, and the transnational connections of Central Europe’s early
modern proto-industries.
Michael Zeuske was educated at the University of Leipzig and obtained
his two German doctorates (Ph.D. and habilitation) at Leipzig in 1984 and
NOTES ON CONTRIBUTORS xix
1991, respectively, with works on revolutionary elites and hegemonial groups
in Latin-American independence movements. He taught as full professor at
the University of Leipzig (1992–1993) and as professor at the University of
Cologne (1993–2018). At Universität zu Köln (University of Cologne/UoC),
he was for 25 years professor of Iberian and Latin-American History. Zeuske
is a leading expert on global slavery; Alexander von Humboldt’s Amer-
ican journeys; the revolutionary elites of the independencia-movements in
Spanish America (Simón Bolívar, Francisco de Miranda, etc.); the history
of the Caribbean (especially second slavery), as well as slaveries i n Cuba,
Santo Domingo, Venezuela and Colombia; and is most widely known for
his studies on slavery and slave trades (as the author or co-author of 20
books; editor/co-editor of 30 books and special issues of journals; and some
220 articles and chapters). Since March 2018 Zeuske is professor emeritus of
Iberian and Latin-American history at the University of Cologne. Since august
2019 he is senior professor and PI of the Bonn Center for Dependency and
Slavery Studies (BCDSS) at the University of Bonn. Zeuske is currently doing
field research in Cuba, island Caribbean, Venezuela, Colombia, ports of the
Atlantic, and archival research in Spain, Portugal, USA, UK, France, Germany,
Cuba, Cape Verde Islands, São Tomé, Dakar (Senegal), and at many small
archives over the Atlantic world. In 2007 he was a research fellow at the
Gilder Lehrman Center for the Study of Slavery, Resistance, and Abolition
(Yale University, New Haven). In 2015, his research took him to Beijing,
where he was a fellow at the BeiDa University (Peking University).
List of Figures
Fig. 5.1 Grabstein aus Nickenich bei Mayen (etwa 50 n. Chr.):
Das Relief zeigt einen Sklavenhändler, der zwei Sklaven
an einer Kette führt, um sie zu verkaufen. Es ist im
LVR-Landesmuseum in Bonn zu sehen. In regard to the use
of pictorial material: use of such material in this press release is
remuneration-free, provided the source is named. The material
may be used only in connection with the contents of this press
release. For pictures of higher resolution or inquiries for any
further use, please contact the Press office publishing this
directly. https://idw-online.de/en/image?id=274118&size=
screen 91
Fig. 5.2 Bronze image of a nude ephebe from Xanten. The
boy, who would have carried an actual tray, is shown
long-haired and garlanded with his nude body right
at the end of prepubescence. Berlin, Staatliche Museen,
Antikensammlung, Sk. 4. Bildarchiv Preussischer
Kulturbesitz/Art Resource, NY 99
Fig. 5.3 Tombstone of Capua depicting the sale of an enslaved person,
stripped and standing on a catasta with auctioneer (winged,
to the right) and buyer, late first century CE. G. Fittschen
Neg. D-DAI-Rom 1983VW1305) 100
Fig. 38.1 Benchmarking contemporary practices against transatlantic
enslavement 685
xxi
CHAPTER 1
Introduction: Historicizing and Spatializing
Global Slavery
Damian A. Pargas
Introduction
Slavery has been a common—if often fluid and complex—condition in most
world societies throughout history. Only very few societies became so econom-
ically, politically, and culturally dependent upon slavery as to ultimately
develop into what Moses Finley famously dubbed “slave societies”—a cate-
gory he reserved for ancient Greece and Rome, and the plantation regions of
the Americas from the sixteenth through the nineteenth centuries. It has been
precisely the latter societies, however, that have long dominated static popular
images and the historical literature on slavery.1 That has begun to change.
The study of global slavery has grown strongly in the last few decades, as
scholars working in several disciplines have actively cultivated broader perspec-
tives on enslavement. Not only has interest in slavery among scholars working
on the Atlantic world reached a high point, but scholars have also intensified
their study of slavery in ancient, medieval, North and sub-Saharan African,
Near Eastern, and Asian and Pacific societies. Practices of modern slavery
and human trafficking from South Asia to Europe and the Americas are also
receiving more academic attention than ever before, and are now being inte-
grated into historical paradigms of slavery. More importantly, scholars are
increasingly looking across borders—temporal, spatial, and disciplinary—to
better understand slavery and slaving throughout world history. The recent
D. A. Pargas (B)
Leiden University, Leiden, The Netherlands
e-mail: d.a.pargas@hum.leidenuniv.nl
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_1
1
2D. A. PARGAS
surge in slavery studies has led to a greater appreciation of slavery as both a
global and a globalizing phenomenon in human history.2
First, scholars increasingly underscore slavery as a global practice that
has existed in innumerable world societies. Historians and anthropologists
working on communities as far apart in time and space as ancient Babylonia,
medieval Venice, Chos˘on Korea, the nineteenth-century American South, and
twentieth-century West Africa have devoted considerable ink to illuminating
local and regional case studies of slavery in extremely diverse settings. To be
sure, practices of slavery differed widely across time and space, and catego-
rization in some settings has proved difficult—scholars indeed continue to
disagree on what constituted “slavery” in some localized settings. Most studies
of slavery, however, converge on situations throughout world history in which
human beings were (or are) treated as property that could be bought, sold,
or transferred; held captive for indefinite periods of time; subordinated to
others in extremely dependent and exploitive power relationships; denied basic
choices (including potentially rights over their bodies, lives, and labor); and
compelled to labor, provide services, or serve various personal, cultural or
societal functions against their will.3
Second, scholars now more fully appreciate the globalizing effects that
slavery has had on world societies. From antiquity to the present day, slaver y
has by definition connected societies through forced migrations, warfare, trade
routes, and economic expansion. Slaving blazed both maritime and land routes
around the globe. Slave-trading routes crisscrossed Africa; helped integrate
the Mediter ranean world; connected China to the Indonesian archipelago;
and fused the Atlantic world. Global and transnational approaches to history
focus heavily upon the global movement of people, goods, and ideas, with a
particular emphasis on processes of integration and divergence in the human
experience. Slavery in various settings straddled all of these focal points, as
it integrated various societies through economic and power-based relation-
ships, and simultaneously divided societies by class, race, ethnicity, and cultural
group.
Both of these developments—the remarkable expansion of slavery scholar-
ship in various settings throughout world history and the greater appreciation
of slavery’s role in connecting societies—have led to new understandings,
definitions, and approaches to the study of slavery. The inevitable cr oss-
pollination of slavery studies from such diverse and global perspectives has
greatly influenced the ways in which historians and anthropologists talk and
think about slavery around the world. Long dominated by scholarship on the
early modern Atlantic and classical Graeco-Roman case studies—which created
the very framework for slavery studies, from its terminology to its theoretical
approaches—slavery scholarship has in recent years been enriched with new
insights into how slavery was understood in various settings, including how
it functioned, how it was meant to function, how and why people moved in
and out of conditions of slavery, how experiences of slaver y were character-
ized, and how practices of slavery affected regional and interregional power
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 3
relationships. Understandings of slavery have moved beyond static snapshots
and abstract definitions. There is now more focus on situating practices of
slavery along a broad continuum of coercion and extreme dependencies;
understanding the constantly developing and changing nature of slavery prac-
tices across time and space; and appreciating what conditions of slavery meant
for real people, both the enslaved and slaveholders.
Put simply, the recent surge in slavery studies has helped scholars to
historicize and spatialize slavery in world history. Historicizing slavery has
entailed moving beyond linear stories that trace slavery from the Graeco-
Roman context directly to Atlantic slavery and abolition, and embracing a
broader appreciation of how widespread and interlinked diverse practices of
slavery were and continue to be around the world, as well as how systems
of slavery have arisen and fallen in localized settings. Spatializing slavery has
entailed recentering the geography of slavery, appreciating for example just
how exceptional and atypical Atlantic slavery was and what made it so, and
illuminating regional contexts of slavery around the world.
The Palgrave Handbook of Global Slavery throughout History aims to intro-
duce students and scholars to the study of slavery across time and space. Its
intention is to historicize and spatialize slavery, providing both emerging and
established researchers with a comprehensive understanding of the current
state of the field, as well as serve as a unique reference work for devel-
oping further lines of inquiry. Providing chapter-length analyses of the most
prominent and widely researched systems of slavery around the world—
from antiquity to the contemporary era—it integrates various strands of
slavery studies and encourages readers to uncover connections, similarities, and
differences between various manifestations of slavery throughout history.
Global Perspectives of Slavery
How do scholars understand slavery, and how do they approach the study
of slavery in light of recent developments in the field? It can be difficult to
find cohesion in the various strands of global slavery research. Different case
studies have necessitated different approaches to establish what exactly consti-
tuted (or constitutes) slavery or slavery-like practices in various settings. They
have necessitated different approaches to the centrality of labor to conditions
of slavery. They have necessitated different approaches to slavery’s relation-
ship to “freedom” or other conditions of non-slavery. As scholarship moves to
global views of slaver y as a human condition, the danger arises that academic
understandings of slavery will ultimately encompass virtually all forms of
oppression and thereby seem so nebulous as to become meaningless. Kostas
Vlassopoulos recently underscored this in an article on the consequences of
global approaches to slavery: “If slavery has an essence, but its historical mani-
festations differ substantially across time and space, how can we study slavery
as a global phenomenon?”4 What indeed can be said about slavery from
4D. A. PARGAS
new global perspectives, and what parameters can be said to constitute the
framework from which most scholars of slavery operate?
Upon closer inspection most approaches to the study of slaver y around the
world differ by degrees rather than fundamentally. There are exceptions, to be
sure, but most new global slavery research does reveal common understand-
ings and approaches that provide a basic framework from which to analyze
slavery across time and space. Three interrelated themes stand out in partic-
ular, all of which are characterized by calls to both broaden our understanding
of slavery in light of its diversity in world history and clarify its position in
relation to conditions of freedom and unfreedom.
First, as stated above, new global slaver y scholarship has gone to great
lengths to situate slavery at the most extreme end of a broad spectrum—or
continuum—of unfree and dependent conditions in various settings. This has led
to a further clarification of what distinguishes slavery around the world from
serfdom, debt bondage, various forms of indentured servitude, imprisonment,
peonage, forced labor, and related asymmetrical dependencies. Scholars of
slavery in various settings agree that the condition of slavery, in virtually every
world society in which slavery existed, transferred to the slaveholder unlimited
and potentially permanent power over the enslaved person, including powers
related to life, reproductive capabilities, entitlements, and all other attributes.
This differed from all other dependent conditions. It is important to note
that, from a global perspective, slavery has not always constituted a clearly
defined category or institution, the way it ultimately did in the slave societies
of the Atlantic or Graeco-Roman worlds, for example. Indeed, as Vlassopoulos
recently argued—partly in an attempt to move beyond more static models
such as Finley’s “slave s ocieties” versus “societies with slaves” conceptualiza-
tion—approaching slavery from a global perspective entails understanding it
as a collection of “practices and processes” in various contexts. This view
is reinforced by Joseph Miller’s call for understanding slavery and slaving as
“historical strategies,” or rather temporally and spatially changing processes,
instead of static institutions.5 The practices and processes that constituted
slavery were everywhere, however, to quote Vlassopoulos, rooted in “two
conceptual tools: the tool of property in human beings, and the tool of domi-
nation in which one human being can exercise theoretically unlimited power
over another.”6 Unlike all other forms of dependency, enslaved people were
denied by their enslavers most—indeed virtually all—rights and privileges asso-
ciated with personhood, which were instead conferred upon the slaveholder,
a situation which Orlando Patterson—one of the first to produce a global
comparative study of slavery—famously referred to as “social death.”7 The
utility of Patterson’s conceptualization has been highly contested by some
scholars of global slavery, partly because in practice enslaved people around
the world most certainly functioned as persons, demonstrated agency, and
were sometimes even relatively well integrated into local communities. The
practice of domination and the attempts at dehumanization never resulted
in the enslaved person internalizing their dehumanized status or condition,
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 5
and everywhere both the theory and practice of slavery were riddled with
holes and inconsistencies. Fr ederick Cooper, a historian of African slavery, for
example, has criticized the model for ignoring slave agency and focusing too
much on slaveholders’ ideas on how they thought slavery should work, rather
than how the system actually played out in specific settings.8 And indeed,
enslaved people could find themselves with certain rights and privileges asso-
ciated with personhood in certain contexts—when contractually or legally
promised manumission at a future date, for example (such as with coartación
in Spanish America, or gradual abolition laws in the northern United States).
In essence, however, “social death” does not refer to a loss of person-
hood or social interaction in any absolute sense, but rather a loss of the
rights and privileges associated with personhood during the condition of slavery,
and in this sense there are more similarities than differences across time and
space. Enslaved people were political and social outsiders—the most extremely
marginalized people in any given society, completely subordinated to the will
of their masters.9 The condition of slavery almost everywhere entailed no legit-
imate claim to the fruits of one’s own labor; one’s own of fspring, family, or
community; one’s own body or the reproductive capabilities of one’s own
body; one’s own life. All such power rested with the slaveholder (or higher
cultural or political authorities that governed the slaveholder), who could
either grant or withhold such “privileges.” Crucially, the enslaved person was
denied the rights and privileges associated with personhood for an indefinite—
potentially permanent and even intergenerational—period of time. No action
by the enslaved person guaranteed a release from the condition of slavery—no
repayment of debt, no expiration of term—except as agreed upon by the slave-
holder or, in some cases, the state or other institutions of authority (as with
legal abolition in the modern period, or cultural dictates regarding manumis-
sion in Islamic law, for example). Entry into the state of slavery was also almost
always coerced, usually through violence (especially capture in wars) or birth—
and in the latter case the condition was maintained through violence or the
threat of violence. Few people in world history volunteered themselves for
enslavement, although there are rare examples of people enslaving themselves
to a more powerful person—usually people in desperate and impoverished
circumstances who opted for bondage for purposes of physical sustenance or
protection. In short, most scholars approach slavery as a collection of practices
and processes that fell at the most extreme along the spectrum of unfree and
dependent conditions, one that distinguished itself in its reduction of human
beings to a state of property, subjected to the theoretically unlimited power of
other human beings.
Second, and very much related to the first theme, global slavery scholarship
has underscored the need to understand practices of slavery from perspectives
that move beyond paradigms of “labor” and that embrace broader views of the
various purposes and functions of slavery in diverse settings. Long identified as
the most extreme solution to labor shortages in societies where productive
resources were available and power relationships made coercion of laborers
6D. A. PARGAS
possible—a theory that goes back to H.J. Nieboer’s important thesis on this
subject in the early twentieth century, and the adaptation of that thesis by
Evsey Domar in 1970—slavery has often been equated with forced labor,
especially forced labor in pr ofit-seeking economic activities.10 Global labor
historians, who by definition deal with questions related to work and labor,
continue to approach slavery first and foremost as a form of highly controlled
or coerced labor, one that was at least comparable to (and on the same spec-
trum as) peonage, debt bondage, indentured servitude, and exploited wage
earners in modern industrial and post-industrial societies. Marcel van der
Linden has indeed called for more comparative studies of all forms of coerced
labor (including slavery) by “dissecting” them into three “moments”: entry
into coerced labor, extraction of labor, and exit from coerced labor condi-
tions.11 This approach comes out of a need to escape a longstanding binary
between slave and free labor. Labor historians correctly argue that non-slave
labor was not necessarily free labor, and that the work experiences of enslaved
laborers often resembled those of other marginalized laborers and oppressed
working classes. Global perspectives of slavery, however, remind us to take a
closer look at the nature and centrality of slave labor in societies in which it
existed. First, it is important to remember that slavery was not—or at least,
not only—a form of labor. Its rootedness in the conceptual tools of prop-
erty in human beings and total power over another, as stated above, set it
analytically apart from all other labor systems. To be sure, the extraction of
labor and the acquisition of resources usually lay at the root of enslavement
in most world societies, and work was a central aspect of virtually all enslaved
people’s lives. But the condition of slavery went beyond work and labor. It
applied to non-productive people, including the very young, very old, injured,
and handicapped. It could not be redeemed through any amount of work
or self-purchase, except as agreed upon by the slaveholder or higher authori-
ties. It accrued not only material wealth and resources to the slaveholder but
also (and sometimes only) immaterial benefits such as prestige, privileges, and
power.12 Even in societies in which slaver y unequivocally served to produce
commodities for capitalist markets and thereby enrich the master class, such
as in the Atlantic world, slaver y entailed more than simply a labor system.
And second, slavery studies remind us to broaden our definition of what
slave “work” entailed, as labor historians have indeed long argued. Global
perspectives of slavery underscore the fact that enslaved people performed
a wide variety of functions that went beyond productive economic activities
and included everything from wet-nursing and childbearing to soldiering to
performing rituals to civil service in the upper echelons of government. In
short, global perspectives of slavery necessitate an understanding of its specific
purposes in various settings and an acknowledgment of its similarities but also
fundamental differences with various coerced labor systems.
A third theme that has arisen in light of new global slavery scholarship has
been the call for a reassessment of the relationship between slavery and freedom,
considering not simply what we mean by such categories but also what they meant
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 7
to the enslaved. Two trends are notable in this regard. The first is a tendency
in the scholarship to underscore that non-slavery in various world societies did
(and does) not necessarily constitute freedom in the sense of a person exer-
cising the power to act and make decisions without constraints. Much like
global labor historians’ efforts to escape the traditional binary between slave
labor and free labor, scholars of slavery around the world have undertaken
similar efforts to escape binary thinking about slavery and freedom. Instead,
as stated above, they place slavery at the extreme end of a broad spectrum
of dependent conditions, and they underscore the fluidity between various
conditions of unfreedom. The conceptual language of freedom was indeed
largely absent in many contexts of slaver y, and even where it did exist, it did
not always constitute the opposite of slaver y.13 In many societies in which
slavery existed, various dependent and coerced conditions could morph into
slavery—debt bondsmen in the Indian Ocean world could eventually become
enslaved, for example, as could ransom captives taken by corsairs in wars in
the early modern Mediterranean. People could move in and out of condi-
tions along the spectrum of dependency. Manumitted African Americans in
the nineteenth-century American South—whose condition and legal status fell
far short of the legal condition of “freedom” enjoyed by most white south-
erners, and whose forced poverty and marginalization often resulted in new
dependent relationships—could be reenslaved as a punishment for crime or
vagrancy. Such cases demonstrate that slavery was not always entered into from
a state of “freedom” and that exiting slavery did not always result in a condi-
tion of “freedom,” unless that term refers exclusively to non-slavery. A second
trend in the scholarship has been an effort to qualify the above by considering
how enslaved people understood non-slavery, and how they strove to attain
it. In other words, enslaved people everywhere understood their condition
and understood the differences between their condition and other conditions
and statuses in their respective societies. For most, exiting the slave status was
an act of personal liberation, even if doing so did not result in considerable
improvements in their daily lives. Exiting slavery may not have necessarily
resulted in radical changes in people’s working conditions, for example, nor
afforded them many rights or privileges, nor even led to a detachment from
their former owners. Everywhere, however, the boundar y between slavery and
non-slavery was perceived as enormously important. Relative to slavery, most
conditions of non-slavery appeared “free” to most enslaved people, even when
they constituted conditions of what scholars would categorize as unfree or
dependent.
Scholars of slavery around the world continue to debate and disagree on
various aspects of slavery in different contexts, and consensus is unlikely given
the enormous variety of its historical manifestations across time and space.
Global slavery scholarship does in a very general sense converge with respect
to certain themes, however. It understands slavery as a temporally and spatially
changing collection of practices and processes, situated at the most extreme
8D. A. PARGAS
end of a broad spectrum of unfree and dependent conditions, whose root-
edness in theories of property in human beings and the exercise of unlimited
power over another person’s body and life set it analytically apart from all other
forms of unfreedom. It acknowledges that slavery cannot be strictly equated
with coerced labor, and seeks to understand the similarities and differences
between slavery and other forms of coerced labor in various contexts. It seeks
to reassess the relationship between slavery and notions of freedom, acknowl-
edging that people did not always enter into conditions of slavery from a state
of “freedom,” nor exit slavery into a state of freedom, and exploring human
experiences of living and moving within and across statuses and social hierar-
chies. All of these themes provide an analytically rich way forward in the years
to come.
This Handbook
The Palgrave Handbook is designed to encourage global perspectives and
simultaneously provide a coherent understanding of slavery as a practice in
a wide variety of settings throughout world history in a single volume. A
number of editorial decisions have been made in order to enhance coherence
and readability.
First, the volume is divided into 5 chronological “parts” that highlight the
development of slavery over time. Part I contains chapters on specific case
studies of slavery in Ancient Societies (to 500 C.E.), examining the earliest
written sources on systems of slavery in the Mediterranean and Near East. Part
II continues with case studies of slavery and slave-trading in various Medieval
Societies (500–1500 C.E.), from the Arabian Peninsula to the Mediterranean
and even South America. Part III deals with Early Modern Societies (1500–
1800 C.E.), a period of unprecedented global interactions and long-distance
slave-trading throughout the world. Part IV delves into practices of slavery in
the Modern Societies (1800–1900 C.E.), characterized as an age of revolu-
tions and emancipation but also significant expansion of slavery in some parts
of the world. Part V explores Contemporary Societies (1900–present), an era
defined by the expansion of human rights and ultimately the universal illegality
of slavery. Each part is prefaced by a very short introduction by the volume
editors.
This chronological division is intended to provide the volume with a
coherent structure, highlight developments over time, and encourage compar-
isons of slaver y practices across space within specific time periods. It should be
noted, however, that in practice world histor y is not easily divided into neat
periods with clear beginnings or ends. The years given in parentheses in each
section title are rough indications, not hard boundaries. For this reason, the
editors decided to title each section with both a name and a rough indication
of the years in a given period. Chronological periodizations also do not always
apply neatly to all world societies. If the Medieval Period (ca. 500–1500 C.E.)
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 9
is largely defined as the period following the breakdown of empires and disrup-
tion of long-distance trading in the Mediterranean and Near East, for example,
it is known as a period of expansion and consolidation of empires in parts
of the Americas. Moreover, the danger exists that the chosen periodization
for this volume may be interpreted as adopting Eurocentric understandings of
world history. The volume editors are aware of these shortcomings. In the end,
the decision was made to follow the scholarship in the field of global slavery
and of world history in general, in which the same 5-pronged periodization is
widely used as a frame of reference. Indeed, most scholars of slavery, including
scholars whose case studies fall well outside of European influence, identify
with these periods and situate their case studies in relation to other practices
of slaver y around the world in the same period.
Second, the volume is subdivided into 32 thematic chapters by both estab-
lished and emerging scholars that illuminate specific case studies of practices
of slavery in different parts of the world, providing readers with the broadest
geographic scope possible. Each chapter constitutes a brief introduction to
slavery in a particular region and context; annotation is necessarily light, and
each chapter ends with a Further Reading section for those who wish to learn
more about a specific case study. A condition as common in world history as
slavery does not allow for a complete or definitive geographic representation
in a single volume, of course, and many potential case studies were necessarily
left out. The volume editors made a selection based on a number of factors.
First, each chronological part contains chapters that zoom in on case studies
from around the world insofar as they are reflected in the scholarship. The latter
constitutes a major challenge for any handbook on global slavery—practices
of slavery in some regions or time periods are simply not yet well studied or
remain unstudied due to a lack of sources. The editors did go to great lengths
to offer as broad a selection as possible, from both the Global North and the
Global South. Second, the volume contains chapters on the most prominent
and studied cases of slavery but also a smattering of chapters on case studies
that may be less familiar to students and scholars who are new to the field, such
as for example slavery in the early modern German Reich, asymmetrical depen-
dencies in the Inca empire, and state-sponsored slavery systems adopted by
totalitarian states in the twentieth century. The intention is to expose readers
to the latest scholarship in the main areas and time periods on which global
slavery studies focus, but also identify some relatively new directions that are
currently being explored and integrated into the field. Third, the editors delib-
erately limited the number of chapters that deal with Atlantic slavery to four;
these explore the rise of slavery in the Americas; plantation slavery in the
British Caribbean; slavery in Latin America (especially Cuba and Brazil) during
the “second slavery”; and slavery in the antebellum US South. The volumi-
nous and exciting scholarship on Atlantic slavery easily surpasses that for all
other case studies, and this volume could have contained many more chapters
on various parts of the Americas, but the editors ultimately limited the space
reserved for the Atlantic in order to help readers place Atlantic slavery—which
10 D. A. PARGAS
was in many respects atypical and exceptional—within a wider global context
and to allow more space for other case studies.
In order to enhance coherence and comparability, each contributor was
asked to explore or shed light on three themes in their respective chapters,
insofar as they are r elevant and applicable to their respective case studies.
The three themes are inspired by historian Marcel van der Linden’s three-
pronged approach to “dissecting coerced labor”—which for this volume has
been modified to account for slavery not being strictly equitable to forced
labor, namely:
a. Entry into slavery (how people became enslaved, including from other
conditions of dependency and coercion);
b. Experiences of the enslaved during slavery (how people lived and worked
as “slaves,” and the nature of exploitation, coercion, and violence in their
lives); and
c. Exits from slaver y (methods by which people ceased to be “slaves,”
including cases in which their new status or condition was one of
dependency or coercion).14
The authors were free to interpret and incorporate these three themes in ways
that made sense for their respective cases. Some cases did not lend them-
selves to one of the themes, for example. The second theme—experiences
of the enslaved—also gave contributors considerable leeway to briefly discuss
the most important or pressing challenges or aspects of enslaved people’s
lives. Many chose to focus on issues related to work, while others included
other aspects of enslaved people’s social lives. Authors were also free to deter-
mine their own internal chapter structure, so that not all of the chapters are
necessarily structured according to the three themes in turn.
A third editorial decision that was made in order to enhance coherence
and provide more general reflections on slavery as a global phenomenon
was to include a thematic injection at the end of each of the five chrono-
logical parts. The injection is a short essay (roughly half the length of a
chapter) that discusses an overarching theme or cross-cutting question that
highlights the connections between slavery practices in different settings, or
how scholars approach the study of slavery in different settings. Catherine
Cameron’s injection essay to Part I, for example, examines how archaeologists
identify “invisible” or marginalized people in world history, and how archae-
ological methods are helping historians understand the lives of the enslaved.
The injection essay to Part II, by Ruth Karras, explores the theme of sexual
exploitation of enslaved people from a gender history perspective. Part III
concludes with an injection essay by Klaus Weber about the interconnected
global commodity chains involved in the development and sustenance of early
modern slave systems. William Mulligan’s injection essay for Part IV exam-
ines the development of global abolitionist networks and movements in the
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 11
nineteenth century. Finally, Part V concludes with an injection essay by Joel
Quirk about modern anti-slavery and human rights movements, and the ways
in which these movements affect how scholars think about slavery as a histor-
ical phenomenon. The injection essays encourage readers to zoom out and
consider a theme that helps bring place the specific case studies in each part in
a wider context.
The structure and approach of this handbook make it a unique addition to
the literature on global slavery in the English language.15 Indeed, this hand-
book complements other recent handbooks and anthologies of global slavery
and provides certain features that others do not. Perhaps most well-known
is the excellent four-volume Cambridge World History of Slavery (4 vols.,
Cambridge University Press). Like the Palgrave Handbook, the Cambridge
history is divided chronologically and offers short essay overviews of slavery
in various contexts throughout world history. Unlike the Palgrave handbook,
however, it is divided into four separate (and lengthy) books, all of which can
be purchased or read separately. This allows readers to delve into a wealth
of case studies on specific time periods and regions, but does not encourage
or facilitate a clear understanding of the development of slavery over time,
from antiquity to the present. The Routledge History of Slavery (2012), edited
by Gad Heuman and Trevor Burnard, offers a more concise and accessible
overview of global slavery. Unlike the Palgrave volume, however, it is rela-
tively short and heavily dominated by essays on Atlantic slavery. The recently
published volume Writing the History of Slavery (2022), edited by David
Stefan Doddington and Enrico Dal Lago, provides readers with an excellent
historiographical and methodological overview of global slavery scholarship,
but is intended to introduce readers to historical approaches to slavery rather
than provide an overview of case s tudies on slavery in specific settings.
The Palgrave Handbook complements these other handbooks by providing a
concise volume that introduces readers to practices of slavery in a wide variety
of settings, as well as a handful of thematic and theoretical essays.16
To understand slavery—why and how it developed, and how it functioned
in various societies—is to understand an important and widespread practice in
world civilizations. The Palgrave Handbook of Global Slavery throughout World
History encourages students and scholars to zoom out and understand the
similarities, differences, and connections between practices of slavery around
the world. As such it hopes to inspire a new generation of slavery studies and
help set the research agenda for years to come.
Notes
1. Moses I. Finley, Ancient Slaver y and Modern Ideology (Princeton: Marcus
Weiner, 1988).
2. The recent surge in global perspectives of slaver y can be seen not only in
a wealth of new publications and handbooks, but also in other academic
projects, such as large international conferences; the launch of the Journal of
12 D. A. PARGAS
Global Slaver y to complement the longstanding journal Slaver y & Abolition;
the opening of the Brown University Center for Slavery and Social Justice in
the US; and the opening of the Bonn Center for Dependency and Slaver y
Studies at the University of Bonn, Germany; among others.
3. Damian Alan Pargas, “Slavery as a Global and Globalizing Phenomenon:
An Editorial Note,” Journal of Global Slavery 1, no. 1 (2016): 1–4; David
Stefan Doddington and Enrico Dal Lago, eds., Writing the History of Slavery
(London: Bloomsbur y, 2022), chs. 1–2.
4. Kostas Vlassopoulos, “Does Slavery Have a History? The Consequences of a
Global Approach,” Journal of Global Slavery 1, no. 1 (2016): 6.
5. Joseph C. Miller, The Problem of Slaver y as History: A Global Approach (New
Haven: Yale University Press, 2009).
6. Vlassopoulos, “Does Slavery Have a History?” 12–13.
7. Orlando Patterson, Slavery and Social Death: A Comparative Study
(Cambridge, MA: Harvard University Press, 1982).
8. Frederick Cooper, Colonialism in Question: Theory, Knowledge, History
(Berkeley: University of California Press, 2005), 17.
9. For classic works on slaves as marginalized outsiders that have greatly influ-
enced current global slavery scholarship, see for example: Igor Kopytoff and
Suzanne Miers, “African ‘Slavery’ as an Institution of Marginality,” in idem,
eds., Slavery in Africa: Historical and Anthropological Perspectives (Madison:
University of Wisconsin Press, 1975); Claude Meillassoux, ed., L’esclavage en
Afrique précoloniale (Paris: François Maspero, 1975).
10. H.J. Nieboer, Slavery as an Industrial System: Ethnographical Researches (The
Hague: Martinus Nijhoff, 1900); Evsey D. Domar, “The Causes of Slavery
or Serfdom: A Hypothesis,” Journal of Economic History 30 (Mar. 1970):
18–32. Orlando Patterson delivered one of the most scathing critiques of
the Nieboer-Domar hypothesis, arguing that rather than viewing slavery as
an economic system, it should be seen in terms of power relationships. See
Orlando Patterson, “The Structural Origins of Slavery: A Critique of the
Nieboer-Domar Hypothesis from a Comparative Perspective,” in Annals of the
New York Academy of Sciences 292 (1977): 12–34; Patterson, Slavery and Social
Death,1.
11. Marcel van der Linden, “Dissecting Coerced Labor,” in Marcel van der Linden
and Magaly Rodríguez García, eds., On Coerced Labor: Work and Compulsion
after Chattel Slavery (Leiden: Brill, 2016), 291–322.
12. Vlassopoulos reminds scholars that in many cases “prestige creation” was the
main—sometimes the only—slaving strategy. See Vlassopoulos, “Does Slavery
Have a History?” 14–15.
13. See for example Vlassopoulos, “Does Slavery Have a History?” 10; as well as
Anthony Reid’s discussion of this concept for Asian slavery systems: Anthony
Reid, “Merdeka: The Concept of Freedom in Indonesia,” in David Kelly and
Anthony Reid, eds., Asian Freedoms: The Idea of Freedom in East and Southeast
Asia (Cambridge: Cambridge University Press, 1998), 141–60.
14. van der Linden, “Dissecting Coerced Labor.”
15. The most recent and ambitious handbook to date is the Fr ench-language Les
mondes des esclavages: Une histroire comparée (Paris: Le Seuil, 2021), edited
by Paul Ismard, Benedetta Rossi and Cecile Vidal. This excellent volume is
twice the size of the Palgrave volume and contains 50 chapters on various case
1 INTRODUCTION: HISTORICIZING AND SPATIALIZING … 13
studies. For a similar (more succinct) handbook of global slavery in German,
see Michael Zeuske, Handbuch Geschichte der Sklaverei: eine Globalgeschichte
von den Anfängen bis zur Gegenwart (Berlin and Boston: De Gruyter, 2013).
16. Keith Bradley and Paul Cartledge, eds., The Cambridge World History of Slavery
1: The Ancient Mediterranean World (Cambridge: Cambridge University Press,
2011); David Eltis and Stanley L. Engerman, eds., The Cambridge World
History of Slavery 3: AD 1420–AD 1804 (Cambridge: Cambridge University
Press, 2011); David Eltis, et al., eds., The Cambridge World History of Slavery
4: AD 1804-AD 2016 (Cambridge: Cambridge University Press, 2017); Craig
Perry, et al., eds., The Cambridge World History of Slavery 2: AD 500-AD 1420
(Cambridge: Cambridge University Press, 2021); Gad Heuman and Trevor
Burnard, eds., The Routledge History of Slaver y (New York: Routledge, 2012);
David Doddington and Enrico Dal Lago, eds., Writing the History of Slavery
(London: Bloomsbury, 2022). For a global view of modern slavery, see Kevin
Bales, Understanding Global Slavery (Berkeley: University of California Press,
2005).
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
PART I
Ancient Societies (to 500 C.E.)
Preface
The first part of this handbook introduces the reader to four different empires
of ancient times and their practices of slavery: Mesopotamia, Egypt, Greece,
and Rome. The chapters cover huge time periods of up to three thousand years
of history and deal with various phases, distinct centers, and diverging aspects
of ancient empire building. It is a period in history for which written sources
are scarce and remains of legislative codification little. Historians of slavery
therefore rely to a large extent on fragmented documentar y and literary texts
as well as on archaeological evidence. And yet, the chapters of this section
unveil some patterns of ancient slavery that show remarkable similarities and
give reason to some general reflections also relevant for later periods in history.
First, all contributors agree that there was no labor division based on status.
Even in the so-called slave societies of Athens and Rome, slaves were deployed
side by side with other workers. Second, in all four societies, slaves were
not only used as unskilled workers in food production and service industries.
Mesopotamian and Egyptian, Greek and Roman slaveholders also relied on
slaves as qualified specialists and persons of trust, serving as courier, educa-
tors or doctors. Third, slaveholding was not only (and often even to a lesser
extent) a matter of economic calculations but served other purposes such as
credit needs and revenue extraction, gratification, and prestige creation as well
as authority and the exercise of power.
Generally speaking, human trafficking and slave trade played a bigger
role in societies that were connected to and embedded in broader supra-
regional networks. Likewise, expansionist phases of each of the ancient empires
usually went hand in hand with large-scale imports of captives as slaves. While
enslavement was basically open to all peoples and races, some authorities and
16 PART I: ANCIENT SOCIETIES (TO 500 C.E.)
legislations introduced mechanisms and rules intending to protect their own
people from slavery within their borders. In all four societies, slaves were
most extensively used in labor- and capital-intensive sectors as well as for elite
surplus production. Also, there was a strong tendency to integrate enslaved
persons into existing family and social structures. The fact that multiple types
of manumission existed side by side illustrates the complexity of the relation-
ship between slaveholder and slave where the exercise of power and protection
and the need to trust interacted with the obligation for subor dination and
obedience and various degrees of agency.
As a matter of fact, the use of archeological knowledge for the study of
social and power relations is still in its infancy. Yet, new biochemical methods
might open new perspectives on old questions, not only for the study of
ancient societies.
CHAPTER 2
Mesopotamian Slavery
Seth Richardson
“What are a slave-woman’s dreams? What are a servant’s prayers?”—Sumerian
proverb
Introduction
This chapter introduces us to Mesopotamian slavery by focusing on the Old
Babylonian period (hereafter, “OB”). This periodization is bounded both
temporally (ca. 2000–1600 BCE) and geographically: Babylonia corresponds
roughly to southern Iraq from Baghdad to the Persian Gulf, as opposed to the
Assyrian north near present-day Mosul/Kirkuk. The OB is a historical time
and place about twice as long and large as Classical Greece, with thousands of
documents available to inform us about our subject. It is necessary to focus
on one period, because to describe a single “Mesopotamian slavery” for the
whole of cuneiform culture (from ca. 3300 to 300 BCE) would require us to
concatenate evidence fr om at least ten major historical periods, and only result
in a Frankenstein image of our subject.1 I will therefore explain the situation
in one time and place which typifies the subject, but also try to give a sense of
where, when, and how specific featur es of slavery in other periods conformed
to or differed from it.
S. Richardson (B)
Oriental Institute, University of Chicago, Chicago, IL, USA
e-mail: seth1@uchicago.edu
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_2
17
18 S. RICHARDSON
OB slavery was primarily a mechanism to provide credit in an emerging
mercantile economy—which otherwise had few if any stable financial institu-
tions—rather than to satisfy labor needs in any other productive sector.2 It
was primarily in this way that Mesopotamian slavery had globalizing effects:
merchants, slaves, and transacted values moved through local, regional, and
supra-regional trading networks: traf ficked slaves were moved to what is now
southern Iraq from places of origin as much as 1000 km away which are now
in Turkey, Syria, and Iran. This has implications for language and cultural
contact, but the most durable consequence was the transmission of the socio-
economic practice of using people as a form of capital. The trade in slaves
began to establish homologies of prices, systems of credit, legal standards, and
even concepts of personhood throughout the Near East.
The OB period was the age of Hammurabi, a time characterized politically
by competition and conflict between many small territorial states; socially, by
an emerging class of private householders and merchants; and economically
by credit markets and great trading networks connecting Mesopotamia to the
Mediterranean, Anatolia, Elam, and the Persian Gulf, with contacts extending
to Oman and the Indus Valley. It is a period with a well-rounded range of
sources for slavery, with evidence from law codes and lawsuits, letters and
contracts, proverbs and bequests. These diverse text types all have their own
conventions and limitations, but they allow us to build a portrait of slavery as a
social, legal, and economic institution and even of the experiences of individual
slaves and masters.3
These records reveal the basic socio-economic contours of OB slavery.4
Some slaves were born into status (i.e., “house-born”); others were brought
from foreign places.5 Some slaves eventually “integrated” into their masters’
families by adoption, inheritance, or marriage; some formed families of their
own. Slaves could own moveable property; some were literate; they had the
capacity to give legal testimony.6 Households which owned slaves generally
had only one or two. Slaves and masters knew each other face-to-face as indi-
viduals, and slaves were virtually always referred to by name.7 They performed
housework and field labor, but no type of work which was not also done by
non-slaves. They were trusted with a wide range of tasks which required their
free movement (even unguarded travel to other cities), independent judg-
ment, and personal knowledge of their masters’ affairs. They acted as legal
and business proxies for their masters, and their reputations reflected on their
households. Yet these positive indications of personhood are tempered by
other features. Slaves were bought and sold as chattel much like oxen and
commodities. They were commonly used as pawns to establish commercial
credit, and a variety of mechanisms existed to control their movement and
behavior, including community surveillance and for ce. Slaves were required to
wear distinctive hairstyles and perhaps garments, partially isolating them from
their previous social identities, and they were “genealogically isolated” by the
limited possibilities to pass on their names as patronyms, though this was not
forbidden.
2 MESOPOTAMIAN SLAVERY 19
This mixture of adverse and positive attributes, broadly accurate for slavery
in all periods of Mesopotamian history, requires us to ask where to inscribe
the boundaries of “social death.”8 The social death which characterizes slavery
cross-culturally, according to Orlando Patterson, entails the loss of name and
identity, punishment by death for flight,9 and the rejection of the slave as a
member of society. But Babylonian slaver y, in its particular admixture of social
recognitions and subordinations, does not meet all these criteria.
It must be stressed that we have no clear semantic or etymological under-
standing of the various Sumerian and Akkadian words for “slave” (either
masculine or feminine) and “slavery.” This is a complicated matter which
cannot be addressed in full here but suffice it to say that although some
proposed meanings are hardly impossible, they are neither linguistically clear
nor as logical as they first appear.10 For instance, Sumerian arad2 and Akka-
dian ardu for “slave” might connote foreignness on etymological and even
graphemic grounds (as “mountain man” or “(one) brought down (the
river?).” But the etymons are neither cer tain nor even semantically exclusive i n
either case, and even the logical premises are complicated by the fact that both
words were routinely applied to “slaves” who were clearly not foreign at all
but (former) citizens of the same city. We can also exclude the idea that racial
or ethnic identity was imbued in the Mesopotamian terms, as “Slav” is within
“slave.” The point is that any translation of “slave” for Mesopotamian terms
(including as I use it here) is purely conventional, and possibly anachronistic
and plain old wrong.
All of this introduces us to the central problem and paradox of
Mesopotamian slavery: the disposition of persons whom we identify in socio-
economic terms as “slaves” because they were legally and commercially chattel,
but who nevertheless had both agency and specific social and legal identities.
Thus, even the evidence from this earliest of literate world cultures requires
us to question how foundational slavery really was and the degree to which
it compromised or erased personhood. Should we understand this Babylonian
slavery to be para-social, in which a degree of personhood was conferred on
slaves, at least to the minimum extent necessary to deputize them as effec-
tive proxies for their masters? Or should we think of that very construction as
slavery’s most fundamental hypocrisy, to create a class of persons who could
be “trusted” to perform as many roles as free persons without recognizing
and permitting basic control over their bodies, places of residence, and life
choices? No less urgent are the questions for the slaveholding society itself:
how did the institution, practices, and individual slaves affect what it meant to
be a Babylonian who was not enslaved?
20 S. RICHARDSON
Entering Slavery
When and how did a person become a slave in OB Mesopotamia? The discus-
sion below moves from dispreferring certain options for mode of enslave-
ment—namely, as consequences of legal punishment, raiding, or commercial
markets—in favor of debt as the broad and underlying condition for most
instances.
Probably no one, to begin with, was enslaved as a punishment by law
in practice. A half dozen OB law collections (such as Hammurabi’s) regu-
lated various matters regarding enslaved persons, but only rarely ventured to
form their status.11 The OB law collections we have, preserving around 450
paradigmatic decisions, were primarily concerned with masters and commer-
cial practice rather than the regulation of slavery as such. These laws addressed
five basic areas of practice:
•The award of damages to a master for the death or injury of a slave.
•Holding slaves as pledges or pawns in distraint for credit (see below).
•Limiting (but not excluding) slaves’ capacity to make commercial trans-
actions.
•Punishment of flight, but largely focusing on the free persons who aided
fleeing slaves, not the slaves themselves; the crime was understood to be
a theft of property by a free person, not a slave’s decision to flee.
•The status and inheritance rights of children born of slaves and free
persons. The laws mostly guaranteed the eventual manumission of such
children, but sometimes only following the death of the enslaved parent.
The law codes differed in supporting, limiting, or nullifying the rights of
these children to inheritance of a free father’s property.
But all of these laws focus on slaves already in status. Only three of ca. 450
extant OB laws can even potentially be read as creating slave status. The first
(SLEx 4’) says that an adoptive son who repudiates his family may be “sold
for his full value,” though it does not say “into slavery.” The second and
third are both found in Hammurabi’s laws. ¶141 states that a wife found at
fault in a divorce case must reside in her husband’s house “as (if she were?)
a slave,” though it is unclear why it is qualified “as (if)” (k¯ıma); ¶146 says
that a slave woman who bears her master’s child and then “aspires to equal
status” with his primary wife cannot be sold by her (presumably out of jeal-
ousy), but should “counted (imman¯uši) among the slave-women”—which is
of course confusing, because she was already a slave to begin with. It is possible
that these decisions created slave status as legal punishments, but none of the
three is crystal-clear on this point. Neither is any of them documented in any
real-life context, which raises considerable doubt that these rules were ever
applied.12 No other crimes or civil infractions were said by the statutory laws
to be punishable by slavery, and no court cases reflect such a punishment.
The laws instead were overwhelmingly concerned with rules for people who
2 MESOPOTAMIAN SLAVERY 21
were already masters and slaves to guarantee property rights, not to create new
slaves.
Significantly, when the laws mention in passing any specific mechanism of
status formation, it is sale.13 This accords with the fact that the most substan-
tial body of evidence we have for enslavement comes from slave sale contracts.
These contracts are the documents in which people are first (to our knowl-
edge) called “slave” in their life histories. The slaves sold in these texts are
people who are not previously or subsequently known to us in any non-slave
status; no sale contract makes mention of any different and preceding status
or cause of enslavement. It is therefore almost impossible to reconstruct the
events by which any individual was sold let alone what happened to them
next. Whether or not sale per se was the mechanism which cr eated slave status
cannot therefore be determined—especially given that many sales of slaves
were demonstrably re-sales. (Other significant modes of transfer included gifts,
dowries, and inheritances, but this was not entry into status per se.) It is har d
to avoid the conclusion that slavery was defined as a contractual and commer-
cial matter between private parties and not institutionally generated. It is worth
noting, given the discussion of self-sale and enslavement for debt below, that
sold slaves, although always named by the contracts, were never parties to the
sale, although sometimes their family members were.
There is little evidence that OB slavery was sourced from the widespread
capture of people through war raiding. That is, although there is occasional
evidence for raiding for people subsequently forced to work for their captors—
a form of unfree labor likely a distinction without a difference for those thus
exploited—none of it points to the sale of those prisoners as chattel slaves.14
Indeed, prisoners of war were typically redeemable through the payment of a
ransom (ipt
.¯eru) by their home country’s family, temple, or palace. That OB
slavery was not substantially stocked with prisoners of war may surprise readers
who are familiar with the ample record of the mass deportation of conquered
populations by many Mesopotamian states since the early third millennium.15
But even during the time of the Neo-Assyrian empire, when literally millions
of people deported during three centuries of war violence, the evidence for
the transformation of war captives, plentiful though they were, into slaves is
“rather limited,” “difficult if not impossible to trace,” and “questionable.”16
Even when OB kingdoms captured prisoners of war, the actual work to
which they were put is not well documented. The only evidence for this in
the OB comes from an archive of texts from Uruk covering about eighteen
months over the years 1742–1740, documenting prisoners of war mostly from
local cities (not faraway places). The archive largely documents the individual
households to which these men were assigned and their food rations, but says
almost nothing about their labor. We need not doubt they did work—they
were called “workers” rather than “slaves,” and none were sold—but there is
almost no documentary attention to the kinds or amounts of work they did.
In fact, the capture and deportation of prisoners are topoi entirely absent
from the royal literature we would expect to mention them, which tracks
22 S. RICHARDSON
with the near-absence of slaves in mass-use by institutions. Palace and temple
administrative texts, documenting properties, production, and people, are in
good amounts throughout the four centuries of the OB, but we do not
find much evidence that they depended on enslaved workfor ces. We do not,
however, have from Babylonia the kinds of coherent royal archives such as
the one recovered from the contemporary Syrian city of Mari (ca. 400 km
northwest of Babylon), where slaves disbursed to private households could
be summoned for seasonal mass labor in the king’s orchards and smaller
numbers of female captives manned textile workshops of the palace.17 It is
possible that such arrangements existed in Babylonia proper, but none of
the surviving letters or work rosters hint at it. On the rare occasions when
groups of slaves do appear as laborers, they were working alongside non-slaves
(soldiers, workers, tribesmen, neighbors, etc.) performing the same tasks.18 In
sum, there is little evidence that OB states organized the forcible acquisition
of people in large numbers for labor exploitation or sale.
Instead, the OB slavery represented in contracts and letters was overwhelm-
ingly characterized by relations between small numbers of specific individuals
known to each other by name. Sale contracts usually conveyed just one person,
occasionally two (often a parent and child sold together); rarely, three. In
letters from prospective buyers asking merchants to procure slaves for them,
again, the typical request was for a single slave, although occasionally up to
six are mentioned, but we never hear about the acquisition of slaves by the
dozens or hundreds. The letters make clear that it was a normal expectation
that slaves could be acquired on the market, sometimes in foreign places by
long-distance merchants, but also that the supply was limited, or that only
“unhealthy” or “skinny” slaves might be available. Sometimes a preference for
certain foreign ethnicities was expressed, perhaps for linguistic reasons (i.e.,
to obtain a slave who at least spoke some Semitic language, as opposed to
Elamite or Hurrian19) or because “house-born” (i.e., native) slaves might
come encumbered with rights to eventual manumission or inheritance. But
nothing suggests that enslavement was thought to be particularly appropriate
to specific ethnicities; many ethnonyms attached to slaves also described free
persons.20 Neither do literary sources present a consistent point of view about
slaves as ideally foreign or not. One may consider two proverbs offering
conflicting advice, even though they derive from the same composition. The
first warns against the acquisition of foreign slaves:
When you bring a slave girl from the hills, she brings both good and evil with her.
The good is in the hands; the evil is in the heart. (Instructions of Šuruppak, ll.
193–197)
while another argues for the superiority of foreign slaves, likening house-born
slaves to “herbs that make the stomach sick”:
2 MESOPOTAMIAN SLAVERY 23
You should rather bring down a foreign slave from the mountains, or you should
bring somebody from a place where he is an alien. My son, then he will pour water
for you where the sun rises, and he will walk before you. (ll. 158–159)
No clear view can be gained on this point, unless it is deep ambivalence.
In any event, physical force does not seem to have played an important
role in the acquisition or transportation of slaves. Contracts and letters speak
of slaves bought in the way that other commodities wer e bought, with no
hint of violence: no chains, shackles, brands, yokes, ropes, guards, or other
apparatus are mentioned. Restraints were sometimes used o n slaves who had
fled, and a few instances of corporal punishment are hinted at.
Letters (which can provide more dependable contexts because they were
composed to deal with specific rather than paradigmatic problems) strongly
indicate that enslavement was a common result from the default of private
household debts (Sum. e2 ur5-ra, a “household under obligation”), where the
sale of selves or family members was done in exigence, but voluntarily.21 The
Babylonian concept of “debt slavery” had a fundamentally conditional char-
acter: persons held in this way were understood to be pledges (Akk. nipûtu,
a “distraint”) for debts to be repaid—parked in the household of a cred-
itor until (someday) redeemed—rather than property whose ownership was
unconditionally and permanently transferred. All such “enslavements,” though
indefinite, were not considered a permanent change of status. That the verb
of transaction in contracts was unambiguously “to sell” (Sum. šam2,the same
verb used to sell a sheep or a chair) may be misleading to the extent that slaves
could later be “redeemed” (Akk. pat
.¯aru, uššuru, t
.ar¯adu; see “exits,” below).
To put this in culturally salient terms, “slavery” was often a matter of a low-
ranking member of a poor household being deputed to go and live and work
(but also to be provided for, i.e., fed and clothed) in another household until
such time as a family debt was repaid.
This form of enslavement could hardly be called a desirable option for
the enslaved person, and we ought not avoid casting a critical eye on the
discursive paternalism which characterized and apologized for it; for instance,
the common Babylonian encomium to “take care” of slaves also carried with
it the sense of “to keep an eye on” them. But like many other invidious
choices, enslavement as a pledge was a rational economic alternative for seller,
buyer, and sometimes even slave. It may have been preferable, in hardship
conditions, to have your son or sister eat poorly as a slave in a creditor’s
household for a year or two than to have them starve in your own. These
choices were made within a cultural logic in which all household members—
selves, wives, children, servants, some animals, and even real property—were
not only social beings but also valuable and saleable according to ideas of
proportional value, expressed in legal and even mathematical terms. To this
extent, the social violence of slavery was not exclusively focused on slaves; the
threat of enslavement hung over most non-slaves, who could in theory be
sold any day, usually by the decision of someone else. The unpalatable choice
24 S. RICHARDSON
of enslavement was created by a world in which commercial and financial
economies were still emergent: one where long-distance, local, and financial
markets existed, but sources of credit and financial instruments were few. As
an economic phenomenon, Babylonian enslavement was what happened when
the inequalities of household hierarchies were commodified for a marketplace
which had few other financial structures to build on.
From the perspective of the creditor/enslaver, taking a person as a pledge
was a mechanism meant to guarantee the repayment of a loan, most often
extended for productive purposes, but it also entailed costs of feeding and
housing someone. That the ideal was for distraint-for-debt to be temporary is
exemplified by one of Hammurabi’s laws, which states:
If a debt is outstanding against a man and he sells or gives into debt service his
wife, his son, or his daughter, they shall perform service in the house of their
buyer or of the one who holds them in debt service for three years; their release
shall be secured in the fourth year. (¶117) 22
The way things were supposed to work was: for the term of a loan, a distrained
slave was sustained in exchange for the same labor they might normally do in
their original household; repayment was the hoped-for outcome on the part
of all parties involved. That this did not always work out this way in practice
is addressed below.
Debt-and-default was also the primary motor for the enslavement of foreign
people. That some slaves were indeed foreign has disguised the fact that they
were not acquired by raiding but were already enslaved when purchased in
distant non-Babylonian cities. An analysis of sale contracts shows that, just like
“native”-born slaves, they too had most likely become enslaved because of
economic hardships in their households.23 The majority were women, perhaps
unmarried or widowed, some with children in tow, or even infants at the
breast; this demographic profile suggests that women were often the “least
valuable” members of poor families precipitating out of their households in
order to alleviate the debts of their fathers, husbands, and brothers. The sale
of individuals in contracts may also obscure that they document only single
members of what were entire households collapsing under debt, with family
members sold one by one and pushed out of local communities, on to market
towns, and thence to Babylonia.
This image is preserved for us in one literary composition referring to “the
banished enemy, the slave from the mountains, or the laborer with a poor wife
and small children comes, bound with his rope of one cubit….” The likelihood
that debt was the major stimulus for enslavement is furthered by the last part
of the second proverb quoted earlier, recommending the acquisition of foreign
slaves; it goes on to count among the benefits of having a foreign slave that:
he does not belong to any family, so he does not want to go to his family; he does
not belong to any city, so he does not want to go to his city.
2 MESOPOTAMIAN SLAVERY 25
In practical terms, debt slavery for re-sold foreign slaves was substantially more
permanent than it ever was for Babylonian ones, even those born into slavery
domestically (wilid b¯ıtim, “house-born”).24 Brought hundreds of miles from
the home which had sold them in the first place, who would (ever) come
to redeem these foreign people? Perhaps slavery was intended only to have
been a temporary status—the short-term solution to a very specific problem—
but the accretion of foreign slaves into the system may have created (mutatis
mutandis) a more durable class of unfree persons over time because they were
sold at a distance. The redemption of foreign slaves occupied a prominent
position in Hammurabi’s laws: the antepenultimate and penultimate legal deci-
sions were concerned with facilitating the return of persons purchased as slaves
in foreign lands, pointing to frequent abuses of the redemption system (itself
guaranteed by law). This may hint that permanent enslavement was emerging
as a problem wanting a solution.
We may make an analogy to modern-day abuses where foreign guest-
workers, made liable for outlandish transit and housing fees by coercive
employment contracts, are effectively transformed into unfree workers, e.g.,
Filipino laborers in Kuwait or Latin American migrant workers in the United
States. In systems of social inequality, the economic nature of exploitation
is most directly elucidated by identifying the means of specification used to
create, regulate, and maintain that inequality. The value of Babylonian slaves
was commodified by law and contract; their status as pledges specified by
a debt theoretically equivalent to the value of a human, redeemable but
sometimes unpayable; and the gradual transformation of foreign slaves into
permanent chattels suggests that a system originally based more on human
trafficking for credit may have moved toward the reation of permanent chat-
tels. The transfer of the debt obligation or the sale of the slave to a different
geographic region could then prevent redemption, even if the original term
of unfree service was finite; effectively, the system may have begun to cr eate a
category of permanently enslaved people. These developments presented new
practical and even moral problems in a system meant to afford only temporary
expediences.
The Labor and Economic Function of Slaves
Here, I distinguish between two topics: the labor of slaves and the economic
function of slavery. The reader may reasonably expect that a report on the
work of Babylonian slaves is the natural answer to the implicit question of
why people bought and kept them. We assume that labor was (and is) the
raison d’être for slavery—that slaves were kept to grind grain, mine silver, and
tend the house. Were they not? Why else would people bother to buy, feed,
and clothe slaves if not to profit from their labor?
But the subject of slave work is oddly subdued in the cuneiform record.
Sale contracts never specified any particular skills of slaves being sold or work
to be done; hire contracts (since slaves could also be hired out on wages basis)
26 S. RICHARDSON
only occasionally did so, usually grain-grinding and weaving for women and
fieldwork for men. I am aware of only a single OB letter specifying skill as a
criterion for purchase: “As for the slave you wr ote me about, buy her if she’s
house-born and a weaver” (AbB VI 4). Stray passages in literary sources reflect
ambivalent stereotypes of slave labor, both negative (that slaves were lazy,
unreliable, fearful, grumbling, misbehaving) and positive (capable, honest,
tireless). Compare, for instance, these two proverbs:
A free weaver equals two slave girls. A free worker equals three slaves.
At harvest time, at the most priceless time, gather like a slave girl, eat like a
queen!25
The first proverb implies that slave labor was inherently less productive; the
second implies diligence. But parsing these brief and disparate characteriza-
tions is less helpful than observing how few literary references there are to
slaves’ work at all.
OB letters paint the same picture. Of 804 references to slaves and servants
in OB letters, only 6 percent were about work expected or performed. (By
comparison, 37 percent of all statements had to do with where, when, and
how slaves were to move freely to different locations to carry and acquire
information, and 19 percent related to issues of pledge and distraint.26)Most
references to labor were unspecific—just that slaves should “do work,” “be
on duty,” or “not be idle.” A few letters speak of herding, cooking, fetching
wood, household chores, or field- or canal-work (including as proxies for
labor expected of their masters); six letters mention specialized skills: weaving,
gardening, grooming. But no form of work was particular to slaves: the
work they did included the same tasks performed by free persons: harvesting,
grinding grain, tending animals, dredging canals, etc. No activity was socially
marked in the OB or any other period as distinctively “slave labor.”
What is more revealing is how typical it was for slaves to work alone, very
occasionally in twos and threes, and virtually never in labor gangs. The few
households that owned slaves rarely seem to have had more than one or two
at a time. Slaves almost always worked without supervision and alongside non-
slave persons. Force and coercion are rarely indicated (and even then, not
uniquely for slaves). And though the dress and distinctive hairstyles of slaves
visibly identified their status publicly, chains and other restraints were used
only as forms of punishment in unusual circumstances; branding, used for
slaves in the mid-third and mid-first millennium,27 is not attested for the OB.
If we put these observations in dialogue with the foregoing discussion about
entry into slavery, the somewhat counterintuitive takeaway is that OB slavery
did not exist in order to aggregate labor power, solve labor shortages, or
extract work in exchange for invested value. Slavery was not really about work,
per se. To be clear: all slaves worked, but only as all other people did. Like
most pre-modern agricultural societies, work was not an optional activity for
2 MESOPOTAMIAN SLAVERY 27
anyone—even the highest priests and officials were obliged to perform corvée
labor, for instance—and there was little opportunity to intensify production
or maximize profits. To think that labor in this socio-economic setting could
be commodified and apportioned to an encumbered class of persons so that
another class could thereby enjoy leisure time is not only anachronistic, but
just simply not in evidence. Mor eover, given the technological and informa-
tional boundaries of the period, the raw compulsion of labor would have been
wildly inefficient. Given the dispersed distribution of slaves throughout the
population, forced work would have required supervision at a ratio too close
to 1:1 to effect compliance. Neither did slavery solve any obvious deficits in
available labor resources, given that other forms of semi-free labor existed
alongside it: servants (s
.uh
˘ ar¯u), dependents (r¯edûtu, i.e., in the status of a
client; miqtu, a “fallen” person), tenants (ikkarar¯atu), hirees (agr¯u), levied
workers (tupšikk¯u, rendering ilku-duty), subordinates (m¯ar¯u, lit. “sons”), and
dozens of other terms for various kinds of workers. Finally, ganged labor under
force and supervision would have hampered the most common task we see
given to OB slaves: to carry messages, gather news, act as proxies for their
masters, and be deputed as pledges for productive loans. In economic terms,
slavery existed because it had these specific functions related to information
and credit, not to labor power.
Free movement was central to these functions. The business of merchants
and landowners required that they be in more than one place at one time, and
few communications technologies were available to solve them. To check on
the plowing of a field, collect a debt, deliver a letter, bring silver to a cred-
itor, herd sheep, or ready a departing caravan, Babylonian masters needed
slaves carrying letters to effect business when and wher e they themselves
could not attend. Whether they were dispatched to other households, out to
the countryside, or even to other cities altogether, it was precisely for their
ability to move freely and exercise independent judgment that slaves were
needed. For this reason, masters prized slaves not for brawn or beauty, but
for being “trustwor thy” (taklu), and socio-economic relations based on trust
are fundamentally incompatible with those based on force. If we assume that
“work” should look like hard labor under a beating sun, our expectations are
confounded by the economic functions slaves fulfilled in a mercantile economy
focused on commerce and credit.
The Experience of Household Slavery
Above, I have made two partially contradictory statements: that it is possible
to say something about the individual experiences of slaves and masters; but
also that we usually only have a historical window through single documents
(usually sale contracts), which do not allow us to reconstruct life histories.28
But a few vignettes and exceptions to the one-document rule allow us to
illuminate social experiences of slavery in markets, worksites, households, and
communities.
28 S. RICHARDSON
To begin from some admittedly crude statistics: the OB urban slave popu-
lation (and slavery seems an emphatically urban institution) hovered around
5 percent; the average number of slaves we see in a household is one; an
average household size was around five non-slave persons. Incidental evidence
suggests that more slaves were women than men, perhaps by as much as a
2:1 ratio.29 Given these parameters, we might find one slave in every fourth
household along a city street, and perhaps 500 slaves in a Mesopotamian
city with a population of ca. 10,000 people. Slaves were therefore common
and familiar (but not typical) members of urban communities. This image of
small-scale household slavery may differ from, e.g., institutional workhouses
known from the third millennium, where large groups of foreign men, women,
and children, some branded in their flesh, toiled in such places as a textile
factory called the é kešda, the “Bound-House.”30 We know that OB slaves
were visually identifiable by a shaved hairstyle called an abbuttu-lock, perhaps
wore distinctive garments, and were routinely watched and observed (espe-
cially when in proximity to city-gates, through which they were forbidden
to pass). It is enough to observe that Mesopotamian cities in the OB were
not particularly large, and that there was a great deal of face-to-face recog-
nition and knowledge of people (including slaves) by name. If the shackles
and racial identification we might think of as classic markers and constraints
of slave life were not therefore features of this particular historical context,
slaves were nevertheless under watch through pervasive community and indi-
vidual knowledge—thr ough the watchful nosiness of village life. (And slaves
were themselves participants in social surveillance, sometimes reporting on the
doings of neighbors and collecting juicy gossip.)
Contracts identified some slaves as “house-bor n,” which acknowledged
them as natally Babylonian. Some laws promised eventual manumission and
sometimes inheritance rights for such children born to slave women, and
sometimes for the mothers as well.31 To this limited extent, our sources
suggest that legal and social distinctions were made between slaves who had
been born into households and those who had not. Foreign slaves, meantime,
as far as we can make out, came from lands across a vast arc of the Near East,
stretching almost 2000 km from the region of modern Aleppo in the north-
west to Khuzestan in the southeast. The linguistic or cultural differences of
these slaves from Babylonians seem to have been unimportant, however, to any
documented dimension of their daily working life; foreignness was noted only
in contexts of procurement, sale, and law, aspects related entirely to issues of
property and jurisdiction. In no other respect is any preference or dispreference
for any particular kind of slave expressed in any source beyond stipulations
about health.
We do know that slaves could (and some did) marry, have children, live
together with their families in their own residences, own moveable property
(but not land or houses), and work fields to produce their own crops. How
common these dimensions of life were remains entirely unknown. Slaves could
marry into the families of their masters or take on a new status as caretakers
2 MESOPOTAMIAN SLAVERY 29
for the elderly, as second wives, or even as adoptees. In such cases, they might
in legal and social terms become freer and fuller social beings with a patronym
and household identity. Some slaves were owned within the community of
unmarried religious votaresses called nad¯ıtus, who conducted financial busi-
ness as independent householders, but it is unclear that life in this community
differed in terms of (m)any social or economic nor ms.
Whether “integrated” into masters’ families or not, slaves represented those
households to the community, and their behavior and reputation, good or
bad, “whether honored or despised” (as AbB III 11 puts it), reflected on
their masters. A slave’s reputation went hand-in-hand with the “trust” (taklu)
vested by a master to carry out household business. Slaves also had limited
legal capacity, for instance, to give testimony though not to bring suit, but
their legal liability was unequal, since they could be punished fully for various
crimes (although it was their masters who were often liable for civil penalties
and fines).
But slaves could exacerbate tensions within a household as well as reflect its
identity. Squabbles between siblings often emerged about who should inherit
a favored slave or whether a manumitted slave had received too much inheri-
tance themselves. As with any family, membership hardly resolved all social,
financial, or even emotional problems; and although manumission might
improve a freed slave’s life chances in social and economic terms, upward
mobility was severely limited in this society even for free persons. From the
point of view of a slave, life in a household probably represented safety
against true immiseration—starvation and homelessness, but “emancipation”
was hardly any protection against the kind of debt which could push them
right back down into slavery. There was much to lose, and not so much to
win; unfortunately, such outcomes can rarely be tracked.
The evidence for structural violence in OB slavery is unclear. Notwith-
standing the existence of children born to slave women, we have close to
zero evidence about sexual pr edation. At most, we find oblique statements
in letters like this rebuke: “With me, you talk about slave-women!” (AbB IX
19: 11–12); it is hardly clear what this means. Sexual abuse may have been so
common that it went unremarked upon; it may have been entirely unusual or
forbidden; we simply do not know. It is worth commenting only that namritu,
an adjective commonly describing female slaves and sometimes translated as
“good-looking” (therefore suggesting their acquisition for sexual use) actually
meant only “healthy.”32 The evidence for physical violence is similarly scant;
a very few references imply the use of beatings to compel obedience, but not
to any degree more than for the corporal punishment of wives, children, or
annoying neighbors. Again, this does not by itself tell us whether violence
was common, everyday, and unremarkable—or highly unusual and shocking.
Imprisonment and deprivation are in evidence here and there, as with a female
servant reporting that “cold and lice are eating me” in the house where she
was distrained (AbB VIII 100), but again the typicality of such complaints
is unclear. All we can say is that open r eferences to sexual abuse, physical
30 S. RICHARDSON
violence, and deprivation are few; nor even idioms or allegories connecting
slaves to tropes of sexuality or punishment.
If we look to literary sources for insights on social view of slavery, we find
a pervasive ambivalence. Slaves appear quite frequently in Sumerian literature
of the OB period; more than a hundred references can be found in dozens of
works. But the scribes composing and copying proverbs, hymns, and stories
were not interested in making sociological observations; they narratively posi-
tioned masters and slaves in symbolic roles to allegorize issues of authority
and position, especially to satirize haughty behavior. Reversals were common
devices, as in the proverb: “When the mistress left the house and the slave
woman came in from the street, away from her mistress the slave woman set
up her own banquet.”33 But no consistent viewpoint emerges: some proverbs
depict slaves as unfortunates (“Left-over clothes are the share of the slave-
woman’s child” [SP 4.43]), while others show them as greedy and overfed
(“Although the chickpea-flour of the home-born slaves is mixed with honey
and ghee, there is no end to their lamentations” [SP 1.47]). One could say
that they reflect an implicitly critical view of slavery insofar as it produced
people who were either pitiable or contemptible; but the divergent archetypes
reveal a profound ambivalence, seemingly to the purpose of explaining (or
explaining away) the moral dissonance of slaves’ simultaneous subjugation and
personhood.
Exiting Slavery
If Babylonian slavery was not by design meant to be an unalterable status,
legal terms of enslavement were nevertheless indefinite and could therefore
sometimes be made effectively permanent. That one might remain a slave
indefinitely is demonstrated by the fact that slaves were awarded as damages
in lawsuits and considered heritable, as shown by inheritance texts (t
.upp¯at
apl¯uti), dowries, and laws. That slaves were distinguished (or at least listed
separately) in inheritance texts from “possessions” (Akk. b¯ušu, i.e., mere
“things”) does little to ameliorate that they were chattels whose possession
could be transferred—by gift, sale, default, or bequest. It seems likely that any
time a s lave was transferred from one household to another—and therefore
to a new master—the term of their enslavement and/or how they might be
redeemed were both r endered less clear, not more. We may compare this to
the more permanent status of foreign slaves branded as temple or royal prop-
erty in the mid-third millennium; or the unfree temple oblates called širk¯u in
the mid-first millennium, who (since they were not slaves as such) “could be
neither manumitted nor sold.”34
We do read that some slaves ran away—but not from slavery, only to be
slaves in other households (sometimes nearby, sometimes in other cities) which
they preferred, and which presumably preferred them. Sometimes distrainers
kept possession of slaves in violation of contracts and/or against the will of
the slaves, but choice of residence by slaves played a remarkably common role
2 MESOPOTAMIAN SLAVERY 31
in resolving household conflicts. The Babylonian idiom, in fact, was not that
slaves had “run away,” but merely “wandered” (h
˘ alqu35) from where they
were meant to be, as a stray ox might; there seems no expectation on anyone’s
part that slaves were attempting any wholesale escape from status. To the
extent that the problem was policed (i.e., not much in this period), this was
aimed squarely at the recovery of the master’s property, not the punishment
of the slave.36 None of this precludes, as I have said, that slaves were watched,
marked, and controlled—only that there was not much of anywhere to run
away to. That being the case, restraint and arrest were exceptional measures
used in an environment where the mobility of slaves was required.
Most commonly we read of slaves being “redeemed.” This practice of
pawning a slave for credit and then redeeming him/her when profits were
realized (on the borrower’s sale of barley or the return of his caravan) is
attested many times. There were a number of ways to say this: often with verbs
denoting the satisfaction of debts (e.g., Akk. pat
.¯aru, wuššuru), but sometimes
only descriptively, i.e., that slaves were “returned,” “sent back,” or “handed
over” (Akk. târu, t
.ar¯adu, nad¯anu), where the reason is unstated. In many
instances, the original debt had been repaid, and the creditor returned the
distrained slave to the original master. We know of slaves who were sold and
re-sold in this way multiple times between different masters, probably among
circles of merchants who were all in debt to one another in turn. But often
we read about distrainers not returning a pawned slave, despite being paid or
going past the normal term of distraint. This is where the law stepped in: if a
slave or a master could bring proof that a creditor had either been compen-
sated or that the original debt was somehow invalid.37 It is clear, however,
that a slave “redeemed” from a debt was not freed from slavery, only from
distraint, because they were returned to the household of an original or earlier
master.
Finally, we m ay contrast redemption with emancipation: cer tain OB laws
mandated the “release” ( andur¯aru38) of slaves or prohibited “claims of slav-
ery” against individuals (e.g., ana ward¯utim ul iraggum39), usually for slave
women who bore children to a master. In such cases, a full manumission was
to be effected, involving a full exit out of slave status. (There were, however,
limits on the liability of the master to recognize the inheritance rights of the
children and/or the marital status of the mother.40) What is less clear is how
often this actually happened outside of the idealizing logic of the laws: most
texts about manumissions using these terms are only model contracts used by
scribes as learning tools; we have few actual practice texts. A few documents
describe manumissions in terms of ritual actions, such as a “purification” (Akk.
el¯elu, including with p¯utu 1a-2´, “to purify the forehead”41) or a declaration
before a divine image; in other contexts, these actions were used to indicate
the clearance of debts, which suggests that manumission was conceived of in
terms of financial solvency. The smashing of a pot at the feet of the slave (Akk.
šeb¯eru 1b), anointing with oil (Akk. šamnu 1f + šap¯aku), and the shaving-off
of the abbuttu-hairlock are also attested.
32 S. RICHARDSON
Manumissions most frequently coincided with a transition to some other
specific status. Many slaves being “freed” were simultaneously obliged by new
contracts to support their former and aging masters as “sons” or “daughters”,
to become votaries in temples, or to marry into the family.42 Few manumis-
sions simply released slaves from servitude, but facilitated some new status
with different obligations. The terms of the new arrangement seem typically
to have impr oved the lot of the former slave—there was almost no way for
things to get worse, after all—but we need to recognize it was usually a “free-
dom” which came at the price of a commitment to another obligation. The
contrast between “free” and “slave” is, in fact, one made by astoundingly few
texts, whether letters or contracts43; manumissions did not contrast slavery to
any other civil-legal status, such as am¯ıl¯utu or mušk¯en¯utu, “free” or “semi-
free.”44 This was a society in which the paramount social good was to ensure
that people were established somewhere within a household hierarchy, not to
award (much less guarantee) anyone legal and political rights in any commu-
nity larger than a household. That household membership (even as a slave)
guaranteed some measure of socio-economic security, however, implied its
opposite: the threat of exclusion (and thus possible starvation) was a form
of structural violence which hung over all members. Just because we are not
faced with a record full of beatings or sexual violence (which may anyway have
been unnoted and normalized aspects of class and gender relations) does not
mean that a very substantial and even existential threat of ostracization did
not lurk behind the façade of paternalism and compliance that characterized
mastery and slavery.
Conclusion: Paradigm and Variation
The foregoing has described a particular form of slavery, one in which force,
restraint, and labor extraction were relatively minor concerns, and reliability,
agency, identity, and mobility were considered advantageous characteristics of
a slave. Above all, slavery was geared toward mobilizing productive credit in an
emerging mercantile economy based on moving agricultural goods to market;
where the commodified value of slaves offered one of the few available financial
instruments in the marketplace.
It is not my purpose to depict Babylonian slavery as inherently mild or
humane, but to point to its economic function as based on social relations
of trust. Given this, a determination of “social death” seems unwarranted;
not all exploitation nullifies social identity. I mean to stress what the institu-
tion was for rather than how it compared to other historical instantiations,
e.g., the mass-labor context of racialized slavery of the antebellum US South.
The more urgent question about Babylonian slavery, then, is not about the
degree of exploitation of the slave, but the larger consequences for families
and communities in which all people were potentially salable—where every
mother, daughter, wife, younger brother, or son could be pledged against a
bargeful of barley. Perhaps they would be redeemed next week, perhaps in
2 MESOPOTAMIAN SLAVERY 33
three years, perhaps never. We have few sources directly commenting on the
social and emotional stresses this system must have put on Babylonian house-
holds, but it would be obtuse not to consider it as a larger context. It reminds
us that enduring structures of inequality are not only harmful to the people
who directly suffer deprivation and exploitation, but also to the people who
supposedly benefit but could also fall prey to them—for whom legacies of
privilege and moral compromise entail profound psychosocial harm. Systems
of inequality are har mful to all who participate in them.
How well does the OB situation (ca. 2000–1600 BCE) reflect slavery in
earlier and later periods of Mesopotamian history? This can be difficult to
answer; other periods produced documentation substantially different in quan-
tity and kind. Corpora prior to the OB, for instance, preserve large numbers of
administrative texts from specific times and places—Uruk ca. 3200, Lagaš ca.
2500, Nippur ca. 2300—but with several centuries remaining effectively silent,
and with few contracts and no letters. This situation begins to improve around
fifty years before the OB begins (i.e., around 2050) under the Third Dynasty
of Ur, when commercial documents become more abundant and letters begin
to appear. The documentary record following the OB, meanwhile, is rela-
tively sparse from ca. 1600–750, after which point it picks up momentum
and widens into a vast river, down to the fourth century, producing texts of
all kinds—letters, contracts, laws, and literature. The three thousand years of
Mesopotamian history thus (very!) briefly profiled are not known to us on
a consistent documentary basis in temporal, spatial, or generic terms, which
makes comparisons difficult; the available evidence may artificially emphasize
or elide different aspects of slavery.
What we can do is note large-type differences in period documentation,
if not necessarily in practice. We may begin by noting that the term(s) for
“slave(ry)” develop somewhat later than the institution’s apparent existence.
The term “sag + arad2/geme2” from the Uruk period (3300–2900) does
not so clearly designate unfree laborers . It is not until the Akkadian period
(2350–2200) that we have our first evidence for the sale of slaves by contract
(a commercial practice extending across the Near East over time)45; only by
the Ur III period (ca. 2112–2004) do we see slaves given as pledges for
loans. During these earlier periods (and later ones as well), not only did other
words for unfree persons persist alongside slaves, but terms distinguishing
them from “free” persons were relatively undeveloped. The OB introduced the
concept of a “gentleman” or “free” class (aw¯ıl¯u) but it remained inchoate in
juridical and social terms. By Middle Babylonian times, a privileged status for
city-dwellers existed (kidinn¯utu, referring to divine protection), though this
substantially meant tax exemptions rather than privileges over other classes.
Only by around 700 was there a clearly marked free class called m¯ar banî (lit.,
“grown son”) openly contrasted to slave status. The legal and social differ-
ences between slaves and non-slaves is therefore not clear to us until relatively
late times. This complicates our thinking, and invites us to think about why
34 S. RICHARDSON
a distinction between slavery and freedom was not made more systematically
than it was.
Another possible major difference between OB and third-millennium
slavery is the question of whether, how often, and how many slaves were
obtained through capture in war. There remains a low-level and sustained
(but perhaps irreconcilable) debate about whether the (undoubted) trains of
prisoners brought back to Mesopotamia from foreign wars and set to work
in large groups on temple and palace estates (as described by Early Dynastic
and Akkadian kings) were saleable chattel. The massed POW laborers (perhaps
even stationed in labor camps) and the individual slaves of private households,
though both victims of force, may have formed distinct sub-populations, casu-
alties of substantially different systems of inequality.46 We should, of course,
stop to ask how relevant our modern distinctions about status would have
been to a captured Subarean or Elamite soldier in, e.g., 2250 BCE: roped
around the neck, marched hundreds of miles away from home to the strange
land of Sumer, branded in his flesh, and forced to work on a temple estate or
gifted to a local grandee by the king, he probably cared little what his precise
juridical status was or whether the people who oversaw his work called him a
“slave” or not. We can only say that slavery in subsequent times, beginning
with the Ur III period, took on a more emphatically financial and commercial
character, with little war capture or massed labor.
A more subtle difference can be seen in the Neo-Babylonian period. The
use of slaves as pledges for loans had remained a long-term feature of Baby-
lonian commerce, but the enslavement of free persons for debt became
effectively illegal after ca. 700, effectively limiting distraint to persons already
in slave or subordinate status. Further, manumissions and other exits from
slavery seem to have become rarer and more conditional, and permanent
marking practices such as branding now obtained (in contrast to the special
haircut of the OB, which could obviously be undone). These changes may
indicate that the greater specification of the category “free” (m¯ar banî)may
have come at the price of a hardening of the category “unfree,” with less social
mobility (either upward or downward).
But Babylonian slavery largely remained a stable set of practices. In all
periods, we see the enslavement of both Babylonians and foreign persons,
and of men, women, and children. Slavery remained a highly individualized
arena of interaction between masters and slaves known to each other by name,
identity, and character. After the third millennium, there are fewer instances
of unfree labor at scale, for any kind of labor being particular to slaves, or for
slaves as creating any outsized economic value in either primary or secondary
production. Despite a possible increased stratification between slave and free
persons in the latest periods, slavery remained unmarked in moral or racial
terms; little suggests that slavery was ever considered natural to any particular
(set of) person(s).
So I briefly return to the very difficult questions posed at the outset of this
essay about slavery’s implications for the personhood of everyone involved in
2 MESOPOTAMIAN SLAVERY 35
this culture—sellers, buyers, creditors, debtors, merchants, judges, witnesses,
neighbors, bystanders, and, of course, slaves. On the one hand, one could
emphasize the human scale and temporary nature of Babylonian slavery as less
harmful than more dehumanizing instantiations of the institution; we do not
find the routine violations of human rights and brutality characteristic of some
other historical cases. The close context of household life, an array of expres-
sions of care and affect, relations of trust, and a seemingly high incidence
of mixed free and slave marriages all suggest some degree of personhood for
slaves. The need to use slaves as an extension of a master’s authority suggests
that violence would undermine their economic purpose. Full “social death”
cannot describe this form of slavery, even as contradictions and double stan-
dards persisted (as they do in all cultures). To this extent, Babylonian slavery
looks “nicer.”
On the other hand, Babylonia could be said to be responsible for a gener-
ative and standardizing role in the globalization of slavery, since it stands at
the opening of a very long, continuous, literate, and self-conscious historical
tradition. We have to wonder about the role of the writing system in partic-
ular, almost obsessively organized to create categories of meaning, to mark for
the first time the social worth of cer tain p ersons in relation to others; writing
itself was a means of specification. It is not difficult to see writing’s categorical
imperatives extending first to the practices of war and commerce that ulti-
mately connected individual Babylonian households with far-flung parts of the
world; then to a regularization of the financial means of specification for debt
which translated categories of personhood into commodities; and, finally, the
hardening of these terms into legal rules and social institutions. The appar-
ently lower levels of violence and inequality visible in the Babylonian record
for slavery was not a product of moral imperatives, but of insufficiencies of
financial and technological means: this version of slavery was not a matter of
being “nice,” just of a social engineering calibrated to a specific economic end.
If this culture had had the capacity to intensify production and control people
at scale, it would have been perfectly happy to do it. If only for normalizing
slavery, the Babylonian case was indeed “foundational” to the globalization
of the institution throughout the pre-modern world—out to other ancient
Near Eastern cultures, which adopted many Mesopotamian standards of price,
status, s ale formulae, and legal practice—hence to Greece, Rome, the rising
Islamic world, medieval Europe, and beyond.
Notes
1. All dates used in this article other than bibliographic contexts are BCE.
The major periods of Babylonian history include: Uruk, ca. 3300–2900;
Early Dynastic, ca. 2900–2350; Akkad, 2334–2192; Ur III, 2112–2004;
Old Babylonian, ca. 2017–1595; Kassite, ca. 1475–1154; Middle Babylo-
nian, 1154–ca. 750; Neo-Babylonian, ca. 750–539; Persian, 539–331; Seleucid,
331–63. “Sum.” refers to terms in Sumerian; “Akk.” to terms in Akkadian.
36 S. RICHARDSON
Translations of literary works follow the Electronic Text Corpus of Sumerian
Literature (ETCSL: https://etcsl.orinst.ox.ac.uk); refer ences to laws follow
Martha Roth’s Law Collections from Mesopotamia and Asia Minor (Atlanta:
Scholars Press, 1995); translations of Akkadian words, The Assyrian Dictionary
of the University of Chicago (CAD) (Chicago: The Oriental Institute, 1964–
2010). The abbreviation AbB refers to the series Altbabylonische Briefe in
Umschrift und Übersetzung (Leiden: Brill, 1964–2005).
2. This is supported by studies of other Mesopotamian periods as well: see, e.g.,
Heather Baker, “Degrees of Freedom: Slavery in Mid-First Millennium BC
Babylonia,” World Archaeology 33, no. 1 (2001): 23, reporting: “The tablets
rarely give us information as to what duties the slaves performed”; and Seth
Richardson, “Walking Capital: The Economic Function and Social Location of
Babylonian Servitude,” Journal of Global Slavery 4, no. 3 (2019): n. 15 for
literature. Cf. Vitali Bartash “Going for the Subar ean Brand: The Import of
Labor in Early Babylonia,” Journal of Near Eastern Studies 77, no. 2 (2018):
263–78. Bartash argues for the economic significance of slave labor in Sargonic
Babylonia.
3. I prefer “master” to “owner” partly because it inclines to the semantics of
Akkadian b¯elu, but also for the argument made below that Mesopotamian debt
slavery typically positioned the slave as a pledge, where possession was not
indentical to ownership (if conditionally).
4. See especially Richardson, “Walking Capital,” 285–342.
5. See F. Van Koppen, “Geography of the Slave Trade,” in Mesopotamian Dark
Age, ed. H. Hunger and R. Pruzsinszky (Vienna: ¨
Osterreichischen Akademie
der Wissenschaften, 2004); Seth Richardson, “Origin of Foreign Slaves in the
Late Babylonian Period,” KASKAL 17 (2020): 53–73.
6. Slave women are attested as witnesses to legal documents as early as the
Akkadian period.
7. Slave names ar e rarely given with a patronym, however, which denoted full and
recognized social identity. As for the re-naming of slaves, the evidence is mixed:
slave names of both undoubted foreign origin and probable “re-christenings”
are attested in many periods. But re-naming does not appear to have been a
consistent practice. For exceptions in the Neo-Babylonian period, see Baker,
“Degrees of Freedom,” 22.
8. Orlando Patterson characterized slavery in many (if not all) world cultures as
“social death” in these terms. See Orlando Patterson, Slavery and Social Death
(Cambridge, MA: Harvard University Press, 1982). Baker engages with Patter-
son’s ideas in her study of Neo-Assyrian slavery (Heather Baker, “Slavery and
Personhood,” in On Human Bondage, eds. J. Bodel and W. Scheidel [Malden,
MA: Wiley-Blackwell, 2017]).
9. Even OB “laws” at most punished slave flight by cutting off an ear rather than
inflicting death. Two points may be made, however: one, we have little indica-
tion of how often these punishments were actually practiced; two, Hammurabi’s
law to punish a slave rejecting his master (¶282, the last in his list) was pr ob-
ably a literary metaphor for the punishment of political rebellion and not the
enforcement of slavery per se.
10. See further my article in prep, “Mesopotamian Words for ‘Slave’: Opacity and
Mutability in Early Terms and Practices.”
11. The “laws” found in compositions like the Code of Hammurabi were not
statutes as such, but collections of the notional and paradigmatic decisions of
2 MESOPOTAMIAN SLAVERY 37
just rulers. That I occasionally refer to them by shorthand as “laws” is simply
to avoid annoying the reader.
12. One may compare with the Neo-Babylonian period, when enslavement is docu-
mented as a punishment for, e.g., infidelity (Baker, “Degrees of Freedom,”
21).
13. With the Akk. verb šâmu, “to sell,” distinct from ag¯aru, “to hire,” including
the hire of labor.
14. See esp. Richardson, “Origin of Foreign Slaves”; cf. Jacob J. de Ridder “Slavery
in Old Assyrian Documents,” in Kültepe International Meetings, Vol. II,eds.
F. Kulako˘glu and G. Barjamovic (Turnhout: Brepols, 2017), 56f.
15. Notably by the Neo-Assyrian (934–612) and Neo-Babylonian (626–539)
empires. There is substantial evidence for the capture and deportation of pris-
oners of war from Early Dynastic and Akkadian inscriptions. Instances in which
those prisoners are called “slaves,” however, are few.
16. Baker, “Slavery and Personhood,” 22; see also Ignace Gelb’s classic study
“Prisoners of War in Early Mesopotamia,” Journal of Near Eastern Studies
32 (1973): 70–98: “Is it possible to turn POWs into slaves? The answer is
theoretically, yes, practically, no!”
17. I thank Hervé Reculeau for this observation.
18. On the poor evidence for POWs/captives as slaves, the absence of
mass/institutional contexts, and the emphasis on individual social relations,
see Richardson, “Walking Capital,” 3 n. 9, 9 with n. 28, 16–17, and 23; and
Richardson, “Origin of Foreign Slaves,” 55–57.
19. See Bartash “Going for the Subarean Brand,” 263.
20. “Elamite,” for instance, was an ethnonym identifying both slaves and free
land-owning citizens. Similarly, V. Bartash (“Going for the Subarean Brand,”
273–74) notes that although most “Subareans” were slaves, some were merely
menial laborers.
21. This practice is attested from as early as Ur III times; see S. J. Garfinkle, “Shep-
herds, Merchants, and Credit,” JESHO 47, no. 1 (2004): 5, 7 n. 16, and
27.
22. See again Garfinkle, “Shepherds, Merchants, and Credit,” 9 and 27, writing of
the Ur III situation: “This does not mean that customar y loans could not result
in debt slavery, but only that this was not the primary intent of the cr editor in
all such arrangements.”
23. Richardson, “Origin of Foreign Slaves.”
24. Cf. the Laws of Hammurabi, ¶¶117–18, which specify that slaves (of any
origin) given into debt service could be kept or sold by the creditor if the
debt was not paid within three years, whereas non-slaves in the same position
were to be released.
25. SP 3.183 and “The Instructions of Šuruppak” ll. 131–133.
26. Richardson, “Walking Capital,” 305–23.
27. Bartash “Going for the Subarean Brand,” passim; Baker, “Degrees of
Freedom,” 22.
28. Cf. Bartash “Going for the Subarean Brand,” 269–70, on the (brief) life history
of a third-millennium slave woman named “Goody.”
29. See, e.g., Bartash “Going for the Subarean Brand.”
30. Bartash “Going for the Subarean Brand.”
38 S. RICHARDSON
31. Cf. Baker, “Slavery and Personhood,” 17–18, who is less sanguine about
apparent Mesopotamian protections against “natal alienation,” at least for
first-millennium contexts.
32. On the material conditions of OB slave life, see Richardson, “Walking Capital,”
passim.
33. Perhaps the most famous composition based on this kind of reversal/mirroring
rhetoric is the later Akkadian composition “The Dialogue of Pessimism,” in
which a slave mechanically endorses a series of contradictory actions proposed
by his master, satirizing his vacillation.
34. Bartash “Going for the Subarean Brand,” 275; Baker, “Slavery and Person-
hood,” 24. Baker notes that manumission is simply not attested in Neo-Assyrian
texts at all.
35. Cf. the less common Akk. ab¯atu, more straightforwardly “to flee.”
36. LE ¶¶33–34, “defraud”; LE ¶¶49–52; LH ¶¶15–16.
37. In which case penalties for the faulty distrainer might apply, including if the
slave had been injured or died while in their custody: LL ¶14 (two-fold); SLHF
viii 11–15 (equal amount); LE ¶¶22–23; LH ¶280.
38. See the note in CAD A/2 s.v. andur¯aru s. (p. 117) on the term’s use for
both release from slavery and cancelation of debt. The Sumerian term for
andur¯aru (ama.ar.gi4) means a “reversion to a previous state,” which may
imply a presumption that free status preceded slavery.
39. See also CAD Q s.v. qabû 3a-1´b´ and qer¯ebu 5a.
40. LL ¶25 (slave mother and children freed; no inheritance share); LL ¶26 (if the
master marries a slave after his first wife’s death, child gets inheritance share);
LH ¶119 (with pat
.¯aru), ¶¶170–71 (the latter with andur¯aru), ¶175; cf. LH
¶176a–b. More ambiguous is LH gap ¶s, which seems to suggest that a slave
whose master beats him need not be returned to him, but it is unclear if the
slave is actually freed.
41. This may indicate that the forehead was where the abbuttu-lock of hair was
placed: see Kraus Texte 25 cited s.v. abbuttu in CAD A/1 p. 49, “if he has a
low growth of hair on his forehead (p¯utišu) as far as his abbuttu…”; cf. SLHF
ii 4–13.
42. See exs. s.v. CAD Š/3 širku As.and šik¯utu s. (esp. A32117), devotion to a
temple, and šitektu s., given in adoption; CAD Q qerû v. 2b; pal¯ah
˘ u v. 5g,
marriage; h
˘ arr¯anu 9, with an adoption.
43. Cf. H. Baker, “Degrees of Freedom,” World Archaeology 33, no. 1 (2001): 21:
Neo-Babylonian slave sale contracts clarified that a sold person was (also) not
a free person; such distinctions were not made in the OB.
44. Cf. the slightly later situation at the nearby city of Nuzi, where manumission
could be qualified as a transfer to a semi-free status termed šalaššu. On concepts
of freedom, see Eva von Dassow, “Freedom in Ancient Near Eastern Societies,”
in The Oxford Handbook of Cuneiform Culture, eds. K. Radner and E. Robson
(Oxford: Oxford University Press, 2011), 205–24.
45. Slave sale contracts are next found in Elam in the twenty-first century; in
Assyria, by the nineteenth century at Aššur and then in its colony in Kaneš;
in Alalakh by the fifteenth century; in Emar by the fourteenth century; etc.
46. This distinction is probably even more valid for the case of war captives in the
much later imperial Neo-Assyrian period, when large-scale deportations mostly
had the character of mass resettlement rather than mass enslavement.
2 MESOPOTAMIAN SLAVERY 39
Further Readings
Baker, Heather. “Degrees of Freedom: Slavery in Mid-First Millennium BC Baby-
lonia.” World Archaeology 33, no. 1 (2001): 18–26.
———. “Slavery and Personhood in the Neo-Assyrian Empire.” In On Human
Bondage: After Slavery and Social Death, edited by J. Bodel and W. Scheidel, 15–30.
Malden, MA: W iley-Blackwell, 2017.
Bartash, Vitali. “Going for the Subarean Brand: The Import of Labor in Early
Babylonia.” Journal of Near Eastern Studies 77, no. 2 (2018): 263–78.
Culbertson, Laura, ed. Slaves and Households in the Near East. Chicago: The Oriental
Institute, 2011.
Englund, Robert K. “The Smell of the Cage.” Cuneiform Digital Library Journal 4
(2009): 1–27.
Garfinkle, Steven J. “Shepherds, Merchants, and Credit: Some Observations on
Lending Practices in Ur III Mesopotamia.” JESHO 47, no. 1 (2004): 1–30.
Reid, John Nicholas. Slavery in Early Mesopotamia from Late Uruk until the Fall of
Babylon in the Longue Durée. Ph.D. dissertation, University of Oxford, 2014.
Richardson, Seth. “Walking Capital: The Economic Function and Social Location of
Babylonian Servitude.” Journal of Global Slavery 4, no. 3 (2019): 285–342.
———. “The Origin of Foreign Slaves in the Late Old Babylonian Period.” KASKAL
17 (2020): 53–73.
de Ridder, Jacob. J. “Slavery in Old Assyrian Documents.” In Kültepe International
Meetings, Vol. II: Movement, Resources, Interaction, edited by F. Kulako˘glu and G.
Barjamovic, 49–61. Turnhout: Brepols, 2017.
Van Koppen, Frans, “The Geography of the Slave Trade and Northern Mesopotamia
in the Late Old Babylonian Period.” In Mesopotamian Dark Age Revisited,editedby
H. Hunger and R. Pruzsinszky, 9–33. Vienna: Verlag der ¨
Osterreichischen Akademie
der Wissenschaften, 2004.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 3
Ancient Egyptian Slavery
Ella Karev
I am the Lord your God, who has delivered you from the land of Egypt, the
house of slaves.
Exodus 20:2
Introduction
Slavery in Ancient Egypt still exerts a powerful hold over imaginations today.
Based on the Exodus narrative, the view of Egypt is often that of a “house
of slaves,” a society founded on slave labor. And yet, Egyptian documentation
reveals little about slaves and slavery, instead providing us with evidence of
marginalized groups, prisoners of war, and ambiguous terminology that blurs
the line between servant and slave.
The term “Ancient Egypt” suggests a static construction, but this picture is
not entirely accurate. Over its three millennia of history, Egyptian society
underwent significant changes: dynasties rose and fell, the kingdom was
broken apart and reunified, foreign rulers dominated the landscape, and the
fabric of society changed dramatically, including the nature of forced labor and
E. Karev (B)
Department of Near Eastern Languages and Civilizations, University of Chicago,
Chicago, IL, USA
e-mail: ellak@uchicago.edu
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_3
41
42 E. KAREV
slavery. And yet, there is some degree of ideological continuity, enforced by the
Egyptians themselves. The idea of an enduring Egypt was important to Egyp-
tian people, especially to the ruling elite; this is why, for example, Egyptian art
appears to change little over time.1 When foreign kings began to rule Egypt,
this ideological and artistic continuity was used as a method of legitimization
for their rule.
This chapter provides an overview of Egyptian slavery and bound labor
in its varying forms throughout Egyptian history, following Marcel van der
Linden’s analysis of coerced labor as dissected into methods of entry, forms
of extraction, and exit from slave status.2 Each of these sections proceeds
through the major phases of Egyptian history:3 the Old Kingdom (c. 2686–
2160 BCE), the Middle Kingdom (c. 2055–1650 BCE, the New Kingdom
(c. 1550–1069 BCE), the Saite and Persian Periods (664–332 BCE), and the
Ptolemaic Period (332 BCE–30 BCE). The Old, Middle, and New Kingdoms
are separated by so-called “Intermediate Periods” of unrest and division: the
First Intermediate Period (c. 2160–2055 BCE), Second Intermediate Period
(c. 1650–1550 BCE), and Third Intermediate Period (1069–664 BCE).
The broad nature of this overview comes with two caveats: first, the scope
of such a work means that the methods of coerced labor and slavery in each
period are only briefly described to provide a useful guide, though each histor-
ical period certainly warrants a full-length work of its own.4 The second
caveat is related to the first; the concision of this chapter suggests conti-
nuity. However, it cannot be stressed enough that each period was socially and
economically distinct from the others. This chapter serves as introduction to,
and synopsis of, certain aspects of slavery and compulsory labor that existed
throughout this broad timespan. However, this work makes no pretence to
be comprehensive, and an updated study on Egyptian enslavement remains a
desideratum.
There are two overarching considerations which remain the same
throughout Egyptian history and are important to note. The first relates to
our sources of information: unlike Mesopotamia, for most of Egyptian history
there is no preserved codification of laws;5 all observations about slavery must
therefore be gleaned from documentary and literary texts. Especially for the
former, quantity and quality vary considerably over time. Additionally, docu-
mentary texts are focused on single cases or events, and so any conclusions
drawn from them—unless these events are repeated—need to be taken in
stride.
Second, and particularly relevant to slavery studies, is the fact that until
the Ptolemaic period, there is virtually no evidence for large-scale Egyptian
slavery,6 agricultural or otherwise. Slavery was small-scale and mostly involved
domestic labor, though other kinds of labor are attested; this was not the
industrial agricultural slavery evidenced in the American South or Republican
Rome. Even in the few forms of large-scale coerced labor attested in Egypt,
it can be difficult to determine whether the subjects were indeed enslaved or
not.
3 ANCIENT EGYPTIAN SLAVERY 43
Some notes on ambiguity and terminology are also in order. Until the
Ptolemaic period and the incursion of Hellenistic forms of slavery, there
was little clear-cut distinction between slaves and other coerced laborers in
Egypt. Seemingly free people could be abducted and coerced into labor by
the state and punished if they attempted to flee; conversely, seemingly unfree
people who had been sold as pr operty could hold rights we would not expect
from slaves such as owning property, testifying in court, and negotiating the
terms of their enslavement. Therefore, a work on slaves in Egypt needs to be
discussed within the broader context of coer ced labor, as the lines between
“slave” and “coerced laborer” were often blurred. For instance, we cannot say
for certain whether women in Deir el-Medina whose labor was bought and
sold were owned in any real sense—either by the state and then assigned as
property to the workmen at Deir el-Medina, or privately—or whether they
functioned as servants who were ultimately paid for their labor; the past does
not provide enough information. This blurred distinction partially stems from
imprecision in terminology referring to slaves; this imprecision is simply a part
of Egyptian society. Unlike the modern world, in which there is a close connec-
tion between the exactitude of economic terminology and law, throughout
most of Egyptian history a precise translation of terms related to slavery,
servitude, and forced labor is not always possible.
For example, the word hem (h
. m) could be translated as “servant” or
“slave,” and appears in the dictionary with both of these possible meanings.7
The wor d itself has a number of uses, and could variously refer to: a priest;
the subject of a king; an agricultural worker for the state who may or may not
have been enslavede; a prisoner of war; a trusted housemate; or the object of a
small-scale slave sale. While these are all clearly positions of subordination, the
inclusion of the same term in these different contexts would seem to suggest
that these positions are identical, when they are not. Indeed, terminology is
so often so inexact and its context so unclear that deciding whether to label
certain individuals as “slave” or “servant” is usually the decision of a translator.
Historical Overview
Period name Date (BCE)8
Old Kingdom c. 2181–2025
First Intermediate Period c. 2160–2055
Middle Kingdom c. 2055–1773
Second Intermediate Period c. 1773–1550
New Kingdom c. 1550–1069
Third Intermediate Period c. 1069–664
Saite and Persian Periods 664–332
Ptolemaic Period 332–30
44 E. KAREV
Before delving into the nature of slavery and bound labor, it is useful to
begin with a broad historical overview of the main periods under analysis
here. Egyptian history does not begin in the Old Kingdom, but rather in
the period known as the Early Dynastic (c. 3000–2686 BCE), which covers
the 1st and 2nd Dynasties. However, there is limited written evidence from
this period, and most transactions were probably conducted orally, so little is
known about slavery or bound labor practices before the Old Kingdom. The
Old Kingdom formed the basis of Egyptian iconography, r eligion, and culture;
the great pyramids at Giza were built for Old Kingdom pharaohs.
The First Intermediate Period (2181–2025 BCE)—so-named because it
was the first of the three transitional periods between kingdoms—saw the
monarchy split into two competing branches. The reunification of Egypt after
nearly a century of conflict marked the end of the First Intermediate Period
and the beginning of the Middle Kingdom. The political division of the First
Intermediate Period ended with the unification of Egypt under the 12th
Dynasty pharaoh Nebhepetre Mentuhotep II. The Middle Kingdom (2055–
1773 BCE), was marked by internal stability and prosperity, but declined
after a period of dynastic chaos. This decline, combined with the inva-
sion of northern Egypt by the Hyksos from the Levant, began the Second
Intermediate Period (1773–1550 BCE).
The Hyksos were defeated and Egypt reunified under the pharaoh Ahmose
I, beginning the New Kingdom. The New Kingdom was stable excepting the
rule of Akhenaten, who instituted religious reform and moved the royal city
to El-Amarna, later abandoned after his reign. The New Kingdom ushered in
a period of political stability and imperialist expansion. Military campaigns of
the period extended Egypt’s influence in the Near East, expanding the borders
from modern-day Syria to the very edge of Nubia, in modern-day Sudan. This
expansion brought extraordinary wealth into Egypt and an influx of foreign
enslaved persons as prisoners of war, booty of conquests, and incursions into
existing slave markets.
The first millennium BCE in Egypt was marked by shifts in the ruling elite
of Egypt, including periods of foreign dominion. The wealth and expansion
of the New Kingdom was followed by the upheaval of the Third Interme-
diate Period (c. 1069–850 BCE) and the Persian invasion of Egypt in 525
BCE. Persian dominion lasted until 332 BCE, with a brief period of native
Egyptian rulers (404–343 BCE). The Persians ruled Egypt from afar with the
aid of provincial governors known as satraps. The constant changes in rule
and state fragmentation undoubtedly led to changes in labor practices and,
indeed, slavery. However, the period provides precious little documentation
with regard to slavery, with only a few isolated cases.
Persian domination over Egypt ended with Alexander the Great’s expan-
sion and subsequent conquering of the territory in 332 BCE. After his death
in 323, his empire was divided among his generals, with Ptolemy taking Egypt.
Ptolemy initially ruled as an extension of the heirs of Alexander, but ulti-
mately declared himself king in 305 BCE, founding the eponymous Ptolemaic
3 ANCIENT EGYPTIAN SLAVERY 45
Dynasty. The Ptolemaic Dynasty ruled from the royal city of Alexandria for
three centuries, until the Roman conquest in 30 BCE.
Although Ptolemaic Egypt retained much of traditional Egyptian culture,
the culture of the Ptolemaic period was heavily influenced by Hellenism,
including its labor practices and approach to slavery. The Ptolemies also intro-
duced Greek as the administrative language of Egypt, which continued well
into Roman dominion. Slavery is well-attested in the Ptolemaic period of
Egypt, and papyrological sources on slavery are published as a collection in
the Corpus der Ptolemäischen Sklaventexte.9
Entry into Enslavement
In his classification of coerced labor, Marcel van der Linden listed ten reasons
for entering a coerced labor relationship, of which the following seven are
relevant for Egyptian coerced labor: sale (between individuals or institutions);
hiring out or leasing of owned persons; self-sale; birth to an enslaved woman;
abduction; debt bondage; and taxation (levied in the form of labor). Though
all are attested, the incidence of these reasons differ over time, as summarized
in the following table, with an X indicating attestation. This table should also
be considered in the context of the varying availability of sources; the Old
Kingdom and the Saite and Persian periods, with fewer sources, are likely to
evidence fewer methods of entry, but that does not necessarily mean that these
methods did not exist, only that they are undocumented.
Old Kingdom
and First
Intermediate
Period
Middle
Kingdom and
Second
Intermediate
Period
New Kingdom
and Third
Intermediate
Period
Saite and
Persian
Periods
Ptolemaic
Period
Abduction ✕ ✕ ✕ ✕ ✕
Taxation ✕ ✕ ✕
Sale ✕ ✕ ✕
Hiring ✕ ✕
Self-sale ✕
Birth ✕ ✕ ✕
Debt ✕
These methods of entry differ ed not just in their incidence, but also in
their manifestation. For example, abduction in the Middle Kingdom took the
form of prisoners of war from Nubia and Libya drafted into the military; in
contrast, abduction in the Ptolemaic Period included native Egyptians enslaved
as punishment for their participation in rebellion.
46 E. KAREV
Abduction
Throughout Egyptian history, the best-attested method of entry into enslaved
status was through abduction as a prisoner of war. This is not to say that it was
the most common method, but simply the one for which there is the most
evidence, partially because the capture of prisoners of war was ideologically
important to the Egyptians as a mark of strength in the region. Autobiogra-
phies boasted of capture of enemies, and scenes of “smiting the enemy”10 and
leading away chained Nubians and Libyans became a mainstay in Egyptian ar t
even when Nubians and Libyans were no longer enemies of the state.
In the Old Kingdom, prisoners of war ( , “bound for life”) included
male Nubians and Libyans.11 Old Kingdom expeditions into Nubia are partic-
ularly well-attested in personal autobiographies of military officials12 which
boast their capture and enslavement of thousands of captives, both men and
women. Native Egyptians could also be abducted into labor, though it seems
that this practice was frowned upon; two autobiographies of officials list
among their accomplishments that they had never forced Egyptians into servi-
tude ( ), with one of these officials specifying that he had never forced any
daughters into enslavement.13
Conscription of prisoners of war into the military continued into the Middle
Kingdom, in which Nubian14 and Levantine troops were conscripted in order
to accompany Egyptian troops on quarrying expeditions as armed support.15
Women and children were also captured and enslaved en masse: P. B rooklyn
35.144616 lists among its eighty laborers forty-five “Asiatics” along with eight
“Asiatic” children. These were presumably captured in the Levant, part of the
Middle Kingdom expansion into that area.17
The New Kingdom experienced greater imperial Egyptian expansion into
the Levant, with the result that there are more records of people entering
enslavement through capture during wartime. The autobiography of General
Ahmose18 evidences how he “brought away” two enslaved women (h
. mwt )as
booty ( ) from his conquests in the Levant. As part of his retirement gift of
land and gold, he was gifted nineteen more slaves, some of whom bore foreign
names and originated from similar campaigns. The stela of Usertat, the viceroy
of Kush, lists among the achievements of the king that he possesses a “female
slave from Babylon, a female slave from Byblos, a little girl of Alalakh, and an
old woman of Arrapkha.”19
Tutankhamun’s Restoration Stela, written after the upheaval of his father
Akhenaten’s reign, mentions among his accomplishments that he had filled the
workhouses of his officials and priests with men and women, brought as booty
and entered into slave status. Papyrus Harris I, a New Kingdom administra-
tive document,20 includes a historical section that details the king’s capture of
persons in military campaigns in Nubia, Libya, and the Levant:
I brought back in great numbers those that my sword has spared, with their
hands tied behind their backs before my horses, and their wives and children in
3 ANCIENT EGYPTIAN SLAVERY 47
tens of thousands […] I imprisoned their leaders in fortresses bearing my name,
and I added to them chief archers and tribal chiefs, branded and enslaved, with
the cartouche of my name, their wives and children being treated in the same
way. (P. Harris I, 77.4–6)
Enslaved persons who entered their status as prisoners of war in the New
Kingdom were branded upon the right shoulder, as attested on a relief at
Medinet Habu. Branding continued into the Late Period, in which the practice
is applied to all enslaved persons.21
After the end of the New Kingdom and the upheaval of the Third Interme-
diate Period, Egypt no longer had the military presence in the Levant nor the
centralized state that would provide for such huge influxes of foreign labor.
However, the Kushite kings of the Late Period (c. 800 BCE) led campaigns
into what is now Gaza, capturing men who appear sporadically in the textual
record as “Men of the North.”22 The Persian satrap Arshames, residing in
Egypt in the fifth century BCE, records in his letters the importation of a
number of foreign slaves, including Cilicians and Iranians23;the former were
likely captured as part of Persia’s expansion into Anatolia.
In the Ptolemaic period, capture in war remained a common source of
slaves, though it is interesting to note that not all captured persons became
enslaved; some of the prisoners of war from the ongoing conflicts in Syria
ended up as free settlers in the Fayum.24 Prisoners of war who were captured
for the purpose of enslavement had to be registered with the state (along with
a 20 drachma payment, C. Ptol. Sklav.4.4–16). Failure to do so would result
in the confiscation of the enslaved person by the state (C. Ptol. Sklav. 3).
This method of entry into enslavement was not limited foreign prisoners; in
198 BCE, a royal decree was issued stating that Egyptians who took part in
rebellious unrest would then be enslaved (C. Ptol. Sklav. 9; 88).
Taxation
It is important to note at this juncture that although taxation levied in the
form of labor was a reason for entering a (possibly coerced) labor force
in Egypt, this does not necessarily mean that this labor could be defined
as enslavement in modern terminology. The aforementioned ambiguity of
terminology means that it is difficult to tell whether the people performing
this labor were coerced into doing so or not.25
Taxation through labor, usually referred to as “cor vée” in Egyptology, first
appears in the Old Kingdom, though the evidence is sparse and mostly indi-
rect. The most detailed information about the corvée actually comes from
the so-called “Exemption Decrees”26 of the Old Kingdom, which delineate
the people who are not eligible for corvée labor. Corvée recruitment of
eligible persons happened on a scheduled day (or series of days) known as
“Day of Corvée Recruitment ( ).” Officials responsible for raising
the corvée workforce were apparently tasked with a quota; a graffito from
48 E. KAREV
Wadi Hammamat evidences a military official who, knowing that the day
was coming, sets out to r ecruit 4350 persons to labor by the order of the
king. Recruitment was likely based on registration; a 4th Dynasty Gebelein
Papyrus,27 for example, preserves a list of men and women recruited for
temple construction. The class of laborers who performed the corvée as well
as agricultural labor28 were known as “merit ().” The meaning of the class
is unclear beyond the fact that they are subject to labor.
In the Middle Kingdom, taxation in the form of labor is attested, as
well as punishment for attempting to avoid such labor. P. Brooklyn 35.1446
includes a list of Egyptians who attempted to avoid their obligatory labor and
as a result are now imprisoned in a building termed “The Great Enclosure
().”29 The “Great Enclosure” of this papyrus was not the only one.30
In addition, records from Lahun evidence that in the event that a person
neglecting their state duties could not be found, the state could also seize
a replacement, usually a family member, and usually determined by sex (i.e.,
a father replaced his son, a mother replaced her daughter).31 Another Middle
Kingdom papyrus, P. Reisner I,32 includes a list of women who have been
“drawn (šd)” from the town for the purposes of weaving and housed in the
“Enclosure.” Fragments of official papers from the temple of Senwosret near
Illahun include letters that indicate that the authorities of the town could, at
will, take a female singer and a child from the “Enclosure” for the purpose of
state works.33
There is considerably less evidence for taxation as labor in the New
Kingdom, but there is evidence that free people were requisitioned alongside
slave laborers for state projects. The Horemheb Decree34 warns against such
a requisitioning by officials, who perhaps did so under the pretense of official
state business. The same decree states that officials who seize other kinds of
property will be imprisoned in the fortress of Sile, though it is not specified
whether these convicts are then impressed into labor or the military.35 From
the Saite and Persian period, taxes were collected by temple officials, in cash
or kind.36
Sale
Sales of slaves are only attested from the New Kingdom onward, but it is
important to note that this does not indicate that sales did not occur before
the New Kingdom; rather, it is likely that sales of slaves were oral and left
behind no written record.37 New Kingdom sales were likely small-scale—there
is no evidence of anything akin to a slave market—and took part between
individuals, though public ownership did occur. One of these private small-
scale sales is evidenced in P. Cairo 65739,38 in which a Syrian girl (very likely
the result of capture abroad) has been sold and re-named;39 the sale itself was
not documented in writing, and the transaction is recorded only as part of the
judicial record of a dispute over the payment.
3 ANCIENT EGYPTIAN SLAVERY 49
Enslaved persons could, by petition, transfer their ownership from private to
public. A petition from the Second Intermediate Period (P. Berlin 10470)40
evidences a slave woman named Senbet who is owned by both the people
of the town of Elephantine and by a man named Hebsy. The text approves
the petition of transferring Senbet to public ownership—“giving her to the
city”41—but the text is elusive on whether this new status grants Senbet
freedom42 or citizenship.43 This private ownership could be contested and
then settled in court.44
In the Third Intermediate Period, slaves are sold in a larger capacity as part
of an itemized list of land and commodities given to the king.45 These slaves
are sold at scale (5–30 people at a time) and are unnamed in the transaction.
The Saite and Persian Periods continue the New Kingdom trend of small-scale
slave sales46 between individuals, but integrate a statement from the slave sold,
perhaps influenced by the self-sales into enslavement from the same period.47
Sales of slaves are only sporadically attested in the Saite and Persian Periods.
In the Ptolemaic period, however, slave sales proliferate and are written exclu-
sively in Greek, which may be the result of an otherwise unattested decree.
These sales are usually direct (i.e., the seller and buyer know one another), but
sometimes include an intermediary who will deliver the slave to their owner
(e.g., Zaidelos of Idumea, who appears in P. Cair. Zen. 3.59374). There is also
evidence of slave auctions, in which the state was involved and collected a fee
for each bid and counterbid.48 Although there is no evidence of a slave market
per se, the involvement of an intermediary and the possibility for auction
suggests the development of a system designed to buy, sell, and deliver slaves.
Hiring
The New Kingdom provides the most sources for the hiring of slaves. The
workmen’s community at Deir el-Medina included female domestic servants,
who were assigned to certain households on certain days known as “days of
labor ( )” which could then be bought, sold, inherited, and traded
between fr ee villagers (e.g., O. Gardiner. 123, a sale of 480 days of service;
O. Gardiner. 90, a bequeathment of days of service from a father to his son).
However, it is somewhat unclear whether these women were owned in any
real sense—either by the state and then assigned as property to the workmen
at Deir el-Medina, or privately—or whether they functioned as servants who
were ultimately paid for their labor.49 Other slave women were also hired out.
Three papyri of the New Kingdom reference the purchase or leasing of slave
women for periods between two and seventeen days.50 These hired women
could pursue legal action if they were abused during their hiring period.51
The Saite and Persian Periods offer scant evidence of hiring. There are
some texts which could potentially refer to a leasing arrangement,52 but the
evidence is inconclusive. In the Ptolemaic Period, slaves with specialized skills
were often rented out for their labor. For instance, enslaved women could also
function as wet-nurses for a contracted period of time, for which their owner
50 E. KAREV
would receive additional payment.53 Interestingly, these contracts can involve
the nursing of a child who is himself a slave, a child borne between a slave and
the owner—and yet the slave who bore the child cannot nurse her own child.
Self-Sale
Self-sale is attested only in the Saite and Persian Periods, and even so, it is
poorly attested: a mere four documents evidence s elf-sale,54 three of them
belonging to the same individual who has sold himself to a new owner. These
self-sales are identical formulaically to sales of other commodities, with the
only difference being that the contracting party is also the object of the sale.
These have often been thought to represent debt bondage, though this is likely
not the case (see below, Debt Bondage).
Birth
In the Middle Kingdom, the hereditary nature of enslavement is implied by
the inclusion of children with their mothers in lists of slaves.55 The first direct
evidence that the children of slaves were also enslaved comes from the New
Kingdom, and it seems that the sale of enslaved children was discouraged until
they reached a certain age. In the archive of Ahmose son of Peniati,56 aman
requests that a female slave of his is returned after her mother complained:
Why is it that the female slave who was with me has been taken away to be
given to someone else? […] Let payment for her be accepted for her to be with
me, because she is only a child and unable to work […] her mother has written
to me, saying: ‘it is you who has allowed my daughter to be taken away […]’.
In the New Kingdom, it is unknown if these children were born to two
enslaved parents or if only the mother was enslaved; however, impregnating a
slave of the household was considered to be an undesirable act. In his tomb,
a man portraying himself as a model son proudly states among his accom-
plishments: “I did not know a slave-woman of [my father’s] house; I did not
impregnate a slave.”57
The self-sales of the Saite and Persian Periods include current and future
children in the transaction, even when it is the father who is selling himself,
suggesting that slave status could be passed on not just from the mother, at
least in the Egyptian record. The fifth-century archives of the Jewish commu-
nity at Elephantine evidences six people born to slave mothers, all of whom
appear to have inherited their slave status from their mother.58
In the Ptolemaic period, inherited slavery through birth was well-known
and even administered by the state. Children born to slave women were
referred to as “house-bred ( o„κoτρϕ`
ην)” or “house-born (o„κoγεν´
η).”59 These
house-born slaves would have to be registered with the state for the purposes
of taxation.60
3 ANCIENT EGYPTIAN SLAVERY 51
Debt Bondage
Debt bondage appears to not have been a method of entry into slavery in
Egypt before the Ptolemaic period, but is often cited as such,61 and warrants
some discussion. A distinction should be made between self-sale as a satis-
faction of debt versus transfer of title following seizure, as the two are often
confused.62 Putting up collateral in the form of the debtor’s own person could
lead to a form of conditional enslavement in which he is seized until the debt
is repaid, but self-sale is a different mechanism in that the transaction in itself
satisfies the debt.63 It is also, by this nature, more final than self-seizure; in
order to exit enslavement, a person who has sold themselves would then need
to pay the debt through self-purchase.64
Slaves, as well as free women and children, could be pledged as collat-
eral against loans from the First Intermediate Period onward. The question
remains as to what could actually happen if the debtor defaulted on his debt.
With slaves, it is relatively certain that they would simply be transferred into
the new owner’s possession, perhaps even without a written transfer of title.65
But what would happen to the children? Would they become the creditor’s
slaves,66 and thereby enter the status of enslavement through seizure? Hypo-
thetically, yes; however, the evidence suggests that this rarely happened in
practice, and there is only one extant case of a man’s seizure following his
father’s failure to return loaned copper tools. It is unclear if he actually did
any work during the time of their bondage, or if he simply served as collateral.
Either way, he was freed when the tools were returned.
There are four instances of self-sale into slavery dating to the Saite and
Persian periods, and it has been suggested that these were intended to satisfy
a debt, contrasting the classical account of Diodorus Siculus which states that
debt-slavery had been outlawed.67 This classical evidence has been shown to
be unreliable,68 but even still: whether as a result of systematic law reform
or not, the four self-sale documents do not seem to refer to a debt of any
kind and more likely represent self-sale in exchange for protection. Only in
the Ptolemaic Period is there direct evidence for debt bondage:69 apapyrus
recording sale tax on slave sales records a list of persons who have become
enslaved as a result of a debt.70 By the Late Ptolemaic Period, however, this
practice had been replaced by imprisonment.71
Extraction of Labor
Once an employer (or owner) has a laborer at his disposal, he must somehow
induce the laborer to actually work. But how? Marcel van der Linden
approached this question from the viewpoint that no one individual can be
forced to work through physical compulsion alone; rather, physical compulsion
is only one of the three factors which motivate a worker to work. These three
factors are compensation (i.e., wages), conditional force (i.e., punishment),
and commitment (e.g., loyalty).72 As with the entry into enslaved status,
52 E. KAREV
the following table summarizes the incidence of various factors motivating a
laborer to work, with an X indicating attestation.
Old Kingdom
and First
Intermediate
Period
Middle
Kingdom and
Second
Intermediate
Period
New Kingdom
and Third
Intermediate
Period
Saite and
Persian
Periods
Ptolemaic
Period
Indirect wages ✕ ✕ ✕ ✕
Conditional
force
✕ ✕ ✕ ✕
Commitment ✕ ✕
The issue here is not determining whether these factors were existent in
Egyptian slavery—as they likely were—but that only one generally warranted
written documentation and therefore is directly evidenced: conditional force.
The written documentation reflects corporal punishment, imprisonment, and
expulsion, but speaks little to the motivation behind the reasons why acoerced
laborer has chosen to work beyond the fear of retribution if they do not.
Nevertheless, there is some implicit evidence of the other motivators. Indi-
rect wages in the form of housing and protection (from practical fears such as
starvation or theoretical evils) was a feature of Egyptian systems of patronage:
a slave knew that an owner was obligated to provide protection, and this obli-
gation was referenced in self-sale into temple slavery. Commitment, the most
difficult of the three to delineate with regard to ancient labor is implied in
some Saite and Persian Period contracts.
Indirect Wages
Although wages appear to be irrelevant to slave labor, “compensation” also
includes “indirect wages,” such as housing, food, and social perks. This was
likely the premier motivation behind working as an enslaved laborer, and
perhaps even entry into enslavement: in return for labor and freedom, an
enslaved person received the obligation of their owner to protect them from
debt and poverty as well as serve as their advocate. In other words, slaves
were compensated in protection, both practical (e.g., food to eat, a roof to
live under) and social (e.g., literacy, the payment of a dowry, or the social
connections to have a complaint heard in a court of law).
It is assumed that coerced laborers were accommodated with a space to live
in the Old Kingdom.73 Direct evidence of (domestic) slaves sharing dwelling
space with their owners only comes in the Middle Kingdom papyri of the
landlord Heqanakht, who expels a slave of his from his home following her
bad behavior ( see below, Conditional Force). Entering a household carried
not just the practical benefit of a roof over one’s head, but also entry into the
3 ANCIENT EGYPTIAN SLAVERY 53
basic unit of social organization.74 Some slaves also learned marketable skills
from their owners, like writing.75
In the New Kingdom, institutional slavery provided slaves with a living
space and protected them from seizure for other types of labor, including state-
imposed taxation through labor. The Nauri Decree of Seti I (c. 1300 BCE),76
a protective decree of the personnel and lands attached to Seti’s temple of
Osiris at Abydos, warns against the “taking” of any of the personnel away from
temple grounds with the intent of employing them in other kinds agricultural
labor, whether by seizure, contract labor, or corvée. Entry into enslavement
was also a form of protection for those with no family ties or connections, such
as abandoned or orphaned children. A passage in P. Harris I,77 explaining the
origins of a priestly order, records the gathering of “children that had been
scatteredinthe serviceofothers” andtheir pledgetoserve thegod Ptah,
guaranteeing them a home and a profession.
In the Saite and Persian periods, there is contractual evidence that a slave-
owner paid the dowry of his female slave when she married a free man.78
This period also provides evidence that a slave could negotiate (or perhaps re-
negotiate) the rations given to him by his owner: in P. Rylands 7, a self-sold
slave requests more grain as compensation for himself and his family. Especially
in the Saite and Persian periods, entering a subordinate relationship with a
superior also involved social protection in the form of access to justice. The
wealthy and socially connected had the upper hand in court: they were more
likely to receive a favorable hearing, and could use their influence to obstruct
the complaints of the less-influential.79 Non-elites, then, traded subordinate
status in exchange for a voice in court.
The same terminology used for human patronage and protection was
employed in Persian-period letters to divinities, requesting protection from
legal threats and physical abuse in exchange for subordinate status. The
word used to indicate this subordinate status, bak (), is thesameone
used for slaves from the New Kingdom onward. In the Ptolemaic Period, at
least 37 Demotic documents evidence petitions to the god, with a suppli-
cant dedicating him/herself—along with a payment of money—to become a
“slave”80 to the god for 99 years in exchange for protection from natural (and
supernatural) threats.81
Conditional Force
Evidence from the Middle Kingdom shows that conditional force in the form
of imprisonment, corporal punishment, expulsion, or even death was likely
a motivating factor in labor extraction. Letters from Illahun attest to the
attempted escape and recapture of a slave, a man named Sobkemhab, who
is then housed in an Enclosure for his hearing and ultimately put to death
for his crime (P. Kahun 34). The punishment for not appearing for coerced
labor was, as noted above, seizure of a substitute; but if a substitute could not
be found, a person was sentenced to coerced labor for life, a status passed on
54 E. KAREV
to children.82 This is the case of one fugitive conscript laborer, Mentuhotep
son of Sabes, who is ordered by the court to be “given to the ploughlands
together with his people (i.e., his family) forever.”83
Household slaves could also seemingly be expelled from their homes as a
punitive measure. The private letters of the landowner Heqanakht attest to his
dismissal of a female slave named Senen for her bad behavior,84 instructing
his family to not even let her spend a single night in the house. Bad behavior
of slaves could also be punished corporally. In the literary tale recorded on
the Westcar Papyrus,85 a slave argues with her owner and receives a corporal
beating both from her owner, and, after she flees, from her own brother as
additional punishment.
From the New Kingdom until the Ptolemaic Period, conditional force may
have been relevant only insofar as the invocation of criminal law, and was not
specific to slaves. A New Kingdom judicial papyrus86 records an interaction
between a slave and a gang of tomb robbers who asked him to participate in
criminal activity. His response—“am I, who came from Syria, one to be sent to
Kush?”87—invokes a punishment also faced by his co-conspirators. At Deir el-
Medina, a slave and a necropolis worker, both thieves, are punished identically
for their crimes.88
In the Saite and Persian periods, Aramaic sources record that slaves would
be liable to punishment if they stole goods,89 and that this punishment was
likely corporal.90 But there is little evidence for specific slave-related crime,
like escape; when the self-sold slave woman of Louvre E706 alludes to escape
(see below, Exit), she does not mention any punishment beyond recapture. If
the motivation to labor lay in protection, then expulsion may have served as a
conditional threat.
In the Ptolemaic Period, conditional force took the form of detainment,
especially as punishment for flight91 or bad behavior.92 This is generally in line
with punishment in Ptolemaic Egypt, during which a broad array of offenses
could lead to detainment.93 Certain offenses—though the record is vague on
what they could be—could also result in corporal punishment, branding, and
deportation.94 While penalties differed for slaves and free men,95 there is once
again little evidence for specific slave-related crime such as escape.
Commitment
In the Saite and Persian Periods, the statement of commitment by the slave in
Demotic and Abnormal Hieratic sources (sales and self-sales) suggests a level
of personal loyalty, at the very least to a contractual relationship. This level
of personal loyalty may have been a factor before, but simply left unspoken,
unwritten, or both. These statements of commitment include a promise: “I
am your slave, forever ( ).” In other words, the workers are motivated to
work by their own oath to do so—sometimes, as in the case of the Demotic
Louvre E706, as reflected by an actual oath to Amun.96
3 ANCIENT EGYPTIAN SLAVERY 55
Exit from Enslavement
According to van der Linden, a person can terminate a coerced labor rela-
tionship on the basis of seven variables, divided into two categories: physical
compulsion (i.e., forced to leave or forced to stay); and constrained choice (can
leave, but chooses not to).97 Some of these methods of exit are not germane to
Late Period Egypt. Methods such as “unconditional exit” would not warrant
a contract of any sort in the Egyptian legal tradition, and therefore we have no
way of knowing if such a practice existed; “exit forced by another power” (as
an example, van der Linden cites the British abolition of slavery) are similarly
not applicable.
Of van der Linden’s seven methods of exit, only three are pertinent to
Egyptian slavery: exit despite impediment (i.e., escape), though the evidence
is minimal; conditional exit (i.e., marriage, adoption, and manumission, the
latter of which listed obligations of its manumitted party); and death, which
is the final and irrevocable termination of a labor relationship.
Escape
There is virtually no documentary evidence of escaped slaves in Egypt before
the Ptolemaic Period. This is not to say that slaves did not escape, but only that
evidence of their escape or their pursuit have rarely been preserved. One such
piece of evidence is from the New Kingdom, a letter recording the pursuit of
two slaves who had escaped from the palace and fled into the desert.98 They
are pursued at top speed99 by a troop commander, suggesting the urgency
of his mission. It is not known if these slaves were particularly impor tant—
prompting their pursuit and its recording—or if this was standard procedure.
In the Persian Period, the self-sold woman of Louvre E706 hints at the
possibility of recapture after escape, suggesting that her new owner may take
her if he finds her in another house: “I am your slave […] you are entitled to
me (lit. behind me) in any house in which you will find me.” The more conclu-
sive Persian-period evidence is from the Arsames archive: a letter dated to
the middle of the fifth century BCE (TADA6.3) records an official complaint
filed by a man named Psamshek. In it, he requests for Arsames to punish
eight slaves—belonging to Psamshek’s father—who he claims have taken his
property and fled. Psamshek is apparently intending to recapture these slaves
though he does not indicate how; his request is that, when Psamshek brings
the (re-captured) men to the official Artavanta, that Artavanta will punish
them as Psamshek sees fit.
In the Ptolemaic Period, slaves regularly escaped, as evidenced in notices
advertising their escape, calling upon authorities to assist in the recovery of
“lost property”100 as well as letters regarding their recapture.101 It is unclear
how successful enslaved persons were in escaping, but they often took valuable
items with them,102 or even the mortgage on a house103 to start a new life.
56 E. KAREV
As evidenced by letters104 and trial records, some freed persons helped these
fugitive slaves in their flight and risked trial to do so.105
Conditional Exit
A conditional exit from coerced labor is one in which a worker has to meet
obligations before being allowed to leave. van der Linden cites indentured
laborers as an example, as they had to first complete the duration of their
labor. In Egypt, conditional exit manifested in marriage to a specific person
(perhaps someone who would be difficult to marry off otherwise) as well as
adoption and manumission (both of which obligated a newly freed person to
“act as a son”).
In the New Kingdom, though it is unclear if marriage provided an exit from
slavery in itself, it was seemingly possible to exit enslavement by meeting the
conditions of a specific marriage, a quid pro quo between the owner and the
slave. This was the case in an inscription evidencing a slave named Sabastet
who had been captured in battle (“taken prisoner with my own arm when I
accompanied the king”). Sabastet agreed to marry his owner’s blind niece in
exchange for the right to “leave the house.”106
Adoption and contractual manumission are near-identical methods of exit
from slavery from the New Kingdom to the Ptolemaic Period. There is only
one extant manumission contract, and much like adoption, it obligates its
subjects to act as children to their ex-owner. A New Kingdom adoption
papyrus (P. Ashmolean Museum 1945.96) details how the children of an
enslaved woman are adopted by her married owners, and her daughter is
married to the owner’s brother. The condition for their manumission is the
adoption, which is intended to reorganise the hierarchy of inheritance within
the family.107
Adoption in the Saite and Persian Periods also obligated the manumitted
parties to act as children in exchange for their freedom.108 In the case of a
male child, this meant both perpetuation of the family line as well as upholding
the religious responsibilities of an eldest son, including presenting the neces-
sary funerary offerings after a parent dies.109 The one surviving pre-Ptolemaic
manumission contract110 evidences a near-identical condition: in exchange for
manumission, the two freed slaves promise to take care of their “father” in his
old age, or be liable for a heavy monetary penalty.
In the Ptolemaic period, manumission through contract occurred more
frequently,111 but evidence is still limited; though only six of these contracts
are attested (C. Ptol. Sklav. 28–34), and the appearance of freed slaves (i.e.,
post-manumission) is rare.112 Manumission was usually embedded into wills,
and came with two conditions: first, that the owner had died; and second,
that the slaves had stayed with their owner “as faithful servants” as long as the
owner lived.113 Once freed, however, the slaves had no obligations to their
previous owner, unlike the adoption/manumission contracts of earlier periods.
3 ANCIENT EGYPTIAN SLAVERY 57
Conclusions
The methods of entry into, exit from, and extraction of enslaved labor differ
substantially throughout the history of Egypt from the Old Kingdom to the
Ptolemaic Period. As the state underwent significant societal and economic
changes, so too did the approach to enslavement and coerced labor.
Despite these considerable variations in slave labor, there is one ideolog-
ical constant in Egyptian slavery: the desire for pr otection, at the cost of
subordination. In all of the periods in which slavery is attested, patronage and
protection are held in high regar d in Egyptian society, whether this protection
originates from the gods, the king, high officials, or simply another person who
can provide a degree of protection. This ideology is apparent in teachings and
wisdom texts from the Old Kingdom to the Ptolemaic Period: instructions
urge to seek out “a strong superior”114 when one has been injured and warn
against taking a superior to “without protection.”115
In return, superiors were morally obligated to take care of their dependents,
whether these dependents were their own children or other members of their
household—including slaves—from illness, lawsuits, and illegal seizure.116 To
be dependent and protected, then, was preferable to being independent and
unprotected, even if it meant becoming a slave. Entering the status of slavery
meant entering a household or an institution, albeit at a lower social stratum,
but in return gave the enslaved person practical protection from debt and
starvation as well as social protection through a connection to a high-status
individual. Thus, we see an overlap between the vocabulary of slavery and
other forms of voluntary subordination for which a person would receive
protection, such as adoption, pleas to a god for patronage, or a fulfilling of
responsibility to the state in exchange for protection by the king.
This aspect of slavery in Egypt throws a wrench in a wholesale accep-
tance of Orlando Patterson’s “social death” theory117; through protection and
patronage, slavery often offered its participants an entry into the protective
fold of society, rather than “killing” them through social alienation. Indeed,
it can be argued that the more socially alienated persons were those who
were not enslaved, but lay outside the norms of Egyptian society and there-
fore lacked protection: unclaimed orphans, vagrants, criminals, and foreigners.
These people would have to seek out some way of entering the social fabric,
including slavery, to find the protection they needed to function as members
of society. This is not to say that Egyptian slavery was an institution with
humane intentions; even at its best—a reciprocal relationship exchanging labor
for protection—it still involved the commodification of human bodies, and the
crucial point is that these were not mutually exclusive concepts. This contex-
tualization of slavery within Egyptian social and legal mores requires us to
recognize both the exploitation inherent to slavery and human commodifica-
tion as well as the Egyptians’ justification for it through language of patronage
and protection. A recognition of these social facts as co-existent in Egyptian
thought points to a critical paradox, hardly unique for cultures with slavery:
58 E. KAREV
an awareness of moral goods along with inequalities which could never be
satisfactorily resolved.
Notes
1. Egyptian art certainly changes over time, but the degree of change is small
enough that we can speak of a relatively uniform and recognizable “character”
of Egyptian art. See H. Schäfer, Principles of Egyptian Art, trans. John Baines
(Oxford: Griffith Institute, 1974), 14–36.
2. M. van der Linden, “Dissecting Coerced Labor,” in On Coerced Labor: Work
and Compulsion after Chattel Slavery, eds. Marcel van der Linden and Magaly
Rodríguez García (Leiden: Brill, 2016), 293–322.
3. Unless otherwise noted, all dates are according to I. Shaw, ed., The Oxford
History of Ancient Egypt (Oxford; New York: Oxford University Press, 2003).
References to primary sources differ by language. For Greek sources, refer-
ences follow the Checklist of Editions, found online at https://papyri.info/
docs/checklist; references for sources in Egyptian follow their museum cata-
logue number (e.g., P. Brooklyn = papyrus housed in the Brooklyn Museum);
references in Aramaic refer to their placement in the four-volume publication
by B. Porten and A. Yardeni, Textbook of Aramaic Documents from Ancient
Egypt (Jerusalem: Hebrew University Department of the History of the Jewish
People, 1986). Each volume is assigned a letter value (1986 = TADA; 1989
= TADB; 1993 = TADC; 1999 = TADD).
4. One needs only to consider that the edited volume on labor in the ancient
Near East—M. Powell, ed., Labor in the Ancient Near East (New Haven:
American Oriental Society, 1987)—included three essays on Egypt, covering
nearly 210 pages.
5. This statement excludes the Codex Hermopolis, a set of laws dated to the
Late Period but written in the Ptolemaic Period; for the majority of Egyptian
history, no such source appears to have existed. For the Codex Hermopolis,
see K. Donker van Heel, ed., The Legal Manual of Hermopolis: (P. Mattha);
Text and Translation (Leiden: Papyrologisch Instituut, 1991).
6. Despite the quote opening this chapter, this includes any reference to the
Exodus narrative; though there is evidence for enslaved persons originating
from the Levant mainstream scholarship largely rejects the narrative as histor-
ical truth. The literatur e is too massive to cite here, but some resources
include D. Redford, “Exodus 1.11,” Vetus Testamentum 13 (1963): 401–
18; D. Redford, “An Egyptological Perspective on the Exodus Narratives,” in
Egypt, Israel, Sinai: Archaeological and Historical Relationships in the Biblical
Period, ed. A.F. Rainey (Tel Aviv: Tel Aviv University Press, 1987), 137–61;
J. Hoffmeier, Israel in Egypt: The Evidence for the Authenticity of the Exodus
Tradition (Oxford: Oxford University Press, 1999).
7. A. Erman and H. Grapow, Das Wörterbuch der ägyptischen Sprache (Berlin:
Akademie-Verlag, 1953), 87.
8. Before 664 BCE, absolute dates are only approximate; it is only after 664
BCE that regnal years, reigns, dynasties, and periods can all be correlated
with absolute dates.
9. R. Scholl, Corpus der Ptolemäischen Sklaventexte (Stuttgart: Steiner, 1990);
abbreviated here and elsewhere as C. Ptol. Sklav.
3 ANCIENT EGYPTIAN SLAVERY 59
10. On this motif, see J. ´
Sliwa, “Some Remarks Concerning Victorious Ruler
Representations in Egyptian Art,” Forschungen und Berichte 16 (1974):
97–117; M. Janzen, “The Iconography of Humiliation: The Depiction
and Treatment of Bound Foreigners in New Kingdom Egypt” (PhD diss.,
University of Memphis, 2013), 23–4, 311–3.
11. Expeditions into the Levant did take place, but are poorly attested. Expe-
ditions to Punt did not seem to have been a significant source of prisoners
of war. On Nubians, see M. Bietak “Zu den nubischen Bogenschutzen aus
Assiut. Ein Beitrag zur Geschichte der Ersten Zwischenzeit,” in Mélanges
Gamal Eddin Mokhtar, eds. Muhammad Mukht¯ar Paule Kriéger. Bibliothèque
d’Étude 97 (Cairo: Institut français d’archéologie orientale, 1985), 87–99.
12. Like Pepynakht; K. Sethe, Urkunden des alten Reichs (Leipzig: J.C. Heinrichs,
1903), 131–5.
13. Sethe, Urkunden, 217.
14. E. Feucht, “The Nubian Mercenaries of Gebelein during the First Interme-
diate Period,” Kush 9 (1961): 44–80; J. Darnell, “A Bureaucratic Challenge?
Archaeology and Administration in a Desert Environment (Second Millen-
nium B.C.E.),” in Ancient Egyptian Administration, ed. Juan Carlos Moreno
García (Leiden: Brill, 2013), 791.
15. Ibid., 817.
16. W. Hayes, A Papyrus of the Late Middle Kingdom in the Brooklyn Museum
(Papyrus Brooklyn 35.1446). (Brooklyn: The Brooklyn Museum, 1955).
17. On this Middle Kingdom expansion, see J.C. Moreno García, “Egypt, Old to
New Kingdom (2686–1069 BCE),” in The Oxford World History of Empire,
eds. Peter Bang, C.A. Bayly, Walter Scheidel (Oxford: Oxford University Press,
2021), 26–7.
18. K. Sethe, Urkunden der 18. Dynastie: historisch-biographische Urkunden
(Leipzig: J.C. Hinrichs, 1927), 1–10.
19. J. Darnell, “The Stela of the Viceroy Usertat (Bostona MFA 25.632), his
Shrine at Qasr Ibrahim, and the Festival of Nubian Tribute under Amenhotep
II,” ENIM 7 (2014): 239–76. In general on prisoners of war in the New
Kingdom, see U. Mati´c, Body and Frames of War in New Kingdom Egypt:
Violent Treatment of Enemies and Prisoners (Wiesbaden: Harrassowitz, 2019).
20. P. Grandet, Le papyrus Harris I (BM 9999) (Cairo: Institut français
d’archéologie orientale du Caire, 1994).
21. On this practice, see E. Karev, “‘Mark Them with My Mark’: Human
Branding in Egypt,” The Journal of Egyptian Archaeology 108 (2022):
191–203.
22. “Men of the North” appear in P. Vatican 10574; P. Leiden 1942/5.15; the
Apanage Stela of Iuwelot (Cairo JdE 31882); P. Louvre E3228e, P. Louvre
E3228d, and P. Louvre E3228c; for identification of the “northern region”
as Gaza, see J. Quaegebeur,“À propos de l’identification de la ‘Kadytis’
d’Hérodote avec la ville de Gaza,” in Immigration and Emigration Within the
Ancient Near East, Festschrift E. Lipi ´nski, eds. Karel Van Lerberghe, Antoon
Schoors, and Edward Lipi ´nski. OLA 65 (Leuven: Peeters, 1995), 245–70. See
also J. J. A. Ramírez, “Papyrus Vatican 38595: A Lease of a Man of the
North during the reign of Py”, Bollettino dei Monumenti Musei e Gallerie
Pontificie 36 (2019): 30–46.
60 E. KAREV
23. J. Ma and C. Tuplin, Aršama and his World: The Bodleian Letters in Context.
3 vols. Oxford Studies in Ancient Documents 14 (Oxford: Oxford University
Press, 2020).
24. E.g., P. Petr. 3. 104.
25. For a detailed look at corvée and people considered “corvéable,” see: J.C.
Moreno García, “La population mrt: une approche du problème de la servi-
tude dans l’Égypte du IIIe millénaire,” JEA 84 (1998): 71–83; C. Eyre,
“Work and Organisation of Work in the Old Kingdom,” in Labor in the
Ancient Near East, ed. Marvin Powell (New Haven: American Oriental
Society, 1987), 18–20.
26. R. Jasnow, “Egypt: Old Kingdom and First Intermediate Period,” in A History
of Ancient Near Eastern Law, ed. Raymond Westbrook (Leiden: Brill, 2003),
94–5; N. Strudwick, Administration of Egypt in the Old Kingdom (London:
KPI, 1985); E. Hornung, Grundzüge der ägyptischen Geschichte (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1978), 40.
27. P. Posener-Kriéger and S. Demichelis, I papiri di Gebelein: Scavi G. Farina
1935 (Geda: Turin, 2004).
28. The hieroglyphic writing of this term includes a hoe and stresses the agricul-
tural nature of these laborers. See Ogden Goelet, “Problems of Authority,
Compulsion, and Compensation in Ancient Egyptian Labor Practices,” in
Labor in the Ancient World, eds. Piotr Steinkeller and Michael Hudson
(Dresden: ISLET, 2015), 547, especially n. 87 and 88. See also S. Allam,
“Une class ouvriere: les merit,” in La dépendance rurale dans l’antiquité
égyptienne et proche-orientale, ed. Bernadette Menu (Cairo: Institut français
d’archéologie orientale, 2005), 123–56 and J. C. Moreno García, “Acquisition
de serfs durant la Première Période Intermédiaire: une étude d’histoire sociale
dans l’Égypte du IIIe millénaire,” Revue d’Égyptologie 51 (2000): 123–39.
29. Goelet, “Problems,” 572; S. Quirke, The Administration of Egypt in the Late
Middle Kingdom: The Hieratic Documents (New Malden: SIA Publishing,
1990), 127–86; C. Eyre, The Use of Documents in Pharaonic Egypt. Oxford
Studies in Ancient Documents (Oxford: Oxford University Press, 2013), 261–
3; W. Grajetzki, “Setting a State Anew: The Central Administration from
the End of the Old Kingdom to the End of the Middle Kingdom,” in
Ancient Egyptian Administration, ed. Juan Carlos Moreno García (Leiden:
Brill, 2013), 233–4.
30. It is unclear how many of these enclosures actually existed—and whether they
were all used to the same end—but throughout the late Middle Kingdom
they do appear to have been limited to the housing (and perhaps also coerced
labor) of individuals attempting to avoid their corvée duties. See S. Quirke,
“State and Labour in the Middle Kingdom: A Reconsideration of the Term
Chenert,” Revue d ‘Égyptologie 39 (1988): 83–106, especially p. 102.
31. On this practice, see Quirke, “State,” 89–90, with citations indem. As Quirke
notes, substitutions were not necessarily gender-based if a task had yet to be
allocated (i.e., conscription for general state services), but once the task had
already been determined, it would require differentiation by gender.
32. W. Simpson, Papyrus Reisner I: The Records of a Building Project in the Reign
of Sesostris I (Boston: Museum of Fine Arts, 1963). See also Quirke, “State,”
86–7.
33. Quirke, “State,” 87–8.
3 ANCIENT EGYPTIAN SLAVERY 61
34. J. Kruchten, Le décret d’Horemheb: Traduction, commentaire épigraphique,
philologique et institutionnel ( Bruxelles: Editions de l’Universite de Bruxelles,
1981); Sethe, Urkunden der 18. Dynastie, 2140–61.
35. Conscription into the military is favored by Kruchten, Le decret, 79.
36. B. Muhs, The Ancient Egyptian Economy, 3000–30 BC (Cambridge:
Cambridge University Press), 184; there was also taxation in the form of
labor, but less attested than in earlier periods.
37. On oral transactions of other goods, see Muhs, The Ancient Egyptian
Economy, 6, 52, 91, 140, 157.
38. A. Gardiner, “A Lawsuit Arising from the Purchase of Two Slaves,” The
Journal of Egyptian Archaeology 21 (1935): 140–6.
39. Gardiner, “A Lawsuit,” 142. Her name is a clue to her origin: gm.n.i’-h
. r-
i’mntt, meaning “I found (her) in the West” or “(she) who I found in the
West.”
40. P. Smither, “The Report Concerning the Slave-girl Senbet,” The Journal of
Egyptian Archaeology 34 (1948): 31–34; see also T. Hofmann, Zur sozialen
Bedeutung Zweier Begriffe für “Diener”: B3k und H
. m: Untersucht an Quellen
vom Alten Reich bis zur Ramessidenzeit (Basel: Schwabe, 2005), 133–9.
41. P. Berlin 10470, 3.5.
42. The conceptualization of “freedom” is nebulous in Egypt. For further, see E.
Karev, “Nemeh in Pharaonic Egypt: ‘Free’ or ‘Miserable’? (A Case Study of
Historical Semantics),” Zeitschrift für Geschichtswissenschaften (forthcoming).
43. For the suggestion that this transfer of ownership was for the purposes of
marriage, see W. Helck, “P. Berlin 10470,” Zeitschrift für Ägyptische Sprache
und Altertumskunde 115 (1988): 35–9.
44. A. Gardiner, “Four Papyri of the 18th Dynasty from Kahun [Berlin P. 9784,
Gurob P. II 1, Gurob P. II 2, Berlin P. 9785],” Zeitschrift für Ägyptische
Sprache und Altertumskunde 43 (1906): 27–47; P. Berlin 9785 on pp. 38–43.
45. So-called “oracular pronouncements.” See R. Ritner, The Libyan Anarchy:
Inscriptions from Egypt’s Third Intermediate Period. Writings from the Ancient
World 21 (Atlanta: Scholars Press, 2009), 166–72, 271–8.
46. For an interpretation of sale and self-sale texts of the Saite and Persian
periods as “cessions of services” rather than sales, see B. Menu, “Cessions de
services et engagements pour dette sous les rois Kouchites et Saïtes.” Revue
d’Égyptologie 36 (1985): 73–87.
47. Self-sales: P. Rylands 3, 5, and 6; Louvre E706; Sales: P. Leiden F1942/5.15;
P. Vatican 10547; P. Louvre E3228e; P. Turin 2122; P. Bibl. Nat. 223; P. Inv.
Sorbonne 1276 + 1277.
48. C. Ptol. Sklav. 5.
49. Favoring the former, see J. ˇ
Cerný, A Community of Workmen at Thebes in the
Ramesside Period (Cairo: Institut français d’archéologie orientale, 1973), 175–
81; for the latter argument, cf. Goelet, “Problems,” 546–47 and Hofmann,
Zur sozialen, 113–8.
50. Gardiner, “Four Papyri”; see also R. Navailles and F. Neveu, “Qu’entendait-
on par ‘journée d’esclave’ au Nouvel Empire?” Revue d’Égyptologie 40 (1989):
113–23.
51. B. Davies and J. Toivari, “Misuse of a Maidservant’s Services at Deir el-
Medina (O. CGC 25237, recto),” Studien zur Altägyptischen Kultur 24
(1997): 69–80.
62 E. KAREV
52. TADD7.9; P. Louvre E3228d. For the latter, see K. Donker van Heel, Dealing
with the Dead in Ancient Egypt: The Abnormal Hieratic Business Archive of
Petebaste Son of Peteamunip (7th Century BCE) (Cairo: American University
in Cairo Press, 2021).
53. As recorded in P. Tebt.2.399. This profession was not limited to enslaved
women; the only difference is that an enslaved nurse was represented by
her owner rather than r epresenting herself. See J. Straus, “Papyrological
Evidence,” in The Oxford Handbook of Greek and Roman Slaveries,eds.
Stephen Hodkinson, Marc Kleijwegt, and Kostas Vlassopoulos (Oxford:
Oxford University Press, 2016).
54. P. Rylands 3, P. Rylands 5, and P. Rylands 6; Louvre E706.
55. As in P. Brooklyn 35.1446; Hayes, APapyrus, 90.
56. P. Louvre 3230b; E. Peet, “Two Eighteenth Dynasty Letters, P. Louvre
3230,” Journal of Egyptian Archaeology 12 (1926): 70–4.
57. A. Gardiner, “The Tomb of Amenemhet, High Priest of Amon,” Zeitschrift
für Ägyptische Sprache und Altertumskunde 47 (1910): 87–97.
58. TADB3.9; TADB3.3; TADB3.6; TADB2.11.
59. Straus, “Papyrological”; I. Bie ˙zu ´nska-Małowist, L’esclavage dans l’Égypte
gréco-romaine: Première partie: période ptolémaïque (Wrocław: Zakład Naro-
dowy im. Ossoli´nskich, 1974), 49–54; Y. Rotman, “Slavery in Gr eco-Roman
Egypt,” in Law and Legal Practice in Egypt From Alexander to the Arab
Conquest: A Selection of Papyrological Sources in Translation, With Intro-
ductions and Commentary, eds. James Keenan, Joseph Manning, and Uri
Yiftach-Firanko (Cambridge: Cambridge University Press, 2014), 443.
60. P. Harr.1.61 = C. Ptol. Sklav. 8.
61. Menu, “Cessions,” 81–87.
62. E.g., F. van Koppen, “Geography of the Slave Trade,” in Mesopotamian Dark
Age, ed. H. Hunger and R. Pruzsinszky (Vienna: Verlag der Österreichischen
Akademie der Wissenschaften, 2004), 11 who equates self-sale with a creditor
claiming the pledges of a previous loan contract—in other words, seizur e
following default.
63. A. Testart, “The Extent and Significance of Debt Slavery,” Revue française de
Sociologie 43 (2002): 179–80.
64. We have no record of this happening in Egypt, but that doesn’t exclude it as
a possibility; we only have three self-sales, of which only two ar e potentially as
a result of an obligation—and they belong to the same individual. For further
on debt bondage in the Late Period, see E. Karev, “Debt Bondage in Late
Period Egypt (8th–5th Centuries BC)” (forthcoming).
65. E.g., in Abnormal Hieratic P. BM. 10113, the debtor states that the creditor
may take any of the property listed “without citing any document (i’w֑ d
¯ dk
. nbt
nb)”; see K. Donker van Heel, “Abnormal Hieratic and Early Demotic Texts
Collected by the Theban Choachytes in the Reign of Amasis. Papyri from the
Louvre Eisenlohr Lot” (PhD diss., Leiden University, 1995), 231.
66. As suggested in T. Markiewicz, “Security for Debt in the Demotic Papyri,”
Journal of Juristic Papyrology 35 (2005): 153, n. 32.
67. Diodorus Siculus, Bibliotheca Historia I.79.3, states that the pharaoh
Bocchoris abolished debt slavery; on this claim, see T. Markiewicz, “Bocchoris
the Lawgiver—Or Was He Really?” Journal of Egyptian History 1 (2008):
309–30.
3 ANCIENT EGYPTIAN SLAVERY 63
68. Markiewicz, “Bocchoris,” 321.
69. Bie ˙zu´nska-Małowist, L’esclavage, 29–49, esp. 43.
70. Sel. Pap.2.205; W. Westermann, Upon Slavery in Ptolemaic Egypt (New York:
Columbia University Press, 1929).
71. J. Bauschatz, Law and Enforcement in Ptolemaic Egypt (Cambridge:
Cambridge University Press, 2013), 236.
72. van der Linden, “Dissecting,” 306–10, graphic on p. 308.
73. Eyre, “Work,” 28–30.
74. J.C. Moreno García, “Households,” in The UCLA Encyclopedia of Egyp-
tology, eds. Elizabeth Frood and Willeke Wendrich (Los Angeles: UCLA
Press, 2012), http://digital2.library.ucla.edu/viewItem.do?ark=21198/zz0
02czx07. The house as a unit was socially important enough that its confis-
cation was employed as a punitive measure; D. Lorton, “The Treatment of
Criminals in Ancient Egypt: Through the New Kingdom,” Journal of the
Economic and Social History of the Orient 20 (1977): 2–64.
75. E.g., P. Kahun 35 = UCL 32210; M. Collier and S. Quirke, The UCL Lahun
Papyri: Letters (Oxford: Archaeopress, 2002), 133.
76. An extensively studied decree; for a partial bibliography cf. Goelet, “Prob-
lems,” 555, n. 109.
77. P. Harris I, 47.8–9.
78. TADB3.3; see also B. Porten and H. Szubin, “The Status of the Handmaiden
Tamet: A New Interpretation of Kraeling 2 (TAD B3.3),” Israel Law Review
29 (1995): 43–64.
79. J.C. Moreno García, “The ‘Other’ Administration: Patronage, Factions, and
Informal Networks of Power in Ancient Egypt,” in Ancient Egyptian Admin-
istration, ed. J.C. Moreno García, Section 1: Ancient Near East, vol. 104
(Leiden: Brill, 2013), 1030.
80. It’s not clear if this is metaphorical or actual enslavement. See K. Ryholt, “A
Self-Dedication Addressed to Anubis: Divine Protection against Malevolent
Forces or Forced Labor?” In Lotus and Laurel: Studies on Egyptian Language
andReligioninHonourofPaulJohnFrandsen, eds. Rune Nyord and Kim
Ryholt (Copenhagen: Carsten Niebuhr Institute of Near Eastern Studies,
Museum Tusculanum Press, 2015), 329–50. The fact that some of the people
dedicated in these contracts were children listed without a father has led some
to suggest actual enslavement or even temple prostitution (H. Thompson,
“Two Demotic Self-Dedications,” The Journal of Egyptian Archaeology 26
[1941)]: 68–78), though the latter is unlikely.
81. E.g., scorpion stings and snakebites, but also “dead men” and spirits; A.G.
Migahid, “Demotische Briefe an Götter von der Spät- bis zur Römerzeit”
(PhD diss., Bayerischen Julius-Maximilians-Universität, 1986).
82. A. Loprieno, “Slavery and Servitude,” in The UCLA Encyclopedia of Egyp-
tology, eds. Elizabeth Frood and Willeke Wendrich (Los Angeles: UCLA Press
2012), http://digital2.library.ucla.edu/viewItem.do?ark=21198/zz002djg3j;
Hayes, APapyrus, 65.
83. Hayes, APapyrus, 52–53; on the practice of substitution, see also Quirke,
“The Administration,” 162–63; K. Kóthay, “La notion de travail au Moyen
Empire. Implications sociales,” in L’organisation du travail en Égypte ancienne
et en Mésopotamie: Colloque Aidea - Nice 4-5 octobre 2004, ed. Bernadette
Menu (Cairo: Institut français d’archéologie orientale, 2010), 155–70.
64 E. KAREV
84. J. Allen, The Heqanakht Papyri (New York: The Metropolitan Museum of
Art, 2002).
85. P. Westcar, 12, 20; her owner is saddened by this beating, referring to her
slave as “the little girl who grew up in this house.” A. Blackman, The Story of
King Kheops and the Magicians: Transcribed from Papyrus Westcar (Reading:
J.V. Books, 1988).
86. P. BM 10052, 12, 20; E. Peet, The Great Tomb-Robberies of the Twentieth
Egyptian Dynasty (Oxford: Clarendon Press, 1930), 155.
87. P. BM 10052 12, 8; Peet, The Great, 154.
88. J. ˇ
Cerný, “Restitution of, and Penalty Attaching to, Stolen Property in
Ramesside Times,” The Journal of Egyptian Archaeology 23 (1937): 186–9.
89. TADA6.3.
90. TADB8.4, referencing a man who had been enslaved and beaten on the
palms and legs following a court dispute. TADB8.6 also involves corporal
punishment, but it is not entirely clear that the person in question was
enslaved.
91. P. Cair. Zen. III 59369; P. Diosk. 9.
92. P. Tebt. III 2.904.
93. Bauschatz, Law and Enforcement, 240.
94. C. Ptol. Sklav. 1; the law makes a distinction between crimes undertaken
with the knowledge of the owner vs. without the knowledge of the owner,
presumably to prevent collusion between the plaintiff and the slave against
the slave’s owner.
95. Rotman, “Slavery,” 449.
96. Standard as part of the Abnormal Hieratic legal tradition; this oath to Amun
is included in other sales as well as a promise to protect title. B. Menu,
Bernadette. “Les actes de vente En Égypte ancienne, particulièrement sous les
rois Kouchites et Saïtes,” Journal of Egyptian Archaeology 74 (1988): 165–81.
97. van der Linden, “On Dissecting,” 310–13; graphic on p. 311.
98. P. Anastasi V, 19: 2–20: 6; this document has attracted considerable attention
due to the apparent similarity between the escape route favored by the two
fugitives and that followed by the proto-Israelites.
99. E. Morris, The Architecture of Imperialism: Military Bases and the Evolution
of Foreign Policy in Egypt’s New Kingdom (Leiden: Brill, 2005), 240.
100. P. Cair. Zen. 1.59070; P. Cair. Zen. 2.59213; P. Cair. Zen. 4.59613; C. Ptol.
Sklav. 61–85.
101. P. Zen. Pestm. 36; P. Hib. 1 54; C. Ptol. Sklav. 72; C. Ptol. Sklav. 69.
102. P. Paris. 10; P. Cair. Zen. 2.59213.
103. P. Oxy. 3.472.
104. P. Oxy. 12.1422.
105. P. Oxy. 14.1643.
106. Sethe, Urkunden der 18. Dynastie, 1369.4–16; see also J. de Linage, “L’Acte
d’établissement et le contrat de mariage d’un esclave sous Thoutmès III,”
BIFAO 38 (1939): 217–34.
107. A much-discussed text. See A. Gardiner, “Adoption Extraordinary,” Journal
of Egyptian Archaeology 26 (1941): 23–9; E. Cruz-Uribe, “A New Look at
the Adoption Papyrus,” Journal of Egyptian Archaeology 74 (1988): 220–3;
S. Allam, “A New Look at the Adoption Papyrus (Reconsidered),” Journal of
Egyptian Archaeology 76 (1990), 189–91; C. Eyre, “The Adoption Papyrus
in Social Context,” Journal of Egyptian Archaeology 78 (1992), 207–21.
3 ANCIENT EGYPTIAN SLAVERY 65
108. P. Louvre E7832 (Demotic), published in Donker van Heel, “Abnormal
Hieratic,” 177–82; TADB3.9 (Aramaic).
109. M. Cannata. Three Hundred Years of Death: The Egyptian Funerary Industry
in the Ptolemaic Period. Culture and History of the Ancient Near East 110
(Leiden: Brill, 2020), 513; Girgis Mattha, “Rights and Duties of the Eldest
Son According to the Native Egyptian Laws of Succession of the Third
Century BCE,” Bulletin of the Faculty of Arts, Fouad I University, Cairo 12,
no. 2 (1950): 113–8.
110. TADB3.6 (Aramaic).
111. Bie ˙zu´nska-Małowist, L’esclavage, 128–29.
112. Straus, “Papyrological”.
113. E.g., P. Petrie I 3, lines 9–38.
114. Teachings of Amenemope 22.1–4; Moreno García, “The ‘Other’,” 1030.
115. Instructions of Anksheshonqy 8, 11.
116. Moreno García, “The ‘Other’,” 1051; M. Chauveau, “Administration centrale
et autorités locales d’Amasis à Darius,” Méditerranées 24 (2000): 99–109; M.
Campagno, “Patronage and Other Logics of Social Organization in Ancient
Egypt during the IIIrd Millennium BCE,” Journal of Egyptian History 7
(2014), 1–33.
117. O. Patterson, Orlando. Slavery and Social Death (Cambridge: Harvard
University Press, 1982), 13; this does not mean that Patterson’s ideas should
be completely rejected, but rather that they are not the best fit for describing
Egyptian slavery, which would be better served by a synthetic approach
such as the one suggested by D. Lewis, “Orlando Patterson, Property, and
Ancient Slavery: The Definitional Problem Revisited,” in On Human Bondage:
After Slavery and Social Death, eds. John Bodel and Walter Scheidel (Sussex:
Wiley-Blackwell, 2017), 31–54.
Further Readings
Bakir, Abd el-Mohsin. Slavery in Pharaonic Egypt. Cairo: L’Institut français
d’archéologie, 1952.
Bie˙zu´nska-Małowist, Iza. L’esclavage dans l’Égypte gréco-romaine: Première partie:
période ptolémaïque. Wrocław: Zakład Narodowy im. Ossoli´nskich, 1974.
Bakir, Abd el-Mohsin. L’esclavage dans l’Égypte gréco-romaine: Deuxième parite:
période romaine. Wrocław: Zakład Narodowy im. Ossoli´nskich, 1977.
Cruz-Uribe, Eugene. “Slavery in Egypt during the Saite and Persian Periods.” Revue
internationale des droits de l’antiquité 29 (1982): 47–71.
Goelet, Ogden. “Problems of Authority, Compulsion, and Compensation in Ancient
Egyptian Labor Practices.” In Labor in the Ancient World, edited by Piotr Steinkeller
and Michael Hudson, 523–82. Dresden: ISLET, 2015.
Helck, Wolfgang. “Sklaven.” In Lexikon der Ägyptologie, edited by Wolfgang Helck
and Wolfhart Westerndorf. Wiesbaden: Otto Harrassowitz. 1984.
Hofmann, Tobias. Zur sozialen Bedeutung Zweier Begriffe für “Diener”: B3k und H
. m:
Untersucht an Quellen vom Alten Reich bis zur Ramessidenzeit. Basel: Schwabe,
2005.
66 E. KAREV
Loprieno, Antonio. “Slavery and Servitude.” In The UCLA Encyclopedia of Egyptology,
edited by Elizabeth Frood and Willeke Wendrich. Los Angeles: UCLA Press, 2012.
https://escholarship.org/uc/item/8mx2073f.
Menu, Bernadette. Nouvelles recherches sur l’histoire juridique, économique et sociale
de l’ancienne Egypte: III . Cairo: Institut français d’archéologie orientale du Caire,
2004.
Moreno García, Juan Carlos. “Acquisition de serfs durant la Première Période Inter-
médiaire: une étude d’histoire sociale dans l’Égypte du IIIe millénaire.” Revue
d’Égyptologie 51 (2000): 123–39.
Morris, Ellen. “Mitanni Enslaved: Prisoners of War, Pride, and Productivity in a New
Imperial Regime.” SAOC 69 (2014): 361–79.
Quirke, Stephen. “State and Labour in the Middle Kingdom: A Reconsideration of
the Term Chenert.” Revue d‘Égyptologie 39 (1988): 83–106.
Thompson, Dorothy J. “Slavery in the Hellenistic World.” In The Cambridge World
History of Slaver y, Volume I, the Ancient Mediterranean World,editedbyKeith
Bradley and Paul Cartledge, 194–213. Cambridge: Cambridge University Press,
2011.
Vittmann, Günter. “Ägypten (vom Alten Reich bis in die Spätzeit).” In Handwörter-
buch der antiken Sklaverei, edited by Heinz Heinen. Wiesbaden: Fran Steiner Verlag,
2006.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 4
Slavery in Ancient Greece
Kostas Vlassopoulos
Introduction
Slavery in the Greek city-states has held a foundational role in the modern
study of global slavery. Moses Finley, an influential ancient historian, coined
the famous distinction between “societies with slaves,” where slaves are few
and slavery plays a limited economic role, and “slave societies,” where slavery
constitutes a dominant economic, social, and political institution and slaves
comprise a substantial proportion of the population. Finley argued that
while societies with slaves have been ubiquitous in global histor y, there had
been only five slave societies: Greece, Rome, the US South, Brazil, and the
Caribbean. In this approach, Greek city-states had the “honor” of being the
first slave societies in world history.1
The traditional approach to Greek slaver y has been based on two major
tenets. The first is that early Greek communities were originally societies with
slaves, where the dependent labor of the free lower classes was the main
source of elite wealth. But in the course of the archaic period (700–500
BCE), the lower classes gained citizenship rights and could no longer be
directly exploited by Greek elites, who turned to the mass importation of
slaves, leading to the emergence of slave societies. The second tenet is the
assumption that Greek slavery is tantamount to Athenian slavery in the clas-
sical period (500–300 BCE), from where most of our evidence comes; the
K. Vlassopoulos (B)
Department of History and Archaeology, University of Crete, Rethymno, Greece
e-mail: vlasop@uoc.gr
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_4
67
68 K. VLASSOPOULOS
other Greek cities are assumed to have essentially the same slave system as
Athens. This idea was further supported by the assumption that servile groups
like the helots of Sparta and the woikeis of Crete were not “proper” slaves,
but should rather be seen as akin to medieval serfs, since they lived in family
groups as dependent peasants. According to the traditional approach, slavery
in the Greek world consisted of those slave systems that resemble the familiar
image of slavery in the US South; any system that diverged significantly can
be explained away as not being “proper” slavery.2
This chapter is based on the revolutionary implications of recent research
over the last decade, which has seriously challenged the major assumptions
of the traditional approach. A careful reading of the Homeric and Hesiodic
epics, our earliest sources for Greek history, has revealed that the elites in the
Homeric world (800–700 BCE) depended overwhelmingly on slave labor;
accordingly, early Greek communities were already slave societies. We do
not know how far back slavery was a dominant phenomenon in the Greek
world, but it is obvious that the traditional narrative of a transition from
societies with slaves into slave societies in the course of the archaic period
is no longer tenable. We need a new kind of narrative to explain the differ-
ences between the forms of slavery attested in different historical periods. At
the same time, scholars have started to accept that Greek slavery was not a
uniform phenomenon, but consisted of various local slave systems, each of
which had developed its own peculiar features. Spartan helots, for example,
were not serfs, but slaves with peculiar characteristics as a result of the partic-
ular historical development of Spartan society. Slaver y does not have some
trans-historical essence, but is the historical outcome of the interplay between
strategies employing human property for various ends and the wider processes
and contexts within which these strategies take place.3
Instead of the traditional narrative and interpretative framework, we should
rather locate the history of Greek slave systems within a number of processes.
We will focus on four distinct but interrelated historical processes. Prime of
place goes to the process of growing connectivity that from the archaic
period onward came to interlink various areas of the Mediterranean and the
Black Sea; this process was partly based on decentralized networks moving
goods, people, ideas and technologies, and partly on attempts by states and
potentates to canalize connectivity for their own ends. Greek slave systems
cannot be understood outside this quantum leap in the connectivity of
the Mediterranean and the Black Sea during the first millennium BCE; in
the same way that early modern slavery is incomprehensible outside the
emergence of the Atlantic world that interlinked European, African, Native
American and Colonial American societies, economies and cultures, Greek
slave systems were intimately related to other Mediterranean and Black Sea
slave systems. Increasing connectivity set the stage for drastic changes in
Greek material culture; it made possible the utilization of Mediterranean
micro-ecological diversity and fragmentation through large-scale processes of
exchange and redistribution. The resulting specialization, production for the
4 SLAVERY IN ANCIENT GREECE 69
market, dependence on exchange and surplus accumulation went hand in hand
with the emergence of the first consumer societies, in which substantial social
strata desired and consumed goods from various areas of the Mediterranean
and the Black Sea. Slavery was deeply inscribed in this process, not only in
terms of the slave trade, but also in terms of producing a major part of
these various goods and creating the surpluses that allowed the emergence
of consumer societies.
At the same time, processes of community formation and claim-making
transformed the socio-political settings of Greek city-states. This process
shaped the institutions of Greek city-states and the meaning of citizenship
and changed the ways in which Greek communities formulated the distinc-
tion between insiders and outsiders. Freedom was no longer the status of not
being property, and started to acquire additional features that tended to turn
into a total and unalterable status; freeborn people could no longer lose their
status within their community and their status protected them from dishonor
and physical punishment. The formalization of free and slave status created
major disadvantages for slaves, but at the same time opened new institutional
settings that slaves could potentially take advantage of. Political communities
could also superimpose their own priorities on masters and slaves, limiting
what could be done to slaves or what slaves could do. Finally, geopolitical
processes redefined how violence and ideology affected enslavement and liber-
ation across the Mediterranean and the Black Sea. Most slaves were produced
through warfare, raiding and international trade. This means that slavery was
directly inscribed in the changing history of the forms of warfare, predation,
exchange, state-building and empire-building that linked together commu-
nities into wider systems of international relations. The emergence of large
states and empires in various parts of the Mediterranean and the Black Sea, as
well as their occasional collapse, enhanced the scale and stakes of warfare and
the extent of slave-making, creating large and interconnected slaving zones;
at the same time, the peculiar form of the Greek geopolitical system had
also important implications for the emergence of various forms of no-slaving
zones.4
It might be worth offering a general typology of Greek slave systems at this
point. The first group of Greek slave systems comprised societies like Sparta
and Crete, whose citizenries consisted of leisured gentlemen exploiting slave
labor. These systems had limited engagement with Mediterranean connectivity,
being agricultural societies geared toward local production and consumption.
Slaves constituted the majority population group in these societies, and their
replenishment was based on natural reproduction. A second group of slave
systems consisted of relatively wealthy societies with deep engagement with
Mediterranean connectivity, highly diversified economies and often a predilec-
tion for marketable crops, like oil and wine. Apart from Athens, which was in
addition an imperial center, this group included primarily island and coastal
communities like Chios, Corfu and Aegina. These slave systems depended
70 K. VLASSOPOULOS
largely on trade for the replenishment of their slave populations. The histo-
rian Thucydides believed that Chios had more slaves than any other Greek
city, even Sparta, presumably as a proportion of the population rather than
an absolute number5; it is possible therefore that some Greek cities like
Chios had slave populations that approached those in some early modern
Caribbean colonies. In the case of Athens, most scholars would accept a
guesstimate between 20 and 50 percent of the population. The third group,
which comprised the majority of Greek societies, was similar to the second, but
with far fewer resources and extent of connectivity; we should therefore expect
a social structure similar to the second group, but with fewer slaves given
their limited wealth; slave populations of 20 percent or less should probably
be expected.6
Entry
Cross-culturally, we can distinguish between four major forms of entry into
slavery: (a) internal enslavement within a community; (b) violent enslavement
through state warfare or piracy; (c) the slave trade and (d) the inheritance of
slave status through the natural reproduction of slave populations. As we shall
see, we can find both common patterns across the Greek world, as well as
major differences.
One of the peculiar features of Greek slave systems is the limited role of
internal forms of enslavement. Solon’s reforms in early sixth-century BCE
Athens prohibited debt slavery; Athenians could no longer be enslaved for
debt within their community. We have no concrete evidence about most other
Greek communities, but we get the impression that enslavement for debt was
marginal, if not equally prohibited. On the other hand, the existence of debt
bondage is attested; free people had to work for their lenders in order to repay
their debts in conditions that were often akin to slavery, although they retained
their free status while in debt bondage. Largely invisible is the right of fathers
to sell their children, as was the case in many other ancient societies. Finally,
penal enslavement is unattested for citizens, although we know that in Athens
it was a possible punishment for free foreign residents who had not paid their
special taxes, or for people who attempted to usurp the right to citizenship.
Internal forms of enslavement were not unknown in Greek societies, but as a
result of community protection of the free status of citizens they were marginal
phenomena in the Greek world.7
Greek city-states had very strong no-slaving zones, but these zones
concerned only their own citizens; the rest of the world, including citizens of
other Greek city-states, were considered potentially enslaveable. The slaving
zones of Greek city-states were therefore enormous; not only could Greek
city-states potentially enslave their neighbors, but, due to the connectivity
expansion we examined above, they could also receive slaves from areas like
Asia Minor, the Levant, Thrace and the Black Sea. Given the limited role of
4 SLAVERY IN ANCIENT GREECE 71
internal for m s of enslavement, it was violence and trade that constituted the
main forms of entry into slaver y in the Greek world.
Generally speaking, in all periods of Greek history it was taken for granted
that violence was a legitimate means of generating slaves; but this general fact
was often hedged in by certain important qualifications. The first important
qualification is the distinction between oligopolistic and monopolistic states in
terms of their recourse to violence; oligopolistic states allowed their citizens
to employ violence against for eigners for their private gain, and required their
citizens to contribute their military equipment and ships only on those occa-
sions in which the state was fighting against another state; monopolistic states
restricted the use of violence to state purposes only, prohibiting or discour-
aging their citizens from using violence for private gain, and often maintaining
substantial state arsenals.8 Until the late archaic period, all Greek states were
oligopolistic; accordingly, elites and commoners habitually engaged in piracy
and other forms of violence that produced movable wealth and captives. The
famous poem of Hybrias presents a Cretan master who attributes his wealth,
leisure and cowering slaves to his military prowess.9
From the late archaic period onward, though, many Greek states made
the transition from oligopolistic to monopolistic forms: although slaves were
still produced by state warfare, their citizens could no longer engage in acts
of private enslavement through violence. In some parts of the Greek world,
like Crete and Aetolia, states remained oligopolistic down to the end of
the Hellenistic period (323–31 BCE); piracy remained an important form of
enslavement in these areas, and, depending on circumstances, it could occa-
sionally contribute substantial numbers of slaves. But in the main areas of
the Greek world (the Peloponnese, central Greece, the Aegean islands and
coastal Asia Minor), the emergence of monopolistic states made private violent
enslavement by elites and commoners a phenomenon of the past. 10
What was the role of state violence in producing slaves in these areas? The
second important qualification is the limited role of transcultural wars in Greek
history, a major aspect of global enslavement. There clearly existed military
conflicts between (some) Greeks and (some) non-Greeks, and the Persian Wars
(490–478 BCE) are the most famous example of them; but, by and large,
most episodes of warfare in the Greek world involved Greeks fighting against
other Greeks. Given this fact, the large numbers of non-Greek slaves in Greek
city-states cannot be attributed to warfare between Greeks and non-Greeks;
not because Greeks would not have enslaved their non-Greek enemies in
large numbers, but because they rarely had the opportunity to do so. Accord-
ingly, enslavement by war in the Gr eek world is effectively tantamount to the
enslavement of Greeks by other Greeks.
As far as the archaic period is concerned, the enslavement of defeated
enemies appears to be a relatively common phenomenon. According to much
later accounts, the helots were populations conquered and enslaved by Sparta,
while similar narratives exist for the conquest and enslavement of the Penestai
of Thessaly or the Mariandynoi in Heraclea Pontica.11 Whether these accounts
72 K. VLASSOPOULOS
are historically reliable is debated by modern historians; but once we reach
the classical period, for which we are much better informed, we encounter
a paradox. On the one hand, the Greeks considered perfectly legitimate the
enslavement of their Greek opponents: among many actual examples, we can
mention the enslavement of the Melians by the Athenians and the enslave-
ment of the Thebans by Alexander the Great. On the other hand, given the
ubiquity of warfare in the Greek world, the enslavement of defeated oppo-
nents appears as a relatively rare outcome of the fate of captives. Furthermore,
Greek texts give the impression that slaves in Greek cities were almost exclu-
sively non-Gr eek, despite the extant evidence for the enslavement of Greeks
by other Greeks.
How should we explain this paradox? Greek city-states could deal with
defeated opponents and captives in a variety of ways: they could exchange
prisoners, release the free captives and keep those who were already slaves,
use the captives as bargain chips for a wider settlement, kill all captives, kill
the male adults and enslave the rest, or enslave all captives. Enslavement of
captives was a form of conspicuous destruction: it was undertaken as a public
statement, whether as revenge for heinous crimes, to discourage future resis-
tance or to exterminate an enemy community. Accordingly, the enslavement
of Greek opponents was only undertaken in particular circumstances and for
specific purposes, rather than as a default policy. The evolution of Greek
interstate relations was also an important factor: the creation of hegemonic
alliances and the development of means of incorporating defeated communi-
ties in state structures offered alternatives to enslavement for victorious states.
Finally, Greek city-states developed robust ransoming mechanisms, by signing
multilateral treaties or encouraging individuals to ransom fellow citizens or
friendly foreigners.12
As a result of all these factors, while the enslavement of Greeks in war
remained a constant factor of entry into slavery, it was rather trade that consti-
tuted the main source of slaves in most areas of the Greek world. The processes
of connectivity that we mentioned above ensured that Greek communities
could draw slaves from all areas of the eastern Mediterranean and the Black
Sea. This constituted an enormous expansion of the reach of slaving zones
in the ancient world. Geopolitical changes in various parts of the Mediter-
ranean and the Black Sea increased significantly the supply of slaves. The
warfare that accompanied the creation of large states and empires, like the
Odrysian kingdom in Thrace or the kingdom of the Royal Scythians in the
Black Sea, provided major opportunities for slave-making; the same applied to
the crisis or collapse of such states. The Greek colonies in the Western Mediter-
ranean, Thrace and the Black Sea and their commercial networks provided
a major, easy and profitable outlet for the thousands of slaves generated by
these conflicts. A single day of campaigning by Seuthes, a Thracian potentate
around 400 BCE who was trying to extend his authority, produced a thou-
sand captives that were quickly disposed of in the nearby Greek colony of
Perinthos.13
4 SLAVERY IN ANCIENT GREECE 73
Furthermore, the maritime character of Mediterranean connectivity offered
significant advantages in relation to the land-based connectivity of most areas
that constituted the earlier civilizations of the Near East. Slaves from the Black
Sea or Thrace could be sold in the port of Athens within a few days of their
capture. The short maritime distances involved meant that there was no need
for specialized slave ships or slave routes; slaves moved in the same ways and
routes as all other commodities.14 As a result, the price of slaves in classical
Athens appears to be far lower in comparison with any other ancient society for
which sufficient evidence exists (Rome, Mesopotamia); on average, Athenian
slaves costed the equivalent of the annual wages of a skilled craftsman. Given
these low prices, it is not surprising that in places like Athens most slaves
appear to be first-generation non-Greeks.15
But what was the role of slave natural reproduction in terms of replenishing
slave populations? For certain Greek communities, like Sparta and Crete, the
answer is simple: irrespective of the origins of their slave populations, natural
reproduction constituted the overwhelming source of their replenishment. As
we shall see in the next section, the leisured lifestyle of Spartan and Cretan
citizens required large numbers of slaves; but the limited engagement of these
societies with Mediterranean connectivity meant that they lacked easy access
to networks of supply and the capital required for the constant replenishment
through trade: slave reproduction was effectively their only feasible choice.
This had important effects on these slave systems: the systemic need for repro-
duction meant that slave families were relatively stable and that slaves were
native inhabitants forming their own communities. This is a major reason why
Spartan helots and Cretan woikeis look more similar to medieval serfs than to
the standard image of first-generation foreign slaves. The woikeis of Cretan
Gortyn had two additional remarkable features. Their slave families had legal
consequences: slave children did not belong to the master of the slave mother
(the widespread principle of partus ventrem sequitur ), but to the master of her
senior male relative (father or brother) or of her slave husband. Furthermore,
Gortyn allowed mixed marriages between free and slave, something unparal-
leled in other Greek societies that we know of. The status of the children of
these marriages depended on the residence of the couple: if the family resided
with the slave husband and was thus under the authority of his master, the
children became slaves, while if it resided with the free mother the children
were free.16
For the rest of the Greek world, the significance of slave reproduction is
difficult to gauge. In the case of Athens, which is better known, the relatively
small numbers of slaves in individual households would have made it difficult
to create slave families within the master’s household, a common phenomenon
in other societies. Slaves would have often needed to find partners from other
households, something that would have created substantial problems in terms
of timetables and living arrangements. On the other hand, cities like Athens
had substantial populations of slaves who lived and worked on their own; they
were more likely to have formed families, and this is often corroborated by the
74 K. VLASSOPOULOS
sources. Accordingly, while for many Greek societies like Athens the role of
reproduction was secondary compared to trade, the existence of slave families
and the role of second-generation slaves was not insignificant.17
The Experiences of Slaves
Slave experience was shaped by the slaving strategies adopted by their masters.
In order to understand these various slaving strategies, it is important to
examine the nature and size of the master class. Greek societies were shaped
by a fundamental distinction between rich and poor: the rich consisted of all
those who were wealthy enough to live without working, while the category of
the poor was highly diverse, from the destitute and those barely able to make
ends meet to those who lived comfortably, but still had to work alongside their
laborers. Leisure was a quintessential aspect of the Greek rich; accordingly, we
can take for granted that in most Greek societies the rich were able to afford
their lifestyle because they possessed a sufficient number of slaves. There are
two main reasons for this: the general absence of institutionalized relations of
dependence among the free population and the unwillingness of free people
to work for somebody else on a long-term basis, as this was considered akin
to slavery. Although free wage laborers were a substantial proportion of some
Greek cities like Athens, they would mostly work on short-term contracts for
successive employers. Accordingly, long-term workers in households, estates
or workshops were almost by definition slaves in the Greek world.
In the case of Athens, where we have s ufficient evidence, it appears that rich
people possessed on average about 10 slaves. We hear of exceptional cases like
the Athenian politician Nicias, who owned a thousand slaves leased to mine
operators, and there are examples of owners of workshops that employed 30,
50 or 60 slaves. But it is fairly evident that there was barely any Greek equiv-
alent to the Roman imperial magnates, who possessed urban households with
hundreds of slaves, let alone slaves in their rural estates. The orator Demos-
thenes accused his wealthy opponent Meidias of arrogance for appearing in
public spaces accompanied by three slaves18; this gives a good impression of
the relative size of Athenian slaveholdings. We can conclude that while a few
very rich people might own tens (and occasionally hundreds) of slaves, most
rich people had much smaller slaveholdings.
While slave ownership among the rich is beyond doubt, it is more diffi-
cult to assess its extent among the rest of the population. During the archaic
period, Sparta and Crete extended the lifestyle of the leisured gentleman to
the whole citizen body; as a result, every citizen at Sparta and most citizens
in Crete were slave-owners who devoted their lives to warfare, politics and
leisure pursuits, because their slaves performed all necessary labor tasks. Sparta
and Crete were exceptional; in most other Greek societies, the overwhelming
majority of the citizen population had to work for a living. Nevertheless, it is
fairly evident that in a rich and powerful society like Athens slave ownership
extended to a significant section of working citizens, perhaps one-third of all
4 SLAVERY IN ANCIENT GREECE 75
citizens. Most of these non-elite citizens would own only one or a few slaves.
Whether this also applies to other Greek communities, i t is impossible to say.
Accordingly, while huge slaveholdings, like those attested in Rome or the New
World, were quite rare in the Greek world, at the same time slave ownership
extended far beyond the elite. The particularly cheap prices of slaves that we
mentioned above are probably a major reason for this phenomenon.19
Households were the key units of social and economic practices in the
ancient Greek world. Most Greek slaves were members of relatively s mall
households, consisting of the nuclear family of the master, a few younger
or older relatives and a couple of slaves. Accordingly, Greek slavery was
deeply shaped by the economic and social strategies of households in terms
of acquiring and maintaining property, ensuring a livelihood, producing heirs
and transmitting property. A crucial parameter in this respect was the extent to
which household heads had free rein to pursue their aims, or were limited by
countervailing tendencies and factors. Were household heads allowed to marry,
recognize as heirs and bequeath their property to whomever they wanted, or
were there rules that imposed, prohibited or prioritized particular courses of
action? While most Greek slaves operated within households, an important
development of Greek history is the emergence of a new context, in which
the economic strategies usually pursued within households were expanded
and transformed into large-scale operations employing hundreds or even thou-
sands. By the classical period we see operations like the Athenian workshops
employing tens of slaves, while the Athenian mining operations in Laureion
used thousands of slaves.20
In what ways were slaves employed? Earlier approaches prioritized the use
of slaves in production, and in particular agriculture, as the key criterion of
the importance of slavery in a society; other uses, like household service, were
considered to be of secondary importance. But this is unnecessarily restrictive.
Without modern technological advances that provide fresh water, electricity
for lighting and cooking, washing appliances for clothes and dishes and dispos-
able nappies, an enormous quantity of labor was required to perform essential
everyday activities, like cutting wood, drawing water and making bread; the
various forms of the sexual exploitation of slaves were crucial parameters of
the gender and sexual structures of ancient Greek societies; the employment
of slaves by Greek states defined their character and activities. At the same
time, the economic role of slavery did not take a single form, but a range of
diverse forms with very different implications. We should therefore pay equal
attention to all the diverse ways in which slaves were employed in ancient
Greek societies; the variety of these uses enables us to escape from the struc-
turalist assumptions that have dominated earlier approaches. Many of these
uses were compatible with each other and even complementary; but they could
also be contradictory and even incompatible. They could therefore generate
important stress points and areas of conflict, while also providing slaves with
opportunities that would otherwise have been impossible. To this end, we can
76 K. VLASSOPOULOS
distinguish between different slaving strategies and examine the full range of
strategies that co-existed within a single society.
We can distinguish between five major slaving strategies in the Greek world.
A first set of strategies focused on the extraction of labor; within this set, we
can further distinguish two subsets: the use of slave labor for maintenance,by
employing slaves for the drudgery required for the everyday maintenance of
households (cooks, cleaners, personal attendants, nannies), and the employ-
ment of slave labor for the production of wealth on rural estates, in workshops
and mines. As we have seen, Greek households, even those of the ver y rich,
usually employed only a few slaves. On the other hand, the Athenian mines
employed thousands of slaves at their peak, while we know of workshops that
employed tens of slaves. Our sources record nothing equivalent for Greek agri-
culture, but this seems to reflect the fact that even rich Greeks possessed a
series of dispersed landholdings, rather than large unified estates. We should
therefore expect that such kinds of landholdings required small group of slaves
for their cultivation.
In this set of strategies, slaves usually worked in labor processes under the
direct control of their masters. Given the unwillingness of free Greeks to be
at the constant beck and call of a long-term employer, the practically exclusive
use of slaves for household maintenance played a crucial ideological role in
Greek societies, by supporting the illusion of citizen equality that was crucial
in particular for democracies like Athens.21 Given the defining role of leisure
for Greek social structures, strategies of labor extraction performed the crucial
task of absolving masters and mistresses from the need to work to produce
wealth and maintain households. Slavery was also crucial for gender structures:
since respectable Greek women were supposed to avoid public spaces and stay
indoors, the possession of slaves who could accompany their mistresses allowed
mistresses to do things that otherwise would have been unacceptable.
Another set of slaving strategies aimed at revenue extraction rather than
labor; in this set masters withdrew from the labor process and used slaves
like other possessions and investments which brought revenue, such as real
property or loans. We can again distinguish between two subsets. In the first
subset masters hired their slaves to other people, who could not afford to buy
their own slaves, or had short-term or temporary labor needs that made hiring
preferable. In the second subset, masters allowed their slaves to work on their
own as cultivators, artisans or traders, on condition that they surrendered part
of their earnings. The strategies of revenue extraction were of crucial impor-
tance for all forms of Greek slave systems. Spartan citizens were obliged to live
in Sparta, but the most fertile part of Spartan territory was Messenia, separated
from Sparta by the Taygetos mountain range and difficult to access. As a result,
Spartan masters were absentee landowners; this allowed their Messenian helots
to effectively operate as dependent peasants, who lived in their own villages,
organized the labor process on their own and surrendered only part of the
crops to their masters.22 In the highly urbanized societies of the second group
of slave systems, independent slaves played major roles as traders and artisans;
4 SLAVERY IN ANCIENT GREECE 77
slave bankers were among the richest inhabitants of Athens. These slaves were
relatively independent and were often indistinguishable from the free lower
classes; they could often use their hard-won savings in order to enhance their
living and ultimately buy their freedom.23
A third slave strategy concerns gratification: the use of slaves to provide
sensory pleasure in all its various forms. Gratification slaving involved musi-
cians, singers, dancers, barbers, masseurs, hairdressers and cooks; but the most
important form of gratification concerned sex. The sexual economy of Greek
societies was organized on the basis of distinct gender roles. Respectable
women could have legitimate access to sex exclusively through marriage; on
the contrary, men could legitimately have sex outside marriage, as long as they
refrained from having sex with respectable women. But women without honor
were fully usable, and slaves constituted the overwhelming majority of women
without honor. Furthermore, while girls were commonly married as soon as
they reached puberty, men in Greek societies normally deferred marriage until
their late twenties or thirties, when they would have received their inheritance
and could support their families. As a result, men faced a window of oppor-
tunity between puberty in the late teens and marriage in the early thirties in
which sexual access took place outside wedlock and was provided by women
without honor.
The casual sexual exploitation of slaves by their masters was a ubiquitous
feature of Greek societies; at the same time, the high levels of urbanization in
the Greek world created the necessary population density for the emergence
of brothels; spurred by the widespread availability of slaves and the gender
prescriptions of Greek sexual economies, prostitution increased exponentially.
The sexual exploitation of slaves fundamentally shaped how gender and sex
operated in the Greek world. At the same time, the use of slaves for gratifica-
tion also led to the creation of r elatively stable relations between masters and
female slaves (relations between mistresses and male slaves were considered
beyond the pale). Concubinage was a common phenomenon. In the Homeric
world, masters could recognize the children of their female slaves; this seems
to have remained the case in certain Greek societies. But in most Greek
societies the intervention of the political community curtailed or prohibited
getting heirs through slaves; in Athens, masters could not recognize their slave
progeny as legitimate heirs even if they wanted to.24
The fourth set of strategies is the most paradoxical, for it employs slaves
for expertise, trust and authority. Slaves were often employed as commer-
cial agents and managers; given the limited scope of Greek legal systems for
the delegation of authority among free people, slavery allowed masters to have
full control over those that run their business interests. Slaving for expertise
was also important for processes of knowledge transmission. In the absence of
institutionalized systems of intergenerational training and knowledge transmis-
sion, like those of the medieval guilds, buying slaves and training them was a
particularly efficient way of creating, maintaining and controlling a specialized
workforce. While free trainees could move or become antagonists, trainers of
78 K. VLASSOPOULOS
slave exper ts could control them in a much more guaranteed way.25 Equally
remarkable is how Greek states employed public slaves as clerks, bureaucrats
and even policemen, like the 300 Scythian archers employed by Athens. The
image of public slaves imposing order on citizens was something inconceivable
in other slave systems, such as the US South. Public slaves offer a valuable
window into Greek politics. In many societies the various groups of the state
sector (bureaucrats, judges, policemen, the military) develop their own inter-
ests and are able to prioritize them over those of the citizenry as a whole;
the use of public slaves allowed Greek states to eschew the development of
a powerful bureaucracy, while also allowing states with high levels of citizen
participation and magistrate turnover to maintain institutional know-how and
continuity.26
The final slaving strategy is that of prestige creation for their masters. This
was certainly not unknown in the Greek world: a satirist describes how a man
of petty ambition in Athens would buy an exotic Ethiopian slave to flaunt
his putative wealth.27 But in contrast to the huge slave retinues that were
the necessary accompaniment of elite Romans in public, the role of slaving
for prestige creation in the Greek world was highly circumscribed. Greek
city-states carefully orchestrated how elite citizens could gain honor and pres-
tige, primarily through acts of public service and munificence; slaving offered
private wealth more influence in gaining honor than Greek communities were
willing to accept. This is a good example of how slaving strategies allow us to
explore the divergent anatomies of different ancient and modern societies.
As the above summary of Greek slaving strategies indicates, the experiences
of Greek slaves could differ significantly.28 There was a world of difference
between the experiences of mining slaves, slave prostitutes, slave bankers, slave
policemen and slave artisans who worked on their own. At the same time, slave
experiences were also shaped by wider features of Greek societies, as well as by
slave agency. Given that Greek slaves did not belong to a single racial group,
it was usually impossible to tell apart slave and free persons by visual means,
though there existed certain bodily characteristics that were more common
among certain groups of non-Greeks, such as red hair, and could therefore
make certain non-Greek slaves easily identifiable. At the same time, Greek
cities, in particular from the later classical period onward, contained signifi-
cant numbers of free immigrants from various parts of the Mediterranean and
the Black Sea. Accordingly, e.g., Athenians did not encounter Thracians solely
as slaves; the Thracian presence in Athens included free immigrants, freedper-
sons and slaves. The co-existence of free and enslaved members of the same
ethnic and cultural groups had important implications for the shaping of slave
identities and ethnicities in the Greek world.29
Another factor that often complicated things was the absence of labor divi-
sions based on status; with the exception of household service and mining,
which were exclusively performed by slaves, in all other professions and tasks
free and slave laborers worked side by side. It was thus difficult to tell the
status of an individual solely on the basis of profession or living standards,
4 SLAVERY IN ANCIENT GREECE 79
as many independent slave artisans or traders lived in conditions identical to
those of their free counterparts.30 As one posh ancient author commented,
in Athens one should avoid hitting a person because he looked poor, as it
was impossible to know whether he was slave or free, and the citizen poor had
power because of the democratic political system.31 Finally, from the later clas-
sical period we see the explosion of associations based on common adherence
to a cult, ethnicity or profession. While most members of these associations
were free, they could also include freedpersons and even slaves. These mixed
associations created networks of solidarity and support that could often play a
crucial role in slave strategies for improving their lot.32 As a result of all these
factors, passing as free was an important strategy adopted by slaves in the
Greek world. In many cases, this was not an attempt to flee from the master,
but merely an effort to avoid prejudice and mistreatment from third parties.
It is telling in this respect that slaves with “white collar” jobs tended to have
standard Greek names shared by the citizens and therefore indistinguishable,
while slaves with “blue collar” jobs often bore foreign or ethnic names that
made them immediately distinguishable.33
Exits
There were diverse forms of exits from slavery across the Greek world; they
include manumission, either by individual masters for their own reasons or
through state intervention for public reasons; and, finally, individual/collective
flight or rebellion. Greek communities employed different forms of manumis-
sion: in some communities, manumissions were private and informal, requiring
little more than a unilateral verbal utterance by the master in the presence of
witnesses, or a written will. In other communities we find the phenomenon
called “sacral manumission”: the master formally sold the slave to a deity,
which paid the master with money provided by the slave on condition that
the slave was then manumitted. The origins and meaning of this practice are
debated, but it is clear that the deity acted as the middleman that ensured
the validity of the contract, as slaves strictly speaking could not contract with
their own masters. We also encounter the phenomenon of masters conse-
crating slaves to deities; such slaves were strictly speaking not free, but the
absence of a concrete human master meant that in practice consecrated slaves
lived as freedpersons, merely offering their services to a local temple on festival
days. While all these forms of Greek manumission were unilateral acts of the
masters, we also encounter Greek communities in which communal assent to
private manumissions was required; in the most extr eme case we know of, that
of Sparta, manumission by masters for private reasons was prohibited, and only
the state could manumit helots for public purposes.34
How did slaves manage to gain their freedom? In certain cases, masters
decided to manumit their slaves free of charge or for a nominal price. This
was usually the result of strong interpersonal relations between masters and
slaves: manumission was a reward for long and faithful service, or because
80 K. VLASSOPOULOS
masters wanted to liberate their slave concubines and their progeny. But in
most cases slaves paid for their freedom, often more dearly than simply in
monetary terms; one of the most heartbreaking aspects of Greek manumission
is the obligation of female slaves to give birth and surrender to their masters
a set number of children who would take their place in slavery. The ability of
slaves to procure money for their manumission highlights two things: on the
one hand, the significance of slaving strategies for revenue, trust and expertise,
which created independent slaves who could keep part of their earnings; on
the other hand, the significance of networks and communities that involved
slaves, freedpersons and free people, who allowed slaves to borrow or pool
resources in order to gain their freedom.
What was the status of freedpersons in the Greek world? In the world
depicted by Homer, incorporation of freedpersons in the free community
appears relatively straightforward: Odysseus promises to reward his faithful
slaves with land and wives, presumably indicating their full incorporation in
the local community.35 But archaic political processes started to change signifi-
cantly the nature of community membership in the Greek city-states; gradually,
citizenship became a highly codified status defined by the political commu-
nity, bestowed through specific rituals and often requiring descent from both
citizen parents. As a result, from the later ar chaic period freedpersons in most
of the Greek world acquired the status of resident foreigners, if they chose to
stay in the s ame place where they had lived as slaves. Like freeborn resident
foreigners, freedpersons could occasionally gain citizenship for major benefac-
tions to the city-state, but that was a very rare phenomenon. It is possible that
in some communities of central and northern Greece in the Hellenistic period
freedpersons could gain political rights alongside their manumission, though
details are unclear.36
A peculiar aspect of the status of Greek freedpersons is the practice of para-
mone. In a significant number of cases, manumitted slaves were obliged to
remain (paramenein) with their former masters for a specified period of time
or until the death of the latter; many manumission contracts explicitly state
that former masters had the right to punish freedpersons as slaves, or even
annul their manumission if they considered them ungrateful or insubordinate,
while other contracts prohibit servile punishment and institute panels of arbi-
trators to settle disputes between former masters and people in paramone.
Were people in paramone slave or free? The best answer seems to be that
they were free in r egard to everybody else, while their status vis-á-vis their
former masters depended on the terms of their manumission, ranging from
continuous servile subordination to free dependence.
Another important means of exit from slavery concerned the role of the
political community. The Greek world was an anarchic geopolitical environ-
ment consisting of hundreds of small, medium and large city-states, in which
warfare and civil war were ubiquitous phenomena. States could resort to public
manumissions in order to enhance their manpower or mitigate a potentially
lethal crisis; in 406 BCE the Athenians manumitted and even enfranchised
4 SLAVERY IN ANCIENT GREECE 81
thousands of slaves in order to man a new fleet in a desperate but successful
attempt to save their fleet from Spartan blockade. In 86–85 BCE Ephesos in
Asia Minor, caught in the middle of the war between King Mithridates and
Rome, resorted to manumitting public and private slaves in order to enhance
its army and avoid the risk of city betrayal under siege. In wars between
different communities or civil conflicts within the same community one or
both sides could attempt to strengthen their side and/or destabilize the oppo-
site side by inviting slaves to flee or rebel, promising freedom as a reward. A
characteristic example is that of the civil war in Corfu in 427 BCE, when both
democrats and oligarchs offered the rural slaves freedom for choosing their
side; the majority of the slaves took the side of the democrats, a pattern that
was repeated on other occasions as well.37
The co-existence of hundreds of neighboring city-states and the ubiquity of
war among them made flight relatively easy, as a different political authority
existed just a few miles down the road. There was obviously no guarantee
that slaves who flew to a different city would remain free there, rather than
being enslaved by a different master, but it was often worth trying; in order to
counter that threat, Greek cities often signed bilateral treaties for the mutual
return of fugitives. The existence of communities of independent slaves and
freedpersons in most urban communities could provide a safe environment for
fugitives. Maroon communities were a common phenomenon in many areas
of the New World, but we are ill informed about Greek examples. The most
famous case is that of Drimacus, who led a maroon community in Chios,
ultimately forcing the Chians to reach a modus vivendi, in which Drimacus
was recognized as leader of a maroon quasi-state.38
Slave revolts appear to be a rare phenomenon in the Greek world. Among
the few examples, the best known is the revolt of the Messenian helots in
the 460s BCE, with the rebels ultimately obtaining a truce allowing them to
go into exile as free persons. The helots of Messenia were ultimately liber-
ated in 368 BCE, when Thebes defeated Sparta and created an independent
Messenian state. Helot revolts profited from the peculiar geopolitical setting
of Sparta; none of its neighbors had a similar form of slave system, and they
were willing therefore to encourage helot revolt without fearing the Spartans
doing the same.
The Peloponnesian War (431–404 BCE) offers a more common example
of how slaves gained their freedom in mass numbers. In the early phase of the
war, the Athenians occupied a fort in Spartan territory and encouraged the
helots to flee; many did, and the Spartan fear of a wider revolt led them to
agree to a humiliating treaty that ended the first part of the war. But the Spar-
tans also decided to use thousands of helots as soldiers by promising them
freedom, and these helot soldiers scored some of the most decisive Spartan
victories. When the Spartans later in the war also occupied a fort in Athenian
territory, thousands of Athenian slaves flew to them, although most of them
did not gain their freedom in this way, but merely changed masters. This mass
flight seriously debilitated Athenian fortunes; at the same time, thousands of
82 K. VLASSOPOULOS
Athenian slaves gained freedom and even citizenship by joining the Athenian
navy, while many others joined the movement that restored Athenian democ-
racy in the aftermath of the war. Athenian slaves and Spartan helots did not act
in defense of collective slave interests; different groups acted in different ways,
according to circumstances and choices. But slave agency played a crucial role
in the outcome of the war, and thousands of slaves gained their freedom in
one way or another.
Slavery was a defining phenomenon of the history of Greek city-states.
The diverse slaving strategies that co-existed in most Greek societies meant
that slavery could be used for a variety of purposes, partly complementary,
and partly contradictory. While Greek elites depended on slaves, substantial
middling strata of Greek societies were also slaveholders; Greek civic institu-
tions and the material culture of Greek consumer societies were fundamentally
shaped by the ubiquity of slavery; the connectivity and market expansion of
the first millennium BCE had slavery in its ver y core. At the same time,
the processes in which slavery was inscribed could lead to ver y divergent
outcomes among Greek slave systems. Studying the history of these diverse
systems is a potent means of understanding the historicity of slavery as a global
phenomenon.
Notes
1. Moses I. Finley, Ancient Slaver y and Modern Ideology (London: Chatto and
Windus, 1980).
2. This is still the dominant approach in Keith Bradley and Paul Cartledge, eds.
The Cambridge World History of Slavery 1: The Ancient Mediterranean World
(Cambridge: Cambridge University Press, 2011).
3. David M. Lewis, Greek Slave Systems in Their Eastern Mediterranean Context,
c. 800–146 BC (Oxford: Oxford University Press, 2018); Kostas Vlassopoulos,
Historicising Ancient Slavery (Edinburgh: Edinburgh University Press, 2021).
4. For the concept of slaving and no-slaving zones, see Jeffrey Fynn-Paul,
“Empire, Monotheism and Slavery in the Greater Mediterranean Region from
Antiquity to the Early Modern Era,” Past & Present 205, no. 1 (2009): 3–40.
5. Thucydides, History, 8.40.2.
6. Nemenja Vujˇci´c, “A Numbers Game: The Size of the Slave Population in
Classical Athens,” Zbornik: Jour n al of Classical Studies 23 (2021): 87–111.
7. Edward M. Harris, “Did Solon Abolish Debt-bondage?” Classical Quarterly
52, no. 2 (2002): 415–30.
8. Vincent Gabrielsen, “Warfare, Statehood and Piracy in the Greek World,” in
Seeraub im Mittelmeerraum: Piraterie, Korsarentum und maritime Gewalt von
der Antike bis zur Neuzeit, ed. N. Jaspert and S. Colditz (Paderborn: Ferdinand
Schöningh/Wilhelm Fink Verlag, 2013), 131–53.
9. Poetae Melici Graeci no. 909.
10. David M. Lewis, “Piracy and Slave Trading in Action in Classical and Hellenistic
Greece,” Mare Nostrum 10, no. 2 (2019): 79–108.
4 SLAVERY IN ANCIENT GREECE 83
11. Hans van Wees, “Conquerors and Serfs: Wars of Conquest and Forced Labor in
Archaic Greece,” in Helots and Their Masters in Laconia and Messenia: Histo-
ries, Ideologies, Structures, eds. N. Luraghi and S. Alcock (Cambridge, MA:
Center for Hellenic Studies, 2003), 33–80.
12. Pierre Ducrey, Le traitement des prisonniers de guerre dans la Grèce antique,
des origines à la conquête romaine (Paris: De Boccard, 1968); Anne Bielman,
Retour à la liberté: Libération et sauvetage des prisonniers en Grèce ancienne.
Recueil d’inscriptions honorant des sauveteurs et analyse critique (Athens: Ecole
Française d’Athènes, 1994).
13. Xenophon, Anabasis , 7.3–4.
14. David M. Lewis, “The Market for Slaves in the Fifth-and Fourth-Centur y
Aegean,” in The Ancient Greek Economy: Markets, Households and City-States ,
eds. E. Harris, D. M. Lewis and M. Woolmer (Cambridge: Cambridge
University Press, 2015), 316–36.
15. Walter Scheidel, “Real Slave Prices and the Relative Cost of Slave Labor in the
Greco-Roman World,” Ancient Society 35 (2005): 1–17.
16. Kostas Vlassopoulos, “Historicising the Closed City,” in La cité interconnectée:
transferts et réseaux institutionnels, religieux et culturels aux époques hellénistique
et impériale, eds. M. Dana and I. Savalli-Lestrade (Bordeaux: Ausonius, 2018),
43–57.
17. Winfried Schmitz, “Sklavenfamilien im archaischen und klassischen Griechen-
land,” in Kindersklaven-Sklavenkinder: Schicksale zwischen Zuneigung und
Ausbeutung in der Antike und im interkulturellen Vergleich, ed. H. Heinen
(Stuttgart: Franz Steiner Verlag, 2012), 63–102.
18. Demosthenes, Against Meidias, 157–8.
19. Jason D. Porter, “Slavery and Athens’ Economic Efflor escence”, Mare Nostrum
10, no. 2 (2019): 25–50.
20. Siegfried Lauffer, Die Bergwerkssklaven von Laureion, I-II (Wiesbaden: Franz
Steiner Verlag, 1956).
21. Robin Osbor ne, “The Economics and Politics of Slavery at Athens,” in The
Greek World, ed. A. Powell (London: Routledge, 1995), 41–57.
22. Stephen Hodkinson, “Spartiates, Helots and the Direction of the Agrarian
Economy: Towards an Understanding of Helotage in Comparative Perspec-
tive,” in Slave Systems Ancient and Modern, eds. E. Dal Lago and C. Katsari
(Cambridge: Cambridge University Press, 2008), 285–320.
23. Jason D. Porter, “The Apophora and the ‘Leasing’ of Property to Slaves and
Manumitted Slaves in Classical Athens,” Historia 70, no. 2 (2021): 185–205.
24. Edward E. Cohen, “Sexual Abuse and Sexual Rights: Slaves’ Erotic Experience
at Athens and Rome,” in A Companion to Greek and Roman Sexualities,ed.
T. K. Hubbard (Chichester: Blackwell, 2013), 184–98.
25. Tracy E. Rihll, “Skilled Slaves and the Economy: The Silver Mines of the
Laurion,” in Antike Sklaverei, Rückblick und Ausblick: Neue Beiträge zur
Forschungsgeschichte und zur Erschließung der archäologischen Zeugnisse,ed. H.
Heinen (Stuttgart: Franz Steiner Verlag, 2010), 203–20.
26. Paulin Ismard, Democracy’s Slaves: A Political History of Ancient Greece
(Cambridge, MA: Harvard University Press, 2017).
27. Theophrastus, Characters , 21.4–5.
28. Hans Klees, Sklavenleben im klassischen Griechenland (Stuttgart: Steiner Verlag,
1998).
84 K. VLASSOPOULOS
29. Peter Hunt, “Trojan Slaves in Classical Athens: Ethnic Identity among Athenian
Slaves,” in Communities and Networks in the Ancient Greek World,eds.C.
Taylor and K. Vlassopoulos (Oxford: Oxford University Press, 2015), 129–54.
30. Kostas Vlassopoulos, “Free Spaces: Identity, Experience and Democracy in
Classical Athens,” Classical Quarterly 57 (2007): 33–52.
31. Pseudo-Xenophon, Constitution of the Athenians, 1.10.
32. Stéphanie Maillot, “Associations dites d’étrangers, clientèles et groups de travail
à l’époque hellénistique,” in Statuts personnels et main-d’œuvre en Méditer-
ranée hellénistique, eds. S. Maillot and J. Zurbach (Clermont Ferrand: Presses
Universitaires Blaise-Pascal, 2021), 285–313.
33. Kostas Vlassopoulos, “Athenian Slave Names and Athenian Social History,”
Zeitschrift für Papyrologie und Epigraphik 175 (2010): 113–144.
34. Rachel Zelnick-Abramovitz, Not Wholly Free: The Concept of Manumission and
the Status of Manumitted Slaves in the Ancient Greek World (Leiden: Brill,
2005).
35. Homer, Odyssey, 21.209–16.
36. Rachel Zelnick-Abramovitz, “Freed Slaves, Their Status and State Control in
Ancient Greece,” European Review of History / Revue européenne d’histoire 16,
no. 3 (2009): 303–18.
37. Karl-Wilhelm Welwei, Unfreie im antiken Kriegsdienst, I-III (Stuttgart: Franz
Steiner Verlag, 1974–1988).
38. Sara Forsdyke, Slaves Tell Tales and Other Episodes in the Politics of Popular
Culture in Ancient Greece (Princeton: Princeton University Press, 2012), 37–
89.
Further Readings
Bradley, Keith, and Paul Cartledge, eds. The Cambridge World History of Slavery 1:
The Ancient Mediterranean World. Cambridge: Cambridge University Press, 2011.
Forsdyke, Sara. Slaves and Slavery in Ancient Greece. Cambridge: Cambridge Univer-
sity Press, 2021.
Garlan, Yvon. Slavery in Ancient Greece. Ithaca NY: Cornell University Press, 1988.
Hodkinson, Stephen, Marc Kleijwegt, and Kostas Vlassopoulos, eds. The Oxford
Handbook of Greek and Roman Slaveries . Oxford: Oxford University Press, 2016–
present. https://doi.org/10.1093/oxfordhb/9780199575251.001.0001.
Hunt, Peter. Ancient Greek and Roman Slavery. Malden, MA: Wiley-Blackwell, 2018.
Ismard, Paulin. La cité et ses esclaves: institution, fictions, expériences. Paris: Le Seuil,
2019.
Klees, Hans. Sklavenleben im klassischen Griechenland. Stuttgart: Steiner Verlag, 1998.
Lewis, David M. Greek Slave Systems in their Eastern Mediterranean Context, c. 800–
146 BC . Oxford: Oxford University Press, 2018.
Taylor, Clair e, and Kostas Vlassopoulos, eds. Communities and Networks in the
Ancient Greek World. Oxford: Oxford University Press, 2015.
Vlassopoulos, Kostas. Historicising Ancient Slavery. Edinburgh: Edinburgh University
Press, 2021.
4 SLAVERY IN ANCIENT GREECE 85
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 5
Slavery in the Roman Empire
Noel Lenski
Introduction
The Roman Empire developed one of the largest and most economically and
culturally integrated systems of slavery in world histor y. It thrived on a remark-
ably robust supply stream that included enslavement by birth, capture, sale
from foreign and domestic sources, the reclaiming of exposed infants, and—
in late antiquity—self-sale, child sale, and debt bondage. Enslavement was
imposed upon people from all r egions, inside and outside the empire, and
was never inflicted exclusively on a particular racial or ethnic group. Those
enslaved to Rome worked in agriculture, industry, service, and even knowledge
production, allowing them to be the primary workforce behind the generation
of elite wealth. Escape from slavery could at times involve resistance, including
everything from open revolt to flight, but Roman society was also remarkably
generous with manumission. This and many other features reflect a hybridity
between ancient patterns of captive integration and modern habits of slave
exclusion.
N. Lenski (B)
Yale University, New Haven, CT, USA
e-mail: noel.lenski@yale.edu
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_5
87
88 N. LENSKI
Entry into Slavery
Roman slaving was practiced in a Mediterranean context where captive taking
and slaveholding were nearly universal. From the Semitic and Greek peoples
that populated the East, to the Germanic and Celtic peoples in the North and
West, to the Phoenicians, Africans, and Egyptians in the South, slaving was an
ancient and entrenched tradition before, during, and after the flourishing of
the Roman Empire between the third centur y BCE and the fifth CE. Aware of
this, the Romans enshrined it in the legal principles by which they structured
their own practices. When Roman jurists described the legal basis on which an
individual might enter slavery, they did so with reference to a bipartite division
between the “law of all peoples” (ius gentium) and their own “civil law” (ius
civile). The former included two of the most common ways to enter servile
status, birth to a slave and captivity to the Romans in enemy combat. For the
Romans, birth into slave status depended entirely on one’s mother, for the
Romans followed the principle of partus ventrem sequitur (offspring follow
the womb): an enslaved mother bore children who were slaves, while a free
woman birthed free children, regardless of the status of the father. We have
no quantifiable data to help us determine the relative importance of various
channels into the Roman slave supply stream, but modern historians agree that
birth was the biggest.1
Nevertheless, captivity by all means played a crucial role in bolstering the
enslaved population. Massive captive-taking events are recorded regularly in
the sources, many of them so sizeable that they temporarily crashed prices
on the slave market. This was largely a result of Rome’s repeated successes
in large-scale combat against foreign enemies and the ruthless efficiency with
which Roman armies disposed of enslaved persons into the market. Julius
Caesar, one of Rome’s greatest conquerors, is said to have enslaved 1,000,000
Gauls in his sweeping campaigns in the territory of what is today France
between 58 and 50 BCE; on a single day in 57 he sold 53,000 members
of the Atuatuci tribe to slave dealers; and when he achieved victory at the
Battle of Alesia in 52, he distributed one Gallic captive as plunder to each of
his 80,000 soldiers.2
Indeed, the Roman Empire attracted captives from all across the Mediter-
ranean—wherever Rome’s armies went, captives were generated for its
markets. These could come in waves as individual territories or polities were
integrated into the imperial machine: when Tiberius Sempronius Gracchus
defeated the rebellious inhabitants of Sardinia in 177 BCE, he brought 80,000
captives into the markets, equal to as much as a ten percent of the Italian slave
population at the time; Aemilius Paulus nearly doubled that number when he
defeated the peoples of Epirus in north western Greece, bringing in 150,000
more.3 The trend continued as Rome shifted from a Republican democracy to
an autocracy under the rule of an emperor in 31 BCE. To take just two exam-
ples, when Titus sacked Jerusalem in 70 CE, he carried off 97,000 Jews into
5 SLAVERY IN THE ROMAN EMPIRE 89
slavery; and Trajan’s campaigns against the Dacians in what is today Romania
probably generated at least 100,000 captives.4
Nor were all these captives put to work in menial labor, for the Romans
were happy to employ victims of enslavement with skills in their respective
fields of expertise. These included craftsmen, but also those trained in service
professions such as medicine, architecture, or music. According to Pliny the
Elder, the satyrist Publilius Syrus, the astronomer Manilius, and the gram-
marian Staberius Eros, each of whom had an important impact on Roman
intellectual history in the first century BCE, all arrived in Rome on the same
ship.5 If younger captives stood out for their aptitude, these might be trained
in a profession rather than being sent directly to the fields. Two captive boys
from the eastern kingdom of Parthia have left inscriptions confirming that one
was trained as a treasurer, the other as a pedagogue to the imperial family.6
All of which is to say that Rome used slaving as part of a larger imperial
apparatus for the integration of conquered peoples into i ts world system. As it
expanded its territorial reach, it generated first captives from the ter ritories it
conquered and only later, through techniques of domination and pacification,
provincial subjects. And neither provincials nor captives were entirely banished
from the levers of power. They were instead absorbed—together with their
skills and to some degree their customs. Roman captive-taking practices offer
only one of the many examples we will see in this essay of the way in which
Rome’s slave system sits squarely between the fluid and integrative captive-
taking systems that predominate in earlier, less complex societies and the more
compartmentalized systems of permanent racial slavery that characterize the
highly stratified and bureaucratized societies of modernity.
If birth into and captivity constituted the two primary ways into slavery
through the ius gentium, the Roman ius civile was also remarkably efficient
at enslaving people or perpetuating their enslavement. The most obvious
example is sale, which allowed one master to transfer the body of a previ-
ously enslaved person to the ownership claims of another. We have over sixty
slave sale contracts preserved on papyri (ancient Egyptian paper) and other
writing materials, such as wood and leather, to ser ve as concrete evidence of
such transfers.7 Because most of these documents were found in Egypt, where
Greek was the official language of the eastern Empire, most are written in
Greek, but we also have sale contracts in Latin and Syriac from find spots
stretching from Britain to the Euphrates River.8 These generally list the ethnic
origin of a slave and thereby confirm the staggering variety of regions from
which Roman slave traders drew—literally every corner to which the empire
stretched and also well beyond it. The most common and indeed the preferred
origin was, however, the “homebred slave” (Greek oikogen¯es;Latin verna).9
Birth into slavery generally rendered a person more docile and manageable
than foreigners or captives because homebred slaves were acculturated in local
customs and languages and had never known any life outside of enslavement.
All of this means that the “slaving zones” that characterize modern slavery did
90 N. LENSKI
not exist for the Romans. The reach of their enslavements was coexistensive
with their world.
Insofar as the Roman Empire favored certain regions to supply it with
slaves, these shifted over the seven centuries during which it remained at its
zenith. Rome’s expansion was always limited by military, financial, and above
all technological constraints, which meant that the Empire was surrounded
by ethnic and political others (“barbarians”), whom the Romans treated as a
potential source of slaves. In the early first century BCE, when the Romans
had conquered only a small slice of southern Gaul, they are said to have traded
for slaves with the Celtic peoples who controlled the remaining Gallic terri-
tory at the rate of one amphora of wine per human.10 After Caesar’s conquest
of Gaul, these same Celtic people became Roman subjects, such that they
were sending their own aristocrats to Rome to serve as senators by the mid-
first century CE. The Roman frontier had simply expanded and continued
to do so into the second century CE, allowing the Empire to draw on new
peoples beyond its frontiers to supply it with slaves, usually by human traf-
ficking among members of their own or neighboring tribes or polities: Scotti
(Irish) and Picti (Scots) in Britain; Frisians and Alamanni along the Rhine;
Sarmatians and Goths on the Danube; Armenians and Arabs on the eastern
frontier; Nubians and Gaetulians in Africa. These were transported, usually
on foot, shackled together at the neck or ankle, over the hundreds or even
thousands of disorienting miles to their new life as slaves to a foreign empire
(Fig. 5.1). We also have textual and archaeological evidence for slave markets
where they were resold. The city of Rome had markets in the Saepta Julia
and near the Temple of Castor, but markets are also attested in most major
cities of the Empire. The Aegean island of Delos—to which Rome had granted
free-port status—became the marketplace par excellence, leaving us abundant
textual and archaeological testimonies to its role as a trafficker of some 10,000
humans per day.11
The Romans also felt comfortable enslaving at least some of those whom
they regarded as provincial subjects. Here it is crucial to note that, as the
Empire expanded and incorporated new territories, it did not instantly grant
all subjects Roman citizenship. This was a privilege reserved for assimilated
members of the local elite, while others were offered more demanding path-
ways to citizenship, such as service in the Roman army. This meant that,
up to the year 212 CE, when the emperor Caracalla issued a sweeping law
granting citizenship to most of Rome’s provincials, less than twenty percent of
Rome’s subjects had attained citizenship. In the centuries before this, Rome
allowed its non-citizen subjects to follow their own regional legal systems,
and many of these permitted enslavement from within their own culture—
especially through the sale of children or debt servitude, neither of which
were permitted to Roman citizens. Once enslaved, such provincials could be
trafficked throughout the empire.
Furthermore, although the Romans of the High Empire never legally
permitted the full enslavement of those born with citizenship, de facto they
5 SLAVERY IN THE ROMAN EMPIRE 91
Fig. 5.1 Grabstein aus
Nickenich bei Mayen
(etwa 50 n. Chr.): Das
Relief zeigt einen
Sklavenhändler, der zwei
Sklaven an einer Kette
führt, um sie zu
verkaufen. Es ist im
LVR-Landesmuseum in
Bonn zu sehen. In regard
to the use of pictorial
material: use of such
material in this press
release is
remuneration-free,
provided the source is
named. The material may
be used only in
connection with the
contents of this press
release. For pictures of
higher resolution or
inquiries for any further
use, please contact the
Press office publishing
this directly. https://idw-
online.de/en/image?id=
274118&size=screen
allowed the enslavement of unwanted infants who were exposed by their
parents after birth. In the absence of effective birth control and safe abortion,
the Romans (whether citizens or provincials) dealt with unwanted pregnan-
cies by carrying a fetus to term, then exposing the newborn—usually on town
garbage heaps. Knowing this, profit-minded enslavers kept watch and snatched
up such infants to raise to four or five years, then sell as slaves. We know this
from multiple sources, but especially the more than forty surviving wetnurse
contracts, which were written to pay lactating women to feed exposed infants
for the first year or two of life and then surrender them as toddlers back to
the enslaver so he could sell them. We have no statistical data, but it appears
that child abandonment p rovided a major stream in the slave supply. Even the
emperor Claudius ordered a child born to his wife Messalina exposed because
he believed it to have been conceived by another man.12
92 N. LENSKI
Finally, it should be emphasized that the Roman Empire was incredibly
long-lived and that its customs and laws changed over time. In the flourish
years from the second century BCE till the second CE, legal protections for
citizens were strong and prevented most from falling into slaver y. But up to
326 BCE, the Romans had been happy to allow citizens to fall into debt
bondage to one another, and this practice became common once again by the
end of the fourth CE. Then too, the introduction of Christianity as the reli-
gion of the emperor after Constantine’s conversion in 312 CE caused a shift
in attitudes toward infant exposure and child sale. Prior to this, the Romans
had left exposure unregulated but forbade the sale of recognized citizen chil-
dren. Constantine began a process that reversed this formula, forbidding
exposure but allowing parents to sell their children into long-term inden-
ture—effectively servitude.13 This he did to protect parents from breaking
the commandment against killing (exposure often led to death) and to protect
children from being raised into a potential life of immorality, especially prosti-
tution. So too the practice of punishing those who committed major crimes by
enslaving them to the emperor (servitus poenae) was normal for the first five
centuries but was eliminated in the sixth century by emperor Justinian, who
based his decision on Christian concerns—because penal enslavement had the
effect of terminating marriages.14
The practice of enslavement thus changed over time, adapting to shifting
territorial, legal, political, and religious landscapes. But the Roman slave supply
was always fortified with multiple streams which ensured that, even as the
circumstances and attitudes that governed slaveholding shifted, supply always
met demand. This fact is confirmed by the stability of documented slave sale
prices throughout the imperial centuries, usually amounting to the equivalent
of about five years of wages for a day laborer.15
Experiences of Enslavement---Labor Extraction
Historians and sociologists have long debated what constitutes “slavery”—
how might it be defined? No consensus has yet been achieved, but two schools
have circled around the definition provided by the 1926 League of Nations
Charter (“Slavery is the status or condition of a person over whom any or all
the powers attaching to the right of ownership are exercised”) and that first
developed by Orlando Patterson in his sweeping 1982 monograph Slavery
and Social Death (“[Slavery is] the permanent, violent domination of natally
alienated and generally dishonored persons”). I have argued that the fullest
definition would combine both, for the former emphasizes the claims and
commitments of the master class while the latter conveys a fuller understanding
of the experience of the enslaved.16 Only in the dialectical space between these
two actors—enslaver and enslaved—can the dynamic of slavery be understood.
With this as background, we can turn to experiences of enslavement as it
was practiced by slaveholders, but also experienced by the enslaved. As we do
so, we must once again emphasize that Roman society functioned in a way that
5 SLAVERY IN THE ROMAN EMPIRE 93
simultaneously shared elements of antiquity and modernity. Like some ancient
societies, Roman society was at times ready to overlook the productive labor
potential of the enslaved in order to capitalize on the numinous or enter-
tainment value of their destruction. Into the middle Republican period, the
Romans occasionally used captives for human sacrifice (usually by live burial)
and they regularly forced enslaved persons to fight to the death in the arena
in spectacles that originated as a kind of funerary rite but took on the primary
role of entertainment by the mid-first century BCE. This pure wastage of
human life and labor never fully ceased even in the later empire, for a selection
of enemy captives was still regularly culled for spectacular execution in public
arenas.
Yet the Romans also developed a remarkably complex economic system that
bore many of the hallmarks of modernity. The Roman economy was highly
monetized; it was supported by a sophisticated institutional framework at the
level of commerce (banks and credit) and law (contracts and corporations);
it developed specialized agricultural markets in trade commodities (wine and
olive oil); it mass produced manufactured goods (tableware, metal work, oil
lamps); it supported specialized service industries (transportation, entertain-
ment, sex work); complex supply networks (capable of maintaining over a
thousand cities); complex labor differentiation and social stratification; and
remarkably developed engineering and technology (paved roads, aqueducts,
watermills). The combination of these features allowed the Roman empire to
compete with early modern west European economies in terms of GDP and
demographic growth, which also meant that its slave system could operate
with many of the same complexities of modern Atlantic slave systems—in fact,
arguably with more, for the openness of the Romans to the enslavement of all
peoples and races and their readiness to deploy slaves in nearly every economic
sector allowed the Romans to develop what was arguably the most socially
integrated slaving system in world history.
First and foremost, slave labor was deployed in primary industries, food
production. This sector commandeered as much as ninety percent of the labor
market, and slaves constituted the major force of production for members
of the elite in the high Roman centuries. Thus, while subsistence farming
continued among the free population throughout the Roman period, slaves
represented the major producers of marketable surplus for elite landholders
between the second century BCE and the second century CE. This we know
above all from the four agricultural manuals surviving from the Empire, those
of Cato (c. 160 BCE), Varro (c. 35 BCE), Columella (c. 60 CE), and Palladius
(c. 470 CE). The first three are filled with advice on the management of slave
labor on the farmstead, where slaves were organized into teams and placed
under a manager (vilicus), who was himself usually a slave. All three manuals
emphasize the need to keep enslaved workers busy throughout the year, but
they focus above all on viticulture, which was particularly capital and labor
intensive. Cato’s manual also lays out the size of rations permitted to slaves:
four and a half modii of wheat per month in summer as well as some wine
94 N. LENSKI
and olive relish for an average of c. 3500 calories per day. Cato also recom-
mends each slave be given a new tunic, a blanket, and pair of wooden shoes
every other year.17 These were, in other words, the barest of living condi-
tions. Animal herding (sheep, goats, cattle) was also labor intensive and was
thus regularly assigned to slaves. Herding slaves were, of course, given freedom
of movement, and because of the danger of rustlers and predators, they were
generally armed. This added to slave agency without generally threatening the
system.
We can catch a glimpse of the lives of agricultural slaves from the archae-
ology. Members of the elite tended to hold land as a collection of geographi-
cally diverse estates, each equipped with a large multi-purpose villa structure.
These were divided into large living areas for the owners, stalls for animals,
functional rooms for food processing and storage, and a series of much
smaller quarters (cellae) measuring 4–9 square meters for the slaves—prob-
ably one cella per slave family. Some, like Villa 34 at Gragnano, also had metal
hooks affixed to interior walls where some slaves could be chained during
sleeping hours. We learn from written sources that many such villae were also
equipped with prison cells (ergastula) in which to pen slave workers, although
archaeological examples have not been found.
It is important to bear in mind that enslaved persons were never the only
labor source for any given economic sector. The size of the enslaved labor
force varied over time but also geographically, with high-imperial Italy being
the period and place of most intensive slave use. Estimates of the percentage
of enslaved persons on the peninsula in this period have ranged as high as
35 percent, but most recently these have been revised downward to 15–25
percent (about 1–1.5 million persons from a total of 6–7 million).18 The
only part of the empire where we can begin to assemble firmer numbers is
Egypt, where it has been shown from extant census declarations that about
11 percent of the population was enslaved.19 This was perhaps representative
of other imperial provinces, which, as mentioned earlier, tended to structure
their societies and economies as they had before the Romans arrived. Thus,
in Asia Minor (modern Turkey) or Syria, where elites had tended to work
their estates using free (or semi-dependent) tenants in Hellenistic times, this
type of production prevailed. The same held of North Africa, which has been
shown always to have relied more heavily on free tenant labor rather than slaves
throughout the Roman period.20 This was also true because of variability in
crop production in different regions. North Africa, for example, was climati-
cally suited to oleiculture, which tended to be less labor intensive and relied
mostly on the development of new tree stocks, which took more time than
effort and was thus best suited to free tenants. Similarly, cereal cultivation was
demanding, but only in the seasons of plowing and harvesting. This meant that
grain crops could be put in by relatively small teams of enslaved plowmen and
farmers, then taken off by hired teams of harvesters each fall. Enslaved persons
were thus one source of labor for elite surplus production, surely the most
important one in High Imperial Italy, but not necessarily so in other places
5 SLAVERY IN THE ROMAN EMPIRE 95
or periods. Already in the second century CE we learn of grand Italian land-
holders using a mix of tenant and enslaved laborers, and by the fourth century
the imperial government tilted the balance in favor of the latter. Beginning in
the 320s, the emperor Constantine per mitted landowners to bind their tenants
(coloni) to the plots of land on which they had been born. This may initially
have represented an effort to pr otect tax revenues by eliminating the mobility
of taxpayers, but over the course of the fourth century the “bound colonate”
came to repr esent a third status between free and slave and to supplant slave
labor on larger estates.21
Enslaved persons were also employed in secondary industries, the manufac-
ture of durable and consumable goods. These included trade commodities
such as metal wares, clay lamps, and table wares. The last is particularly
well attested archaeologically, allowing us to trace both shifts in the centers
of production and networks of distribution. Much of the Mediterranean in
Roman times preferred to eat and drink from “red slipware” vessels, also
known as terra sigillata. These were mass produced to high-standards using
clay molds and then transported throughout the Mediterranean basin via water
transport on seas or rivers. Production was initially concentrated in Italy, then
moved to Gaul in the late first century CE, and shifted again in the third
to North Africa. Workshops employed up to sixty slaves each, but normally
tended to function with ten to twenty. The material excavated from La Grafe-
senque in Gaul is particularly well studied and offers rich testimony to the
potters involved, who proudly signed their wares.22 Yet the material from
La Graufesenque makes clear that free laborers were also involved in pottery
production. Indeed, while some industries relied heavily on slave labor, mining
for example, others appear to have been predominated by free workers, espe-
cially building construction.23 Enslaved persons were also heavily involved
in the processing and production of consumables. Baking offers rich testi-
mony. Because watermills only came into use in the second century CE, grain
processing was highly labor intensive, involving the turning of heavy horizon-
tally rotating stone mills that required tremendous effort on the part of animals
or humans. The second century BCE comic playwright Plautus often has
enslaved characters lament the possibility of being relegated to a bakery, and
the second century CE novelist Apuleius famously depicts a troop of emaciated
slaves pushing a grain mill with their whip-scarred bodies.24 But we also know
that a fair number of enslaved bakers eventually attained freedom, and some
went on to become owners and managers of their own bakeries and to amass
giant fortunes. The star example is Marcus Vergilius Eur ysaces, a freedman of
first century BCE Rome who gained enough wealth as a baker to construct
a gigantic (10 meters tall) monument for himself just outside the city gates
fashioned in the shape of a bread oven.25 Baking was thus quintessentially
Roman in its approach to labor, for it entailed the violent professional induc-
tion of enslaved laborers, some of whom then escaped from their enslavement
to enrich themselves using the professional skills they had learned.
96 N. LENSKI
Slaves were also involved in “tertiary” or service industries. Elite self-
construction in the Roman world hinged upon the control of enslaved bodies
that could be deployed to assist, accompany, curate, and enhance the lives and
prestige of slave owners. Many of the tasks moderns accomplish with tech-
nology were more labor intensive in antiquity, which meant that those with
ambitions to “live well” chose to exploit enslaved persons to do so. Elite
houses had enslaved cooks, housekeepers, chamber servants, doorkeepers,
watchmen, table servants, silver polishers, lamplighters, etc. This we know
from their funerary inscriptions, sometimes preserved in the family colum-
baria of Rome’s super elite. These include the empress Livia (59 BCE–29
CE), whose cadre of enslaved workers included at least 49 discreet kinds of
operatives, among them hairdressers (ornatrices) and a pearl-keeper (margar-
itarius).26 Enslaved personnel also shouldered the burdens of child-rearing as
wetnurses (nutrices), nannies (tattae), and educators (paedagogi).27 House-
hold slaves also played an important role in the daily rituals of the elite.
Each time they went to the public baths, wealthy slaveholders would bring
a large entourage of slaves who could carry their bathing implements, or wash
and massage them. On such trips the slaveowner regularly traveled in a litter
(lectica) born upon the shoulders of as many as eight specialized carrier slaves
(lecticarii).
Finally, in contrast with most modern slave systems, the Roman Empire
regularly employed slaves in quaternary industries, that is to say knowl-
edge production and curation. Slaves served regularly as doctors, midwives,
architects, astrologers, secretaries, accountants, property managers, business
managers, ship captains, etc. Some of this arose from the reality, already
discussed, that the Romans gladly capitalized on the skills and training of
enslaved persons in any economic sector. It also arose from the fact that
Roman law—like most premodern legal systems—had difficulty conceptual-
izing the assignment of agency to third party actors: a person was responsible
for their own dealings, and these responsibilities were not easily transferred to
an agent in ways that left the principal in an organizational chain legally liable.
But a workaround was provided by enslaved persons, who were considered
an extension of their master’s legal personality.28 Thesamelegal fiction also
led to the deployment of an imperial bureaucracy staffed by the emperor’s
own slaves (servi Caesaris), who performed most managerial and secretarial
functions of state for the first century of the Empire before being replaced
by freeborn bureaucrats.29 The same was true of individual cities, including
Rome itself, which owned their own teams of slaves (servi publici)toperform
tasks ranging from street sweeping and aqueduct maintenance to bookkeeping
and accounting.30
Particularly prized among household and imperial staffs were eunuchs,
males castrated in childhood to serve as household attendants. Castration itself
was distasteful to the Romans such that more than one emperor sought to
forbid it by law, but the Roman appetite for eunuch servants meant that the
practice continued inside the empire, and where supplies fell short, eunuchs
5 SLAVERY IN THE ROMAN EMPIRE 97
were imported from foreign markets.31 Eunuchs were especially prized as
chamber servants since they could not pose a threat to the reproductive
capacities of the household.
As this discussion indicates, the sexuality of enslaved persons was always a
primary concern and will be discussed in greater detail below. In regard to
labor output, however, enslaved persons were also regularly deployed as pros-
titutes. Evidence for prostitution abounds in all periods of Roman history,
and particularly from the first century BCE through the third CE. The most
vivid testimony comes from the city of Pompeii which had an estimated 35
brothels to service a town of 10,000 persons.32 The use of female sex workers
by male clients was greatly encouraged by Roman marriage customs, which
tended to favor early first marriage for females (ages 12–18) and later for
males (ages 20–25). Male surplus sexual energy in the years of adolescence and
early adulthood was thus channeled to sex workers, and the problem was only
exacerbated by mar riage regulations introduced by the emperor Augustus,
which strictly punished illicit extra-marital sex. But there were no legal conse-
quences for males fornicating with prostitutes, whether free or enslaved, and
the profits to be made led masters regularly to exploit the bodies of their
enslaved females—and males—as sources of income.33
Experiences of Enslavement---Violent Domination
If the discussion thus far has focused on the use of slave bodies to perform
labor services, we must still explore the question of the experience of enslave-
ment by those who endured it as a regime of physical, social, and psychological
repression. This problem was omnipresent, for even if some of the enslaved
persons discussed above were at times able to escape their fetters and some-
times to benefit from the training or status imparted to them while enslaved,
there was never a slave in the Roman Empire who did not experience slavery
as a relationship of violent domination, natal alienation, and general dishonor.
The final example in the previous section offers excellent proof of this concept.
While some enslaved female sex workers gained fame as professionals and were
even rewarded with freedom and wealth, they were never able to escape the
anguish of a life of serial assault and the stigma of enforced bodily exploitation
imposed on them by the master class.
Indeed, sexual assault was a regular experience for Roman slaves, both male
and female, whether or not they were exploited in the sex trade. Owners of
enslaved persons could and did have sex with them as they wished with no
legal consequences. Romans of the means were thus less likely to use public
prostitutes than simply to purchase sex slaves for their own exclusive exploita-
tion. This had the consequence that male masters fairly commonly freed and
married their slaves, a phenomenon well attested in funerary epigraphy and
one that further points to the integrative role enslavement sometimes played.34
Sex between female masters and male slaves was generally stigmatized, though
it too is attested, and some classes of elite male slaves, especially slaves owned
98 N. LENSKI
by the emperor, are known to have been sought as marriage partners by
free women. Sexual abuse was also common with same sex partners, partic-
ularly male masters who sexually abused their male slaves. This was especially
common with male youths, who were often expected to grow their hair long
and depilate body hair, sometimes well after the onset of puberty, in order
to be ogled and sexually assaulted by their master and his friends (Fig. 5.2).
The slave body, male and female, was thus a target for masters, who feared no
consequences for what would today be considered felony behavior.
The bodies of enslaved persons suffered domination not just as objects of
sexual exploitation but also in myriad other ways. The most basic of these was
hunger, for food was always used as a tool of control. We have already seen
that Cato prescribed set rations for his agricultural slaves, and other sources
report that starvation was used to punish uncooperative slaves.35 The sources
also confirm that the food given to slaves was distinctly inferior to that eaten
by their masters (coarse bread, sour wine, table scraps). The hyper-frugal Cato
recommended selling older slaves or otherwise getting them off the books
since they could no longer earn their keep with their labor power, and some
owners are known to have abandoned their slaves on the Tiber Island if they
regarded them as too sick to survive without the expense of a doctor’s visit—
leading the emperor Claudius to allow any such abandoned slave to go free
if they survived their illness and abandonment.36 Slaves who were inclined to
flee or whose behavior threatened or displeased a master could be bound with
shackles and chains, a regular feature of the archaeological record. And those
who fled habitually were often tattooed on the face so they could be identified
at a glance and returned. After Constantine forbade this practice, the Romans
turned instead to slave collars, permanently bonded to the neck with dog-tags
reading “Hold me for I am in flight.”37
Above all, however, slave bodies were tortured and physically abused, even
unto death, with no consequences for masters. Plautus’ second century BCE
plays regularly feature slaves terrified over an impending whipping, a trope
that was meant to elicit laughs from the audience. Similarly disturbing insou-
ciance about physical abuse is found in the epigrams of the first century CE
poet Martial: “You think me cruel and too fond of my stomach, Rusticus,
because I beat my [enslaved] cook on account of a dinner. If that seems to
you a trivial reason for lashes, for what reason then do you want a cook to be
flogged?”38 And assaults were often much worse than a beating. The physi-
cian Galen speaks of his experience of masters, including his own mother,
biting their slaves or gouging out their eye with a writing stylus.39 Ulti-
mately, the master could even kill his slaves with impunity. This he sometimes
did by contract, especially through the brutal punishment of crucifixion. An
inscription of Puteoli (modern Pozzuoli) lays out prices set by a company that
specialized in torturing and crucifying slaves on contract, allowing the master
to hire out this messy and physically demanding affair to specialized profes-
sionals.40 Here again Constantine became uneasy with this level of violence
and issued a law forbidding the deliberate killing of slaves in 319 CE, but in a
5 SLAVERY IN THE ROMAN EMPIRE 99
Fig. 5.2 Bronze image of a nude ephebe from Xanten. The boy, who would have
carried an actual tray, is shown long-haired and garlanded with his nude body right
at the end of prepubescence. Berlin, Staatliche Museen, Antikensammlung, Sk. 4.
Bildarchiv Preussischer Kulturbesitz/Art Resource, NY
subsequent law he granted tr emendous leeway for masters who happened to
kill a slave in the course of “corrective punishment.”41
Even when slaves were not openly abused, they lived in constant fear of
violence. They also lived in a world of “natal alienation,” which meant that
they were permanent outsiders, excluded from civic or political rights and
privileges, excluded from control over their own birth families and offspring,
and excluded from final control over their very bodies and personhood. Their
100 N. LENSKI
names could be assigned to them by a master and could be changed at any
time, particularly when they were sold to a new master. Their children could
be exposed or sold by their master at will. And they themselves could be liqui-
dated for their cash value at any moment. We have evidence of this process
from multiple sources which reveal enslaved persons intended for sale were
usually stripped down to a loincloth, displayed on a raised platform (catasta),
made to wear a garland if they were war captives and/or marked with chalk
on their feet if they were imported from overseas, their “defects” (disabilities,
diseases, habits) were publicly proclaimed on placards hung round their necks,
and they were subject to humiliating physical inspections by potential buyers
(Fig. 5.3).42 They were, in other words, treated in the manner of livestock at
market, with all of the attendant dehumanization and degradation.
Despite the repression inherent in this system, many enslaved persons
managed to salvage a remarkable degree of agency in their own lives. Much
of this occurred within the context of the paternalistic framework established
by the master class, which offered a clear if not always reliable set of path-
ways to recover personal subjectivity. Roman masters consider ed slaves part
of their familia, a wor d which included not just the nuclear family but also
its enslaved dependents. As such the Romans associated their slaves with chil-
dren, often referring to them as “boys” and “girls” (pueri, puellae), granting
rights of agency (noted above) similar to those enjoyed by children, and
offering both slaves and children the right to control some property in quasi-
ownership—peculium, which was wealth over which they could dispose even if
the father/master claimed bare ownership of it. This last in particular encour-
aged productivity, self-expression, and sometimes even freedom, albeit on
terms ultimately controlled by the master. Here again, we see a system poised
between the assimilative practices of captive-taking societies and the capital-
istic practices of modern chattel slavery, for masters did allow slaves to amass
Fig. 5.3 Tombstone of Capua depicting the sale of an enslaved person, stripped
and standing on a catasta with auctioneer (winged, to the right) and buyer, late first
century CE. G. Fittschen Neg. D-DAI-Rom 1983VW1305)
5 SLAVERY IN THE ROMAN EMPIRE 101
fortunes large enough to buy their own freedom and even the freedom of their
loved ones—spouses, children, parents, friends. And yet, even these freed-
persons (liberti) were bound to offer obedience and often labor services to
their former owner—their patronus, a word derived from the word for father.
Within this framework, we find numerous instances of enslaved persons who
embraced the profession assigned to them wholeheartedly and made it part of
their identity, so much so that, apart from their name, their profession often
constitutes their main or only identifier in the funerary inscriptions they have
left.43 Slaves also formed family groups of their own which, while not recog-
nized by law, represented a reality of their humanity that masters permitted
and even encouraged.
But it is by no means the case that all slaves chose to operate within
the guardrails established by the master class. This was in part because the
actual living and working circumstances created for slaves only sometimes
conformed to the ideal presented in the previous paragraph. Much depended
on the temperament of any given master, but also on the structure of the
work regime, which sometimes favored coercion, sometimes incentives. In less
skilled professions, chief among them agriculture, coercion was the default,
which meant that work regimes could be incredibly harsh and could foster
resistance and even revolt. Three large-scale slave uprisings erupted in the
late Roman Republic, the first two overrunning Sicily in 135–132 and 104–
101 and the third rampaging thr ough Italy in 73–71, the famous Spartacus
revolt.44 All three necessitated the commissioning of full-scale Roman armies,
thus showcasing the fact that Roman slaver y was anything but a benign insti-
tution. Nor was revolt the only option for slaves to resist the hegemony of
Roman slave society. Maroons of escaped slave brigands are attested across the
Mediterranean, most famous among them that of Bulla Felix, who is reported
to have sent a message to the emperor Septimius Severus, “Feed your slaves
so that they may not turn to brigandage.”45 Roman slaves are also known
to have murdered individual masters, a reality which led the Romans to issue
a law ordering the execution of all slaves present in a household if any one
of them murdered the master.46 At most times, however, slaves resisted their
masters in quieter ways: shirking their work, wandering off for long periods,
and damaging or stealing the master’s property.47
Roman slavery was thus fundamentally a system of domination and natal
alienation, equipped with guardrails and safety valves, but never far from raw
aggression. Chief among the mechanisms for relieving tension was Rome’s
practice of relatively frequent manumission, but there were also structures of
paternalism that granted enslaved persons some degree of agency in their lives
even without achieving freedom. But the violence at the core of the system was
always evident and made it such that Rome’s slave system could never be char-
acterized as humane. It was the enslaved themselves who had to assert, uphold,
defend, and recapture their humanity, at times through calculated cooperation,
at others through resistance and open revolt.
102 N. LENSKI
Exits from Slavery
It is a sad reality that the most common exit from bondage for those enslaved
to the Romans was death. For most this meant natural death, often due
to overwork or abuse, but some enslaved persons chose to hasten death’s
approach by taking their own lives. Seneca reports that one captive chose to
avoid being forced to fight in the arena by suffocating himself with a toilet
sponge and another by inserting his head into the spokes of the oxcart on
which he was being transported.48 Another escape route from slaver y was
flight. This was common in Roman society and was assisted by the relatively
undeveloped systems of communication available in antiquity. In the absence
of printing presses and a public postal system (let alone wireless communica-
tion networks), it was more complicated for masters to track down escaped
slaves, but this did not stop them from trying, often with the support of
the state. We have, for example, a series of letters from the politician Cicero
begging friends to help him find and return a person named Dionysius over
whom he claimed ownership and who had fled his household, stealing a
number of Cicero’s books on his way.49 In the imperial period, most cities had
some sort of public police force one of whose major tasks was to track down
and capture runaway slaves (fugitivi).50 We have already seen that tattooing
and collaring were also common safeguards, but the collective weight of the
evidence indicates that flight was common, and many surely managed to escape
enslavement permanently.
The only legitimate means out of slavery was, however, manumission. We
have already indicated that this was common in Roman society, especially when
compared to other slave cultures, like those of the Atlantic world. Attempting
to assign numbers and percentages is difficult given the nature of our evidence,
but careful analysis of the sources (especially inscriptions) points to remark-
ably high rates of manumission. Perhaps more than 30 percent of urban slaves
above the age of 25 could have expected to be given freedom, and many have
argued the numbers were even higher.51
The openness of the system to manumission can also be measured in the
variety of ways it could be achieved, ways which were adapted and shifted over
the course of the Roman centuries. Manumission is perhaps best described
as a formal ending to natal alienation and thus integration of the formerly
enslaved into the enslaving society as a member with subjective rights. In early
Roman society, this happened by literally enrolling a slave into the formal
census roles and thereby rendering them a Roman citizen. Of itself this is
remarkable since most slave societies never permit this level of integration,
but here too we find Rome’s liminal position between earlier captive-taking
societies, which were strongly inclined to the full integration of the formerly
enslaved, and modern societies, which have tended to resist it. The Romans
also permitted a state official or magistrate to manumit a slave in a formal
ceremony called vindicta, which was based on transactions of sale and thus
emphasized the property aspect of slavery, and which also conferred citizen
5 SLAVERY IN THE ROMAN EMPIRE 103
status. Over time, the Romans came to practice “informal” modes of manu-
mission such as freeing a slave before a group of friends or by a letter, attesting
to a desire for streamlining of the formal processes—and for public display.
But the emperor Augustus cracked down on informal modes in the first
century BCE and imposed rules excluding informally manumitted persons
from full citizenship. In the fourth century CE, a new mode arose, “man-
umission in church,” which was instituted by Constantine and also granted
full citizen status but incorporated manumission into Christian ceremonial,
thereby elevating it to the level of a “charitable act.”52
Masters were motivated to manumit slaves for a variety of reasons. These
included social pressure, for Roman society had a stated “preference for free-
dom” (favor libertatis ) which played itself out in law and practice. It was thus
a matter of prestige to manumit slaves, particularly in one’s will—a practice
so common that it too was restricted under Augustus, to no more than 100
slaves per decedent. Female slaves were often manumitted if they produced at
least four viable offspring, thereby reproducing their labor power since all such
children would have been born to slavery. Often loyal and productive slaves
could expect manumission by a certain age, c. 25–30, although this was never
guaranteed, and some made agreements with their masters to purchase their
freedom using money they had saved in their peculium. In every instance these
freedmen remained connected to their former masters through their family
names, which were always those of the patronus¸ and they were also bound to
their former enslavers in other ways. We have already seen that they owed the
patronus obedience and labor services, and most also owed part of their inher-
itance. By the generation following, however, the children of freedmen had
no further restrictions or obligations—even if they too retained the patronus’
family name. This meant that the children of freedmen could and did some-
times attain quite a lofty status—at least two children of freedmen became
emperors, Macrinus and Diocletian.53
Exits from slavery were thus in part controlled by slaves, when these fled or
otherwise resisted through revolt or suicide. Mor e commonly, however, they
were governed by the master class, which doled out manumission liberally
but also in calculating fashion to serve as an incentive and a mechanism of
control which could drive enslaved persons toward cooperation and eventually
assimilation into their hegemonic system.
Conclusion
Roman slavery was thus a peculiar institution. Because Rome was an ancient
society but also a remarkably precocious one, its slaving practices show traces
both of much older patterns and of quite modern ones. Modern are Rome’s
highly specialized and economically integrated uses of enslaved workers across
labor sectors and its ability to convert slave power into money capital thr ough
rationalist investment and profit accumulation. Ancient are its tendencies to
integrate enslaved persons into family and social structures and its readiness
104 N. LENSKI
to assimilate the enslaved (almost) fully into the dominant culture. Pervasive
throughout the system were the unmistakable hallmarks of all slave systems,
violent domination, and natal alienation. These realities were crucial to the
maintenance of Rome’s truly colossal slaving apparatus. Growing out of the
Empire’s inherently militaristic ethos, Roman slaving flourished in an envi-
ronment where violent power could be refined, controlled, and apportioned
into quanta of domination that enmeshed not just the enslaved but also the
enslavers in a nearly unbreakable social cage.
Notes
1. Walter Scheidel, “The Roman Slave Supply,” in The Cambridge World History of
Slavery, vol. 2, eds. K.R. Bradley and Paul Cartledge (Cambridge: Cambridge
University Press, 2011), 287–310.
2. Plutarch, Life of Caesar, 15; Caesar, Gallic Wars, 2.33; 7.89.
3. Livy 41.28.9; 45.34.3–7; Plutarch, Life of Aemilius, 29.4–5.
4. Josephus, Jewish Wars, 6.410; cf. Keith R. Bradley, “On Captives under the
Principate,” Phoenix 58 (2004): 298–318.
5. Pliny, Natural Histories, 35.199–201.
6. H. Dessau, Inscriptiones Latinae Selectae, nos. 1836; 1980.
7. J.A. Straus, L’Achat et la vente des esclaves dans l’Égypte romaine: Contribution
papyrologique à l’étude de l’esclavage dans une province orientale de l’empire
romain (München: K.G. Saur, 2004).
8. Syriac: Dura Papyrus no. 28. Latin: S. Riccobono, Fontes Iuris Romani Ante-
justiniani, vol. III, nos. 87–89, 132, 134; R.S.O. Tomlin, “‘The Girl in
Question’: A New Text from Roman London,” Britannia 34 (2003): 41–51.
9. Elisabeth Herrmann-Otto, Ex ancilla natus: Untersuchungen zu den “haus-
geborenen” Sklaven und Sklavinnen im Westen des Römischen Kaiserreiches
(Stuttgart: F. Steiner, 1994).
10. Diodorus Siculus, 5.26.2–3.
11. John Bodel, “Caveat Emptor: Towards a Study of Roman Slave-Traders,”
Journal of Roman Archaeology 18 (2005): 181–95.
12. Wetnurse contracts: Mariadele Manca Masciadri and Orsolina Montevecchi, I
contratti di baliatico (Milano: Tipolitografia Tibiletti, 1984). Suetonius, Life
of Claudius, 27.1. More on the importance of expositi in the supply stream
at William V. Harris, “Demography, Geography and the Sources of Roman
Slaves,” Journal of Roman Studies 89 (1999): 62–75.
13. Judith Evans Grubbs, “Church, State, and Children: Christian and Imperial
Attitudes toward Infant Exposure in Late Antiquity,” in The Power of Religion
in Late Antiquity, eds. Andrew Cain and Noel Lenski (Farnham: Ashgate,
2009), 119–31.
14. Novels of Justinian, 22.8.
15. Kyle Harper, “Slave Prices in Late Antiquity (and in the Very Long Term),”
Historia 59 (2010): 206–38.
16. Noel Lenski, “Framing the Question: What Is a Slave Society?” in What Is a
Slave Society? The Practice of Slavery in Global Perspective, eds. Noel Lenski
and Catherine M. Cameron (Cambridge: Cambridge University Press, 2018),
15–57.
5 SLAVERY IN THE ROMAN EMPIRE 105
17. Ulrike Roth, Thinking Tools: Agricultural Slavery Between Evidence and Models
(London: Institute of Classical Studies, 2007), ch. 2.
18. Walter Scheidel, “Human Mobility in Roman Italy, II: The Slave Population,”
Journal of Roman Studies 95 (2005): 64–79.
19. R.S. Bagnall and B.W. Frier, The Demography of Roman Egypt (Cambridge:
Cambridge University Press, 1994), 70–71.
20. Noel Lenski, “Peasant and Slave in Late Antique North Africa, c. 100–600
CE,” in Late Antiquity in Contemporary Debate, ed. R. Lizzi Testa (Newcastle:
Cambridge Scholars, 2017), 113–55.
21. See the studies in Domenico Vera, I doni di Cerere: storie della terra nella tarda
antichità (strutture, società, economia) (Turnhout: Brepols, 2020).
22. Luciano Lazzaro, Esclaves et affranchis en Belgique et Germanies Romaines,
d’après les sources épigraphiques (Paris: Belles Lettres, 1993), 419–25.
23. On mining: A.M. Hirt, Imperial Mines and Quarries in the Roman World:
Organizational Aspects 27 BC–AD 235 (Oxford: Oxford University Press,
2010) 222–28, 271–78, 314–18. On building: P.A. Brunt, “Free Labour and
Public Works at Rome,” 70 (1980): 81–100.
24. Plautus: Amy Richlin, Slave Theater in the Roman Republic: Plautus and
Popular Comedy (Cambridge: Cambridge University Press, 2017) 90–104.
Apuleius, Metamorphoses, 9.12.
25. On baking see Jared T. Benton, The Bread Makers: The Social and Professional
Lives of Bakers in the Western Roman Empire (Cham: Palgrave Macmillan,
2020), ch. 4.
26. Susan Treggiari, “Jobs in the Household of Livia,” Papers of the British School
at Rome 43 (1975): 48–77.
27. K.R. Bradley, Discovering the Roman Family: Studies in Roman Social History
(New York: Oxford University Press, 1991), 37–102.
28. Jean-Jacques Aubert, Business Managers in Ancient Rome: A Social and
Economic Study of Institores, 200 B.C.-A.D. 250 (Leiden: Brill, 1994).
29. P. R. C. Weaver, Familia Caesaris: A Social Study of the Emperor’s Freedmen
and Slaves (Cambridge: Cambridge University Press, 1972).
30. Alexander Weiss, Sklave der Stadt: Untersuchungen zur öffentlichen Sklaverei in
den Städten des römischen Reiches (Stuttgart: Franz Steiner, 2004).
31. Helga Scholten, Der Eunuch in Kaisernähe: zur politischen und sozialen
Bedeutung des “praepositus sacri cubiculi” im 4. und 5. Jahrhundert n. Chr.
(Frankfurt: P. Lang, 1995).
32. Thomas A.J. McGinn, The Economy of Prostitution: A Study of Social History
and the Brothel (Ann Arbor: University of Michigan Press, 2004).
33. Thomas A.J. McGinn, Prostitution, Sexuality, and the Law in Ancient Rome
(Oxford: Oxford University Press, 1998).
34. Matthew J. Perry, Gender, Manumission, and the Roman Freedwoman
(Cambridge: Cambridge University Press, 2014); Katharine P.D. Huemoeller,
“Freedom in Marriage? Manumission for Marriage in the Roman World,”
Journal of Roman Studies 110 (2020): 123–29.
35. Richlin, Slave Theater, 126–36; Kyle Harper, Slavery in the Late Roman
World, AD 275–425: An Economic, Social, and Institutional Study (Cambridge:
Cambridge University Press, 2011), 236.
36. Plutarch, Life of Cato, 4.4–5; Suetonius, Life of Claudius, 25.2; Dio Cassius,
60.29.7.
106 N. LENSKI
37. Hugh F. Thompson, The Archaeology of Greek and Roman Slavery (London:
Duckworth, 2003), 217–40.
38. Martial, Epigram, 8.23.
39. Galen, De propriorum animi cuiuslibet affectuum dignatione et curatione [Kühn
V. 17, 40–41].
40. François Hinard and Jean Christian Dumont, eds. Libitina: pompes funèbres et
supplices en Campanie à l’époque d’Auguste (Paris: De Boccard, 2003).
41. Theodosian Code , 9.12.1–2.
42. Pliny, Natural History, 35.200; Aulus Gellius, 6.4.
43. Sandra R. Joshel, Work, Identity, and Legal Status at Rome: A Study of the
Occupational Inscriptions (Norman: University of Oklahoma Press, 1992).
44. Theresa Urbainczyk, Slave Revolts in Antiquity (Stocksfield: Acumen, 2008).
45. Dio Cassius, 76.10.
46. Noel Lenski, “Violence and the Roman Slave,” in The Topography of Violence in
the Greco-Roman World, eds. Werner Riess and Garrett G. Fagan (Ann Arbor:
University of Michigan Press, 2016), 275–98.
47. K.R. Bradley, Slavery and Society at Rome (Cambridge: Cambridge University
Press, 1994), 107–31.
48. Seneca, Letter, 70.20–26.
49. Cicero, Letters to Friends, 5.9, 5.11.3, 13.77.3, and Letter to his Brother
Quintus, 1.2.14.
50. Christopher J. Fuhrmann, Policing the Roman Empire: Soldiers, Administration,
and Public Order (New York: Oxford University Press, 2014), ch. 2.
51. Henrik Mouritsen, The Freedman in the Roman World (Cambridge: Cambridge
University Press, 2011), ch. 5.
52. Justinianic Code, 1.13.1–2; Sozomen, Ecclesiastical History, 1.9; cf. Noel
Lenski, “Constantine and Slaver y: Libertas and the Fusion of Roman and
Christian Values,” Atti dell’Accademia Romanistica Costantiniana 18 (2011):
235–60.
53. Macrinus: Historia Augusta, Life of Macrinus, 4.3. Diocletian: Epitome de
Caesaribus, 39.1; Eutropius, 9.20.
Further Readings
Andreau, Jean, and Raymond Descat. The Slave in Greece and Rome. Translated by
Marion Leopold. Madison: University of Wisconsin Press, 2011.
Bradley, Keith R. Slavery and Society at Rome. Cambridge: Cambridge University
Press, 1994.
———. Slavery and Rebellion in the Roman World, 140 B.C.–70 B.C. Bloomington:
Indiana University Press, 1989.
Bradley, Keith R., and Paul Cartledge, eds. The Cambridge World History of Slavery,
Vol. 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press,
2011.
Harper, Kyle. Slavery in the Late Roman World, AD 275–425: An Economic, Social,
and Institutional Study. Cambridge: Cambridge University Press, 2011.
Joshel, Sandra R. Slavery in the Roman World. New York: Cambridge University Press,
2010.
Lenski, Noel, and Catherine M. Cameron, eds. What Is a Slavey Society? The Practice
of Slavery in Global Perspective. Cambridge: Cambridge University Press, 2018.
5 SLAVERY IN THE ROMAN EMPIRE 107
Mouritsen, Henrik. The Freedman in the Roman World. Cambridge: Cambridge
University Press, 2011.
Thompson, F. Hugh. The Archaeology of Greek and Roman Slavery. London:
Duckworth, 2003.
Vlassopoulos, Kostas. Historicising Ancient Slavery. Edinburgh: Edinburgh University
Press, 2021.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 6
Injection: An Archaeological Approach
to Slavery
Catherine M. Cameron
Introduction
Orlando Patterson’s transformative global study of slavery opens with the
observation: “Probably there is no group of people whose ancestors were not
at one time slaves or slaveholders.”1 The remainder of the book confirms the
truth of this statement. People have been captured and enslaved everywhere, as
far back in time as we can see. Yet, until relatively recently, these vast millions
of enslaved people were largely forgotten. They were marginal in life and
they became invisible in death. The role of archaeology is to recover peoples
and cultures of the past, reconstruct their lifeways, and use those reconstruc-
tions to explore cultural development through time. Our understanding of the
past is skewed, however, if we overlook a significant proportion of the past’s
peoples—in fact, an entire category of people: the world’s slaves. Just as clas-
sicists, historians, and other scholars have awakened to the need to investigate
slavery, archaeologists have become aware of the need to identify slaves in the
archaeological record and explore their lives and the effects they had on the
societies in which they toiled.
Finding slaves in the archaeological record first requires acknowledging that
they existed and then devising methods to identify them. Slaves, like women
and children, were invisible to archaeologists until relatively recently. Part of
this obliviousness results from a remarkable period of forgetting that occurred
C. M. Cameron (B)
Department of Anthropology, University of Colorado, Boulder, CO, USA
e-mail: catherine.cameron@colorado.edu
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_6
109
110 C. M. CAMERON
during the late nineteenth and twentieth centuries as laws against slavery were
enacted around the world. Where the presence of slavery was undeniable, as
in the southern United States, there was an embarrassed silence. The world’s
most brutal slave regime was referred to as the South’s “peculiar institution.”
It took the Civil Rights Movement of the 1960s and the empowerment of
African Americans to waken archaeologists to the potential of an archaeology
of slaver y.2
This injection reviews the lines of evidence that archaeologists are devel-
oping to identify these largely “invisible” people and explore the lives they
lived. The discussion is organized by societal scale. The first section, on state-
level societies, focuses on ancient Greece and Rome, emphasizing the paucity
of archaeological study of slavery in classical societies. Classical scholars have
been advised to look to the extensive archaeological study of the African
diaspora, the best-studied slave population. Methods and examples from the
American South and Caribbean are explored as a guide for archaeological
study of classical societies. The second section explores recent work on the
archaeology of slaver y in small-scale societies.
The Archaeology of Slavery in State-Level Society
There is little doubt that slaves existed in state-level societies from earliest
times; many early states were built and operated using slaves. Slaves appear
in early cuneiform tablets in Mesopotamia and they are depicted in bas relief
at Angkor Wat. Slaves are characters in ancient Greek and Roman plays and
were portrayed on Greek vases and on Roman mosaics and murals. Figurines
of slaves are found in ancient China, sometimes in graves, where they served
their master in the afterlife. Although they appear in art and literature, until
the past few decades, slaves were rarely the subject of archaeological study in
most parts of the world.
In contrast, in the American South archaeologists have studied slavery for
more than sixty years. The Civil Rights Movement of the 1960s, historic
preservation legislation, and the involvement of African-American organiza-
tions and individuals combined to incr ease awareness of the African-American
past and the role that archaeology can play in recovering that past.3
Classical Archaeology
Texts describing slaves and images of slaves found on pottery decorations,
murals, sculptures, and more make clear that slaves were numerous in both
ancient Greece and Rome. Nevertheless, archaeological studies of slavery in
the classical world are not well developed. This is in part because classicists
see textual evidence as more valuable and material evidence of slaves difficult
to identify.4 Human remains can provide evidence of slave status, as with the
two men discovered at Pompeii in 2020.5 The younger man’s body showed
6 INJECTION: AN ARCHAEOLOGICAL APPROACH TO SLAVERY 111
evidence of hard work and the older man’s did not; they were assumed to be
slave and master.
In spite of the obvious presence of slaves, efforts to identify slave dwellings
have met with difficulty. At times, structures have been proposed as slave
houses based on their location or layout, but there proves to be little confirma-
tory evidence. For example, Greek agricultural sites often have a rectangular or
circular tower which has been purported to be an accommodation for female
slaves based on only a single text that mentions their use for that purpose.
The island of Delos was well-known as a major slave market in the first few
centuries B.C. and had large estates worked by slaves. Small cell-like lower-
story rooms in buildings on these estates have been suggested as slave quarters
with the further assumption that males and females were spatially separated,
but there is no material evidence that supports this suggested use.6 Many slaves
are known to have lived and worked at the Greek mines of Attica making
this an obvious place to find evidence of slave lives. Given the likelihood that
slaves were procured from distant places, pottery and habitations in this mining
region might logically have been made in the style of the areas from which
slaves came.7 Yet this proved not to be the case. It is almost certain that slaves
were present at the settlements near the mines of Attica, but to date there is
little supporting material evidence.
Material evidence for slaves during the centuries of the Roman Republic and
Empire is somewhat greater, but not robust. Recent excavations in a suburb
of Pompeii have uncovered a small, cramped room that excavators interpret
as slave quarters; it had three beds, two adult sized and one apparently for a
child.8 Texts report underground slave prisons called “ergastula” where slaves
could be held, often in chains, and sometimes were worked there. These struc-
tures were described in texts as semi-subterranean with high barred windows,
but rooms fitting this description have rarely been found.9 Other material
evidence of slaves, found during the time of the Roman Empire, are slave
collars.10 These objects involved a metal collar that fit around a slave’s neck
and was inscribed with the slave’s name and instructions to hold or return
them if found; alternately the collar was uninscribed and used to suspend a
small metal plate (bulla) on which information about the slave was engraved.
Slave collars and bulla are infrequently found.
Archaeologists have recognized that likely many thousand slaves were sold
every year across the Roman empire but identifying slave markets has been
elusive. Using texts, images, and building layout, Fentress explores four build-
ings in Roman Italy for evidence of the sale of slaves at the front of these
buildings.11 She looked especially for porticos (chalcidica) where slaves could
be displayed, likely on platforms (catastae ), and an architectural layout that
would have controlled the flow of movement to and from this location.
Although evidence for the sale of slaves was not conclusive, she reasoned that
slave sales at the front of these buildings were likely and in all cases occurred
in central, monumental areas of the cities. In other words, the sale of human
112 C. M. CAMERON
beings in Imperial Rome was not hidden, but was a public activity, a part of
the daily life of the citizens of these towns.
The material evidence for slavery in classical Greece and Rome is not exten-
sive and some scholars argue that more effort needs to be put into developing
this field of study. Jane Webster argues persuasively that a comparison of
slavery in the classical world with that in the New World or elsewhere would
provide avenues for research that could reveal slaves in the ancient classical
world.12 For example, rather than imagining foreign slaves continuing to
create material culture like that of their birthplace, as Ian Morris suggested for
slaves in the Greek mines of Attica, classical archaeologists should recognize
that slaves might blend the material culture of their masters with that of their
homeland, creating a distinctive new identity expressed in classical material
culture—as was observed by archaeologists of the African diaspora. In other
words, before determining that an archaeology of slavery in the classical world
is unimportant, new approaches to finding slaves in the material record need
to be developed.
African Diaspora
Archaeologists studying the African diaspora have demonstrated the rich
understandings of the slave experience that archaeology can uncover. They
have set the tone for an archaeology of slavery elsewhere, and their work can
help classical scholars as they develop an archaeology of ancient Greece and
Rome. Archaeologists studying slavery of the African diaspora had an advan-
tage over classical scholars in that slave housing and associated artifacts are
generally easily identified, at least for large plantations. Extensive descriptions
of slave cabins exist for these plantations, and in the nineteenth century there
are even photographs. The living spaces and associated artifacts are much more
poorly known in holdings that had only one or a few slaves because their
housing and material culture blurred with that of their masters.
Archaeological studies of the enslavement of Africans in the New World
initially attempted to explore the lifeways of slaves. Zooarchaeological and
paleoethnobotanical studies of animal and plant remains from slave cabins on
plantations have been used to reconstruct the types and quantities of foods
slaves ate, including both domestic and wild sour ces. Although masters nomi-
nally provided enslaved peoples with food, archaeological remains demon-
strated that their diet was often extensively supplemented with wild foods that
they pr ocured themselves. Houses occupied by enslaved people were mostly
built using designs and materials provided by the master. But archaeology
has revealed modifications to these structures in response to the needs of
their occupants. For example, during excavations at the Seville Plantation in
Jamaica, slaves used the yard surrounding house structures for gardening, food
preparation, and social and ritual activities.13 In other words, slaves reoriented
their living space in ways that addressed their needs and took advantage of the
6 INJECTION: AN ARCHAEOLOGICAL APPROACH TO SLAVERY 113
environment (such as cooling sea breezes), even though forced to accommo-
date to structures dictated by their masters. In some parts of the American
South, slaves dug pits in the floors of their dwellings to store food and valu-
ables, in spite of opposition of their masters to this practice.14 Slaves used their
houses to create personal space in their otherwise tightly controlled lives.
New World archaeologists use artifacts to reconstruct the social identities
and ritual practices of slaves. Early studies aimed to link African-American
slaves to their African origins by attempting to locate objects from Africa
or objects made in an African fashion. Eventually, archaeologists recognized
that African Americans, like any other migratory group, blended the cultural
elements they brought with them with the cultural practices of the people
with whom they interacted in the New World, creating a distinctive African-
American culture. One of the most widely known artifact types associated with
enslaved African Americans is a handmade earthenware pottery type called
Colonoware. Colonoware is found on sites in the American South that had
large populations of African slaves, dating to the colonial period. A study
of Colonoware from nineteenth-century archaeological sites in Manassas,
Virginia, demonstrates that only the enslaved used Colonoware and that it
was associated with slave status.15 Although some archaeologists have argued
that Colonoware provides evidence for efforts by slaves to continue an African
identity, more recent studies reject that view. With emancipation, Colonoware
was quickly abandoned as liberated African Americans were able to exercise
consumer choice in the pottery they used.
In other studies, artifact caches and designs on artifacts have been used to
recover distinctive African-American ritual practices. Scholars have found that
African Americans used symbols on everyday objects as a method of communi-
cating beliefs that drew on both West African and Christian European elements
and those were being passed on to future generations.16 One example is the
BaKongo cosmogram, which likely served as an initial model for crossroads
symbols that are found on a variety of artifacts in the remains of many slave
communities, including pottery, buttons, spoons, metal objects, and more.
This “X” symbol was also reproduced and hidden beneath the floors of impor-
tant buildings (churches, midwife cabins) with caches of ritual objects buried
in each cardinal direction. Rather than simply a carry-over of African rituals,
such symbols were distinctively African-American and linked enslaved people
across the plantation South in a common symbolic understanding that was
“hidden in plain sight” from the dominant society.
Finding Slaves in Small-Scale Societies
Archaeologists have only begun to study slaver y in small-scale societies within
the past two decades, but already new methods are being introduced that
promise to illuminate marginalized peoples that, until recently, few archae-
ologists would have imagined existed. The recognition that slaves existed in
114 C. M. CAMERON
small-scale societies in the past is not insignificant. Archaeologists have envi-
sioned such gr oups as egalitarian with social differentiations based largely on
age and sex. The presence of slaves, of course, means that marginalized people
were common in small-scale societies and that significant social variability was a
normal part of life in such communities. Furthermore, the ownership of slaves
provided an opportunity for the individuals that held them to gain power. In
state-level societies, slaves clearly provided enormous power for their owners,
and they almost certainly did in small-scale societies, too. However, the inves-
tigation of these issues requires archaeologists to acquire the ability to detect
slaves in the archaeological record.
Material culture distinctive to slaves in small-scale societies is difficult to
identify. While images of slaver y bring to mind plantations with rows of slave
cabins, in small-scale societies slaves generally lived in their master’s house,
although they often slept in the least desirable parts. Fur thermore, they were
typically occupied with the same tasks as other members of the group. Wealthy
and powerful individuals tended to own the largest number of slaves, and
slaves might have freed the upper strata from quotidian labor. But since slaves
worked in productive activities alongside non-slaves, it is dif ficult to distinguish
their activities in the archaeological record.
Slaves are commonly discussed in ethnohistoric, historic, ethnographic, and
other written accounts of small-scale societies, and these texts have been used
as a starting point for archaeological identification of slavery in similar soci-
eties in prehistoric times. Common patterns in these accounts allow us to look
for contexts in which captive-taking and enslavement might have occurred.
In small-scale societies, captives were most often taken in raids or warfare,
although one or a few individuals were sometimes kidnapped and slaves may
have also been traded or sold from group to group. Worldwide, women and
children were most commonly taken captive, as adult males were difficult to
control and transport, and might pose a threat to the captor’s settlement.
These patterns allow archeologists to construct a series of expectations for
the presence of slavery in the small-scale societies of the past. Evidence of
warfare should alert archaeologists to the presence of captives and, poten-
tially, slaves. Such evidence includes defensive sites built on high or otherwise
inaccessible places; settlements surrounded by stout walls; empty “no-man’s
lands” between groups of settlements; weapons of war; iconography (rock art,
figurines, etc.) that shows violence; and trauma to human remains. Because
cross-cultural studies have suggested that women and children were most
commonly captured, skewed sex ratios in burial populations may be used to
identify the presence of captives or the absence of people taken captive: more
females may indicate groups that successfully took women, while more men
might indicate a society that had been raided for its women.
Human remains provide the strongest evidence for slavery in small-scale
societies. Slaves were subject to violence, and indicators of violence include
cranial fractures, signs of trauma and pathology in various stages of healing
6 INJECTION: AN ARCHAEOLOGICAL APPROACH TO SLAVERY 115
indicating repeated beatings (called “injury recidivism”), and fractures to fore-
arms (from warding off blows to the head or from violent falls), hands, feet,
ribs, or leg bones. Injury recidivism identifies subordinate individuals who may
be subject to frequent punishment. Blows to the head can cause neurological
trauma that results in a diminished ability of the victim to control her actions,
perhaps inviting further punishment.17
Slaves in small-scale societies were often foreigners to the societies in which
they found themselves, captives from distant regions. Consequently, another
place to start the search for slaves is by identifying non-local people. Isotope
analysis provides one of the best ways of identifying human movement in the
past. The two types of isotopic analyses most often used in studies of human
mobility are strontium (87Sr/86 Sr) and oxygen (18O). Different environments
have different isotopic signatures, and as people eat and drink, these isotopes
are incorporated into their bones and teeth. Tooth enamel is created between
ages four and twelve, while bone is remodeled throughout an individual’s
life. As a result, an individual whose bones and teeth have different isotopic
signatures, or whose teeth have an isotopic signature different from the envi-
ronment in which their body was found must have moved at some point in
their life. Ancient DNA (aDNA) is also used to explore human movement in
the past, but is most useful for identifying broad patterns of genetic ancestry
rather than the movement of individuals.
Although material culture distinctive to slaves in small-scale societies is
rare, other aspects of artifacts can help us to “see” captives. For example, at
the fourteenth-century site of Grasshopper Pueblo in the American South-
west, an excess of females in the burial population has been interpreted as
migrants fleeing war-torn areas to the north,18 although others have suggested
they were captives.19 The burial population also included a much higher
than normal number of children. Interestingly, non-local material culture at
Grasshopper was female-linked, including potter y and hearth style. Male-
linked material culture, including ceremonial architecture and projectile points
did not change after the influx of migrants, indicating that males at the site
were local.
A study of early twentieth-century artifact distributions along the Ucayali
River in western Amazonia found that, while most objects (weapons, utility
objects, clothing, ornaments, tools for body modification) became less
common as the distance to their center of manufacture increased, a number
of similar female-linked objects did not follow this pattern.20 Instead, objects
typically used by women were found in widely spaced tributaries of the Ucayali.
These were not places likely to exchange women as marriage partners. The
pattern is consistent with raiding for women. In other words, practices of
artifact manufacture had been introduced to these remote regions by captive
women. The presence of out-of-place female-linked non-local material culture
at archaeological sites should alert archaeologists to the potential presence of
captives.
116 C. M. CAMERON
Conclusions
The archaeology of slavery is a relatively new field but it promises to change
our understanding of many cultures in the past. When we think of times and
places in the past, whether a small tribal society in a remote landscape or an
ancient state, like the city-states of Mesopotamia, we should acknowledge that
some proportion of the people who made up these societies were marginalized,
dependent, probably enslaved. We should envision these people as important
actors in the societies of which they were a part. We know that in many times
and places in the past, marginalized people not only made many of the tools,
utensils, artwork, and more, but they also continued to handle, use, clean,
and care for these objects. They used tools to build houses, shrines, temples;
as well as to hunt or fish, manufacture canoes, process food, and dig irrigation
canals. They served their master’s food on dishes that they may have made,
painted the murals on the walls of the room in which the master dined, grew
and transported the food that was eaten. They made clothing, even though
they may have been denied substantial clothing themselves.
This injection suggests some of the avenues that archaeology is taking to
identify slaves in the past, reconstr uct their lifeways, and recognize the contri-
butions they made to the societies in which they lived. In state-level societies,
where slaves often lived and worked separately from their masters, finding
slave houses and material culture is easier. Such studies have been especially
successful in the American South and the Caribbean. But admittedly, for these
places there is abundant textual and iconographic documentation of the loca-
tions in which slaves lived and worked. Such detailed descriptions of slave lives
are lacking for ancient Gr eece and Rome, and archaeological studies of slave
lives in these societies are, as a result, more poorly developed. Greater efforts
by classical archaeology to explore slavery in the past may eventually open a
new window to our understanding of slave lives in ancient societies.
In small-scale societies where slave lives are closely entangled with those
of their masters, finding slaves in the archaeological record is difficult. Cross-
cultural ethnohistoric and ethnographic accounts suggest that captives were
most often women and children taken fr om other groups, so the identifica-
tion of non-local people provides a starting place for finding marginalized
or enslaved individuals. Slaves tend to suffer lives of violence, and studies of
human remains can uncover such treatment. These steps allow us to begin
to identify slaves in prehistoric small-scale societies, but scholars have yet to
develop robust means of studying the lives they lived. Such studies should
transform our understanding of small-scale societies that we once assumed
were egalitarian.
Notes
1. Orlando Patterson, Slavery and Social Death: A Comparative Study
(Cambridge, MA: Harvard University Press, 1982), vii.
6 INJECTION: AN ARCHAEOLOGICAL APPROACH TO SLAVERY 117
2. Theresa A. Singleton, “The Archaeology of Slavery in North America,” Annual
Review of Anthropology 24 (1995): 119–40.
3. Singleton, “The Archaeology of Slavery.”
4. Jane Webster, “Archaeologies of Slavery and Ser vitude: Bringing ‘New World’
Perspectives to Roman Britain,” Journal of Roman Archaeology 18 (2005):
161–79. Jane Webster, “Less Beloved: Roman Archaeology, Slavery, and the
Failure to Compare,” Archaeological Dialogues 15, no. 2 (2008): 103–23.
5. Associated Pr ess, 22 November 2020. https://www.nbcnews.com/news/
world/remains-man-his-slave-unearthed-ashes-pompeii-after-almost-2-n12
48541.
6. Hugh F. Thompson, The Archaeology of Greek and Roman Slaver y (London:
Duckworth, 2003).
7. Ian Morris, “Remaining Invisible: The Archaeology of the Excluded in Classical
Athens,” in Women and Slaves in Greco-Roman Culture: Differential Equations,
eds. Sandra R. Joshel and Sheila Murnaghan (London: Routledge Publishing,
1998), 193–220.
8. “Slaves’ Room Found in Pompeii,” The Past, 16 January 2022, https://the-
past.com/news/slaves-room-found-in-pompeii/.
9. Thompson, The Archaeology of Greek and Roman Slavery, 242–44.
10. Thompson, The Archaeology of Greek and Roman Slavery, 238–40.
11. Elizabeth Fentress, “On the Block: Catastae, Chalcidica and Cryptae in Early
Imperial Italy,” Journal of Roman Archaeology 18 (2005): 220–34.
12. Webster, “Less Beloved,” 103–23.
13. Douglas V. Armstrong, “Cultural Transformation Within Enslaved Laborer
Communities in the Caribbean,” in Studies in Culture Contact: Integra-
tion, Culture Change, and Archaeology, ed. James G. Cusic (Carbondale, IL:
Southern Illinois University Press, 1998), 378–401.
14. Singleton, “The Archaeology of Slavery in North America,” 119–40.
15. Laura J. Galke, “Colonowhen, Colonowho, Colonowhere, Colonowhy:
Exploring the Meaning Behind the Use of Colonoware Ceramics in
Nineteenth-Century Manassas, Virginia,” International Journal of Historical
Archaeology 13, no. 2 (2009): 303–26.
16. Kenneth L. Brown, “Retentions, Adaptations, and the Need for Social Control
Within African and African American Communities Across the Southern United
States From 1770 to 1930,” in The Archaeology of Slavery: A Comparative
Approach to Captivity and Coercion, ed. Lydia Wilson Marshall (Carbon-
dale IL: Southern Illinois University Press, 2015), 166–91. Leland Ferguson,
Uncommon Ground: Archaeology and Early African America, 1650–1800
(Washington, DC: Smithsonian Institution Press, 1992).
17. Debra L. Martin, “Ripped Flesh and Torn Souls: Skeletal Evidence for Captivity
and Slavery From the La Plata Valley, New Mexico, AD 1100–1300,” in Invis-
ible Citizens: Captives and Their Consequences, ed. Catherine M. Cameron
(Salt Lake City: University of Utah Press), 159–80; Debra L. Martin, Ryan
P. Harrod, and Misty Fields, “Beaten Down and Worked to the Bone: Bioar-
chaeological Investigations of Women and Violence in the Ancient Southwest,”
Landscapes of Violence 1, no. 1 (2010): art. 3.
18. Julia C. Lowell, “Women and Men in Warfare and Migration: Implications of
Gender Imbalance in the Grasshopper Region of Arizona,” American Antiquity
72, no. 1 (2007): 95–123.
118 C. M. CAMERON
19. Warren R. DeBoer, “Wrenched Bodies,” in Invisible Citizens: Captives and
Their Consequences, ed. Catherine M. Cameron (Salt Lake City: University of
Utah Press, 2008), 233–61.
20. Warren R. DeBoer, “Deep Time, Big Space: An Archaeologist Skirts the
Topic at Hand,” in Ethnicity in Ancient Amazonia: Reconstructing Past Iden-
tities From Archaeology, Linguistics, and Ethnohistory, eds. Alf Hornborg and
Jonathan D. Hill (Boulder: University Press of Colorado, 2011), 75–98.
Further Readings
Donald, Leland. Aboriginal Slavery on the Northwest Coast of North America.
Berkeley: University of California Press, 1997.
Fentress, Elizabeth. “On the Block: Catastae, Chalcidica and Cryptae in Early Imperial
Italy.” Journal of Roman Archaeology 18 (2005): 220–34.
Ferguson, Leland. Uncommon Ground: Archaeology and Early African America, 1650–
1800. Washington, DC: Smithsonian Institution Press, 1992.
Martin, Debra L. “Ripped Flesh and Torn Souls: Skeletal Evidence for Captivity and
Slavery from the La Plata Valley, New Mexico, AD 1100–1300.” In Invisible Citi-
zens: Captives and Their Consequences, edited by Catherine M. Camer on, 159–180.
Salt Lake City: University of Utah Press, 2008.
Morris, Ian. “Remaining Invisible: The Archaeology of the Excluded in Classical
Athens.” In Women and Slaves in Greco-Roman Culture: Differential Equations,
edited by Sandra R. Joshel and Sheila Murnaghan, 193–220. London: Routledge,
1998.
Patterson, Orlando. Slaver y and Social Death: A Comparative Study. Cambridge, MA:
Harvard University Press, 1982.
Reséndez, Andrés. The Other Slavery: The Uncovered Story of Indian Enslavement in
America. Boston: Houghton Mifflin Harcourt, 2016.
Singleton, Theresa A. “The Archaeology of Slavery in North America.” Annual
Review of Anthropology 24 (1995): 119–40.
Thompson, F. Hugh. The Archaeology of Greek and Roman Slavery. London:
Duckworth Books, 2003.
Webster, Jane. “Less Beloved: Roman Archaeology, Slaver y, and the Failure to
Compare.” Archaeological Dialogues 15, no. 2 (2008): 103–23.
6 INJECTION: AN ARCHAEOLOGICAL APPROACH TO SLAVERY 119
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
PART II
Medieval Societies (500–1500 C.E.)
Preface
The second part addresses the slaving practices in five wider regions before
the rise of the modern world: Byzantium, the Arabian peninsula, and Latin
Europe as the three monotheistic ruling powers in the Mediterranean, the
Black Sea region as one of the main slaving zones of the period, and the pre-
Columbian Inka state on the other side of the still unexplored Atlantic Sea.
Between the disintegration of the ancient empires and the integration of the
Americas into a new colonial system led by Western Europe, the trade and use
of slaves continued under different circumstances.
As in ancient times and no matter on which side of the Atlantic, in the
period from 500 to 1500 C.E., the border between capture, kidnapping, and
trading was often blurred. Slaves were used side by side with other coerced
people and entrusted with all sorts of duties connected to the needs of the
ruling elites. They served in the household, in the manufacture, and on the
field, they were traded as precious commodities, as symbols of prestige and
financial assets, and some of them held high positions of trust based on their
skills and education.
Four elements, however, prove characteristic for the “old world’s” millen-
nium before 1500. First, starting from the leading powers around the Mediter-
ranean Sea, the question of religious affiliation took center stage in discourses
on legal and illegal enslavements, in discussions on slaves as commodities
or human beings as well as in practices of slaving and manumission. The
mapping of the world in religious cultures of distinct monotheistic belief struc-
tured not only trading networks and political borders but also slaving and
no-slaving zones and patterns of social mobility and integration in Europe,
Asia, and Africa. Second, and closely linked to the first point, the sexual
122 PART II: MEDIEVAL SOCIETIES (500–1500 C.E.)
exploitation of female slaves took different forms according to the respec-
tive religious cultures. Sexual gratification aside, in Christianity, sex with slave
women produced property while in Islam it produced heirs, and whereas in
Judaism, it was forbidden but tolerated. Third, the male counterpart of the
concubine, the castrate, appears as an important figure and symbol for luxury
and loyalty in the court cultures of the medieval Mediterranean world. Fourth,
related to the formation of slaving and no-slaving zones, ethnic attributions
given to the most desirable groups of slaves such as “the Ethiopians” or “the
Caucasians” paved the way for discourses and practices of racing that little later
replaced religious affiliation as main determinant for enslavability.
CHAPTER 7
Slavery in the Byzantine Empire
Youval Rotman
Introduction
The importance of Byzantium to the history of global slavery stems from its
geographic and historical position. Byzantium boasts a history of more than
a millennium, longer than any other Mediterranean empire. As an offspring
of the Roman Empire, it inherited the Roman institution of slavery and its
legal definition. Yet, both proved to be in constant movement in view of the
changes that the medieval world underwent. The Byzantine Empire offers an
ideal historical context to examine questions about global slavery, questions
that pertain to continuity and change, conditions to entry slavery, the living
conditions of the enslaved, conditions of manumission, the destiny of ancient
slavery, and thanks to its geopolitical position, also to connectivity between
different medieval societies. It offers, in addition, a framework to examine the
states of enslavement and questions pertaining to the labor of the enslaved,
and its place within the socioeconomic organization.
Nowadays we address and emphasize the contradiction between the
humanity of the enslaved and the treatment they receive as a commodity in
the labor market. This contradiction stems from the perspective that sees all
human beings as having rights, and is a product of the modern age and the
human rights movement. To resolve this contradiction, activists today employ
two distinct forms of action. The first uses legislative means to enhance the
Y. Rotman (B)
Tel Aviv University, Tel Aviv, Israel
e-mail: yrotman@tauex.tau.ac.il
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_7
123
124 Y. ROTMAN
human rights of enslaved and bounded persons in order to eliminate the
enslavement conditions and exploitation. The second way of action uses the
status of the enslaved in the labor market to empower them in order to change
their human condition and the violation of their human rights. The present
chapter takes the case of Byzantine slavery to examine the relationship between
the status of the enslaved as a human being and as a commodity in pre-modern
society. In Byzantium this relationship was particularly important.
Contrary to popular belief, slavery did not decline in the transition from
the ancient period to the Middle Ages and was as much a part of medieval
societies as it was part of the society of the Roman Empire. This current
study of slavery in Byzantium will examine the status of the enslaved in the
Byzantine labor market and social organization, and will reveal the centrality
of the phenomenon of enslavement and the contribution of slavery to the
private household in the urban and agricultural economy. This analysis will
show the role that the institution of slavery played in social organization.
Moreover, examining slavery in view of the evolution of the Byzantine social
structures in the central Middle Age, particularly the growing socioeconomic
polarization of the Byzantine society starting from the ninth century, will
reveal the enslavement of human beings as a means to maintain economic
independence through the accumulation of “human property.” The other
side of the phenomenon of slavery in Byzantium was the relationship estab-
lished between this “human property” and the proprietors of humans, i.e.,
between the enslaved and the enslaver. Based on inequality, force, and asym-
metric dependency, in Byzantium this relationship was aimed nevertheless at
the integration of the enslaved, integration that carried legal and socioeco-
nomic consequences. A pivotal factor in this integration stemmed from the
religious identity of the enslaved persons and their conversion to Christianity.
This identity was both enforced on the enslaved persons, and contributed to
their integration into Byzantine society by granting them accessibility to legal
institutions and open ways for their manumission.
An analysis of the relationship between the enslaved status as a person and
as a commodity in Byzantine society may therefore clarify the tension between
the two different ways of dealing with slavery today, and will offer a new and
fresh perspective. Although there was no human rights discourse in Byzantium
in its modern form, the enslaved were perceived as human beings and were
attributed with the agency to act as independent persons. These possibilities
were not always in line with their exploitation, enslavement, and commodifi-
cation. In fact, enslavement, conversion, and manumission were three phases
in creating dependent agents to enlarge the socioeconomic position of the
household. They were therefore also means for the empowerment and liber-
ation of enslaved persons and for their integration into Byzantine society.
This chapter analyzes both sides of the phenomenon of slavery: the central
role of the enslaved persons in the private economic organization, and their
integration into the social organization. In Byzantium these two sides were
interdependent.
7 SLAVERY IN THE BYZANTINE EMPIRE 125
Byzantium was a direct successor of the Roman Empire in the eastern
Mediterranean starting from the four th century, when Emperor Constantine
I built a new capital on the r uins of the ancient Gr eek city of Byzantium and
named it after him, Constantinople (“Constantine’s City,” modern Istanbul).
The Byzantines, who continued to refer to themselves as Romans (and as
“Byzantines” in referring to the inhabitants of Constantinople), also inher-
ited the legal and social institutions of the Roman Empire. Their language
was Greek, and the religion of most of them was Greek Christianity. This
became the official religion, “the right faith” (Greek: Orthodoxy), and one
of the characteristics of the Byzantine state. Byzantium inherited slavery as a
legal institution from the Roman society. Under Roman-Byzantine law both
male and female slaves had no juridical persona, in a similar way to children
for example. They could not own property, become a party of legal contracts
(including marriage), serve as guarantors, give legal testimony, or sue or be
sued.
Although there is no record of a decline in the use of slaves in the late
Roman period, historians have tended to connect the idea of the “decline of
the Roman Empire” in late antiquity (fourth-seventh centuries) to the idea of
“decline of slavery.” This idea is not supported by historical sources. Indeed,
studies of the last two decades have revealed the existence of various forms
and institutions of slavery in medieval societies.1 In the seventh century, the
Mediterranean entered a new stage in its history, with its division into three
distinct civilizations: the Islamic Caliphate in the south, Byzantium in the
northeast, and Latin Western Europe in the northwest. Each had its own
language, culture, and religious identity. Slavery continued to play an impor-
tant role in the socioeconomic life of Byzantium and the Caliphate. Following
the Crusades, a new element was added to this map with the creation of local
Latin kingdoms in the Eastern Mediterranean. In addition to the important
role it played in local socioeconomic structures, slavery became also a means
of connectivity between the different states.
Entry into Slavery
Like the Roman law, Byzantine law determines that the free status of the child
came from the mother side. If she had free status during her pregnancy (even
if she was later reduced to slavery), the child received the status of free born.
In the same way, the children of a slave woman, no matter who was their
father were slaves from birth. Cases of self-sale of free persons who sold them-
selves for debts or other reasons were known since antiquity and were legal
provided that the person was sui iuris. Children, who did not have a legal
persona, could be sold or exposed by their parents. Child exposur e could lead
to enslavement if the children were abducted by slave traders who then sold
or prostituted them illegally. In the fourth through sixth centuries Byzan-
tine emperors promulgated laws to limit such cases, and allowed it only for
parents in dire economic circumstances. Reducing a free born person, was
126 Y. ROTMAN
finally prohibited by Leo VI (886–912), and in 1095 by Alexios I Komnenos
(1081–1118). We have nearly no evidence of enslavers who sold the chil-
dren of their slaves to a third party. In fact, most of the documented cases
of enslavement in Byzantium point to manumission of the enslaved as the
norm, even the objective of Byzantine slavery. These left two ways as the main
means to procure slaves in Byzantium: war and trade. Both cases concerned
the enslavement of foreigners.
Slavery became dependent on the enslavement of captives. Wars and
conquests filled this demand in particular during the period of extensive
wars: the sixth through the eighth centuries. By the end of the eighth
century the medieval geopolitical map stabilized. It was no longer a period
of large conquests and geopolitical annexations. Piracy, captivity, abduction,
and enslavement of people became the dominant forms of enslavement espe-
cially in frontier zones. The Byzantine sources often depict the Byzantines
as the main victims of raids of pirates coming from the Caliphate. But Arab
sources of the ninth and tenth centuries reveal that raids were also practiced
by Byzantine forces in both land and sea. And yet, the significant part of the
enemy population which was captured was not sold as booty, but was kept for
prospective acts of ransoming or exchanges of prisoners of war. This left as the
main way to acquire starting from the ninth century the slave trade. In fact,
it is impossible to distinguish between the medieval slave trade, piracy, and
captivity. Indeed, pirates by land and sea were also slave traders and vice-versa.
The medieval slave trade is normally not referred to as trafficking. And yet the
evidence of the lives of the enslaved, their abduction, the violent ways they
entered slavery, their forced migration, and the scale of the international slave
trade all point to a new international dynamic centered on human trafficking.2
Sources and documents from the period reveal wide-ranging itineraries of
slave traders that connect Eastern Europe, the Eurasian Steppe, and the African
Sahel to the markets and economies of the Mediterranean, where the demand
for slaves was high and the financial means for their purchase were available.3
In this commercial dynamic Eastern Europe and the Slavic countries were
the main source of slaves for Mediterranean societies, Byzantium, and the
Caliphate in particular. This orientation of the slave trade marks demographic
and economic differences in the Middle Ages between the richly populated
areas in the south and east of the Mediterranean and the undeveloped areas
northwest of the basin and the African desert and the Sahel. This economic
imbalance was the main engine for the medieval human trafficking.
The term “Slavs” became in the Middle Ages a generic name for slaves
both in the Arabic of Muslim Al-Andalus (s
.ak
.¯aliba) and in Greek in Byzan-
tium (sklavoi, σκλ ´αβ
o
ι).4 The term later penetrated most of the Western and
Central European languages. The enslaved were mostly victims of slave traders
and pirates, including Vikings, who operated along the rivers between the
Baltic Sea, the Caspian Sea, and the Black Sea, and between Eastern Europe
and southwestern Europe. Human trafficking was based on raids by merchants
and private and military militias that captured the local population, either
7 SLAVERY IN THE BYZANTINE EMPIRE 127
through abduction or for a fee paid to local leaders, and led them far from
their country of origin. Slavery in the Middle Ages was therefore dependent
on the enslavement of abducted and forced migrants.5
The average price of a man in Byzantium was around 20–25 gold coins,
and remained fairly constant in Byzantium.6 This was the price of a house in a
county town; an average price of three shops in the capital, Constantinople; a
wage of one year for an employee in the public ser vice, or five to eight years of
an employee. It was a serious financial investment, worthwhile only for wealthy
households that could afford it.
TheEnslavedinthe Labor
Market and Social Organization
Slaves were employed in Byzantium in every possible economic and social
role, in both rural and urban milieu, in the private and public sector, in the
service of the emperors, the socioeconomic elite as well as by less wealthy
people. The basic economic unit was the private household. Its level deter-
mined the number of enslaved persons it included. Enslavement was a means
of increasing the economic power and the social position of the family’s house-
hold in both the rural and urban economic systems. Regulations from the
eighth-tenth centuries dealing with the economic-legal organization in the city
and in the Byzantine village include references of slaves as part of the economic
organization of the private household of peasants and urban enterprises alike.7
Although slaves are mentioned alongside waged/hired workers (Greek:
misthioi, misth¯otoi ergatai) in both the city and the countryside, their mode
of employment was different from the second. Waged workers were employed
under a specific labor contract (misthosis ) concluded between employer and
employee. This form of work differed from that of slaves: it was limited to one
month and the salary had to be paid in advance.8 A household, rural or urban,
could not therefore employ a wage worker over time. These regulations of the
labor market encouraged the growth of economic organizations that were not
dependent on hired labor but on slavery. They were particularly critical for
economic enterprises that required trained professionals, such as goldsmiths,
money changers, animal traders, shopkeepers, carpenters, builders, painters,
and in various types of candle, soap, and silk fabrication.9 Slaves could be
employed as a long-term, even life labor force. Moreover, unlike wage workers,
slaves could become guild members and serve as managers of private enter-
prises such as shops and workshops. Normally, five guarantors were required
to open a private enterprise in Constantinople. But in case of slaves the guar-
antee of the owner was sufficient. No one would have agreed, presumably, to
be a guarantor of another man’s slave, and slaves could not stand as guaran-
tors because they had no legal persona.10 This made slaves the ideal business
managers. The economic consequences of this situation were far-reaching. In
order to set up a business, a slave could be appointed as responsible for life.
The social consequences were also far-reaching: on the one hand, a potential
128 Y. ROTMAN
weakening of financial relations of inter-socioeconomic dependency. On the
other hand, the strengthening of independent households that gained their
independence by acquisition of enslaved agents. A person who was interested
in setting up workshops of various kinds (which was prohibited by law) could
use slaves for this purpose, and appoint them at the head of numerous work-
shops. The socioeconomic dynamics, then, relied on the financial ability to
acquire and enslave people in order to employ them as managers and workers
for life. The fact that the enslaved were the property of the enterprise’s owner,
meant that all profit and control was in the hands of the second. Slavery was
thus a means of increasing the economic independence of the family house-
hold. In this way, the socioeconomic rationale of slavery in Byzantium fits in
with anthropological theories that see slavery as a means of expanding the
family organization.11
The same rationale also applied in the Byzantine rural organization, which
was composed of landowners, slaves, employees, and working animals.12 In the
ninth-century Life of Philaretos the merciful, a historical figure who possessed
rich lands in Asia Minor and became a saint, Philaretos’ household deteriorates
from wealth to poverty. He loses his large estate and retains only a modest plot
of land around his house which he cultivates himself with his son and daughter.
The loss of his slaves indicates his economic decline. This description clarifies
what the reverse process was: how a small family household could be devel-
oped into a large and rich estate. Indeed, information about this comes from a
Byzantine document, dated to the tenth through twelfth centuries, about the
tax organization of the Byzantine village.13 It shows that the expansion of the
rural household from a modest land to a large estate depended on the acquisi-
tion of manpower. Slavery served precisely this objective. Moreover, enslavers
linked their enslaved persons in couples and profited from their offspring who
were enslaved from birth and continued to maintain the family’s agricultural
enterprise. In this way slavery provided a means of increasing private economic
independence, a means that was accumulated and managed by the family unit.
In the central medieval period, Byzantine society experienced a transfor-
mation in its economic organization, and saw the creation of a new socioe-
conomic elite who gained its richness from the control over farmer lands,
hitherto independent. Byzantine sources from the ninth through eleventh
centuries refer to this new elite as “the powerful” (hoi dunatoi, oƒ δυνατo´
ι in
Greek). Families close to the imperial government gained authority over large
tracts of land by receiving control over the taxes of the land.14 The farmers,
either owners or state tenants, became dependent on by private powers who
controlled their land taxes and as a consequence also their far mers’ socioeco-
nomic position. A social dependency was created between those who worked
the land and those who controlled it, that helped to establish the second as
a new elite. Against the background of this new socioeconomic dynamics,
slavery gained a new role. Acquiring and enslaving people in order to use
them in farming became the main option through which independent farmers
could enrich their estates and improve their economic situation in view of the
7 SLAVERY IN THE BYZANTINE EMPIRE 129
growing influence of the new elite of “the powerful.” Maintaining economic
independence in Byzantium was dependent on the ability to accumulate
“human property” to use it in both work and management. Enslavement was
therefore a means to acquire and maintain economic independence.
Private testaments from the period show how widespread these dynamics
were in the organization of rural family units. This is the case of Gemma’s
1049 testament from Puglia in southern Italy.15 Gemma, a widow without
children, left her land and houses to her four nephews. She bequeathed other
houses, plots of land, and cattle to her manumitted slaves: three men and
three women, along with three more persons whose juridical status. Her rural
household, therefore, consisted of a number of lands and houses inhabited
and cultivated by relatives, slaves, and freedmen. This was also the case of
larger landowners, for example, Eustathios Boïlas who drafted his testament
in 1059.16 His lands in southeast Anatolia included the eleven villages he
founded himself. He left most of his property to his two daughters, their
husbands, and the churches he founded. Along with them he mentioned three
orphans he raised, who r eceived two of his villages. He also mentioned fifteen
slaves alongside their families and children, all of whom he had previously
manumitted. They all received plots of land. He bequeathed his other slaves,
along with the lands and cattle, to his daughters. These were probably the
main labor force of the estates and enabled him to build his eleven villages as
an independent economic unit.
Another detailed picture of Byzantine household management comes from
the testaments of the Pakorianoi couple (Symbatios Pakourianos and his
widow Kal¯e Pakourian¯e).17 The couple lived in Constantinople in the eleventh
century and belonged to the social elite close to the emperor. They owned the
lands of four villages. Thirty-one men and woman are mentioned by name
in Symbatios Pakourianos’ testament, eighteen among them are slaves. Upon
his death he manumitted all his enslaved men and bequeathed them clothes,
bedding, horses, weapons, and modest sums of money. He bequeathed his
enslaved women to his wife, who, in her later testament, manumitted all her
slaves, women, and men. The couple referred to their entire staff in the testa-
ments by the overall term “my people” (hoi anthr¯opoi mou; ¥νθρωπoι μoà in
Greek): all those who are in their service. These were not what modern schol-
arship term “domestics.” Their function was not limited to domestic roles
within the house, but they sustained, supported, and maintained the entire
economic organization of the household of this aristocratic family.18 Upon
their manumission, the enslaved men and women remained attached to the
household and its owners, and continued to sustain and maintain the family
unit as a private economic system. The term “my people” indicates that the
strength of a household depended on its economic independence, and this
meant the number of people who maintained it. This enabled the socioeco-
nomic mobility of the entire unit, which included the socioeconomic mobility
of the enslaved themselves. These remained a part of the private household of
their enslaver, according to Byzantine customs, also after their manumission.
130 Y. ROTMAN
Manumission of the enslaved acquired a paramount importance in Byzan-
tium as a legal means to integrate the enslaved into Byzantine society. In fact,
enslavement and manumission were two sides of the same coin. Together they
ensured the dependency and the integration of the enslaved. The new reli-
gious identity that the enslaved acquired in Byzantium played a key role in the
process of “flipping the coin” toward their integration.
Exit from Slavery, Economic
Dependency, and Social Integration
As was shown above, human trafficking in the Middle Ages was unprecedented
in its geographical scope. Its victims were usually local children, women,
and men who were abducted, trafficked, and sold into slavery very far from
home and country. In the medieval world this meant that they were different
also in their religion. The medieval world was divided between different
political blocs with distinct religions: Greek Christianity in Byzantium, Islam
in the Caliphate, Latin Christianity in western Europe, Jewish communities
throughout these regions, and populations that were still pagan in the Slavic
world, northern Europe, and the Sahel. The result was that the women and
men who were abducted, enslaved, and trafficked to the Byzantine markets
were foreign in origin and faith. The process of enslavement included the
conversion of the enslaved to the religion of the enslaver: in Byzantium to
Christianity, in the Caliphate to Islam, or to Judaism in the Jewish commu-
nities (conversion to a religion other than the state religion was forbidden in
both the Caliphate and Byzantium). A series of laws, regulations, and treaties
from Byzantium, Venice, Rome, and Francia, from the ninth through the
twelfth centuries, restricted and prohibited the trade in Christians, and the
sale of slaves to Jewish and Arab slave traders.19
Foreigners who were enslaved by Byzantines were usually not Christian
and were converted to Christianity by their enslavers. This separation between
enslaved and enslaver according to faith gave moral justification for enslave-
ment: the act of enslavement itself being regarded as an outgrowth of religious
superiority and a sense of religious mission to convert. Indeed, starting
from the fourth-century Christian writers developed different justifications for
slavery. Some saw it as a product of war, others as a crime, sin, or stupidity. At
the basis of all these justifications was a worldview that saw slavery as part of
the existing divine order and therefore legitimate and justified.20 At the same
time, conversion was also a means of integration. The conversion of slaves
made them part of the religious community.21 This too was the meaning of
the conversion process of the enslaved: a religious and social conversion that
made the foreigner “one of us” and therefore trustworthy. This is reflected
in the two Byzantine legal customs of slave manumission: manumission in
church and manumission by baptism.22 The first was introduced in the fifth
century and was performed in the church in front of a bishop who acted as
the magistrate, with no reference to the religion of the manumitted slave. The
7 SLAVERY IN THE BYZANTINE EMPIRE 131
second, attested from the eighth century, aimed at manumitting non-Christian
slaves through their baptism by their enslaver, act that also created legal kinship
between the enslaver and the freedman.
The new religious identity of the Byzantine enslaved person also opened up
possibilities for personal empowerment and social mobility. In fact, the new
religious identity of the enslaved changed their legal status. Although slaves
continued to be defined by law as property, the very fact that they were Chris-
tians turned them from objects to subjects because they were perceived and
considered also as believers. Moreover, their religious identity as Christians
conferred a legal personality through which they could realize their status
as believers: marry, create a family, become churchmen or monks, and even
act independently to be liberated.23 These fundamental changes were due to
the fact that the enslaved Christian was perceived not only as a subject of
his enslaver, but also a subject of God. As such the question of his loyalty
whether to his material master or to the heavenly master (ho kurios, Ð κ
´ριoς
in Greek) was open to interpretation. A number of Byzantine writers addressed
this question in great detail, especially in light of the sentence from the Gospel
according to Matthew 6:24 that it is not possible to worship two masters, God
and Mammon. This verse was interpreted as a contradiction between loyalty to
God and any material/corporeal master. Thus, for example, Gregory Bishop
Nisa explains that the very fact that a person owns other persons is a violation
of God’s property right over all of humanity.24 Although this view did not lead
to anything close to an abolitionist movement, Byzantine Christianity never-
theless dealt with the question of authority: to whom man, whether enslaved
or not, owed his primary loyalty: to an earthly or an eternal master.25 This
perspective also saw the enslaved as a subject: God’s subject. An outgrowth of
this approach was a legal development in the status of the enslaved.
One of the most interesting features of slavery in the medieval world
concerns the development of access to the law for slaves. The roots of this
approach can be traced back to the late Roman period.26 It became more and
more common with the recognition of the religious identity of the slave as
a believer and a part of the religious community. So it was for example in
regards to marriage. By its very definition as a legal contract, the institution
of marriage was impossible for slaves, and in fact meaningless. Moreover, the
enslavement of a married person immediately entailed the annulment of the
marriage. However, the recognition of Christian marriage as a legal institution
in Byzantium as an unbreakable legal relationship meant that the enslave-
ment of a married person did not change the marriage, for example in case
of prisoners of war. The Byzantine legislator has intervened in such cases by
allowing marriage between spouses when one of them was enslaved by a third
person.27 Moreover, the Byzantine legislature increasingly interfered with a
person’s authority over his human property, authority that was traditionally
considered private. For example, an eleventh-century law prohibited any possi-
bility of marrying slave couples outside the Christian institution of marriage.28
132 Y. ROTMAN
This law severely restricted enslavers who wanted to unite their slaves in fami-
lies outside of Christian marriage. The legislative application of the Christian
institution of marriage also in the case of slaves made these ties unbreakable.
The sale of married slaves became impossible, and the manumission of part of
the enslaver’s family could become legally problematic.29 The religious status
of the enslaved made them therefore part of the religious society, and gave
them a legal status that allowed them agency and opened up possibilities in
regards to their private life.30
Byzantine slaves therefore were defined by a legal status that we can under-
stand as civil status, that is a category defined by legal status. By civil status I
mean a legal definition that constitutes a distinct group of people in terms of
duties, privileges, or other criteria. A legal definition indicates which criteria
set a group of people as a civil category. The purpose of such a legal delimita-
tion is to give a special status to this group of people. We need to distinguish
between legal status and civil status, since the second can apply only to human
beings as society members. In this way too, Byzantine law delimited the free-
born and the enslaved by determining the criteria by which the enslaved was
distinguished from all other members of society. In the same way, for example,
the age criterion for minors, or that of sex for women, defined their respective
civil status. Moreover, the civil status of Byzantine slaves was in movement
because of their religious identity as Christians. This movement was a means
of their social integration into the society of believers; sometimes it weakened
the enslaver’s property rights. This was already manifested, for example, in the
asylum law of Justinian (527–565), which gave runaway slaves the possibility
to become a monk or a clergyman without the permission of their owner.31
The owner could only demand them back for a short period of time and only
if they proved that they had caused damage.
These legal changes that started from the sixth century and increased in
the tenth century, reflect the development of the civil status of the enslaved.
They stem from a new approach regarding the authority of the Byzantine state
and law: the expansion of the civil status of the slave was done in parallel with
the strengthening of the authority of the public authority at the expense of
restricting private authority over human property. This was not a deliberate
empowerment of the enslaved, but a result of the Byzantine imperial policy
to increase the authority of the state and its legal regulation in a way that
restricted private authority over “private subjects,” meaning Byzantines who
were not under the authority of the state, such as slaves. Slavery continued to
exist, but those who were enslaved by private enslavers were not exclusively
the private property of the enslaver, but also subjects of the authority and laws
of the state and therefore of the emperor.
This legal process did not lead to an abolitionist attitude, but to a new
definition of the enslaved: not merely as property, but as men and women
who are part of the private household, and as Christians also part of the reli-
gious society. Even if the enslaved persons still had inferior legal status, were
restricted in their movement and cruelly treated, the dependence of the family
7 SLAVERY IN THE BYZANTINE EMPIRE 133
organization on them along with the changes in their d efinition as members
of the religious society, often led to their manumission. Manumission did not
make the enslaved independent or free to go their own way. On the contrary,
as freedmen the manumitted slaves continued to be part of the household of
their enslaver/manumitter. They received economic autonomy, the autonomy
that according to Byzantine wills was part of the expansion of the family’s
economic organization. The manumitted slaves were still very much depen-
dent on their former enslaver, whom they continued to refer as “master”
(kurios, κ ´
ρι
o
ς in Greek). The option to be “free” and to go wherever one
wanted was destructive, both economically and socially. It would have left
the manumitted slave without a socioeconomic attachment and any means of
subsistence. In fact, the dependency of the manumitted slaves on their enslaver
opened up opportunities for their socioeconomic integration.
Conclusion
Enslavement entails an ongoing act of violence, and is always accompanied by
the exploitation of human beings by other human beings. At the same time,
conditions were created in Byzantium for interdependency between enslaved
and enslaver: the enslaved were dependent on the enslaver for every detail
of their personal life, and the enslaver depended on the enslaved on the
success of their economic independence. The integration and empowerment
of the enslaved became an interest to both enslaved and enslaver and created
dynamics that led to the social integration of the first, an integration from
which the second benefited. Moreover, this interdependency continued after
manumission which opened more options for both sides. Manumitted slaved
acted as empowered agents in the family household to which they belonged
and of which they were a part. We would be wrong to think that manumitted
slaves had better conditions if they lived a “free life” independently of their
enslaver’s household. A “free life” meant a hard and detached life from any
social and economic framework, a homeless life, with no source of living and
minimal living conditions. Such was the situation of the poor who lived in a
daily war of survival on the margins of society. It is precisely dependency ties
that have provided living conditions and opportunities of empowerment. The
uniqueness of the case of Byzantine slavery lies in the fact that the manumis-
sion of the enslaved was worthwhile to both enslaved and enslaver because the
first remained dependent on the household of the second. The empowerment
of the first contributed to the empowerment of the second. These dynamics
point to the transformation of the enslaved from being a passive victim into an
active agent. In other words: in Byzantium, the empowerment of the enslaved
was beneficial to the enslaver. Development in the legal status of the enslaved
gave them more and more options when it came to their private lives, and
was the engine behind their empowerment. The analysis of the case of Byzan-
tine slavery provides a unique perspective on questions regarding slavery in
134 Y. ROTMAN
general. It shows that structural economic, social, and legal elements are what
shapes the civil status of the enslaved.
Notes
1. Youval Rotman, Byzantine Slavery and the Mediterranean World, trans. Jane
Marie Todd (Cambridge, MA: Harvard University Press, 2009); Alice Rio,
Slavery after Rome, 500–1100 (Oxford: Oxford University Press, 2017); Carl
Hammer, A Large-Scale Slave Society of the Early Middle Ages: Slaves and their
Families in Early Medieval Bavaria (Aldershot: Ashgate Publishing, 2002);
Matthew S. Gordon, The Breaking of a Thousand Swords: A History of the
Turkish Military of Samarra (AH 200–275/815–889 CE) (Albany, NY: State
University of New York Press, 2001); Y¯usuf R¯ag¯ıb, Actes de vente d’esclaves et
d’animaux d’Egypte médiévale, 2 vols. (Cairo: Institut français d’archéologie
orientale, 2002–2006); Kurt Franz, “Slavery in Islam: Legal Norms and Social
Practice,” in Slavery and the Slave Trade in the Easter Mediterranean (c. 1000–
1500 CE), eds. Reuven Amitai and Christoph Cluse (Brepols: Turnhout, 2017),
51–142; Mohamed Meouak, S
. aqâliba: eunuques et esclaves à la conquête du
pouvoir. géographie et histoire des élites politiques “marginales” dans l’Espagne
umayyade (Helsinki: Academia scientiarum Fennica, 2004); Youval Rotman,
Slaveries of the First Millennium (Leeds: ARC Humanities Press, 2021).
2. Youval Rotman, “The Medieval Slave Trade: Map, Data, Sources” (H-Slavery
Resources, D. Prior, ed., 11 Jan. 2019, https://networks.h-net.org/system/
files/contributed-files/yrotman2cmedievalhumantrafficking2cmapanddata.pdf);
Christopher Paolella, Human Trafficking in Medieval Europe (Amsterdam:
Amsterdam University Press, 2020).
3. Michael McCormick, Origins of the European Economy, AD 300–900
(Cambridge: Cambridge University Press, 2001); Salah Trabelsi, “Commerce et
esclavage dans le Maghreb oriental (VIIe-Xe siècles)” and Mohamed Meouak,
“Esclaves noirs et esclaves blancs en al-Andalus umayyade et en Ifr¯ıqiya
f¯atimide,” in Couleurs de l’esclavage sur les deux rives de la Méditerranée (Moyen
Age - XXe siècle), eds. Roger Botte and Alessandro Stella (Paris: Karhala, 2012),
9–53.
4. Helga Köpstein, “Zum Bedeutungswandel von ‘sklabos’/sclavus”, Byzan-
tinische Forschungen 7 (1979): 67–88. Marek Jankowiak, “What Does the
Slave Trade in the Saqaliba Tell Us About Early Islamic Slavery?” International
Journal of Middle East Studies 49, no. 1 (2017): 169–72; Meouak, S
. aqâliba.
5. Youval Rotman, “Migration and Enslavement: The Medieval Model,” in
Migration History of the Medieval Afro-Eurasian Transition Zone, eds. Lucian
Renfandt, Johannes Preiser-Kapeller and Iannis Stouraitis (Leiden: Brill, 2020),
387–412; Paolella, Human Trafficking.
6. Rotman, “The Medieval Slave Trade.”
7. Das Eparchenbuch Leons des Weisen, ed. Johannes Koder (Vienna: Verlag der
Österreichischen Akademie der Wissenschaften, 1991); “Nomos ge ¯orgikos”,
ed. Walter Ashburner, in Jus Graecoromanum,eds.I¯oann¯es D. Zepos and
Panagi¯ot¯es I¯oannou Zepos (Athens: Georgion Phexis & uiou, 1931), vol. 2,
67–71.
7 SLAVERY IN THE BYZANTINE EMPIRE 135
8. Das Eparchenbuch Leons des Weisen; Alexander Kazhdan, “Slaves and misthioi
in Byzantium ninth-eleventh century,” Uchenye zapiski Tul’skij gosudarstvenyj
pedagogicheskij instituta 2 (1951): 63–88 (in Russian).
9. Das Eparchenbuch Leons des Weisen. See the thorough study of Daphne Penna,
“The Role of Slaves in the Byzantium Economy, 10th–11th Centuries: Legal
Aspects,” in Slavery in the Black Sea Region, c.900–1900: Forms of Unfreedom
at the Intersection between Christianity and Islam, ed. Felicia Ro¸su (Leiden:
Brill. 2021), 63–89.
10. Das Eparchenbuch Leons des Weisen, §§2.9; 8.13.
11. Suzanne Miers and Igor Kopytoff, eds., Slavery in Africa: Historical and
Anthropological Perspectives (Madison: University of Wisconsin Press, 1977).
12. “Nomos ge¯orgikos…”. John of Ephesus, Lives of the Easter n Saints,ed. E. W.
Brooks, Patrologia Orientalis (Paris: Firmin-Didot et Cie, 1923), vol. 17, 191–
95. Rotman, Byzantine Slaver y, ch. 3; Ghislaine Noyé, “Quelques observations
sur l’évolution de l’habitat en Calabre du Ve au XIe siècle,” Rivista di studi
bizantini e slavi 25 (1988): 57–138.
13. Franz Dölger, Beiträge zur Geschichte der byzantinischen Finanzverwaltung,
besonders des 10. und 11. Jahrhunderts (Hildesheim: Teubner, 1964), 3–9, 113–
156; Charles M. Brand, “Two Byzantine Treaties on Taxation,” Traditio 25
(1969): 35–60.
14. Eric McGeer, The Land Legislation of the Macedonian Emperors (Toronto:
Pontifical Institute of Mediaeval Studies, 2000); Nikolaos Oikonomides, “The
Social Structure of the Byzantine Countryside in the First Half of the
10th Century,” Byzantina Symmeikta 10 (1996): 105–25; Leonora Neville,
Authority in Byzantine Provincial Society, 950–1100 (Cambridge: Cambridge
University Press, 2004); Paul Lemerle, Cinq Études sur le XI e s. byzantin (Paris:
CNRS, Paris, 1977).
15. Gertrude Robinson, Cartulary of the Greek Monastery of St-Elias and St-
Anastasius of Carbone (Rome: Pontificium Istitutum Orientalium Studiorum,
1929), 179–84.
16. Lemerle, Cinq Études, 15–63; Claudia Rapp, “Zwangsmigration in Byzanz:
Kurzer Überblick mit einer Fallstudie aus dem 11 Jahrhundert,” in Erzwun-
gene Exile: Umsiedlung und Vertreibung in der Vormoderne, ed. Thomas Ertl
(Frankfurt A.M.: Campus, 2017), 59–80.
17. Jacques Lefort, Nikolaos Oikonomidès and Denise Papachryssanthou, eds.,
Actes d’Iviron II: du milieu du XI e siècle à 1204 (Paris: Lethielleux, 1990),
150–56, 170–83.
18. Paul Magdalino, “The Byzantine Aristocratic oikos,” in The Byzantine Aris-
tocracy IX-XIII Centuries, ed. Michael Angold (Oxfor d: B.A.R., 1984),
92–111 (repr. in Paul Magdalino, Tradition and Transformation in Medieval
Byzantium, Aldershot, Hampshire: Variorum, 1991, pt. III).
19. Agobardi Lugdunensis Opera omnia, ed. Lieven van Acker (Turnhout: Brepols,
1981), 189–95; Urkunden zur Älteren Handels-und Staatsgeschichte der
Republik Venedig, eds. G.L.F. Tafel and G.M. Thomas, 3 vols. (Amsterdam:
Hakkert, 1964), vol. 1, §§3, 7, 12–14; I trattati con Bisanzio 992–1198,ed.
Marco Pozza and Giorgio Ravegnani, Pacta Veneta 4 (Venice: Il Cardo, 1993),
§1.
20. Peter Garnsey, Ideas of Slavery from Aristotle to Augustine (Cambridge, UK:
Cambridge University Press, 1996); Jennifer A. Glancy, Slavery in Early
Christianity (Oxfor d: Oxford University Press, 2002).
136 Y. ROTMAN
21. Rotman, Byzantine Slavery and the Mediterranean World, 139–52.
22. Codex Theodosianus, ed. Theodor Mommsen (Berlin: Weidmann, 1962),
IV.7.1; Fabrizio Fabbrini, La manumissio in ecclesia (Milan: Pubblicazioni
dell’Istituto di diritto romano e dei diritti dell’Oriente mediter raneo, 1965);
Fabrizio Fabbrini, “Un nuovo documento relativo alla manumissio in ecclesia,”
Rendiconti della Classe di Scienze morali, storiche e filogiche 16, nos. 5–6
(1961): 214–15. Ciro Giannelli, “Alcuni formulari relativi alla ‘manumissio
in ecclesia’ tratti da eucologi italo-greci e slavi,” Rivista di cultura classica
e medioevale 1, no. 2 (1959): 127–47. Ecloga: Das Gesetzbuch Leons III.
und Konstantinos’ V , ed. Ludwig Burgmann (Frankfurt A.M.: Lowenklau-
Gesellschaft, 1983), §8.1; Jus Graecoromanum,vol.2,139 (Prochiros Nomos,
§7.28).
23. See below fn. 27–30. Basile, Lettres, ed. Yves Courtonne, 3 vols. ( Paris: Les
Belles Lettres, 1957–1966), §115. Grégoire de Nazianze, Lettres, ed. Paul
Gallay (Paris: Les Belles Lettres, 1964–1967), §79.
24. Gregory of Nyssa, Homilies on Ecclesiastes, ed. Stuart G. Hall (Berlin: De
Gruyter, 1993), Homily §4.1 (ed. Lionel Wickham); Chris L. De Wet, The
Unbound God: Slavery and the Formation of Early Christian Thought (London:
Routledge, 2018). See also Ilaria L.E. Ramelli, Social Justice and the Legiti-
macy of Slavery: The Role of Philosophical Asceticism from Ancient Judaism to
Late Antiquity (Oxford: Oxford University Press, 2016) for the proliferation
of these ideas among monks.
25. Life of St. Andrew the Fool, ed. Lennart Rydén, 2 vols. (Uppsala: Uppsala
University Library, 1995), chs. 1–5.
26. The Digest of Justinian, eds. Paul Krüger and Theodor Mommsen, trans. Alan
Watson, 4 vols. (Philadelphia: University of Pennsylvania Press, 1985), XLVIII,
8, 4.
27. “Die Novellen des Kaiserin Eirene”, ed. Ludwig Burgmann, in Fonter Minores
4(Forschungen zur Byzantinischen Rechtsgeschichte 7) (1981), 26; Les Novelles
de Léon le Sage, eds. Pierre Noailles and Alphonse Dain (Paris: Les Belles
Lettres, 1944), §§100–101.
28. Jus Graecoromanum, vol. 1, 341–46. Köpstein, “Zur Novelle.”
29. H Πε
‹
ρα—The Peira: Ein juristisches Lehrbuch des 11. Jahrhunderts aus
Konstantinopel—Text, Übersetzung, Kommentar, Glossar, eds. Dieter Simon and
Diether Roderich Reinsch (Berlin: De Gruyter, 2022), 342.
30. Les Novelles de Léon le Sage, §§11; 37; 38; 40; 49; 60; 69; 100; 101.
31. Novellae Justiniani, §123.
Further Readings
Pahlitzsch, Johannes. “Slavery and the Slave Trade in Byzantium in the Paleologan
Period.” In Slavery and the Slave Trade in the Eastern Mediterranean (c. 1000–1500
CE), edited by Reuven Amitai and Christoph Cluse, 163–184. Turnhout: Brepols,
2017.
Penna, Daphne. “The Role of Slaves in the Byzantium Economy, 10th-11th Centuries:
Legal Aspects.” In Slavery in the Black Sea Region, c.900–1900: Forms of Unfreedom
at the Intersection between Christianity and Islam, edited by Felicia Ro¸su, 63–89.
Leiden: Brill, 2021.
7 SLAVERY IN THE BYZANTINE EMPIRE 137
Prinzing, Günter. “Sklaven oder freie Diener im Spiegel der ‘Prosopographie der
mittelbyzantinischen Zeit’ (PmbZ).” In Prosopon Rhomaikon: Ergänzende Studien
zur Prosopographie der mittelbyzantinischen Zeit, edited by Ralph-Johannes Lilie,
Alexander Daniel Beihammer, Bettina Krönung, and Caludia Ludwig, 129–173.
Berlin: De Gruyter, 2017.
Rotman, Youval. Byzantine Slavery and the Mediterranean World, translated by Jane
Marie Todd. Cambridge, MA: Harvard University Press, 2009.
———. “Byzantium and the International Slave Trade in the Central Middle Ages.”
In Trade in Byzantium: Papers from the Third International Sevgi Gönül Byzan-
tine Studies Symposium, edited by Nevra Necipoglu and Paul Magdalino, 129–142.
Istanbul Koç University Publications, 2016.
Rotman, Youval. “Migration and Enslavement: The Medieval Model.” In Migration
History of the Medieval Afro-Eurasian Transition Zone, edited by Lucian Renfandt,
Johannes Preiser-Kapeller and Iannis Stouraitis, 387–412. Leiden: Brill, 2020.
———. Slaveries of the First Millennium. Leeds: ARC Humanities Press, 2021.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 8
Slavery in Medieval Arabia
Magdalena Moorthy Kloss
Introduction
This chapter discusses slavery in medieval Arabia1 by focusing on Yemen, the
Arabian Peninsula’s southwestern part. Sources from the eleventh to fifteenth
centuries CE feature slaves in diverse life situations: as human commodities
sold on public slave markets, commanders of cities, mothers of their masters’
children, resistors to enslavement, laborers in kitchens and workshops, and
as freed persons. The following case study will characterize the life trajec-
tories of slaves in medieval Yemen and highlight three central aspects: their
enslavement (entry into slavery ), how they lived and worked as slaves (experi-
ences of slavery), and the possibilities of altering their unfree status (exits from
slavery). The evidence presented comes mostly from the Najahid (1021–1158)
and Rasulid (1229–1454) eras of Yemen’s medieval history. While offering a
detailed account of slavery in these two Yemeni polities, this chapter will also
discuss the phenomenon of slavery throughout medieval Arabia more broadly,
highlighting temporal and geographical continuities and differences.
M. Moorthy Kloss (B)
Institute for Social Anthropology, Austrian Academy of Sciences, Vienna, Austria
e-mail: magdalena.moorthy-kloss@oeaw.ac.at
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_8
139
140 M. MOORTHY KLOSS
Entry into Slavery
In the early 2000s, the Yemeni scholar Muh
. ammad J¯azim published a fasci-
nating collection of administrative documents from late thirteenth-century
Yemen that had survived in a private library in Sana’a. Known as N¯ur al-
ma ‘¯arif (“The light of knowledge”),2 this source provides rare information
on slave trading practices across the Red Sea to Yemen, including routes and
procedures, as well as prices and taxes paid for different categories of slaves.
The following passage from N¯ur al-ma ‘¯arif is part of a broader section on
trade between Ethiopia and Yemen. Tellingly, slaves are listed here among
many other “products” to be evaluated, sold, and taxed:
The good eunuch (al-kh¯adim al-jayyid) is expensive up to a hundred wiq¯ıya [a
common weight measure] in coins, and that is precious. […] The eunuch of
medium quality is for 50 or 60 wiq¯ıya. The regular eunuch is for 40 wiq¯ıya.
The uncastrated slaves (al-
ab¯ıd al-fuh
. ¯ul ): the good slave is the pure Ethiopian
slave boy (al-was
.¯ıf al-h
. abash¯ı al-s
.ar¯ıh
. ), such as the jizl¯ı and the amh
. ar¯ı,orthe
sah
. art¯ı or any kind (jins ) as long as he is flawless. His price in Ethiopia is 20
wiq¯ıya, and the one of medium quality 15 or 14, and the one of lesser quality
12, 11 or 10.
As for female slaves (al-jaw¯ar): the good slave girl of excellent quality (al-
jayyida al-was
.¯ıfa al-
¯al ) is for 20 wiq¯ıya, and of medium quality for 15 or 16,
and of lesser quality for 12 or 10.3
This short passage highlights many of the complexities characterizing the
medieval slave trade to Arabia and thus requires some context. The Rasulids at
whose court this document was written ruled over large parts of lower Yemen
from 1229 until 1454. Of Turkoman descent, they had come to the country
with the Ayyubids, their predecessors, whom they served as high military
and administrative officers. Yemen’s geographical location on both the Indian
Ocean and the Red Sea meant that important maritime trade networks inter-
sected there, linking up with caravan routes leading to Mecca and Medina, the
holy sites of Islam. The Rasulids exploited this strategic advantage by actively
supporting trade through the provision of administrative, infrastructural, and
security services. The state in turn shared in the merchants’ wealth by taxing
them heavily.
According to medieval Islamic law, slaves could be bought and sold like
any other pr operty, although their humanity was acknowledged. In the above
passage, slaves are divided into categories—female, male, and castrated male
slaves (eunuchs)—and ranked by their perceived “quality”—low, medium,
and high. Eunuchs were up to five times more expensive than other slaves,
matching accounts from the Roman, Byzantine and Ottoman empires, all of
which considered eunuchs to be the ultimate luxury possession.4 The reasons
for the high value placed on eunuchs were twofold. The castration procedure,
8 SLAVERY IN MEDIEVAL ARABIA 141
despite being carried out in specialized centers in Ethiopia and elsewhere, was
very dangerous, and many boys did not survive this mutilation. Furthermore,
courts in the Islamic world had come to rely on the service of eunuchs, who
were considered to be the perfect servants. In the Rasulid case, as we shall see
below, eunuchs were important players at all levels of government and admin-
istration. From the perspective of slave traders, eunuchs were thus a high-risk
commodity that potentially yielded high profits.
Ethnicity also impacted a slave’s price: the highest value was attributed to
a “pure Ethiopian slave boy” who was “flawless.” The author of the above
passage distinguished between three different Ethiopian ethnic affiliations—
jizl¯ı, amh
. ar¯ı,and sah
. art¯ı—suggesting he had a fairly nuanced knowledge
about the country’s inhabitants. Ethnonyms used for slaves in Arabic sources
underline the finding that most slaves in medieval Arabia were of African
origin. Yemeni authors writing between the eleventh and fifteenth centuries
mainly used three ter ms to describe African slaves: habash¯ı, denoting individ-
uals from today’s Ethiopia and Eritrea, zanj¯ı, roughly pointing to modern
Somalia and coastal regions further south, and n¯ub¯ı, indicating origins in
today’s southern Egypt and Sudan. The geographical meaning of these terms
was approximate and fluctuated over time. What is more, it appears that many
individuals were captured from populations living further in the African heart-
land, rather than in the coastal areas from where they were later trafficked
to Arabia.5 The exact origins of enslaved Africans who labored in medieval
Arabia therefore remain unknown. A new body of evidence that has recently
been introduced into slavery research are genetic studies. This type of data
shows that female African contributions to the gene pool of the Arabian Penin-
sula are on average triple that of the male contributions, pointing to slavery
and specifically to concubinage (the sexual exploitation of female slaves by
their masters). Most African genetic input into Arabia is of Ethio-Somali and
Nilo-Saharan origin, roughly validating the geographical information given by
medieval Arabic authors.6
N¯ur al-ma ‘¯arif contains several other passages that offer striking insights
into slave trading practices during the Rasulid period. The source shows that
children were captured or bought from populations living in today’s Ethiopia
and Eritrea, and then transported by Ethiopian traders along commercial
routes to Zayla
, a port city and slave trading hub commonly identified with
today’s Saylac in Somalia. There, they were sold to Yemeni merchants with
the help of an Ethiopian intermediary known as naz¯ıl. Yemeni traders then
transported the slaves on boats carrying mixed cargo to Aden, Yemen’s major
port on the country’s southernmost tip, and to the ports of Zab¯ıd, one of
two Rasulid capitals. Upon arrival, some slaves were immediately selected
for government service by state officials and taken overland to Ta’izz, the
dynasty’s second capital, while all others were sold on the public slave market.
A graphic description of s ale procedures on the slave market of Aden was
recorded by Ibn al-Muj¯awir, a thirteenth-century traveler in Yemen:
142 M. MOORTHY KLOSS
Selling slave girls. The slave girl is fumigated with an aromatic smoke,
perfumed, adorned and a waist-wrapper fastened round her middle. The seller
takes her by the hand and walks around the souk [market] with her; he calls out
that she is for sale. The wicked merchants appear, examining her hands, feet,
calves, thighs, navel, chest and breasts. He examines her back and measures her
buttocks in spans. He examines her tongue, teeth, hair and spares no effort.
If she is wearing clothes, he takes them off; he examines and looks. Finally, he
casts a direct eye over her vagina and anus, without her having on any covering
or veil. When he has examined, expressed his approval and bought the slave girl,
she remains with him for about ten days. When [the buyer] has taken care of
her, had his fill, become bored and tired of her and got what he wanted from
her, his lust is at an end. Zayd, the buyer, says to Amr, the vendor, “Indeed,
sir, we have a case to settle in court!” So they attend in front of the judge and
one makes a claim against the other, [suggesting there is] a defect [in the slave
girl].7
Ibn al-Muj¯awir’s distressing account closely matches the advice given in
medieval Arabic slave-buying manuals, which recommend that prospective
buyers inspect slaves’ private parts before making their choice. Doing so in
public went against the principles of Islamic law and morality, but nevertheless
seems to have been quite common in the medieval era. The latter part of Ibn
al-Muj¯awir’s report reveals his penchant for scandal, while also addressing a
specific provision of Islamic law pertaining to the slave trade: the buyer had a
right to annul the acquisition of a slave if a defect was discovered that had not
previously been disclosed by the seller.8 Zayd, the buyer in this story, makes a
mockery of this right by sexually exploiting the enslaved girl he had purchased
and then attempting to return her. The author does not reveal how this legal
dispute ended, but recounts a similar case that was dismissed in court. What is
most striking about this description, however, are not the insights into customs
and legal provisions surrounding the sale of slaves. The main character of this
vignette is the anonymous enslaved girl, who first suffers objectification and
humiliation through the actions of the seller and the prospective buyer, in
full view of the public, and is then sexually exploited by her new owner. As
Hannah Barker has pointed out, slaves were considered objects to be bought
and sold, but at the same time, their human capacity to think and act inde-
pendently was recognized and exploited.9 Themomentofsalemarkedthe
point in the life of enslaved persons when they were almost fully reduced
to commodities and their humanity was suppressed through humiliation and
coercion. Ibn al-Muj¯awir’s account does not tell us anything we do not know
from other sources—enslaved girls and women were sold to satisfy the sexual
appetites of their masters—but the detail and crudity of his description make
the human suffering of this particular girl palpable. In light of this source, the
common scholarly assertion that Islamic slavery was of a relatively harmless
nature becomes difficult to sustain.10
Combined evidence from N¯ur al-ma ‘¯arif and Ibn al-Muj¯awir has allowed
for a vivid reconstruction of slave trading practices between East Africa and
8 SLAVERY IN MEDIEVAL ARABIA 143
Yemen during the Rasulid era. Looking at the medieval Arabian Peninsula
more broadly, it is important to note that slave trading routes were not stable
over time, but fluctuated according to political and economic developments.
Overland slave trading networks in late antiquity connected Syria, Jordan, and
Arabia. In fact, slave raiding practices of nomadic Arabs during that time prob-
ably ser ved as a model to the early Muslim conquerors.11 In pre-Islamic times,
African slaves were shipped to Yemen and then transported northwards to the
Hijaz12 on caravan routes, rather than being “imported” directly through Red
Sea ports in Arabia.13 Mecca was already a major commercial hub before the
advent of Islam, and slaves were among the many commodities that changed
hands on markets in and around the city. Most slaves were brought to the
peninsula from East Africa via the Red Sea. As the Periplus Maris Erythraei
attests, by the first century CE, the trafficking of humans across the Red Sea
was an established practice.14 Slaves were usually carried on mixed-cargo ships
in small numbers, and their acquisition was often arranged through personal
networks rather than on larger slave markets. During the eighth and ninth
centuries, two dynamics spurred the demand for African slaves in Arabia. A
boom in mineral exploitations in today’s Asir region of Saudi Arabia increased
the demand for cheap labor.15 At the same time, local rulers began assembling
slave soldiers (mamluks), a practice first introduced by the Abbasid caliph al-
Mutas
.im (r. 833–842). In Yemen, the Ziyadids (ca. 818–981) were the first
dynasty to establish a mamluk army. According to the Yemeni historiographer
‘Um¯ara b. Al¯ıal-H
. akam¯ı (b.1120 or 1121), the Ziyadids imposed a tribute
on the ruler of the Dahlak archipelago in the Red Sea that included 1000 male
and female slaves identified as Ethiopian and Nubian.16 An eleventh-century
traveler observed enslaved workers from Zanzibar and Ethiopia in the oasis of
Al-Ah
. s¯ain today’s eastern Saudi Arabia, suggesting that African slaves were
not only traded across the Red Sea, but also directly to the Persian Gulf.17 This
evidence is also a rare glimpse into agricultural and rural slavery, both of which
have likely been underestimated by modern scholars due to the urban and elite
focus of most available sources.18 In the twelfth century, the Arab geographer
Muh
. ammad al-Idr¯ıs¯ı describes how Arab traders lured children on the coast of
present-day Kenya with dates, in order to capture and enslave them.19 Around
the same time, letters from Jewish merchants attest that slaves were shipped
from the Horn of Africa to Aden.20
Not all slaves in medieval Arabia were African or of African origin. In
fact, the enslavement of Arabs by Arabs was not uncommon until Muslim
jurists ruled that only persons born into slavery and enemies captured in
warfare could legally be enslaved.21 Although the medieval slave trade from
India to the Arabian Peninsula still awaits thorough scholarly exploration,
the appearance of Indian slaves in sources on medieval Arabia proves its exis-
tence. The Yemeni evidence shows that female slaves from India were prized
as entertainers and concubines.22 In the thirteenth century, the traveler Ibn
al-Muj¯awir reports that enslaved boys (ghilm¯an) were imported from India to
the Yemeni ports of Aden and Al-Shih
. r, even specifying the customs due on
144 M. MOORTHY KLOSS
slaves who were likely Goan.23 Aden also served as an entrepôt for enslaved
Indians destined for Egypt.24 A fifteenth-century source describes the eunuchs
guarding the Prophet Muhammad’s tomb in Medina as being mostly Indian,
but also of East and West African as well as of Byzantine origins.25 Finally,
Turkish slaves are occasionally mentioned in medieval Yemeni sources, and a
handful of references to slaves from other parts of the world further complicate
the picture.26
Slave trading was not the only way in which people wound up as slaves in
medieval Arabia. In pre- and early Islamic times, most slaves on the penin-
sula seem to have been Arab prisoners of intertribal warfare.27 Between the
third and sixth centuries, the Ethiopian kingdom of Aksum suffered a number
of military defeats in South Arabia, likely increasing the African slave popula-
tion there.28 Military expansions during the early Islamic period (seventh to
eighth century) provided many opportunities for the capture and subsequent
enslavement of prisoners of war. Although figures are difficult to extrapolate
from the available sources, the scale of these practices was undoubtedly signif-
icant. Once conquer ed populations had either embraced Islam or accepted
the authority of their Muslim rulers, enslaving them became illegal. In this
respect, it is instructive to briefly consider how the entry into slavery was regu-
lated and restricted by Islamic law. Islam’s view of slavery rests on two main
sources, namely the Quran and the h
. ad¯ıth. The principles stipulated in these
sources are congruent and form the basis of Islamic law (fiqh), which was
codified in the eighth and ninth centuries. Islamic law recognizes two ways
of entering into slavery: only children born to enslaved parents and enemies
captured in warfare could rightfully be slaves, meaning that previous prac-
tices such as debt bondage, enslavement as punishment for crimes, or self-sale
were forbidden. Furthermore, as will be discussed below, medieval Islamic law
promoted manumission and granted the children of concubines free status.
These legal principles caused a steady reduction in the number of slaves but
left open a loophole, since the supply of slaves from beyond the Islamic realm
remained unregulated. Hence, while the normative framework established by
Islamic law arguably sought to reduce the number of slaves in medieval Islamic
societies, it ultimately led to the development of a commercialized slave trade.
Experiences of Slavery
We know that in medieval Yemen, enslaved persons belonged to royal courts
in large numbers and were also found in the households of local elites, but
there is only minimal evidence for slave ownership among the broader popu-
lation. Due to a strong bias toward a historiography of the elites, only slaves
who closely associated with influential individuals appear regularly in narra-
tive sources from medieval Yemen. This imbalance of evidence means that the
lives of eunuchs and concubines can be reconstructed in much greater detail
than those of low-ranking slaves. Administrative documents partly compen-
sate for this shortcoming, offering glimpses into the lives of enslaved menial
8 SLAVERY IN MEDIEVAL ARABIA 145
workers. It is important to consider that slaves in medieval Yemen lived under
vastly different conditions, but that they nevertheless shared the same state
of unfreedom or partial freedom which set them apart from the free popu-
lation. The study of slaves in relatively privileged positions thus also offers
clues on the experience of enslaved persons in medieval Yemen more gener-
ally. Harnessing the complementary strengths of sources from the Najahid
and Rasulid era allows us to gain important insights into the lived reali-
ties of slavery in medieval Yemen, illuminating a broad range of aspects of
this phenomenon. For example, Rasulid works are rich in information on
the lives of eunuchs serving at the Sultanic court. Furthermore, the admin-
istrative documents contained in N¯ur al-ma
¯arif not only describe the slave
trade across the Red Sea, but also offer rare information about slaves at the
bottom of the social ladder. The surviving Najahid chronicle, on the other
hand, offers a detailed account of the lives of (formerly) enslaved girls and
women, a demographic that remains largely unmentioned in Rasulid sources.
Concubines
In the twelfth century, the Yemeni poet and scribe Um¯ara b. Al¯ıal-H
. akam¯ı
recorded a number of anecdotes featuring an enslaved singer known as Warda
(Flower). Trained as an entertainer by her first owner, a well-known slave
trader, she then became the concubine of a high-ranking military commander.
Later, she attracted the attention of an influential vizier who devised an
intricate plot in order to gain possession of her. In the following passage,
Warda’s second owner describes the moment when she appears amid other
enslaved girls to entertain guests at a banquet, and later informs her that she
will be given to the vizier as a present:
Then we order ed [Warda’s] presence, as the tenth of ten [enslaved girls]. They
kissed the hands of the vizier and began to sing in his presence, with uncovered
faces. […]. Then [the vizier] Uthm¯an became dr unk and slept, and the women
became drunk, except Warda. I had wanted her to be alert. I went to the
privy, called for Warda and informed her of the stor y. She said: “I don’t desire
anything except my master.”29
After a number of scenes in which Warda merely appears as the voiceless object
of men’s desire, she finally speaks. Although the above quote is at best a
re-narration by the author, perhaps even mere fiction, it neatly encapsulates
the precariousness inherent in the lives of concubines. Warda has just discov-
ered that she will be handed over from her second owner to her third. Her
statement “I don’t desire anything except my master” expresses her lack of
choice and is masterfully diplomatic, in that it could be read as a statement
of submission to any one of her previous or future owners. Medieval Islamic
law legitimized and strictly regulated concubinage, the sexual exploitation of
female slaves by their male masters. These relations were licit, and the children
146 M. MOORTHY KLOSS
born out of them were free and their father’s legitimate offspring. A concu-
bine who bore her master a child acquired the status of umm walad (literally
“mother of the boy/child”), which meant that she could no longer be sold and
automatically acquired her freedom upon her master’s death. Statistical anal-
yses of prosopographic evidence have recently shown that concubinage, which
was rare in pre-Islamic Arabia, expanded dramatically in the conquest era due
to the ready availability of female captives.30 This dataset also shows that
the number of concubines dropped significantly by the time of the Umayyad
caliph Hisham’s reign (724–43), when the major Islamic military expansion
cametoanend.
31 The practice of concubinage exposed enslaved girls and
women to sexual exploitation, while offering some of them opportunities for
social advancement that were unavailable to other slaves. Al-H
. akam¯ı’s account
reveals that enslaved girls such as Warda were trained by slave traders and
then sold to the elites as entertainers and concubines.32 This trajectory paral-
lels practices in the Hijaz during the Abbasid era. Around the ninth century,
Mecca and Medina became centers for the training of quiy¯an, enslaved girls
and women who were then sold onto the Abbasid elites, whom they enter-
tained with musical performances, poetry, and quick-witted banter.33 For
instance, an early eleventh-century slave trader relates that he bought nine-
year-old girls and trained them for three years each in Medina and Mecca,
upon which they arrived in Iraq perfectly equipped to carve out careers as
quiy¯an.34 What followed—both in Abbasid Iraq and in Najahid Yemen—was
a life of precariousness in which concubines were frequently moved from one
owner’s household to the next. In Warda’s case, she was owned by three men
consecutively (that we know of) before finally gaining her freedom. After the
vizier’s death, Warda is described as choosing her own marriage partner. It
is therefore likely that she had born the vizier a child and was manumitted
according to the umm walad laws. A careful analysis of the available evidence
reveals that Warda displayed remarkable ingenuity enabling her to endure the
coercion of enslavement and to navigate the complex dependencies tying her
to her former and current owners, before she finally attained greater agency
and social standing later in life.
It is thanks to Um¯ara b. Al¯ıal-H
. akam¯ı’s twelfth-century chronicle Al-
Muf¯ıd f¯ıakhb¯ar Zab¯ıd, the only surviving source from the Najahid period,
that we learn about the experiences of concubines like Warda. This work is a
striking description of a society deeply impacted on all levels by enslaved and
freed persons. Concubines and former concubines figure prominently in this
work, allowing for a reconstruction of their lives in some detail. The oppor-
tunities for social advancement open to some concubines such as Warda were
harnessed in an even more impressive fashion by a woman named Alam (d.
1150). Her story is tightly intertwined with that of the Najahid Sultans, a
dynasty founded by Ethiopian slave soldiers who had usurped power from
their masters in 1021 and ruled parts of Southern Yemen until around 1158.
The Najahids routinely took concubines as wives, likely because local elites
shunned them due to their African slave origins. In this respect, Alam’s story
8 SLAVERY IN MEDIEVAL ARABIA 147
is not uncommon. She was an enslaved singer, concubine, and later wife of
the fifth Najahid Sultan Mans
.¯ur (d. 1130), whom she bore a son. What sets
Alam apart was her strong political engagement, which continued even after
the death of her husband and son had deprived her of any direct link to the
throne. The early twelfth century was a period of vicious strife at the Najahid
court, with different factions of enslaved viziers vying for influence while the
sultan’s role slowly grew obsolete. As al-H
. akam¯ı notes, Alam became one of
the most powerful political figures of her time by placing her own male slaves
in strategically important government and military positions:
Men from among the slaves (
ab¯ıd) of the free lady, queen Umm F¯atik b.
Mans
.¯ur [i.e.
Alam] were raised in the palace of king F¯atik b. Mans
.¯ur. They
were S
.aw¯ab, Rayh
. ¯an, Yumn, Azz, and Rayh
. ¯an the Elder. These governors were
dignitaries, important personalities. And among the non-castrated ones [were]
Iqb¯al, Masr¯ur, B¯arih and Sur¯ur. He [Sur¯ur] was the amir of the two parties,
given his capabilities and affluence. This group were those who spoke with the
sultan’s tongue.35
The deployment of trusted slaves enabled Alam to exercise political influ-
ence in spheres that were inaccessible to women, such as the army and
government. Al-H
. akam¯ı mentions elsewhere that Alam had raised some of
these slaves in the royal palace, thereby nurturing their careers from the very
beginning.36 Alam’s social standing is further underlined by the fact that she
became the patroness of the yearly pilgrimage caravan from Yemen to Mecca
and Medina, ensuring its safety through her presence.37 The life stories of
Alam and Warda exemplify broader trends in the biographies of concubines in
medieval Yemen. While they might have enjoyed a relatively privileged lifestyle
compared to the conditions of low-ranking slaves, concubines had to endure
sexual exploitation, usually by several consecutive owners, and their wellbeing
was largely dependent on the goodwill of these men. At the same time, the
biographies of individual concubines show that some of them were able to
incrementally increase their status and independence by bearing their masters
children, making strategic use of their intimate association with influential
men, and using their own slaves to expand their political r each.
Eunuchs
At the end of the fourteenth century, the court historian Al¯ıb.H
. asan al-
Khazraj¯ı recorded the death of Ahyaf, the most influential eunuch of the
Rasulid era:
In this year [1385], the chief eunuch Am¯ın al-D¯ın Ahyaf al-Muj¯ahid¯ıdied. He
was resolute, courageous, tough, wayward, blood-shedding, murderous, crude,
uncouth, resolute, determined, canny, haughty, of great prestige and of severe
spirit. He was brave, fearless in battle, and a good advisor to the sultan.38
148 M. MOORTHY KLOSS
Ahyaf was the figurehead among a number of eunuchs belonging to Rasulid
Sultans who made stellar careers in the Rasulid army and administration. He
served four consecutive Sultans, rising to the very top of the Rasulid mili-
tary apparatus. Al-Khazraj¯ı portraits him as a skilled military commander who
was often charged with heading punitive campaigns against rebellious tribes.
Rasulid sources show that promising young eunuchs were selected by govern-
ment of ficials as soon as they arrived in Yemen, and were then trained by
senior eunuchs in the royal palaces. Eunuchs were thought to display undi-
vided loyalty to their owners, since they had been torn from their families
of origin and were physically unable to start ou their own families later in
life. For this reason, they were considered the perfect servants worthy of their
masters’ trust. In the Rasulid era, the importance of eunuchs in the polit-
ical and military apparatus greatly increased. They supervised slave soldiers,
led militar y campaigns, and acted as governors of cities, castles, and admin-
istrative regions. A number of them even attained the position of amir, the
highest military rank in Rasulid Yemen. Sultans also occasionally sent eunuchs
as diplomatic envoys to Egypt. It is remarkable that enslaved persons of foreign
descent were trusted to return from a mission abroad, rather than simply disap-
pearing. Eunuchs also performed key roles in relation to Rasulid women and
children. Administrative and narrative sources reveal that the households of
Rasulid noblewomen comprised dozens of eunuchs whose roles included safe-
guarding the women’s quarters, educating royal offspring, supervising sultanic
kitchens and storehouses, and likely also training and supervising female slaves.
Furthermore, each Rasulid noblewoman had a eunuch administrator (zim¯am)
who was usually appointed by a male family member and managed her house-
hold. The importance of this relationship is epitomized by the fact that Rasulid
royal women were known not by their given names, but by the name of
their eunuch administrator. The wife of the seventh Rasulid sultan al-Ashraf
Ism¯a‘¯ıl, for instance, was known by the name of Jihat Mutab, after her zim¯am
Jam¯al al-D¯ın Mutab al-Ashraf¯ı. This naming practice powerfully illustrates the
role that eunuchs played in representing their mistresses in male-dominated
spheres beyond the royal residences. Rasulid sultans often appointed their
former educators and tutors as their wives’ administrators. In this way, eunuchs
accompanied their royal masters from childhood throughout their adult life,
serving their wives and children as well. It is easy to imagine that these inti-
mate responsibilities also enabled eunuchs to exercise a subtle influence on the
ruling family. In sum, eunuchs were the only slaves serving in all spher es of
Rasulid private and public life—managing the households of royal women,
educating royal children, supervising male and female slaves, commanding
mamluk armies, and occupying the highest political positions. These elite
eunuchs must be seen as important players in the Rasulid system of govern-
ment. Their potential success, however, came at the price of physical mutilation
and was contingent on their absolute loyalty to their masters.
8 SLAVERY IN MEDIEVAL ARABIA 149
Slave Soldiers
Slave soldiers (also known as mamluks) are among the most commonly
encountered slaves in medieval Yemeni sources. Their deployment by local
rulers is attested from the ninth century onwards. The Najahid dynasty, itself
established by Ethiopian slave soldiers who overthrew their masters, relied on
Ethiopian and Nubian mamluks as well as on mercenaries from Central Asia
known as Ghuzz to fend off rivals.39 During the Rasulid era, the mamluks were
essential in fighting internal and external enemies, but they also destabilized
their masters’ rule through frequent plots and revolts. Despite their centrality
to the medieval history of Yemen, it is impossible to establish the basic facts
about the lives of mamluks during that time. Not only is information on their
training and living conditions in the military barracks virtually inexistent in
the sources, their origins are also shrouded in mystery. The fact that they bore
Turkic names prompts the assumption that they were of Central Asian descent,
as was the case for most slave soldiers serving the Mamluk rulers in Egypt
during the same time period. 40 However, it is also entirely possible that the
Rasulids relied on African mamluks, just as preceding Yemeni dynasties had
done, while giving them Turkic names to mark them off as military slaves.41
The mamluks of Rasulid Yemen not only elude scholarship, but they are also
remarkably often described as revolting against their masters. For example, in
1322, a contingent of Rasulid mamluks rebelled, arrested the young sultan
al-Muj¯ahid and attempted to install his uncle al-Mans
.¯ur Ayy¯ub and later his
cousin al-Z
. ¯ahir on the throne. After these events, it took al-Muj¯ahid a decade
to restore his sovereignty.42
Female Slave Attendants and Domestics
The role of female slaves as attendants and domestics of elite women is amply
attested, even though details on the work and life conditions of these girls and
women remain unknown. As noted above, Islamic legal texts present domestic
slavery as the norm, and it was likely the most common form of slavery in
medieval Islamic societies.43 Because their presence was taken for granted and
considered to be irrelevant to historiography, enslaved domestics hardly figure
in written sources and have thus frequently been underrepresented by modern
historians.44 On rare occasions, however, these women and girls are thrown
into the limelight, allowing us to gain insights into their lives. The story of
Nukhba is a case in point here. She was an enslaved girl (j¯ariya)who worked
in the Rasulid castle of Taizz, which at the time was under siege. Nukhba
had just brought the reigning Sultan Al-Muj¯ahid Al¯ı (r. 1321–1363) water
to perform his ablutions when he was almost struck by a ballista and barely
escaped death.45 It was merely by being the only eyewitness to this event
that Nukhba’s existence was recorded, and that we learn about her service in
the most intimate realms of the Sultanic residence. How many others like her
toiled in the households of the elites and are forever lost to history? While
150 M. MOORTHY KLOSS
Nukhba is described here as serving the sultan, she actually belonged to the
sultan’s powerful mother, queen Jihat S
.al¯ah
. . The households of elite women
comprised large numbers of female slaves. The labor of these slaves enabled
their mistresses to live the largely secluded lifestyle that was expected from
women of high rank and honor. Women-owned domestics were not allowed
to perform sexual work, but anecdotal evidence shows that they were at times
gifted to men and thereby turned into concubines.46 It should also be noted
that although sex with male slaves was forbidden, both male and female slaves
were at risk of sexual exploitation.47
Other Types of Labor Performed by Slaves
It is likely that the labor of enslaved women and girls comprised many tasks
in the household setting, as well as in agriculture and craftsmanship, that
were simply not recorded by contemporary historians and scribes. Even male
slaves who were engaged in menial work are rarely mentioned. Nevertheless,
medieval Yemeni sources do offer occasional clues on those slaves toiling at
the bottom of the social ladder. Scattered hints, most of them from N¯ur al-
ma
¯arif, give some indication of the breadth of slave labor at the time. This
information largely survived because at least some categories of slaves owned
by the Rasulids received salaries and were given gifts during religious festivals.
However, while male slaves at the Rasulid court frequently appear in salary and
gift lists in N¯ur al-ma
¯arif , female slaves are virtually absent, suggesting that
they likely were not paid for their work but were instead maintained within
the royal court. A fascinating exception concerns enslaved girls who received
payment for producing buttons made of silk thread, under the super vision of
a male slave, likely a eunuch.48 A handful of references also attest to the pres-
ence of female cooks and bread-makers in the royal kitchens. The work of male
slaves in these kitchens is much better documented, and the range of salaries
given to them suggests a diversification and hierarchization of tasks performed
by them. The Yemeni sources analyzed also provide meager evidence that
slaves were deployed in farming, especially in cattle husbandry.49 It is likely
that slaves also worked in the royal stables as well as in mining and hunting.50
Furthermore, evidence from Geniza sources, a corpus of documents from the
medieval Jewish community in Cairo, attests to the fact that merchants active
in the Red Sea and the wider Indian Ocean world owned male slave agents
who supported their business endeavors, sometimes even traveling abroad to
trade on their behalf. The same practice existed among the K¯arim¯ı, a group
of merchants active in the trade between Egypt and the Indian Ocean during
the Ayyubid and Mamluk periods.51 Due to a lack of sources, the level of
exploitation and violence suffered by slaves working in kitchens and work-
shops, on fields and ships is impossible to gage. The available evidence only
paints a partial picture that obscures almost every detail about the lives of these
children, women, and men.
8 SLAVERY IN MEDIEVAL ARABIA 151
Exits from Slavery
According to medieval Islamic law, slavery was considered to be a legal state,
not an innate characteristic of the slave, and could therefore be reversed
through manumission. The freeing of slaves was encouraged as an act of
piety and recognized as a way of atoning sins, such as perjury or the viola-
tion of the Ramadan fast. Detailed legal provisions distinguished different
kinds of manumission, such as a slave owner’s testamentary provision that her
slaves be freed upon her death, or contractual manumission, in which slaves
bought their freedom for an agreed sum of money paid in installments. In the
foundational texts of Islam, manumission was understood as a way for slaves
to regain their freedom and become full members of the umma; for slave
owners, it offered the opportunity to perform a good deed and atone for sins.
However, the way manumission developed in Islamic legal thought and lived
social practice effected slaves in ways that were more complex. Manumission
in the medieval Islamic context was not simply the reversal of enslavement.
Rather, it is better understood as a long journey on which the slave trav-
eled toward greater freedom, gaining more agency and rights along the way.
The promise of freedom could also be used by masters to command obedi-
ence from their slaves, thereby becoming a tool of suppression rather than
liberation. Not all slaves who were on a path to manumission were eventu-
ally freed. For example, if a master’s testament stated that his slaves be freed
after his death, his heirs could still sell off these slaves to pay outstanding
debts. Even achieving the status of freedman or freedwoman did not constitute
full freedom in any modern sense. Rather, freed slaves remained connected
to their former masters through bonds of clientage (wal¯a
). This relationship
was passed on from generation to generation, binding the families of former
masters and former slaves together through responsibilities and rights toward
each other. Wal¯ais modeled after kinship relations and is based on protection
in exchange for submission. At its best, this relationship ensures that former
slaves and their descendants are integrated into mainstream society through
a connection to powerful patrons; at its worst, it can perpetuate exploitation
and dependency for generations to come. Surprisingly, manumission is rarely
discussed in medieval Yemeni sources. Exceptions include the fourteenth-
century report that the Rasulid queen Jihat Tagh¯a manumitted upon her death
large numbers of slaves,52 and a mention of the freeing of a slave to atone for
asin.
53 Biographical dictionaries occasionally feature manumitted slaves who
became saints or merchants. For example, Ab¯ual-D
. iy¯aJawhar ibn Abd All¯ah
al-S
. ¯uf¯ı was a slave who engaged in trade on behalf of his master, a merchant.
He was later manumitted and became a famous Sufi saint.54 If one traces the
lives of individual concubines through al-H
. akam¯ı’s chronicle of the Najahid
era, it becomes clear that they were indeed manumitted—either after their
master’s death, as prescribed by Islamic law, or earlier. Whether and when
the high-ranking eunuchs that figure so prominently in medieval sources on
Yemen were manumitted remains a mystery. Only one eunuch named Niz
. ¯am
152 M. MOORTHY KLOSS
al- D¯ın Mukhtas
.s
. (d. 1267) is known to have been freed by his master.55 This
remarkable silence around manumission could mean that it was rarely practiced
in medieval Yemen, which however would be unlikely given its centrality in
Islamic legal texts. Alternatively, it might have been such a standard occurrence
that it deserved little mention by court chroniclers and scribes.
Conclusions
Slavery has long constituted a blind spot in the study of medieval Islamic soci-
eties, but a number of important works published since the 2000s have greatly
contributed to our understanding of the phenomenon. Yet, few of these works
focus on slavery in Arabia, where the birth of Islam in the early seventh century
brought about changes in the practice of slavery that continued throughout
the medieval period. Our understanding of slavery in medieval Arabia largely
rests on the work of a few scholars, which are listed in the suggestions for
further reading below. Elizabeth Urban has studied the changing roles of
Muslims of slave origin in the early Islamic community. Majied Robinson has
fruitfully applied statistical approaches to Arab genealogical literature, thereby
illuminating the role of concubinage in the tribe of Muhammad from the
sixth to the mid-eighth century. Hend Gilli-Elewy has analyzed a wealth of
Arabic source material to uncover the origins of slaves living in Hijaz during
the Prophet Muhammad’s lifetime. Chase Robinson has examined practices
of enslavement during the Islamic conquest period. Shaun Marmon’s study of
the eunuchs who guarded the Prophet Muhammad’s tomb in Medina from the
twelfth century onwards offers insights into slavery practices in both Egypt and
the Hijaz. My own research focuses on the roles played by slaves at the Najahid
and Rasulid courts of Yemen. Two recent works by Craig Perr y and Jonathan
Miran offer the larger transregional context for slave trading to Arabia in the
medieval p eriod. Additionally, scholars studying related topics such as historical
commercial connections across the Red Sea and the Indian Ocean, the history
of slavery in Africa, and slavery in Islamic law also offer important insights into
the subject.
The story of how slavery developed in medieval Islamic societies begins in
Arabia. It was here that the religious principles related to slavery were revealed
through the Quran and first put to practice by the Prophet Muhammad and
his disciples. The way in which the early community of Muslims (umma)
related to slaves was both influenced by the region’s previous history of slavery
and constituted a significant break from it. The early umma consisted not
only of free Arabs, but also included slaves and former slaves. A prominent
example among them was Abu Bakra, an Ethiopian slave who joined the
nascent Islamic community as an equal member after having been freed by
the Prophet Muhammad. The h
. ad¯ıth relates that the Prophet Muhammad
did not reject slavery outright, but urged his supporters to treat their slaves
kindly and to manumit them. Urban has argued that the early umma was a
“radically inclusive, faith-based community,” a character that was lost by the
8 SLAVERY IN MEDIEVAL ARABIA 153
time the Islamic scriptural tradition became codified in the eighth to ninth
centuries.56 Slaves were first unconditionally welcomed to the early commu-
nity of believers, but as the growing umma sought to define its boundaries,
it became more difficult for slaves, former slaves, and their descendants to be
considered full and equal members.
The conquest period of the seventh and eighth centuries saw a dramatic
increase in the number of slaves owned by Muslim elites, due to the capture of
large numbers of enemies in warfare. The proliferation of concubinage espe-
cially, which had been a modest phenomenon in pre-Islamic Arabia, would
have far-reaching consequences for the nature of Islamic slavery, as well as
for the composition of Muslim families and households. Islamic law, which
was codified in the eighth and ninth centuries, reflects these developments,
as jurists sought to translate the religious principles pertaining to slavery
defined in the Quran and the h
. ad¯ıth into a legal apparatus that would regulate
lived practice. This apparatus established who could be a slave—only children
born into slavery or enemies captured in warfare—and urged masters to treat
their slaves kindly. Slave owners were obliged to secure the upkeep of their
slaves and provide them with medical treatment. However, no legal sanctions
punishing abusive slave owners were intr oduced. Islamic legal texts presuppose
a setting of domestic slavery in which masters possessed a limited number of
slaves whom they interacted with personally. This scenario was likely the stan-
dard in the formative period of Islam and throughout the medieval period,
although a lack of sources precludes us from gaining a full picture of all
aspects of slavery. The normative framework around concubinage and manu-
mission in particular was elaborated in minute detail, while other aspects of
slavery were largely disregar ded in the legal literature. The legal texts failed to
account for the fact that the reality of slaver y expanded and changed over time
and depending on the geographical context, resulting in a growing mismatch
between norms and practice.57
Yet, in the case of medieval Yemen, the sources prove that people were
aware of the legal framework around slaver y and largely sought to follow it.
Around the mid-eighth century, after the great conquests were completed and
Islamic polities had stabilized, slave trading replaced slave raiding as the main
strategy for the acquisition of new slaves. Most slaves reached medieval Arabia
from East Africa, having been trafficked as children first via overland routes
that connected the African hinterland to the coast, and then across the Red
Sea on mixed-cargo ships. Slaves from India appear regularly in sources on
medieval Arabia, and enslaved individuals from Byzantium and other parts of
the world are also occasionally mentioned. Evidence on the experiences of
slavery in medieval Arabia is scattered and fragmentary. A heavy source bias
toward male elite perspectives means that the lives of most slaves remained
unrecorded. What is however clear is that slaves sustained the lavish lifestyle
of Yemen’s upper classes for centuries. As in other historical and geographical
contexts, slavery in medieval Yemen was highly gendered. This fact is most
starkly on display in the case of concubines and eunuchs, who were valued
154 M. MOORTHY KLOSS
for their sexuality and presumed lack thereof respectively. While concubines’
bodies were exploited to generate pleasure and offspring for their masters,
eunuchs were mutilated to increase their versatility and perceived trustworthi-
ness as servants. Female slaves cooked and cleaned, worked in agricultur e and
crafts, provided entertainment and sexual gratification to their masters, and
bore them children. Male slaves also performed menial, domestic, and agri-
cultural duties, fought and died for their masters, traded on their behalf, and
represented their interests as high army and government officials. Most slaves
lived within their masters’ households, a fact that should not automatically be
considered as limiting the level of exploitation and violence they endured. The
legal avenues to freedom were many, but the extent to which they were avail-
able to individual slaves depended on their status and on the benevolence of
their masters.
Research for this article was conducted within the program “Visions of
Community - Comparative Approaches to Ethnicity, Region and Empire in
Christianity, Islam and Buddhism,” funded by the Austrian Science Fund
(SFB 35). I am grateful to my colleagues at the Austrian Academy of Sciences’
Institute for Social Anthropology for their support throughout this project. I
thank Hannah Barker and Craig Perry for their valuable feedback on the
draft chapter. Special thanks to Juliane Schiel and Damian Pargas for bringing
together this volume and for organizing the writers’ workshop.
Notes
1. For the purposes of this chapter, the t erm “Arabia” will be used to denote
the geographical area of the Arabian Peninsula, and the medieval period is
defined broadly as the time between 500 and 1500 CE. Daniel M. Varisco
and Thomas Bauer have convincingly challenged the term “medieval” in refer-
ence to Islamic history, but I consider the term to be useful in a comparative
context such as this handbook (Daniel M. Varisco, “Making ‘Medieval’ Islam
Meaningful,” Medieval Encounters 13, no. 3 (Sept. 2007): 385–412. Thomas
Bauer, Warum es kein islamisches Mittelalter gab - Das Erbe der Antike und der
Orient (München: C. H. Beck, 2018).
2. Anonymous, N¯ur al-ma¯arif f¯ı nuz
.um wa-qaw¯an¯ın wa-ar¯af al-Yaman f¯ıal-
ahd
al-muz
.affar¯ıal-w¯arif (Lumière de la connaissance. Règles, lois et coutumes du
Yémen s ous le règne du sultan rasoulide al-Muz
.affar), ed. Muh
. ammad Abd
al-Rah
.¯ım J¯azim, 2 vols. (Sana’a: Centre français d’archéologie et de sciences
sociales, 2003).
3. Anonymous, N¯ur al-ma ‘¯arif , I.362. Medieval Yemeni sources often give the
plural of j¯ariya as jaw¯ar instead of the standard plural form jaw¯ar¯ı.
4. David Ayalon, Eunuchs, Caliphs and Sultans: A Study in Power Relationships
(Jerusalem: Magnes Press, 1999), 63; Ehud Toledano, “The Imperial Eunuchs
of Istanbul: From Africa to the Heart of Islam,” Middle Eastern Studies 20,
no. 3 (1984): 380.
5. Ayda Bouanga, “Gold, Slaves, and Trading Routes in Southern Blue Nile
(Abbay) Societies, Ethiopia, 13th–16th Centuries,” Northeast African Studies
17, no. 2 (2017): 31–60.
8 SLAVERY IN MEDIEVAL ARABIA 155
6. For an overview of this research, see Craig Perry, “Slaver y and the Slave Trade
in the Western Indian Ocean World,” in The Cambridge World History of
Slavery, eds. Craig Per r y et al. (Cambridge: Cambridge University Press, 2021),
149.
7. Y¯usuf b. Ya‘q¯ub Ibn al-Muj¯awir, A Traveller in Thirteenth-Century Arabia: Ibn
al-Muj¯awir’s T¯ar¯ıkh al-mustabs
.ir, ed. G. Rex Smith (Burlington, UK: Ashgate,
2008), 162.
8. Yah
. y¯a b. Sharaf Al-Nawaw¯ı, Minh¯aj al-t
.¯alib¯ın wa
umdat al-muft¯ın,ed.
Muh
. ammad T
. ¯ahir Shab¯an (Beirut: D¯ar al-Minh¯aj, 2005), 220–1.
9. Hannah Barker, That Most Precious Merchandise: The Mediterranean Trade in
Black Sea Slaves, 1260–1500 (University of Pennsylvania Press, 2020), 98.
10. E.g. Jerzy Zdanowski, Speaking with their Own Voices. The Stories of Slaves in the
Persian Gulf in the 20th Century (Cambridge: Cambridge Scholars Publishing,
2014), 50.
11. Noel Lenski, “Captivity and Slavery Among the Saracens in Late Antiquity (ca.
250–630 CE),” AnTard 19 (2011): 237–66.
12. The Hijaz (also spelled Hejaz) is a region of Arabia located in the northwestern
part of what is today Saudi Arabia. The term usually designates the coastal areas
on the Red Sea, r oughly between the Gulf of Aqaba to the North and As¯ır to
the South.
13. Hend Gilli-Elewy, “On the Provenance of Slaves in Mecca during the Time
of the Prophet Muhammad.” International Journal of Middle East Studies 49
(2017): 166.
14. Richard Pankhurst, “Across the Red Sea and Gulf of Aden: Ethiopia’s Historic
Ties with Yaman,” Africa: rivista trimestrale di Studi e documentazione
dell’istituto italiano per l’Africa e l’Oriente 57, no. 3 (2002): 398.
15. Timothy Power, The Red Sea from Byzantium to the Caliphate: AD 500–1000
(Cairo: American University Cairo Press, 2012), 127–32.
16. Um¯ara b. Al¯ı al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, ed. Henry Cassels Kay
(London: Arnold, 1892), 6.
17. Nasir-i Khusraw, Nasir-i Khusraw’s Book of Travels (Costa Mesa: Mazda
Publishers, 2001), 112.
18. Chase Robinson, “Slavery in the Conquest Period,” International Journal of
Middle East Studies 49 (2017): 159.
19. Ab ¯u ‘Abd All¯ah Muh
. ammad b. Muh
. ammad al-Idr¯ıs¯ı. Kit¯ab nuzhat al-musht¯aq
f¯ıikhtir¯aq al-¯af¯aq, ed. Alessio Bombacci et al., vol. 3 (Napoli: Istituto
universitario orientale di Napoli, 1970), 58–66.
20. Shelomo D. Goitein and Mordechai A. Friedman. India Traders of the Middle
Ages: Documents from the Cairo Geniza (India Book) (Leiden: Brill, 2007),
453.
21. Robinson, “Slaver y in the Conquest Period,” 158; Gilli-Elewy, “On the
Provenance of Slaves,” 164–8.
22. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 65; Goitein and Friedman, India
Traders, 10, 481.
23. Ibn Al-Muj¯awir, T¯ar¯ıkh al-mustabs
.ir, 140–3.
24. Perry, The Daily Life of Slaves, 63; Ibn al-Muj¯awir, T¯ar¯ıkh al-mustabs
.ir, 146.
25. Al-Sakh¯aw¯ı in Shaun E. Marmon, Eunuchs and Sacred Boundaries in Islamic
Society (Oxford University Press, 1995), 39.
156 M. MOORTHY KLOSS
26. E.g. Badr al-D¯ın Ibn H
. ¯atim, Al-simt
. al-gh¯al¯ı al-thaman f¯ı akhb¯ar al-mul¯uk min
al-ghuzz bi-l-Yaman, ed. G. Rex Smith (London: Luzac and Co., 1974), 179;
Al¯ıb.H
. asan al-Khazraj¯ı, Al-Uq¯ud al-luluiyya f¯ıta
r¯ıkh al-d¯awla al-ras ¯uliyya
(The Pearl-Strings; a histor y of the Resúliyy Dynasty of Yemen), ed. James W.
Redhouse and Alexander Rogers, 5 vols. (Leyden: Brill and London: Luzac,
1906), II.294.
27. Gilli-Elewy, “On the Provenance of Slaves”, 164.
28. Power, The Red Sea from Byzantium t o the Caliphate, 95.
29. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 81.
30. Majied Robinson, “Statistical approaches to the rise of concubinage in Islam”,
In Concubines and Courtesans: Women and Slavery in Islamic History,eds.
Matthew S. Gordon and Kathryn A. Hain (Oxford University Press India,
2017), 17.
31. Elizabeth Urban, Conquered Populations in Early Islam: Non-Arabs, Slaves and
the Sons of Slave Mothers (Edinburgh: Edinburgh University Press, 2020), 112.
32. E.g. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 38, 72, 78, 88.
33. E.g. Julia Bray, “Men, Women and Slaves in Abbasid Society,” In Gender in
the Early Medieval World. East and West, 300–900, eds. Leslie Brubaker and
Julia M. H. Smith (Cambridge: Cambridge University Press, 2004); Matthew
S. Gordon, “The Place of Competition: The Careers of ‘Ar¯ıb al-Mam¯uniya and
‘Ulayya Bint al-Mahd¯ı, Sisters in Song,” in Occasional Papers of the School of
‘Abbasid Studies, vol. 135 (Cambridge: Orientalia Lovaniensia Analecta, 2004),
62–81.
34. Ibn But
.l¯an in Antonella Ghersetti, Trattato generale sull’acquisto e l’esame degli
schiavi (Catanzaro: Abramo, 2001), 76–7.
35. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 82.
36. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 86.
37. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 71.
38. Al-Khazraj¯ı, Al-
Uq¯ud al-lu
lu
iyya, II.183.
39. Al-H
. akam¯ı, Al-Muf¯ıd f¯ıakhb¯ar Zab¯ıd, 77.
40. In scholarship on Islamic history, slave soldiers are usually referred to as
mamluks, while the eponymous Sultanate (648–922/1250–1517) is capitalized
(Mamluks).
41. Similarly, non-Turkic mamluks in m edieval Egypt were given Turkic names
as markers of group identity. David Ayalon, “Names, Titles and Nisbas of
the Mamluks,” in The Mamluk Military Society, Collected Studies Series 104
(London: Variorum reprints, 1979), 194–95.
42. Al-Khazraj¯ı, Al-
Uq¯ud al-lu
lu
iyya, 1906, II.5 ff.
43. E.g. Al-Nawaw¯ı, Minh¯aj al-t
.¯alib¯ın.
44. Craig Perry, The Daily Life of Slaves and Global Reach of Slavery in Medieval
Egypt, 969–1250 (PhD diss., Emory University, 2014), 68.
45. Al-Khazraj¯ı, Al-
Uq¯ud al-lu
lu
iyya, II.20.
46. For example, the daughter of a Najahid prince tried to evade the sexual
advances of a powerful vizier by handing over to him “forty virgins from among
her jaw¯ar¯ı.” (Al-H
. akam¯ı, 73).
47. Shaun Marmon, “Intersections of Gender, Sex, and Slavery: Female Sexual Slav-
ery”, in The Cambridge World Histor y of Slavery, Volume 2: AD 500–AD 1420,
ed. Craig Perr y, David Eltis, Stanley L. Engermann, and David Richardson
(Cambridge University Press, 2021).
8 SLAVERY IN MEDIEVAL ARABIA 157
48. Anonymous, N¯ur al-ma
¯arif , I.104.
49. Ibn H
. ¯atim, Al-simt
. al-gh¯al¯ı, 110–2; Anonymous, N¯ur al-ma
¯arif , I.373;
Al-Khazraj¯ı, Al-
Uq¯ud al-lu
lu
iyya, II.217; and Ibn al-Muj¯awir, T¯ar¯ıkh al-
mustabs
.ir, 186.
50. Ibn al-Muj¯awir, T¯ar¯ıkh al-mustabs
.ir, 145; Muh
. ammad at
.-T
. ayyib b. Abd All¯ah
B¯a Makhrama, “T¯ar¯ıkh taghr ‘Adan”, in Arabische Texte zur Kenntnis der Stadt
Aden im Mittelalter, ed. Oscar Löfgren (Leipzig: Harrassowitz, 1936), A.9;
and Anonymous, N¯ur al-ma
¯arif , I.294. II.353, I.126, I.175.
51. Goitein and Friedman, India Traders, 66 ff. Shelomo D. Goitein, Letters
of Medieval Jewish Traders (Princeton: Princeton University Press, 1973),
13; S
.ubh
.¯ıLab¯ıb, Handelsgeschichte Ägyptens im Spätmittelalter: 1171–1517
(Wiesbaden: Steiner, 1965), 112, 287, 489.
52. Al¯ıb.H
. asan al-Khazraj¯ı, Al-
Iqd al-f¯akhir al-h
. asan f¯ıt
.abaq¯at ak¯abir ahl al-
Yaman, wa huwa t
.ir¯az a‘l¯am al-zaman f¯ıt
.abaq¯at a‘y¯an al-Yaman, ed. ‘Abd
All¯ah b. Q¯a’id ‘Abb¯ad¯ı et al. (Sana’a: Maktabat al-j¯ıl al-jad¯ıd, 2009), 2502.
53. Al-Khazraj¯ı, Al-
Uq¯ud al-lu
lu
iyya, II.175.
54. Ab ¯u l-‘Abb¯as Ah
. mad b. Ah
. mad al-Sharj¯ı al-Zab¯ıd¯ı, T
. abaq¯at al-khaw¯as
.s
. ahl al-
s
.idq wa-l-ikhl¯as
., ed. ‘Abd All¯ah al-H
. ibsh¯ı(S
.an¯a:Al-d¯ar al-yamaniya li-l-nashr
wa al-tawzi ‘, 1992), 120–21.
55. Al-Khazraj¯ı, Al-
Iqd al-f¯akhir, 1688.
56. Urban, Conquered Populations, 50.
57. Kurt Franz, “Slavery in Islam: Legal Norms and Social Practice”, in Slavery and
the Slave Trade in the Eastern Mediterranean (c. 1000–1500 CE), eds. Reuven
Amitai and Christoph Cluse (Turnhout: Brepols Publishers, 2018), 51–141.
Further Readings
Gilli-Elewy, Hend. “On the Provenance of Slaves in Mecca during the Time of the
Prophet Muhammad.” International Journal of Middle East Studies 49 (2017):
164–68.
Marmon, Shaun E. Eunuchs and Sacred Boundaries in Islamic Society. Oxford: Oxford
University Press, 1995.
Marmon, Shaun E. “Intersections of Gender, Sex, and Slavery: Female Sexual Slavery.”
In The Cambridge World History of Slavery: Volume 2: AD 500–AD 1420,editedby
Craig Perry, David Eltis, David Richardson, and Stanley L. Engerman, 185–213.
Cambridge: Cambridge University Press, 2021.
Miran, Jonathan, “Red Sea Slave Trade.” In The Oxford Encyclopedia of Slavery, the
Slave Trade, and the Diaspora in African History, edited by Martin Klein. Oxford:
Oxford University Pr ess, forthcoming.
Moorthy Kloss, Magdalena. “Eunuchs at the Service of Yemen’s Ras¯ulid Dynasty
(626–858/1229–1454).” Der Islam 98, no. 1 (2021): 6–26. https://doi.org/10.
1515/islam-2021-0002
Moorthy Kloss, Magdalena. Race and the Legacy of Slavery in Yemen. Journal article
in History and Anthropology. London: Taylor & Francis, 2023. https://doi.org/10.
1080/02757206.2023.2164927
Perry, Craig. “Slavery and the Slave Trade in the Western Indian Ocean World.” In The
Cambridge World History of Slavery, vol. 2: AD 500–1420, edited by Craig Perry,
David Eltis, Stanley Enger man, and David Richardson. Cambridge: Cambridge
University Press, 2021.
158 M. MOORTHY KLOSS
Robinson, Chase. “Slaver y in the Conquest Period.” International Journal of Middle
East Studies 49 (2017): 158–63.
Robinson, Majied. “Marriage in the Tribe of Muhammad: A Statistical Study of Early
Arabic Genealogical Literature.” Berlin: De Gruyter, 2020.
Urban, Elizabeth. “Conquered Populations in Early Islam: Non-Arabs, Slaves and the
Sons of Slave Mothers.” Edinburgh: Edinburgh University Press, 2020.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 9
Slavery in the Black Sea Region
Hannah Barker
Kiev
Bursa
Constantinople
Pera Samastro
Sinope
Tre biz on d
Simisso Batumi
Sevastopol
Tiflis
Saray
Saray-Berke
Tan a
Porto
Pisano
Copa
Caffa
Solgat
Cembalo
Soldaia
Kherson
Moncastro
Kilia
Varn a
Kerch
Chios
Caspian
Sea
Sea of
Azov
Aegean
Sea
Hajji Tarkhan
© S.Ballard (2022)
N
D
a
n
u
b
e
R
.
D
n
i
e
s
t
e
r
R
.
D
o
n
R
.
V
o
l
g
a
R
.
D
n
i
e
p
e
r
R
.
Nicopolis Black Sea
H. Barker (B)
Arizona State University, Tempe, AZ, USA
e-mail: hannah.barker.1@asu.edu
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_9
159
160 H. BARKER
Introduction
Slavery pervaded the medieval Black Sea, as it did most parts of the medieval
world, and was practiced ther e in a variety of forms. In order to trace the full
range of slave-related activity in the region, one may begin with Joseph Miller’s
call to treat slavery as a matter of strategic decision-making on the part of
both enslavers and enslaved within specific historical contexts that shaped their
actions and that they intended their actions to shape. In other words, slavery
should be considered both a cause and an effect of historical change, not a
static system of power relations. According to Miller, “the definable and distin-
guishing position of slavers is their marginality;” from this position, slavers
adopt slaving as a strategy “to convert their marginality toward centrality.”1
The experiences of the enslaved are characterized by “isolated helplessness, or
helpless isolation;” their primary strategies, therefore, aim “to overcome their
initial isolation, to make human contacts with whomever they find accessible,
to build committed relationships of whatever sorts, and to defend whatever
connections they manage to make with whatever means may be available.”2
From this perspective, the Black Sea may seem a slaving zone par excellence,
a region on the margins of conquests (Arabs, Magyars, Mongols), empires
(Byzantine, Abbasid, Ottoman, Russian), and trade routes (the northern arc,
the silk roads, the Italian shipping networks). Making slaves and trading them
were certainly strategies that inhabitants of the Black Sea used to center them-
selves and gain leverage over the powers that surrounded them. Yet inhabitants
of the Black Sea also found themselves targeted for enslavement, isolated and
scattered to the far ends of the medieval world in service to the strategies of
others. As Miller intended, this strategy-focused perspective forces us to ask
which people in the Black Sea benefited from slaving and which people were
targeted.
In the chapter that follows, I will outline some of the strategies associated
with slaving in the Black Sea from about 500 to about 1500 CE. Although
the survey format emphasizes commonalities, it is essential to remember that
the strategies of Black Sea slavers changed over time and varied across cultures.
Rus’ merchants of the tenth century lived in a different world than Mongol
soldiers of the thirteenth century, and their slaving strategies differed accord-
ingly. In addition, because the surviving source base is richer after 900 CE,
my survey will skew toward the end of the period in question.
A greater challenge is to address the experiences of the enslaved. Although
the majority of surviving sources from medieval Black Sea were written by
enslavers, there are a few exceptions. I have chosen to highlight three that
describe entrances into, experiences of, and exits from slavery in some detail.
The first was written by Kirakos Gandzakets’i, an Armenian monk and chron-
icler who described his own capture, enslavement, and escape during the
Mongol conquest of Armenia in 1236.3 The second was Johann Schiltberger,
a Bavarian soldier captured at age sixteen during Bayezid I’s victory over
Sigismund of Hungary’s crusading army at the battle of Nicopolis in 1396.4
9 SLAVERY IN THE BLACK SEA REGION 161
Johann served as a military slave for over twenty years, passing from court
to court as a gift or prize until he made his escape via the Black Sea and
recounted his experiences in writing. The third was Giorgio, kidnapped at
age six from the Crimean port of Caffa and enslaved as a domestic in Chios
and Ancona. Ten years later, in 1460, he narrated his story in a petition for
freedom addressed to the government of Siena. 5 Note that although women
constituted the majority of those enslaved within and exported from the Black
Sea, the written record of their experiences is mor e fragmentary than that of
enslaved men. This is unfortunate because it was the fetishization of enslaved
women in the Ottoman harem that led early racial scientists to choose “Cau-
casian” as the generic name for their newly imagined white, and therefore
inherently beautiful, race.
Entrance into Slavery
Free people might become enslaved in the Black Sea in a number of ways.
Violent capture was the most common. Taking captives in war as well as
targeting human beings alongside cattle and other valuables in raids were
widespread practices throughout the medieval world. Every state in the region,
as well as various non-state-based societies and small groups of raiders acting
on their own initiative, shared in this predatory attitude toward the vulnerable.
A few examples will illustrate the point. In the ninth century, Magyars habit-
ually raided coastal Slavic settlements, bringing their captives to the Byzantine
port of Kerch to sell in exchange for brocade, woolen cloth, and other goods.6
Riverine Slavic settlements were attacked by the Rus’, who took their captives
to Khazar and Bulghar entrepots to sell.7 But the Rus’ did not consider them-
selves bound either to the rivers as routes or to the Slavs as victims; in 943,
they carried out a particularly large and violent raid on the south coast of the
Caspian Sea, carrying off people who would normally have been slave-buyers.8
In the twelfth century, as the polity of Kievan Rus’ began to collapse,
Rus’ princes seized and enslaved each other’s subjects in the course of their
infighting. So did their occasional allies, the Polovtsy (also known as Cumans
or Kipchaks), who also took slaves in campaigns against their Turkic neighbors.
Hunter-gatherer groups known as the Ves and the Yughra, living on the Kama
River and in the Belozersk region respectively, raided for slaves to exchange
for swords imported from the Islamic world via the Bulghars.9 The khan of
the Bulghars conducted his slave raids seasonally, “in the winter, [when] the
cold is so intense that wood splits. It is at this season of great cold that the
king sets out on raids against the infidel and captures his women, his sons, his
daughters and his horses.”10 The cold may have prevented his targets from
evading capture or fleeing during the subsequent forced march.
The most notorious slavers of the thirteenth century were the Mongols.
Kirakos Gandzakets’i, an Armenian monk, gave a detailed account of his own
capture during the Mongol conquest of Armenia in 1236. He had been
studying with a senior monk and historian named Vanakan when villagers
162 H. BARKER
fleeing a Mongol unit led by Molar-noyin took refuge in Vanakan’s cave. They
had no food or water, and the summer heat was intense. After several days,
the villagers pleaded with Vanakan to “go and save all of our lives, descend to
them and make friends with them.”11 He agreed and went with two priests to
persuade Molar-noyin that they were “neither soldiers nor lords of goods, but
exiles and foreigners assembled from many lands for studying our religion.”
Then the rest surrendered: “We descended, quaking, like lambs among the
wolves, frightened, terrified, thinking we were about to die, each person in his
mind repeating the confession of faith in the Holy Trinity.” The Mongols gave
the captives water, confined them in a church, then forced them to march for
several days to the main encampment.
In the early fourteenth century, according to a Franciscan friar appointed
bishop of the Circassian port of Sevastopol, raiders were “selling Christians
for a price on market days, where they are dragged with a rope tied from
the tail of a horse to the neck of those who are sold.”12 A Dominican friar
appointed archbishop of Sultaniyyah in the late fourteenth century explained
where these captives came from: Circassian nobles “go out from one village to
another publicly, or else secretly if they can, and violently seize children and
adults of the other village, and immediately sell [them] to merchants by the
sea. And in the same way as the Tatars were accustomed to sell theirs, so too
these wretched people.”13
Unlike the unfortunate Circassian villagers, Johann von Schiltberger was
captured as a combatant in the Battle of Nicopolis in 1396. After the battle,
when Bayezid “saw that so many of his people were killed, he was torn by
great grief, and swore he would not leave their blood unavenged.”14 Johann’s
captor bound all three of his captives with the same rope and brought them
before the emperor, where he was ordered to kill them. “Then they took my
companions and cut off their heads, and when it came to my turn, the king’s
son saw me and ordered that I should be left alive, and I was taken to the other
boys, because none under twenty years of age were killed, and I was scarcely
sixteen years old.” As part of the spoils, Johann was eventually claimed by the
sultan and taken to the imperial palace in Bursa.
Even during peacetime, the inhabitants of the Black Sea were vulner-
able to kidnapping. A tenth-century treaty between the Rus’ and Byzantines
banned the enslavement of shipwreck survivors discovered along the coast near
Kherson; almost five hundred years later, enslavement still threatened ship-
wreck survivors on the Circassian coast.15 In the fourteenth century, Italian
shippers, already involved in the export of slaves, sold their free passengers
too. For example, a group of Tatars who thought they had arranged passage
from Porto Pisano to Caffa on a Venetian ship were instead sold as slaves by
the pilot and two sailors in 1373.16 This case is documented only because
the enslavers were denounced to the Venetian authorities, found guilty, fined,
imprisoned, and banned from future voyages in the Black Sea.
Most kidnappers were not punished. In 1460, a boy named Giorgio testified
in Siena about the circumstances of his enslavement. At the age of six, he had
9 SLAVERY IN THE BLACK SEA REGION 163
been playing on the shore near Caffa with a group of boys. “A ship of Genoese
being in port, it sent a gondola to land with several men and secretly captured
me and another boy, who was with me, age ten years or so, and then we rose
and betook ourselves to Chios of the Levant, and there I submitted, or indeed
my masters assigned me, to one Lorenzo da Richasole of Florence.”17 What
else could a six-year-old do? Although there was a Genoese statute against
enslaving the free inhabitants of Caf fa, Giorgio may not have been aware of it.
Even if he had been, he could not file a petition for freedom until he turned
fourteen. In the end, he presented his petition at age sixteen. The outcome
is unknown, but if he were successful, his original kidnappers would still have
remained anonymous and unpunished.
Non-violent modes of enslavement were spelled out in the various legal
codes that governed Black Sea communities. These included the expanded
redaction of the Russkaia Pravda, the four major schools of Sunni Islamic law,
the ius commune that prevailed in Latin communities, and the Mongol Yasa.18
Marriages between enslaved people were legally and religiously recognized,
and the children of enslaved parents were automatically born into slave status.
In general, the children of an enslaved woman and a free man were consid-
ered free, either immediately (Islamic and Mongol law) or at the death of
their father (the Russkaia Pravda). The possibility of an enslaved man having a
child with a free woman was not acknowledged. Islamic and Mongol law also
permitted the children of enslaved women and free men to inherit, but the
Russkaia Pravda conflicted with Russian ecclesiastical law on this point.19
The exception was the Latin ius commune, which stipulated that a child
must always follow the status of its mother. Free Latin men who wanted to
claim their children from enslaved women therefore found ways to circum-
vent the law. For example, in a letter from 1345, the Venetian merchant
Francesco Bartolomei asked his correspondent to alter the testament of his
deceased brother Petro. Petro had been a merchant in Tana, where “he bought
a slave with whom he slept and so a child was born,” a boy whom they named
Pascuale. A life in slavery was not what Petro wanted for his son, but two
months after the boy’s birth, Petro fell seriously ill. Before his death, Petro
first drew up a testament, then decided to “marry his slave, the mother of the
boy, in the presence of good witnesses… he did this for his soul and because
the boy was legitimate.”20 Yet Tana was experiencing political turmoil, and
Francesco feared that the witnesses to Petro’s marriage might die. He there-
fore asked his correspondent to alter Petro’s testament. Instead of “I leave to
my natural son Pascuale,” could it be changed to say “I leave to my legitimate
son Pascuale”? After all, despite the penalties for tampering with a notarial
document, Francesco argued that this second statement was the truth,21 “and
it is a thing that does not turn to the detriment of anyone.”
People born free could be enslaved through legal means as well. Allega-
tions of child sale were levied against parents in the Black Sea, but internal
evidence to support it is limited.22 Under certain circumstances, free people
could also choose to enslave themselves. Self-sale was possible under the Latin
164 H. BARKER
ius commune. The Russkaia Pravda recognized three forms of self-enslavement
for free men: selling oneself, marrying an enslaved woman, or undertaking
to work as an estate manager or household steward.23 The Yasa forbade free
Mongol men from becoming domestic servants, perhaps to prevent them from
evading military service. Islamic law forbade the sale of free people into slavery,
though in practice there were exceptions. The most famous was Qaws
.¯un, a
young man who traveled to Cairo in the entourage of the daughter of Özbek
Khan of the Golden Horde.24 One day he went up to the citadel, either as a
merchant selling leather goods or in the company of slave traders. There he
encountered the sultan and was persuaded to sell himself into military slavery.
His price was sent to his br other S
.us
.¯un in the Black Sea. Later, when Qaws
.¯un
had become well-established in Cairo, he sent for his brother and cousin and
appointed them military commanders.
Enslavement was also used as a punishment, individual or collective. The
Árpád kingdom of Hungary enslaved clerics convicted of theft, women who
left their husbands three times, and those who could not pay judicial fines.
In the Russkaia Pravda, enslavement was the penalty for various kinds of
debt, including merchants who lost the goods of others through drinking
or gambling; merchants who borrowed money from foreigners and then
went bankrupt; and indentur ed laborers who stole or tried to escape their
contracts.25 Early Muscovite law allowed enslavement for murder and for
thieves on their second offense.26 After the Mongol conquest of Rus’ prin-
cipalities in the early thir teenth century, Mongol tax farmers enslaved those
who could not pay what they owed. This led to a revolt in 1262, after
which enslavement for tax debt seems to have ceased. Nevertheless, in 1348, a
Venetian merchant was threatened with enslavement by Tatar authorities after
having been imprisoned twice for debt in Tana.27
Experience of Slavery
Enslaved people in Black Sea societies were used for a wide range of purposes.
In addition to performing domestic, sexual, reproductive, artisanal, agricul-
tural, administrative and military labor, slaves functioned as commodities,
financial assets, and symbols of prestige. Slave ownership was not limited to
wealthy elites. Slaves appeared in urban and rural settings; in sedentary and
nomadic cultures; and in the possession of women and men.
Female slaves who belonged to women were expected to provide them
with personal service, companionship, and assistance in their work. Enslaved
women performed domestic tasks such as preparing food, making and washing
clothes, cleaning, and child care. When the Moroccan traveler Ibn Bat
.t
.¯ut
.a
visited the wives of Özbek Khan of the Golden Horde, he found one cleaning
cherries with fifty slaves and another embroidering cloth with twenty slaves.28
Even the wives of traders and ordinary people had three or four slave atten-
dants to carry the trains of their garments when they went out. He also
9 SLAVERY IN THE BLACK SEA REGION 165
observed that ordinary Mongol women owned male slaves who accompanied
them to the market with sheep and milk to sell.29
In the Black Sea, as in most of the medieval world, an enslaved woman who
belonged to a man was understood to be sexually available to him regardless
of her consent. As discussed previously, any children that she bore him would
most likely be born free and raised as his heirs. As a result, the boundaries
between different kinds of labor could become blurred. Nursing a fr ee infant,
for example, was only possible if an enslaved woman had already given birth to
a child of her own. While the companions of the king of the Rus’ in the tenth
century were each reported to own two slave women, one for sex and the
other “to wait on him, wash his head, and provide him with food and drink,”
a Rus’ merchant would use just one woman for sex, carrying the washbasin,
serving food, and as a commodity to sell.30
Slaves also engaged in agricultural and artisanal work. In nomadic societies,
male slaves herded cows and sheep but not horses. In sedentary societies, slaves
worked on farms. Even the Mongols moved captured farmers to devastated
areas to rebuild them. The princes of the Rus’ used elite slaves to manage
their estates; those slaves could also be authorized to trade on behalf of
their owners. They appear most fr equently in princely testaments, which often
provided for the manumission of enslaved estate managers, household stew-
ards, and treasur ers. Boyar households probably also used slaves as stewards
and estate managers but on a smaller scale.
The production of wax and honey, two major exports of the Black Sea
region, intersected with slavery as well. In addition to estate managers, Rus’
princely testaments mentioned slaves as beekeepers. Ab¯uH¯amid al-Garnat¯ı,
a traveler from Granada who visited Hungary in the mid-twelfth century,
purchased an eight-year-old slave girl whom he set to process honey and wax:
“one day I bought for half a dinar two jars full of honeycomb with its wax and
I said to her: ‘I want you to purify this honey and extract the wax.’ Then I
went out and sat on a bench at the door of the house, where people were gath-
ered. After sitting with them for a while, I went back into the house and saw
five disks of wax as pure as gold and two jars full of liquid honey that seemed
like rose water. The honey had been purified and returned to the two jars, all
within an hour.”31 Finally, in 1360–1361 in the port of Kilia, merchants from
Hungary, Caffa, and Piacenza pledged slaves as surety against the loans which
they used to buy wax and honey for export.32
Rulers had many additional uses for slaves. Female slaves were sometimes
sacrificed during elite funerals, most notably by tenth-century Rus’ leaders.33
Thirteenth-century Mongol commanders absorbed captured units into their
own forces; employed slaves received as tribute from the Rus’ for military
service; and sent captives ahead of their main forces as arrow fodder, to test
the safety of river and swamp crossings, or to inflate the size of their armies.
Johann Schiltberger, the captive taken at the Battle of Nicopolis, served the
Ottoman sultan Bayezid I for twelve years, first running before him, “it being
166 H. BARKER
the custom that the lords have people to run before them,” then riding with
him as part of his guard.34
Eunuchs were comparatively rare and expensive because of the high
mortality rate associated with castration. In the Abbasid and Ottoman courts,
eunuchs acted as gatekeepers between the inner and outer parts of the house-
hold. In the Byzantine court, they mediated access to the sacred person of the
emperor as well as guarding imperial women. According to the sixth-century
Byzantine historian Prokopios, the kings of Abkhazia enriched themselves by
castrating Abkhaz boys and selling them to the Byzantine court until Justinian
sent Euphratas, one such eunuch, to stop the practice.35 The Khazar court
operated on a similar model, with eunuchs attending the king and his wives
and concubines. Mongol khatuns also had eunuchs among their attendants.
But because they were traded as slaves across cultural and linguistic bound-
aries, eunuchs wer e equipped to mediate in other ways too. In the ninth
century, Slavic-speaking eunuchs acted as interpreters between Rus’ merchants
and their customers in Baghdad.36
Slaves, especially those from far-away places, could be publicly displayed
or exchanged as gifts to demonstrate prestige. When the Grand Prince Igor’
received Byzantine emissaries to ratify a treaty in Kiev in 944, he gave them
gifts of fur, wax, and slaves.37 Upon the occasion of Berke Khan’s conver-
sion to Islam, the Mamluk sultan Baybars sent him gifts including slaves of
African origin and enslaved cooks. Berke’s successors in the Golden Horde
reciprocated with gifts of Black Sea slaves. 38 When elite women married, their
dowries included people as well as livestock and goods. The human dowry
or inje of a Mongol khatun might include slaves and domestic servants as
well as a share of her father’s free subjects who formed part of her retinue.39
According to the fourteenth-century traveler Ibn Bat
.t
.¯ut
.a, the wives of Özbek
Khan of the Golden Horde had retinues of several hundred slave soldiers; four
hundred slave girls; three hundred slave boys; ten or fifteen Greek and Indian
eunuchs; and eight or ten slave girls as attendants.40 His third wife Bayal¯un,
herself a Byzantine princess, had Nubian slave women in her retinue as well
as Greeks and Turks.41 Özbek Khan’s daughter, in turn, brought hundreds
of Black Sea slaves to Egypt when she arrived to marry the Mamluk sultan
al-N¯as
.ir Muh
. ammad ibn Qalaw¯un.
The Mongols were known to seek out educated people and skilled artisans
among their captives. Those who learned the Uighur script were incorpo-
rated into the bureaucracy. For instance, when the Armenian monk Kirakos
and his fellow captives reached the Mongol encampment in 1236, “they took
me from my companions to serve their secretarial needs, writing and reading
letters. During the day they made me travel with them and in the evening they
would bring us to the vardapet [Vanakan], with a pledge.”42 Poets, musi-
cians, chemists, astronomers, physicians, and others with unusual skills were
often sent to the capital cities of the Mongol empire. Perhaps the most inter-
esting example of this phenomenon was the creation of workshops in which
enslaved artisans produced luxury goods such as nas¯ıj (cloth of silk and gold)
9 SLAVERY IN THE BLACK SEA REGION 167
to be distributed by the Mongol khans. Artisans captured in different areas
were brought together in these workshops, where they learned new techniques
and created distinctive styles.43 These enslaved artisans and bureaucrats were
usually allowed to bring their families or create new ones, receive money for
their work, and enjoy a certain degree of freedom of movement.
Finally, a significant number of slaves moved through the Black Sea as
commercial and financial assets. Soldiers and raiders took people, as they took
silver, gold, and cattle, in the expectation that these forms of loot could be
easily converted into money through sale or ransom. After a raid or battle,
captives were divided among the participants to compensate them for their
work. Slaves could be used in lieu of money to make other payments too. Until
1262, Russian vassals paid tribute in silver, fur, and slaves to their Mongol
lords. In 1415, the scribe of a Venetian ship in Tana used a female slave to
make partial payment on a loan that would come due when he returned to
Constantinople.44 The commander of the Burgundian fleet during the crusade
of Varna in the 1440s used five female slaves to repay money which he had
borrowed in Trebizond to purchase supplies for his galleys.45 As mentioned
previously, slaves could also be pledged as security for loans, such as the silver
borrowed by merchants in Kilia to finance the honey and wax trade. And they
could be rented: in 1448, a woman identified as “Chexum Bicha Usdena, a
Goth or Circassian, called Caterina in our idiom, an inhabitant of Tana” rented
her male slave Semen to a Venetian merchant for 120 bezants per year.46
Merchants tr eated slaves as commodities to be traded for profit, sometimes
over long distances. In the early ninth century, captives taken in northern
Europe passed via Scandinavia and the Baltic Sea into the river systems of
eastern Europe, through the Black and Caspian Seas, and thence to the slave
markets of Constantinople and Baghdad. Rus’ traders dominated the northern
sections of this route, while Khazars dominated the southern sections. The
traveler Ibn Fad
. l¯an observed that when Rus’ merchants reached Khazar settle-
ments, their first act was to offer food and drink to wooden figures of their
gods, praying: “Lord, I have come from a distant land, with such and such
a number of female slaves and such and such a number of sable pelts… and
I have brought this offering… I want you to bless me with a rich merchant
with many dinars and dirhams who will buy from me whatever I wish and
not haggle over any price I set.” 47 In the late ninth century, control over the
southern parts of the route passed to the Bulghars.
At every stage, local raiders added new slaves to the supply and local buyers
purchased some of the slaves passing through. For example, an episode from
the Laxdæla Saga portrays an Icelandic farmer buying an enslaved Irish woman
from Gilli the Russian, a slave trader with a Gaelic name and a Russian epithet,
at an assembly called by the Norse king.48 The woman was very beautiful and
Gilli was reluctant to sell her. Presumably he meant to take her south to the
Volga, where she would fetch a high price, but the Icelander had enough silver
on hand to make an immediate sale worthwhile.
168 H. BARKER
The Mongol propensity to move slaves around their empire has already
been discussed. Thus, people captured in eastern Europe were taken long
distances through the Silk Road trade network, and vice versa. Indian eunuchs
appeared in the retinues of the khatuns of the Golden Horde, and an Indian
girl was sold in Caffa in 1289.49 In 1302, a woman identified as Chinese (de
partibus Catajo) was sold in Genoa.50 On the other hand, William of Rubruck,
a traveler to the court of Möngke Khan in the mid-thirteenth century, met
a woman there, a slave of one of the khatuns, who had been captured in
Hungary and found a husband among the other slaves, a Ruthenian who built
houses.
Finally, the long-distance trade in slaves between the Black Sea, Egypt, and
Africa during the thirteenth through fifteenth centuries deserves comment.
The close relationship between the Golden Horde and the Mamluks, including
the exchange of slaves as gifts, has already been mentioned. Those exchanges
brought small numbers of African slaves, male and female, to the Mongol
court, where their presence undoubtedly contributed to the khan’s prestige
by illustrating the long range of his influence. But slaves moved in the other
direction as well. The Mamluk rulers themselves were former slaves, many with
origins in the Black Sea. Baybars, the first Mamluk sultan, had been captured
as a child during the Mongol conquest of the Kipchak steppe and sold to the
Ayyubid sultan of Egypt. From the markets of Egypt, slaves from the B lack Sea
region were exported as far away as the West African kingdom of Mali, where
the ruler’s retinue included about thirty military slaves, “Turks and others
who are bought for him in Egypt.”51 As in the Mongol court, the presence
of “exotic” slaves confirmed the ruler’s prestige by displaying his power over
distant peoples.
Rulers surrounding the Black Sea treated the long-distance slave trade as a
rich source of tax revenue. Slavic kings taxed Rus’ slave traders by taking one of
every ten slaves.52 In the second half of the fourteenth century, the Genoese
began to collect a head tax on slave sales and possession in their principal
colony of Caffa, generating annual revenues of 13,666 aspers (in 1465) to
219,332 aspers (in 1446).53 They also taxed slave sales and possession through
the por t of Copa. At the same time, Genoa created the Office of Saracen
Heads of St. Antony to tax Muslim travelers. In this way they raised significant
sums by taxing, among others, Muslim merchants taking Muslim slaves to the
Mamluks.
Exit from Slavery
The status of people enslaved in the Black Sea could be changed in several
ways, but only a few of them were within the enslaved person’s control.
Change of status was an individual matter; enslaved groups were usually
not given their freedom collectively. Also, in almost all cases, release from
slavery was not automatic but required a conscious act of intervention. The
9 SLAVERY IN THE BLACK SEA REGION 169
following discussion will therefore focus on the different actors and types of
interventions that could change an enslaved person’s status.
Perhaps the most common type of status change was manumission. This
act could be initiated only by the enslaved person’s owner, but all Black Sea
societies recognized the concept of manumission and the right of slave owners
to perform it. Some legal systems (Rus’) enacted manumission by oath, while
others (Latin) enacted it in written documents. Manumission could also be
performed posthumously through a testament.
Some manumissions were unconditional. Such acts might express piety,
gratitude, or celebration. Tomaxius Zariexa, a Venetian inhabitant of Tana in
1407, freed two slave women, Agnes (formerly Saraimelich) and Magdalena
(formerly Suer), who seem to have been the mothers of his two daughters,
“for the remedy of my soul.”54 In 1436, Antonelus Crescono freed his slave
Magdalena with bequests of money and land for her dowry.55 Muscovite
princes often manumitted enslaved estate managers and their families in
testaments for pious reasons.
Other manumissions imposed conditions of varying severity and complexity.
Hungarian manumissions usually required the enslaved person to repay their
price or make regular gifts to a church. In 1290 in Caffa, Iacobus was freed on
the condition that he serve Stephanus the Armenian and his wife for the dura-
tion of their lives.56 In 1362 in Tana, a Venetian stipulated that his slave Aza
should be freed if she agreed to become Christian; otherwise, he instructed
his executors to sell her at their discretion.57
Another common method for changing an enslaved person’s status was
ransom. The distinction between captivity and slavery was blurry, but as a
rule of thumb, captivity was understood by both the captive and the captor
to be temporary, while slavery was understood by both the enslaved and the
enslaver to be permanent. Turkic languages made a distinction between slaves
that could be sold and slaves held as political hostages, pledges, or prisoners.58
However, these understandings sometimes turned out to be mistaken. Thus
people who believed themselves permanently enslaved were sometimes freed,
while people who expected to be ransomed sometimes fell into permanent
slavery.
Captives and slaves could attempt to facilitate their own ransoms by writing
to family members, business partners, and state agents, though there was no
institutionalized system of ransom as in the western Mediterranean or during
the Ottoman-Russian period.59 Nicholeto Gata, a Venetian merchant, was
threatened with sale by Mongol authorities after having been imprisoned twice
for debt in Tana. He turned to his business partners for help, believing that
he could settle his affairs for 20 sommi.60 Maria, a Russian woman enslaved
by a Venetian in Tana, contacted her brother Samuel. By the time he made a
down payment on her ransom, however, three years had passed and Maria had
given birth to her enslaver’s daughter. He refused to release her until she had
nursed the child for an additional two years.61 For unexplained reasons, an
170 H. BARKER
Italian crossbowman and a shopkeeper named Usayno (Husain) the Saracen
collaborated to ransom an enslaved Alan man in Tana in 1451.62
If no one agreed to pay a ransom or if the captor refused to accept it,
then that route to freedom was closed. Returning to the story of Kirakos, the
Armenian monk captured and employed as a scribe by the Mongols, a ransom
by pious Armenians was arranged for his teacher Vanakan but not for Kirakos
because “we [the Mongols] need him to read and write letters. No matter
what sum you offer, we will not give him up.” Molar-noyin instead offered
Kirakos a wife, his own tent, and a horse to help reconcile him to slave status.
But Kirakos was not satisfied and ended up regaining his freedom through
escape.
When large groups of people were taken en masse, especially during a mili-
tary conflict, state officials might pay their ransom or negotiate their release as
part of a peace treaty. On the other hand, some captors refused ransoms, either
to make a political point or because they needed to raise money more quickly
than the ransom process would allow. Aleksandr Nevskii, saint and prince first
of Kiev and then of Novgorod during the Mongol invasion, made great efforts
to ransom the Rus’ taken captive by Batu’s army. Italian merchants captured by
Janibek, khan of the Golden Horde, in Tana in 1343 were released four years
later as a result of peace negotiations. Yet when Timur (Tamerlane) conquered
Tana in 1395, he preferred to keep its Italian residents as slaves and refused
to accept a ransom for them. The Ottomans agreed to release captives taken
during their conquest of Caffa in 1475 in a context of territorial expansion,
but not after their victory at the battle of Nicopolis in 1396 in the context of
holy war.
The third method of changing an enslaved person’s status, escape, could
be initiated by the enslaved without the cooperation of their enslavers. Most
records of escape in the Black Sea concern people who crossed jurisdictional
boundaries. This makes sense both as a strategy of the enslaved to gain free
status and as an artifact of the archival process, since escape across jurisdictions
generated conflict and paperwork.
States around the Black Sea addressed the potential for conflict over fugitive
slaves in their treaties and legal codes. Slaves who escaped from Rus’ owners
or merchants in the Byzantine territory would be returned, and vice versa.63
Within Rus’ territory, the escape of a slave was to be announced in the market
place. If the slave had taken refuge with someone, that person had three
days from the time of the announcement to return the slave.64 People who
voluntarily returned fugitive slaves to their owners were rewarded. Otherwise,
there were detailed provisions governing the reclamation of fugitive slaves and
punishment for those who helped them.65
Slaves who fled from Caffa to Solgat, the Golden Horde’s regional capital in
Crimea, and vice versa, were covered by a treaty dating to 1380–1381.66 They
should be returned to their owners for a fee of 35 aspers, with any disputes to
be adjudicated by the Genoese consul of Caffa. Slaves who escaped to Caffa
“from the countryside or the Ordo, but not from Solgat” entered a grey area.
9 SLAVERY IN THE BLACK SEA REGION 171
They were understood to be free, but if their owners managed to find them,
“the sindicatori are held to sell the said slave at a public auction and give his
price to the said master of the said slave.” Slaves who escaped within the city
of Caffa could take refuge in the home of the bishop. The bishop was required
“to give notice to the sindicatori [Genoese authorities] without delay… and
to baptize them within three days, then to present them immediately before
the said sindicatori, who ought to sell such male and female slaves and pay
their proceeds to their masters.” Thus, at least in theory, escape within the
city of Caffa or its immediate hinterland entailed a change of enslaver but not
of status.
I have found one case in which this law was tested. In 1450, the Genoese
consul in Tana, Iohannes Spinola, issued a decision in the case of “a certain
Ivan, Ruthenian or Russian, about twenty years old, formerly the slave of a
certain Saracen named Ari Gozza, baker, inhabitant in this place of Tana,”
who had appeared before him to be resold.67 “On account of [Ivan’s] zeal
for Christianity, which he had enjoyed from his youth, born from worship-
pers of Christ, and… marked by baptism and the sign [of the cross],” he
had fled from his Muslim owner to Erasmo Salomone, the Franciscan chap-
lain for the Genoese in Tana. Under Erasmo’s protection, he took the name
Franciscus “as is customary.” Then the speaker (it is not clear whether this
was Ivan/Franciscus, Erasmo, or someone else) requested that the consul,
“according to law as much as custom, attributed to him in such things and
similar things… carry out and manage the sale of his slave.” In other words,
the friar, acting in place of a bishop, had fulfilled the same legal obligations
that applied to fugitives between the Tatar and Genoese communities in Caffa.
The consul in Tana, however, decided not to sell Ivan/Franciscus. Instead, he
granted him an unconditional manumission. Although failing to compensate
a slaveowner for the loss of his slave might have had serious repercussions, in
this case it apparently did not.
Enslaved people being exported sometimes found opportunities to escape
while in transit, a situation in which the forms of coercion and surveillance that
enslavers normally used to control slaves’ behavior might slip. In 1395, the
Dominican bishop of Caffa arranged for two slaves, a man, and a woman, to be
shipped to a contact in Genoa, but after they had crossed to the southern coast
of the Black Sea, the man jumped from the ship and fled inland.68 In 1437,
two male slaves in transit from Caffa to Genoa also seized the opportunity
to flee along the southern coast when plague broke out on their ship.69 The
merchant charged with their transport searched for three days and nights but
could not find them.
Other slaves sought ships to carry them away from their places of enslave-
ment. After two decades as a military slave, Johann Schiltberger escaped with
four companions at a moment when their owner was fleeing into political exile
along the Circassian coast.70 The fugitives first headed to the port of Batumi
and “begged that we should be taken across [the sea], but it was not granted
to us.” Then they rode along the coast for four days until they saw a cog
172 H. BARKER
about eight miles out to sea. That night they made a fire, and the captain sent
a skiff to investigate. “They asked what sort of people we were? We said we
were Christians, and were made prisoners when the king of [Hungary] was
defeated at Nicopolis, and had come so far with the help of God; therefore,
might we not go over the sea, as we had dependence and hope in God, that we
should yet return to our homes and to Christianity. They would not believe us,
and asked if we could repeat the Pater Noster, the Ave Maria, and the Belief?
We said, ‘Yes’, and repeated them. They then asked how many of us there
were? We said, ‘Five’. They told us to wait on the mountains.” In the end the
captain agreed to take them to Samastro and eventually to Constantinople. A
Turkish boy who escaped from the Moroccan traveler Ibn Bat
.t
.¯ut
.amay have
had a similar strategy; he fled from Saray Berke toward the Caspian coast at
Hajji Tarkhan but was recaptured after three days.71 An even more daring
Tatar man named Georgio escaped from Bussana, a small port west of Genoa,
with the intention of traveling all the way back to the Black Sea.72 It is not
clear whether he succeeded.
Finally, enslaved people were sometimes able to gain freedom by petition.
This process could be initiated by a slave, but the final decision concerning
freedom rested in the hands of a judge. The petition of Giorgio, the boy
kidnapped from the coast near Caffa, has already been mentioned. Another
case was that of Cecilia, “the daughter of Theodorus the Greek, an inhabitant
of Caffa.”73 Her petition was phrased in a distinctive way: she instructed her
representative “to proclaim freedom on her behalf against Georgius Stella,
notary, who, as she asserts, is striving to hold her as a de facto slave.” Whoever
was responsible for this wording was careful not to use any phrase that would
reify Cecilia’s enslavement or imply even the most provisional acceptance of it.
Conclusion
Slavery i n the medieval Black Sea had an enduring legacy. The Ottoman
takeover of port cities during the 1470s reoriented the slave trade again, this
time to serve the needs of the court at Istanbul and the broader empire.
Crimean Tatar raids on the Polish and Ukrainian populations increased. As
the emerging Russian state challenged Ottoman control of the north coast of
the Black Sea and the Caucasus, these rivals developed a system for managing
the ransom and exchange of captives. Meanwhile, Mediterranean slaveholders
from Egypt to Iberia turned decisively toward Africa, East and West, for
domestic slaves and eventually for military slaves.
It was in this context that one of the early racial scientists, Johann Friedrich
Blumenbach, selected the skull of an enslaved Georgian woman as his exem-
plar of white beauty and therefore supremacy.74 Womenfromthisregionhad
been enslaved, sexually exploited, and fetishized both locally and globally for
centuries before the Ottoman or Russian empires existed. Men from the Black
Sea had also been enslaved for centuries, sought out especially as eunuchs for
9 SLAVERY IN THE BLACK SEA REGION 173
the Byzantine court and as soldiers for the Mamluk army. Yet by the eigh-
teenth century, western Europeans had come to focus on enslaved women,
especially Georgians and Circassians, in association with the Ottoman harem.
They became objects of orientalist sexual fantasy, most famously in artistic
depictions of odalisques but also in verbal descriptions composed by European
travelers to the Caucasus. These attitudes were reinforced by the immersion
of educated western European men in ancient Greek texts, like Euripides’
Medea and Aristotle’s Politics, that associated Scythians from the Black Sea
with slavery, sexuality, and barbarism.75
Blumenbach used this fetishization of enslaved Georgian and Circassian
women to promote his racial theories. Although he could not access the
corpses of women from the Ottoman court, he was aware that women from
the Caucasus were marketed in more than one place and managed to procure
a sample for his collection from Russia. In his treatise On the Natural Variety
of Mankind, Blumenbach introduced the skull of “a young Georgian female,
made captive in the last Turkish war by the Russians, and brought to Muscovy.
There she died suddenly, and an examination was made of the cause of death
by Hiltebrandt, the most learned anatomical professor in Russia. He carefully
preserved the skull for the extreme elegance of its shape, and sent it to St
Petersburg to de Asch,” who forwarded it to Blumenbach at the University
of Göttingen.76 The image commissioned to accompany the skull was “an
Eastern scene… the whole breathing as much as possible the finest volup-
tuousness… [like] something out of Niebuhr’s Travels or still mor e precisely
out of Russel’s Natural History of Aleppo.”77 In his note stored with the
skull, Blumenbach wrote that it confirmed “the beauty of the Georgians…
[and] the delightful charms of their women.”78 In this way, by drawing on the
fantasy of the beautiful enslaved Georgian woman to center his classification
system within the eighteenth-century scientific discourse on race, Blumenbach
transformed the geographical term “Caucasian” into a generic term for white
people.
The association between “Caucasian,” whiteness, and beauty emerged from
a web of enslavers’ strategies: the writers and artists who used sexualized depic-
tions of Circassian and Georgian women to attract an audience; the Russian
soldiers who demonstrated their power over both Georgians and Ottomans
by taking this particular Georgian woman captive; the anatomy professor
in Moscow who honed his expertise by dissecting her body; the baron in
St. Petersburg who cultivated his scholarly network by sending her skull to
Germany; and Blumenbach, the professor who presented her skull as the
elegant centerpiece of his racial classification system. In turn, the association
between “Caucasian,” whiteness, and feminine beauty was used strategically
by enslavers in other contexts to justify their oppression of other groups in
Blumenbach’s racial hierarchy, especially black Africans.
Recentering this story on the enslaved Georgian woman rather than her
skull highlights the gap between the racialized meaning of “Caucasian” and
the reality of people targeted for enslavement in the Caucasus. We know
174 H. BARKER
nothing about this woman’s name, parents, self-ascribed identity, or the
community in which she grew up. Her experience of slavery is also obscure:
a violent experience of capture, a journey from Georgia to Moscow, a sudden
death caused by venereal disease,79 and a ghost value attached to her bones
which made them vulnerable to public display as late as the 1980s.80 The
identity of this woman as an individual, her unique life story, was erased by
Blumenbach to make her a fitting representative of his newly invented group.
Yet if the characteristic strategy of the enslaved is to seek connection, then at
a minimum we can render this anonymous woman less isolated by connecting
her experiences not only to those of her contemporaries taken captive in the
Russo-Turkish wars, but also to the long history of slaving and enslavement in
the Black Sea.
Notes
1. Joseph Miller, The Problem of Slavery as History: A Global Approach (New
Haven: Yale University Press, 2012), 24, 29.
2. Miller, The Problem, 31, 33.
3. Kirakos Gandzakets’i, History of the Armenians, trans. Robert Bedrosian (New
York: Sources of the Armenian Tradition, 1986), Chapter 24.
4. Johann Schiltberger, The Bondage and Travels of Johann Schiltberger, A Native
of Bavaria, in Europe, Asia, and Africa, 1396–1427, trans. J. Buchan Telfer,
ed. Philip Brunn (Cambridge: Cambridge University Press, 2010).
5. Giulio Prunai, “Notizie e documenti sulla servitù domestica nel teritorio senese
(secc. VIII–XVI),” Bulletino senese di storia patria n.s. 7 (1936): 415–6, doc.
45.
6. Paul Lunde and Caroline Stone, eds., Ibn Fadl¯an and the Land of Darkness:
Arab Travellers in the Far North (London: Penguin Books, 2012), 122–123,
citing Ibn Rusta.
7. Lunde and Stone, eds., Ibn Fadl¯an, 126, citing Ibn Rusta.
8. Lunde and Stone, eds., Ibn Fadl¯an, 145–152, citing Mas¯ud¯ı and Miskawayh.
9. Lunde and Stone, eds., Ibn Fadl¯an, 72, citing al-Garnat¯ı.
10. Lunde and Stone, eds., Ibn Fadl¯an, 67, citing al-Garnat¯ı.
11. Gandzakets’i, History,Chapter 24.
12. Friedrich Kunstmann, “Studien über Marino Sanudo den aelteren,” Konigliche
bayerische Akademie der Wissenschaften; Philosophische philologische und
historische Klasse 7 (1855): 817–8.
13. Anton Kern, “Der ‘Libellus de notitia orbis’ Iohannes’ III. (de Galonifon-
tibus?) O.P. Erzbischofs von Sulthanyeh,” Archivium fratrum praedicatorum 8
(1938): 110.
14. Schiltberger, The Bondage,4–7.
15. Daniel Kaiser, The Laws of Rus’—Tenth to Fifteenth Centuries (Salt Lake City:
Charles Schlacks Jr., 1992), 11; Kern, “Der ‘Libellus’,” 111.
16. Sergei Karpov, La navigazione veneziana nel Mar Nero XIII–XV sec. (Ravenna:
Edizioni del Girasole, 2000), 61.
17. Prunai, “Notizie,” 415–6, doc. 45.
18. Byzantine law is addressed in another chapter. Kaiser, The Laws, xlv, 20–34;
Robert Brunschvig, “Abd,” in The Encyclopaedia of Islam, 2nd ed. (Leiden:
9 SLAVERY IN THE BLACK SEA REGION 175
Brill, 1960), 1:24–40; R. H. Helmholz, “The Law of Slavery and the European
Ius Commune,” in The Legal Understanding of Slavery: From the Historical to
the Contemporary, ed. Jean Allain (Oxford: Oxford University Press, 2012),
17–39; George Vernadsky, “The Scope and Contents of Chingis Khan’s Yasa,”
Harvard Journal of Asiatic Studies 3 (1938): 337–60.
19. Kaiser, The Laws,31vs. 63.
20. Raimondo Morozzo della Rocca, Lettere di mercanti a Pignol Zucchello (1336–
1350) (Venice: Comitato per la pubblicazione delle fonti relative alla storia di
Venezia, 1957), 33–6, doc. 14.
21. This is debatable. A child born to married parents was legitimate. A child born
to parents who could have been married but were not was natural. Marriage
between free and enslaved people was legal as long as the free partner was
aware of the enslaved par tner’s status, but Pascuale’s parents had married after
he was born. It was possible to have a natural child declared legitimate by an
ecclesiastical authority after the fact, and this was clearly Petro’s wish, but he
had not had time to carry it out before his death.
22. Barker, That Most Precious Merchandise, 125–8.
23. Kaiser, The Laws, 32.
24. Khal¯ıl ibn Aybak al-S
.afad¯ı, Kit¯ab al-W¯af¯ı bi-al-Wafay¯at (Wiesbaden: Franz
Steiner, 1962), 24: 277–80; Taq¯ıal-D¯ın Ah
. mad al-Maqr¯ız¯ı, Kit¯ab al-khit
.at
. wa-
al-ath¯ar f¯ıMis
.r wa-al-Q¯ahirah, ed. Ayman Fu’ad Sayyid (London: Al-Furqan
Islamic Heritage Foundation, 2002–2004), 4: 224–6. See also Donald Little,
A Catalogue of the Islamic Documents from Al-H
. aram aš-Šar¯ıf in Jerusalem
(Beirut: Orient-Institut der Deutschen Morgenländischen Gesellschaft, 1984),
380, doc. 683 in which a man named Tugha Timur ibn Karlabi sold himself
and his wife into slavery. Tugha Timur’s father received their price.
25. Kaiser, The Laws, 26–28, 32; Lunde and Stone, eds., Ibn Fadl¯an, 76, citing
al-Garnat¯ı.
26. Kaiser, The Laws, 108, 112.
27. Morozzo della Rocca, Lettere, 118, doc. 63.
28. H. A. R. Gibb, trans., The Travels of Ibn Bat
.t
.¯ut
.a, A.D. 1325–1354 (Cambridge:
Cambridge University Press, 1959), 2: 487–488.
29. Gibb, trans., The Travels, 481.
30. James Montgomery, trans., “Mission to the Volga by Ah
. mad ibn Fad
. l¯an,” in
Two Arabic Travel Books, ed. Philip Kennedy and Shawkat Toorawa (New York:
New York University Press, 2014), 243, 253.
31. Lunde and Stone, eds., Ibn Fadl¯an, 81, citing al-Garnat¯ı.
32. Michel Balard , Gênes et l’Outre-mer (Paris: École des Hautes Études en Sciences
Sociales, 1980), 2: 193–4, doc. 122; Geo Pistarino, Notai genovesi in oltremare:
Atti rogati a Chilia da Antonio di Ponzò (1360–61) (Genoa: Università di
Genova, Istituto di Paleografia e Storia medievale, 1971), 63–65 and 72–73,
docs. 39 and 43.
33. Lunde and Stone, eds., Ibn Fadl¯an, 127, 159; Montgomery, trans., “Mission,”
245–53.
34. Schiltberger, The Bondage,7.
35. Prokopios, History of the Wars, Books VII and VIII,trans.H.B.Dewing
(London: William Heinemann, 1928), book VIII.
36. Lunde and Stone, eds., Ibn Fadl¯an, 112.
37. Kaiser, The Laws, 12.
176 H. BARKER
38. Barker, That Most Precious Merchandise, 71.
39. Evans, “The Womb of Iron and Silver,” 110.
40. Gibb, trans., The Travels, 484–86, also 498.
41. Gibb, trans., The Travels, 488.
42. Gandzakets’i, History,Chapter 24.
43. Thomas Allsen, Culture and Conquest in Mongol Eurasia (Cambridge:
Cambridge University Press, 2001). The earlier Türk state also created commu-
nities of captured weavers and blacksmiths.
44. Nina Prokofieva, “Akti venetsianskogo notariya v Tane Donato a Mano (1413–
1419),” Prichernomorie v srednie veka 4 (2000), doc. 73.
45. Henri Taparel, “Un Épisode de la politique orientale de Philippe le Bon: Les
bourguignons en Mer Noire (1444–1446),” Annales de Bourgogne 55 (1983):
21.
46. Venice, Archivio di Stato di Venezia (ASVe), Cancellaria inferiore, Notai, b.148,
N.6, reg. 2, fol. 39r–v. Notary Pietro Pelacan (1448 May 29). The term uzden
suggests that she was a free peasant from the North Caucasus. Thanks to John
Latham-Sprinkle for this information.
47. Montgomery, trans., “Mission,” 243–5.
48. https://medievalslavery.org/europe/source-anonymous-laxdaela-saga-early-
1200s/.
49. Balard, Gênes, 1:150–151, doc. 388.
50. Genoa, Archivio di Stato di Genova (ASG), Notai Antichi 137, fol. 49v–50r.
Notary Conrado Castello de Rapallo (1302 May 7).
51. N. Levtzion and J. F. P. Hopkins, eds., Corpus of Early Arabic Sources for West
African History (Princeton: Markus Wiener, 2000), 265, citing al-Umar¯ı.
52. Montgomery, trans., “Mission,” 239.
53. Barker, That Most Precious Merchandise, 138–9. Aspers were silver coins issued
by various polities surrounding the Black Sea. In 1436, a Venetian ducat was
worth 35 aspers. Peter Spufford, Handbook of Medieval Exchange (London:
Offices of the Royal Historical Society, 1986), 290. In 1447, slaves auctioned
by the commune of Caffa cost about 1000 to 1500 aspers. ASG, Banco di San
Giorgio, Sala 34, 590/1235 (Caffa Massaria), fol. 21v, 54v, 57v.
54. Sandro de’ Colli, Moretto Bon, Notaio in Venezia, Trebisonda e Tana (1403–
1408) (Venice: Comitato per la pubblicazione delle fonti relative alla storia di
Venezia, 1963), docs. 26, 27.
55. ASVe, Cancellaria inferiore, Notai, b.231, N.3, reg. 1, f.8v–9v. Notary Nicolaus
de Varsis (5 June 1436).
56. Balard, Gênes, 1:347, doc. 846.
57. Charles Verlinden, “Le recrutement des esclaves à Venise aux XIVe et XVe
siècles,” Bulletin de l’institut historique belge de Rome 39 (1968): 195, doc.
116.
58. Golden, “The Terminology,” 42.
59. Tenth-century Rus’-Byzantine treaties did address the possibility of mutual
ransom. Kaiser, The Laws, 6, 10.
60. Morozzo della Rocca, Lettere, 118.
61. Prokofieva, “Akti,” doc. 20.
62. ASVe, Cancellaria inferiore, Notai, b.148, N.6, reg. 2, fol. 60v–61r. Notary
Pietro Pelacan (12 July 1451).
63. Kaiser, The Laws,6,10.
9 SLAVERY IN THE BLACK SEA REGION 177
64. Kaiser, The Laws, 23.
65. Kaiser, The Laws, 33.
66. C. Desimoni, “Trattato dei genovesi col Chan dei Tartari nel 1380–1381 scritto
in lingua volgare,” Archivio storico italiano 20 (1887): 164; Amadeo Vigna,
“Statuto di Caffa: Codice diplomatico delle colonie tauro-ligure,” Atti della
società ligure di storia patria 7, no. 2 (1879): 634–5.
67. ASVe, Cancellaria inferiore, Notai, b.148, N.6, reg. 2, fol. 55v–56r. Notary
Pietro Pelacan (16 Dec. 1450).
68. ASG, Notai Antichi 402, fol. 108r-v. Notary Bartolomeo Gatto (1 Sept. 1395).
69. Domenico Gioffrè, Il mercato degli schiavi a Genova nel secolo XV (Genoa:
Fratelli Bozzi, 1971), 157.
70. Schiltberger, The Bondage, 99–100.
71. Gibb, trans., The Travels, 517.
72. ASG, Notai Antichi 397, fol. 203r–v. Notary Bartolomeo Gatto (31 Oct.
1376).
73. ASG, Notai Antichi 497, doc. 50. Notary Giacomo di Camulio (26 Mar. 1397).
74. Nell Irvin Painter, The History of White People (New York: W.W. Norton,
2010), 43–90.
75. Shelley Haley, “Self-Definition, Community, and Resistance: Euripides’ ‘Medea’
and Toni Morrison’s ‘Beloved’,” Thamyris 2, no. 2 (1995): 177–206.
76. Johann Friedrich Blumenbach, “On the Natural Variety of Mankind, ed. 1795,”
in The Anthropological Treatises of Johann Friedrich Blumenbach, ed. and trans.
Thomas Bendyshe (Boston: Longwood Press, 1978), 162. The war in question
was Catherine the Great’s second Caucasian campaign against the Ottomans.
77. F. W. P. Dougherty, ed., Commercium epistolicum J.F. Blumenbachii
(Göttingen: Niedersächsische Staats- und Universitätsbibliothek Göttingen,
1984), 156.
78. Dougherty, ed., Commercium, 148.
79. Ibid.
80. Dougherty, ed., Commercium, 148, no. 178 is the skull itself, displayed in the
foyer of Göttingen’s university library in 1984. Daina Ramey Berry, The Price
for Their Pound of Flesh: The Value of the Enslaved, from Womb to Grave, in the
Building of a Nation (Boston: Beacon Press, 2017), 7.
Further Readings
Balard, Michel. La Romanie génoise (XIIe – début du XV siécle). Genoa: Società ligure
di storia patria, 1978.
Barker, Hannah. That Most Precious Merchandise: The Mediterranean Trade in Black
Sea Slaves, 1260 –1500. Philadelphia: University of Pennsylvania Press, 2019.
Biran, Michal. “Encounters among Enemies: Preliminary Remarks on Captives in
Mongol Eurasia.” Archivum Eurasiae Medii Aevi 21 (2014–2015): 27–42.
Evans, Nick. “The Womb of Iron and Silver: Slavery in the Khazar Economy.” In
From the Huns to the Turks—Mounted Warriors in Europe and Central Asia. Von
den Hunnen zu den Türken—Reiterkrieger in Europa und Zentralasien,editedby
F. Daim, H. Meller and W. Pohl, 107–115. Halle: Landesamt für Denkmalpflege
und Archäologie Sachsen-Anhalt, 2021.
178 H. BARKER
Golden, Peter. “The Terminology of Slavery and Servitude in Medieval Turkic.” In
Studies on Central Asian History in Honor of Yuri Bregel, edited by Devin DeWeese,
27–56. Bloomington: Research Institute for Inner Asian Studies, 2001.
Langer, Lawrence. “Slavery in the Appanage Era: Rus’ and the Mongols.” In Eurasian
Slavery, Ransom and Abolition in World History, 1200–1860, edited by Christoph
Witzenrath, 145–170. Farnham: Ashgate, 2015.
Noonan, T. S. “Why Dirhams First Reached Russia: The Role of Arab-Khazar Rela-
tions in the Development of the Earliest Islamic Trades with Eastern Europe.”
Archivum Eurasiae Medii Aevi 4 (1984): 151–282.
Perry, Craig, David Eltis, Stanley Engerman, and David Richardson, eds. The
Cambridge World History of Slavery,vol.2, AD 500–AD 1420. Cambridge:
Cambridge University Press, 2021.
Ros
,u, Felicia, ed. Slavery in the Black Sea Region, c.900–1900: Forms of Unfreedom at
the Intersection between Christianity and Islam. Leiden: Brill, 2022.
Tardy, Lajos. Sklavenhandel in der Tartarei: Die Frage der Mandscharen. Translated by
Mátyás Esterházy. Szeged: Universitas Szegediensis de Attila József nominata, 1983.
Zevakin, E. S., and A. Penˇcko. “Ricerche sulla storia delle colonie genovesi nel
Caucaso occidentale nei secoli XIII-XV.” Translated by Maria Teresa Dellacasa. In
Miscellanea di studi storici, 7–98. Genoa: Fratelli Bozzi, 1969.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 10
Slavery in the Western Mediterranean
Juliane Schiel
Introduction
The history of the Mediterranean Sea has always been closely linked to the
history of human trafficking and slavery. From ancient times until today,
people have been transported across the sea against their will to serve others
with their bodies and labor. They have been used as goods, as currency for
other goods, or as gifts for the purpose of establishing and maintaining social
networks and diplomatic relations. The naming and legal framework as well
as the social settings and contexts of exploitation have changed over time.
From slavery in the wealthy city-states of ancient Greece and the Roman
colonate system through the use of enslaved workers in the urban and agri-
cultural economy of Byzantium and the military and imperial slavery system
in early Muslim societies to the import of slaves to Southwestern Europe
as cheap labor or status icons and the ransom economy between Christian
and Muslim parties: The Mediterranean has continually represented a hub for
human exploitation and a central point of reference for interconnected prac-
tices of slaving between the religious and political cultures shaping the region
and linking the ancient to the medieval, the medieval to the modern, and the
modern period to the world of today. This chapter focuses on the social real-
ities of slaves in the late medieval Western Mediterranean by showcasing the
interconnectedness of individual slave biographies with the slaving practices of
J. Schiel (B)
University of Vienna, Vienna, Austria
e-mail: juliane.schiel@univie.ac.at
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_10
179
180 J. SCHIEL
the neighboring empires, as well as the links between these practices and those
in earlier and later periods.
From around 1300 onwards, the number of persons transported across
the Mediterranean Sea to Southwestern Europe as slaves increased signifi-
cantly. While slaves traded over long distances were a rar e and high-priced
merchandise only accessible to the political and economic elites of the Western
Mediterranean prior to the thirteenth century, unfree servants from Central
Asia—and to a lesser extent from sub-Saharan Africa—had already begun to
populate the urban and aristocratic households of the Italian and Iberian
peninsulas. By 1400, between 2000 and 5000 slaves were being sold in Genoa
each year. Every moderately wealthy household included at least one or two
slaves working and living side by side with local domestic servants and wet
nurses, agricultural laborers and artisanal apprentices, carters and messengers.
Members of the ruling classes of Renaissance Europe often employed up to ten
or fifteen slaves under their roofs. Based on fundamental research performed
by early experts in the field, the percentage of slaves in Western Mediterranean
societies during this period is estimated to have been between 1 and 2 percent
in ruralareas,between 3and5percent in theurban centers, andupto10
percent on the Mediterranean islands such as Mallorca, Malta, and Crete.1
The reasons for this sharp increase are still heavily debated among histo-
rians. One of the traditional explanations, the so-called Ehrenkreutz thesis,
has recently been challenged by an interdisciplinary group of researchers. In
his Strategic Implications of the Slave Trade, Andrew Ehrenkreutz depicted the
revival of the European slave trade as a secondary byproduct of power relations
in Central Asia and the Black Sea region.2 According to Ehrenkreutz, when
the rising European maritime powers expanded their trading routes toward the
Eastern Mediterranean and the Black Sea, Genoese merchants came to func-
tion as intermediary shippers between the slave markets on the Black Sea coast
and the imperial ports of the Mamluk state. As Italian merchants provided the
rulers of the Sultanate in Egypt and Syria with thousands of boys and young
men for their armies of slave soldiers, they gradually began to bring the sisters
and mothers of these young recruits back home to the booming urban centers
of the Latin Western Mediterranean. N ew research has now shown, however,
that the so-called Mamluk system relied on the support of Italian shippers
much less than previously assumed, while European participation in the slave
trade followed its own logic from the very beginning.3 Another explanation
provided in relevant literature refers to the Black Death as a major factor in
the increase of slaves from Central Asia and sub-Saharan Africa. Jeffrey Fynn-
Paul, for example, has argued that town dwellers in Catalonia and Italy had
strong economic reasons to keep slaves after 1348, when wages had incr eased
dramatically and profits had stagnated or declined for many urban elites. As
the normative power of the Roman Catholic Church had made Europe a “no
slaving zone” where the enslavement of fellow believers was forbidden, slave
labor had to be imported over long distances from non-Catholic regions.4
Other voices have contested the assumption that the loss of life from the Black
10 SLAVERY IN THE WESTERN MEDITERRANEAN 181
Death played a decisive factor, instead pointing to the dynamics of the late
medieval urbanization and colonization processes that gave rise to a new sex
industry and settlement policy heavily reliant on human traf ficking and slave
trading.5
No matter what the reasons were for the increase in slaves, the late medieval
period brought about a particular type of household slavery that was clearly
distinct from earlier forms of slavery as a means of increasing agricultural
production, as well as from later phenomena such as aristocratic court slavery
and ransom slavery that began to play a role around 1500. Intimately inter-
twined with slaving practices and discourses found in the neighboring Greek
Orthodox and Muslim spheres,6 the late medieval form of Latin household
slavery can be characterized as very ambivalent and fluent, with slaves not
representing a clear-cut social or legal category of its own. The labels used
to designate them, the status applied to them in legal documents, and the
daily tasks they fulfilled overlapped with the ones assigned to other dependent
members of the Latin household. Contemporaries nevertheless knew very well
who worked as a slave and who was an ordinary servant, nurse, serf, or appren-
tice—and in cases of conflict, the fine lines of distinction became apparent
and were sometimes controversially debated in cour t. Unlike other domes-
tics, slaves were bought and sold by oral agreement or written contract, and
their services were more easily rented out to third parties than those of other
dependent household members. Slaves were associated with non-Latin origins
not adhering to the Roman rite. They usually arrived as individuals by sea
and were baptized upon entering the Latin sphere regardless of their orig-
inal religious background. Sexual intercourse, marriage, and the production
of offspring with a person outside the master’s household was forbidden. In
practice, however, it could either be the reason for some of the severest forms
of persecution and punishment by criminal courts or serve as an opportunity
to escape the slave status and leave the master’s household. In general, slaves
in the Western Mediterranean rarely remained slaves for the entirety of their
lives—though almost none of them managed to return home or live their lives
as free citizens of European society.
This ambivalent and fluent position of the household slave in the Western
Mediterranean was embedded in a multi-layered discourse on slavery. On
the one hand, slavery as a legal institution was associated with Muslim
practices that were clearly condemned and needed to be combatted: Any
activity supporting the Muslim slave trade was officially forbidden and penal-
ized. Christian crusaders often legitimated their actions by proclaiming to be
fighting Muslim slavery and freeing or protecting “poor Christians” or “inno-
cent pagans” in the eastern part of the Mediterranean from enslavement.7 On
the other hand, buying slaves on the slave market, converting them to the
Roman Catholic faith, and integrating them into a Christian household was
perceived as a pious act—and sometimes even as a Christian duty. The Church
and the Pope increasingly encouraged merchants and travelers to buy slaves
on their journeys to the East in order to rescue them from Muslim slavery
182 J. SCHIEL
and offer them protection under a Christian roof.8 Latin clerics and mission-
aries living in Central Asia and the Near East bought slave children specifically
to raise them in the Christian faith.9 The language use reflects this ambiva-
lence in the Latin discourse: The abstract noun schiavitù or esclavitud in the
vernacular documents of the Western Mediterranean designated the Muslim
institution of slavery and was never used for slaving practices in Latin house-
holds. The slaves living in Latin households, however, were naturally called
schiavo/schiava or esclavo/esclava. A Muslim master was considered a slave-
holder, while a Christian master acted as a pater familias taking care of all
members of the household including slaves. The Latin discourse on slavery
can thus only be understood in the broader context of late medieval crusades,
maritime expansion, and early Mediterranean colonialism.
Given the ambivalence and fluidity of slave biographies and slavery
discourses, this chapter will focus on following the fragmented traces of indi-
vidual slaves in order to explore the range of scenarios people could experience
(a) while becoming a slave and entering the Western Mediterranean, (b) while
living in a Latin household as a slave, and (c) upon exiting (or attempting to
exit) the condition of slavery.
Following the traces of individual slaves means depending on the frag-
mentary tradition of the time, however. Luckily for us, the number of
administrative records increased dramatically in the late medieval period.
Starting in Italy, more and more people began to confir m their businesses
and last wills with contracts and testaments written and signed by (or in the
presence of) official notaries. Even though the overall level of literacy remained
low, people were able to write and read their names and referred to written
documents as evidence for agreements they had made. Writing was no longer
limited to clergymen and high officials of the lordship. Parishes and municipal-
ities, craft guilds, and trading communities now produced their own records.10
For us as historians, this means that besides official chronicles, legal codes,
and deeds from secular and ecclesiastical rulers, a huge number of administra-
tive documents are available for research on household slavery. Sale contracts,
letters of manumission, account books, tax lists, parish registers, court records,
and of course private wills, letters, and travel accounts help to reconstruct the
lives of the slaves of the time. Ego documents from slaves themselves still do
not exist, however, and in most cases, only brief references or incidences per
slave rather than any form of continued documentation of one and the same
enslaved individual have been handed down to us. All existing r eferences and
representations of slaves were written from the perspective of those who were
not enslaved, and they need to be read with this aspect in mind. Furthermore,
the amount of available material varies greatly from place to place. To explore
the variety of slave lives in Latin households thus requires us to make sense
of scatter ed fragments by aligning and combining them with one another—
like an enormous puzzle where most of the pieces have been lost over time, or
were never even produced. In the following three sections, the reader will learn
the names of individual slaves whose lives stand for the assumed experiences
of many others.
10 SLAVERY IN THE WESTERN MEDITERRANEAN 183
Entrance into Slavery
In terms of historical tradition, the act of enslavement took place outside
the realm of Latin textual sources. Based on this inexistence of written
evidence, historians have long been convinced that Western Europeans were
not involved in the late medieval enslavement processes themselves. It is
only recently that this assumption has been contested by interdisciplinary
research. Bringing together archeological findings and implicit references in
travel accounts of Latin and non-Latin tradition, it has been shown that Latin
and Arab settlers could indeed be involved in inland manhunting. Scholarship
has also determined that the moment in which a person’s name appeared on
a Latin or Arab sale contract and in the trade register of a merchant vessel
disambiguated their status into that of a slave. While people sold on markets
or in ports could be used under many labels and in a variety of roles in the
local context, transshipment over the Mediterranean Sea explicitly transformed
them into slaves.11
Around two thirds of the slaves in the Western Mediterranean were women
and one third were men. Male slaves were more often bought as minors, while
women were on average five years older than their male counterparts at the
time of purchase. What is more, female slaves were generally more expensive
than men. Prices could vary significantly from case to case, ranging from 30 to
135 yperperi depending on the age, sex, origin, and health status of the slave
as well as on the social relation between seller and buyer. Given the rising
demand, the overall price level rose continuously during the late medieval
period. By 1430, the average price was 70 yperperi (or around 23 Venetian
gold ducats) for male slaves and 100 yperperi (or around 33 ducats) for female
slaves. Most of the time, slaves were traded in smaller numbers alongside other
goods transported on larger cargo vessels. The Italian merchant Giacomo
Badoer, for example, who lived in Constantinople for three and a half years,
trading textiles, cotton, linen, and furs for spices, wheat, wine, and copper,
also accepted individual orders from family members and business partners for
a total of 62 slaves.12 Only few merchants specialized in the trading of larger
numbers of several hundreds of slaves at once on ships specifically dedicated
to their transport.
The biggest group of slaves serving in Latin households were purchased
from the slave markets on the Black Sea coast. While Caffa and Tana, the main
trading posts of the Genoese and the Venetians in the region, had the busiest
slave markets at their disposal, Trebizont, Sinope, and other smaller ports also
served as points of departure for the transpor t of enslaved people to the West.
In the purchase contracts signed in the notary offices of the Latin adminis-
tration, these people were identified as tartari, circassi,or rossi, often stating
their original and/or baptismal names, their age and sex, and their physical
and mental status. Most of them arrived there as a consequence of the unstable
political situation in Central Asia following the decline of the Golden Horde.
Riots, border wars, and pillages continuously produced captives of all ethnic
184 J. SCHIEL
and linguistic affiliations fr om the Mongol, Circassian, and Russian principal-
ities, who were then sold to Muslim and Christian merchants operating on
the M editerranean Sea. Others were sold by their parents or relatives (or sold
themselves) to escape poverty and famine, hoping for a better fate away from
home. But even prominent members of the political elites could end up as
slaves in Latin households, with the example of Chebechzi serving as a case
in point: Having served as a close advisor to Cazadahuch, the political repre-
sentative of the Golden Horde in Tana in the 1440s, Chebechzi was captured
and sold into slavery when his lord lost his position due to political unrest. He
eventually arrived in Catalonia, where he served as a slave for several years. It
was only thanks to a chance meeting with a Venetian nobleman and his high
level of education that Chebechzi was ultimately able to make his voice heard
and regain his freedom.13
The second group of slaves populating the Latin households of the Western
Mediterranean were of Greek, Turkish, or Slavic origin, hailing from the Pelo-
ponnese, the Aegean islands, and most importantly from the Balkans, Bulgaria,
and sometimes even Hungary. They were sold in the Peloponnese ports of
Modon and Coron, in Famagusta on Cypr us, in Candia on Crete, and at the
trading posts of the Adriatic coast such as Ragusa, Spalato (today: Split), and
Zadar. In the case of the Adriatic Sea, most ports on its eastern coast were
dominated and controlled by the city of Venice. Nevertheless, cargo vessels
from Apulia, Sicily, Catalonia, and even Palestine and Egypt frequented the
trading ports of Dalmatia and Istria as well, exporting Bosnian and Serbian
slaves to their respective home regions or other parts of the Mediterranean.
The Slavic hinterlands of the Dalmatian and Istrian coasts were contested
between the Roman and Greek Orthodox Churches as well as between the
competing dominions on the western side of the Balkans and the Ottoman
territories in the east.14 The mountainous regions were poor areas difficult
to access and control by ecclesiastical and political authorities. Besides orga-
nized and spontaneous raids, many poor Slavic families sought their fortunes
on the Istrian and Dalmatian coast—sometimes by selling the services of their
children to foreign merchants across the sea. The numerous cartae servitutis
handed down to us note their respective geographic origin (de Bosna, de
Verbase , de Trebotich) and specify whether the person in question was sold
for life (deffinite ad mortem) or for a certain period of time. The young
girl Stana, for example, was sold to the Italian settler Jacobo of Talava in
Spalato for three years. Her mother Liuba r eceived 24 denarii for her daugh-
ter’s services—roughly the value of a cow.15 Some of the children serving in
Latin households on the eastern Adriatic coast were lucky enough to return to
their families with some savings and training in a handicraft. Others were less
fortunate, becoming separated from their families for good when a merchant
from Apulia or Sicily, from Aragon or Valencia shipped them to his hometown
together with other precious goods and sold them into an Italian or Catalan
household for life.
10 SLAVERY IN THE WESTERN MEDITERRANEAN 185
The third group of slaves to be found in Latin households of the Western
Mediterranean were people from sub-Saharan Africa, mainly Ethiopia. These
individuals usually arrived through the agency of Muslim middlemen on the
northern coast of Africa and in Muslim Iberia. Captured in the small Islamized
parts of Africa and traded to the Maghreb region over land, they were sold to
Aragonese and Catalan merchants. Some of them entered Italy via Barcelona
or Mallorca, others were brought from Tunis to Sardinia and Corsica via Sicily.
While slaves from the Black Sea region and the eastern Adriatic coast were
usually traded as individuals, African slaves often entered the Western Mediter-
ranean in groups. Besides Rialto in Venice, proper slave markets primarily
existed in the larger port cities of the Iberian Peninsula.16
It was only when the Ottomans controlled access to the Black Sea after
having conquered Constantinople in 1453 and the Portuguese subsequently
discovered the sea route to Western Africa and India that a shift occurred.
From the 1490s onwards, most of the slaves living in Latin households were
of African origin, while Russian and Mongol slaves only rarely reached the
southern part of Europe.
Experiences of Slavery
Being a slave in a late medieval Latin household could mean many different
things and assume various forms. The most important common denominator
of all slave experiences in the Western Mediterranean was that slaves could be
employed as unskilled workers entirely arbitrarily by their masters, in whatever
position the latter desired or required their help and presence. Slaves could
be assigned to any form of domestic work, side by side and interchangeably
with other domestic workers that were not enslaved. They cooked, cleaned,
looked after children, and nursed infants. They performed courier services,
ran errands, and acted as porters and carters. Outside the domestic sphere,
they worked the land alongside serfs and other dependent agricultural workers.
They helped in craftsmen’s workshops together with apprentices and jour-
neymen. They supported journeying merchants with various tasks, sometimes
even acting as dragomen. And they were at their masters’ disposal for physical
pleasures and sexual services, as well as for the production of offspring.
The second important common denominator was that a slave was subject to
the direct jurisdiction of his or her master. Other jurisdictions only came into
play when the master seized the competent courts. The practical guides for
household masters urged the pater familias to ensure the physical and mental
integrity of their slaves.17 While the latter thus lived under the protection of
the household’s master, they were also completely and utterly at his or her
mercy.18
These two components of slave existence in the late medieval Western
Mediterranean are the primary reason why the actual experiences of slaves left
very few traces. Once slaves had been bought, shipped, and baptized, their
186 J. SCHIEL
various unskilled activities in and outside the household as well as their experi-
ences of encounter and negotiation, of conflict and violence remained invisible
in historical tradition unless the master or an external institution indicated a
problem.
Only letter correspondences provide insight into the daily lives of slaves.
The long-term exchange of letters between the merchant couple Margherita
and Francesco Datini, who maintained two households in Tuscany, is a promi-
nent example.19 Here we learn that the slave Bartolomea enjoyed a low level
of trust and remained under the direct observation of her mistress, never
allowed to leave the house. Other slaves bore babies to Francesco, whose own
marriage with Margherita remained childless. The estimation of slave Lucia,
for example, clearly improved when she gave birth to Francesco’s daughter
Ginevra, who survived infancy and grew up as the Datini’s only child.
Where we have no such telling narrative sources, the range of activities
and experiences can only be deduced from contested cases—from situations of
conflict and debate. The most important sources for the reconstruction of slave
experiences are therefore court records and council or guild orders mentioning
slaves. Throughout the late medieval period, more and more regulations were
concerned with the danger of criminal slaves being prepared to commit theft
or use violence. In the case of Venice, the supposed poison murder of a
prominent nobleman by a slave woman called Bona Tatara not only triggered
an entire series of conspicuously similar cases in court, it also persuaded the
political class to regulate access to toxic substances as well as the granting of
licenses to run pharmacies and practice natural medicine. Accor ding to the
court records, the slaves accused of poisoning their masters complained that
they had been beaten or mistreated or that they could no longer endure the
hard work requested of them. At the same time, the reader learns from the
court documents that slave women moved freely within the town and were
often well acquainted with other neighboring women of the popolo.20
A further informative source is the guild prescripts of the time. Besides
regulating access to and training in crafts, medieval guilds also protected their
members and their families in the event of illness and death, as well as against
competition from foreign craftsmen. Employing slaves as unskilled workers
in a workshop constituted a cheap alternative for masters practicing a craft,
but it also went against the idea of clearly regulated pathways of education
and training and created a competitive situation for the apprentices and jour-
neymen. In theory, the employment of slaves not officially registered in the
guild register was forbidden. In practice, however, it was largely tolerated
except when higher interests were at stake, as the various city council orders
show. For example, the Venetian senate pronounced a general ban on slave
labor in the production of luxury goods destined for export, arguing that the
specialist know-how of a goldsmith or glassmaker could leave the Venetian
dominion if an appropriately trained slave were manumitted or resold to a
non-Venetian buyer.21
10 SLAVERY IN THE WESTERN MEDITERRANEAN 187
Finally, the gradual shift in inheritance law allowing children begotten by a
household master with a female slave to inherit the free status of their father
(rather than their mother’s slave status, as in the Roman law tradition) and
the simultaneous significant rise in foundlings in the foundling hospitals of
Southern Europe demonstrates another important element of slave experi-
ences of the time. The sexual power of household masters over their female
slaves prevented the noble lines of European aristocracy from becoming extinct
while at the same time establishing a new branch of human exploitation: the
wet nursing industry. In Florence, like in other Italian cities, the priors of
the Ospedale degli Innocenti complained that the number of slave babies had
become unmanageable. Household masters were obligated to announce the
birth of a slave baby in advance and asked to pay compensation to the hospi-
tals upon delivery. On the flip side, employing and renting out slave women
as wet nurses began to become such a lucrative business that the r eproductive
functions of female slaves were increasingly the subject of purchase negotia-
tions and slave contracts. In Genoa, the Officium Mercantie regularly noted
cases in which female slaves were returned or money was reclaimed when the
purchased woman did not menstruate.22
Exit from Slavery
In the late medieval Western Mediterranean, almost all slaves exited the state of
slavery at one point or the other. This rarely meant a substantial improvement
of their social status, however: While opportunities for social advancement
were plentiful for former slaves in Muslim societies at the time, ex-slaves in
the western part of the Mediterranean region mostly remained dependent on
someone else’s mercy.
Generally speaking, there were four possible scenarios for escaping the
state of enslavement. The first and most common of these was manumission.
Household masters disposed of their belongings and decided about the future
fate of their slaves in their wills or, on special occasions or for specific reasons,
by way of notarial deeds. The exact wording of such testaments and deeds
proves very telling with regard to the social relationship between the respective
master or mistress and his or her slave. Slaves could become a bequest them-
selves when the master or mistress transmitted them to a close relative, or they
could become the recipient of a bequest when their manumission was affirmed
with the allocation of movable goods or a certain amount of money. The Vene-
tian doge Antonio Venier, for example, decreed that his slave Lucia was to be
set free upon his death with a financial endowment of 20 gold ducats, while
slave Lena was to serve his wife for another five years before being manumitted
with an endowment of 10 ducats. Another of Venier’s slaves was to serve his
daughter Valvina until the latter’s death, at which point she was to be released
if she had behaved well.23 In many cases, manumitted slaves continued to
serve in the same household—sometimes in a more favorable position, some-
times not. Francesco Datini’s slave Argomento was rewarded for his excellent
188 J. SCHIEL
services with manumission and became one of the most important carters—
and thus a person of trust—in the Datini business.24 Others continued to fulfill
similar tasks as before, with the sole difference that they now worked under
regularly renewed six- or twelve-month contracts and were remunerated with
a small wage. The standard justification stated in wills and notarial deeds for
granting a slave freedom was the salvation of the testator’s soul. Manumission
was deemed to be a pious act of Christian charity. The real motives of the
testators were often much less honorable, however, and the social reality of
the manumitted was sometimes quite sad. Many slaves were set free with very
little financial or material endowment, or none at all—for in practice, manu-
mission was also an easy means of getting rid of a disagreeable or aging slave
no longer willing or able to work. Released from the state of enslavement,
these people were forced to accept any type of work and working conditions.
In the absence of alternatives, some of them resorted to begging, prostitution,
or crime.25
The second possible scenario for exiting the state of slavery was marriage,
an option that mainly applied to female slaves. Generally speaking, marriage
offered the greatest potential for upward social mobility for slaves in the late
medieval Western Mediterranean world. Marrying a burgher with citizen rights
could open up new social and economic possibilities and allow former slaves
to live in an economically autonomous household. In some cases, marriage
came as an add-on to standard manumission. Francesco Datini’s slave Lucia,
for example, who had given birth to his daughter Ginevra, was later compen-
sated with marriage to Nanni da Prato, a laborer in the Datini business, along
with a generous dowry that allowed them to live on the Datini estate in their
own small house. In other cases, however, the marriage of a slave to a person
outside the master’s household was the result of a court settlement or an extra-
judicial arrangement following an incident of sexual intercourse between the
slave and their eventual spouse. As a matter of fact, slaves were not allowed
to have sexual contact with persons outside their master’s household at all—at
least not without the master’s consent. Voluntary or involuntary sexual contact
with slaves was considered a property crime against the master’s property. Not
the potential sexual violence against the slave, but the misappropriation of their
body as the master’s property was considered a crime to be brought before a
court.26 Irrespective of a female slave’s own will, her marriage to a man from
the neighborhood was thus often the outcome of a conflict between her master
and her lover or rapist.
The third exit scenario was absconding. Some striking differences between
escape practices can be noted here. Generally speaking, enslaved men tried to
escape more often than women, who were more vulnerable to violence and
sexual exploitation than men. Furthermore, the rate of attempted escapes was
considerably higher on the Iberian Peninsula than in Italy. While the majority
of the slaves living in Italian households had been traded and sold as indi-
viduals at a relatively early age, slaves arriving on the Iberian Peninsula often
came in groups or as families from the same place. The fact that they knew
10 SLAVERY IN THE WESTERN MEDITERRANEAN 189
each other from the time before their enslavement and/or transfer to Western
Europe greatly increased their incentive to flee. Related to this aspect, slaves
in Italian households mainly fled from a situation that had become unbearable
for them to another household in a nearby town or region in hopes of finding
better conditions there. On the Iberian Peninsula, however, at least some of
the runaway slaves also hoped to return to their homeland and regain their
former lives. In general, slaves’ attempts to escape were more successful when
they relied on a network of accomplices who often spoke the same language,
no matter whether they had known each other before their enslavement or
not.
The final possibility for slaves to attain freedom was court decisions: In
rare cases, slaves could be manumitted by a court order because they had
been illegally enslaved. Thus we read that the Mongolian slave Chebechzi and
his compatriot were freed by the Venetian Signori di Notte and sent back to
their homeland when it became known they had fled Catalonia in a small
boat and been caught by a Venetian vintner who clandestinely kept them in
chains in his wine cellar.27 Other cases concerned the illegitimate enslavement
of Catholic Christians from the Balkans. A Sicilian court decision of 1417
returned 25 Bosnians to their homeland who, according to the magistrates,
were “true Christians” and had been wrongly “sold as slaves.”28 The other exit
scenario by way of a court decision was, of course, the condemnation of a slave
for a crime he or she was accused of. In these cases, the vulnerable position
of the slave becomes most obvious: Slaves appear in criminal court records
as witnesses and as accused persons, but never as plaintiffs. Their testimony
before a court served to administer justice in a conflict involving their master,
but never to enact justice for wr ongs they had suffered themselves. A trial
involving a slave could thus easily end with a sentence of death. The case of
the previously mentioned Bona Tatara is a striking example here: According
to Bona, her master had beaten her almost to death with a leather strap when
he learned that she was pregnant. In revenge, she said, she had bought arsenic
in a nearby pharmacy and poisoned her master. The criminal court sentenced
her to death by being dragged first through water and then along the street,
tied to a boat and a horse respectively, before being burned at the stake.29
Conclusion
Placed in a broader context, slavery in the late medieval Western Mediter-
ranean relied on long-distance trafficking networks that mainly traded people
across religious and cultural borders. In the context of the crusading age,
religious and cultural otherness had become the primary marker for “enslave-
ability,” and human trafficking as well as a theologically loaded discourse on
slavery had become a means of religious conflict. At the same time, military
struggles went hand in hand with economic and political expansion, paving the
way for a new form of Mediterranean colonialism. Western European maritime
powers experimented with different forms of colonial rule while exploring the
190 J. SCHIEL
coasts of the Mediterranean, the Adriatic, and the Black Sea along with their
hinterlands, as well as the Mediterranean islands. Their colonizing practices
ranged from the imposition of fees and taxes or economic sanctions to the
implementation of a Latin administration and the establishment of a central-
ized settlement policy with a transregional redistribution system. The trade
in slaves was an integral part of this new E uropean expansionism driven by
the Western Mediterranean Sea powers, since it enabled a new form of labor
recruitment. Even though most of the slaves sold into the Western Mediter-
ranean served as individual unskilled workers in urban Latin households, the
new labor requirements of the recently colonized islands also presented an
opportunity to use coerced labor on a larger scale: The use of slaves for agri-
cultural work on the plantations of Mallorca, Malta, or Crete formed the basis
for the new plantation economy of the early modern age. The transatlantic
slave trade and the implementation of a colonial economy based on slave
labor would not have been possible without the experience gathered in the
late medieval Western Mediterranean.30
Notes
1. Still relevant for a general overview: Charles Verlinden, L’esclavage dans l’Eu-
rope médiévale. 2 vols. (Bruges/Ghent: De Tempel, 1955/1977); Sergej P.
Karpov, L’Impero di Trepisonda Venezia, Genova e Roma. 1204–1461: Rapporti
politici, diplomatici e commerciali (Rome: Veltro Ed., 1986).
2. Andrew Ehr enkreutz, “Strategic Implications of the Slave Trade between
Genoa and Mamluk Egypt in the Second Half of the Thirteenth Century,”
in The Islamic Middle East, 700–1900, ed. Abraham Udovitch (Princeton:
Princeton University Press, 1970), 335–45.
3. Reuven Amitai and Christoph Cluse, eds., Slavery and the Slave Trade in the
Eastern Mediterranean (c. 1000–1500 CE) (Turnout: Brepols, 2017), 401–69.
4. Jeffrey Fynn-Paul, “Empire, Monotheism and Slavery in the Greater Mediter-
ranean Region from Antiquity to the Early Modern Era,” Past & Present 205,
no. 1 (November 2009): 3–40.
5. Christopher Paolella, Human Trafficking in Medieval Europe: Slavery, Sexual
Exploitation, and Prostitution (Amsterdam: Amsterdam University Press,
2020), 215–45; Peter Feldbauer, Gottfried Liedl, and John Morrissey, eds.,
Mediterraner Kolonialismus: Expansion und Kulturaustausch im Mittelalter
(Essen: Magnus-Verlag, 2005).
6. Hannah Barker, The Most Precious Merchandise: The Mediterranean Trade in
Black Sea Slaves, 1260–1500 (Philadelphia: University of Pennsylvania Press,
2019).
7. As a representative example of Crusade sermons, see Guillelmus Adae, “De
modo Saracenos extirpandi (1317),” in Recueil des historiens des Croisades:
Documents arméniens, vol. 2 (Paris: Société des amis de la Romanie, 1906),
521–55.
8. As a representative example of Church decrees, see S. Franco, H. Fory, and H.
Dalmazzo, eds., Bullarium Romanum, 24 vols. (1857–1872), here vol. 4, no.
16–17 (3 June 1425), 718–21.
10 SLAVERY IN THE WESTERN MEDITERRANEAN 191
9. Representative for other accounts, see “Epistolae Fr. Iohannis de Monte
Corvino,” in Sinica Franiscana,I: Itinera et relationes fratrum minorum saeculi
XIII et XIV , ed. P. A. v.d. Wyngaert O. F. M. (Firenze: Apud Collegium S.
Bonaventurae, 1929), 340–55, esp. 347–8.
10. Andreas Meyer, Felix et inclitus notarius: Studien zum italienischen Notariat
vom 7. bis zum 13. Jahrhundert (Tübingen: Niemeyer, 2000); Irmgard Fees,
“Ein venezianischer Kaufmann des 12. Jahrhunder ts: Romano Mairano,”
in Il mito di Venezia: Una citt `a tra realt`a e rappresentazione,ed. Peter
Schreiner (Rome—Venice: Ed. Di Storia e Letteratura, 2006), 25–59; Linda
Guzzetti, Venezianische Vermächtnisse: Die soziale und wirtschaftliche Situa-
tion von Frauen im Spiegel spätmittelalterlicher Testamente (Stuttgart: Metzler,
1998).
11. Marie Favereau, La Horde d’Or et l’islamisation des steppes eurasiatiques (Aix-
en-Provence: Presses Universitaires de Provence, 2018); Barker, The Most
Precious Merchandise; Juliane Schiel, “Tatort Tana: Die Rolle Lateineuropas
in der Sklavenökonomie des Schwarzmeerraums (ca. 1300–1500),” Historische
Zeitschrift 313, no. 1 (2021): 32–60; Juliane Schiel, “The Ragusan ‘Maids-of-
all-Work’. Shifting Labor Relations in the Late Medieval Adriatic Sea Region,”
Journal of Global Slavery 5 (2020): 139–69.
12. Umberto Dorini and Tommaso Bertelè, eds., Il libro dei conti di Giacomo
Badoer (Constantinopoli, 1436–1440) (Rome: Ist. Poligr. dello Stato, 1956).
13. “Di messer Iosafa Barbaro, gentiluomo veneziano, il viaggio della Tana e nella
Persia,” in Navigazioni e viaggi: A cura di Marica Milanesi, ed. Giovanni
Battista Ramusio, vol. 3 (Turin: E inaudi, 1980), 485–574, at 504–5.
14. Emir O. Filipovi´c, “The Ottoman Conquest and the Depopulation of Bosnia
in the Fifteenth Century,” in State and Society in the Balkans Before and After
Establishment of Ottoman Rule, eds. Srdan Rudi´c and Selim Aslantas (Belgrade:
The Institute of History Belgrade, 2017), 79–101.
15. Gregor ˇ
Cremošnik, Acta Cancellariae et Notariae Annorum 1278–1301
(Belgrade, 1932), no. 79, 46–7.
16. Debra Blumenthal, Enemies and Familiars: Slavery and Mastery in Fifteenth-
Century Valencia (Ithaca: Cornell University Press, 2009); Aurelia Martín
Casares, La esclavitud en la Granada del siglo XVI: Género, raza y religion
(Granada: Univ., Centro Prov. de la Mujer, 2000).
17. Dennis Romano/Guido Ruggiero Leon Battista Alberti, eds., I libri della
famiglia (Torino: Einaudi, 1989). Vera Ribaudo, ed., Benedetto Cotrugli
Raguseo. Libro de l’arte de la mercatura (Venice: Edizioni Ca’Foscari, 2016).
18. Stanley Chojnacki, “Crime, Punishment, and the Trecento Venetian State,” in
Violence and Civil Disorder in Italian Cities. 1200–1500, ed. Lauro Martines
(Berkely: University of California Press, 1972), 184–228; Dennis Romano,
Housecraft and Statecraft: Domestic Service in Renaissance Venice 1400–1600
(Baltimore: Johns Hopkins University Press, 1996).
19. Elena Cecchi, ed., Le lettere di Francesco Datini alla moglie Margherita (1385–
1410) (Prato: Società Pratese di Storia Patria, 1990); Valeria Rosati, ed., Le
lettere di Margherita Datini a Francesco di Marco (1384–1410) (Prato: Cassa di
Risparmi e Depositi, 1977); Carolyn James/Antonio Pagliaro, eds., Margherita
Datini. Letters to Francesco Datini (Toronto: Centre for Reformation and
Renaissance Studies, 2012).
20. Archivio di Stato di Venezia, Avogaria di Comun, Raspe, reg. 3646, fol. 84v–
85r (19 June 1410); for more detailed information on the case of Bona
192 J. SCHIEL
Tatara, see Juliane Schiel, “Mord von zarter Hand: Der Giftmordvorwurf
im Venedig des 15. Jahrhunderts,” in Mediterranean Slavery Revisited (500–
1800)/Neue Perspektiven auf Mediterrane Sklaverei (500–1800), eds. Stefan
Hanß and Juliane Schiel (Zurich: Chronos, 2014), 201–28.
21. Juliane Schiel, “Die Sklaven und die Pest: Überprüfung eines Forschungsnarra-
tivs am Beispiel Venedigs,” in Schiavitù e servaggio nell’economia europea, secc.
XI–XVIII/Serfdom and Slavery in the European Economy, 11th–18th Centuries,
ed. Simonetta Cavaciocchi, vol. 1 (Florence: Firenze University Press, 2014),
365–75, at 371–74.
22. Christoph Cluse, “Frauen in Sklaverei: Beobachtungen aus genuesischen
Notariatsregistern des 14. und 15. Jahrhunderts,” in Campana pulsante convo-
cati: Festschrift anlässlich der Emeritierung von Prof. Dr. Alfred Haverkamp,ed.
Frank G. Hirschmann and Gerd Mentgen (Trier: Kliomedia, 2005), 85–123;
Sally McKee, “Inherited Status and Slavery in Late Medieval Italy and Venetian
Crete,” Past and Present 182 (2004): 31–53.
23. Sergio Perini, ed., “Testamento del doge Antonio Venier: 1400, 24 ottobre,”
Archivio Veneto 138–139 (1992): 126–33; Juliane Schiel, Isabelle Schürch,
and Aline Steinbrecher, “Von Sklaven, Pferden und Hunden: Trialog über
den Nutzen aktueller Agency-Debatten für die Sozialgeschichte,” Schweiz-
erisches Jahrbuch für Wirtschafts- und Sozialgeschichte/Annuaire suisse d’histoire
économique et sociale 23 (2017): 17–48.
24. See endnote 19.
25. Romano, Housecraft and Statecraft ; Guido Ruggiero, The Boundaries of Eros:
Sex Crime and Sexuality in Renaissance Venice (New York: Oxford University
Press, 1985).
26. Ruggiero, The Boundaries of Eros.
27. See endnote 13.
28. Ferdinando Lionti, ed., Codice diplomatico di Alfonso il Magnanimo,vol.1:
1416–1417. Palermo 1891, 159, n. 297.
29. See endnote 20.
30. Giulia Bonazza, “Connecting the Mediterranean and the Atlantic,” Journal of
Global Slaver y 3 (2018): 152–75.
Further Readings
Amitai, Reuven, and Christoph Cluse, eds. Slavery and the Slave Trade in the Eastern
Mediterranean (c. 1000–1500 CE). Turnhout: Brepols, 2017.
Barker, Hannah. The Most Precious Merchandise: The Mediterranean Trade in Black
Sea Slaves, 1260–1500. Philadelphia: University of Pennsylvania Press, 2019.
Blumenthal, Debra. Enemies and Familiars: Slavery and Mastery in Fifteenth-Century
Valencia. Ithaca: Cornell University Press, 2009.
Guillén, Fabienne, and Salah Trabelsi, eds. Les esclavages en Méditerranée: Espaces et
dynamiques économiques . Madrid: Casa de Velàzquez, 2012.
Hanß, Stefan, and Juliane Schiel, eds. Mediterranean Slavery Revisited (500–1800):
Neue Perspektiven auf Mediterrane Sklaverei (500–1800). Zurich: Chronos, 2014.
Martín Casares, Aurelia. La esclavitud en la Granada del siglo XVI: Género, raza y
religion. Granada: Univ., Centro Prov. De la Mujer, 2000.
McKee, Sally. Uncommon Dominion: Venetian Crete and the Myth of Ethnic Purity.
Philadelphia: University of Pennsylvania Press, 2000.
10 SLAVERY IN THE WESTERN MEDITERRANEAN 193
Paolella, Christopher. Human Trafficking in Medieval Europe: Slavery, Sexual
Exploitation, and Prostitution. Amsterdam: Amsterdam University Press, 2020.
Rio, Alice. Slavery after Rome, 500–1100. Oxford: Oxford University Press, 2017.
Verlinden, Charles. L’esclavage dans l’Europe médiévale. 2 vols. Bruges/Ghent: De
Tempel, 1955/1977.
Open Access This chapter is licensed under the terms of the Cr eative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 11
The Question of Slavery in the Inca State
Karoline Noack and Kerstin Nowack
Introduction
The question of whether or not there were slaves and slaver y in the Inca
state (or Tawantinsuyu) has long been disputed in historiography, as has the
general question of the dependence of people in the Inca state and the nature
of their dependency. This debate is an expression of the continuous search
for theoretical frameworks for Inca history, which have been based largely on
various European models and terminologies. The various forms of dependency
existing in the Inca state have been interpreted in the context of evolutionar y
models of economic and social evolution, where one category of dependents,
the yanacona, were regarded as possible slaves. Accor ding to John Murra,1 the
dispute about the yanacona is as old as the anthropology of the Inca state.2
Yanacona can be defined as people detached from their family group, freed
from tribute to the Inca, and serving the Inca and provincial elites. However,
this is not the only social and labor-related category of dependency in the
Inca state. Other main social and labor categories are mitimaes, camayoc and
K. Noack (B) · K. Nowack
University of Bonn, Bonn, Germany
e-mail: knoack@uni-bonn.de
K. Nowack
e-mail: Kerstin_Nowack@gmx.de
K. Nowack
Independent Scholar, Bonn, Germany
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_11
195
196 K. NOACK AND K. NOWACK
mitayos. Paraphrased as relocated people (mitimaes), occupational specialists
(camayoc), and commoners taking turns of ser vice to fulfill their tribute obliga-
tions (mitayos), all of these are deeply rooted, but also historically evolving and
intersecting social groups. These categories include men and women, while
an additional category, acllacuna/ mamacuna, comprised exclusively women,
the first category referring to young, the second to elder women. These are
considered very close to the yana category. Given this situation, it is dif ficult
to isolate just one of the social groups to describe, but we will focus our atten-
tion on the yanacona as the group that in previous works had been seen as
experiencing an especially strong form of asymmetric dependency. However,
we will also consider the overlapping of the yana category with other groups.
A systematic study of the yanacona, but also of the other social and labor-
related categories of the Inca state, i.e., on the dependent population, is still
pending. After the seminal works of the so-called Andean Studies of the 1960s
to 1980s, established principally by John Murra (1916–2006), no scholars
have given much attention to these topics, generally, and also not to the yana-
cona, in particular.3 We focus here on a source-saturated approach that follows
from Andean Studies and incorporates new anthropological, historical, and
archaeological perspectives on the Inca state.
All written sources about the yanacona come from the Spanish colonial
period. These were eyewitness reports, chronicles, administrative documents as
visitation reports (“visitas”), and also early Quechua dictionaries. The meaning
of the Quechua word yana in an early seventeenth-century dictionary is given
as “criado moço de seruicio” (“servant, servant youth”), yanacona is the plural
form.4 Generally, Spanish terms like “criado,” “indios de servicio” or some-
times just “servicio” (service), can be assumed to mean yanacona as well, but
might also cover other individuals. Not everyone called a “criado” (“servant,”
but also “raised”) was necessarily a yana. In addition, yanacona could also be
occupational specialists (termed camayoc), and some of them occupied leader-
ship positions and the term was applied to members of local elites. The degree
of dependency is therefore not easy to determine, as the yanacona could cover
a whole spectrum of different status positions. After a few general reflections
on Inca society and economy, we will describe the yanacona as a dependent
social group and labor category in three main sections. The first section will
examine how people became yana; the second will explore personal service to
the ruler, as well as working for the Inca and provincial elites; and the third
section will consider how people exited the yana status.
11 THE QUESTION OF SLAVERY IN THE INCA STATE 197
The Inca Rulers
Name According to Miguel Cabello
Valboa [1586]a
According to Juan de Betanzos
[1551–57]b
Manco Inca
Sinchi Roca
Lloque Yupanqui
Mayta Capac
Capac Yupanqui
Inca Roca
Yahuar Huacac
Viracocha ?–1438 ?–1420/30
Pachacuti Inca Yupanqui 1438–1471 1420/30–1470/75
Topa Inca Yupanqui 1471–1493 1470/75–1485/90
Huayna Capac 1493–1527 1485/90–1525
Huascar 1527–1532 1525–1532
Atahualpa 1532–1533
Table (elaborated by Kerstin Nowack)
aMiguel Cabello Valboa, Miscelánea Antártica, ed. Isaías Lerner (Sevilla: Fundación José
Manuel Lara, 2011 [1586])
bJuan de Betanzos, Suma y Narración de los Incas: seguida del discurso sobre la descendencia y
gobierno de los Incas, ed. María del Carmen Martín Rubio (Madrid: Polifemo, 2004 [1551–57])
The Tawantinsuyu encompassed parts of present-day southern Columbia,
Ecuador, Peru, Bolivia, and from northwest Argentina up to central Chile,
making it the largest state in the Americas in terms of g eographic extent. Its
manifold landscapes ranged from a desert coast through hot and fertile valleys,
to arid and cold highlands down to the eastern slopes of the Andes and the
eastern rainforest. It was inhabited by between six and thirteen million people
under the control of Inca rulers at its peak extent.5 The Inca state arose
through expansion during the last hundred years or so before the Spanish
conquest of Peru beginning in 1532.
How did Inca rulers manage to establish a state of such great dimensions
in such a short time? The explanation for this phenomenon lies in the inter-
locking structures Andean communities, such as the mincca, reciprocal or
mutual aid that was inseparably linked to the ayllu, the basic, household-based
social organization of hatun runa or “commoners.” Social relations within
the ayllu were structured on the basis of collective land ownership and the
related principle of reciprocity. Members of local communities benefited from
the mincca in agricultural work or, for example house construction. Gradu-
ally and in a manner almost imperceptible to Andean peoples, these structures
were used to further the Inca state’s interests and transformed into the mita,
a system of rotating labor obligations to the state based on the mincca model,
for which new redistributive forms and ideological-religious justifications were
created. The state needed this form of extended labor obligation of the
community members, or mitayos, to expand its productive agricultural base
and its infrastructure (e.g., roads, road stations, storehouses, administrative
centers).
198 K. NOACK AND K. NOWACK
The Inca succeeded in achieving a subtle balance between diversity through
the maintenance of local structures and unification on a higher level. In
the diverse Andean ecosystem with its micro-landscapes and micro-climates,
local agriculture was not only practiced at a single altitude level, for example
growing maize at about 3000 meters or potatoes in the Puna highlands.
Instead, local communities often controlled several vertical levels by estab-
lishing permanent colonies or “archipelagos” in order to obtain a wider range
of agricultural products. These levels were often inhabited by mitimaes,relo-
cated people or “colonists” of different origins.6 Inca rulers expanded this
institution by systematically relocating and resettling families, or even entire
communities to new, strategic locations, thereby developing unused or under-
used land and other resources, but also contributing to the stability of their
rule by dividing conquered polities. In a similar manner, they took the existing
practice of having yanacona and expanded and adapted it for the support of
their elite and of state-sponsored cults.
Becoming a
Yana
The yanacona have their origins in the time before the Inca. There were
different ways of falling into this form of dependency. In pre-Inca times, the
yanacona were apparently an integrated part of the local social organization.
In Inca times, one can assume that the yanacona are to be seen in connection
with the wars of conquest of the Inca and generally with the progressive consti-
tution of the Tawantinsuyu. At the same time, the pre-Inca practices continued
to exist in the local organizations. That is to say that often different forms
of becoming yana were intertwined. During Inca rule, capture in war was
probably a common way of becoming a yana. In the main Andean languages,
Quechua and Aymara, the words for prisoner of war and their Spanish transla-
tions indicate that there was an overlap in meaning between “war captive,”
“prisoner,” and “slave.”7 Juan de Betanzos, an important chronicler who
could communicate in Quechua, mentions that some yanacona had been pris-
oners of war, mostly men or families abducted from their provinces during
campaigns of conquest. He exemplifies this when he describes specific military
campaigns and gives examples of people taken as booty and individuals settled
as yanacona.8 After the initial transformation of war captives into yanacona,
themajorityofthe yanacona was then provided by the local population as
part of its tribute obligations after the wars of conquest. These are described
as being taken or recruited from the provinces and allocated to the descent
groups of the former rulers,9 who continued to be present in the life of their
descendants as mummified ancestors, and whose land, servants, and buildings
remained in their possession, serving to maintain their descent group.
Accordingly, yanacona could be provided for the Inca ruler himself and
his family; including the former rulers; the Inca elite;orthe local elites of the
subjugated provinces—in this last case as a reward for subjugation, but also
11 THE QUESTION OF SLAVERY IN THE INCA STATE 199
as part of the existing practices. The Inca elite expected subjugated popu-
lations to provide it with yana laborers in a similar manner as provincial
lords requested these permanent dependents from their subjects. The yana-
cona themselves could come from socially completely different status groups,
namely both from the local elite of the subjugated region, but also from
the group of the commoner population. As yanacona from the elite, they
either remained at their status or were able to raise it which was sometimes
also true for commoners who became yanacona. Yanacona thus at times
were an important instrument of Inca rule and the maintenance of their
power in the conquered territories. They could exercise positions of high
responsibility as government officials, for example, but the case of the ten
thousand Chachapoyas, the “Warriors of the Clouds” on the eastern slope
of the northern Andes in Peru, stands out, who due to their military skills
were turned into specialized soldiers by the Inca, some of them yanacona.10
After the Spanish conquest, these military skills could be turned against the
Inca. Condorguacho, the female ruler of the Huaylas province, had been
given yanacona by her husband, the Inca ruler Huayna Capac, respectively his
military leaders. In 1536, she personally led troops to support the Spaniards
during their fight against the Inca and “brought many Indians Chachapoyas
[and other] with her that were hers.” They might have been the war captives
awarded to her as yanacona.11 Other Chachapoya yanacona and/or miti-
maes were assigned to the Inca Huayna Capac personal guards and for his
son Huascar together with other groups. Especially from mitimaes we know
that they were able to maintain their high status during the turbulent times of
Spanish conquest, when the Chachapoya became allies of the Spaniards. This
guaranteed them a high social status with access to honor and privilege even
during the colonial period.12
One specific example of becoming a yana is given by don Joan Puyquin
during the General Visitation of viceroy Francisco de Toledo, the “reformer”
of the viceroyalty of Peru (1570–1575). He described to the Spanish informa-
tion gatherers that his father had been captured as a child during a campaign
against the Chachapoya. While most members of his local group were killed,
the father of don Joan Puyquin was allowed to live because he was so young,
and was made a servant or yana of Topa Inca.13 Later in life, the father
married, had children on his own, and even acquired a position of local leader-
ship in the Cusco region. Remarkably, it is documented that the Chachapoya
turned war captives into yanacona as well, since Inca Garcilaso de la Vega,
famous author of the “Royal Commentaries of the Incas” (1609) mentions
a supposed Chachapoya rebellion, where they took Inca soldiers as prisoners
“with the intention of using them as slaves.”14
Although we know something about wars being the context in which
people became yanacona, in the other cases we know nothing about the selec-
tion process by which yanacona were drafted. Because the sources on the
yanacona are so scattered, they seem to give the impression of random deci-
sions. The exception is the yanacona of elite origins who came “from the
200 K. NOACK AND K. NOWACK
best people and most [were] children of curacas [local rulers],” as the Spanish
administrator Hernando de Santillán explains.15 They served the Inca ruler in
special capacities and progressed to become part of a patronage network in the
Inca state. Their fate differed completely from that of yanacona drafted from
the commoner population.
The other accounts tell of the way yanacona were chosen in the provinces
by the Inca nobles, but not who exactly was chosen. In the Chincha valley
in south coastal Peru, Inca officials are reported to have come inspecting the
performance of the local rulers, collecting data on the province’s demographics
as well as selecting individuals who would become aclla (“chosen women”)
and yana. A local ruler in Huanuco in the northern-central Andes reported to
the Spanish crown officials about the work his people had done for the Inca
state. For example, he sent a number of yanacona to live in Cuzco in service
of the Inca ruler. As he underlined, their children as adults continued to serve
as yanacona. This is an indication that the status of the yanacona may have
been hereditary. If there were no children, the home province had to send a
replacement.16 Nothing is said about the fate of yanacona who had become
too old for work.
This form of recruitment explains why even an Inca ruler who had not yet
led a successful campaign of conquest could nevertheless distribute yanacona
to the descendants of his predecessors, the “dead lords”. Similarly, high-
ranking members of an Inca ruler’s family like his mother and principal wife
were allowed to get yanacona from the provinces to work on the parcels of
land they had been allocated.17 Possibly, yanacona of the Inca ruler and the
Inca elite were selected mostly from small and non-influential kinship g roups
(ayllus), but it is also imaginable that, as in the case of general labor service
(mita), a percentage of every given unit was chosen.18 The only other infor-
mation on the origins of yanacona again comes from Betanzos. He provides a
list of Inca laws, one of them stating that thieves either had to recompense the
owner of stolen property by giving it back in double. If they did not have the
means for making amends, they had to become the yanacona of the property
owner. Not only would the thief stay the yana of property owner until his
death, he would also bequeath his status on his children. A second law implies
that unwanted children exposed as infants could be turned into yanacona.19
In these cases, it is evident that people already marginalized in their societies
were in danger of becoming yanacona.
As stated above, the Inca built their institutions and practices on those
existing in the Andes prior to their rule. Having yanacona was widely practiced
by the local elite in the pre-Inca Andes. Local rulers from highland Ecuador,
Peru and Bolivia as well as from coastal Peru describe that they enjoyed the
service of yanacona.20 They expected to be provided with yanacona from their
subjects. That is to say that, unlike as in the above-mentioned cases, the depen-
dents originated from the same group as the people for whom they worked.21
In a village near the Lake Titicaca, one of the Lupaca rulers still powerful in the
colonial epoch, explained that he had both Aymaras and Uros in his service,
11 THE QUESTION OF SLAVERY IN THE INCA STATE 201
referring categories used to define people around the Lake Titicaca by occupa-
tion, lifestyle, and language.22 The ruler governed members of both groups,
and both provided him with yanacona. Yanacona could also be “given” within
the hierarchy of the provincial elite from one lord to a higher-level lord.23
A very few examples indicate that yana status could be the result of a break-
down of social relations, as for example in the case of orphaned children who
became members of a local elite household, and as adults found that they
were ascribed to the yana category for their lifetime. In the aftermaths of
the Spanish invasion, documents from the central highlands of Ecuador and
Peru refer to individuals and families who apparently accepted yana status in
exchange for a livelihood and the protection of local elites.24
While coercion by economic and social circumstances might be a factor
in becoming a yana in local social organizations, direct violence is hardly
ever mentioned. Although the inhabitants of the central Andes experienced
a period of regular conflict among neighboring polities before the Inca ruled,
and although warfare was also before the Inca one of the most common ways
to acquire dependents, there is little evidence that yanacona in those times
were war captives. At the same moment, the testimonies of local elites reflect
the situation after one to four generations of Inca rule, and the correlating
suppression of conflicts among the Tawantinsuyu’s subjects.
The term yana encountered in the colonial sources thus covers a great
variety of statuses and historical developments. What interests here most is
the employment of dependents in the Inca empire, starting with the largest
group, those working for the Inca elite.
Personal Service to the Ruler: Working
for the Inca and Provincial Elites
In the Inca state, the obligation to work applies to all who depended on the
Inca state or the local political elites, or a ruler, whether or not they are in
servitude. Therefore, most of the population as a whole worked to provide
basic subsistence, and this is true for the yanacona as well. Lamentably, the
information about the circumstances of their life does not go far beyond these
references to their labor obligations. Based on the sources, we can again distin-
guish between labor and personal services of the yanacona for the Inca ruler
and his family as well as for the Inca elite and for the local elites.
Most of the Inca yanacona were workers on the Inca elite’s rural estates.25
These estates, mainly located in the envir ons of the Inca capital of Cuzco, were
settlements and r esource systems, created by a ruler and then managed in the
name of his descendants. Many of them included more than one palace as well
as private lands and resources for their inhabitants and for the sun, the prin-
cipal deity. They also encompassed storehouses, pastures, forests, gardens or
moyas, irrigated fields, coca fields, salt resources, bridges, and villages.26 Some-
times thousands of yanacona worked there, nearly all resettled from other
provinces (as were mitimaes ), producing resources to maintain the legacy of
202 K. NOACK AND K. NOWACK
the deceased nobles. Sometimes, subsequent rulers augmented the number
of yanacona of their predecessors in exchange for the political support of
the descent group. Machu Picchu, for instance, housed dwellings for 500 to
750 nobles and servants, apparently yanacona. Bioarcheological studies of the
laborers at Machu Picchu show that most spent their childhood in remote
coastal or high-altitude regions, meaning that they had been resettled from
diverse locations of the Inca state.27
On other estates, the yana settlers lived in small villages and were super-
vised by other yanacona, and sometimes by inhabitants of villages nearby.28
Cheqoq, part of Yucay estate of Huayna Capac’s lineage, for instance, was
mostly inhabited by yanacona (who also were resettled, that is, mitimaes )
rather than nobles. During Huayna Capac’s rule, 150,000 mita workers were
temporarily transferr ed to channel the river and create irrigable land for maize
production in the Yucay valley, as well as 2,000 yanacona, one half coming
from the northern and the other half coming from the southern part of the
state, with very diverse geographical and cultural origins.29 At the same time,
his grandfather Topa Inca’s descendants in Cuzco disposed of the services
of another 1,000 yanacona.30 Apart from fieldwork, some yanacona on the
estates had specialized occupations and worked as weavers, salt makers, or
builders.31 In this case, the yana category was overlapping with the camayoc
category, occupational specialists. In addition to the dependents in service of
the Inca elite, major supernatural beings both in Cuzco and the provinces
were endowed with yanacona, although in lower numbers than those ser ving
the Inca elite, together with fields and camelid herds.32
Service for the Inca elite was not associated with great material benefits,
as archaeological studies show.33 Practically no information is found about
the family life, social organization, work conditions, lifestyle, and religious
practices of the yanacona on the estates. It was the pre-Inca institution of
yanacona that resulted in their being able to marr y and form households.34
As we already know that the father of don Joan Puyquin from the Chachapoya
region had been captured in war when he was a child, and became a yana, but
we do not know anything about don Joan Puyquin’s mother. Was she a captive
of the same campaign, the child of a yanacona couple with similar or different
regional origins, or did she descent from the local non-yana population? And
who became don Joan Puyquin’s wife? Offspring inherited the status of a yana
and had to continue working for their elite lords. Were male and female able to
find partners among the yanacona of the same master? If marriages between
members of different yanacona groups occurred, where did the couple live,
and for whom did they and their eventual children work?
As yanacona can be defined as people detached from their kinship group
(ayllu), this nearly always meant that they had to leave their homeland and lost
their familial and communal networks. They often became members of a larger
and more anonymous workforce, and it can be supposed that their quality of
life was lowered, and that individuals lost the ability to make choices about
significant aspects of their life.35 However, being a yana did not mean that
11 THE QUESTION OF SLAVERY IN THE INCA STATE 203
these individuals were publicly degraded. No specific rituals of humiliation
are known to mark the transition to yana status and, although the Inca state
valued such symbols to make a status visible, no outward signs on the body
or a special attire denoted yana status.36 In contrast to mitimaes , no special
rules applied to the yanacona, except that legally yanacona fell under the juris-
diction of their Inca superiors, not of provincial rulers.37 Yanacona apparently
could not be killed randomly or as sacrificial victims, and no special corporal
punishments were reserved for them.38 Most importantly, they retained the
right to have a family.39
A very different group of yanacona were those taken from the provincial
elites to serve the Inca ruler directly. Their office was to work in ruler’s house-
hold, overseeing the preparation of food and caring for his clothes, insignia,
and weapons, tasks associated with a great responsibility, since everything in
contact with the Inca’s body could endanger his ritual purity and be used
for harming him by magical means.40 Their relationship with the Inca ruler
could become very intimate, best exemplified by one of the already mentioned
Chachapoya yanacona, Chuquimis, who sucked blood from a wound of Inca
Huayna Capac. As a reward for his service, as it is claimed, Chuquimis later
became a local ruler in his home province.41 It is, however, more likely that the
time spent in the Inca ruler’s household prepared young men for future lead-
ership positions in provinces. During their turn in the Inca ruler’s household,
they learned how the administration of the empire functioned and acquired
valuable contacts among its elite—this is what we understand as patronage.
Furthermore, having served the Inca ruler was prestigious and enhanced their
status after returning home. For the Inca, the presence of offspring of local
elites in the imperial center facilitated the flow of information between them
and the provinces, gave the ruler and his advisors insights into the politics and
personalities of the local elites, and continued to weave the network what was
becoming and strengthening the “fragile” Inca state.42
When local rulers from the central coast of Peru like Taulichusco of the
Rimac valley and Francisco Yayvi of the Chillón valley described themselves as
yanacona of Inca rulers, they refer to a patronage relationship resulting from
their stay at the Inca court. They ended up governing members of their own
local group. Most yanacona described as having acquired local rulership were
not outsiders, but members of the group they came to rule. There is no clear
evidence that the appointment of a former Inca yana as a local ruler can be
interpreted as the start of an administration based on merit and abilities.43
A possible exception is the case of don Alonso Condor, interviewed during
the General Visitation of viceroy Toledo. His father (or perhaps grandfather?)
came from the Soras province (today Ayacucho, Peru) and was brought to the
Xaquixaguana valley close to Cusco by the Inca ruler Pachacuti, to become the
curaca of Pomaguanca, where mitimaes had been relocated to.44 As Pacha-
cuti had subjugated the Soras, don Alonso’s progenitor might have been a war
captive. When Alonso Condor’s father was dying, he had entrusted Alonso and
his elder brother to the Inca ruler Huayna Capac. Alonso Condor became a
204 K. NOACK AND K. NOWACK
page (a term applied to a youth serving a member of the elite) of Huayna
Capac, as probably his brother as well. Apparently, the brother was later
installed as a curaca in Tomebamba (perhaps heading another group of miti-
maes with unknown regional extraction) in Ecuador, while Alonso Condor
came to rule Pomaguanca as the successor of his father.45 So Huayna Capac
took the fatherless sons of a loyal subject lord into his household to serve him
and train them for future positions in the Inca hierarchy and administration.
In this example, members of the provincial elites advanced from a position in
the Inca ruler’s household to local rulership of mitimaes.
The yanacona ser ving the local elite in the provinces were mostly active
in either agricultural work or herding, some were craft specialists (camayoc),
and a few supervised specific tasks or people.46 Depending on the nature of
their occupations, these men and women lived in or near the households of
their elite overlords or at separate locations, which probably determined the
degree of autonomy they enjoyed. In general, they remained in their home-
land, in contrast to Inca yanacona, but like them lost their ayllu affiliation and,
possibly, had to identify with the elite’s descent group and very likely partici-
pate in its religious practices. The case of Alonso Poma Cochache (see below)
shows that some yana superiors physically mistreated their dependents; other
hints that (physical?) intimidation was sometimes used can be found as well.47
The number of yanacona a local leader could claim depended on mostly
unknown historical conditions and customs, and the proportion of yana-
cona in relation to the number of a leader’s subjects varied. In the Bolivian
highlands, the rulers of 10,000 households stated that they had the right
to have 50 yanacona, that is half a percent of their subjects became their
dependents. Additional yanacona in decreasing numbers were allocated to the
lower-ranking leaders of units of 5000, 1000, 500, and 100 taxpaying house-
holds, so that the total of yanacona expected to be provided by these leaders’
subjects might have amounted to five percent of all households. It cannot be
said if these numbers reflect pre-Inca ratios or if the Inca put a limit on the
number of yanacona local leaders could recruit (the decimal units refer to Inca
division of the subject population).48 In an example from highland Ecuador,
dependents constituted up to ten percent of the population, which seems to be
on the higher end of the range. Murra discusses the question of the yanacona’s
share of the total population in the context of whether they can be considered
slaves. According to his calculation using the example of the visitation report
of the Lupaca, yanacona made up a share of two to three percent. He left
open the question of whether such a society could be referred to as a slave
society.49
Leaving the
Ya n a
Status
As in the case of becoming a yana, which varied depending on whether they
were on-going local or regional practices stemming from the pre-Inca period
11 THE QUESTION OF SLAVERY IN THE INCA STATE 205
or practices related to the Inca expansion and state-building process, the possi-
bilities for leaving the status, if that was at all possible, varied as well. However,
there are even fewer sources on this question than regarding the first two
topics. Criteria that determined these differences were the social origins of
the yanacona, i.e., whether they were commoners from the communities or
belonged to the regional or local elites themselves, and whom they served,
i.e., superiors at the level of the Inca state or at a subordinate level of the local
political elite.
It is mentioned that yanacona of the Inca elite meant a lifelong ser vice
which to our knowledge ended only in the case of severe political upheaval.
For the yanacona of Topa Inca’s descendants, the outcome of the succes-
sion war between his grandsons Atahualpa and Huascar for the position of
Inca ruler resulted in the dissolution of the descent group, and the death of
many of their members. Perhaps it also gave some of the yanacona of that
kinship group the opportunity to escape their status. In a similar manner, the
yanacona on a newly founded estate of Huascar were sent home by the mili-
tary leaders of the opposite party during that war, so that they could revert
to be ordinary community tribute payers.50 Other sources referring to yana-
cona of commoner background serving the Inca elite make clear that people
designated to become yanacona remained so for the rest of their lives. This is
true for those drafted from the local population as well for example for those
punished to become yanacona forcriminaloffensesastheft,especially toward
a property owner, generally a person of higher status.
In general, there was no need for “manumission” for those yanacona of
the Inca elite as in the case of slaves in other parts of the world. Yanacona
never completely lost their position in their provincial community or the Inca
state, and therefore they were not viewed as outsiders. They did not need to
be reintegrated into Inca society, as in the case of slaves who were often seen
as standing outside of the society where they had brought to serve.
The experience of elite yanacona for the Inca r uler was the opposite. For
them, service at the ruler’s court was a transition in their life, a form of
training for their future positions as provincial leaders. They continued to
refer to themselves as yanacona of the ruler, but this meant a prestigious
status denoting their close political ties to the Inca sovereign, not a contin-
uance of personal service. It is not known if this special relationship could be
transmitted to the successors of rulers who had been yanacona.
Colonial-era wills of local elite and documents from tribute inspections
show that yanacona service to provincial elites—as part of local social orga-
nization—was permanent. Only the case of Alonso Julca Guaman, who had
been physically mistreated, as mentioned above, indicates that a yana could
leave his position. During a quite early Spanish inspection of the local popu-
lation in central Peruvian Huaraz (1558), the inhabitants were asked if their
rulers abused them in any way. Alonso Julca Guaman, one of the witnesses,
explained that he had been given by his superior (“principal”) to a higher-
level ruler, don Alonso Poma Cochache, as a yana, and complained about his
206 K. NOACK AND K. NOWACK
treatment. He had been physically abused by Alonso Poma Cochache, while
customary clothing and food were withheld from him.51 As a consequence,
Alonso Julca Guaman left the service of his abusive lord, apparently without
any punishment.
The previously mentioned criteria that mark the differences in the possi-
bilities of leaving yana status also make it clear that there are differences in
terms of the heritability of that status. These are determined by the same
principles. The commoner yana status will have been hereditary. Yanacona
lived in families. The children of those families continued to serve as yana-
cona as adults. As we have seen, if there were no children, the home province
had even to send a replacement. It will have been different in cases that we
call patronage. Since the yanacona, who came from the provincial elites, went
through a kind of educational process at the ruler’s home, it was completed
at a certain point. These differences regarding the heritability of the yanacona
probably also explain their different, even contradictory presentation in the
literature.52 The problem is that the differentiations in the large group of the
yanacona had not been perceived and therefore was not addressed.
Conclusion
Since the first research on yanacona was based on little and one-sided source
material, it had come to the assumption that the yanacona were slaves. In
the context of European historical concepts, an Inca state without slaves was
inconceivable. More recent research since the 1960s, such as in the Andean
Studies initiated by John Murra, based on the analysis of new source mate-
rial like for example the visitation reports, brought new insights and made
clear that the yanacona were subject to a strong form of asymmetrical depen-
dency, but could not be considered slaves according to the terminology
and the models of economic and societal evolution developed in Euro-
pean historiography.53 After the 1980s, the topic of yanacona was addressed
only sporadically. For example, in the Oxford Handbook of the Incas (2018)
the social and labor-related categories of dependency in the Tawantinsuyu,
including the yanacona, are hardly dealt with in detail.
Andeanist John H. Rowe clarified that the categories of mitimaes, yana-
cona,and camayoc were “not three contrasting categories of men,” but that
they could overlap. As we have seen, Inca subjects could be everything
together, a combination of two or only one category. “Which status or statuses
are mentioned, may depend on the context of discourse.” Following Rowe,
the difference between them should not be described as a degree of servility
or dependency. Rather, the distinction between social and labor-related cate-
gories “was not more or less freedom but more or less access to honor and
privilege,” the source of which was the Inca government.54 Accordingly, it was
the status of the dependents in Inca society, not the degree of dependency,
that was directly related to the privileges to be transferred by the Inca ruler
to his subjects. As we have seen, the line between those who were endowed
11 THE QUESTION OF SLAVERY IN THE INCA STATE 207
with privileges and those who were not passed right through the group of
yanacona.
Instead of assuming greater or lesser dependency of groups or individuals,
they are distinguished by the privileges they have received from the Inca ruler,
who else they had to serve like local rulers or supernatural beings, by the
respective work they had to perform and by the characteristics associated with
the yanacona such as heritability, the possibilities to leave the status or not,
i.e., the degree of their separation and uprooting from their original groups,
and others.
This makes clear that the complexity and diversity of the dependent popu-
lation, which was involved in multiple areas of the Inca labor system, implied
a whole range of different types of dependency and of different mechanisms of
direct and indirect coercion. During the colonial period, the Spanish Crown
“saw great benefit in persisting with and resignification servitude practices that
were later institutionalized as yanaconazgo,” as Paola Revilla Orías has argued.
Thus, the “yanaconazgo institution was preserved and re-signified along colo-
nial period,” colonial yanacona gradually lost their fr eedoms, like the liberty
of movement, in that they found themselves more and more tied to the land
for whose owners they worked, and became “almost slaves.”55
Notes
1. John Murra, Formaciones económicas y políticas en el mundo andino (Lima:
Instituto de Estudios Peruanos, 1975), 226.
2. Karoline Noack, “Mobilization as Dependency: the Case of Mitimaes of the
Inka State as a Hotspot of Early Glocalization” in Comparative and Global
Framing of Enslavement , eds. Youval Rotman, Ehud Toledano and Rachel
Zelnick-Abramovitz (Berlin: De Gruyter, 2023), forthcoming.
3. For a short summary, see Teresa Vergara, “Tahuantinsuyo: el mundo de los
Incas,” in Historia del Perú (Barcelona: Lexus editores, 2000), 282.
4. Diego Gonçález Holguín, Vocabulario de la lengua general de todo el Peru
llamada lengua Qquichua, o del Inca (Quito: Corporación Editora Nacional,
1993 [1608]), 363-64.
5. John Verano and Douglas H. Ubelaker, Disease and Demography in the
Americas (Washington: Smithsonian Institution Press, 1992), 16.
6. Murra, For maciones económicas y políticas en el mundo andino, 59-115.
7. Gonçález Holguín, Vocabulario de la lengua general de todo el Peru llamada
lengua Qquichua, o del Inca, vol. 3: bk. 1, 286; bk. 2, 450, 512; Ludovico
Bertonio, Vocabulario de la lengua Aymara. [Juli (Provincia de Chucuito)
(Cochabamba: CERES/IFEA/MUSEF [1612] 1984), vol. 2, 242, 316, 328,
and vol. 1, 122, 223.
8. Betanzos, Suma y Narración de los Incas, pt. 1, ch. 20, 140; pt. 1, ch. 21,
150, pt. 1, ch. 48, 234; pt. 2, ch. 1, 244; María Rostworowski, Histor y of the
Inca Realm, translated by Harry B. Iceland (Cambridge: Cambridge University
Press, 1999), 176–77.
9. Hernando de Santillán, “Relación del origen, descendencia, política y gobier no
de los Incas,” in Crónicas peruanas de interés indígena, ed. Francisco de Esteve
Barba (Madrid: Ediciones Atlas, 1968 [1563]), núm. 36, 114.
208 K. NOACK AND K. NOWACK
10. Terence D´Altroy, “Inca Political Organization, Economic Institutions, and
Infrastructure,” in The Oxford Handbook of the Incas, ed. Sonia Alconini and
R. Alan Covey (New York: Oxford University Press, 2018), 218–19. See also
Inge Schjellerup, “Inca Transformations of the Chachapoya Region,” in The
Oxford Handbook of the Incas, ed. Sonia Alconini and R. Alan Covey (New
York: Oxford University Press, 2018), 435–51.
11. Probanza de Martin and Francisco de Ampuero, in: Informacion hecha en
cumplimento de una cedula se su magestad en su rreal audiencia de la Ciudad
de los Reyes a pedimiento de Francisco de Ampuero vezino della e de Martin
de Anpuero y Francisco de Ampuero sus hijos (Archivo General de Indias:
Lima, 1572), 38v.
12. D´Altroy, “Inca Political Organization, Economic Institutions, and Infrastruc-
ture,” 218–19; Carolyn Dean, Inka Bodies and the Body of Christ: Corpus
Christi in Colonial Cusco, Peru (Durham, NC; London: Duke University Press,
1999), 186–92.
13. Informaciones hechas por orden de don Francisco Toledo [1570–72], in Don
Francisco de Toledo. Supremo organizador del Perú. Su vida, sus obras (1515–
1582), 2. Sus informaciones sobre los Incas (1570–1572), ed. Roberto Levillier
(Buenos Aires: Espasa-Calpe, 1940), 55–57.
14. Garcilaso de la Vega, Comentarios Reales de los Incas (México: Fondo de
Cultura Económica, [1609] 1995), bk. 9, ch. 7, 574.
15. Santillán, “Relación del origen, descendencia, política y gobierno de los Incas,”
núm. 36, 114.
16. Cristóbal de Castro y Diego de Ortega Morejón, “Relaçion y declaraçion
del modo que este valle de chincha,” in Quellen zur Kulturgeschichte des
präkolumbischen Amerika, ed. Hermann Trimborn (Stuttgart: Strecker und
Schröder, [1558] 1936), 237–41; Visita de Huánuco, ed. John Murra
(Huánuco: Universidad Nacional Hermilio Valdizán, [1562] 1967), 237–40.
17. Betanzos, Suma y Narración de los Incas, pt. 1, ch. 41, 217; ch. 43, 220;
“Informaciones hechas por orden de don Francisco Toledo” [1570–72], 112–
13, 142; Testimonio dado por Benito de la Peña [1552], in “Documentos
sobre Yucay en el siglo XVI” ed. Horacio Villanueva Urteaga, Revista del
Archivo Histórico del Cuzco, 13 (1970): 149–52.
18. Catherine Julien, “How Inca Decimal Administration Worked,” in Ethnohistory
35 (Summer 1988): 257–79.
19. Betanzos, Suma y Narración de los Incas , pt. 1, ch. 21, 148–50; pt. 1, ch. 21,
f. 54v.
20. Visita de Huánuco [1562], 108, 127, 146, 168. See also John Murra, “New
Data on Retainer and Ser vile Populations in Tawantinsuyu,” in Actas y Memo-
rias, XXXVI Congreso Internacional de Americanistas , vol. 2 (Sevilla: Cathólica
Española, 1966), 40–41; Visita de los valles de Songo [1568–70], ed. John
Murra (Madrid: ICI/Instituto de Estudios Fiscales, 1991); Testamento de
Melchior Carorayco [1565], Estratificación social y el hatun curaca en el mundo
andino, in Ensayos de historia andina. Elites, etnías, recursos, ed. María Rost-
worowski (Lima: Instituto de Estudios Peruanos, 1989), 239; Pleitos por los
cacicazgos de Torata y Moquegua [1590–94], in La visita de Juan Gutiérrez
Flores a Colesuyo y Pleitos por los cacicazgos de Torata y Moquegua,ed. Teresa
Cañedo-Argüelles (Lima: PUCP, 2005), 25–47, 148.
11 THE QUESTION OF SLAVERY IN THE INCA STATE 209
21. “El Memorial de los Mallku y principales de la provincia de los Charcas,” in
Qaraqara-Charka — Mallku, Inca y Rey en la provincia de Charcas (siglos XV–
XVII): Historia antropológica de una confederación aymara, ed. Tristan Platt,
Thérèse Bouysse-Cassagne and Olivia Harris (La Paz: IFEA, 2006), 834–36.
22. Visita hecha a la provincia de Chucuito, ed. Waldemar Espinoza Soriano (Lima:
Casa de la Cultura del Perú, [1567] 1964), 111. See also Murra, “New Data
on Retainer and Servile Populations in Tawantinsuyu,” 37–39.
23. “Visita de los repartimientos de Ichoc y de Allauca Guaraz…” [1558], in
Huaraz: Poder, sociedad y economía en los siglos XV y XVI. Reflexiones en torno
a las visitas de 1558, 1594 y 1712, ed. Waldemar Espinoza Soriano (Lima:
UNMSM, 1987), 113.
24. “Pleitos por los cacicazgos de Torata y Moquegua” [1590–94], 25–47, 148;
Testamento de Gonçalo Taulichusco, ed. Guillermo Lohmann Villena, Revista
del Archivo General de la Nación, 7 (Lima, [1562] 1984), 270–71; Testamento
de doña Francisca Sina Sigchi, in Sancho Hacho. Un cacique mayor del siglo
XVI, ed. Udo Oberem (Quito: CEDECO / Abya-Yala, [1580] 1993), 137;
Visita de Huánuco [1562], 196–201; see Murra, “New Data on Retainer and
Servile Populations in Tawantinsuyu,” 41.
25. See for example Autos seguidos por el capitan don Martin Garcia de Loyola y
la coya doña Beatriz… [1574] 2008), in Imperial Transformations in Sixteenth-
Century Yucay, Peru, ed. Alan Covey and Donato Amado González (Ann
Arbor: Michigan University Press), 78; and “Informaciones hechas por orden
de don Francisco Toledo”.
26. Kylie Quave, “Royal Estates and Imperial Centers in the Cuzco Region,” in
The Oxford Handbook of the Incas, ed. Sonia Alconini and Alan Covey (New
York: Oxford University Press, 2018), 102–104.
27. Lucy Salazar, “Machu Picchu. Mysterious Royal Estate in the Cloud Forest,”
in Machu Picchu. Unveiling the Mystery of the Incas Burger, ed. Richard Burger
and Lucy Salazar (New Haven: Yale University, 2004); and Lucy Salazar and
Richard Burger “Lifestyles of the Rich and the Famous: Luxury and Daily
Life in the Households of Machu Picchu’s Elite,” in Palaces of the Ancient
New World, ed. Susan T. Evans and Joanne Pillsbur y (Washington: Dumbarton
Oaks Research Library and Collection, 2009), quoted in Quave “Royal Estates
and Imperial Centers in the Cuzco Region,” 108–109.
28. Betanzos, Suma y Narración de los Incas , pt. 1, ch. 43, 220; for yana as rulers
of other yanacona see “Informaciones hechas por orden de don Francisco de
Toledo,” 112–18; for adjacent groups overseeing the yanacona see Probanza
de Sancho Usca Paucar y Alonso Auca Poma [1569], in “¿Incas de privilegio?
La probanza de Sancho Usca Paucar y Alonso Auca Poma, caciques principales
de Maras y Mullaca (4–12 de mayo de 1569),” ed. Laurent Segaline, Revista
Andina 55 ([1569] 2017): 47.
29. Betanzos, Suma y Narración de los Incas, pt. 1, ch. 43, 220; Autos seguidos
por el capitan don Martin Garcia de Loyola y la coya doña Beatriz…,” 117,
206–96 y “Testimonio dado por Benito de la Peña” [1552].
30. Cabello Valboa, Miscelánea Antártica, bk. 3, ch. 31, 529. See also Sarmiento
de Gamboa, “Segunda parte de la Historia general llamada Indica… [1572],
in Geschichte des Inkareiches, ed. Richard Pietschmann (Berlin: n.p. 1906) ch.
66, 123; Martín de Murúa, Historia general del Peru, ed. Manuel Ballesteros
(Madrid: Historia, [1613] 1987), bk. 1, ch. 57, 202.
210 K. NOACK AND K. NOWACK
31. “Autos seguidos por el capitan don Martin Garcia de Loyola y la coya doña
Beatriz…,” 78; Informaciones hechas por orden de don Francisco de Toledo, 108–
18, 125–33.
32. Betanzos, Suma y Narración de los Incas , pt. 1, ch. 11, 89, 91; ch. 45, 226. See
also Castro/Ortega Morejón, “Relaçion y declaraçion del modo que este valle
de chincha…,” 240–41; John Rowe, “Inca Policies and Institutions Relating
to the Cultural Unification of the Empire,” in The Inca and Aztec states The
Inca and Aztec states 1400–1800, ed. George A. Collier, Renato I. Rosaldo and
John D. Wirth (New York: Academic Press, 1982), 101; Pedro Cieza de León
[1548–54], El señorío de los Incas , ed. Manuel Ballesteros (Madrid: Historia 16,
1985) ch. 28, 101–102; Titu Cusi Yupanqui, “Ynstruçion … para el muy ilustre
señor el liçençiado Lope Garcia de Castro,” in Titu Cusi Yupanqui, History
of how the Spaniards arrived in Peru, ed. Catherine J. Julien (Indianapolis:
Hackett Publishing, [1570] 2006), 128–31; Cristóbal de Albornoz, “Instruc-
cion para descubrir todas las guacas del Piru y sus camayos y haciendas, in Pierre
Duviols “Albornoz y el espacio ritual andino prehispánico,” Revista Andina 2,
no. 1, [1583–84] 1984: 198; Marco Curatola Petrocchi, “La función de los
oráculos en el Imperio Inca,” in Adivinación y oráculos en el mundo andino
antiguo, ed. Marco Curatola Petrocchi and Mariusz S. Ziółkowski (Lima: Insti-
tuto Francés de Estudios Andinos/Pontificia Universidad Católica del Perú,
2008).
33. Di Hu and Kylie E. Quave, “Prosperity and Prestige: Ar chaeological Reali-
ties of Unfree Laborers under Inka Imperialism,” Journal of Anthropological
Archaeology 59, no. 2 (September 2020): 10.1016/j.jaa.2020.101201; Ian
Farrington, Cusco: Urbanism and Archaeology in the Inka World (University
Press of Florida: 2013), 234–49.
34. Murra, For maciones económicas y políticas en el mundo andino, 229.
35. Noel Lenski, “Framing the Question: What Is a Slave Society?,” in What Is
a Slave Society? The Practice of Slavery in Global Perspective, ed. Noel Lenski
and Catherine M. Cameron (Cambridge: Cambridge University Press, 2018),
49–53.
36. Fernando Santos-Granero, “Slavery as Structure, Process, or Lived Experience,
or Why Slave Societies Existed in Precontact Tropical America,” in What Is a
Slave Society? The Practice of Slaver y in Global Perspective, ed.NoelLenskiand
Catherine M. Cameron (Cambridge: Cambridge University Press, 2018), 201–
02, 208–15; Christina Snyder “Native American Slavery in Global Context,” in
What Is a Slave Society? The Practice of Slavery in Global Perspective, ed. Noel
Lenski and Catherine M. Cameron (Cambridge: Cambridge University Press,
2018), 180–82.
37. Castro/Ortega Mor ejón, “Relaçion y declaraçion del modo que este valle de
chincha… [ 1558],” 240; José Toribio Medina, “Probanza sobre la forma y
orden que los Ingas tenían en el juzgar [1582],” in La imprenta en Lima
(1584–1824), 1 (Santiago de Chile: n.p. 1904), 191.
38. Noel Lenski, “Ancient Slaveries and Modern Ideology,” in What Is a Slave
Society? The Practice of Slaver y in Global Perspective, ed. Noel Lenski and
Catherine M. Cameron (Cambridge: Cambridge University Press, 2018)2018),
139; Catherine M. Cameron, “The Nature of Slavery in Small-Scale Societies,”
in What Is a Slave Society? The Practice of Slavery in Global Perspective, ed. Noel
11 THE QUESTION OF SLAVERY IN THE INCA STATE 211
Lenski and Catherine M. Cameron (Cambridge: Cambridge University Press,
2018), 162–63.
39. “Autos seguidos por el capitan don Martin Garcia de Loyola y la coya doña
Beatriz…”; Betanzos, Suma y Narración de los Incas , pt. 2, ch. 9, 267–68.
40. Informacion sobre las chacras… que cultivaban los mitimaes … en la provincia
de Abancay, in “Colonias de mitmas múltiples en el valle de Abancay. Siglos
XV y XVI. Una información inédita de 1575 para la etnohistoria andina,” ed.
Waldemar Espinoza Soriano, Revista del Museo Nacional, 39 [1575] 1973),
278; “Informaciones hechas por orden de don Francisco de Toledo,” 78,
113, 115, 152; Pleito de don Pedro Ocxaguaman [1562–64], in La crónica
de Ocxaguaman, ed. Jorge Zevallos Quiñones (Trujillo: Fundación Alfredo
Pinillos Goicochea, 1994), 61; Primera información hecha por don Juan
Colque Guarache, in “El reino aymara de quillaca-asanague, siglos XV y XVI,”
ed. Waldemar Espinoza Soriano, Revista del Museo Nacional 45 ([1575] 1981),
241–42.
41. Memoriales… de Leimebamba y Cochabamba, in “Los señoríos étnicos de
Chachapoyas y la alianza hispano-chacha, siglos XV–XVI,” ed. Waldemar
Espinoza Soriano, Revista Histórica, 30 ([1572–74] 1967): 294.
42. Norman Yoffee, “Introducing the Conference: There Are No Innocent Terms,”
in The evolution of Fragility: Setting the Terms, ed. Norman Yoffee (Cambridge,
UK: McDonald Institute for Archaeological Research, 2019), 1 –8; Noack,
“Mobilization as Dependency,” forthcoming.
43. Prouança de don Gonçalo [1555], in Dos probanzas de Don Gonzalo, Curaca
de Lima (1555–1559), ed. María Rostworowski, Revista Histórica, 33, 1981–
82, 115, 120; Tierras y chacaras de coca del valle de Quivi [1558–1570],
in Conflicts over Coca Fields in XVIth-Centur y Per u, ed. María Rostworowski
(Ann Arbor: University of Michigan, 1988), 105; Rowe, “Inca Policies and
Institutions Relating to the Cultural Unification of the Empire,” 100–02;
Rostworowski, History of the Inca Realm, 148–150.
44. Alan Covey and Kylie Quave, “The Economic Transformation of the Inca
Heartland (Cuzco, Peru) in the Late Sixteenth Century,” in Comparative
Studies in Society and History, 59, no. 2 (2017): 277–309.
45. “Informaciones hechas por orden de don Francisco de Toledo,” 108; Betanzos,
Suma y Nar ración de los Incas., pt. 1, ch. 18, 126–31; See also Rowe, “Inca
Policies and Institutions Relating to the Cultural Unification of the Empire,”
98.
46. Visita de Huánuco [1562], 108, 127, 146, 168. See also Murra, “New Data
on Retainer and Servile Populations in Tawantinsuyu,” 40–41; Visita de los
valles de Songo [1568–70]; “Testamento de Melchior Carorayco” [1565] 239;
“Pleitos por los cacicazgos de Torata y Moquegua” [1590–94], 25–47, 148.
47. “Testamento de doña Francisca Sina Sigchi” [1580], 137.
48. El Memorial de Charcas [1582], 833–34 y “Probanza de don Fernando Ayavire
Cuysara” [1583–84], 885, in Qaraqara-Charka. Mallku, Inca y Rey en la
provincia de Charcas (siglos XV–XVII), ed. Tristan Platt, Thérèse Bouysse-
Cassagne and Olivia Harris (La Paz: Instituto Francés de Estudios Andinos,
2006); Visita … del valle de los Chillos [1551–59], in Visita y numeración de
los pueblos del valle de los Chillos 1551–1559, ed. Cristóbal Landázuri (Quito:
Abya-Yala /Marka 1990); and Visita de los valles de Songo [1568–70].
212 K. NOACK AND K. NOWACK
49. Visita … del valle de los Chillos [1551–59]. See also Frank Salomon, Native
Lords of Quito in the Age of the Incas (Cambridge, MA: Harvard Univer-
sity Press, 1986), 127–29 and Murra,”New Data on Retainer and Servile
Populations in Tawantinsuyu,” 39–40.
50. Información sobre las tierras del inca … en … Pinagua [1571], in El habitat
de la etnia Pinagua, siglos XV y XVI ed. Waldemar Espinoza Soriano, Revista
del Museo Nacional 40 (1974): 201–203, 206–207.
51. “Visita de los repartimientos de Ichoc y de Allauca Guaraz…” [1558], 113.
52. Rowe, “Inca Policies and Institutions Relating to the Cultural Unification of
the Empire,” 100; Quave, “Royal Estates and Imperial Centers in the Cuzco
Region,” 114; D’Altroy, “Inca Political Organization, Economic Institutions,
and Infrastructure,” 218–19.
53. Paola Revilla Orías, Entangled Coercion: African and Indigenous Labour in
Charcas (16th–17th Century) (München: De Gruyter Oldenbourg, 2021), 119.
54. Rowe, “Inca Policies and Institutions Relating to the Cultural Unification of
the Empire,” 96–97.
55. Revilla Orías, Entangled Coercion, 28 (fn. 86), 120.
Further Readings
Alconini, Sonia and R. Alan Covey. The Oxford Handbook of the Incas.New York:
Oxford University Pr ess, 2018.
D’Altroy, Terence N. The Incas. Malden, Massachusetts: Willey/Blackwell, 2015.
Dillehay, Tom D., and Steven A. Wernke. “Fragility of Vulnerable Social Institutions in
Andean States,” in The Evolution of Fragility: Setting the Terms, edited by Norman
Yoffee, 9–23. Cambridge, UK: McDonald Institute for Archaeological Research,
University of Cambridge, 2019.
Guaman Poma de Ayala, Felipe. Nueva corónica y buen gobierno, edited by John V.
Murra, Rolena Adorno and Jorge L. Urioste. Madrid: n.p. [1615] 1987.
The Guaman Poma Website http://www5.kb.dk/permalink/2006/poma/info/en/
frontpage.htm
Julien, Catherine. Reading Inca History. Iowa City: University of Iowa Press, 2000.
Murra, John V. The Economic Organization of the Inka State. Greenwich, Conn.: Jai
Press, 1980.
Niles, Susan A. The shape of Inca History: Narrative and Architecture in an Andean
Empire. Iowa City: Univ. of Iowa Press, 1999.
Pease G. Y., Franklin. The Incas , translated by Simeon Tegel. Lima: PUCP, 2011.
Rostworowski de Diez Canseco, María. History of the Inca Realm, translated by Harry
B. Iceland. Cambridge: Cambridge University Press, 1999.
Rowe, John H. “Inca Policies and Institutions Relating to the Cultural Unification of
the Empire,” in The Inca and Aztec states, 1400–1800: Anthropology and History,
edited by George A. Collier, Renato I. Rosaldo, and John D. Wirth, 93–118. New
York/London: Academic Press, 1982.
11 THE QUESTION OF SLAVERY IN THE INCA STATE 213
Open Access This chapter is licensed under the terms of the Creative Commons Attri-
bution 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits use, sharing, adaptation, distribution and reproduction in any medium
or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were
made.
The images or other third party material in this chapter are included in the chap-
ter’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the chapter’s Creative Commons license and
your intended use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright holder.
CHAPTER 12
Injection: A Gender Perspective on Domestic
Slavery
Ruth Mazo Karras
This “injection” comes from the point of view of a historian of slavery turned
historian of sexuality. The history of sexuality sits very uneasily with the history
of enslavement, however, because of the problem of agency and its relation-
ship to consent. Historians must not erase the personhood of the enslaved
and their ability to make choices; but the realm within which they make those
choices is so constrained that recognizing their agency can be akin to blaming
the victim or implying consent when it cannot really be given. Can we really
say that every sexual encounter between an enslaved person and their enslaver
is an act of rape, if the enslaved person has been the one to initiate it in the
hope that it could better their life in some way? Indeed, we can, by analogy
with the laws about rape of minors or others who are incapable of consent.
Even if a 16-year-old initiates a sexual encounter or agrees to one, sex with
them is legally rape because their consent is not valid. Of course, sex with
one’s own human property was not legally considered rape under Western
and some other legal systems. The enslaved person had no ability to withhold
consent from their enslaver, in the same way a wife until quite recently (1990
in Ireland; 1991 in England; and 1993 in some US states) could not be raped
by her husband because her marital vows constituted permanent consent. But
although sex between enslaved people and their enslavers would not be consid-
ered rape under the law of the time, it is entirely reasonable for us to call it
R. M. Karras (B)
Trinity College Dublin, Dublin, Ireland
e-mail: ruth.karras@tcd.ie
© The Author(s) 2023
D. A. Pargas and J. Schiel (eds.), The Palgrave Handbook of Global Slavery
throughout History, https://doi.org/10.1007/978-3-031-13260-5_12
215
216 R. M. KARRAS
that. Even if the encounter was not in proximate terms violent or coercive,
even if the enslaved person participated enthusiastically, the coerciveness of the
entire situation was such that they cannot be considered independent actors.
This essay focuses largely on the actions and choices of enslavers and treats
enslaved people as victims. This does not deny their humanity, their choices,
or their desires, but they were nevertheless victims of a violent system in which
they had no redress. There are three factors in the sexual exploitation of the
enslaved by their enslavers, although they overlap a good deal. One is, simply,
desire. If enslavers wished to have a variety of sexual partners—or even one
new sexual partner—systems of enslavement allowed them to choose from
among their enslaved people, or to purchase additional enslaved people, with
whom to have sex. Already the enslaver’s property, they did not need to be
courted, or paid, or endowed. They were practically and legally available, and
if there were no social sanctions there was no reason for the enslaver not to
engage in sex with them if he so desired (usually he, because sex between free
women and enslaved men was a serious offense in most cultures).
But, of course, there is no such thing as simple desire without issues of
power being involved, and there is a significant overlap between the factors of
desire and domination. The assumption that whenever one feels the wish to
have a sexual encounter, that encounter is immediately available is a result
of a huge power differential and unquestioned privilege. For men of the
enslaving class to assert that privilege is an act of domination over people
of other groups, notably people of whatever ethnic or racial group or social
class was being enslaved in that particular culture. It may not feel that way to
the one who holds the power, but privilege is not any less the case for being
unexamined.
Sex as domination is both the result of unexamined power and privilege
that manifests in an individual, and the result as well as the cause of larger
group processes. Sex with one’s enslaved women or men could carry with it
a message of dominance over other men: men who are raped; enslaved men
who are less masculine because they are unable to protect the victims; men
whose wives, daughters, sons, sisters, brothers have been captured and raped
(even if the men in question are dead, their posthumous reputation may still be
dominated); men within the enslaving society who are not wealthy or powerful
enough to have slaves or perhaps even any sex partners, and are dominated by
the assertion of a monopoly. These messages need not always be deliberate,
but sometimes they are. And, of course, for a man to have multiple enslaved
sex partners can also be an assertion of domination over other women as well,
that is, his free sex partners, whether wives or concubines, whom he may be
trying to make jealous or more eager to retain his favor.
There is a third factor in men’s having sexual relations with their enslaved
women, which may not be obvious to those of us living in a culture where
the fathering of childr en out of wedlock is often secret or shameful and where
birth control is commonly practiced: precisely the wish to beget children. Even
when law or religious teaching r estricted the inheritance rights or the social
12 INJECTION: A GENDER PERSPECTIVE ON DOMESTIC SLAVERY 217
standing of children born out of wedlock, the begetting of such children could
still be a proof of masculinity. Children born of an enslaved woman and her
enslaver could often be expected to be loyal and trusted supporters even if they
could never take the full role of a child born in wedlock of a free woman. In
many premodern societies such children were either automatically considered
free or could be freed. Such a child, if recognized, might have full inheritance
rights equivalent to those accruing to a child of formal marriage. Enslaved
women were thus often valued for their reproductive labor.
For medieval Christian Europe the status of a child born of an enslaved
woman impregnated by her enslaver was legally problematic. First, relation-
ships between men and women other than within mar r iage were condemned
by the church, regardless of whether there was a status differential between the
parties. In practice, the definition of what relationships constituted marriage
took a long time to work out. Historians have placed the triumph of church
control over marriage, and the consequent definition of those born outside
of it as illegitimate, anywhere from the ninth to the thirteenth century. Even
when marriage came to be considered as within ecclesiastical jurisdiction, what
constituted it was not necessarily a blessing by a priest (not required until the
sixteenth century). Marriages were meant to be conducted publicly but were
not invalid if they were not. By the twelfth century it was generally agreed that
marriage was constituted by the consent of the parties, and that of their fami-
lies was not required. Secular law might judge the validity of a marriage by the
payment of a dowry or other marital assigns, which did require the consent of
the family from which the money came. More importantly, however, whether a
particular union was considered a marriage by the community, and sometimes
by various legal systems, could depend on the relative status of the parties.
Pope Leo, I had written to Rusticus of Narbonne in 458–59 that “Marriage
is a legitimate agreement between freeborn and equal persons,” and suggested
that someone who wished to marry his daughter to a man who had an enslaved
concubine need not wor r y that the man could be considered already married.1
Leo’s opinion was based on Roman law, which did not consider any union
involving an enslaved person as a marriage. The enslaved could have only
contubernium and not coniugium which only existed between Roman citi-
zens. By the twelfth century the chur ch would come to reject Roman law on
this point.2 The jurist Gratian held that a marriage between an enslaved and a
free person or between two enslaved people was valid as long as there was no
fraud and the status of the enslaved party was known to the free party.3 But the
older attitude that a marriage could not be valid one unless the parties were
of approximately equal status still persisted, particularly in areas whose legal
systems wer e based on Roman law. Bartolo of Sassofer rato (1314–1357), one
of the great legal authorities of the fourteenth century, enunciated a general
principle that any son of an unmarried man who is acknowledged by his father
should be considered legitimate unless the father calls the child “natural.”
Children born to servants, however, were exceptions to this rule: “if some
honorable and noble citizen should have children by some servant who served
218 R. M. KARRAS
him or another, then by those words he cannot say he is legitimate because
marriage cannot happen with that woman, at least honorably.”4
The Roman law principle for the children of the enslaved had been partus
sequitur ventrem, or “the child follows the womb”: because there was no
marriage of the enslaved, children born to them were by definition illegitimate,
and illegitimate children did not take the status of their father. As an enslaved
woman was legally property, her children were considered the property of her
enslaver as well. This was the case whether or not her enslaver was the father;
fatherhood brought with it no rights here, as indeed motherhood did not.
The principle that the children of an enslaved woman and a free man were
themselves enslaved did not hold throughout medieval Europe, however. In
Scandinavia, for example, the laws held that children of the enslaved followed
the status of the higher-status parent. And even in places where Roman law
was generally followed, such as medieval Italy, by the later Middle Ages both
legal commentators and judges were discussing whether a man could make
legitimate the child he had with an enslaved woman, without referring to any
preceding manumission. In other words, they were assuming that the acknowl-
edged child of a free man was already free. Sally McKee has argued