Slavery and Empire in Central Asia

The Central Asian slave trade swept hundreds of thousands of

Iranians, Russians, and others into slavery during the eighteenth
and nineteenth centuries. Drawing on eyewitness accounts, autobio-
graphies, and newly uncovered interviews with slaves, this book
offers an unprecedented window into slaves’ lives and a penetrating
examination of human trafficking. Slavery strained Central Asia’s
relations with Russia, England, and Iran, and would serve as a major
justification for the Russian conquest of this region in the 1860s–70s.
Challenging the consensus that the Russian Empire abolished slavery
with these conquests, Eden uses these documents to reveal that it was
the slaves themselves who brought about their own emancipation by
fomenting the largest slave uprising in the region’s history.

Jeff Eden is a Mellon Postdoctoral Research Fellow at Cornell

University. He is the author of The Life of Muhammad Sharif
(2015) and co-editor of Beyond Modernism: Rethinking Islam in
Russia, Central Asia and Western China (19th-20th Centuries)
(2016).
Cambridge Studies in Islamic Civilization

Editorial Board
Chase F. Robinson, The Graduate Center, The City University of New York
(general editor)

Michael Cook, Princeton University

Maribel Fierro, Spanish National Research Council
Alan Mikhail, Yale University
David O. Morgan, Professor Emeritus, University of Wisconsin-Madison
Intisar Rabb, Harvard University
Muhammad Qasim Zaman, Princeton University

Other titles in the series are listed at the back of the book.
Slavery and Empire in Central Asia

JEFF EDEN
Cornell University
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
79 Anson Road, #06-04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781108470513
DOI: 10.1017/9781108637329
© Jeff Eden 2018
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2018
Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A.
A catalogue record for this publication is available from the British Library.
ISBN 978-1-108-47051-3 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Contents

Acknowledgments page vi
A Brief Note on Transliteration viii
Map ix

Introduction 1
1 The Setting: Russia, Iran, and the Slaves of the Khanates 13
2 Beyond the Bazaars: Geographies of the Slave Trade
in Central Asia 55
3 From Despair to Liberation: Mı̄rzā Mahmūd Taqı̄
˙
Āshtiyānı̄’s Ten Years of Slavery 88
4 The Slaves’ World: Jobs, Roles, and Families 115
5 From Slaves to Serfs: Manumission along the Kazakh
Frontier 144
6 The Khan as Russian Agent: Native Informants
and Abolition 163
7 The Conquest of Khiva and the Myth of Russian
Abolitionism in Central Asia 183

Bibliography 213
Index 222

v
 Acknowledgments

This book began as my doctoral dissertation at Harvard, under the direc-

tion of my adviser Roy Mottahedeh and committee members David
J. Roxburgh and Ali Asani. Suffice to say the book and I have both
benefited very much from their valuable advice and recommendations.
Several other colleagues and friends read earlier drafts of this book in its
entirety and offered invaluable insights: Allen J. Frank, James Millward,
Alexander Morrison, William A. Wood, and the anonymous reviewers of
the Cambridge Studies in Islamic Civilization series. I owe each of them my
profound gratitude. Many thanks are also due to the series editors and
dedicated production team at Cambridge; to Kevin Hughes for his excel-
lent work copyediting the manuscript; and to Maria Marsh for all her help
along the way.
My research trip to the archives of Kazakhstan in 2013 was sup-
ported by a grant from the Eurasia Program of the Social Science
Research Council with funds provided by the State Department under
the Program for Research and Training on Eastern Europe and the
Independent States of the Former Soviet Union (Title VIII). (Sadly, the
Eurasia Program has since been defunded and shut down.) I would
like to thank the many patient and generous archivists who helped me
at the Central State Archive of the Republic of Kazakhstan, the
National Library of Kazakhstan, and the Kazakh Academy of
Sciences.
For their time, encouragement, and enlightening conversations about
the book project and/or the subject of slavery, I wish to thank Naofumi
Abe, Gulnora Aminova, Christopher Atwood, Patrick Baron, David
Brophy, Devin DeWeese, Michael Feener, Baber Johansen, Scott C. Levi,

vi
 Acknowledgments vii

Christine Nölle-Karimi, Jürgen Paul, Paolo Sartori, Ron Sela, Akifumi

Shioya, Elena Smolarz, Wheeler M. Thackston, and Rian Thum.
To all my students at Cornell, Harvard College, Harvard Extension
School, the University of Maryland, Baltimore County, and the
University of Maryland, College Park: Being in class with you has been
my greatest inspiration, as well as a joy beyond words.
This book is dedicated to my parents, Lawrence and Susan Eden.
Finally, I thank Ashley, the light of my life.
 A Brief Note on Transliteration

In the interests of keeping diacritics to a minimum, I have omitted them

for most place-names (e.g. Bukhara, Khwarazm) as well as for common-
place titles that are widely recognized in English (e.g. Shah, Khan). I have
retained diacritics for personal names as well as for transliterated samples
of Persian and Turkic sources. In transliterating Persian, I have followed
the IJMES standard. There is no commonly accepted standard for trans-
literating nineteenth-century Central Asian Turkic, but I have tried to be
as consistent as possible on that front. In transliterating Russian, I have
used the “Modified Library of Congress” system.

viii
 Kazan

N E M P I
I A R E

ol
S

Tob
S
U Samara
R

Irt
ïs h
im
Ish
Saratov

Orenburg
Ural’sk
lg
a E
D

l)
Vo
Semipalatinsk
Q

i q (Ura
Tsaritsin
R A
O Z E
H D

y
Ya
A R
R O
ba H ag
uz
Em O E
L. Zaysan

Ay
Astrakhan I
N D L Q
U I D Ba l k h a s h L. Ala-Kul
M EJunghar

ke
a

Ka r
La
M D
S e

Gates
an Ar a l R L. Ebi Nor

at
g

al
Pe ishl O

S
nin aq Aq-Masjid
sul
a
Ust-Yurt Sea Chu H Qulja Turfan
Plateau Suzak Il i
n

Ta R

Sy
BUKHA
Sïghnaq lis
O

r- D
NI
p i a

ary
Qongrat a Turkestan L. Issïq- Qarashahr
K H I VA Awliya-Ata
Chimkent SE Pishkek
Kul
Z E
L. Sarïqamïsh K H A N AT E QOQ TE I
Qucha
AND KHANA R
C a s

Baku Urgench Khanqah Aqsu

I
y

Hazarasp
H rim
Uzbo

RA
Khiva Tashkent ik Narïn
Namangan irch Ta P L. Lob Nor
T UR Qoqand Andijan Ch
Cheleken I. U Ghijduvan Kermine
A-
Khojend Margilan G M
R E

TÜ
Ogurchin I. Qizïl- Bukhara Jizak Ura-Tübe k hab

BE
K S ur

R
Arvat
Qara-qala
Qaraköl
M Charjuy Qarshi
Samarqand
Z erafshan KH
TEGIN
Kashghar E
rek Shahrisabz QARA S

Ï
At A Yarkand
Ashqabad E A Hisar h D A RV Murghab E
N Kerki u -D aryaA
N
s N

AZ

Q
T
m

Merv
I

kh
Gorg an Kulab Khotan
E a
H
V
Astarabad S
Mu r g h

C
Ted

Tehran Serakhs Aqcha

j
Nishapur Andkhoy Pan
zhe n

Balkh Qunduz
Mashhad
Central Asia:
ab

I R A N Maymana
In First Half of the 19th Century
A F G H A N I S TA N

du
Herat Hari-r

s
u Scale 0 100 200 300 400 km
Kabul SIKH STATES
d

Yuri Bregel, Historical Central Asia Maps (Bloomington, IN: SRIFIAS Papers on Inner Asia, 2000)
 Introduction
A Forgotten Slave Trade

The indolent, enervated Orientals may still regard with bitter resentment and
rancor the efforts of Europe in the cause of humanity; but the sale and purchase
of human beings is everywhere practiced with a certain reserve arising from
a sense of shame, or, to speak more correctly, of fear of European eyes. This
trade is now to be found unfettered and unembarrassed only in Central Asia.
Arminius Vambery, Sketches of Central Asia, 18681
By the time of the Russian conquest of Central Asia in the 1860s and
1870s, the region’s social landscape had been shaped by a millennium of
slavery. Slaves served as farmworkers, herdsmen, craftsmen, soldiers, con-
cubines, and even, in rare cases, as high-ranking officials in the region
between the Caspian Sea and westernmost China. The institution of
slavery in the region had never been seriously challenged by any internal
or external forces down to the nineteenth century. It thrived especially in
the khanates of Khwarazm and Bukhara. As the nineteenth century wore
on, however, negotiations over the release of slaves began to factor heavily
in these khanates’ relationship to Iran, Russia, and Great Britain. By the
end of the century, tens of thousands of slaves would be free.
This book examines the period from 1750 to 1873, which saw both the
flourishing of Central Asia’s slave trade and its collapse, and it focuses in
particular on the region extending from Khurasan in the south to the
Kazakh–Russian frontier in the north, and encompassing Khwarazm,
Bukhara, and their environs.2 It is not a political history of Central Asia,
1
Arminius Vambery, Sketches of Central Asia (London: W.H. Allen, 1868), 205.
2
Slavery was also prominent in other regions of Central Asia, such as Afghanistan and East
Turkistan, but I have chosen not to cover those regions in the present work, in part because,
as we shall see, the region extending from Khurasan north across the Caspian coast and

1
2 Introduction: A Forgotten Slave Trade

nor another diplomatic history of the so-called Great Game. Rather, the
purpose of this book is to advance two arguments about slavery and
abolition in the region. First, drawing on slaves’ testimonials as well as eye-
witness accounts and official sources, I challenge the historiographical
consensus that Russian military force ended the slave trade and show
how Russian efforts toward fostering abolition often had ulterior motives
as well as wildly mixed results. Second, I argue that slaves influenced the
nature of their captivity through their own initiatives and ingenuity, and
I show how slaves in the khanate of Khwarazm launched an uprising, little-
known even among historians of Central Asia, which served as the catalyst
for abolition in the region as a whole.
While evidence of slavery and the slave trade in Central Asia is plentiful,
scholarship on it is nearly nonexistent.3 Despite its extraordinary scale,

along the Russian–Kazakh frontier can be considered a distinct and bounded (albeit
roughly) ecosystem in which slaves circulated. The slave trade in East Turkistan, for
example, which revolved around Tarim Basin trade networks and also involved Chinese
slaves and Chinese traders, is deserving of separate study, and Laura Newby (see below) has
broken ground in that effort. A recent dissertation by Benjamin Levey has offered ground-
breaking insights into the fate of slaves along China’s Kazakh frontier: “Jungar Refugees
and the Making of Empire on Qing China’s Kazakh Frontier, 1759–1773”
(Ph.D. dissertation, Harvard University, 2013).
3
The only monograph on the topic remains a slim but important Soviet-era volume in Uzbek,
by Turgun Faiziev: Buxoro feodal jamiyatida qullardan foidalanishga doir hujjatlar (XIX asr)
(Tashkent: Fan Nashriyoti, 1990). Some related works of note include Laura Newby,
“Bondage on Qing China’s North-Western Frontier,” Modern Asian Studies 47:3 (2013),
968–994; Scott Levi, “Hindus Beyond the Hindu Kush: Indians in the Central Asian Slave
Trade,” Journal of the Royal Asiatic Society 12:3 (2002), 277–288; Benjamin Hopkins,
“Race, Sex, and Slavery: ‘Forced Labor’ in Central Asia and Afghanistan in the Early 19th
Century,” Modern Asian Studies 42:2 (2008), 629–671; Alessandro Stanziani, Bondage:
Labor and Rights in Eurasia from the Sixteenth to the Early Twentieth Centuries (New York,
NY: Berghahn, 2014), 63–100; Elena Smolarz, “Speaking about Freedom and Dependency:
Representations and Experiences of Russian Enslaved Captives in Central Asia in the First
Half of the 19th Century,” Journal of Global Slavery 2 (2017), 44–71; and Yuan Gao,
“Captivity and Empire: Russian Captivity Narratives in Fact and Fiction” (M.A. thesis,
Nazarbayev University, 2016). Valuable information on slavery along the Chinese–Central
Asian frontier is provided in James Millward’s Beyond the Pass: Economy, Ethnicity, and
Empire in Qing Central Asia, 1759–1864 (Stanford, CA: Stanford University Press); see, for
example, 305–306n28, 238, 123. On slavery in premodern Central Asia, see Peter
B. Golden, “The Terminology of Slavery and Servitude in Medieval Turkic,” in
Devin DeWeese, ed., Studies on Central Asia in Honor of Yuri Bregel (Bloomington, IN:
Research Institute for Inner Asian Studies, 2001), 27–56. The paucity of published work on
the subject is not entirely surprising, as the scarcity of research on slavery in Islamic Central
Asia mirrors the broader scarcity of research on slavery anywhere the Muslim world.
In Joseph C. Miller’s comprehensive bibliography of scholarly works on slavery, published
in 1999, we find a table showing the distribution of works on the subject according to their
geographical focus: Among the thousands of works on slavery published between 1900 and
1991, a mere 3.3 percent focused on the Muslim world. Between 1992 and 1996, the
 Introduction: A Forgotten Slave Trade 3

Central Asia’s slave trade has largely been forgotten. This is no great
surprise: Central Asia remains among the least-studied regions of the
world. As Alexander Morrison has observed, even the Russian conquest
has been the subject of surprisingly little research.4 There is no consensus
on what motivated it. This does not mean, however, that Russian imperial
officials and propagandists failed to articulate a justification at the time.
Ending the slave trade was at the heart of their justification.
European abolitionists, meanwhile, had high hopes for the conquest.
Herbert Wood, writing soon after the Russian military took the town of
Khiva in 1873, praised the Russian “civilizing mission” in the most
generous terms: “Though Russia’s position in the Central Asian
Khanates may not yet be assured,” he writes, “it is certain that without
her leave no dog may bark in the bazaars of Khiva, Bokhara, and Kokand.
And if a strong government which preserves social order and has put
down brigandage, slavery, and man-stealing is worthy of sympathy, it is
impossible not to feel that in undertaking the thankless and costly task of
introducing civilisation into Turkestan, Russia is fully entitled to the
good wishes and gratitude of every Christian nation.”5 It is natural,
perhaps, that European observers would have expected abolition to
constitute a major feature of the Russian “civilizing mission” in the
East, given that Western European powers tended to cast many of their
own conquests and interventions in this era as emancipating enterprises.
Indeed, the nineteenth century was an age of global abolitionist interces-
sions, even if not all interventions were successful, and even if many were
mere foils for more pressing (and more selfish) motivations. The British
Empire led the way, officially abolishing the slave trade throughout its
imperial holdings by 1807, and other colonial powers soon followed:
Portugal signed a treaty stifling the importation of slaves into its colonial
possessions in 1810; Sweden banned the trade in 1813; the Netherlands
did the same the next year; and Spain and France followed soon after, the
former promising to abolish the trade by 1820 and the latter by 1819.
In the decades to come, the freeing of Christian captives was presented as
a major incentive for the French conquest of Algiers (though, as
W. G. Clarence-Smith observes, “they failed to extend their liberality”

proportion dipped slightly, to an even 3 percent (Joseph C. Miller, Slavery and Slaving in
World History: A Bibliography, 2 vols. [Armonk, NY: M.E. Sharpe, 1999], xi–xii).
4
Morrison, “Introduction: Killing the Cotton Canard and getting rid of the Great Game:
rewriting the Russian conquest of Central Asia, 1814–1895,” Central Asian Survey 33:2
(2014), 131–142.
5
Wood, The Shores of Lake Aral (London: Smith, Elder, & Co., 1876), 182–183.
4 Introduction: A Forgotten Slave Trade

to non-Christian slaves),6 while the British Empire attempted to foster

abolition throughout the Ottoman world, sometimes with the assistance
of Ottoman rulers and elites and sometimes in spite of their opposition.7
The British likewise led the way in efforts to undermine the Atlantic slave
trade, though the last slave ship to arrive in the American South did so as
late as 1858.8 Indeed, colonial emancipation projects often achieved
their ends, if at all, only very gradually, and sometimes over the course of
decades.9 Nevertheless, in this climate of global abolition, the Russian
efforts in Central Asia, which came not long after the abolition of serf-
dom within the Empire itself, were naturally regarded by many Western
observers as yet another mission to end the misery of bondage, whatever
its other motivations may have been.10
While historians have long cast a critical gaze on Western empires’
moral pretenses for conquest, the Russian Empire has generally been
sheltered from similar scrutiny. Historians within the Soviet Union shared
a tendency to interpret the conquest of Central Asia purely as a means of
extending Russian industry into the region, while historians in the West
have generally preferred a “Great Game” narrative that explains the con-
quest as a simple race for regional dominance between England and Russia.
Proponents of both approaches tend to avoid the question of the “civilizing
mission” entirely. Meanwhile, some of the (few) recent works addressing
the issue of slavery directly have tended to concur with the Russian
“abolitionist” narrative: Liubov Kurtynova-D’Herlugnan has recently
argued that Russian abolitionism in Eurasia culminated in the eradication
of slavery in the Caucasus, suggesting that the “civilizing mission” was
both sincere and effective, and M. D. Farah has argued that the Central

6
Clarence-Smith, Islam and the Abolition of Slavery, 100.
7
Ehud R. Toledano, Slavery and Abolition in the Ottoman Middle East (Seattle, WA:
University of Washington Press, 1998); Toledano, The Ottoman Slave Trade and Its
Suppression: 1840–1890 (Princeton, NJ: Princeton University Press, 1982);
Ismael M. Montana, The Abolition of Slavery in Ottoman Tunisia (Gainesville, FL:
University Press of Florida, 2013).
8
W. E. B. DuBois, The Suppression of the African Slave Trade to the United States of America,
1638–1870 (New York, NY: Longmans, Green, and Co., 1904), 181–185.
9
Clarence-Smith, Islam and the Abolition of Slavery, 99–218.
10
Several recent works have offered important insights on the ulterior motives—and,
oftentimes, the evident insincerity—of abolitionist efforts among Western powers: see,
for example, Matthew S. Hopper, Slaves of One Master: Globalization and Slavery in
Arabia in the Age of Empire (New Haven, CT: Yale University Press, 2015); Ehud
R. Toledano, “Abolition and Anti-Slavery in the Ottoman Empire: A Case to Answer?”
(forthcoming); and Behnaz A. Mirzai, A History of Slavery and Emancipation in Iran
(Austin, TX: University of Texas Press, 2017).
 Introduction: A Forgotten Slave Trade 5

Asian slave trade essentially ended with the Russian military conquest of
Khiva.11 In this book, I come to the opposite conclusions concerning the
Russian “civilizing mission” in Central Asia.
Indeed, Russian abolitionism in Central Asia is a myth. The overall
mechanism of abolition in Central Asia, and the means by which the
region’s new Russian rulers would patrol and monitor the slave trade
were evidently never articulated even among the tsar’s top generals and
officials. If indeed there had been a grand, overarching plan for general
emancipation in Central Asia (and we have no evidence such a thing
existed), it was certainly implemented in a scattershot – perhaps even
improvised – manner. Russian demands for abolition were applied
unequally across the conquered territories, and enforcement seems to
have been non-existent – a fact that, as we shall see, agitated some
Russian officers who had anticipated more active antislavery efforts.
As I show in Chapter 7, the most important force behind the liberation
of Central Asia’s slaves was the slaves themselves – particularly the slaves
of Khwarazm,12 who seized the occasion of the Russian invasion to launch
a courageous uprising against their masters. Witnessing this uprising, the
Russian general in charge of the invasion hung two rebel slaves in the town
square as a warning to other rebels. Evidently, this general preferred to
seize an orderly town of slaveholders rather than a chaotic town of self-
emancipated slaves. Russian threats notwithstanding, the slave uprising
continued until the general had no choice but to support it. It was abolition
sans abolitionism, in other words.
Notwithstanding these dramatic events, which culminated in an
“official” abolition decree, the slave trade continued. Even if the
Russians had attempted to patrol it, which they did not, the trade could
not have been stopped without a relentless and wide-ranging enforcement
strategy. This is because, as Chapter 2 shows, the slave trade was not
11
Liubov Kurtynova-D’Herlugnan, The Tsar’s Abolitionists: Languages of Rationalization
and Self-Description in the Russian Empire (Leiden: Brill, 2010); M. D. Farah, “Autocratic
Abolitionists: Tsarist Russian Anti-Slavery Campaigns,” in William Mulligan, ed., A Global
History of Anti-Slavery Politics in the Nineteenth Century (New York, NY: Palgrave, 2013),
97–117. While I do not always agree with Kurtynova-D’Herlugnan’s conclusions, her
work deserves recognition as a groundbreaking and important study. It is interesting to
contrast this book with Irma Kreiten’s valuable article: “A Colonial Experiment in
Cleansing: The Russian Conquest of Western Caucasus, 1856–65,” Journal of Genocide
Research 11:2–3 (2009), 213–241.
12
In most Russian and Western European sources, the khanate of Khwarazm is referred to as
Khiva, often resulting in ambiguity as to whether an author is referring only to the capital
city or to the khanate as a whole. In Central Asian sources, however, the khanate was
always referred to as Khwarazm, and I will maintain that usage here.
6 Introduction: A Forgotten Slave Trade

confined to the urban centers and their bazaars; Central Asia’s slave trade
was decentralized and widely dispersed, with many slaves traded in villages
and nomadic encampments. Indeed, the nomads of the hinterlands played
a key role in the slave trade, not only as raiders but also as merchants and
slave-owners. In contradiction to many previous studies, which allege that
nomads did not need or use slaves, I will argue here that slave ownership in
Central Asian nomadic communities was commonplace. However, while
nomadic raiding was a key feature of the slave trade, helping to ensure its
survival after the fall of major towns to Russian armies, it was hardly the
only means by which individuals were seized for enslavement. As I argue in
Chapter 1, selling captives into slavery was a normative part of warfare
across the region; this practice was not limited to nomads, and the
Russians themselves sometimes took part in the slave-dealing economy
in Central Asia.
When it comes to the nomads, slave-raiding and slave-trading served
more than just their immediate commercial interests. Turkmen and Kazakh
nomads developed complex, symbiotic relationships with merchants in the
towns of Khwarazm and Bukhara that revolved around the commerce in
slaves. Turkmens traded slaves for grain, for example, which these desert-
dwellers had no way of growing for themselves. The slaves were often put
to work on plantations, growing grain – a cyclical commercial system.
Slave-raiding was also a form of resistance. In the nineteenth century,
Turkmen and Kazakh nomads occupied territories encircled by sedentary
powers, with the Russian Empire encroaching inexorably from the north
and Qajar Iran threatening always to encroach (and occasionally invading
Turkmen territories) from the south. Destabilizing sedentary borderland
territories through raiding helped to keep these powers at bay. Central
Asian sedentary states used this dynamic to their own advantage:
Khwarazmian khans – threatened simultaneously by Iran and Russia –
alternately allied with Turkmens hostile to Iran and with Kazakhs hostile
to Russia, benefiting from whatever limited, proxy defensive capacity these
borderland nomads provided. As Chapter 6 shows, Russia later adopted
this “nomadic proxy” strategy against Iran, continually prompting “client”
Turkmens to raid Iranian territory, or at least offering protection to those
who did.
In the decades before the Russian conquest, the issue of the slave trade
was at the heart of Russian and Iranian diplomacy with the khanates of
Khwarazm and Bukhara. As we shall see in Chapter 1, diplomats from
Russia, Iran, and Britain even converged simultaneously in Khiva at one
point and issued a joint request – hopelessly, of course – for the
 Introduction: A Forgotten Slave Trade 7

emancipation of all the slaves in Khwarazm. Both Russia and Iran had
expansionist ambitions in Central Asia; Both protested the enslavement of
their citizens in the khanates; and both raised the issue of slavery in order
to justify threats.
In other respects, however, the positions of these two empires differed
dramatically. First, the total number of Russian slaves held in Central Asia
was miniscule compared to the total number of Iranian slaves. Second,
Russia’s military resources were vastly greater than Iran’s; by the mid-
nineteenth century, the threat of a Russian conquest was palpable, while
the threat of a major Iranian conquest anywhere in Central Asia (at least
after the early 1830s) probably seemed remote. Finally, while Iranian
travel literature from the nineteenth century often casts Central Asia as
a historically Iranian-dominated region, and sometimes even as a rightful
Iranian possession, one scarcely perceives in these “imperial” literatures
the pretenses of a “civilizing mission.”
British activities in Central Asia, meanwhile, were peripheral in every
sense, and the British Empire will factor very little in this book. This
omission may surprise some readers, given the extensive literature concern-
ing British “players” in the Great Game. For all their great adventures,
miseries, and ambitious pretenses – Captain Arthur Conolly, for example,
planned in 1838 to concoct an Anti-Slavery Confederation in the region13 –
the impact of British officers and adventurers on the Central Asian slave
trade was negligible. The greatest British accomplishment on this front was
the alleged role of the officer Richmond Shakespear in convincing the
Khwarazmian khan to release over 400 Russian slaves following General
Persovskii’s failed campaign against the khanate in 1839.14 Certain other
British efforts were characterized by a grandiosity that smacks of madness,
such as the petition penned by the independent traveler Joseph Wolff while
imprisoned in Bukhara in 1844; he addressed his dispatch “To all the
monarchs of Europe”:15
Sires!
. . . I do not supplicate for my own safety; but, Monarchs, two hundred thousand
Persian slaves, many of them people of high talent, sigh in the kingdom of Bokhara.
Endeavour to effect their liberation, and I shall rejoice in the grave that my blood

13
See, for example, Evgeny Sergeev, The Great Game, 1856–1907: Russo-British Relations in
Central and East Asia (Baltimore, MD: Johns Hopkins University Press, 2013), 4.
14
Russian accounts tend overwhelmingly to deny Shakespear credit for the release of these
slaves.
15
Wolff, Narrative of a Mission to Bokhara, in the Years 1843–1845, Vol. 2 (London: John
W. Parker, 1845), 104.
8 Introduction: A Forgotten Slave Trade

has been thus the cause of the ransom of so many human beings. I am too much
agitated, and too closely watched, to be able to say more.

Suffice to say, Wolff received no reply from the monarchs of Europe.

British and American travel writing on slavery is nevertheless pre-
cious, since visitors like Conolly, Wolff, Alexander Burnes, and Eugene
Schuyler provide us with some of the most detailed ethnographic
reportage in existence concerning slaves’ experiences. The abolitionist
bias of these authors is undisguised, and should always be borne in
mind, but this fact by no means proves that the eyewitness evidence
they provide is falsified. When it comes to the role of the British
Empire in nineteenth-century Central Asia, however, this book leaves
its officers and adventurers at the periphery – which is precisely where
they ought to be, considering their minimal impact on Central Asian
slavery.16
This book is not concerned only with empires, however. It is concerned
with slaves’ lives too, and I have tried here to illuminate something of
slaves’ experiences. For this effort, I have found two types of source
especially useful. First, I have relied on interviews conducted by Russian
border officials with slaves who escaped from the Kazakh steppe. These
interviews are preserved in the archives of the Central State Archive of the
Republic of Kazakhstan. Chapters 2, 5, and 6 make extensive use of these
documents and shed light on their contents and implications. These inter-
views, while precious for the rare glimpse they provide into the nature of
the slave trade, are typically quite concise, rarely running to more than
a few pages, and sometimes comprising no more than a few lines. A far
more detailed source – indeed, the greatest single source for the study of
slavery in the region – is the Persian-language memoir of Mı̄rzā Mahmūd
˙
Taqı̄ Āshtiyānı̄,17 whose extraordinary story of survival and hardship is
recounted in Chapter 3.
Another key source for this book is a rare manuscript history of Central
Asian slavery composed in Chaghatay Turkic in the early Soviet period.
Called the Āzādnāma, this work offers details on the fate of Khwarazm’s

16
See Alexander Morrison’s insightful recent comments on this subject: “Killing the Cotton
Canard and getting rid of the Great Game: rewriting the Russian conquest of Central Asia,
1814–1895,” Central Asian Survey 33:2 (2014), 131–142 (see especially 132–133). For
those interested in the British role in Central Asia and in Anglo-Russian relations, there is
no shortage of literature on the subject; Morrison’s up-to-date bibliography in the above-
mentioned article directs readers to the major works.
17
Mı̄rzā Mahmūd Taqı̄ Āshtiyānı̄, ʿIbratnāma: Khatirati az dawran-i pas az jangha-yi herat va
˙
merv, ed. Husayn ʿImadi Ashtiyani (Tehran: Nashr-i markaz, 1382/2003).
 Introduction: A Forgotten Slave Trade 9

freed slaves that are found nowhere else.18 Chapter 7 makes use of this
manuscript in reconstructing the conquest of Khiva and its aftermath.
T. Faiziev’s collection of nineteenth-century legal documents concern-
ing the slaves of Bukhara is another crucial source for the study of Central
Asian slavery. These documents, which include numerous manumission
deeds, provide critical documentary proof that the slave trade continued
well after the Russian conquests – and even into the late 1880s. They also
demonstrate that slavery was formalized and regulated, at least to some
extent, by the region’s Hanafi Muslim legal system.
Russian imperial officers, diplomats, and travelers in Central Asia, like
their British counterparts, were keenly interested in the issue of slavery,
and their eyewitness reportage, while sometimes lurid, is likewise vital for
any study of the subject. In citing Russian witnesses, I have tended to
privilege reports of a “military-statistical” or diary-like nature over the
more overtly “literary” products of the captivity-narrative genre, whose
target readership was resolutely popular and whose details sometimes
strike me as fantastical and suspect.
While I will focus here on slavery in the Muslim societies of Central
Asia, it is important to observe that slavery was not introduced to the
region by Muslims, and neither were captive-taking and slave-owning
exclusive to Muslims. Furthermore, as I will show, Muslim Central Asia
was home to a great diversity of slave systems, some of which invite
comparisons beyond the Muslim world.19 In Khwarazm, for example,
where we find slaves laboring on large agricultural estates, the prevailing
system of slavery shares more common features with plantation slavery in
the American South than it does with urban slavery in Istanbul. Slavery
among the nomadic Kazakhs, meanwhile, shares more in common with
slavery among nomadic non-Muslim groups such as the Mongols than
with either Khwarazm’s plantation slavery or Istanbul’s urban slavery.
Aside from its diversity of forms, the extensiveness of slave-owning also

18
MS IVAN Uz No. 12581. I am grateful to Paolo Sartori for providing me with a copy of this
manuscript.
19
The terminology of slavery in Central Asia is vast – ghulām, qul, chūrı̄, and mamlūk are just
a few of the many terms for slaves that we shall encounter over the course of this book –
and the word “slave” is hardly adequate in reflecting that diversity. What unites the roles
defined by all of these terms is best captured in Seymour Drescher’s definition of slavery:
“The most crucial and frequently utilized aspect of the condition [of slavery] is
a communally recognized right by some individuals to possess, buy, sell, discipline, trans-
port, liberate, or otherwise dispose of the bodies and behavior of other individuals”
(Drescher, Abolition: A History of Slavery and Anti-Slavery [Cambridge: Cambridge
University Press, 2009], 4–5).
10 Introduction: A Forgotten Slave Trade

varied by region: Even as the trade thrived in Khwarazm and Bukhara, it

remained strikingly small in scale in the neighboring Muslim khanate of
Kokand.20 Given this, it is possible to discuss Central Asian slavery as
a regional phenomenon, extending beyond the borders of Muslim-
majority territories.21 A longue durée view of slavery in the region might
inspire the view that slavery among Muslims was merely a continuation of
a region-wide practice that can be traced into the distant pre-Islamic past,
as well as into neighboring, contemporary, non-Muslim societies.22
However, we must keep in mind two facts that demonstrate that slavery
among Central Asian Muslims was not regarded as a mere “pre-Islamic
survival.” First, Hanafi Muslim law both governed the slave system and
justified it. It is clear that the Muslim jurists of nineteenth-century Bukhara
and Khwarazm did not see the institution of slavery as “un-Islamic”; if
they had, they would not have countenanced it by producing bills of sale
for the slave trade or adjudicating disputes concerning slaves and their
masters. Second, religious differences typically distinguished slaves from
their Sunni Muslim owners, and the usual justification for the enslavement
of non-Sunnis in Central Asia was explicitly religious. Iranian Shiʿites and
Russian Orthodox Christians, among others, were licit to enslave because
they were not Sunni Muslims.
Finally, a note on terminology: When I refer to Central Asian “slaves,”
I have in mind those people classed as qul and/or ghulām – two words that
appear commonly in Turkic and Persian sources from the region. These
two words are almost always rendered in English as “slave,” and for good
reason: both imply a condition of unfreedom in which an individual can

20
The main reason for this is likely Kokand’s relative distance from Khurasan, the region that
supplied most of the slaves kept in Khwarazm and Bukhara.
21
This approach contrasts with that taken by a number of recent works on “Muslim” slavery,
most notably William Gervase Clarence-Smith’s ambitious comparative synthesis, Islam
and the Abolition of Slavery (Oxford: Oxford University Press, 2006).
22
The last two decades have seen the steady development of research on Eurasian slavery
beyond Central Asia, and I will mention just a few notable works here. On slavery in early
modern China, see Pamela K. Crossley, “Slavery in Early-Modern China,” in David Eltis
and Stanley Engerman, eds., The Cambridge World History of Slavery, Vol. 3 (Cambridge:
Cambridge University Press, 2011), 186–216; on South Asian slavery, see
Indrani Chatterjee and Richard M. Eaton, eds., Slavery & South Asian History
(Bloomington, IN: Indiana University Press, 2006). On slavery in Iran (still a relatively
little-studied topic), see Thomas Ricks, “Slaves and Slave-Trading in Shi’i Iran, AD
1500–1900,” Journal of Asian and African Studies 36:4 (2001), 407–418; and Mirzai,
A History of Slavery and Emancipation in Iran. Valuable studies of slavery and the slave
trade in regions to the west and southwest of Central Asia can be found in
Christoph Witzenrath, ed., Eurasian Slavery, Ransom and Abolition in World History,
1200–1860 (Aldershot: Ashgate, 2015).
 Introduction: A Forgotten Slave Trade 11

typically be bought and sold licitly, and in which an individual requires

a formal deed of manumission to become free.23 By the nineteenth century,
Hanafi legal norms had been used to manage issues of slavery for nearly
a millennium; as one might expect, there is little evidence of confusion
over something as basic as which individuals could legally be bought and
sold, or which individuals required deeds of manumission to gain the same
formal legal rights as those characterized as “free” in the Hanafi legal
nomenclature. There were, however, a few exceptional cases in Central
Asia for which I suspect that formal legal procedure was hazier: Military
servitude among Kazakhs and Mongols, for example, has sometimes been
described by outsiders using the language of slavery, but I have not seen any
record of such military “slaves” being processed or defined as unfree
people by Muslim jurists. Another ambiguous case is discussed at length
in Chapter 5—namely, slaves among the Kazakhs who are alleged either to
have been “adopted” as children or who have been given their own
independent living-space and means as adults (typically comprising both
a tent and livestock). Nevertheless, for the vast majority of the victims
discussed in this book – who are generally Iranian Shiʿites taken captive in
Khurasan and forced to labor on plantations or other private estates –
I consider the word “slave” an appropriate description, both because it
approximates the formalized free/unfree distinction made by the region’s
native legal system and because, as we shall see, their experiences and roles
were in many respects remarkably similar to those of “slaves” in the
English-speaking world.
In short, this book is an attempt to consider Central Asian slavery both
from the “bottom-up” and from the “top-down.” I aim to provide
a window on slaves’ experiences while locating their activities in the
broader geopolitical framework of Central Asia in the age of imperial
expansion. I hope to show how slaves’ agency and resistance not only
impacted their experiences, but also influenced the slave system itself,
forming a pattern of autonomous activity that culminated in the Khivan
slave uprising of 1873.
Slaves were certainly the most powerless, subaltern population in
the region. And yet it is these slaves – as well as the trade that ensnared

23
Mı̄rzā Mahmūd, our Persian memoirist, prefers to describe himself as an ası̄r, a term
˙
generally better translated as “captive,” and which can suggest a kind of temporary
military imprisonment as opposed to the ten years of private ownership and forced labor
that he endured. But he too required a letter of manumission to gain his freedom, and this
fact – along with the other major details of his ordeal – has inclined me to translate ası̄r as
“slave” in his case.
12 Introduction: A Forgotten Slave Trade

them – that bound together the “core” and the “periphery,” the nomadic
and the sedentary: Russian tsars and Iranian shahs were repeatedly drawn
to their respective Central Asian “peripheries” on account of the trade,
while for the Central Asian khanates themselves, the trade linked major
towns (Bukhara, Urgench, Khiva) and remote oases via nomadic transit
and caravan routes. The story of colonization, local resistance, caravan
commerce, diplomatic rivalries, and imperial conquests in nineteenth-
century Central Asia all converge around the problem of the slave trade.
Chapter 1 introduces historical, social, and political settings of slavery
in early modern Central Asia, arguing that slavery and slave-taking were
not unique to borderland nomads, but rather that they were normative
features of Eurasian warfare over the course of centuries; that slavery
gained particular prominence in the region with the expansion of agricul-
tural plantations in the late Timurid period; and that, in the centuries to
come, it would become a key issue in diplomacy between Iran, Russia, and
Central Asia. Chapter 2 explores the geography of slavery, using evidence
from slaves’ testimonials to argue that slavery was a largely rural, agricul-
tural phenomenon in the region, and that the slave trade was intimately
connected with overland caravan routes. Because of the trade’s decentra-
lized nature, I argue, it was nearly impossible to police. Chapter 3 focuses
on the experiences of Mı̄rzā Mahmūd, who spent nearly a decade as
˙
a slave, first among the Turkmens and then in Bukhara, and has left us
the richest firsthand account of slavery in the region. Chapter 4 draws on
other autobiographical sources as well as eyewitness reports to describe
slaves’ occupations and roles, revealing that slaves could be found at all
levels of Central Asian society.
Chapter 5 considers the curious fate of the many slaves who fled their
masters for the Russian border, only to be pulled into serfdom as part of the
Tsar’s plan to settle and cultivate the borderlands. Here, I challenge the
notion of Russian “abolitionism” in the region while further exploring
slaves’ means of resistance. Chapter 6 reveals how imperial powers
employed Central Asian “native informants” in an attempt to pacify the
borderlands and liquidate captive-taking among nomads. I weigh the
mixed results of these efforts, further challenging longstanding assump-
tions about the Russian “abolitionist” program. Chapter 7 concludes the
book by showing how slaves throughout Khwarazm joined together in the
largest slave uprising in Central Asian history. I argue that this revolution-
ary, little-known episode triggered the abolition of slavery in the region as
a whole.
 1

The Setting
Russia, Iran, and the Slaves of the Khanates

The slave trade played a major part in Russian and Iranian diplomacy
with Central Asia throughout the eighteenth and nineteenth centuries.
The rulers of Khwarazm and Bukhara in particular were blamed for
encouraging captive-taking, for providing markets for slaves, and for
keeping tens of thousands of slaves in bondage. When it came to the
actual taking and selling of captives, however, the blame cast by foreign
statesmen, diplomats, and adventurers fell squarely upon nomadic
Turkmens and Kazakhs. The near-exclusive focus of foreign
observers – particularly Russian and British ones – on slave-trading
and captivity among the Turkmens and Kazakhs has obscured a larger
truth about slavery and Eurasian warfare before the twentieth century:
Captive-taking was central to it, and significant armed conflicts almost
invariably involved the phenomenon of mass captivity and enslavement.
This chapter introduces the origins, development, and major features of
the region’s slave trade, as well as Russian and Qajar Iranian efforts to
end the trade through diplomacy in the nineteenth century.
Ironically, the Qajars and Russians themselves engaged in both captive-
taking and in the ransom economy, and even while deploring the barbarity
of enslavement among the nomads, they made little effort to hide their
own parallel efforts to deprive rivals, including noncombatants, of their
freedom. Examples are legion, especially in “official” Persian chronicles,
which often brag of captive-taking as if it was a hallmark of victory in
battle.1 At times the Turkmens would complain of Qajar captive-taking to

1
The chroniclers had much to brag about on that front, and Turkmens – the most notorious
captive-takers in the region’s history – were often themselves the captives in question. In the

13
14 The Setting

Russian officials, who were simultaneously hearing similar complaints

about the Turkmens from the Qajars.2
Russian sympathies tended overwhelmingly to fall on the side of the
Iranians, however, as have the sympathies of historians in the generations
since. The idea of Turkmens as perennial persecutors and Iranians as their
perennial victims is etched so deeply in the historiography of nineteenth-
century Central Asia that it has become a kind of leitmotif. We see this, for
example, in the kitschy title of Charles Marvin’s earnest and classic work
on Merv in the nineteenth century: Merv, Queen of the World, and the

Ma’āsir-i sultānı̄ya of ʿAbd al-Razzāq Dunbulı̄, the court historian of ʿAbbās Mı̄rzā
˙ ˙
(1789–1833), multiple assaults by the Iranian Shah’s armies on Yomuts, Göklengs, and
Tekes – invariably characterized as reprisals for their plundering – result in captivity for
these Turkmens. In 1792, according to the chronicle, during a Qajar assault on the Yomuts
and others in the vicinity of Astarabad, “great numbers of them were put to death or
reduced to slavery and captivity, and on the highways were built minarets constructed
with their heads” (Harford Jones Brydges, The Dynasty of the Kajars [London: John Bohn,
1833], 22). In 1803–1804, according to the same chronicle, Yomuts and Göklengs living
along the banks of the Gurgen River, who had formerly been tax-paying subjects of the
Qajars, were allegedly making trouble in the region and refused to pay taxes; the Shah’s
retinue attacked the Turkmens, burning their tents and taking their wives, daughters, and
sons into captivity. Those left alive agreed to pay their dues (Materiali po istorii turkmen
i turkmenii, t. II. XVI–XIX vv. Iranskie, bukharskie, i khivinskie istochniki [Moscow: 1938–
1939], 217). Ten years later, in 1813, the Shah’s troops intercepted a host of Tekes, claimed
by the chronicler to be in the employ of Khwarazm, that had been staging attacks on the
towns of Sabzavar and Juvayni. The Tekes were stripped of their loot and their prisoners;
forty of them were killed, and 100 were taken captive by the Qajar army and sent to Tehran.
Considering the damage insufficient, the Shah delegated Muhammad Qulı̄ Mı̄rzā, who was
˙
at that time governor of Mazandaran and Astarabad, to launch a massive assault on the
Tekes. In the attack which followed, the Ma’āsir-i sultāniya claims that some 2,000 Teke
˙ ˙
men, women, and children were taken captive, and that 50,000 camels, oxen, sheep, horses,
and mares were stolen. Some Iranians who had been held among the Tekes – including
pilgrims to Mashhad and merchants – were freed in the course of the attack (Materiali po
istorii turkmen i turkmenii, t. II, 218). Similar incidents can be observed in nearly every
major Persian chronicle concerning Khurasan in the eighteenth and nineteenth centuries.
In 1841, for example, a Qajar force of 22,000 attacked more than 20 Turkmen auls, making
off with 1,200 camels, 30,000 sheep, 7,000 horses, and 150 captives (Russko-turkmenskie
otnosheniia v XVIII–XIX vv., 339, doc. no. 245). In 1861, a Qajar unit assaulting Yomut
settlements along the Atrek River stole 1,000 camels, 12,000 sheep, and 15 elderly
Turkmen women; the reason, evidently, was that these Yomuts refused to become Qajar
citizens and to surrender the Iranian captives held among them (ibid., 505–506, doc.
no. 374).
2
Thus, in 1828, some Turkmen leaders from the Caspian coast informed the Russians that
many of their people were still held in Iran even after a peace agreement had been
concluded, inspiring resentment in their communities (Russko-turkmenskie otnosheniia
v XVIII–XIX vv., 259, doc. no. 174). It is revealing that, in this document, the Turkmens
kept in Iran are referred to as being held “under arrest” (pod arestom) – a markedly different
terminology than the language of captives (plenniki) and slaves (raby, nevol’niki) typically
used to describe Iranians taken by the Turkmens.
 The Setting 15

Scourge of the Man-Stealing Turcomans,3 and in the terrifying Turkmen

brigands of The Adventures of Hajji Baba of Ispahan, James Justinian
Morier’s once-beloved picaresque novel.4 Depictions of Iranians in
nineteenth-century European literature are likewise generally negative,
but they are relatively multifaceted, drawing on a wider array of stereo-
types; the Turkmens, however, were either barbarous slave-raiders or
slave-raiding barbarians, and nothing more, though on occasion one
finds an appreciative word or two concerning their audacity.
Rare European accounts exist, nevertheless, which detail the horrific
cruelty suffered by some Turkmen captives in Iran. N. G. Petrusevich,
a Russian officer and scientist who was killed at the siege of Dengil
Teppe, reported that in 1861, not long after the disastrous Qajar attempt
to take Merv from the Teke Turkmens, these Turkmens assembled a large
force and began raiding the villages around Mashhad. Qajar troops inter-
vened and many of the Tekes were killed; up to 100 were taken captive.
Petrusevich describes their terrible fate:5
There followed an order by the Shah to deliver the captives to Tehran, and so, in
groups of several individuals, they were shackled by the hands and feet to a single
iron rod and driven on foot to Tehran, which was over 1,000 versts from
Mashhad. The Shah, wishing to reassure the populace, which was displeased
with the shameful showing of the substantial army [which had been defeated at
Merv], ordered that all the captives be executed in front of the gates of the city.
His ministers decided, for the public’s great entertainment [dlia bol’shego
naslazhdeniia publiki], to tie the captives to the city walls and begin shooting
them at a distance of 300 paces. It should be clear enough that the regular infantry
[sarbazy], never having been trained in arming and firing flintlocks, could not
manage to hit the living targets which were laid out so far from them, such that
the pleasure of shooting could last until evening, subjecting the unfortunate
Turkmens to the torments of hell. Learning of these barbaric orders, all of the
ambassadors immediately appealed for the abolition of such executions. But it
was too late: the execution took place, though only when the infantry were
summoned closer. Regardless, the shooting still continued until evening. Some
of the bullets hit not the captives, but the ropes by which they were tied. Thus
untied, they would come forward and sit before the infantrymen, in hopes that

3
Marvin, Merv, Queen of the World, and the Scourge of the Man-Stealing Turcomans
(London: W.H. Allen, 1881).
4
Morier, The Adventures of Hajji Baba of Ispahan (London: MacMillan, 1902).
5
Petrusevich, “Turkmeny mezhdu starym’ ruslom’ Amu-Daryi (Uzboem’) i severnymi okrai-
nami Persii,” Zapiski kavkazskago otdela imperatorskago russkago geograficheskago
obshchestva, Vol. 11 (Tbilisi, 1880), 53–54; see also A. Rzhevuskii, “Ot Tiflisa do Dengil’-
tepe,” Voennyi sbornik 8 (1884), 285.
16 The Setting

they would depart from this life more quickly if they were closer to the infantry,
since they had no hope for mercy.

Another revolting “entertainment” was observed in 1875, and is worth

retelling here both because it is one of the few accounts concerning the fate
of Turkmen captives in Iran and because it speaks volumes about the extra-
ordinary bitterness that could characterize the Turkmen–Qajar relationship:6
In 1875, the brother of the current Shah was appointed governor of Khurasan.
The administration of Khurasan decided to honor his arrival at Mashhad,
Khurasan’s capital, by sacrificing Turkmen captives. They prepared twenty of
them for this, and when the new governor arrived the captives were raised up on
bayonets before him, one after another, so they could be seen by him and by the
majority of the people who had gathered to welcome the Shah’s brother. What
must have been the mortal agony of the captives can be ascertained from the fact
that the last of them, when it was his turn to be raised on bayonets, tried to offer
2,000 tumans (8,000 rubles at the going rate) to ransom himself. But his offer was
not accepted. His hopes were dashed, and he had to go forth to be impaled. He did
not make it, however: death overtook him beforehand, and he fell dead in front of
the governor and the whole gathering.

Turkmens as well as Kazakhs were also taken into captivity in Khwarazm

and Bukhara, and according to the chronicles produced in both of these
domains, the Khwarazmians and Bukharans also very frequently seized
one-another’s subjects in warfare.7 The Khivan chronicle Firdaws al-iqbāl
contains no fewer than nineteen references to conflicts in which “innumer-
able,” “countless,” or “numerous” prisoners were taken (not including one
which resulted in “an unspeakable number” of captives),8 as well as dozens

6
Petrusevich, “Turkmeny mezhdu starym’ ruslom’ Amu-Daryi (Uzboem’) i severnymi okrai-
nami Persii,” 54; Rzhevuskii, “Ot Tiflisa do Dengil’-tepe,” 285. Petrusevich meditated on
the implications of this repulsive scene: “It stands to reason that, having neighbors char-
acterized by such a disposition – and despite the fact that Persia is still seen as a beacon of
civilization, tending to mitigate human cruelty and inspire a more rational vision of human
life – the Turkmen have had nowhere else from which to receive examples of kind-
heartedness, and it is for that reason that they remain primitive in their savagery”
(Petrusevich, “Turkmeny mezhdu starym’ ruslom’ Amu-Daryi (Uzboem’) i severnymi okrai-
nami Persii,” 54).
7
Baron Meyendorff recounted meeting a group of Uzbeks in Bukhara who boasted, “if the
Khan would only give us permission, we would revenge ourselves on the Khivans by
conquering them, killing or taking them prisoners, as we did 10 years ago” (Georg von
Baron Meyendorff, A Journey from Orenburg to Bukhara in the Year 1820 [Calcutta, 1870],
41).
8
Shı̄r Muhammad Mı̄rāb Munis and Muhammad Rizā Mı̄rāb Āgāhı̄, Firdaws al-iqbāl:
˙ ˙
History of Khorezm, trans. Yuri Bregel (Leiden: Brill, 1999), 118, 194, 196, 205, 210,
211, 220, 221, 297, 305, 307, 309, 331, 389, 397, 398, 448, 503, 645.
 The Setting 17

of other mentions of captives being taken in known or estimable numbers.

Naturally, the elaborate bravado of the genre must be kept in mind, and
some of these numbered or numberless victims may have been simply
rhetorical embroidery. A typical passage informs us that in one battle
between the Khivan army and a Qongrat detachment, “the ever-
victorious [Khivan] army captured rich booty and innumerable prisoners.
Thanks to his royal majesty’s valor and courage this large army, as numer-
ous as ants and locusts, was utterly destroyed and the late khan gained
victory. No brave warrior except his majesty ever gained such a marvelous
result.”9
Whatever the exaggerations involved, the nearly constant reference to
captive-taking in the region’s chronicles is revealing. It reveals the nor-
mative quality of captivity as a result of warfare, as well as the prestige
related to taking captives. Prisoners-of-war were listed among the other
spoils of battle – such as camels, horses, and sheep – and were a hallmark
of victory.10 Combatants and noncombatants alike were captured,
though the capture of young women seems to have held particular appeal
in, for example, the world of the Firdaws al-iqbāl, where we learn that
spoils – women among them – were divided between warriors and elites.
Concerning the aftermath of one battle, the chronicle relates that “When
the raiders gathered under the victorious banner, his majesty divided the
booty amongst the troops, and everyone received a great amount of
property . . . Among the prisoners were thirty seven virgin girls of perfect
beauty and exceptional slenderness. His majesty entrusted the prisoners
to reliable and virtuous people and went home victorious and
triumphant.”11 After a particularly successful struggle in which Khivan
and Yomut forces teamed up against the Kazakhs, we learn that “the ever-
victorious troops captured plenty of booty and innumerable prisoners,
including more than 100,000 sheep and more than 40,000 camels; and
from this one may have some idea about the rest of the booty. There were
500 virgin girls among the prisoners, and from this one may estimate the
number of other prisoners. The daughters and harems of Jantu Töre and
Ayten Töre, sons of [the Kazakh Lesser Horde leader] Bölekey Khan, we
also captured. One wife of Burkut Bay Biy and his young sons were also
taken prisoner.”12

9
Firdaws al-iqbāl, 211.
10
E.g. “They continued to pursue the fleeing [Turkmens] till the time of night prayers, killed
many of them, took some prisoners, captured many women and children, seized ten
thousand sheep as booty and went back” (Firdaws al-iqbāl, 197).
11
Firdaws al-iqbāl, 172. 12 Firdaws al-iqbāl, 398.
18 The Setting

While we may assume that young women taken as captives tended to

end up in the harems of elites, there is little clear information on what
became of the other captives taken by Khivan, Bukharan, or Iranian
armies. Arminius Vambery, in one of the only extant descriptions concern-
ing Turkmen captives in Central Asia, claims to have seen three hundred
Chawdur “prisoners of war” awaiting their fate in Bukhara. “They were
separated into two divisions,” he writes, “namely, such as had not yet
reached their fortieth year, and were to be sold as slaves, or to be made
use of as presents, and such as from their rank or age were regarded as
Aksakals (grey beards) or leaders, and who were to suffer the punishment
imposed by the Khan.” The former were led away in groups of ten or
fifteen, chained together by their iron collars. The latter, he claims, suf-
fered a shocking, horrific punishment:13
Whilst several were led to the gallows or the block, I saw how, at a sign from the
executioner, eight aged men placed themselves down on their backs upon the earth.
They were then bound hand and foot, and the executioner gouged out their eyes in
turn, kneeling to do so on the breast of each poor wretch; and after every operation
he wiped his knife, dripping with blood, upon the white beard of the hoary
unfortunate. Ah! cruel spectacle! As each fearful act was completed, the victim
liberated from his bonds, groping around with his hands, sought to gain his feet!
Some fell against each other, head against head; others sank powerless to the earth
again, uttering low groans, the memory of which will make me shudder as long as
I live.

Russian officers and subjects likewise took captives, and could some-
times be found raiding nomadic communities in a manner not dissimilar
from that by which nomads raided sedentary districts. In the summer of
1841, the Orenburg governor-general received a complaint from repre-
sentatives of a Kazakh community that Siberian Cossack bands had been
attacking their village and others nearby. In one alleged assault on some
twenty-five villages, the Cossacks killed forty people, including fifteen
women (all of whom are named in the document), and stole 621 camels,
555 horses, 263 cattle, and 7,770 sheep. This was followed by
a confrontation in which the Cossacks captured twenty-four people
(including eleven women and girls and ten boys between three and ten
years old), killing another eighteen.14 As we shall see in the chapters to
come, the Russian military also took part in the “ransom economy,”

13
Vambery, Travels in Central Asia, 138.
14
Sbornik materialov dliia istorii zavoevaniia turkestanskogo kraia, ed. A. G. Serebrennikov,
Vol. 3 (Tashkent: Tip. Sht. Turkestanskogo V.O., 1908–1912), 84–85, doc. no. 48.
 The Setting 19

trading in captives as well as holding hostages in exchange for political

promises and concessions.
Mass resettlement was another option for disposing of captives, and
sometimes the number of people forcibly resettled at one time could total
in the thousands. Entire towns could be created out of thin air in this
manner, the most famous of which was Khurasan’s “Khivaqābād,”
a Khurasanian town under Qajar jurisdiction consisting entirely of people
forcibly resettled by Nādir Shah after his conquest of Khiva in 1740.15
Turkmens were sometimes the victims of this sort of strategy as well: 1,500
families of Gökleng Turkmens were resettled by the Qajar Najaf ʿAlı̄ Khan
after battling with the Khwarazmians in 1837.16 Resettlement could also
be a defensive strategy: This same Najaf ʿAlı̄ Khan resettled 150 house-
holds of Arabs from the vicinity of the Atrek to the walled village of Katlish
in Khurasan in order to protect them from the inevitable attacks by the
Turkmens.17
In short, captive-taking in Central Asia and its borderlands was not
limited to the Turkmens and Kazakhs; it was a normative tactic of war.
Nevertheless, foreign observers imposed a remarkable double-standard on
this phenomenon, whereby violence done by non-nomads was dignified
with a formal military language that was denied to nomadic combatants.
Russian, Qajar, or Bukharan violence was described with the orderly voca-
bulary of campaigns, expeditions, offensives, operations, detachments, sol-
diers, regiments, and prisoners-of-war. Turkmen or Kazakh violence,
meanwhile, was described using the contemptuous language of raiding,
pillaging, plundering, looting, and slave-taking. At the heart of this division
is the idea that the violence perpetrated by settled peoples – and especially
by European empires – has a legitimacy that nomadic violence lacks.
These Turkmens and Kazakhs are very often called, in our Russian and
English sources, “pirates” (piraty), “brigands” (razboiniki), and “preda-
tors” (khishchniki). Their activities were compared to those of the Barbary
pirates from the North African coasts and the caravan-robbers of the
Sahara. Thus, Ferrier writes that “A horse is to the Turcoman what a ship
is to the pirate.”18 For Vambery, “What the Portuguese slave traders and
15
These events are described in, for example, Mı̄rzā Mahdi Khan Astarābādı̄’s Tārı̄kh-i nādirı̄
(cf. Materiali po istorii turkmen i turkmenii, tom 2. XVI–XIX vv. Iranskie, bukharskie,
i khivinskie istochniki [Moscow, 1938], 146).
16
Muhammad Hassan Khan, Tārı̄kh-i muntazam-i nādirı̄ (cf. Materialy po istorii turkmen
˙ ˙ ˙
i turkmenii, tom 2, 235).
17
C. E. Yate, Khurasan and Sistan (London: Blackwood and Sons, 1900), 202–203.
18
J. P. Ferrier, Caravan Journeys and Wanderings in Persia, Afghanistan, Turkistan, and
Beloochistan (London: J. Murray, 1857), 94. Cf. also Mikhail Khodarkovsky, Russia’s
20 The Setting

the Arabian ivory merchants are in Central Africa, that are the Turkomans
in the north-eastern and north-western portions of Iran, indeed we may
say in all Persia. Wherever nomad tribes live in the immediate neighbour-
hood of a civilised country, there will robbery and slavery unavoidably
exist to a greater or less extent.”19 Vambery’s comment is typical in
associating nomadism itself with brigandage. The perceived independence
and statelessness of nomads was often seen as an inherent threat to “civi-
lization,” as the nomads’ seemingly unbounded migrations were thought
to undermine borders and citizenship, as well as taxation, agriculture,
urbanism, and all the allegiances and systems of loyalty that were regarded
as hallmarks of “settled” statehood.
The nomads were also alleged to be impervious to diplomacy, or even to
reason. Joseph Wolff, the eccentric, globetrotting missionary, wrote that
“The Turkomauns of the desert of Mowr and Sarakhs are a people of such
a perfidious disposition, and of such great rapacity, that one could not
depend for a moment on their promises, or on any treaties entered into
with them; for the Turkomauns, as well as the Beduins in the deserts of
Arabia, do not consider consequences, but are only restrained by instant
infliction of punishment.”20 Alexander Gorchakov, the Tsar’s chancellor
in the era of Central Asia’s conquests, would echo these sentiments in his
famous circular from November of 1864, in which he reasoned that “the
tribes on the frontier have to be reduced to a state of more or less perfect
submission” – and reduced violently, since, he wrote, “It is a peculiarity of
Asiatics to respect nothing but visible and palpable force; the moral force

Steppe Frontier: The Making of a Colonial Empire, 1500–1800 (Bloomington, IN: Indiana
University Press, 2004), 29; Gao, “Captivity and Empire,” 3–4.
19
Vambery, Sketches of Central Asia, 206. Vambery elaborates: “The poverty-stricken
children of the desert are endowed by nature with an insatiable lust for adventure, and
frames capable of supporting the most terrible privations and fatigues. What the scanty
soil of their native wilderness denies them, they seek in the lands of their more favoured
neighbours. The intercourse between them, however, is seldom of a friendly character.
As the plundered and hardly used agriculturist cannot, and dare not, pursue the well-
mounted nomad across the pathless deserts of sand, the latter, protected by the nature of
the country, can carry on his career of plunder and rapine without fear of chastisement.
In former times the cities on the borders of the Great Sahara and of the Arabian desert
were in the same plight. Even at the present day the caravans in the latter country are
exposed to the greatest dangers. But Persia has to suffer from these evils to a still greater
extent, as the deserts which form her northern boundary are the most extensive and the
most savage in the world, while their inhabitants are the most cruel and least civilised of
nomads.”
20
Joseph Wolff, Narrative of a Mission to Bokhara in the years 1843–45, to Ascertain the Fate
of Colonel Stoddart and Captain Conolly, Vol. 1 (London: John W. Parker, 1845), 272.
 The Setting 21

of reason and of the interests of civilization has as yet no hold upon

them.”21
While the nomads were typically singled out for a special degree of
contempt, these same traits were sometimes attributed to Khwarazm and
Bukhara as well. But even then the nomads were implicated: One of the
most common critiques of these two states – and particularly of
Khwarazm – is that they benefitted from the thievery of nomads, forming
alliances with them or employing them as agents and mercenaries.
Thus M. Ivanin, deploring Khwarazm’s rulers for “inciting” the
Turkmens and Kazakhs to abduct Russians and sell them into slavery,
writes that “Khiva, by its actions in relation to its neighbors, could be
dubbed the Algeria of Central Asia, as the foremost enterprise of its
government has been to rob their neighbors and traffic in people.”22
Khwarazmian elites did indeed benefit from this traffic directly, by owning
the slaves themselves and putting them to work on their estates.
Even so, Ivanin wildly exaggerates the centrality of the slave trade (not
only for Khwarazm, but also for Algeria), and, in a typical fashion, sim-
plifies the relationship between Turkmens, Kazakhs, and Khwarazmian
rulers. The nomads were not just “agents” of the khan, but an essential
part of the fabric of Khwarazmian society; the Turkmen and Kazakh
populations of Khwarazm, though not entirely nomadic, may at times
have added to as much as 40 percent of the total population of the khanate.
The khan benefitted not only from their military participation, which was
considerable, but also from their tax revenues and their commerce more
generally. Relations between these populations and the khan were often
strained; the years 1855–1867, for example, witnessed continual uprisings
among Yomuts and other Turkmen groups who had formerly been con-
sidered Khwarazmian subjects.23 Baron Meyendorff succinctly summed
up a general predicament faced by the khans with respect to the Turkmen
and Kazakh populace: “The Nomads, who wander about all over the
country,” he writes, “could easily leave it altogether, so that their Chiefs
are compelled to treat them gently, and have even, sometimes, to flatter

21
Demetrius Charles Boulger, England and Russia in Central Asia, vol. 1 (London: W.H.
Allen and Co., 1879), 319.
22
M. Ivanin, Opisanie zimnogo pokhoda v khivu v 1839–1840 g. (St. Pb, 1874), 18. Here,
Ivanin echoes the sentiment articulated by General Perovskii, who, in 1835, wrote to
St. Petersburg to urge a “punitive” expedition against the Khivans: “The guilt of the Dey of
Algiers against the King of the French pales into insignificance in comparison with the
crimes carried out by whole generations of Khivan Khans against the Emperors of Russia”
(Morrison, “Twin Imperial Disasters,” 284).
23
Yuri Bregel, Khorezmskie turkmeny v XIX veke (Moscow, 1961), 197–228.
22 The Setting

them. The present khan has not had enough regard for this precept, and
has therefore lost many Turcomans, who, having subjected themselves to
the khan of Khiva, show their fidelity by wasting and plundering the
countries belonging to their former master.”24

MAJOR FEATURES AND ORIGINS OF THE SLAVE TRADE

What did the khanates stand to gain from the slave trade, which became by
the mid-nineteenth century such a flashpoint in diplomacy with Russia,
Iran, and Great Britain? They gained quite a lot, as we shall see, and these
gains were not limited to free labor. First, tax revenues could be gained
from the trade in slaves. According to Ismāʿı̄l Sarhang Mı̄rpanja, who was
imprisoned for ten years in Khiva, a tax of one tilla was extracted for every
˙
single sale of a slave.25 N. N. Murav’ev, whose travelogue of a mission to
Khiva provides some of the most detailed reportage on Khwarazmian
slavery, writes that all subjects of the khan who engaged in raids into Iran
would owe in taxes one-fifth of any spoils they brought back to the
khanate.26 Vambery, traveling in a caravan full of emancipated slaves,
observed that the transport tax for this cargo was extracted beyond
Khwarazm as well, and sometimes the levies could be considerable.27
Slaves purchasing their own freedom from their owners were also taxed
some portion of their value.28
The ransom economy also served as a significant financial incentive for
perpetuating the slave trade. Of those captives taken in the relatively
affluent district of Mazandaran and along the southern Caspian coast,
Vambery estimated that one-third would be ransomed back into the care
of relatives rather than being sold north to Khwarazm or Bukhara.
24
Meyendorff, A Journey from Orenburg to Bukhara in the Year 1820, 50.
25
Mı̄rpanja, Khātirāt-i asārat: ruznāma-yi safar-i khwārazm va khiva, ed. Safā’ ad-Dı̄n
Tabarrā’iyān (Tehran: Muʿssassa-yi Pazhuhish va Mutālaʿat-i Farhangı̄, 1370/1991), 119.
26
N. N. Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 1819–20
(Calcutta: Foreign Department Press, 1871), 140.
27
“Here, as everywhere,” Vambery writes, while entering Jamshidi country, “our difficulties
began and ended with questions respecting the customs. It had been said, all along, that
with the left bank of the Murgab Afghanistan began, and that there the slave tax would
cease to be exacted. It was a grievous mistake. The Khan of the Djemshidi, who treated in
person with the Kervanbashi concerning the taxes, exacted more for goods, cattle, and
slaves than the former claimants, and when the tariff was made known, the consternation,
and with many the lamentation, knew no bounds” (Vambery, Travels in Central Asia, 260).
28
James Abbott, Narrative of a Journey from Heraut to Khiva, Moscow, and St. Petersburgh,
during the Late Russian Invasion of Khiva, Vol. 2 (London: W.H. Allen and Co., 1843),
288; Arminius Vambery, Travels in Central Asia, 235.
 Major Features and Origins of the Slave Trade 23

Captives taken from poorer regions of Khurasan and Sistan rarely had
relatives who could pay a suitable ransom price, however, and so most
were sold into slavery: “I have heard,” Vambery writes, “out of the mouth
of a slave dealer who had grown grey in his trade, that from these districts
scarcely a tenth part are ransomed, the remaining nine-tenths being for-
warded for sale in the markets of the khanats.”29 Ransom was no doubt the
outcome of first resort from the perspective of the captive-takers them-
selves, not only for its greater financial rewards but also because it elimi-
nated the need to transport captives across long distances. The bounty
gained from ransom was also free from the fluctuating prices of the north-
ern markets and independent of the captive’s physical traits; if slave prices
were down in Khiva, or if the captive in question was ill-suited to labor,
relatives would still pay top dollar to liberate a loved one.
Mı̄rzā Mahmūd Taqı̄ Āshtiyānı̄, who was himself held for ransom
˙
among the Sarı̈q Turkmens, records the system by which these ransom
exchanges would take place. Sometimes, the families of captives would
send coins and promissory notes (called barāt) via caravans traveling from
Iran, which merchants from the caravans would exchange for their loved
ones. Otherwise, the merchants themselves might pay out-of-pocket for
the captives, having received a promise of reimbursement from their loved
ones back in Iran.30 Families with the means to do so could also hire
Turkmen agents to “kidnap” their enslaved relatives in their places of
bondage and bring them back home.31 Once within Khwarazm or
Bukhara, however, ransom would have been a much greater challenge, as
rulers of these domains – as we shall see – were disinclined to allow the
repatriation of slaves into non-Sunni environs. The process of ransoming
or otherwise extracting slaves from the khanates was thus always done
covertly.32
Our sources abound with tales of former slaves who, having purchased
or otherwise gained their own freedom, attempt to buy the freedom of
family members who had likewise been taken into slavery. In 1804–1805,
a woman named Akulina Krivobokova approached Orenburg border
authorities after having lived for thirty-one years in slavery, having

29
Vambery, Sketches of Central Asia, 213–214.
30
Mı̄rzā Mahmūd Taqı̄ Āshtiyānı̄, ʿIbratnāma: Khatirati az dawran-i pas az jangha-yi herat va
˙
merv, ed. Husayn ʿImadi Ashtiyani (Tehran: Nashr-i markaz, 1382/2003), 29–32; cf.
Amanat and Khazeni, “The Steppe Roads of Central Asia and the Persian Captivity
Narrative of Mahmud Mirza Taqi Ashtiyani,” 125.
31
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 156.
32
“Nevol’niki v Khive,” Vestnik evropy 80:7 (1815), 245.
24 The Setting

apparently been owned by at least one Khivan khan (v nevol’nichestve

khivinskogo khana). Relatives had purchased her freedom the preceding
summer, but her two sons and two daughters remained in captivity.
The cost of buying their freedom was more than the family could afford,
and she had come to the Russian government in order to petition them to
provide the necessary funds.33 Vambery reports having met in a single
caravan several people who had been in a similar position, seeking to
purchase or having purchased their loved ones’ freedom:34
There were in the karaván, as I remarked at the first station, many others besides
myself who were longing to reach the southernmost frontiers of Central Asia.
These were the emancipated slaves, with whom Hadjis were intermixed, and
I had an opportunity of witnessing the most heart-rending incidents. Near me
was an old man – a father – bowed down by years. He had ransomed, at
Bokhara, his son, a man in his thirtieth year, in order to restore a protector to his
family left behind – that is to say, to his daughter- in-law a husband, to his children
a father. The price was fifty ducats, and its payment had reduced the poor old man
to beggary. “But,” said he to me, “rather the beggar’s staff than my son in chains.”
His home was Khaf in Persia. From the same city, not far from us, was another man,
still of active strength, but his hair had turned grey with sorrow, for he had been
despoiled by the Turkomans, some eight years ago, of wife, sister, and six children.
The unfortunate man had to wander from place to place a whole year in Khiva and
Bokhara, to discover the spot in which those near members of his family were

33
TSGKaz 4.1.499 67a–69a.
34
Vambery, Sketches of Central Asia, 235–237. Elsewhere, Vambery records other affecting
narratives concerning the vicissitudes of the ransom economy: He writes a letter to
a relative on behalf of a young Iranian domestic slave “praying them for God’s sake to
sell sheep and house to ransom him” (Travels in Central Asia, 60); he meets a five-year-old
who, having been captured and sold two years before along with his father, reports that his
father had bought his own freedom and that he expected his father would free him too
before long (ibid., 163); and he comments on the challenging position of the newly
captured Iranian of wealthy background, who, on the one hand, wishes to be ransomed
but, on the other, hopes to hide evidence pointing toward the vastness of the ransom that
might be gotten for him: “This poor Persian was transferred, for chastisement, to [the
Turkmen] Kulkhan, who had the peculiar reputation of being able most easily to ascertain
from a captive whether he possessed sufficient means to enable his relatives to ransom him,
or whether, being without relatives or property, he ought to be sent on to Khiva for sale.
The former alternative is much the more agreeable one to the Turkomans, as they may
demand any sum they please. The Persian, who is cunning even in his misfortune, always
contrives to conceal his real position; he is therefore subjected to much ill-treatment until
by the lamentations which he forwards to his home his captors have squeezed from his
friends the highest possible ransom, and it is only when that arrives that his torment
ceases” (ibid., 75–76). The ordeal described by Vambery, in which ransoms are gradually
extracted by tormenting a wealthy captive, are consistent with Āshtiyānı̄’s experiences as
well (cf. Amanat and Khazeni, “The Steppe Roads of Central Asia,” 126).
 Major Features and Origins of the Slave Trade 25

languishing in captivity. After long search, he found that his wife, sister, and two
youngest children had succumbed under the severity of their servitude, and that, of
the four children that survived, he could only ransom half. The remaining two
having besides grown up, the sum demanded for them was beyond his means.
Farther on sat a young man from Herat, who had ransomed his mother. Only two
years ago, this woman, now in her fiftieth year, was, with her husband and eldest
son, surprised by an Alaman [raid]. After seeing those near relatives both fall, in
self-defense, under the lances and swords of the Turkomans, she experienced
herself unceasing sufferings until sold for sixteen ducats in Bokhara. The owner,
discovering a son in him who sought to ransom her, exacted a double amount, thus
turning filial piety to cruelly usurious account. Nor must I omit to mention another
unhappy case – that of an inhabitant of Tebbes. He was captured eight years ago,
and after the lapse of two years he was ransomed by his father. They were both
returning home, and were three leagues from their native city, when they were
suddenly attacked by the Turkomans, taken prisoners, led back to Bokhara, and
again sold as slaves. Now, they were a second time freed, and were being conveyed
to their homes.

Still more important than the tax and ransom revenues, however, were the
captives themselves: by all accounts, a substantial proportion of the agri-
cultural labor both in Khwarazm and in Bukhara was performed by slaves,
and slaves constituted a significant demographic in these states’ militaries
as well. For their part, those Turkmens and Kazakhs bringing slaves to the
market would benefit from payment not only in cash, but also in goods
and – most importantly – in the crucial stocks of grain that nomads of the
arid zones were unable to produce themselves. This “grain-for-slaves”
commerce was one aspect of the symbiotic relationship between slave-
traders and the settled states.
It would be wrong to see captive-taking in purely economic terms,
however, especially in light of the overarching culture of captivity dis-
cussed earlier, in which it was a standard part of the strategic vernacular of
warfare. Though foreign observers did not dignify Turkmen and Kazakh
raids as part of an ongoing political struggle, this is precisely what they
were. For both groups, the long-term goal was often independence – either
a degree of it within the confines of fealty to Khwarazm or Russia, or
complete independence in regions that were increasingly encroached upon
by expansionist imperial neighbors. Attacks on the villages of Khurasan
provided revenue, but they also served to create a buffer zone of weakened,
destabilized territory between Tehran and the Turkmen deserts. Kazakh
attacks on caravans and settlements likewise came with financial benefits,
but they also asserted dominance over the steppe in an era when Russian
26 The Setting

power was increasingly visible at its peripheries. Raiding was a form of

resistance.
As I will show in the chapters to come, the slave trade also fulfilled many
other functions for the captive-takers and slave-owners. For the slaves
themselves, captivity could be as brief as a matter of days before being
ransomed or as long as several decades in bonded labor. Many would die
during the brutal ordeal of being transported across the deserts by their
captors. Those who survived the journey would join tens of thousands of
other slaves in Khwarazm, Bukhara, and the Kazakh steppe, performing all
manner of work. Agricultural labor predominated, especially for Iranian
slaves, but we also find slaves working as soldiers, teachers, carpenters,
blacksmiths, musicians, drovers, prostitutes, miners, herders, and dancers.
Most were converted to Sunni Islam. Many were married, sometimes
forcibly, and most often to other slaves. Fortunate slaves were permitted
to perform independent labor in their limited free time, the proceeds of
which were their own to keep (if their masters did not expropriate them);
by this means, a great many slaves purchased their own freedom. A very
fortunate few, either during or after their period of bondage, managed to
achieve prominent positions in the royal dı̄vān or in the military.
The estimated total number of slaves in Khwarazm and Bukhara varied
throughout the nineteenth century from tens of thousands to hundreds of
thousands. There has been consensus on one key point: The overwhelming
majority of slaves were Iranians. The earliest estimates concerning the
number of Iranian slaves in Khwarazm alone were generally in the range of
30,000–40,000, though on the higher end we find estimates of up to
140,000. Similarly, for Bukhara, estimates on the number of enslaved
Iranians tend to range from 30,000 to 40,000; one estimate – by Joseph
Wolff – put the figure as high as 200,000, a sum which seems impossible but
which we should hesitate to dismiss entirely, given Wolff ’s claim that the
estimate was offered by the Bukharan Amı̄r himself (albeit in passing).35
By the mid-late nineteenth century, the claim that Iranian slaves in
Khwarazm numbered as many as 50,000–60,000 became more common.
By contrast, the estimated number of Russian slaves plummeted as the
century wore on. Unsurprisingly, the number of Russian slaves, despite
being by all estimates a tiny fraction of the total slave population,
inspired much more commentary among Russian and British travelers,
officials, and observers. Estimates from the early decades of the century
posited as many as 15,000 Russian slaves in Khwarazm and Bukhara

35
Wolff, Narrative of a Mission to Bokhara in the years 1843–45, 226.
 Major Features and Origins of the Slave Trade 27

combined. The best-known and most widely reported estimate was

relayed by Murav’ev, who, during his mission to Khiva, discovered
a haunting secret letter from a representative of the Russian slave
population:36
As I was preparing to clean this gun, I discovered a slip of paper in one of the
barrels, on which the following was written in Russian: “We venture to inform your
Honor that there are over 3,000 Russian slaves in this place, who have to suffer
unheard-of misery from labour, cold, hunger, &c. Have pity on our unhappy
situation and reveal it to the Emperor. In gratitude we shall pray to God for your
Honor’s welfare.” The perusal of these lines deeply affected me, and I thanked God
that I should, perhaps, have the fortune to serve as an instrument of help.

These numbers align more or less with the estimate offered by Lieutenant
Gladyshev, who proposed that there were 3,000 Russians, Kalmyks, and
“foreigners” in the khanate, a number of whom he saw cleaning the canals
when he visited Khiva in 1740–1741.37 They also align with the estimate
of an Orthodox priest named Khrisanf, who traveled in the region near the
end of the eighteenth century, estimating some 4,000 Russian slaves in
Khiva (as well as 6,000 in Bukhara).38 In 1840, however, when the Khivan
khan, as a concession to Russia, released what he claimed were all the
remaining Russian slaves in the khanate, they totaled fewer than 500.
Another twenty-one were freed on the eve of Khiva’s conquest in 1873,
and not many more appear to have been liberated during that event. It is
possible that the early estimates were exaggerated; or that a great many of
the Russian slaves had since been manumitted and returned home or
assimilated totally into Central Asian communities; or – most likely –
that they simply died and were not replaced, as the overall trade in
Russian captives decreased over the course of the nineteenth century.
When it comes to Bukhara, meanwhile, there is general agreement,
and among a greater number of eyewitnesses, that very few Russian slaves
remained there by the middle of the nineteenth century. Jan Prosper
Witkiewicz, a Polish exile who entered Russian service and ventured to
Bukhara as a diplomat for the Tsar, met some twenty-five of them per-
sonally, and estimated their total number at no more than fifty.39 Burnes
put the number at 130,40 and Kostenko observed simply that they were
36
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 1819–20, 77.
37
Gladyshev and Muravin, Poezdka iz Orska v Khivu i obratno, sovershennaia v 1740–41
godakh poruchikom Gladyshevym i geodezistom Muravinym (St. Pb, 1851), 18.
38
Puteshestviia po Vostoku v opokhu Ekateriny II, 279.
39
Vitkevich, Zapiski o Bukharskom khanstve (Moscow: Nauka, 1983), 115.
40
Burnes, Travels into Bokhara, Vol. 2 (London: Carey and Hart, 1835), 115.
28 The Setting

few.41 Nevertheless, the freeing of Russian slaves continued to play

a significant role in Russian diplomacy with both Bukhara and
Khwarazm down to the age of the region’s conquest.
The abundance of slaves increased during times of heavy warfare along
the Iranian frontier, and the increase could reportedly be substantial.
Baron Meyendorff claimed that a particularly violent struggle for Merv
increased the number of Iranian slaves in Bukhara by 25,000 (bringing the
total, in his estimation, to 40,000 overall).42 Naturally, any great influx of
slaves resulted in lower prices at the markets.43 Regardless of the abun-
dance of slaves for sale, however, prices could vary considerably. The most
commonly cited market price for a slave, which may be taken as an overall
average throughout the nineteenth century, is between thirty and sixty
gold coins (tilla, chervonits),44 in both Bukharan and Khwarazmian cur-
˙
rency. Indeed, many former slaves who reported their own original sale
price to Russian border officials were sold for a price in the twenty-five to
forty coin range. But the price of a slave depended on a variety of factors,
including age, origins, gender, and physical condition. Major
Blankenagel’, a physician who visited Khiva in 1792 to treat the khan for
an eye malady, reported that the Kazakhs were selling Russian captives at
the Khivan and Bukharan bazaars at a rate of forty to fifty coins for a man
and anywhere from 50 to 100 coins for a woman.45 Meyendorff observed
precisely the opposite, claiming that “The women, as a rule, are cheaper
than the men, except those still young and handsome.” He writes that the
price of a “well-built” man of any background averaged forty to fifty tillas,
˙
though knowledge of a useful craft, such as blacksmithing, could raise
a slave’s price to 100 tillas. Young and attractive women, meanwhile, could
˙
sometimes fetch 100–150 tillas.46 Murav’ev writes that Iranian men
˙
fetched a lower price than Russian men – generally twenty to thirty tillas,
˙
as opposed to sixty to eighty tillas for a “young and healthy” Russian – but
˙

41
L. F. Kostenko, Puteshestvie v Bukharu russkoi missii v 1870 godu (St. Pb, 1871), 107.
42
Meyendorff, A Journey from Orenburg to Bukhara in the Year 1820, 61–62; cf. also
M. Alikhanov-Avarskii, Pokhod v Khivu (Kavkazkikh otriadov) 1873. Step’ i oazis (St. Pb,
1899), 280.
43
Cf., for example, A. Maslov, “Rossiia v Srednei Azii. (Ocherk nashikh noveishikh
priobretenii),” Istoricheskii vestnik 5 (1885), 386.
44
I estimate that one tilla in nineteenth-century Bukhara was generally worth about 6.4
British pounds sterling, or 26 francs, though there was significant variation in the relative
value of silver over the course of the century.
45
Blankenagel’, Zamechaniia maiora Blankenagelia, vposledstvie poezdki ego iz Orenburga
v Khivu v 1793–94 godakh (St. Pb, 1858), 12–13.
46
Meyendorff, A Journey from Orenburg to Bukhara in the Year 1820, 61–62.
 Major Features and Origins of the Slave Trade 29

that Iranian female slaves sold for much higher prices than Russian
women; Kurds fetched the lowest sums.47 Witkiewicz wrote that the
ruler of Kunduz was at the time of his visit “constantly conducting raids
on the peoples of the district, taking captives; and his merchants bring
them to Bukhara. 20–50 tillas – or Bukharan coins – are paid for them.
˙
A pleasant maiden fetches up to 70 tillas; a comely (prigozhii) boy, up to
˙
40. But workers usually fetch no more than 30 tillas.”48
˙
Slaves were sold and traded extensively beyond the bazaars, however,
and they were very often traded for objects or livestock, especially among
the Turkmens and Kazakhs. Former slaves report having been traded for
a wide range of goods and animals. Trading slaves for sheep was evidently
quite common; they could be sold for as few as thirty-six sheep,49 though
a higher valuation is more customary. One slave who had previously been
sold for forty-one tangas50 was traded for eighty sheep;51 two others for
100 sheep each;52 another for 150 sheep;53 another for twenty-two sheep
and a horse;54 another for twenty sheep and three horses;55 and another
for ten “big sheep,” one camel, one horse, and three large felt carpets.56
Others report having been traded for twenty Khivan silk robes (khalat);57
one camel and one horse;58 one camel and two horses;59 and nine horses,
one fleece and fur coat, and a gun.60 Many slaves were sold or traded
multiple times, and the price for the same slave could vary dramatically
from one sale to the next. The slave who had been traded for twenty
Khivan silk robes had earlier been sold by Turkmens to a Khivan for nine
tillas, and he was later sold to a Kazakh for twenty-nine tillas.61 A slave
˙ ˙
who was traded to a Kazakh for forty mares was, twenty-one years later,
traded to another for just ten mares – the difference owing no doubt to his
advanced age.62
Among the first indignities suffered by many slaves in the region was the
loss of their name. New owners were at liberty to change their slaves’
names, and they very often did. Sometimes the masters proved themselves

47
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 1819–20, 57–58,
148.
48
Ia. P. Vitkevich, Zapiski o Bukharskom khanstve (Moscow: Nauka, 1983), 102.
49
TsGAKaz 4.1.3573, f. 132a.
50
A tanga is a silver coin, its value roughly twenty kopeks in this period.
51
TsGAKaz 4.1.2821, ff. 6a–b. 52 TsGAKaz 4.1.198, f. 19a; TsGAKaz 4.1.3646, f. 47a.
53
TsGAKaz 4.1.3646, f. 49a. 54 TsGAKaz 4.1.3646, f. 46a.
55
TsGAKaz 4.1.3646, ff. 75b–76b. 56 TsGAKaz 4.1.3646, ff. 74a–75a.
57
TsGAKaz 4.1.3646, ff. 77a–78a. 58 TsGAKaz 383.1.89, ff. 14a–b.
59
TsGAKaz 4.1.3646, f. 47b. 60 TsGAKaz 4.1.198, ff. 36b–37a.
61
TsGAKaz 4.1.3646, ff. 77a–78a. 62 TsGAKaz 4.1.499, ff. 104a–b.
30 The Setting

creative in this effort, as with the Kalmyk slave who was renamed “Manas”
(the legendary hero whose chief occupation was the slaying of Kalmyks).63
More often, however, they proved themselves remarkably uncreative.
In July of 1852, a Russian border official logged the arrival of eight escaped
slaves, whom he identified (using the names given to them in captivity) as:
“Nazar, Dawlat, Dawlat, Dawlat, Nazar, Mustafa, the women Summanaz,
and Dawlat.”64
Having been sold, transported, and often renamed in the khanates or in
the steppe, accounts differ wildly concerning the sort of treatment slaves
could expect in their period of bondage. The earliest Russian reports of
slaves’ treatment in Khwarazm are almost invariably grim. An article from
1815 reports that “[t]hese sufferers’ food consists of two unleavened flat-
breads per day, sometimes some gruel, and very rarely a piece of lean meat.
The only vegetables and fruits allowed to them are those beginning to
spoil. By way of clothing they get one shirt per year, and one robe (khalat)
every two years. They are rarely given shoes, and the ones they are given
are worn out. Their bed is of straw and reeds.”65 Khwarazmian slaves’
starvation was likewise reported by Murav’ev, who writes that the “diet of
the slaves and servant class is very bad, the latter have to be content with
what is left from the tables of their masters, and they struggle and fight
amongst themselves for the fragments . . . These wretched creatures fre-
quently go a whole day without a meal, and keep soul and body together by
what they can beg or steal.”66 Conditions were reported to be similarly
harsh in Bukhara and among Teke Turkmens to the south.67
Many travelers reported on horrific punishments that were visited upon
slaves who attempted escape, or otherwise earned their owners’ ill-will.
63
TsGAKaz 4.1.198 f. 54a. 64 TsGAKaz 4.1.3641 f. 56a.
65
“Nevol’niki v khive,” Vestnik evropy 80:7 (1815), 245.
66
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 161.
67
On Bukharan slavery, Desmaisons writes that “Russian captives, although treated less
barbarously than in Khiva, nevertheless live in very tough circumstances. In addition to
the hard labor which they are forced to perform, they are kept in conditions demonstrating
the loathsome stinginess generally characteristic of Bukharans. They are often deprived of
basic necessities; they receive meager sustenance, and sometimes die of chronic malnutri-
tion” (Zapiski o Bukharskom khanstve, N. A. Khalfin ed. [Moscow, 1983], 25). V. A. Tugan-
Mirza-Baranovskii offers a similar appraisal of slavery among the Tekes: “The position of
slaves is generally unenviable, but it is particularly so among the Tekes. To say nothing of
those who languish for years shut up in some mud hut with heavy logs tied to their legs and
shackles on their arms, the lives of other slaves, who enjoy more freedom relative to these
ones, working the fields and grazing the herds and serving their owners, are very difficult
owing to the scarcity of food given to them and to the beatings and insults constantly
inflicted upon them. Turkmen or Kirghiz slaves, however, are treated much better than the
Persians, [other] Shiʿites, and Russians.” (Russkie v akhal teke [St. Pb, 1881, 71).
 Major Features and Origins of the Slave Trade 31

Some of these punishments were witnessed firsthand, some reported by

slaves, and some only rumored.68 The most terrible punishments, accord-
ing to broad consensus, awaited fugitive slaves who had been recaptured.
Several reports contend that such slaves were punished by impalement.
“Before our arrival,” one Russian observer writes, “an unfortunate one was
subjected to such a fate, with the spike entering through his flank, and this
Persian lived in such a situation and in terrible suffering for two days. He
pleaded in vain to the people around him for a sip of water to quench the
terrible thirst that was consuming him, but under penalty of death they
were forbidden from carrying out the request of the convicted person, and
he died, cursing the Khivans and the day they were born.”69 Another
claims to have witnessed such a terrible spectacle with his own eyes:
“[A]long the road” outside Khiva, “a few Persians were impaled by
stakes . . . With their arms tied parallel to their legs, these unfortunate
people were finishing their lives in terrible pain, filling the air with loud,
pathetic cries: ‘Su! Su! Su!’ [Water! Water! Water!]. The Khivans accom-
panying [our] embassy explained that the crime of these unfortunate
people was that, having been captured by [Yomut] Turkmen robbers and
sold into slavery in Khiva, they conspired and fled. The Khivans caught
them the next day, and now, in order to teach a lesson to other Persian
slaves, and for the edification of departing Russians, the cruel khan
ordered all of these unfortunates staked down on the same day our
embassy was accomplished, and along its route of travel.”70 One widely-
reported method of impeding slaves’ ability to flee involved cutting their
feet or heels and stuffing horsehair in the wound.71

68
Meyendorff, for example, reports seeing a slave “whose master had cut off his ears, pierced
his hands with nails, and, taking the skin off his back, had poured boiling oil on his arms, so
as to force him to tell by what means a comrade of his had escaped” (Journey of the Russian
Mission from Orenbourg to Bokhara, 61). Murav’ev writes of slaves being nailed to doors
by an ear or having an ear cut off, being deliberately starved, having an eye gouged out, or
being stabbed (Muraviev’s Journey to Khiva through the Turcoman Country, 49, 136).
69
N. Zalesov, ‘Pis’mo iz Khivy’, Voennyi sbornik 1 (1859), 287.
70
Zakhar’in, “Posol’stvu v Khivu v 1842 godu,” Istoricheskii vestnik 11 (1894), 445.
71
See, for example, Lucy Atkinson, Recollections of Tartar steppes and their inhabitants
(London: John Murray, 1863), 290; Zakhar’in, “Posol’stvu v Khivu v 1842 godu,” 445;
MacGahan, Campaigning on the Oxus, and the Fall of Khiva (London: Sampson Low,
1874), 311. This manner of punishment is also inflicted on Leskov’s Ivan in The Enchanted
Wanderer during his captivity among the Kazakhs: “Some ten men threw me down on the
ground and said, ‘Shout, Ivan, shout louder when we start cutting. It’ll be easier for you.
And they sat on me, and in a trice one master craftsmen of theirs cut the skin open on my
soles, put in some chopped-up horse-hair, covered it with the skin, and sewed it up with
string. After that they kept my hands tied for a few days, for fear I’d harm the wounds and
the bristles would come out with the pus; but once the skin healed, they let me go: ‘Now,’
32 The Setting

Notwithstanding horrors such as these, a number of visitors reported

that cruel treatment was the exception rather than the rule for slaves in the
khanates. The British adventurer Alexander Burnes “heard from every
quarter that slaves were kindly treated,” and he writes that he “never
heard these [slaves], in my different communications with them, complain
of the treatment which they experienced in Toorkistan . . . [T]hey are never
beaten, and are clothed and fed as if they belonged to the family, and often
treated with great kindness.”72 Later observers, including Arminius
Vambery, Januarius Aloysius MacGahan, and N. Zalesov would follow
suit in offering optimistic appraisals of slaves’ treatment.73
Later chapters will explore slaves’ treatment in terms of the opportu-
nities available to them and the limitations imposed upon them. When it
comes to general patterns of mercy or cruelty, the variation in travelers’
accounts probably indicates a diverse range of experiences that were
possible for slaves. To characterize slaves’ treatment by emphasizing only
the horrific cruelty of certain punishments – as some nineteenth-century
commentators did – obscures the fact that slaves did not spend the entirety
of their captivity being punished for particular offenses. On the other
hand, claims of “good” treatment at the hands of slave owners obscure
the fact that the slave system itself was inherently cruel and degrading even

they say, ‘greetings to you, Ivan, now you’re our real friend and you’ll never go away and
leave us’” (Leskov, The Enchanted Wanderer and Other Stories, trans. Pevear and
Volokhonsky [New York: Alfred A. Knopf, 2013], 150).
72
Burnes, Travels into Bokhara, Vol. 1, 283, 342.
73
Vambery suggests, rather naively, that among Iranian slaves the greatest injury was suffered
by their pride: “The Persian is,” Vambery writes, “from his childhood accustomed to the
most refined politeness, and to a flowery, elegant conversation; and must of course suffer
mentally a great deal when first introduced to the savage manners and habits of Turkestan.
His physical sufferings are by no means so great. The majority of them, destined for
agricultural labour, generally gain the confidence and affection of their master by their
good behavior. If a slave has during a year not incurred punishment, he is soon looked
upon as a member of the new family” (Sketches of Central Asia, 223). He allows, however,
that disobedient slaves could be punished very harshly indeed (Sketches, 224). MacGahan,
who witnessed the conquest of Khiva, writes that the slaves he observed there were not, “so
far as I could learn, treated so badly. They get enough to eat and drink; and as to their
clothes, there is no difference in this respect between the master and his slave. They would
not seem to have been overworked, for many of them were able to purchase their liberty by
doing extra work” (Campaigning on the Oxus, and the Fall of Khiva, 300). Zalesov writes
that the purportedly mild treatment of slaves in Khwarazm was a purely pragmatic
strategy: “The Uzbeks’ treatment of their slaves, except in punishment for running away,
is generally quite gentle, all the more so since their owners, including the Khan himself,
understand quite well that, of all the disasters which could befall the Khanate, one of the
most horrible would be a general uprising of Persian slaves, of which there are here up to
10,000” (“Pis’mo iz khivy,” Voennyi sbornik 8 [1858], 286).
 The origins of Iranian Slavery in Central Asia 33

for those slaves who suffered minimal punishment. Finally, given that few
foreign observers spent much time in the presence of slaves, accounts
generalizing about slaves’ treatment may reveal more about the observers’
expectations and susceptibilities than they do about slaves’ ordeals.

THE ORIGINS OF IRANIAN SLAVERY IN CENTRAL ASIA

Iranian Shiʿites would constitute the majority of Central Asia’s enslaved

people for three centuries, and the origins of their enslavement can be
traced to particular developments in the early sixteenth century. This was
an age of bitter Uzbek–Safavid warfare, during which Sunni jurists in
Transoxania issued multiple fatwas that made licit not only the raiding
and plundering of Shiʿite communities but also the capturing of Shiʿite
individuals, and by extension their enslavement. Iskandar Beg, a Turkmen
historian employed at the court of the Safavid Shah ʿAbbās I, provides
a striking and rare account of the origins of Shiʿite enslavement in his
Tārı̄kh-i ʿālamārā-yi ʿAbbāsı̄. It is worth noting his emphasis on the reci-
procal nature of enslavement as a tactic of war:74
I should comment here on a point I have mentioned before, but which can bear
repetition. In earlier periods, when the Uzbeg rulers invaded Khurasan and the
Ottoman sultans invaded Azerbaijan, either with the object of annexing territory or
of plunder, they did not take captives from Shῑʿite lands, nor did qezelbas armies
commit this heinous crime in Sunni territory. However, in the reign of the Ottoman
Sultan Morad, an Ottoman and Tartar army invaded Azerbaijan and Sirvan and was
guilty of this practice. When the Ottomans occupied Tabriz, many children of
seyyeds, who were descendants of the Prophet himself, were carried off into
captivity and sold to Frankish infidels in Istanbul. The Uzbegs adopted this practice
during their invasions of Khurasan under ʿAbdollah Khan and his son ʿAbd al-
Mo’men Khan. For instance, at Mashhad, they took captive many descendants of
the Imam Reza, and many children of the nobility, of the ʿolama, of ascetics and
other honorable men, and of the military and civilians in general – several thousand
altogether. These captives were sold in Turkestan and Transoxania, and even as far
away as Kabul and India.
Because the heavens so decreed, the Shah was forced to overlook these crimes at
the time; this world is a vale of woe, and revenge for these heinous acts could safely
be left to the Lord of vengeance. At the urging of his commanders, however, the
Shah allowed several thousand prisoners to be taken on this campaign and exiled
from their homelands. Subsequently too, the Safavid governors of Astarabad on

74
Iskandar Beg Munshı̄, Tārı̄kh-i ʿālamārā-yi ʿAbbāsı̄, trans. Roger Savory as History of Shah
ʿAbbas the Great, Vol. 2 (Boulder, CO: Caravan Books, 1978), 819–820.
34 The Setting

several occasions led punitive expeditions against rebel groups of the Ūklū and
Göklen tribes, which were Muslim only in name, and many prisoners were taken
on these occasions. But if impartial critics will take a searching look at the Shah’s
actions in this regard, they will discover that he has earned the approval of the
religious authorities, because these prisoners were not taken into slavery but were
treated as prisoners of war. Several thousand women and children were brought up
in Shῑʿite and God-fearing homes, and adopted the Shῑʿite faith.

The name of the Sunni jurist who first made licit the enslavement of
Shiʿites, while absent from our chronicle sources, was preserved in
Turkmen oral historical tradition down to the nineteenth century: Both
Iranian and European observers report having met Turkmens who identi-
fied the main jurist responsible as one Shams al-Dı̄n Herātı̄.75 Considering
that captive-taking was a standard feature of warfare among Muslim
groups in the region during Herātı̄’s lifetime (in the sixteenth century),
this jurist’s formal legal licensing hardly seems necessary. Nevertheless, it
aligned with and legitimized the formalized discourse of sectarian struggle
that was employed by both Sunni and Shiʿite statesmen and jurists.
As for the age of conflict in which these legal developments took place,
a very brief overview of its major features must suffice here. As Shah Ismāʿı̄l
(r. 1501–1524) consolidated Iran as a Shiʿite domain, Khurasan – a region
divided from Transoxania during the partition of the Timurid empire –
became the focus of struggles between the Shah and the Sunni Uzbeks to
the north, which only increased after Ismāʿı̄l’s death. Uzbek armies led by
ʿUbaydullāh Khan made five major forays into Khurasan between 1524
and 1538, but were unable to retain control of any major towns in the
region other than Balkh. Significant Uzbek campaigns in Khurasan were
resumed with the conquest of Herat in 1589, led by ʿAbdullāh Khan II, and
subsequently Mashhad and much of the rest of Khurasan was captured by
that ruler’s son, ʿAbd al-Mu’min. Uzbek raids in this period proceeded
deep into Iranian territory, reaching as far south as Yazd, and relented only
with the death of both ʿAbdullāh Khan and ʿAbd al-Mu’min in 1598, after
which the Iranian ruler, Shah ʿAbbās, was able to retake most of Khurasan.
It was likely during this century of Sunni–Shiʿite conflict that we find the
initial proliferation of Shiʿite slaves in Central Asia.
The work performed by slaves in Central Asia during this period, as we
shall see, appears to have been primarily agricultural, and the new influx of
75
See Āshtiyānı̄, ʿIbratnāma, 84 (here, the name given is Shams al-Dı̄n Muhammad);
˙
Vambery, History of Bokhara from the Earliest Period Down to the Present (London:
Henry S. King, 1873), 266n1; Faiziev, Buxoro feodal jamiyatida qullardan foidalanishga
doir hujjatlar (XIX asr), 8.
 The origins of Iranian Slavery in Central Asia 35

slaves corresponds compellingly to a revolution in agriculture in the

region. Maria Subtelny has shown how large plantations had developed
during the late Timurid period thanks to a shrewd revenue-generating
strategy by a Central Asian ruler facing the prospect of fiscal disaster.
Caught up in the conflict between a largely Turkic military class covetous
of its traditional landholding privileges and an Iranian bureaucratic class
interested in imposing centralizing reforms, the last Timurid ruler, Sultan
Husayn Mı̄rzā (r. 1469–1506), was forced to find a new source of tax
˙
revenue that would alienate neither faction. He found it in the expansion
of hydrological agriculture, which he encouraged elites to develop both for
their own enrichment and the enrichment of the state treasury. At the same
time, he rapidly expanded the system of pious endowments (waqf) such
that Sufi shrine-complexes could flourish into massive agricultural
estates.76 It was through these developments that a Sufi master such as
Khwāja ʿUbaydullāh Ahrār could become one of the wealthiest land-
˙
holders in the region. The Khwāja’s private correspondences reveal that
he also owned slaves, who worked on his estates.77
We have reason to believe that the new agricultural “plantations” that
emerged in this period, such as those owned by Khwāja Ahrār and others,
˙
made extensive use of slave labor. R. K. Mukminova’s study of sixteenth-
century waqf documents has revealed that most of the slaves mentioned in
these pious endowments were used for agricultural labor or animal
husbandry.78 Other sources reveal large numbers of slaves working agri-
cultural jobs on the estates of prominent Sufi leaders such as the Juybāri
shaykhs.79 Thanks to the influx of captives from conflicts further south,
such landholders would hardly have needed to worry about labor scarcity
in the hinterlands occupied by their expanding estates.
In the seventeenth century, most slave laborers in Central Asia would
likely have been drawn from the pool of captives seized in more limited
campaigns into Khurasan. The series of conquests by Uzbek rulers

76
Subtelny, Timurids in Transition: Turko-Persian Politics and Acculturation in Medieval Iran
(Leiden: Brill, 2007).
77
See The Letters of Khwāja ʿUbayd Allāh Ahrār and his Associates, ed. Jo-Ann Gross and
˙
Asom Urunbaev (Leiden: Brill. 2002), 19, 25, 97 (see especially 97n9. The slaves men-
tioned here are of Indian descent, and Scott Levi has argued that the slave demographic in
Central Asia remained predominantly Indian throughout the medieval period: “Hindus
Beyond the Hindu Kush: Indians in the Central Asian Slave Trade,” Journal of the Royal
Asiatic Society 12:3 [2002], 277–288.).
78
Mukminova, Sotsial’naia differentsiatsiia naseleniia gorodov Uzbekistana v XV–XVI vv.
(Tashkent, 1985), 122–123.
79
Levi, “Hindus Beyond the Hindu Kush,” 278.
36 The Setting

ʿAbdullāh Khan and ʿAbd al-Mu’min that concluded the sixteenth century
marked the last time for at least another two hundred years that forces
from Transoxania would endeavor to permanently occupy Khurasanian
towns. Smaller-scale military forays into the region, however, resumed just
two decades later under Imām Qulı̄ Khan (r. 1611–1642), and they inten-
sified under ʿAbd al-ʿAzı̄z Sultān (r. 1645–1680) after Shah ʿAbbās’ death,
˙
though ʿAbd al-ʿAzı̄z seems to have tempered his raiding once he became
khan.80
During these formative centuries in the development of the Shiʿite slave
economy – from which, unfortunately, sources on slavery in the region are
scarce – it would probably have been Uzbek soldiers, rather than nomadic
Turkmens, who engaged in most of the captive-taking. There is, moreover,
no evidence that raiding of Khurasanian villages for captives was at this
time the sort of near-constant phenomenon that it would later become.
The eighteenth century witnessed the increasing influence of Turkmen
tribes in the affairs of Khwarazm and Bukhara, and also their migration
closer to what would become the main routes of the slave trade: The Tekes
migrated to northern Khurasan, while a large branch of the Yomuts, along
with the Chawdurs, moved into Khwarazm. Meanwhile, the khanates
were embroiled in ongoing political crises. In Khwarazm, the years
1685–1715 saw the succession of as many as thirteen khans, many of
whom were installed by nomadic Uzbeks, sometimes in alliance with
Kazakhs or Turkmens. The latter began at this time to factor more promi-
nently in the khans’ raiding forays into Khurasan. Slave-trading in the
Bukharan khanate, meanwhile, was likely at a low ebb due to internal
turmoil: Kazakhs – fleeing the steppe before the advancing Kalmyks –
raided Bukharan towns and villages, leaving Samarqand and all but two
districts of Bukhara itself depopulated by the time they returned to the
steppe in the late 1720s. The regions of Ferghana, Hisar, and Shahrisabz
were all more-or-less independent of Bukhara at this time, and the influ-
ence of the Bukharan ruler – nominally the khan, but in reality his ataliq –
was confined to the capital and its adjacent provinces.81

80
On the political history of this period and the conflicts in question, see Yuri Bregel,
“Bukhara iv. Khanate of Bukhara and Khurasan,” Encyclopedia Iranica, Vol. 4, fasc. 5,
521–524; Martin Dickson, “Shāh Tahmāsb and the Uzbeks: The Duel for Khurāsān with
ʿUbayd Khān: 930-946/1524-1540” (Ph.D. dissertation, Princeton University, 1958);
Iskandar Beg Munshı̄, Tārı̄kh-e ʿĀlamārā-yi ʿAbbāsı̄.
81
Yuri Bregel, “The New Uzbek States: Bukhara Khiva and Qoqand: c. 1750–1886,” in
Di Cosmo et al., eds., The Cambridge History of Inner Asia: The Chingghisid Age
(Cambridge: Cambridge University Press, 2009), 392–393.
 The origins of Iranian Slavery in Central Asia 37

In 1740, Nādir Shah of Iran launched an ambitious raid into the region,
winning submission (as well as the services of a 10,000-strong cavalry
force) from the Bukharan khan before marching on Khiva. The
Khwarazmian ruler, Ilbars Khan, surrendered and was promptly executed,
and Khiva suffered an intense assault. The city was surrendered after just
a few days. In the aftermath of these attacks, the Shah reportedly liberated
several thousand Iranian slaves as well as a smaller number of Russian
slaves. The former were sent to Khurasan, and the latter were turned over
to the Russian officers Gladyshev and Muravin, who conducted them back
to Russia. The Shah, who has since often been depicted as a liberator of
captives, also took some corvée laborers for himself: somewhere between
4,000 and 18,000 Khwarazmian cavalrymen were pressed into service in
the Iranian military.82 The presence of so many Iranian slaves in the
khanate indicates that the slave system that would become well-known
to foreign observers in the nineteenth century, by which Iranians from
Khurasan were pressed into bondage on Khwarazmian and Bukharan
estates, may already have been firmly in place by the mid-eighteenth
century.
For the remainder of the eighteenth century, Iran would launch no
major campaigns into the khanates, even as raids on Khurasanian towns,
often carried out by Turkmens, continued to supply the Khwarazmian and
Bukharan markets with Iranian slaves. The first few decades of the
nineteenth century saw a renewed effort by Iran – now led by the dynasty
of the Qajars – to secure Khurasan against these incursions, and to estab-
lish a military vanguard against the khanates and the nomads who some-
times served as the khans’ allies.
In the 1820s and 1830s, Iranian troops made considerable gains in
reducing the trade in Qajar slaves. In 1831–1832, acting as the Qajar
ruler Fath ‘Alı̄ Khan ’s governor of Khurasan, the crown prince ‘Abbās
Mirzā led successful campaigns against defiant chieftains in Quchan,
Amirabad, Turshiz, and Turbat, and he inspired the formerly hostile Salar
Turkmens of Herat to make a plea for peace. In 1832, he launched
a successful attack on Sarakhs, sacking the Salar stronghold and liberating
hundreds of Iranian captives. After his death in 1833, however, Qajar
fortunes in Khurasan quickly and decisively turned. From their base at
Merv, the Khwarazmians patronized Turkmens who patrolled and plun-
dered the border region, taking captives when they could; their spoils also
included heavy-laden caravans bound for Iran. An Iranian campaign in

82
Bregel, “The New Uzbek States,” 393.
38 The Setting

1837 to take Herat was a disaster, due in part to the ability of the Merv-
based Khwarazmian forces to interrupt Iranian supply lines at will.
In 1841, Allāh Qulı̄ Khan, ruler of Khwarazm, ordered a direct offensive
against Iran which was evidently a success; among the results was the
deportation of some 15,000 Jamshidi tribesmen from Badghis to Merv.
In the years to come, Khwarazm would lose its foothold in Merv, as many
of the Sarı̈q Turkmens of the oasis threw off and continually rejected their
former state of subjection to the khan. Unwilling to abandon Merv com-
pletely, however, Khwarazmian rulers would spend the next decade launch-
ing annual assaults on the Sarı̈qs, draining both their own resources and
those of their entrenched Turkmen adversaries. Khwarazm was sometimes
supported in these efforts by confederations of Turkmens, which included
members of the Yomut, Chawdur, Yemreli, Qaradashli, and Taze Qongrat
tribes. Meanwhile, the Khwarazmians and their Turkmen clients continued
to plunder caravans bound for Iran, and in 1851 the khan’s troops were
among the forces that repulsed Iranian efforts to take Sarakhs.83 In short,
the constant presence of Khwarazmian troops and mercenaries in Khurasan
was a major obstacle against Iranian efforts to exert control over the region,
and the lack of control – as well as the continual armed conflict fomented
both by the Qajars and their opponents – allowed for new victims to be
channeled constantly into the Central Asian slave trade.

QAJAR MISSIONS TO END THE SLAVE TRADE

Qajar military efforts in Khurasan during the nineteenth century were

inspired not only by a desire to end the plague of captive-taking, but also
by the Iranian government’s consistent ambition to expand its control
northward.84 This fact may come as a surprise, since, for decades, histor-
ians have presented this era of Central Asian history as a contest between
England and Russia, in which local powers like Iran, Bukhara, and
Khwarazm were alternately pawns, victims, or bystanders.85 This Qajar
project was played out on both military and diplomatic fronts, and gave
rise to a genre of travel literature – combining intelligence-gathering,

83
William A. Wood, “The Sariq Turkmens of Merv and the Khanate of Khiva in the Early
Nineteenth Century” (Ph.D. dissertation, Indiana University, 1998), 137–226.
84
James M. Gustafson. “Qajar Ambitions in the Great Game: Notes on the Embassy of
‘Abbas Qoli Khan to the Amı̄r of Bokhara, 1844.” Iranian Studies 46:4 (2013), 536, 551.
85
Alexander Morrison, “Killing the Cotton Canard and getting rid of the Great Game:
rewriting the Russian conquest of Central Asia, 1814–1895,” Central Asian Survey 33:2
(2014), 131–142.
 Qajar Missions to End the Slave Trade 39

ethnography, geography, political commentary, and history – which

thrived especially from the 1840s through the1870s.86 Some of the most
important and well-known works within this genre arose from missions
that were tasked with freeing Iranian slaves from Khwarazm and Bukhara
and ending the slave trade.
The two best-known missions tasked with ending the trade in Iranian
slaves are those of ‘Abbās Qulı̄ Khan and Riżā Qulı̄ Khan Hidāyat.87 ‘Abbās
Qulı̄ Khan undertook an embassy to the Bukharan ruler Nasrullāh Khan
˙
between May and August of 1844, apparently for the purpose of asserting
Qajar dominion over Merv, for seeking Bukharan cooperation in ending
Iranian slavery in the khanate, and – at the behest of the British – for
retrieving the missionary and adventurer Joseph Wolff from his temporary
imprisonment. The envoy successfully stakes his rhetorical claim to Merv
and retrieves the ailing Wolff, but when it came to general emancipation
‘Abbās Qulı̄ Khan’s efforts came up short. The Bukharan ruler denies
responsibility for his subjects’ enslaving of Iranians, claiming that the
blame rests exclusively upon the ʿulamāʾ who continued to countenance
the practice. Therefore, the only solution – according to the Nasrullāh – is
˙
for the ʿulamāʾ of Iran and Bukhara to discuss the matter together and
arrive at some kind of joint resolution. Even then, the ruler cautions, old
habits die hard, and it would be impossible to reverse this unsavory custom
right away.88
Riżā Qulı̄ Khan Hidāyat traveled to Khwarazm in 1851 to meet with the
ruler, Muhammad Amı̄n Khan (r. 1845–55), and his bid to free Iranian
˙
slaves in Khiva would be no more successful than that of his counterpart,
‘Abbās Qulı̄ Khan, in Bukhara. Before setting off on his trip, Riżā Qulı̄

86
Arash Khazeni. “Across the Black Sands and the Red: Travel Writing, Nature, and the
Reclamation of the Eurasian Steppe circa 1850,” International Journal of Middle East
Studies 42 (2010), 594–595; Abbas Amanat and Arash Khazeni, “The Steppe Roads of
Central Asia and the Persian Captivity Narrative of Mirza Mahmud Taqi Ashtiyani,” in
Nile Green, ed., Writing Travel in Central Asian History (Bloomington, IN: Indiana
University Press, 2013), 113–134; Gustafson. “Qajar Ambitions in the Great Game,” 541.
87
‘Abbās Quli Khān, Safarnāma-yi Bukhārā, ed. Hussain Zamani (Tehran, 1373/1995); Rizā
Quli Khān Hidāyat. Sifāratnāma-yi Khwārazm (Relation de l’Ambassade au Kharezm
[Khiva] de Riza Qouly Khan. Texte Persan), ed. Charles Schefer (Paris: Ernest Leroux,
1876). Both works have been the subject of insightful recent studies: Nölle-Karimi.
“‘Different in All Respects’: Bukhara and Khiva as Viewed by Kāǧār Envoys,” in Yavuz
˙
Köse, ed., Şehrâyı̂n: die Welt der Osmanen, die Osmanen in der Welt: Wahrnehmungen,
Begegnungen und Abgrenzungen; Festschrift Hans Georg Majer (Wiesbaden: Harrassowitz,
2012), 435–446; Khazeni. “Across the Black Sands and the Red”; Gustafson, “Qajar
Ambitions in the Great Game.”
88
‘Abbās Quli Khān, Safarnāma-yi Bukhārā, 36–38.
40 The Setting

Khan had received the names of a number of men and women from
Astarabad and elsewhere who were held as slaves in Khwarazm. Some
individuals whose parents were among the enslaved had begged Riżā Qulı̄
Khan for permission to accompany him on the journey.89 The envoy
recounts meeting with a great number of Iranian slaves over the course
of his travels in the khanate. He hosts them at his residence, serves them
tea, and learns of their backgrounds and circumstances. Some had been
enslaved for more than fifty years, and some for less than three.90 At one
point he overhears a cacophony of cries and moans from two of the slaves
coming to visit him: It turns out that they are cousins who had not had any
news of one-another for some time before meeting by coincidence at Riżā
Qulı̄ Khan’s residence and realizing that they shared the same awful fate.91
Riżā Qulı̄ Khan collects many slaves’ names, and he thrills them with the
news that he will request their freedom from the khan. Slaves of diverse
backgrounds gather to follow him through the streets. Their excitement is
such that Riżā Qulı̄ Khan nearly expected the start of a slave uprising as he
and his retinue move through the city.92
But all of the excitement is in vain. He is unable to free the slaves.
The Khwarazmian ruler, Muhammad Amı̄n Khan, after consulting with
˙
some of his top officials, declines the envoy’s emancipation request on the
grounds that it would embolden the “qizilbāsh” (Iranians), reinforcing
among them the notion that the concession was a result of their troop-
movements around Astarabad and in Khurasan. The result, the khan says,
would be an onslaught of new demands on the part of the Qajars.93 Simply
releasing thousands of Iranian slaves at the Qajars’ behest was out of the
question.
Along with the two well-known missions discussed previously, there
was at least one other Qajar embassy to free slaves in this period – one that
appears to be largely forgotten. This is the mission of Muhammad
˙
ʿAlı̄ Khan Ghafūr, who was dispatched to Khiva in 1842.94 His primary
aim was to negotiate for the release of the lieutenant and nephew of

89
Rizā Quli Khān Hidāyat, Sifāratnāma-yi Khwārazm, 52.
90
Hidāyat, Sifāratnāma-yi Khwārazm, 85–86.
91
Hidāyat, Sifāratnāma-yi Khwārazm, 86.
92
Hidāyat, Sifāratnāma-yi Khwārazm, 86–87.
93
Hidāyat, Sifāratnāma-yi Khwārazm, 111.
94
Muhammad ʿAlı̄ Khan Ghafūr, Ruznāmah-i safar-i Khwārazm (Tehran: Daftar-i
˙
mutāla‘āt-i siyāsı̄ va bayn al-milalῑ, vizārat-i umūr-i khārija, 1373/1994). An important
˙
recent article assesses the motivations of this embassy: Christine Nölle-Karimi, “On the
Edge: Eastern Khurasan in the Perception of Qajar Officials,” Eurasian Studies 14 (2016),
135–177.
 Qajar Missions to End the Slave Trade 41

Mashhad’s Qajar governor, who had been captured by Turkmens while

hunting and brought to the Khwarazmian capital. Muhammad ʿAlı̄ Khan
˙
was also tasked with making a more ambitious request – the same one that
Riżā Qulı̄ Khan had made: that the slave trade in Khwarazm be ended, and
that the Iranian slaves there be freed.
Muhammad ʿAlı̄ Khan arrives in Khiva at the same time as embassies
˙
from Russia and England, whose ambassadors he regards as amenable to
his cause in liberating the slaves. He has good reason to assume this;
indeed, Russian accounts claim that Muhammad ʿAlı̄ Khan had been sent
˙
to Khwarazm at Russia’s behest, alleging that the shah, having petitioned
the Russians to intercede with the khan for the liberation of Iranian slaves,
had begun preparing for war, “and only the combined efforts of Russian
and British ambassadors in Tehran convinced him to postpone military
action and to send to Khiva, as an envoy, Muhammad ʿAlı̄ Khan, offering
˙
him [also] the services of an official of the British mission [named]
Thomson.”95 As we shall see, the Iranian envoy would leverage these
alliances to little effect.
At his first meeting with the khan in Khiva, Muhammad ʿAlı̄ Khan
˙
assures the ruler that ending the slave trade and emancipating Iranian
slaves would “bring friendship and brotherhood” between Iran and
Khwarazm. On the other hand, he threatens, if the khan declined, he
would find that “nothing comes of enmity but the ruin of a country and
the trampling of its people.”96 The envoy alludes to previous Iranian
missions to free captives that had been unsuccessful, and alleges that the
khan had broken promises to stop Turkmens under his sway from taking
Iranian captives. In response, the khan alleges transgressions by Iranian
governors and elites, but at first he appears to agree to take a firmer hand
with the offending Turkmens. As an underwhelming sign of goodwill, he
consents to release fifteen slaves. The Iranian envoy replies, “O glorious
khan, it is not a matter of ten slaves or a hundred. The Shah of Iran wants
all of his people [returned], and his wishes must be fulfilled.”97 The envoy
then meets with the khan’s mihtar and outlines his demands in more detail:
He asks for nothing less than the liberation of all the slaves that had been
taken since the time of Fath ‘Alı̄ Khan (d. 1838).
What follows this initial meeting with the khan is a series of negotia-
tions with both the mihtar and the ruler that are practically comical in their
95
N. Zalesov, “Posol’stvo v Khivu podpolkovnika Danilevskogo v 1842 g,” Voennyi sbornik
3 (1866), 46.
96
Ghafūr, Ruznāmah-i safar-i Khwārazm, 19.
97
Ghafūr, Ruznāmah-i safar-i Khwārazm, 20.
42 The Setting

elaborate futility. First, the mihtar suggests that only those slaves taken
since the relatively recent conclusion of peace between Khwarazm and Iran
should be freed, but that Muhammad ʿAlı̄ Khan would be welcome to buy
˙
the rest of the slaves, just as the Khwarazmians had once ransomed their
own people from the Qajars. Muhammad ʿAlı̄ Khan replies that this would
˙
not suffice, and stresses again that all the slaves must be freed. He asks the
mihtar how many, then, he was willing to offer. The mihtar replies with
a story that, if true, reveals something of the high-handed manner in which
the Qajars had formerly dealt with the khanate: Some time previously, the
khan had sent a Khivan ambassador to Tehran along with some Iranian
slaves who had been freed as a gesture of goodwill. In exchange, the mihtar
claims, the ambassador asked for the liberation of two or three Sarakhsi
slaves that were held in Iran, but was told that this was impossible, since
they had become close to the Shah’s mother, and that respecting one’s
mother was as necessary as respecting the holy kaʿba in Mecca.98 Thus, the
mihtar explains, the last time Khwarazm had offered concessions on the
issue of slavery, their goodwill had not been returned; instead, the khan
had been insulted.
Nevertheless, the mihtar agrees to bring Muhammad ʿAlı̄ Khan’s
˙
entreaty before the khan. Because the khan was busy with some other
pressing issues, however, Muhammad ʿAlı̄ Khan would in the meantime
˙
have to carry out further negotiations with the mihtar. At their next meet-
ing, Muhammad ʿAlı̄ Khan and the mihtar carry out their discussions in the
˙
presence of the imprisoned notable that the Iranian envoy had come to
liberate – the nephew of Mashhad’s governor. This time, Muhammad
˙
ʿAlı̄ Khan evokes the specter of Russian and British disapproval, warning
the mihtar that “three countries [i.e. Iran, England, and Russia] that are
friends of the Khan ask that the slaves of Iran be given leave to depart, and
there are 30,000 slaves in Khwarazm.” Of these, he asks, how many would
the khan be prepared to emancipate “so that the Pādishah of Islam [i.e. the
Shah of Iran] will be satisfied” with him? “The slaves,” the mihtar replies,
“are in the hands of the citizens and the populace. Each was bought for 30
tillas, for 40 tillas. It is not our custom to force or harass our people.” He
˙ ˙
reiterates that Khwarazm had previously been forced to buy its citizens
back from Iran, and that the Iranians were welcome to do the same. This
offer does not appease Muhammad ʿAlı̄ Khan: “The Shah,” he replies,
˙
“does not give money to buy back his own slaves.”99

98
Ghafūr, Ruznāmah-i safar-i Khwārazm, 22.
99
Ghafūr, Ruznāmah-i safar-i Khwārazm, 23.
 Qajar Missions to End the Slave Trade 43

Muhammad ʿAlı̄ Khan’s claim to negotiate on behalf of a troika of

˙
great empires was only partly accurate. While the Russians and
British may have been sincere in their hopes of averting warfare
between Iran and Khwarazm, they had little reason to be optimistic
about the emancipation of tens of thousands of Iranian slaves, espe-
cially considering that a previous Russian embassy, less than two
years prior, had made no headway on that front – which is no
surprise, considering the disastrous results of the military expedition
recently launched against the khanate. The ambassador at that time,
Nikoforov, had been coached on his mission in a revealing debriefing
from March 19, 1841:100
In addition to the instructions which you received directly from the Asian
Department, the Ministry of Foreign Affairs entrusts you with another mission,
of no small importance. Though it does not concern our own relations with Khiva,
successful execution of [the mission] may nevertheless have a beneficial influence
on the course of our affairs in Asia generally . . . [R]ecently the Persian government
appealed to us with the earnest request that we render our assistance in Khiva
toward the liberation of Persian subjects there, adding that if for some reason the
Russian state finds it inconvenient to take part in this matter, then in such a case the
Persians will need to resolve upon their last resort, namely the achievement of their
aim by force of arms.
We have no doubt that the fulfillment of the Persian Court’s request will meet
strong resistance in Khiva, especially since Persians are the most numerous class of
people being used as slaves. Nevertheless, we consider it necessary to make you
aware that the liberation through your mediation of even some number of Persian
captives would be extremely good for us. For that reason, do not fail to take
advantage of any opportune moment to try to achieve that goal. Explain to [the
khan] Allah Kuli what the consequences could be for him with respect to Persia if he
continues to follow the same hostile system in his relations with her and, on the
other hand, what he would gain in personal esteem if he takes as his principal
neighborliness (dobroe sosedstvo) and justice. And while the complete release of the
Persian captives cannot be hoped for, it may be possible for you to at least persuade
the khan to send an envoy to Tehran with the proper authorization to work with the
Persian Court on mutually-agreed terms, on the basis of which the liberation of
Persians from Khiva may be enacted . . .
If, however, the demands of the Khivans should be entirely unmanageable, and
your advice concerning abolition be in vain, then in that case the Khivans can by no
means hope for the assistance of our mission in Tehran, and will have to come to an
agreement with the Persians at their own discretion.

100
Sbornik materialov dlia istorii zavoevaniia Turkestanskogo kraia, ed. A. G. Serebrennikov,
Vol. 3 (Tashkent, 1912), doc. no. 24, pp. 41–44.
44 The Setting

On the basis of this quote, one might conclude that the issue of Iranian
slaves was of great significance, a crucial corollary to the primary goals of
Nikoforov’s mission. But following this, the memorandum delivers
Nikoforov a strong word of caution in raising the subject with the khan,
urging him to remember that emancipation was not, in fact, one of the
mission’s foremost concerns, but merely a supplementary issue to be
discussed after other matters had been concluded, and then only if
a convenient opportunity presented itself:101
It goes without saying that the matter of the Persians should not harm the arrange-
ment of our own affairs with Khiva. For that reason, do not initiate negotiations on
this subject before fulfilling the other political errands with which you have been
entrusted, and in any case only after being well received, because on top of an
inauspicious reception this would serve only to excite greater antipathy toward you
on the part of Allah Kuli. Consequently, it would be completely contrary to the
purposes for which you were sent to Khiva.

As it happens, Nikoforov did not find a suitable moment to broach the

subject with the khan. He later concisely summarized the matter in a letter
to his comrade, Captain Khanykov: “Concerning negotiations on the
Persian captives . . . The agent did not initiate [these negotiations] for the
following reasons: 1) Up to now, there is nothing positive about our
relationship with Khiva . . . ”102
Danilevskii, the Russian ambassador who joined Muhammad ʿAlı̄ Khan
˙
in Khiva the very next year, had no more reason to expect success on the
emancipation front. Still, Danilevskii was commissioned to “clarify” to the
khan that permitting the enslavement of Iranians in his domain was not
worthwhile.103 As for Danilevskii’s British counterpart, a man named
William Taylour Thomson, hardly any information exists in English
sources concerning his experiences,104 but the mission debriefing provided

101
Sbornik materialov dlia istorii zavoevaniia Turkestanskogo kraia, doc. no. 24, p. 44.
102
“Do sego vremeni net nichego polozhitel’nogo otnositel’no otnoshenii k Khive.” Sbornik
materialov dlia istorii zavoevaniia Turkestanskogo kraia, doc. no. 58, p. 111.
103
Zalesov, “Posol’stvo v Khivu podpolkovnika Danilevskogo v 1842 g,” 45.
104
Strangely, the only publication featuring any substantial information on Thomson’s
mission appears to be Lady Sheil’s memoir of her time in Iran, which includes extracts
from Thomson’s diary in an appendix. These extracts contain no useful information
concerning the emancipation of slaves, however, nor any useful details concerning the
content of Thomson’s negotiations with the khan: Lady Sheil, Glimpses of Life and
Manners in Persia (London: 1856), 358–370. Thomson observes, in any case, that
“The number of Persian slaves imported and also bred in the country [of Khiva] is
immense, and in almost every house where servants are kept, one or more, according to
the means of the proprietor, are to be found” (Lady Sheil, Glimpses of Life and Manners in
 Qajar Missions to End the Slave Trade 45

for Danilevskii at least describes his position vis-à-vis the other ambassa-
dors. Thomson, “according to the instructions given to him,” was charged
“exclusively with arranging matters relating to Persian slaves,” and “has
been given the fundamental duty to consult with the Russian agent in
Khiva on this matter and act together with him.” He was, moreover,
“positively forbidden from intervening in our agent’s negotiations on
matters relating to Russia,”105 and it appears that Danilevskii did indeed
use the occasion of his visit to advance a number of Russian positions,
including the liberation of Russian slaves, who were far fewer in
number.106
After raising the dubious specter of Russian and British anger over the
issue, and declining the opportunity to buy Iranian slaves at their market
price, Muhammad ʿAlı̄ Khan finally wrings a small concession from the
˙
mihtar: the latter agrees to release some captives, albeit only a small
number of them. “If you want ten or fifteen people,” the mihtar tells
him, “I will give them over so that you will not go away empty-handed.”
Muhammad ʿAlı̄ Khan balks at this, and demands 5,000 slaves. This back-
˙
and-forth continues for three or four hours, and Muhammad ʿAlı̄ Khan
˙
eventually reduces his demand from 5,000 to 2,000 slaves, but this does
not satisfy the mihtar. The imprisoned Khurasanian notable, Muhammad
˙
Walῑ Khan, who was present as an intermediary, pleads with the Iranian
envoy to reduce his demand still further, to 1,000. Muhammad ʿAlı̄ Khan
˙
comes down to 1,000, and the mihtar agrees to bring the proposition
before the khan.
After this, the khan summons Muhammad ʿAlı̄ Khan before him once
˙
again, this time for a meeting which would prove to be, in the envoy’s
retelling, far tenser than the first.107 After exchanging pleasantries, the
khan asks the envoy what he had been discussing with the mihtar. “I was
explaining to him my mission,” Muhammad ʿAlı̄ Khan replies. “The
˙
circumstances of its resolution are for you to decide.” The khan extends
his initial offer: He is willing to give over those slaves who had already
been manumitted, but any others would have to be purchased. Just as he
had in his negotiations with the mihtar, Muhammad ʿAlı̄ Khan objects that
˙
the Shah does not buy slaves. The khan then makes a still less generous
proposition: “The viziers of Russia and England wrote to me that the Shah

Persia, 365). Cf. also Wood, “The Sariq Turkmens of Merv and the Khanate of Khiva in
the Early Nineteenth Century,” 41.
105
Zalesov, “Posol’stvo v Khivu podpolkovnika Danilevskogo v 1842 g,” 45.
106
Zalesov, “Posol’stvo v Khivu podpolkovnika Danilevskogo v 1842 g,” 49.
107
Ghafūr, Ruznāmah-i safar-i Khwārazm, 24.
46 The Setting

wanted to come [and attack Khwarazm], but that they did not permit it . . .
If the Shah comes, we will be here [waiting for him]. Whatever God wishes
will come to pass.”108
In Muhammad ʿAlı̄ Khan’s telling, he meets this threat with one of his
˙
own: “God forbid the retinue of the Shah of Islam should venture hither.
In such a case, neither your country nor your people would remain.”109
The khan scoffs at this: “You think you can frighten me?” “I am not trying
to scare you, Your Benevolence,” the envoy replies. “I came to converse
with you.” With negotiations thus deteriorating, the khan finds an oppor-
tunity to stall: He asks Muhammad ʿAlı̄ Khan to contact the Shah and ask
˙
him to convey precisely how many slaves he would like the Khan to
surrender. Muhammad ʿAlı̄ Khan objects that it could take two or three
˙
months to go back to Tehran and return with a courier.110
Thus the talks drag on, and finally they conclude in a manner thor-
oughly unsatisfying for the Iranians: The slaves were not to be freed.
Danlievskii, for his part, petitioned for the release of “even just 1,500”
Iranian slaves, “but unfortunately all of [Danilevskii’s] admonitions on this
subject to the khan were useless.” The one significant concession offered
by the khan was the release of Muhammad Wali Khan, the imprisoned
˙
nephew of the Mashhad governor. This, the Khan allegedly told the
Russian envoy, “would have in the eyes of the Persian government more
value than the liberation of 5,000 [other] people”111 – a dubious assump-
tion, of course.
The Qajars soon resumed their hostilities toward Khwarazm, and it is
worth considering whether Muhammad ʿAlı̄ Khan’s demands were merely
˙
a pretext to justify the impending conflicts. If, as the khan claimed, there
were in fact Khwarazmians held in Iran that the Shah had shown no inten-
tion of freeing, it is reasonable to sympathize with the khan’s reluctance to
put himself in a position of such blatant weakness at the negotiating table.
More importantly, however, the freeing of even 1,000 slaves from private
hands would have been no simple matter: If the khan managed it by force, he
would inspire the anger of the aggrieved slave-owners, and if he purchased
their freedom he could do so only at great expense. The objection, more-
over, that liberating slaves would be a sign of Khwarazmian weakness and
could inspire further Iranian demands – an argument articulated by the khan
to Riżā Qulı̄ Khan – does not seem far-fetched.
108
Ghafūr, Ruznāmah-i safar-i Khwārazm, 25.
109
Ghafūr, Ruznāmah-i safar-i Khwārazm, 25.
110
Ghafūr, Ruznāmah-i safar-i Khwārazm, 25.
111
Zalesov, “Posol’stvo v Khivu podpolkovnika Danilevskogo v 1842 g,” 52.
 Russian Interventions in Central Asian Slave Trade 47

All of this would have been well-understood by Muhammad ʿAlı̄ Khan

˙
and his Qajar patrons. They would also have known that Khwarazm’s
degree of influence over those Turkmens funneling victims into the slave
economy was highly variable. The khan’s troops may have been unable to
patrol the Turkmens of the southern deserts, and even policing the caravan
routes to Khwarazm would have been only a half-measure against the slave
trade; the markets of Bukhara would surely receive any slaves that the
Khwarazmians did not purchase. Finally, neither the shah nor the tsar
showed any inclination to offer the khan any significant compensation
for his efforts. In other words, there is good reason to suspect that the
failure of these negotiations was a foregone conclusion even for the
ambassadors that took part in them. The main accomplishment of
Muhammad ʿAlı̄ Khan’s mission, then, along with the rescue of the
˙
Khurasanian notable, was perhaps the very fact of his emancipation
demand and its legitimation by the British and Russian ambassadors.

RUSSIAN INTERVENTIONS IN THE CENTRAL ASIAN SLAVE TRADE

DOWN TO THE NINETEENTH CENTURY

The slave trade in Central Asia was not limited to the Iranian frontier.
Many Russian citizens and others were also caught up in the trade, and the
Tsars – no less than Qajar Shahs – made a significant effort to safeguard and
liberate their own subjects. This targeted effort was generally limited to the
liberation of Russian subjects alone, and should not, as we shall see, be
mistaken for an “abolitionist” campaign in the region.
Between the mid-sixteenth century, a period marked by Muscovy’s
eastward expansion, and the mid-nineteenth century, the risk that
Russian diplomats in Central Asia would be killed, taken captive, or sold
into slavery made such missions relatively rare. Those ambassadors who
did venture to the region usually had the liberation of Russian slaves high
on their state-mandated agendas, and scarcely an envoy reached Bukhara
or Khiva without broaching the subject of Russians held in captivity. In the
effort of liberating Russian slaves, these missions failed unanimously and
miserably. A brief overview will suffice to make the point that the Russian
Empire had hardly any impact on the slave trade or on the phenomenon of
slavery in Central Asia down to the nineteenth century.
In the mid-sixteenth century, an English explorer and merchant named
Anthony Jenkinson returned from a grueling voyage to Central Asia,
leaving us a personal account of what would remain the most ambitious
48 The Setting

Western mission to the region for decades to come. Jenkinson traveled as

a member of the Muscovy Company, a mercantile venture with both
English and Russian backing, and he has left us with the earliest account
of slaves being sold at a Central Asian bazaar. Surveying the wares on
display at a Bukharan market in 1558, Jenkinson observed that both Indian
and Persian merchants were buying slaves, though of their origin he
records only that they were “of divers Countreis.”112 He brought twenty-
five of these slaves back with him, all of them Russians.113 He also
described the awful and perilous landscape that his mission traversed on
the way to Central Asia. In the region around Kazan and Astrakhan, he
witnessed Tatars suffering from extreme poverty and plague, and Noghay
children being sold as slaves.114 The mission was later stalled by raiders
who searched the party for non-Muslims to take as captives, as well as
terrible storms, unwelcoming locals, and a hard passage through the
desert. There would scarcely be a Russian mission to Central Asia for
decades to come; there would not be another British mission to the region
for nearly two centuries.115
In the early seventeenth century, the Bukharan ruler Imām Qulı̄ Khan
attempted to make contact with the newly crowned Russian tsar, the first
of the Romanovs. After the tsar failed to receive the first Bukharan ambas-
sador, the Khan sent another, requesting the establishment of permanent
diplomatic relations. This time, he offered in exchange to return some
number of Russian slaves that had been purchased from Crimean and
Noghay traders who were residing in Bukhara. The tsar received this
ambassador, and promptly sent an ambassador of his own to collect the
slaves and to learn more about conditions in Bukhara. As Ron Sela
observes, this ambassador, Ivan Khokhlov, would have an experience
much like that of his predecessor Anthony Jenkinson. He visited both
Khiva and Bukhara, and reported that circumstances in the region were
extremely perilous for travelers. His party faced hostile nomads, grueling
travel conditions, extortion-minded officials, and poor hospitality.

112
Anthony Jenkinson, Early Voyages and Travels to Russia and Persia by Anthony Jenkinson
and other Englishmen, ed. E. Delmar Morgan and C. H. Coote, Vol. 1 (London: Hakluyt
Society, 1886), 88–89.
113
Jenkinson, Early Voyages and Travels to Russia and Persia, 95.
114
Jenkinson, Early Voyages and Travels to Russia and Persia, 57. Jenkinson arrived soon
after the Russian conquest of both khānates, during a time when Kazan was racked with
rebellions.
115
Ron Sela, “Prescribing the Boundaries of Knowledge: Seventeenth-Century Russian
Diplomatic Missions to Central Asia,” in Nile Green, ed., Writing Travel in Central
Asian History (Bloomington, IN: Indiana University Press, 2014), 71–73.
 Russian Interventions in Central Asian Slave Trade 49

It would be another two decades before a Russian diplomat set foot once
more in the region, though the Bukharans maintained consistent diplo-
matic contact in the meantime.116
In the 1640s, another Russian ambassador – Anisim Gribov – was
dispatched to Central Asia with the mission of freeing and retrieving
Russian slaves, visiting both Khwarazm and Bukhara. He managed to
ransom only three slaves. Forty others came before him in order to beg
his intervention, but he had no choice but to disappoint them.117 In the
years 1669–1673, an embassy led by two brothers – Boris and Semen
Pazukhin – was dispatched to Bukhara and Khwarazm, as well as to
Balkh, and again the task of ransoming Russian slaves was high on the
list of priorities. In Bukhara, the ambassadors were received warmly and
treated to lavish banquets and entertainments, but negotiations for the
release of captives were scarcely more effective than in previous missions.
Upon demanding the release of all Russian Christians who had been
enslaved, the Bukharan khan countered that there were no Christian slaves
in Bukhara: All had converted to Islam. Finally, the khan allowed for the
free release of nine slaves and offered that the Pazukhins could purchase
another twenty-two. Discouraged, the ambassadors nevertheless managed
to make contact with some of the Russian slaves in the khanate and learn
valuable information about the workings of the slave trade. They learned
that the Bukharans purchased many of their Russian slaves from Khiva;
that they were often sold by Kalmyks or Bashkirs; and that some slaves
were sold even to India by Persians or Kalmyks. The Pazukhins’ report
estimated that 150 Russians were kept in the city of Bukhara, 100 in Balkh,
and 50 in the khan’s palace complex in Khiva; many more were assumed to
be residing in the suburbs and hinterlands.118
The reign of Peter the Great witnessed the most disastrous Russian
mission to date: the infamous Bekovich-Cherkasskii expedition of 1717,
which was dispatched to Central Asia in order, among other things, to gain
the Khwarazmian khan’s submission and to establish possible trade routes
with India. Arriving with a considerable military retinue, the Russian
officer Alexander Bekovich-Cherkasskii reportedly found himself under
siege some 120 kilometers from Khiva, facing a Khwarazmian army that
some of the expedition’s survivors estimated to be more than 20,000

116
Sela, “Prescribing the Boundaries of Knowledge,” 77–79.
117
Sela, “Prescribing the Boundaries of Knowledge,” 79–80. Gribov resolved to return on
a second mission just a few years later, but decided not to travel any further than the
Iranian frontier once he realized how dangerous conditions in Central Asia had become.
118
Sela, “Prescribing the Boundaries of Knowledge,” 83.
50 The Setting

strong. The expedition survived the attack, and Bekovich-Cherkasskii met

with the khan, who welcomed him peaceably and offered to accommodate
his retinue in five different encampments. Once divided, the expedition
was again attacked, and the majority of Bekovich-Cherkasskii’s party
was reportedly killed or sold into slavery. Future Russian ambassadors to
Khiva would have to contemplate the rumor that Bekovich-Cherkasskii
himself had been flayed alive, his skin made into the head of a ceremonial
drum.119
The remaining decades of the eighteenth century saw relatively few
Russian diplomatic missions to Central Asia, and as a result little in the
way of new information emerged for Russian observers concerning slavery
in the region’s southern expanses. It was known, however, that some
number of Russians was continually being captured, particularly Caspian
fishermen and merchants, the latter sometimes taken in caravan raids.
In 1819–1820, a Russian officer and envoy named N. N. Murav’ev
reported learning of 3,000 Russians in Khiva alone, though he – like his
predecessors – was unable to free them.120 In 1833, Russian foreign
minister Karl Nesselrode corresponded with the chief of the Orenburg
Border Commission, General G. F. Gens, concerning the possibility of
launching a military mission to liberate these Russian slaves, but the
minister expressed skepticism over the feasibility of such a mission. That
same year, a well-connected general named V. A. Perovskii, former aide to
Tsar Nicholas I, was appointed Orenburg’s military governor, and soon
requested that a list be drawn up of all the Russians known to be held
captive in Khiva. The list was some 599 names long, and Perovskii wrote to
St. Petersburg to advocate for a military expedition against the khanate.121
Allāh Qulı̄ Khan, the ruler of Khwarazm, was not insensible to the
growing Russian anger over the question of Russians held in the khanate.
In 1836, after some 500 Khwarazmian merchants were detained in
Orenburg and Astrakhan along with their goods, the khan freed eighty
Russian captives, having freed another twenty-five that same year.122
The remaining Russians, however many there were, would be fewer in

119
N. N. Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 1819–20,
trans. Philipp Strahl (Calcutta: Foreign Department Press, 1871), 136.
120
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 77. More on this
below.
121
Alexander Morrison, “Twin Imperial Disasters: The Invasions of Khiva and Afghanistan
in the Russian and British official mind, 1839–1842,” Modern Asian Studies 48:1 (2014),
283–285.
122
V. I. Dal’, “Pis’ma k druz’iam iz pokhoda v khivu,” Russkii arkhiv 3 (1867), repr.
“Khivinskii pokhod,” Gostynnyi dvor 2 (1995), 168.
 Russian Interventions in Central Asian Slave Trade 51

number than when Murav’ev visited the khanate: According to

Zakhar’in, their numbers had been halved during a cholera epidemic in
1829.123 Notwithstanding the Russians given over by the khan, Russian
authorities found other reasons to proceed with Perovskii’s plan, and in
any event the number of captives freed was deemed insufficient.
Intercession by British envoys likewise came to naught, not because the
khan was unwilling to release Russians as a concession to the tsar, but
rather because these concessions, along with the khan’s repeated efforts
to negotiate and come to some agreement short of violence, were
regarded with suspicion or outright disdain by Perovskii and other
Russian officials. Claiming that there were 3,000 Russians held in
Khwarazm – a number he may have derived from Murav’ev’s travelogue,
then nearly twenty years old – Perovskii demanded that this same number
of Russians be freed.124 Plans for the military expedition to Khiva would
proceed.
The expedition, which set out in the winter of 1839–1840, was
a spectacular disaster. Blindsided by an especially harsh desert winter,
Perovskii’s force of 4,000 to 5,000 men and as many as 10,000 camels
did not even reach Khiva. The entire expedition was forced to retreat, and
hundreds of men perished in the cold. Despite his easy, default “victory,”
the khan hoped to stave off future Russian campaigns: As open as ever to
negotiating, Allāh Qulı̄ passed a decree forbidding his subjects from rob-
bing or capturing Russian citizens, under penalty of death. After this, he
freed all of the Russian slaves that had been in his own service, and ordered
that other Khwarazmians follow suit. The Russians that were liberated in
this effort were brought to the Russian ambassador so that he could verify
that they had indeed been captives. Once this was done, each freedman
was given a Khivan gold coin (worth about four rubles at that time) and
a bag of flour. One camel was provided for every two men and women, and
by mid-October some 416 freed Russians had arrived in Orenburg.125
E. M. Kosyrev has left an affecting description of these Russians’
condition:126

123
I. N. Zakhar’in, Khiva. Zimnii pokhod v Khivu Perovskago v 1839 godu,—i ‘Pervoe
posol’stvo v Khivu’ v 1842 godu (St Petersburg: Tip P.P. Soikina, 1898), 202.
124
Morrison, “Twin Imperial Disasters,” 285, 291.
125
M. I. Ivanin, Opisanie Zimnego Pokhoda v Khivu v 1839–40g (St. Pb: Tip. Tov.
“Obshchestvennaya pol’za,” 1874), 157. Cf. also Dal, “Khivinskii pokhod,” 169. These
Russians were escorted by the British envoy Richmond Shakespeare, whose role in freeing
them is generally omitted from Russian accounts.
126
E. M. Kosyrev, “Pokhod v Khivu v 1839 godu (Iz zapisok uchastnika),” Istoricheskii
vestnik 8 (1898), 544.
52 The Setting

The captives were a pitiful sight. In ragged Khivan robes, with shaven heads or with
shaggy hair, they barely resembled people. On their faces one could read the
suffering they endured in captivity among the wild Khivan tribes. These unfortu-
nate martyrs discussed the horrid details of the torments they bore. They were sold
into slavery, beaten with whips and lashes, held in vermin-infested places without
food, left at the mercy of horrible insects. Deep scars on their shoulders and backs
gave testimony to the tortures these unlucky ones had borne. Several people had
had their eyes gouged out and, returning their homeland, which they could no
longer see, sobbed and hugged their countrymen. What was in the hearts of these
unfortunates – God only knows. The women were terrible to behold: exhausted,
and with mercilessly defiled honor. It was impossible to look upon them without
crying. The soul seethed for vengeance on behalf of this group of brethren, for
whom life had already withered, and for whom no rosy future opened up, as their
best years and the best of their strength had withdrawn under slavery, violence and
abuse. Many did not survive this crushing slavery, and left their bones in Khiva;
many were executed for trying to escape. Some managed to flee, but their fortune
was no better: they died of hunger and thirst in the endless steppe, to be covered
over in sand or snow. There were also those who converted from their Christian
faith and began to live in high esteem among the Muslims, but there were very few
such cases – most longed violently for their homeland, keeping their faith in Christ
and their love for Russia.

Russian accusations concerning the presence of Russian captives in

Khwarazm – and, more generally, Khwarazmian interference in Russian
efforts, mercantile or otherwise, to travel through the region – did not end
upon the liberation of these 416 captives. But it would be thirty years
before another major military expedition would be launched on the pre-
text of ending the alleged Khwarazmian threat to Russian subjects and
Russian interests in the region.
While Turkmens took scores of Russians into captivity along the
Caspian shores or at sea, Kazakh raiders found easy prey further north,
along the steppe frontier. In 1766, an envoy and military officer named
Bogdan Aslanov reported the presence of some 200 Russian slaves in Khiva
and Bukhara who had originally been taken by Kazakhs near Orenburg.
Hundreds more were taken during the chaos of the Pugachev Rebellion,
and some estimates – unsubstantiated, but noteworthy – posited that the
khanates held anywhere from 2,000 to 5,000 Russian slaves in total.127
These included officers and local officials, serfs, Chuvash, Mordvins,
Poles, and Maris. An unknown number of captives, rather than being

127
M. A. Terent’ev, Istoriia zavoevaniia Srednei Azii, Vol. 1 (St. Pb, 1906), 106; I.I. Kraft,
Sudebnaia chast’ v Turkestanskom krae i stepnykh oblastiakh (Orenburg, 1898), 33–34.
 Russian Interventions in Central Asian Slave Trade 53

sold southward, were held among the Kazakhs in the steppe. Early efforts
to affect the return of these captives included a decree, passed in the winter
of 1767, commanding Russians to take “Asian” captives of their own, so
that these captives might be ransomed in trade for their Russian
counterparts.128 These efforts were in vain, however, and eventually the
governors of Orenburg and Astrakhan resolved to set aside a yearly alloca-
tion of 3–6,000 rubles for paying off the ransom. The standard rate of
ransom in the mid-eighteenth century was 150 rubles per captive, though
some could fetch a higher price.129
These allocations, according to I. I. Kraft, were not sufficient to ransom
all of the captives, and many “laid down their lives in that distant land”
(slozhili svoi golovy na dal’nei chuzhbine). Kraft writes of the new ransom
economy as marking a new era in Russian–Kazakh relations (albeit, unfor-
tunately, without citing his sources):130
[These circumstances] also spawned a new view of trade between Russians and
Asians and of that most shameful trade, as history knows it, the trade in living
people. The Kazakhs took Russian people not so much in battle or as revenge for
some injustice, but simply for profit in the slave trade. Into that shameful trade
came intermediaries and large enterprises [krupnye predprinimateli], competing
with one-another. It is impossible to pass over in silence the fact that numbered
among the large enterprises with respect to the trade in Russian people were
sometimes also Russians, although, fortunately, these fanatics are considered but
few . . .

The numbers of Russians taken by Kazakhs doubtless rose and fell accord-
ing to the state of Russian–Kazakh diplomacy, the character and tenor of
which fluctuated considerably throughout the eighteenth and nineteenth
centuries. As we shall see in coming chapters, however, the capture and
enslavement of Russians would be an ongoing issue in frontier diplomacy
down to the mid-nineteenth century.
In this chapter, I have briefly summarized more than a century of failed
diplomatic efforts on the part of Russia and Iran to negotiate for the return
of their captured subjects and to end captive-taking along their contested
frontiers. Iranian efforts in this respect were generally ineffective, but they

128
Terent’ev, Istoriia zavoevaniia Srednei Azii, Vol. 1, 106. This, Terent’ev writes, was the
natural outcome of circumstances by which Khivan traders could move about freely in
Russia but Russians could not go anywhere in Khiva “without the danger of falling into
perpetual slavery” (“opastnosti popast’ v vechnoe rabstvo”; p. 106).
129
Kraft, Sudebnaia chast’ v Turkestanskom krae i stepnykh oblastiakh, 34–35; cf. also
Terent’ev, Istoriia zavoevaniia Srednei Azii, Vol. 1, 106.
130
Kraft, Sudebnaia chast’ v Turkestanskom krae i stepnykh oblastiakh, 35.
54 The Setting

are crucial to consider because the ongoing crisis of captive-taking in

Khurasan served as a major justification for Iranian military action in the
region, defined Iranian relations with the khanates for decades, and in
several key instances – the campaigns of ʿAbbās Mı̄rzā, for example – had
an immediate impact on the lives of many slaves. Russian efforts tended
likewise to be ineffective, but they too are crucial to consider because the
liberation of Russian slaves (and, more generally, the “pacification” of
Bukhara and especially of Khwarazm) helped justify Russia’s conquest of
the region. In short, in the nineteenth century, the issue of slavery became
the nexus of competing imperial claims over Central Asia.
 2

Beyond the Bazaars

Geographies of the Slave Trade in Central Asia

By the late nineteenth century, as we have seen, the Central Asian khanates
of Bukhara and Khwarazm had long been notorious among Russian and
Western European observers for their part in the region’s thriving slave
trade. This notoriety would have severe consequences: As Alexander
Morrison has recently argued, the ongoing slave trade, as a pretext for
war, “played a crucial role in legitimizing and motivating the Russian
advance” into the region, which culminated in the conquest of Bukhara in
1866–1868 and of Khwarazm in 1873.1 The conquest of the latter was
followed by official proclamations of emancipation and abolition in both
khanates, issued by the khans at the behest of the Russian Governor-
General. These events were widely assumed to have brought about the
end of the slave trade in this part of Central Asia, which has been char-
acterized – according to the standard narrative – by the phenomenon of
nomadic Turkmen and Kazakh slave-raiders transporting their captives for
sale to the region’s major urban centers. There has been little research into
how the trade actually functioned, however, and little evidence to confirm
that the Russian conquests brought about its demise. In this chapter, I will
argue two points: (1) that the urban centers of Khwarazm and Bukhara
were, for many slaves, merely transit points in a decentralized network of
trade extending well beyond the major towns and their bazaars; and (2)
that the decentralized nature of the slave trade demands that we reject
long-held notions about Russian and local authorities effectively

1
Alexander Morrison, “Twin Imperial Disasters. The Invasions of Khiva and Afghanistan in
the Russian and British Official Mind, 1839–1842,” Modern Asian Studies 48/1 (2013),
282–283.

55
56 Beyond the Bazaars

abolishing the slave trade in the 1870s. I will draw evidence for these
arguments from eyewitness accounts of the trade, as well as from the
firsthand experiences of Central Asian slaves themselves, preserved in
unpublished interviews found in Central Asian archives.

A DECENTRALIZED SLAVE TRADE

According to a longstanding consensus among historians of Eurasian

economies, the rise of maritime trade linking Western Europe with India
and China in the fifteenth and sixteenth centuries had a devastating effect
on the caravan trade across Central Asia, rendering it largely obsolete.2
This overland trade had formerly brought prosperity to Central Asia’s
cities by promoting the urban-centered production of everything from
paper to textiles. Isolation from the new oceanic trade routes was thus
(the story goes) catastrophic for the regional economy as a whole, and it
resulted in decentralization and rapid deurbanization. In recent decades,
however, research into overland trade in the region after the sixteenth
century has revealed that it was far from stagnant. Audrey Burton’s land-
mark work on Bukharan trade networks, for example, reveals how the
khanate engaged in extensive trade with nomadic, sedentary, rural, and
urban peoples throughout Eurasia, thanks to overland commercial net-
works extending through Khwarazm, East Turkistan, Iran, China,
Muscovy, Siberia, the Ottoman Empire, and India.3 James Millward has
shown in remarkable detail how Qing China engaged neighboring Central
Asian regions by maintaining overland trade networks in a multitude of
commodities throughout the eighteenth and nineteenth centuries.4 Trade

2
Recent examples include Christopher I. Beckwith, Empires of the Silk Road: A History of
Central Eurasia from the Bronze Age to the Present (Princeton, NJ: Princeton University
Press, 2009). For summaries and critiques of this approach, see especially the work of Scott
C. Levi, The Indian Diaspora in Central Asia and Its Trade, 1550–1900 (Leiden: Brill,
2002); “Early Modern Central Asia in World History,” History Compass 10:11 (2012),
866–878; “India, Russia and the Eighteenth-Century Transformation of the Central Asian
Caravan Trade,” Journal of the Economic and Social History of the Orient 42:4 (1999),
519–526. See also Morris Rossabi, “The ‘Decline’ of the Central Asian Caravan Trade,” in
James Tracey, ed., The Rise of Merchant Empires (Cambridge: Cambridge University Press,
1990), 351–370; and Robert D. McChesney, Central Asia: Foundations of Change
(Princeton, NJ: Darwin Press, 1996), 41–42.
3
Audrey Burton, The Bukharans: A Dynastic, Diplomatic, and Commercial History
1550–1702 (New York, NY: St. Martin’s Press, 1997).
4
Millward, Beyond the Pass: Economy, Ethnicity, and Empire in Qing Central Asia,
1759–1864; Millward also offers important comments challenging the Silk Road “decline”
hypothesis on pp. 98–101.
 A Decentralized Slave Trade 57

by Indian merchants in Central Asia during this period has, meanwhile,

been the subject of important studies by Scott C. Levi, Stephen Dale,
Muzaffar Alam, Claude Markovitz, and others.5 While the mere existence
of overland trade, as Ron Sela has observed, is not sufficient to prove that
Central Asia as a whole was a thriving, flourishing region during these
turbulent centuries,6 it is nevertheless worthwhile to consider the nature of
the existing commerce and how some types of trade managed to flourish.
The trade in slaves can be counted among those types of commerce that
thrived in the eighteenth and nineteenth centuries, and it was bolstered by
continual conflict in Khurasan and the Kazakh steppe as well as by demand
for slaves in Bukhara, Khwarazm, and among rural nomads.
As we shall see, much of the slave trade took place outside major urban
centers, with slaves being transported along caravan routes across rural
Central Asia and traded in rural caravanserais and settlements. They were
often traded by and among nomads, and it may be for precisely this reason
that studies describing this slave trade remain so rare. Nomadic commerce,
after all, has most often been considered not in relation to the wide-
ranging caravan trade but rather in relation to specific urban centers, and
the most common approach has followed a core/periphery model: The city
is the core and the nomads populate the periphery, like spokes radiating
from a hub. In the decades since Anatoly Khazanov published his influen-
tial Nomads and the Outside World, the relationship between Eurasian
nomads and the urban “core” has frequently been described as a symbiotic
one, with the former providing the cities with goods or services in
exchange for food, handicrafts, and luxury items; the urban “core” con-
stituted the final destination of the nomads’ goods, as well as the source of
goods and revenue they gained in trade.7
In Central Asia, slaves were undoubtedly a major source of trade
revenue for many nomads, particularly Turkmens, and they were most
5
Levi, The Indian Diaspora; Stephen Dale, Indian Merchants and Eurasian Trade, 1600–1750
(Cambridge: Cambridge University Press, 1994); Muzaffar Alam, “Trade, State Policy and
Regional Change: Aspects of Mughal-Uzbek Commercial Relations, c. 1550–1750,”
Journal of the Economic and Social History of the Orient 37:3 (1994), 202–227;
Claude Markovitz, The Global World of Indian Merchants, 1750–1947: Traders of Sind
from Bukhara to Panama (New York, NY: Cambridge University Press, 2000); Levi and
Alam, eds., India and Central Asia: Commerce and Culture, 1500–1800 (Oxford: Oxford
University Press, 2007).
6
Ron Sela, The Legendary Biographies of Tamerlane: Islam and Heroic Apocrypha in Central
Asia (Cambridge: Cambridge University Press, 2013), 122.
7
Anatoly Khazanov, Nomads and the Outside World (Madison, WI: University of Wisconsin
Press, 1994 [second edition]). See also Khazanov and Andre Wink, eds., Nomads in the
Sedentary World (New York, NY: Routledge, 2001).
58 Beyond the Bazaars

certainly traded and sold in the cities. But the kinds of bustling slave-
bazaars found in Istanbul or Timbuktu, in which hundreds or even
thousands of individuals were displayed for purchase at any given time,
simply did not exist in the region. While a city such as Khiva was a site of
slave-trading and of slave labor, it was hardly the “core” of the slave
trade; it was merely one possible transit point among many. In fact, as
I will argue here, slaves were sold and traded at caravanserais and in rural
settlements throughout the region. Rather than searching for a “core” or
“periphery,” it would be better to think of the slave trade in terms of
“circulation.”8 Slaves and their sellers circulated in decentralized zones
of commerce.
Two aspects of the Central Asian trade contributed to its decentralized
character. First, a significant portion of the trade took place between
nomads and other nomads. Second, the nature of slavery itself in the
region was distinctly rural: Eyewitness accounts as well as studies of the
region’s urban economies have confirmed that the predominant type of
work that slaves performed was agricultural.9 The primarily rural char-
acter of slavery was a consistent trait over the course of centuries, although
the trajectory of the trade in slaves shifted over time. The demographic of
slaves from India, once prominent in the region, was fast disappearing by
the end of the seventeenth century, with Iranian Shiʿite slaves from

8
Claude Markovits, Jacques Pouchepaass, and Sanjay Subrahmanyam have recently offered
an expansive definition of this term that touches upon the circulation not only of people
and objects in trade-networks but also of “information, knowledge, ideas, techniques,
skills, cultural productions (texts, songs), religious practices, even gods” (“Introduction:
Circulation and Society under Colonial Rule,” in Markovitz, Pouchepaass, and
Subrahmanyam, eds., Society and Circulation: Mobile People and Itinerant Cultures in
South Asia, 1750–1950 (London: Anthem, 2003), 2–3. Here, I discuss only the circulation
of people and merchandise, but where people go, their ideas, songs, and gods go with them,
and one of the goals of this chapter is to reveal how people circulated, both with merchan-
dise and as merchandise.
9
For example, “Nevol’niki v Khive,” Vestnik Evropy 80:7 (1815), 245; Major Blankenagel’,
Zamechaniia maiora Blankenagelia, vposledstvie poezdki ego iz Orenburga v Khivu
v 1793–94 godakh (St. Pb, 1858), 13, 17; N. N. Murav’ev, Muraviev’s Journey to Khiva
through the Turcoman Country, 1819–20 (Calcutta: Foreign Department Press, 1871), 144;
N. Zalesov, “Pis’mo iz Khivy,” Voennyi sbornik 1 (1859), 288; Georges de Meyendorff,
A Journey from Orenburg to Bukhara in the Year 1820 (Calcutta: Foreign Department Press,
1870), 62. R. K. Mukminova’s study of sixteenth-century waqf documents has revealed that
most of the slaves mentioned therein were used for agricultural labor or animal husbandry
(Mukminova, Sotsial’naia differentsiatsiia naseleniia gorodov Uzbekistana v XV–XVI vv.
[Tashkent, 1985], 122–123; see also Levi, “Hindus Beyond the Hindu Kush,” 278n5).
Other sources reveal, for example, large numbers of slaves working in agricultural on the
estates of prominent Sufi leaders such as the Juybari shaykhs (ibid., 278).
 Evidence from the Lives of Slaves 59

Khurasan taking the Indian slaves’ place.10 The Turkmen deserts north of
Khurasan became increasingly prominent zones of captivity, transport,
trade, and sale. By the nineteenth century, Iranian Shiʿites would, by all
accounts, constitute an overwhelming majority of the region’s slaves.
Through the northward trajectory of these trade networks, they came to
form a significant demographic in Khwarazm and Bukhara. Estimates of
the number of slaves in Khwarazm in the nineteenth century generally
range from 30,000 to 60,000 individuals; similar numbers were estimated
for Bukhara.11

EVIDENCE FROM THE LIVES OF SLAVES

As the region’s slave traders have left us no ledgers or diaries, the best way
to reconstruct the trade itself is through eyewitness accounts, as well as
through the declared experiences of the slaves themselves. For much of the
Muslim world, detailed information on the lives of individual slaves, and
especially accounts authored by individual slaves, appear to be few and far
between.12 Based on the recent ethnographic literature on Ottoman

10
Levi, “Hindus Beyond the Hindu Kush,” 277–279.
11
For example, among our nineteenth-century estimates for Khwarazm, A. P. Khoroshkhin,
who had been involved in a Russian census project based at Orenburg after the conquest of
Khiva, estimated that 29,291 Iranian slaves had been living in the khānate, along with 6,515
manumitted slaves (Khoroshkhin, Sbornik statei: Kasaiushchikhsia do Turkestanskago kraia
[St. Pb., 1876], 486); Murav’ev estimated at least 33,000 slaves, 30,000 of them Iranians
(Murav’ev, Muraviev’s Journey to Khiva, 57–58); Blankenagel’ estimated 25,000 slaves
(Blankenagel’, Zamechaniia maiora Blankenagelia, 12–13; Herbert Wood estimated
50,000 Iranian slaves and manumitted former slaves (Wood, The Shores of Lake Aral
[London: Smith, Elder & Co., 1876], 219; Basiner estimated 52,000 Iranian slaves (see
Emil Schmidt, The Russian Expedition to Khiva in 1873 [Calcutta: Foreign Department
Press, 1876], 12); Vambery estimated 40,000 Iranian slaves (ibid., 12); Alikhanov-Avarskii
estimated 25,000–40,000 total slaves (M. Alikhanov-Avarskii, Pokhod v Khivu (Kavkazkikh
otriadov) 1873. Step’ i oazis [St. Pb, 1899], 280); and Hassan Muhammad Amı̄n Oghlı̈,
˙ ˙
author of an unpublished Chaghatay tract on slavery from the early Soviet period, estimates
that there had been 58,500 slaves and manumitted former-slaves in 1873, including Iranians,
Kurds, Afghans, Azeri Turks, and Russians (MS IVAN Uz No. 12581, f. 51b). For Bukhara,
estimates are scarcer; Meyendorff estimated 30–40,000 Iranian slaves, and the Greek
metropolitan Khrisanf Neopatraskii, writing near the end of the eighteenth century, esti-
mated 60,000 Iranian slaves (in S. G. Karpiuk, ed., Puteshestviia po Vostoku v opokhu
Ekateriny II [Moscow: Vostochnaia literatura, 1995], 279). Joseph Wolff, the eccentric
missionary and adventurer, estimated 200,000 slaves, though this figure seems unlikely
(Wolff, Narrative of a Mission to Bukhara in the Years 1843–1845, Vol. 2 [London: J.W.
Parker, 1845], 104).
12
On the relative lack of firsthand slave narratives in the Ottoman context, see, for example,
Y. Hakan Erdem, “Slavery and Social Life in Nineteenth-Century Turco-Egyptian
60 Beyond the Bazaars

slavery, for example, it seems that there are precious few enslaved persons
for whom we know the full complement of basic information: their birth-
place; their age at capture, if captured, along with the location and cir-
cumstances of their capture and who captured them; their sale price and
the location(s) of their sale(s); the length of their captivity; the sort of work
they did; the identity and character of their owner(s); their treatment at the
hands of their owner(s); their marital status; and the manner and means of
their manumission, if freed, as well as what became of them afterwards.13
This was precisely the sort of data collected from Central Asian slaves
and former slaves by the officials of the Orenburg Border Commission and
other Russian border authorities,14 however, and the scarcity of such
information for the broader Eurasian world would render even a few
such interviews precious. For the present work, I have collated data on
forty-five of these slaves and former slaves for whom the documentary
record is particularly rich; for each of these individuals, most or all of the
above information was collected at the Russian border.15 In some

Khartoum,” in Terence Walz and Kenneth M. Cuno, eds., Race and Slavery in the Middle
East: Histories of Trans-Saharan Africans in 19th-Century Egypt, Sudan, and the Ottoman
Mediterranean (Cairo: American University in Cairo Press, 2010), 125; and Toledano, As If
Silent and Absent, 52, 57.
13
Recent examples of an ethnographic approach to Ottoman slavery include Ehud
R. Toledano, As If Silent and Absent (New Haven, CT: Yale University Press, 2007), and
Eve Troutt Powell, Tell This in My Memory: Stories of Enslavement from Egypt, Sudan, and
the Ottoman Empire (Palo Alto, CA: Stanford University Press, 2013).
14
Established in 1799, the Commission was tasked with extending legal and administrative
control over the Kazakhs of the steppe; the Junior Horde was subjected to particular
scrutiny, as its territories of migration abutted the Russian border. The reasons why border
officials would have been tasked with recording former slaves’ testimonies and autobio-
graphical details are not entirely clear, but a couple possibilities come to mind. First, the
former slaves, once they had arrived at the border, would either be received as citizens of
the Russian Empire or returned to the steppe, and their testimonials, once written, would
serve as proof that protocol had been followed in “processing” these individuals. Second,
the testimony of the former slaves often contained details concerning violence and ten-
sions between nomadic groups that could be of interest in formulating the Empire’s steppe
diplomacy and policies.
15
The information collated is found in the following documents held by the Central State
Archive of the Republic of Kazakhstan (henceforth TsGAKaz): TsGAKaz 4.1.3646, ff.
45a–49b; ff. 73a–78a; TsGAKaz 4.1.198, f. 3a; ff. 19a-b; 36a–37a; 53a–54b; 104a; 137a;
ff. 173a–b; TsGAKaz 4.1.195, ff. 10a–11a; TsGAKaz 4.1.3641, ff. 36a–38b; TsGAKaz
4.1.2821, ff. 2a–3a; 6a–b; TsGAKaz 383.1.89, ff. 14a–b; TsGAKaz 4.1.3573, ff. 40a–41b;
132a; 227a–b; TsGAKaz 4.1.197, f. 3a; 44a–b; 81a–b; TsGAKaz 4.1.3730, ff. 5a–b;
17a–18a; TsGAKaz 4.1.499, ff. 104a–b. Most of these documents date from between
1850 and 1861, with a few dating from as early as 1800. Some of these records are quite
concise, comprising merely a few short sentences, while others are significantly more
elaborate, but all of them provide much of the information listed in this section.
 Evidence from the Lives of Slaves 61

documents, their lives are recounted in the first person (albeit usually
translated into Russian by interpreters, who were present and named).
Most of these biographies and autobiographies follow a common formula,
as if in response to a questionnaire, and the formula remained consistent
throughout the nineteenth century. The accounts are short on sentiment,
but the consistent nature of the information they provide lends itself well
to a comparative assessment. With tens of thousands of slaves in the region
at any given time, such a small sample cannot be taken for a comprehensive
prosopography of Central Asian slaves in general. But certain striking
commonalities that emerge from this small sample cannot be dismissed
as mere coincidence; and at times, these commonalities defy common
conceptions. The evidence provided by these biographies, when combined
with evidence offered in other sources, points to a re-envisioning of the
nature of nomadic trade in Central Asia, as well as to a revision of the
standard narrative of how slavery came to an end in the region.
The origins of these forty-five slaves and former slaves are recorded
as follows: thirteen were described as Persian; seven as Kalmyks; four as
Afghans; three as Arabs; two as Hazaras; two as Turkmens; two as
“Bukharans”; two as Kirghiz; two as Karakalpaks; two as “Central
Asians”; one as a Tatar; one as an Uzbek; and four as of unknown
origin.16 These forty-five individuals – forty-one men and boys, and four
women and girls – had lived a combined 683 years in captivity. It is claimed
in some sources that slaves in the Muslim world could very often expect
their freedom after a period of roughly seven years, since it was at this
point that slaves became eligible to purchase their own freedom.17 Indeed,
this alleged peculiarity of slavery in the region has been used to argue that
Central Asian slaves – and slaves in the Muslim world more generally – had
an easier time of it than slaves elsewhere, as they were rarely held in
perpetuity. Of the thirty-five slaves in the present cohort whose total
16
Typically, those of unknown origin had been taken captive very early their childhood.
Here, I have specifically selected accounts given by Muslim slaves, as these slaves formed
the overwhelming majority of slaves in Central Asia as a whole. Russian slaves were
likewise received, liberated, and interrogated, however, and some excellent recent
research offers coverage of their fates as well as the (sometimes evidently “dramatized”)
accounts of their captivity: Smolarz, “Speaking about Freedom and Dependency,” 44–71;
Gao, “Captivity and Empire: Russian Captivity Narratives in Fact and Fiction.”
17
For example, Toledano, As If Silent and Absent, p. 91. William Gervase Clarence-Smith
provides a comparative overview of slaves’ periods of captivity throughout the Muslim
world: Islam and the Abolition of Slavery (Oxford: Oxford University Press, 2006), 67–68.
The period of seven years’ captivity before manumission appears to have had deep roots as
a tradition in the Judeo-Christian world; it is mentioned in the Old Testament
(Deuteronomy 15: 1–18); Clarence-Smith, 222.
62 Beyond the Bazaars

period of captivity is recorded, however, only three individuals had spent

fewer than fifteen years in bondage. Only two had been enslaved for
a period of fewer than seven years before gaining their freedom. (Both of
these two had been presented at the Russian border by their owners after
the circulation of Imperial decrees threatening punishment for all Kazakhs
under Russian dominion who did not bring forward their slaves.18) Sixteen
others, having been presented to the Russian authorities, requested to be
sent back with their owners in order to continue in their present circum-
stances – their captivity was open-ended at the time of their interview, in
other words. In all, the mean period of captivity among this cohort of
slaves was 19.51 years, with a median of fifteen years. This tells us, among
other things, that these slaves had substantial experience of captivity by the
time they were interviewed.
Though nearly all of these forty-five slaves had come to the Russian
authorities from among the Kazakhs, their generally long duration of
captivity cannot be considered a trait unique to the Kazakh context:
These statistics alone do not reflect the fact that a great many of these
slaves had several owners over the years. Indeed, the prominence of traffic
from owner to owner is another striking commonality that emerges in their
biographies. This “horizontal” transit is especially pronounced among
those slaves who were initially taken captive by the Turkmens: For this
cohort of sixteen individuals, ten were sold more than once, and on the
whole they were sold an average (mean) of 2.43 times per individual. This
mobility from one owner to another is worth bearing in mind as we come
to consider slaves’ transit across the region. We will now turn to the
autobiography of one of these slaves, whose story unfolds across
a remarkably broad geographical and cultural range.
Received by Russian border authorities in November of 1852, a man
named ʿAbdullāh, son of one Mullā Walı̄ Yār, described himself as forty
years old, a Shiʿite, and originally from Herat. Fourteen years prior, he had
been living with his parents, his elder brother, his wife, his two sons
(Begrashı̄d and Sayfullāh), and six servants (prislugi) on a farm the family
owned near Herat. One day, they were attacked by a host of some 2,000

18
On these policies and decrees, see, for example, TsGAKaz 383.1.184, ff. 11b–19a;
TsGAKaz 383.1.89, ff. 10a–14b. The earliest of the decrees that ordered general manu-
mission among Kazakhs under Russian dominion appears to have been issued in 1859,
with similar decrees following in 1860, 1861, and 1869 (G. I. Semeniuk, “Likvidatsiia
rabstva v Kazakhstane,” repr. in Zh. O. Artykbaev et al., eds., Raby i tiulenguty
v kazakhskoi stepi [Astana: Altyn Kitap, 2006], 238). Slaves manumitted at the border
before this period were generally escapees.
 Evidence from the Lives of Slaves 63

Teke Turkmens, who were in some sort of relationship of dependency to

Merv, and had been – according to ʿAbdullāh – at the command of Tabsad
Khan [sic]. The attacking party robbed the family of their property, making
off with 200 sheep, 60 horses, and 50 head of cattle. The Turkmens took
the family captive, along with their servants, numbering thirteen people in
all. (ʿAbdullāh’s third son, Iskandar, and his daughter, A___,19 had been
living with his father-in-law, who was an official in Herat, so they were
spared from the attack.) The captive family was taken to the Turkmens’
camp, southwest of Merv, where they remained for four days. During that
time, the Khan freed ʿAbdullāh’s parents and sent them home, returning to
them from among their stolen goods a carpet, some cattle, a cooking-pot,
a blanket, and all of his father’s books (his father had a reputation in the
region, being a well-known mullā). ʿAbdullāh’s brother was sold to some
Turkmens for the very low sum of seven tillas, and ʿAbdullāh was sold for
eight tillas to a Turkmen named Berdi-Kilich, who in turn sold him at
a profit to a Khivan two months later, earning twelve tillas. The sale
evidently took place in the Turkmens’ camp near Merv. The Khivan,
named D___, immediately took ʿAbdullāh to the town of Urgench and
sold him to another Khivan, this one named Palvān Niyāz, for eighteen
tillas. After five days, Palvān Niyāz sold him to another Khivan, identified
as “yuzbāshi Yakub,”20 for twenty-five tillas. In less than three months,
ʿAbdullāh had already been sold four times.
Soon after arriving at the yuzbāshi’s house, Abdullah received the news
that his father-in-law, Arbab Zulfikar [sic], the Herati official, had written
to Tabsad Khan regarding the manumission of ʿAbdullāh’s wife and two
sons, who had also been sold off. He learned also that their servants had
been sold to Khiva, but he did not find out where they and his elder brother
were currently residing.
ʿAbdullāh lived with the yuzbāshi for the next four years, evidently in
Urgench, doing agricultural labor for him, and after four years his master
sold him to a Khivan named Yusuf Khop-Khar [sic] for twenty-five tillas,

19
The Russian transliterations of non-Russian names that appear in these documents often
appear strange or ambiguous to me, and, rather than continuing to garble these indivi-
duals’ names for posterity, I have opted to use first-initials wherever I have had especially
strong reservations about proper identification and rendering of a name.
20
In Khwarazm, the title of yuzbāshi – literally meaning “head of one hundred” – did not,
according to Murav’ev, indicate a formal position, but was rather an honorary title
“bestowed by the Khan in war time on the officers commanding divisions of his forces”
(N. N. Murav’ev, Muraviev’s Journey to Khiva through the Turcoman County, 1819–20,
trans. Philipp Strahl [Calcutta: Foreign Department Press, 1871], 50). Those upon whom
this title was bestowed tended to retain it indefinitely, and not only at times of war.
64 Beyond the Bazaars

earning back the sum he had initially paid for ʿAbdullāh four years prior.
After five days, this Khivan sold ʿAbdullāh to a Karakalpak named Mullā
Bishbay [sic] for thirty-seven tillas, and he lived with the Mullā for the next
seven years, performing domestic labor (domashnaia rabota) for him.
At the end of seven years, the Mullā sold ʿAbdullāh for thirty tillas to
a Kazakh of the Tabı̈n clan who was a nomad in Khivan territory. He lived
with this nomad for a month before being sold once again, this time for
three horses and twenty sheep, to a Kazakh named Chin-Timur [sic], who
was from the Shekti clan and lived as a nomad along the river Emba. He
lived with Chin-Timur for a little over a year before this owner presented
him to Sultan Arslan Jan-Turin [sic], who gave him over to the Russian
authorities. ʿAbdullāh reported that seven of his owners subjected him to
no manner of abuse or mistreatment – implying, however, that one of his
eight total owners did mistreat him. He requested that the Orenburg
Border Commission convey him back home to Herat, at long last.21
Over the course of fourteen years, ʿAbdullāh was enslaved by a remark-
ably diverse cross-section of Central Asian society. He was held and sold by
Turkmen raiders, by Khivan traders, by a Karakalpak mullā, by an Urgench
yuzbāshi, and by two different Kazakh nomads. We know that he worked
both as a domestic servant and as an agriculturalist, and one can only
imagine the range of households to which he became accustomed – from
Turkmen camps to Kazakh tents to the yuzbāshi’s estate. His price ranged
from seven tillas, near the site of his capture, to thirty-seven tillas as an
“import” at an urban center. “Horizontal” mobility and demographic
diversity characterize his ordeal, and in both of these traits ʿAbdullāh’s
narrative is no great exception in our cohort of forty-five. His story also
follows a general pattern among our biographies in that a significant por-
tion of his experience took place beyond the “urban” sphere.
The first half of ʿAbdullāh’s story is quite in keeping with common
perceptions of how the slave trade functioned: archetypically, Turkmens
raided Khurasan and brought Iranian captives to the bazaars in Khiva or
Bukhara, where they would enter the service of urban notables. Here we
find the standard model of nomadic economic networks, with nomads
supplying cities whose markets were the nexus of the slave trade.
The nomadic and sedentary economies thus become inseparable, as in
the widely repeated observations of Murav’ev: “The practice of catching
human beings and selling them to the Khivans has become an absolute
necessity to the nomadic tribes; that is to say, the latter have to depend for

21
TsGAKaz 4.1.3646, ff. 75b–76b.
 Evidence from the Lives of Slaves 65

grain on Khiva, and grain cannot be grown there without extraneous

labour, so that this abominable trade has become an institution for the
mutual benefit of Khiva and the predatory tribes, without which neither
could exist.”22
Soon after, however, ʿAbdullāh passes from the cities back into the
hinterlands, where he is traded among nomads. Rare indeed is the study
that emphasizes trade among and within nomadic groups, and scholars
attuned to the many reminders that nomadic and sedentary economies
were inextricable may begin to wonder whether there was any such thing
as an intranomadic economy, and whether there is any alternative model
for nomads’ trade-relations other than symbiosis (or, according to older
models, parasitism) with respect to settled regions. For our purposes, two
questions emerge in light of ʿAbdullāh’s story: First, did a significant slave
trade exist beyond the urban bazaars? And, second, could the slave trade
persist without the bazaars?
At first glance, the travelers and diplomats who provide much of our data
on the nineteenth-century slave trade do not give us much cause to depart
from the standard “urban core/nomadic periphery” models sketched earlier,
or even to elaborate on them. In source after source, we are told of a linear
progression: captives were dragged from the peripheries to the urban bazaars
of Khiva and Bukhara, with Turkmens – and, less commonly, Kazakhs – as
their vector. The center of the trade was the urban bazaar, and the nomads
themselves were ever the captivators, armed chaperons, sellers, and nothing
more. Thus, we have it since the very earliest Russian appraisals of the slave-
dealing economy, such as the following, dating from 1815:
The Kirghiz [Kazakhs], marauding raiders in Russian lands, and the Turkmen in
Persian lands, abduct no small number of people, and sell them as slaves to the
Khivans, who, being squarely between Russia, Persia, and Bukhara, have for some
time maintained for themselves a healthy trade. Without their mediation, the
thieves would not be able to have so much success and [such a level of] dissemina-
tion; for the [Kazakhs], if they had not always such ready and generous bidders in
the Khivans, would not be so eager to go and take captives, inasmuch as they are but
little kept as slaves [for themselves].23

22
N. N. Murav’ev, Muraviev’s Journey to Khiva through the Turcoman County, 1819–20,
trans. Philipp Strahl (Calcutta: Foreign Department Press, 1871), 148. He elsewhere
observes, “The Turcomans buy their grain at the markets, and dispose of slaves, so useful
in the cultivation of the land, as they are generally the ploughmen; were the trade with the
Turcomans to cease, Khiva would lose the chief source of its prosperity, and probably sink
back into insignificance” (ibid., 144).
23
“Nevolniki v khive,” Vestnik Evropy 80:7, (1815), 244.
66 Beyond the Bazaars

The idea that Kazakhs and Turkmens kept few slaves among themselves
plays into the notion that they were exclusively raiders, transporters, and
dealers, and it is likewise a longstanding misconception. One can trace the
idea from early Russian travel reportage through Soviet scholarship.
A. I. Levshin, who traveled among the Kazakhs in the first half of the
nineteenth century, was perhaps the earliest to propose that they kept no
slaves among them – other than their women, who, Levshin characteristi-
cally claims, generally took the place of slaves, given the alleged laziness of
Kazakh men.24 In the many years since, some scholars have flatly denied
that slavery could develop in nomadic societies in general, and particularly
among the Kazakhs. Their reasoning usually concerns: (1) the relative lack
of labor-intensity involved in herding; (2) the weak development of agri-
culture and crafts among the Kazakhs; (3) the availability of alternative
sources of inexpensive or free labor; and (4) the prohibitive expense of
slaves when compared with the average Kazakh’s worldly holdings.25
The first three points presume that owning slaves was merely a matter of
perceived labor necessity – a simplistically “rationalist” vision of slavery
which overlooks the many tasks performed by slaves, throughout human
history, which have been neither rational nor necessary nor always readily
identifiable as “labor.” This leaves us with the notion of prohibitive
expense. Semeniuk, for one, argues that a slave typically cost, among the
Kazakhs, an average of seventy-five sheep, while, according to my collation
of his data, only 21.1 percent of Kazakh households in his sample (from
mid-nineteenth-century Omsk) had more than fifty sheep, and only
9.6 percent had more than 100.26 By considering ʿAbdullāh’s biography,
however, among others, we can see that prices fluctuated wildly.
ʿAbdullāh’s price rose by nearly 500 percent – from eight tillas to thirty-
seven tillas – in different contexts and at different times. At a mere eight
tillas, ʿAbdullāh would have been an obtainable commodity for many. And
some slaves were free: They could be captured, and they could reproduce.

24
A. de Levchine, Description des hordes et des steppes des Kirghiz-Kazaks ou Kirghiz-
Kaissaks (Paris: Impr. royale, 1840), 354.
25
On these points, see Semeniuk, “Istochniki rabstva v Kazakhstane v XV–XIX vekakh,” in
Raby i tiulenguty v kazakhskoi stepi, 140–41, 151, 176; K. Kraft, “Unichtozhenie rabstva
v kirgizskoi stepi,” in Iz kirgizskoi stariny (Orenburg, 1900), 95. David Sneath has con-
nected the claim that Central Asian nomads had few slaves to a more general preoccupa-
tion among scholars, especially in the Soviet Union, with the idea – one Sneath combats at
length – that nomadic societies were fundamentally egalitarian (Sneath, The Headless
State: Aristocratic Orders, Kinship Society, & Misrepresentations of Nomadic Inner Asia
[New York: Columbia University Press, 2007], 152–154, 186).
26
Ibid., 150–151.
 Evidence from the Lives of Slaves 67

If one is still inclined to question whether the Kazakhs kept significant

numbers of slaves, the Border Commission’s tally of freedmen – those
who either fled captivity in the steppe or were freed by their masters, and
arrived safely at the border – can be illuminating. The earliest numbers
I have seen are already impressive: between 1749 and 1753, no fewer
than 1,024 former slaves from among the Kazakhs turned up at Russian
defensive posts.27 Of these individuals, 862 claimed to have fled their
owners.
The most compelling evidence concerning slavery among the Kazakhs,
however, is the testimony of the slaves themselves. It is with this in mind
that we return to ʿAbdullāh, whose journey eventually took him in pre-
cisely the opposite direction from the standard model of nomadic trade: he
went from the city of Khiva to the domains of nomadic Kazakhs and
Karakalpaks, rather than vice-versa. Here, his story is similar to the others
in his cohort: all spent some portion of their captivity among the Kazakhs,
as one might expect, given that they emerged from the steppe; more
surprisingly, most spent nearly the entirety of their captivity among
Kazakhs. The sedentary centers of Khiva and Bukhara emerge in these
biographies as merely two transit-points in a series of sales. Generally,
major towns are rarely mentioned in these documents. Another example
will further illustrate the common feature of broad mobility beyond the
bazaars and urban centers.
Like ʿAbdullāh, Kalbay Hamzin [sic] was interviewed by Russian border
officials in November of 1852. He was then twenty-four years old, a Shiʿite
from the town of Kalyan, one-and-a-half days’ caravan journey from
Mashhad, where his parents and brother were evidently living at the
time of his interview. Six years prior, at the end of a winter night, Kalbay
and twelve companions were grazing sheep about twelve versts (eight
miles) from town, when they were attacked by forty Teke Turkmens who
had been based at Ashgabat. These Turkmens took them captive and
brought them to the town of Alkabad, where they were divided up, and
Kalbay was sold to a Turkmen with whom he lived for one month. This
Turkmen then sold Kalbay to another Turkmen, named Khwāja Bulak [sic],
for thirty-seven tillas. Khwāja Bulak was from one of the tribes which had
accompanied the raiding party. He immediately took Kalbay to the Khivan
village of Pars, two days’ caravan journey from the town of Urgench, where
he sold him to a Khivan named Kut-Muhammad [sic] for thirty tillas. He
˙
lived with Kut-Muhammad in Pars for the next five years, serving him by
˙
27
“Iz istorii Kazakhstana XVIII v.,” Krasnyi arkhiv 2:87, (1938), 167–168.
68 Beyond the Bazaars

planting grain and working the land. Finally, he was sold by Kut-
Muhammad to a Khivan trader of Persian slaves named Palvān, for thirty
˙
tillas. This trader turned around and sold him for the same amount to
a Kazakh of the Tabı̈n clan, living within the domain of Khiva. Kalbay lived
with this Kazakh for two months before he was traded to another Kazakh,
this one named Murat-Bek [sic], of the Karakesek, in exchange for ten
large sheep, a camel, a horse, and three large felt carpets. Kalbay stayed
with Murat-Bek for seven months, and he was then sold to a Kazakh
named Butunbay [sic] of the Nazar clan,28 with whom he lived for another
four months. This last master finally presented Kalbay to the sultan of the
Middle Horde. At the conclusion of this narrative, Kalbay adds the details
that all of his masters treated him well, that they did not beat or abuse him,
and that he possessed no livestock or holdings of his own, or of his former
masters.29
As with ʿAbdullāh, Kalbay spent much of his captivity in the steppe. It is
not clear that he ever passed through the urban bazaars of Khiva, Urgench,
and Bukhara, and his time in the Khwarazmian khanate was evidently
spent in remote, rural quarters. This rural orientation is, as mentioned,
something his narrative shares with nearly all others in the cohort, and in
order to appreciate the magnitude of this commonality, we will now step
back briefly from particular narratives and have a look at trends related to
acts of sale.
Of these forty-five slaves, seven were kept by their original captors, and
the other thirty-eight were sold – so far as their biographies reveal – a total of
sixty-seven times altogether. For sixty-three of these recorded sales, some-
thing of the origins of the buyer and seller is recorded, i.e. whether they were
“Kirghiz” (Kazakh), “Khivan” (a category which seems to exclude Turkmens
and Kazakhs within Khivan domains), Karakalpak, Turkmen, and so on.
Given the urban-centered model offered by our eyewitness accounts and by
later secondary literature, one would naturally expect that most of the sales
took place between Turkmens and Khivans, between Khivans and other
Khivans, and – because these slaves issued from the Kazakh steppe –
between Khivans and Kazakhs. After all, the best-known sources are unan-
imous in emphasizing that the bazaars of Khiva and Bukhara served as the
crux of the traffic in slaves. But in fact only five of the sixty-three detailed
sales took place between Turkmens and sedentary Khivans. Another four
28
The Nazars were a large subdivision of the Shektis, living along the Emba River. As of 1848
they reportedly numbered some 1,200 yurts. I am grateful to Allen J. Frank for providing
this information.
29
TsGAKaz 4.1.3646, ff. 74a–75a.
 Evidence from the Lives of Slaves 69

took place between Khivans and other Khivans. In other words, only
14 percent of the recorded sales involved a sedentary Khivan buyer, and
only eleven sales (17 percent) involved a sedentary Khivan seller. Bukharans
were involved in six of the sales, and in four of these cases they were
recorded selling slaves to the Kazakhs. On both the buying end and the
selling end, Turkmens and Kazakhs predominate. Sixteen of the sales
involved Kazakhs selling slaves to other Kazakhs, with another eight being
sales in which the buyers were Kazakhs and the sellers were, based on
context clues, very likely other Kazakhs. Another eleven sales involved
Turkmens selling slaves to Kazakhs. There were two sales by Turkmens to
other Turkmens, two by Kazakhs to Karakalpaks, and two by Karakalpaks to
Kazakhs. In all, among the sixty-three detailed sales, these sources document
fifty buyers from nomadic or semi-nomadic groups (Turkmens, Kazakhs,
Karakalpaks); at least thirty-six sellers from these groups; and no less than
thirty-three cases in which both seller and buyer were from one of these
(generally nonsedentary) groups. The predominance of urban centers –
Khiva, Bukhara, or others – is nowhere in evidence, nor is the predominance
of urban, sedentary traders.
Four brief examples, all of which will be drawn, like the biographies of
ʿAbdullāh and Kalbay, from the autumn of 1852, will give some further
idea of how these slaves changed hands beyond the urban, sedentary
sphere: (1) A Turkmen whose name is recorded as Kul-Murat Sagdiev
was twenty-five years old when he was interviewed by border officials.
After the death of his parents, he passed into the care of his brother, who
quickly sold him to a Kazakh for a camel and two horses. This owner
surrendered him to the Russian authorities, and Sagdiev requested of them
that he be sent back to his original homeland.30
(2) An Afghan, whose name was illegibly recorded, was forty-five
years old when he was interviewed by border officials. He did not
know the details of his original captivity, only that Turkmens sold
him to a man named Khuday-Berdi for sixty tillas, and that this man
resold him, when he was four years old, to a Kazakh, for the same
amount. He lived with this owner for the next thirty-seven years. He
was married to a Kazakh of the Bayulı̈ clan, and he had three sons,
two daughters, and his own kibitka (yurt) among the Kazakhs. His
owner presented him to the Russian authorities, who formally freed
him, but he requested to be sent back to live with his present family
among the Kazakhs.31

30 31
TsGAKaz 4.1.3646, f. 47b. TsGAKaz 4.1.3646, ff. 47b–48a.
70 Beyond the Bazaars

(3) A Karakalpak whose name is recorded as Mamandat Imamaliev was

thirty years old when he was interviewed by border officials. After the
death of his parents, he likewise passed into the care of his brother, who
sold him for 100 sheep to a Kazakh, with whom he “traveled” for ten years.
At the time of interview, he had a wife and a son among the Kazakhs. His
owner surrendered him to the Russian authorities, who granted him his
freedom, but he requested that he be returned to his home among the
Kazakhs.32
(4) An Afghan whose name is recorded as Midman was thirty-four years
old when he was interviewed by border officials. He relates that he was
originally from Herat. Nineteen years prior, he had been tending to some
cattle when he was taken captive by Turkmens. These Turkmens carried
him along with them for the next two years before finally selling him to
a Kazakh for forty [?] sheep. He lived with this owner for the next fourteen
years before being presented by that same owner to the Russian autho-
rities. Midman did not wish to be returned to his original homeland, and
requested that he be sent back among the Kazakhs.33
Despite the scarce mention of Khiva and Bukhara in these sources, it
may be wise to assume that those sales involving sedentary Khivans and
Bukharans took place in or near a town, for the simple fact that townsmen
did not often venture out into the hinterlands unless joining a caravan. But
in those cases when Kazakhs sold slaves to other Kazakhs (for example), it
is equally likely that these sales took place in the steppe. Transport to
a town prior to sale is never mentioned in such cases, nor would it be
necessary: not only did neighbors trade with one-another beyond the
bazaars, but it turns out that the bazaars themselves – and urban spaces
more generally – were not as essential to the slave trade as they were
elsewhere in the Muslim world. As we shall see, this hypothesis is borne
out by the available descriptions of urban slave-trading in Central Asia.

THE SLAVE TRADE OBSERVED

Most of the eyewitness reportage on urban slave-trading in the cities of Khiva

and Bukhara concern the latter, given the relative inaccessibility of Khiva
(and Khwarazm as a whole) for European travelers. Notorious as it was as
a hub of the slave trade, the city of Khiva remained fundamentally obscure to
most Western explorers and diplomats well into the mid-nineteenth century.

32 33
TsGAKaz 4.1.3646, f. 47a. TsGAKaz 4.1.3646, ff. 48a–48b.
 The Slave Trade Observed 71

A powerful indication of this obscurity comes in the form of an annotated

map of the town that was published in the Russian journal Syn otechestva in
1842. Here, this map, which in its sparseness resembles a Piet Mondrian
painting sans color, is presented on the pretext that it should help provide
some military knowledge in support of Russian expeditions, since the
Russians knew so little of the layout of the town. The problem – such as it
is – is that the map itself was drafted more than a hundred years prior, not
long before Nādir Shah’s troops caused significant damage to the city.
The idea that this map from 1740 should offer valuable military intelligence
to the armies of 1842 may lead us to suspect that few Russians had, in the
intervening years, become intimate with the city (except under circumstances
of captivity), and that fewer still would have personally observed slave-
dealing there.34
The earliest eyewitness account which I have yet seen of an actual
Khivan slave-bazaar is that of Murav’ev, published in 1822. His descrip-
tion is very concise indeed, and it is not clear if he is describing a scene he
saw for himself. Of Iranian captives coming to Khiva, he writes,
“On arrival at Khiva their owner sets himself down with them in the
market, and purchasers surround him, inspecting and examining the
poor wretches, and haggling about their prices, as if they were buying
horses. Sometimes the Turcomans kidnap them out of Khiva and take them
back to their parents in Persia, who are often able to pay them handsomely.
During my stay in Khiva several such batches were brought into the
market, sold, and taken off to the villages.”35 Though Murav’ev had
much to say about the trade in slaves and their experience of captivity,
he had nothing more to say about the spaces where they were sold. As far as
I have seen, there would not be anything like another eyewitness report on
a Khivan slave-bazaar for nearly four decades thereafter.
The next account I have found – and perhaps the first which truly
resembles eyewitness reportage – is by N. Zalesov, from his 1859 “Letter
from Khiva.” The passage is striking for its luridness; for our purposes, it is
still more striking for the modest scale of the bazaar itself, and its scant
offerings:
Not far from the court there is a small pool, bounded on all sides by high buildings,
among which a narrow passage serves as the only throughway. Arriving at this pool,

34
“Khiva za sto let nazad,” Syn otechestva 1 (1842), 33–39. Many Tatars and Bashkirs from
Russian dominions had been there, either as residents or traders, but I am not aware of any
substantial written records of Khiva by nineteenth-century Tatar or Bashkir observers.
35
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman County, 58.
72 Beyond the Bazaars

you will find yourself in a slave-market, which is mostly supplied with Persian
captives, male and female, recently delivered here by Turkmens, though the trade
for this commodity decreased significantly during the Turkmen revolt, and
throughout the period of our stay in Khiva, we saw in the market only an old
women of around 60 years old, two boys between 9 and 12, and a lovely girl who
was about 14 but already perfectly formed. All of these unfortunates sat in the
aforesaid lane leading to the pool, which was fenced off with sticks. With the arrival
of a purchaser for a male or female slave, [this purchaser] would be conducted to an
inner [part of] the market, where, in specially-arranged rooms, they would undress
and examine the goods in every detail. Among the male [slaves], attention is paid to
the firmness of their muscles; among the women, to their beauty and litheness of
figure. When we examined the market, the girl for sale looked at us coquettishly
and it seemed that she wanted very much to strike the fancy of the young “Urus”
[Russian] rather than some old Uzbek. The girl was assigned a price of 30 tilla (60
silver rubles), but the acquisition of her – as with the other slaves – was permitted
only for a Muslim.36

Another decade would evidently pass before any traveler offered

further observations about the specifics of a Khivan slave-bazaar, and we
have reason to doubt that the traveler in question – Arminius Vambery –
actually saw the market with his own eyes. While listing the various
bazaars of Khiva in his Travels in Central Asia, Vambery writes: “I must
also class amongst the bazaars the Kitchik Kervanserai, where the slaves
brought by the Teke and the Yomuts are exposed for sale. But for this
article of business Khiva itself could not exist, as the culture of the land is
entirely in the hands of the slaves.”37 This is all Vambery has to say about
the bazaar, though he elsewhere dedicated ample space to discussing the
plight of Iranian captives in the region. Indeed, he is one of our most
impassioned commentators on the subject. Had he seen the bazaar for
himself, we can be sure he would have provided more detail than this. His
comment is helpful nevertheless, as it identifies the site of sales not as
a typical bazaar, but rather as a caravanserai – a hint, perhaps, that slave-
trading was the sort of thing which tended to take place in venues with
interior rooms and out-of-the-way dwelling spaces.
Amazingly, this appears to be the full extent of our eyewitness reportage
on Khiva’s slave-bazaars. In all, firsthand evidence regarding the nature of
those market-spaces is inconclusive – despite a general sense of certainty
among sources (especially those who never saw the bazaar) that, as
V. A. Tugan-Mirza-Baranovskii put it, “Until 1873, the main locale for

36
N. Zalesov, “Pis’mo iz Khivy,” Voennyi sbornik 1 (1859), 285.
37
Arminius Vambery, Travels in Central Asia (New York: Harper & Bros., 1865), 380.
 The Slave Trade Observed 73

the sale of slaves was in Khiva, in whose markets one could always meet
with Tekes trafficking in this commodity.”38
We are more fortunate when it comes to eyewitness reportage on the
slave-bazaar of Bukhara. One of the earliest accounts of the bazaar,
offered by the English explorer Alexander Burnes, gives a wealth of
detail:
I took an early opportunity of seeing the slave-bazar of Bokhara, which is held
every Saturday morning. The Uzbeks manage all their affairs by means of slaves,
who are chiefly brought from Persia by the Toorkmuns. Here these poor wretches
are exposed for sale, and occupy thirty or forty stalls, where they are examined
like cattle, only with this difference, that they are able to give an account of
themselves viva voce. On the morning I visited the bazar, there were only six
unfortunate beings, and I witnessed the manner in which they are disposed of.
They are first interrogated regarding their parentage and capture, and if they are
Mahommedans, that is, Soonees. The question is put in that form, for the Uzbeks
do not consider a Shiah to be a true believer; with them, as with the primitive
Christians, a sectary- is more odious than an unbeliever. After the intended
purchaser is satisfied of the slave being an infidel (kaffir), he examines his body,
particularly noting if he be free from leprosy, so common in Toorkistan, and then
proceeds to bargain for his price. Three of the Persian boys were for sale at thirty
tillas of gold apiece; and it was surprising to see how contented the poor fellows
sat under their lot. I heard one of them telling how he had been seized south of
Meshid, while tending his flocks. Another, who overheard a conversation among
the by-standers, regarding the scarcity of slaves that season, stated, that a great
number had been taken. His companion said with some feeling, “You and I only
think so, because of our own misfortune; but these people must know better.”
There was one unfortunate girl, who had been long in service, and was now
exposed for sale by her master, because of his poverty. I felt certain that many
a tear had been shed in the court where I surveyed the scene; but I was assured
from every quarter that slaves are kindly treated . . .39

In this account, it sounds as if the bazaar in question was capable of

accommodating a fair amount of merchandise, with thirty or forty stalls
available. But again, as with the extant descriptions of the Khivan markets,
our eyewitness saw just a few captives for sale. We see not the faintest
glimmer of the kinds of teeming slave-bazaars for which Istanbul (for
example) was known. As with Zalesov’s visit to Khiva, it appears that
Burnes visited Bukhara during a low point in the trade, a season when
slaves for sale were generally scarce.

38
V. A. Tugan-Mirza-Baranovskii, Russkie v Akhal-Teke (St. Pb, 1881), 71.
39
Alexander Burnes, Travels into Bokhara, Vol. 2 (London: John Murray, 1839), 241–242.
74 Beyond the Bazaars

Another traveler of the early 1830s – P. I. Demaisons, sent to Bukhara

to secure the release of Russian captives – offers some further information
about the structure of a Bukharan slave-bazaar, though there is no indica-
tion that he saw one for himself. He writes: “In the dirty yard by the
Registan there is a slave-market. It is open at dawn on Mondays and
Thursdays, usually [staying open] for no more than three hours. Here
they merely exhibit the slaves, and the commercial transactions are almost
always concluded in the caravanserais, where those who come to Bukhara
to sell these unfortunates reside.”40 Demaisons later identifies the cara-
vanserai by name: the Sarāy-i pā-yi astāna (literally, “Palace at the Foot of
the Threshhold,” or simply “Gateside Palace”).41
Concise as it is, the information Demaisons offers here is significant:
slaves, he tells us, were not actually sold at the bazaar, which served merely
as a showroom. The fact that much of the trade took place in peripheral
spaces, beyond the public eye, is likewise confirmed by Vambery, who
writes:
The sale [of slaves] takes place either in the dealers’ magazines, or in some
market-place outside the town, to which place the goods are removed some
days previous. The most important depots are to be found in the Khanat of
Khiva, first of all at the capital, then in Hezaresp, in Gazavat, in Giirlen, and in
Kohne. Besides these, every place of any pretensions has a retail dealer, who is in
connection with the large wholesale dealers, or sells goods on commission.
In Bokhara is to be mentioned first of all Karakul, and next the capital; besides
these, Karshi and Tchihardjuy.42

The cumulative impression here is of a trade which was perhaps deliberate

in its remove from the public eye. The general absence of slave traders
from the main bazaars is nowhere more palpable than in the account of
Mohan Lal, the brilliant Kashmiri munshı̄ who served as an aid to
Alexander Burnes during his voyage to Bukhara. Lal, like Burnes and
Vambery, was repulsed by the trade in Iranian slaves, and dedicated exten-
sive commentary to the subject in his travelogue. He too interviewed
a number of Iranian slaves, and he extensively describes the circumstances
of their captivity, both in specific and general terms. He was also

40
P. I. Demezon and I. V. Vitkevich, Zapiski o Bukharskom khanstve, N. A. Khalfin ed.
(Moscow, 1983), 57. (The French-born Demaisons and the Polish-born Witkiewicz pub-
lished their travel accounts in Russian, and their reportage can thus be found under their
“Russified” names, Demezon and Vitkevich.)
41
Zapiski o Bukharskom khanstve, 59. This information was confirmed in the account by
Demaisons’ contemporary, Witkiewicz: Zapiski o Bukharskom khanstve, 101.
42
Arminius Vambery, Sketches of Central Asia (London: W.H. Allen & Co., 1868), 217.
 The Slave Trade Observed 75

a connoisseur of Central Asian bazaars: in the course of his travels, Lal

visited bazaars in nearly every major town he passed through; he com-
ments on the bazaars of Charjui, Mashhad, Bala Khayaban, Turbat, Herat,
Shikarpur, Ghazni, Jalalabad, Balahisar, Kabul, Kulum, Balkh, Karasan,
and, of course, Bukhara. But for all his time in the bazaars, he reports only
a single eyewitness experience of slave-dealing, which he observed in
Qarshi, a town identified by Vambery as one of the major slave “depots”
of the region. Lal writes:
On my return from the bazar, I asked my companion to shew me the house of
a slave-dealer; so I was conducted through numerous hot streets, and after a short
walk, I got into the caravansarae where the merchant resided. He received me with
courtesy and sent for three women from the room next to his own. They sat
unveiled, and their master asked me which of the three I liked the best.
I pretended to select the younger one; she had regular features and most agreeable
manners, her stature was elegant, and her personal attractions great. On my
choosing her, the others retired to their lodgings, and she followed them, but sat
in a separate room guarded by an old slave. The merchant told me to go to her,
speak to and content her. After a good deal of conversation, she felt pleased with
my choice; but told me to swear not to sell her again. She was thirteen years of age,
and an inhabitant of Chatrar, a place near Badakhshan. She said that she belonged
to a large family, and had been carried off by the ruler of the country, who reduced
her to slavery. Her eyes filled with tears, and she asked me to release her soon from
the hands of the oppressive Uzbeg. As my object was only to examine the feelings of
the slave-dealer, and also to gratify my curiosity, and not to purchase her, I came
back to my camp without bidding farewell to the merchant.43

In this account too, we find the trade based not at the bazaars, but in the
chambers of a caravanserai, seemingly well out of the public eye. Again, the
scale of the trade is strikingly small: here, only three women for sale.
One might expect to gain further details on Khivan and Bukharan slave-
markets from freed Russian slaves, many of whose biographies were pre-
served either in the Russian press or in Russian travelogues. But here too,
eyewitness reportage is scarce, and by the early 1840s the trade in captive
Russians had subsided so much that the Russian officer I. F. Blaramberg
wrote of it in his journals as a distant memory: “August 21st, [1843]. I went
to Orsk. Upon arrival at that place, I visited the exchange yard, the
hospital, and the Cossack settlements, and I met Major Lobov, who was
for many years a slave in Bukhara. When he was just a young man, he was
abducted by the Kirghiz and sold into slavery. He told me horrifying details
43
Mohan Lal, Travels in the Panjab, Afghanistan, and Turkistan, to Balk, Bokhara, and Herat;
and a Visit to Great Britain and Germany (London: W.H. Allen, 1846), 122.
76 Beyond the Bazaars

about the treatment of Russian slaves in Bukhara and Khiva in those days.
Thank God those days are long gone!”44
In the 1860s and 1870s, the Russian conquests further reduced the
visibility of slave-trading. Captain L. F. Kostenko, reporting on the Russian
mission to Bukhara in 1870, evidently learned that the slave-market there
was no longer in existence: “There had previously in Bukhara been a slave-
bazaar, in which male and female Persians were sold like beasts of burden. But
with the Russian occupation of Samarkand, when many slaves fled from
Bukhara to this city, where they were freed by the Russian government, the
slave market in Bukhara has been discontinued, owing to bidders’ fears that
whatever goods they purchase can escape.”45 Emil Schmidt reported
a similar state of affairs in Khiva, where slavery was formally abolished in
1873: “Latterly . . . the slave trade had considerably diminished, and
a thousand individuals per annum were no longer brought to the Khivan
markets.”46 The open trade in slaves, then, which had evidently never been
especially visible to foreign travelers in the region, was by the late nineteenth
century assured to remain mostly out of sight.

THE SLAVE TRADE CONTINUES

According to some Russian officials as well as foreign observers, the con-

quest of Khiva in 1873, which resulted in a proclamation of abolition
jointly prepared by General Kaufman and the Khivan khan, signaled not
only the emancipation of Khwarazm’s slaves but the effective end of the
region’s slave trade. In fact, the reality was not so simple.
First, we must observe that, although the Russian conquest of Bukhara
in the late 1860s had resulted in a number of Bukharan concessions
rendered in ambitious treaties, an order of general emancipation was not
among them. It was only in the autumn of 1873 – some five years after
Bukhara’s conquest – that, by Russian demand, the universal emancipation
of slavery became a law in the territory. The enforcement of the law
involved a striking concession on the part of the Russian Empire, however:
General Kaufman, the architect of the treaty, allowed the Bukharan Amı̄r
an entire decade to provide complete emancipation, during which time all

44
I. F. Blaramberg, Vospominaniia (Moscow: Glavnaia redaktsiia vostochnoi literatury;
Tsentral’naia Aziia istochnikakh i materialakh XIX-nachala XX veka, 1978), 252.
45
Kostenko, Puteshestvie v Bukharu russkoi missii v 1870 godu (St. Pb, 1871), 94.
46
Schmidt, The Russian Expedition to Khiva in 1873, trans. P. Mosa (Calcutta: Foreign
Department Press, 1876), 122.
 The Slave Trade Continues 77

slaves would stay with their present masters. Attempts to escape to free-
dom were punishable by death.47 In all, we find here a remarkably hesitant
approach to emancipation for an “abolitionist” Empire.
Moreover, while it was the liberation of specifically Russian slaves that,
as Morrison observes, was frequently evoked to justify military expansion,
multiple eyewitness accounts confirm that Russian slaves were already
quite scarce in the khanates by the mid-nineteenth century. Jan Prosper
Witkiewicz wrote that “There are not many Russians in the Khanate;
perhaps one could count up to 50 of them, and these ones are elderly,
imported some time ago.” Some twenty of them, he estimated, belonged to
the Khan.48 Alexander Burnes, who likewise visited the khanate in the
1830s, put the number of Russian slaves at not more than 130.49
The Russian officer L. F. Kostenko, reporting on Bukhara circa 1870,
wrote: “Aside from fugitives, there are still Russian captives in Bukhara,
but now they are not many – two or three people. Previously there had
been many more. In 1869 the Amı̄r sent them to Charjui where they all
died, due to the swampy climate and foul accommodations.”50 As for
Khwarazm, as Kaufman and his forces approached the city, the Khan
made a desperate attempt to placate him and stave off invasion by round-
ing up and sending to the General what he claimed to be all of the Russian
slaves that remained in the khanate. They totaled twenty-one people
in all.51
Notwithstanding the scarcity of Russian slaves to liberate, popular
opinion that conquest was necessary to end Russian enslavement in the
khanates was rooted in a widespread feeling in Russia of perennial victim-
hood on the frontiers. This feeling manifested itself, as Bruce Grant has
shown, in a vast popular literature, consisting of everything from short
stories and novels to popular songs and ballads, all concerning Russian
captivity in the East, generally among Muslims.52 To the extent that these
captivity narratives had historical roots, these roots might be found in the
long, traumatic period during which Russians were regularly captured and

47
Seymour Becker, Russia’s Protectorates in Central Asia: Bukhara and Khiva, 1865–1924
(New York: Routledge, 2004 [1968]), 67–68.
48
Ia. P. Vitkevich, Zapiski o Bukharskom khanstve (Moscow: Nauka, 1983), 115.
49
Alexander Burnes, Travels into Bokhara, Vol. 2 (London: Carey & Hart, 1835), 115.
50
L. F. Kostenko, Puteshestvie v Bukharu russkoi missiiv 1870 godu (St. Pb, 1871), 107.
51
MS IVAN Uz No. 12581, f. 42a; cf. also MacGahan, Campaigning on the Oxus, and the
Fall of Khiva (London: Sampson Low, 1874), 20.
52
Bruce Grant, The Captive and the Gift: Cultural Histories of Sovereignty in Russia and the
Caucasus (Ithaca, NY: Cornell University Press, 2009); see also Kurtynova-D’Herlugnan,
The Tsar’s Abolitionists, 37–72.
78 Beyond the Bazaars

carried off by the soldiers and raiders of the Crimean Khanate. Up to

200,000 Muscovites were estimated to have been taken into slavery by
Crimean Tatar and Noghay captors between 1600 and 1650.53 The total
number of Ukrainians, Russians, and Poles captured by Tatars between
1468 and 1694 has been estimated at 1,750,000.54 As recently as the
eighteenth century, Russians were frequently captured along the steppe
frontier and the eastern Caspian shores; many were ransomed, but some
were sold into slavery.
By the mid-nineteenth century, the age of Russian enslavement had
mostly passed. Meanwhile, as we shall see, the trade in Iranian slaves,
who constituted the vast majority of slaves held in the khanates, seems to
have continued in some quarters even after the fall of Khiva. Nevertheless,
Russian military presence in the sedentary heartland of Central Asia does
appear to have occasioned some changes in the volume of the slave trade.
One significant change evidently concerned the demographics of the high-
est registers of power in Bukhara, as many functions of state were carried
out by slaves. P. Shubinskii described the alleged inconveniences the
Bukharan Amı̄r ‘Abd al-Ahad (r. 1885–1910) occasioned for himself in
˙
proposing general abolition:
Sayyid ‘Abd al-Ahad passed this measure, and brought upon himself some very
significant challenges, since a considerable portion of the Bukharan military and
almost all of the minor court officials and palace servants were slaves. Receiving
their freedom, all of these people hastened to return to their homeland, and in their
place unfamiliar salaried people had to be recruited, the maintenance of whom
brought about substantial new expenditures.55

It is unlikely that “all” of the slaves who had been serving as court officials –
however minor – would have departed upon receiving their freedom, and
we have no further evidence, so far as I have seen, to support the claim.
Nevertheless, it is plausible that manumission, if it had really been effec-
tive, would have had a cumulatively dramatic effect on the military, on the
government, and on the domestic sphere among the nobility. But a trade
that occurred largely in out-of-the-way spaces – rural nomadic settlements,
backrooms in caravanserais – must have been difficult to abolish while
demand for slaves still existed. Indeed, according to N. P. Stremoukhov,
53
Brian L. Davies, Warfare, State and Society on the Black Sea Steppe, 1500–1700 (New York:
Routledge, 2007), 25.
54
Alan W. Fisher, “Muscovy and the Black Sea Slave Trade,” Canadian-American Slavic
Studies 6:4 (1972), 575–94; see also Clarence-Smith, Islam and the Abolition of Slavery,
13.
55
Shubinskii, “Ocherki bukhary,” Istoricheskii vestnik 7 (1892), 125.
 The Slave Trade Continues 79

who traveled to Bukhara in 1873, demand still existed, and it was still
satisfied:
Officially, the trade in slaves was banned in Bukhara by the command of the Emir,
and the caravanserai where slaves used to be sold is now permanently closed.
Violators of this command are subject to punishment, a fine of 1000 tenga, and
six months in prison. Even so, a large number of people (mostly merchants) engage
secretly, in their own homes, in the buying and selling of slaves. As before, the main
suppliers of slaves can be identified as Turkmens, whose raids have not stopped.
Most of those who fall into captivity are Persians (I have already said that there are
very many Persians among the troops of the Bukharan Emir, and these are all
unfortunate ones who had been captured by Turkmens). But especially vast is the
traffic in women: even those from within our own borders are taken by Bukharans
[sic], by means of deception, using all sorts of tricks; and they are secretly sold in
Bukhara, and this is done so discretely that there is absolutely no way to keep track
of them. The traffic in women is conducted predominantly by Tatars.56

Given the occult nature of the practices Stremoukhov describes, it is

unfortunate that he provides no hint as to how he gathered the foregoing
information. He testifies, in any event, that the trade continued not only
among merchants, but at the highest levels of state:
Although a formal ban exists, it exists only nominally, as the very Emir himself
patronizes this despicable trade. Two things impel him in this respect: first, this
trade provides him with recruits for the army; and second, by these means he can
acquire for himself young and beautiful women, and get rid of those of whom he’s
grown tired. For the purchase and sale of slaves, Muzaffar maintains certain secret
agents, who he pays very handsomely.57

Along with these bold claims, Stremoukhov provides one anecdote con-
cerning the continuance of the trade which was drawn from his own
personal experience:
On the morning of the 6th of June, some poor old man dressed in rags . . . ran to my
tent and began asking for my protection. He turned out to be an Uzbek, and
complained that two of his daughters, despite the publicized prohibition against
selling slaves, had been forcibly captured and sold to a wealthy Bukharan who,
paying no attention to the order of the kadis and the command of the Emir, did not
want to return them to freedom at any cost.58

Elsewhere in Central Asia, travelers, military personnel, and other eye-

witnesses were turning in similar reports. In Badakhshan, where Shiʿites

56
N. P. Stremoukhov, “Poezdka v Bukharu,” Russkii vestnik, 6 (1876), 690.
57
Stremoukhov, “Poezdka v Bukharu,” 690. 58 Stremoukhov, “Poezdka v Bukharu,” 655.
80 Beyond the Bazaars

had long been taken into captivity, and where groups of them had long
been sent off into slavery as tribute to local Afghan rulers, the announce-
ment of abolition from Kabul evidently had mixed effects. According to
Thomas Edward Gordon, the more visible forms of trafficking had ceased,
but the trade went on:
Slavery still continues to be the curse of many of the Shiah states round about
Badakhshan. Notwithstanding its prohibition by the Amir of Kabul, the disgraceful
trade in human beings, with all its attendant crime and cruelty, still flourishes . . .
The open slave-market certainly is closed, but beyond that nothing seemingly is
done to suppress the shameful and horrible traffic, which is otherwise carried on as
briskly as ever. The Affghan occupation of Badakhshan has had the good effect of
abolishing the tribute in slaves which used to be demanded and enforced by the
ruling Sunni Mirs from their feudatories with subjects professing the heretical
Shiah creed. Futteh Ali Shah of Wakhan told me that the tribute he paid
in September 1873 was the first ever given of which slaves did not form a part.59

It would not be long before a traveler attempted to conduct a more

intensive and personal investigation into the persistence of the slave
trade. The traveler in question was the American diplomat Eugene
Schuyler, who visited Bukhara in his Central Asian travels, having departed
St. Petersburg for the region in March of 1873. He made his first attempt to
observe the trade when he arrived at Qarshi, the town whose slave-dealing
had been noted by Vambery and Mohan Lal. Here, he saw nothing:
“I asked to see the slave market and was shown the sarai, but saw no slaves,
though I was told that the next day (Tuesday), being bazaar day, some
would probably be brought in for sale.”60 He apparently never had
a chance to return the next day, but in the town of Bukhara he would
resume his search. His curiosity was piqued by a disparity between the
claims of some Russian officials and those of local merchants and others he
had met in the region. He writes:
In visiting Bukhara I was especially anxious to learn something about the slave
trade, and if possible to see for myself what was going on. The Russian authorities
had expressed their desire that the slave trade should cease, and had been of course
informed by the Bukharians that it had long since come to an end. Nearly all the
Russian officials who had been in Bukhara had been deceived in this respect, and an
official report had been made to General Kaufmann that the slave trade no longer

59
Edward Gordon, The Roof of the World: Being a Narrative of a Journey over the High
Plateau of Tibet to the Russian frontier and the Oxus Sources on the Pamir (Edinburgh:
Edmonston and Douglas, 1876), 147.
60
Schuyler, Turkistan: Notes of a Journey in Russian Turkistan, Khokand, Bukhara, and
Kuldja (New York: Scribner, Armstrong, 1876), 79.
 The Slave Trade Continues 81

existed there. Merchants, however, told me that they had frequently seen public
sales of slaves in the bazaar, and my interpreter said that, on two visits to Bukhara
during the preceding year, he had seen the slave market filled with Persians who
were dying of cholera and hunger, for, in the panic caused by the epidemic, they had
not been fed; and the Agent of the Ministry of Finance had been able, in the spring
of 1872, to see slaves publicly exposed for sale. He had made a report of this, but
the matter had been passed over without notice by the Russian authorities.61

Schuyler knew that he could not reveal his intentions to the Bukharan
officials around him, since they would be likely to deny the existence of the
trade. Seizing an opportune moment, he asked someone less beholden to
the Bukharan government whether he might be shown a slave-market, and
he was promptly taken to “a large sarai,” where he proceeded up the stairs
into a gallery and found “several rooms, some of which were locked, and
a number of slaves – two little girls of about four years old, two or three
boys of different ages, and a number of old men – all Persians.” Schuyler
reports that there were no women in sight, as both young women and old –
excepting, apparently, the very young girls – were purchased immediately
upon their arrival in Bukhara. The slaves were shown off by “an old
Turcoman, who acted as a broker,” and who explained to Schuyler that
“the market was rather dull just then, but that a large caravan would
probably arrive in the course of a few days.” At this point, curious to
learn about how such sales were conducted, Schuyler began bargaining for
a “lively looking lad of fifteen” who, he was told, had been taken captive
near Astarabad five months previously.62
The ensuing negotiations drew a small crowd: “I was immediately asked
to take a seat on a mat,” Schuyler writes, “and the room soon filled with
people, all of whom seemed to take much more interest in the sale than did
the boy himself, who did not understand what was being said, the con-
versation being in Turki. The first price asked was more than 1,000 tengas,
which I gradually reduced to 850 tengas; the seller constantly dilating on
the good points of the boy, what an excellent jigit he would make, and so
on, the bystanders joining in on one side or the other.”63 At first, Schuyler
vacillates on the price, and searches the building’s other rooms for other
61
Schuyler, Turkistan, 100–101. 62 Schuyler, Turkistan, 100–101.
63
The comparatively astronomical price of the slave will be noticed: unless we can assume
severe currency depreciation in 1870s Bukhara – the likes of which I have not heard – the
price of 1000 tengas for a slave is roughly thirty-three times greater than the historical
market norm. Perhaps the dealer was relying on Schuyler’s unfamiliarity with the usual
price-range. Schuyler shows at least some awareness of the odd pricing, however:
“I thought that 850 tengas was too much to pay for the lad, especially as I had no desire
to buy him; at the same time, the wistful looks of the boy, who seemed very anxious to be
82 Beyond the Bazaars

slaves. When he and his guide find none, he finally agrees to buy the boy,
evidently for 850 tenga, planning to take him to Russia and, eventually, to
secure his safe return to Iran. At this point he is confronted with “a broker,
a swarthy, thick-set fellow from Kara-kul, a well-known slave dealer,” who
tells him that a rival bidder had agreed to pay 900 tenga for the boy, along
with two “gowns” on top of that. An additional complication, according to
this “broker,” was that the boy’s real owner was not present, so that he
himself had no right to part with the boy on his owner’s behalf – the result
being that Schuyler agreed to pay a portion of the price to the man as an
advance in order to secure the later delivery of the boy, whom he failed to
take with him.
In short, Schuyler was swindled. He never received the boy, nor, appar-
ently, did he ever get back the money he had already paid for him. He sent
an acquaintance back to the place of sale two days later in order to seek out
the boy, but this man found only two very young girls and one other boy for
sale.64 Later, Schuyler mentions his adventure to a Bukharan official, who
insists that the slave trade had indeed been ended, and that these few items
of sale were nothing more than remnants. “I told him,” Schuyler writes,
“that I did not doubt his words, although, at the same time, it appeared
very strange to me in this case, that when a caravan of sixty slaves had
arrived at Bukhara the night before, at nine o’clock, he himself had given
order that it should remain outside the Kara-kul gate, in order that I should
not see it.” Schuyler did not see it, nor would he ever see it, but by his own
account the official did not deny the accusation.65
Undeterred, Schuyler resolves to buy another slave. He sends his
wagon-driver, Pulat, to search the town for one to buy. Pulat spends
the day searching, and returns the same evening with the news that he
had found a boy for sale, about seven or eight years old, offered for 700
tenga “and a good gown.” Schuyler buys him, and he turns out to be “very
small and feeble, although intelligent,” a Persian from Maymana who had
been taken captive by Salar Turkmens some three years earlier. According
to Schuyler, the boy had little recollection of his parents, and seemed not to
know his own name, in light of which he “took the liberty, which in these
countries is always allowed on the purchase of a slave, and named him
Hussein.” He managed to retain this boy, despite the efforts of the boy’s
former owner to steal him back (fearing, after the sale, that he would be

bought, smote my conscience a little, and I asked for the refusal of him at that price, which
was given” (Schuyler, Turkistan, 102).
64
Schuyler, Turkistan, 105–106. 65 Schuyler, Turkistan, 104–105.
 The Slave Trade Continues 83

punished by officials for selling a slave to a foreigner who could expose the
ostensibly illegal trade).66 Schuyler later tells us that his purchase caused
a stir in Samarqand and Bukhara, with some notables further denying the
existence of the trade and others glad to have it exposed.67 Ultimately,
Russian authorities seem to have taken the reports of slave-trading in
Bukhara seriously: after the fall of Khiva, the Russian Court Councilor
K. V. Struve negotiated a treaty with the Amı̄r, which resulted in a new
injunction, signed in the autumn of 1873, that the slave trade be ended in
Bukharan domains. This action, according to Schuyler, was still insuffi-
cient: “Unfortunately, the Russians have always found it more easy to make
treaties in Central Asia than to enforce their observance, and I have
received information from Russians as well as from natives that since this
treaty the slave-trade has rather increased than diminished, although slaves
are no longer sold publicly in the open market, as was done when I was in
Bukhara.”68
Schuyler was perhaps the most deeply invested among foreign observers
investigating the persistence of the slave trade. There is likewise, however,
significant evidence from native Central Asian sources that confirms the
continuation of the trade after the Russian-fostered “official abolitions” of
the 1870s. When it comes to Bukhara, for example, we can observe the
persistence of slavery through the lens of Islamic law, thanks to a document
collection published by Turgun Faiziev. This collection comprises some
fifty-two documents, most of a legal nature, including everything from bills
of sale to manumission contracts and correspondences among Bukharan
notables concerning slaves. The compilation contains no fewer than seven-
teen such documents dating from the year 1886 alone.69 One document
contains a brief register of the names and jobs of slave-boys – fifteen in all –
owned by the Amı̄r, some serving in the royal household and some in the
government chambers.70 In another document, two Muslim notables write
to the Amı̄r notifying him that one Sardar Ishāqjān, residing along with
“several households” of people in Qubadiyan province, remains in posses-
sion of three male and seven female slaves, despite the fact that these slaves
had been freed by official decree. Evidently, the slaves had come before

66
Schuyler, Turkistan, 108–109. Later, he brought “Hussein” back to Russia, where he saw to
his education.
67
Schuyler, Turkistan, 310. 68 Schuyler, Turkistan, 310–311.
69
T. Faiziev, Buxoro feodal jamiyatida qullardan foydalanishga doir hujjatlar (XIX asr)
(Tashkent: Fan, 1990), 115–127, documents no. 26–42.
70
Faiziev, Buxoro feodal jamiyatida qullardan foydalanishga doir hujjatlar (XIX asr),
117–118, document no. 30.
84 Beyond the Bazaars

a Bukharan general and a Russian official and explained that their masters
had agreed to set them free, but that these masters were declining to issue
them manumission documents in order to make their freedom legitimate
within the context of Islamic law. The masters were duly ordered to issue
the necessary paperwork.71 The implications are remarkable: the official
decree was not enough; for these slaves, and perhaps also for their owners,
in order for manumission to be valid, it needed to issue – in the traditional
fashion – from legitimate Islamic juridical authorities.72
Other documents preserve petitions requesting that the Amı̄r validate
and ensure the release of slaves, in keeping with his general manumission
decree. There are two ways of understanding such petitions: on the one
hand, the fact that denizens of Bukhara felt that they could affect change by
appealing to the highest authority might hint at a degree of action on the
part of the Amı̄r and other high officials when it came to enforcing
manumission; on the other hand, the fact that these same residents felt
they needed to address their petitions to the royal court indicates the
failure of more “local” organs of change, and for each petition naming
a particular slave, we must wonder how many other slaves went unmen-
tioned by any appeal.

CONCLUSION

Acknowledging that the slave trade appears to have persisted after the
period of “official abolition,” it is nevertheless difficult to get a sense of the
scale of the commerce. As we have seen, even Schuyler – who, among our
eyewitnesses, sought it out most intensively – saw no more than a few
slaves being sold, and, like many others before him, he was informed that
he had chosen an inauspicious day if he had wished to see many slaves for
sale. Better that he should come back the next day; better that he should
have come earlier. Not a single eye-witness, so far as I know, reported
seeing a caravanserai or marketplace overflowing with victims of the trade.
And yet none denied that the region itself was flush with slaves: However

71
Faiziev, Buxoro feodal jamiyatida qullardan foydalanishga doir hujjatlar (XIX asr), 127
(document no. 43).
72
Similarly, concerning the French-sponsored abolition of slavery in the Comoro Islands in
1904, Gill Shepherd writes that “slaves were not freed, in Comorian eyes, by European
emancipation decrees, but only by the individual action of their masters.” Shepherd,
“The Comorians and the East African Slave Trade,” in James L. Watson, ed., Asian and
African Systems of Slavery (Oxford: Basil Blackwell, 1980), 96; see also Clarence-Smith,
Islam and the Abolition of Slavery, 147.
 Conclusion 85

invisible their sale and exchange, they were visible at all levels of society,
from the royal court to the nobles’ estates to the tents of nomads.
In this disparity between the low visibility of the trade and the
evident wealth of slaves, we can ascertain something of why the com-
merce would have been so difficult to stop. First, it appears that slave-
selling did not tend to take place openly, in public markets and bazaars.
Slaves were sometimes exhibited in such spaces, but it appears to have
been more common for sales to take place in out-of-the-way spaces: in
backrooms, or in the chambers of caravanserais. Granted, a shipment
of sixty slaves may indeed have been waiting beyond the gates as
Schuyler did his shopping, but the balance of our evidence points to
the conclusion that slave-dealing took place predominantly in caravan-
serais, at least some of which were specially appointed for this sort of
business.
If the trade in slaves was a business for caravans and their members –
a proposition which makes intuitive sense, given the great distances the
“merchandise” had to travel between, say, Khurasan and Bukhara, and
the dangers of the road in between – then we might look to caravanser-
ais, rather than the main bazaars of major towns, as our nexuses and
points-of-sale as we attempt to recreate a “geography” of the slave trade
in Central Asia. The region, moreover, was traversed by caravan routes,
and dotted with so many caravanserais that, in the last decade, the
United Nations Educational, Scientific and Cultural Organization
(UNESCO) launched a multi-national project to inventory all of the
caravanserai structures that can still be found there.73 A map released by
UNESCO reveals, by my count, more than two dozen known caravan-
serais stretching from Khurasan across the territory of the Khanates.74
Presumably, dealers in slaves – or any other merchandise – could have
stopped at any one of these that may have been operational in the
nineteenth century, and plied their trade while in residence. This
could help explain the dispersal of slaves across the rural landscape
observed by Burnes as he traveled through a small village in the region:
“Though the village in which we were now residing could not boast of
more than twenty houses, there were yet eight Persian slaves; and these
unfortunate men appear to be distributed in like proportion throughout
the country.”75
73
Inventory of Caravanserais in Central Asia (UNESCO / Ecole d’Architecture Paris Val de
Seine EVCAU Research Team, 2004).
74
www.unesco.org/culture/dialogue/eastwest/caravan/countries.htm.
75
Burnes, Travels into Bokhara, 342.
86 Beyond the Bazaars

We may now recall the former slave Kalbay Hamzin whom we met
earlier in this chapter. He was held for some time at the Khwarazmian
village of Pars, which he describes as a two-day caravan journey from
Urgench. Kalbay never mentions having been sold at Urgench, however,
and when he was resold by his owner at Pars – to another “Khivan,” and
then to a Kazakh – we may ask ourselves whether it seems likely that each
subsequent owner needed to travel to a major town in order to buy or sell
him, when caravans of slave-dealers and other merchants were passing
through the hinterlands all the while. The more likely site of sale would
have been the caravanserai closest to where the owner lived, and only for
some owners would this have meant towns like Khiva, Urgench, and
Bukhara. All caravanserais were sites of commerce, and there is no reason
to believe that slaves were different than any other commodity in terms of
where they could be sold. Stopping the trade in slaves, in other words,
would have involved, at the very least, patrolling the caravanserais.
Beyond the caravanserais, moreover, there was slave-trading among
nomads. Kazakhs of the remote steppe evidently exchanged slaves with
other nomadic Kazakhs, and Turkmens did the same further south.
In short, the generally accepted view that the Russians ended the slave
trade by enforced decree from the towns of Khiva and Bukhara becomes
still harder to believe in light of contrary eyewitness evidence, and espe-
cially in light of native documents, dating from well beyond the 1870s,
which prove that Islamic jurists were still receiving cases concerning slaves
at that time and dealing with slavery in the familiar legal framework. When
we turn to the geography and mobility of the trade itself, we can cast
further doubt on the effectiveness of Russian-sponsored abolition. In light
of the evidence that caravanserais served as points where slaves were
trafficked, and in the absence of any evidence that the vast network of
Central Asian caravanserais was patrolled by any “abolitionist” force
(whether Russian or native), we should wonder how probable it is that
dealers would have been caught with any frequency. Finally, in light of
evidence suggesting that nomadic groups traded and sold slaves among
themselves, we can easily understand why Russian authorities had to rely
increasingly on native informants to track down offenders and free cap-
tives. While we cannot accept, for the reasons set out in this chapter, the
proposition that Russian and locally fostered abolition actually stopped
the slave trade, we cannot dismiss the numerous reports which suggest –
contrary to Schuyler’s claim – that the volume of the trade did indeed
decline. To find the reasons for this decline, however, we must look beyond
the bazaars, turning our attention from the urban centers and markets to
 Conclusion 87

those steppe and Iranian borderlands from which the supply of captives
had long been taken.
Here, we have considered some major aspects of the trade in slaves
across Central Asia. In the next chapter, we will explore the experience of
slavery in the region through the eyes of a man who spent a decade in
bondage.
 3

From Despair to Liberation

Mı̄rzā Mahmūd Taqı̄ Āshtiyānı̄’s Ten Years of Slavery
˙

When Mı̄rzā Mahmūd Taqı̄ Āshtiyānı̄, an Iranian scribe, artist, and

˙
accountant (karguzar), arrived at his family home in Tehran in 1870, his
loved ones may have thought they were looking at a ghost. Mı̄rzā Mahmūd
˙
had been away for nearly a decade. He had fallen into captivity among
a group of Sarı̈q Turkmens while accompanying the Qajar military in
a campaign against Merv in 1860–61, and had spent the next ten years
as a slave. After making his way back to freedom, he authored the richest
firsthand account of Central Asian slavery in existence.1 Having been
enslaved both as a laborer in a Turkmen desert village and as a well-
compensated servant of Bukharan elites, his decade-long ordeal offers us
a window onto the remarkable diversity of experiences possible for slaves
in the region. As we shall see, the many clever strategies Mı̄rzā Mahmūd
˙
used in guiding his own fate, whether by manipulating his owners or by
shrewdly deploying his artistic talents, likewise reveal something of the
individual initiative slaves could exert to survive their ordeals and better
their positions.
For the study of slaves’ experiences in Central Asia, there is no better
point of entry than Mı̄rzā Mahmūd’s account. Rather than breaking up the
˙
momentum of his narrative with analytical asides, this chapter will provide

1
The first half of the work has been the subject of a recent paper by Arash Khazeni and
Abbas Amanat: “The Steppe Roads of Central Asia and the Persian Captivity Narrative of
Mahmud Mirza Taqi Ashtiyani,” in Nile Green, ed., Writing Travel in Central Asian History
(Bloomington, IN: Indiana University Press, 2014), 113–133. The authors highlight several
important aspects of Mı̄rzā Mahmūd’s early ordeals, although they offer no discussion of
˙
the eventful years he spent in and around Bukhara, which constituted the majority of his
time as a slave.

88
 Mı̄rzā Mahmūd’s Story 89
˙
a detailed, uninterrupted retelling of Mı̄rzā Mahmūd’s ordeal, concluding
˙
with some thoughts on what his experiences can tell us about slavery in the
region more generally.

MĪ RZĀ MAḤM Ū D’S STORY

Soon after taking Mı̄rzā Mahmūd into captivity, the Sarı̈q Turkmens take
˙
his clothes. His proper hat is swapped for an ill-fitting hat that hangs over
his eyes, and he is given an old green shawl – the shawl of a sayyid (an
esteemed direct descendant of the Prophet Muhammad). At first he
˙
declines to wear it, on the grounds that donning the attire of a sayyid is
inappropriate for a man of a lesser lineage. For several days he attempts to
make do without any covering at all, but eventually he caves in to the
elements and takes to wearing the shawl, leading many of his fellow-
travelers to assume – notwithstanding his vehement denials – that he is
of sacred lineage.2 As he marches through the desert in his green shawl, he
has no way of knowing his ultimate destination. He knows only that slaves
were sent off to all parts, urban and rural: Bukhara, Shirgen, Urgench,
Labab, Khiva, Balkh, Qaraqul, and “other, worse places.” Day by day, 200
or 300 individuals were sent off to slavery as if they were “ugly wares and
dirty commercial goods.”3 For the time being, he would be the property of
a Sarı̈q named Khan Muhammad, who is determined to ransom him back
˙
to Iran for no less than 100 tillas, a very large sum for any slave.
˙
Khan Muhammad refuses lesser offers, and he begins tormenting Mı̄rzā
˙
Mahmūd in order to force him to correspond with Iranians in Mashhad to
˙
secure a higher ransom. Mı̄rzā Mahmūd is shackled at night with stocks
˙
(bukhaw-i pā) on his legs, weighted ring-cuffs on his hands (tās-i halqa),
˙ ˙
and a chain on his neck (zanjı̄r). The chains wound him where they dig into
his skin, and he cannot sleep due to the excess of filth and lice (chirk
o shipish). He is given no carpet or blanket to use at night, and his hat,
which blackens from the smoke of the dwelling’s wood-fire, is all he has
with which to cover himself. He is given stale barley bread to eat, bread so
hard that it cuts his mouth, though he observes that the Turkmens them-
selves partook of the same.4
2
Āshtiyānı̄, ʿIbratnāma, 27–28. To one supplicant he objected, “Do not call me a sayyid; for
both you and me it is a sin” (mārā sayyid magūyɩ̄ d ke az barā-yi man o shomā har du taqsɩ̄ r
o gunāh ast), but, despite his objections, some among the Sarı̈qs took to calling him “Sayyid
Taqi.”
3
Āshtiyānı̄, ʿIbratnāma, 28–29. (kālā-i zisht va mutā’-i kasɩ̄ f.)
˙
4
Āshtiyānı̄, ʿIbratnāma, 29–30.
90 From Despair to Liberation

Mı̄rzā Mahmūd soon begins planning his escape. He and six others pool
˙
their resources to hire a guide, figuring that it is too dangerous to try to
make their way alone to Herat, Mashhad, or Sarakhs. Mı̄rzā Mahmūd
˙
secures a loan of 14 tillas from his master, promising to pay him back 20
˙
tillas in one month’s time – a hint of the privileges that could be enjoyed
˙
when a slave convinced a master of his or her talents. Altogether, the group
of seven slaves gather over 90 tillas, and a man named Rustam agrees to
˙
guide them. At the captives’ prompting, Rustam swears on a Qur’ān and
vows to divorce his own wife (sawgand-i zan talāqı̄) if he fails to live up to
his end of the bargain. He receives 82 tillas with which to buy horses for all
˙
involved, and another 10 tillas as payment for his services. He returns with
the horses, and the group sets off toward freedom. Just a few miles down
the road, however, Rustam stops them. He asks for more money, for
reasons that Mı̄rzā Mahmūd cannot ascertain. Then Rustam disappears
˙
into the desert, leaving the group stranded. They return to their masters,
who are fortunately unaware of the attempted escape.5
A month passes, and Mı̄rzā Mahmūd’s debt to his master comes due,
˙
but – to the slave’s good fortune – Khan Muhammad is thoroughly dis-
˙
tracted: He is enraptured by a “beautiful” ten- or twelve-year-old slave boy
named Shahbāz, whom he had recently purchased. Khan Muhammad
˙
dresses the boy in such fine clothes that it is impossible to tell that he is
a slave (hālat-i ası̄rı̄ dar u hı̄ch maʿlūm nabūd). He takes the boy every-
where, and, in Mı̄rzā Mahmūd’s words, intended to “tame” him and use
˙
him for sodomy (livāt). He prepares the boy’s bedding close to his own,
˙
and sleeps beside him. Mı̄rzā Mahmūd, meanwhile, finds it difficult to
˙
sleep more than a couple of hours a night, and on one sleepless night he
hears the boy weeping. He attempts to comfort the boy, but suddenly Khan
Muhammad arrives and beats Mı̄rzā Mahmūd for bothering his beloved
˙ ˙
slave. The next morning, Shahbāz comes to Mı̄rzā Mahmūd and explains
˙
how Khan Muhammad mistreats him at night (shabhā-rā bā man ʿamal-i
˙
bad nimāyad), and that he sometimes suffers punishment when he refuses
his master’s advances.6
Mı̄rzā Mahmūd’s heart aches for Shahbāz, and he hatches a plan to
˙
liberate him from their master. He conspires to inform Khan Muhammad’s
˙
wife of her husband’s doings. One night, while Khan Muhammad is busy
˙
abusing the boy, his wife rushes in and yanks the blanket from them.
In Mı̄rzā Mahmūd’s words, she “saw his pantless ass and gave him
˙
a kick” (kūn-i bı̄-shalvār dı̄da lagadı̄ bar u zāda). She shouts at him, pulls

5
Āshtiyānı̄, ʿIbratnāma, 32–34. 6
Āshtiyānı̄, ʿIbratnāma, 36–37.
 Mı̄rzā Mahmūd’s Story 91
˙
his beard and rips his clothes. In the end, Khan Muhammad sells Shahbāz
˙
to a man from another village.7
Soon after, Mı̄rzā Mahmūd would come to the aid of a fellow slave held
˙
in the village. The slave, named Muhammad, was purchased by a wealthy
˙
shepherd named Aqqi who, despite his wealth, was so stingy that his usual
meal was nothing more than an onion and a dish of water. This tight-fisted
master had turned down multiple offers of ransom for Muhammad, and it
˙
was clear that he – like Mı̄rzā Mahmūd’s own master – was holding out
˙
indefinitely for an unrealistically high price. This inspired anger among the
people from Muhammad’s native village, and in the meantime
˙
Muhammad was made to herd sheep and work the land. One day,
˙
a Sarı̈q named Baylı̈ tells Mı̄rzā Mahmūd of his intentions to rob the stingy
˙
slave-owner. Baylı̈ was famous for his thieving, but he evidently had
disdain for the slave trade; he was also competent in shoemaking
(kaffāshı̄), and he earned his keep both as a shoemaker and as a thief.
At dawn, he robs Aqqi’s residence and carries off the goods to Herat in
order to sell them. When Baylı̈ returns, he announces to Mı̄rzā Mahmūd
˙
that he has come out 100 tillas richer than expected, and that his next plan
˙
was to free Muhammad from his master. He provides the slave with
˙
money, bread, food, and water, and Muhammad prepares to make his
8 ˙
getaway.
On the night of his escape, Muhammad meets Baylı̈ at an agreed-upon
˙
place and they ride off to Herat. It is a windy night, with dust blowing all
around, and Aqqi shouts out to Muhammad, reprimanding him for failing
˙
to close the tent-flaps. He soon realizes that his slave is gone. A month
later, Mı̄rzā Mahmūd receives a letter from Muhammad, who reports that
˙ ˙
he managed to make it safely back home to Mashhad. His successful escape
has the incidental effect of worsening Mı̄rzā Mahmūd’s own servitude,
˙
however: Khan Muhammad, paranoid of losing his valuable slave in
˙
a similar manner, keeps a stricter watch on him, rising twice each night
to check the tightness of Mı̄rzā Mahmūd’s chains.9
˙
With the coming of summer, the Turkmens spend their time working the
land, and Mı̄rzā Mahmūd is forced to lug stones and tend to the chickens.
˙
He suffers terribly from the heat and from his shackles, and finds that he
cannot endure the labor. He cries out behind Khan Muhammad’s resi-
˙
dence at night like a man possessed, and at one point he throws himself to
the ground and faints (khod-rā be zamı̄n afkānda va bı̄hūsh shoda). Khan

7
Āshtiyānı̄, ʿIbratnāma, 37–38. 8
Āshtiyānı̄, ʿIbratnāma, 39–41.
9
Āshtiyānı̄, ʿIbratnāma, 42.
92 From Despair to Liberation

Muhammad and his brother come and carry Mı̄rzā Mahmūd into the
˙ ˙
house, where he behaves as if insane, striking out at anyone who comes
near and bashing his head against the ground. His master, understandably,
laments that he had not sold Mı̄rzā Mahmūd when he had the chance.10
˙
Mı̄rzā Mahmūd soon realizes the benefits of madness: He is excused
˙
from lugging stones. He keeps up the act for twenty-three days, and word
spreads among the village children about the “crazy” slave. He frightens
them away whenever they gather to see him. He also takes to scaring Khan
Muhammad’s wife away from her meals, snatching the bread she leaves
˙
behind. With the extra sustenance and exemption from labor, Mı̄rzā
Mahmūd’s health improves, and he hatches a new plan to escape. He
˙
takes advantage of his free access to the “postal” system, by which his
master had formerly made him contact possible ransom-payers, as well as
his connections to other slaves and former-slaves he had met in the village.
One of these slaves was a Qandahari named Mullā Riżā, who had done
taqı̄ya (claiming to be a Sunni rather than a Shiʿite) in order to improve his
lot, but was still treated harshly by his master. Mullā Riżā’s captivity came
to the attention of the Afghan chieftain Dost Muhammad Khan, and he
˙
was ransomed for 44 tillas. Mı̄rzā Mahmūd contacts the liberated Mullā
˙
Riżā, and makes an unusual request: He asks him to send some arsenic
(sam al-fār) and a file (sūhān). Mullā Riżā sent him two misqāls of
arsenic.11
It would not be long before Mı̄rzā Mahmūd gets his chance to deploy
˙
the poison. One day, Khan Muhammad asks Mı̄rzā Mahmūd to prepare
˙ ˙
some tea for him. Seizing the opportunity, he poisons the tea-water with
half a misqāl of arsenic. But just as he is adding it to the water, Khan
Muhammad’s brother arrives and requests milk-tea instead. Mı̄rzā
˙
Mahmūd pours out some of the water, since milk-tea requires less of it,
˙
and then adds milk to the poisoned water remaining in the pot. Mı̄rzā
Mahmūd had no intentions of poisoning Khan Muhammad’s brother
˙ ˙
along with Khan Muhammad himself, but soon enough the situation
˙
becomes still more complicated: His master’s sister, son, brother, wife,
and one daughter all arrive to enjoy some of the freshly made tea.12
Khan Muhammad and his sister sip the poisoned tea, and they complain
˙
about its strange flavor. Khan Muhammad’s brother echoes the complaint,
˙
and spills his cup on the ground. The others, too, declare it to be bad tea,
and soon enough a change comes over them all. They go pale, grow weak,

10
Āshtiyānı̄, ʿIbratnāma, 44. 11
Āshtiyānı̄, ʿIbratnāma, 45–46.
12
Āshtiyānı̄, ʿIbratnāma, 47.
 Mı̄rzā Mahmūd’s Story 93
˙
and begin to vomit. They all complain of suffering great thirst, and they
begin to speculate that they have been poisoned.13
Khan Muhammad accuses Mı̄rzā Mahmūd of trying to kill the lot of
˙ ˙
them, and the hākim of the Turkmens is summoned along with a Herati
˙
Jewish doctor, who examines the tea and tests it with instruments.
In response to their accusations, Mı̄rzā Mahmūd objects that he had
˙
been a slave there for some two years and lacked both the money and the
means to get hold of any poison. He claims that the dirty copper teapot was
to blame for the poisoning, and he wins enough support for this hypothesis
that his life is spared. He is nevertheless beaten brutally (for good mea-
sure), and he is never again permitted near the cups and pitchers. His
master learns to fear him, and Mı̄rzā Mahmūd reaps the benefits of being
˙
feared: for two or three months, he is exempted from some of his former
14
labors.
Not long after, a number of slaves from the village would make another
attempt to escape. After spending the night hiding in a pit that had been
dug at a local residence, the group ventures toward Balamurghab along
with a group of Turkmens. They are intercepted in the desert by a band of
alamanchis (raiders), however, and after a fierce fight in which several
Turkmens are killed, the slaves are dragged back to Panjdih.
Although Mı̄rzā Mahmūd is not among them, he again suffers incidentally
˙
from the escape-plans of other slaves; once again, Khan Muhammad grows
˙
paranoid about losing his slave, and he tightens Mı̄rzā Mahmūd’s chains.15
˙
One of Mı̄rzā Mahmūd’s duties had been to give fodder to Khan
˙
Muhammad’s camel, a task he seems usually to have performed within
˙
sight of his master. One day, however, Khan Muhammad gives him a rope
˙
and a sack, unchains him, and instructs him to go retrieve more fodder on
his own. Mı̄rzā Mahmūd seizes the opportunity: once he is out of sight, he
˙
sets off toward Balamurghab. He ties his shoes on backwards in order to
deceive any pursuers (kaffashı̄ ki dar pā dāshtam ān-rā vārūna pūshı̄da).
Eventually, the cord with which he had strapped on the shoes frays, and he
continues barefoot through the desert. He soon loses his way. That night,
he finds himself at a Salar settlement. He stays out of sight, and attempts
stealthily to steal a horse. He cannot get the animal untethered, however,
and so he travels on, still barefoot. By morning he is tired, hungry, and
terribly thirsty. He comes to a deep canyon with a river at its bottom, but
the river is unreachable. Mı̄rzā Mahmūd climbs a hill to get his bearings
˙
13
Āshtiyānı̄, ʿIbratnāma, 47–48. 14
Āshtiyānı̄, ʿIbratnāma, 49.
15
Āshtiyānı̄, ʿIbratnāma, 49–51.
94 From Despair to Liberation

and sees, in the distance, the village of Murghab; with freedom in sight, he
resolves to press on toward the village under the cover of night.16
Suddenly, he sees five Turkmens some distance down the hillside – three
on horseback, two on foot. They see him too. Mı̄rzā Mahmūd’s hair had
˙
grown long, “like an Afghan’s,” and the Turkmens on foot take him for an
Afghan from Balamurghab. But the horsemen knew better: They were
from Panjdih, and they recognize Mı̄rzā Mahmūd, shouting out to the
˙
others that he is an escaped slave. They capture him and Mı̄rzā Mahmūd
˙
pleads that they not bring him back to the village; he promises a reward if
they sell him instead in Balamurghab. The Turkmens on foot immediately
agree, but the horsemen from Panjdih at first refuse. Eventually, however,
they too consent to the idea. Settling in for the night, they swear an oath
that they will sell Mı̄rzā Mahmūd somewhere nearby the very next day.17
˙
Their promise turns out to be a ploy, probably intended to placate their
prisoner. Khan Muhammad is informed of Mı̄rzā Mahmūd’s recapture, and
˙ ˙
the slave is taken to the Salar settlement of Marchaq, where Khan
Muhammad is waiting to retrieve him. There, Mı̄rzā Mahmūd finds his
˙ ˙
master “sitting like an angry boar” (chūn gurāz-i khashmgı̄n nishasta). Mı̄rzā
Mahmūd is beaten and chained. The next morning he is taken back to
˙
Panjdih, where he is beaten yet again until a local resident intervenes,
counseling Khan Muhammad that there is no point in beating a slave,
˙
since a caged bird will inevitably think of flying to freedom.18
Not long after, Mı̄rzā Mahmūd, remarkably undeterred, makes yet
˙
another attempt to gain his freedom. He slips from his chains one night
while Khan Muhammad is asleep, and heads again toward Balamurghab.
˙
He stays some distance from the main road, hoping to stay out of sight, but
once again he loses his way. Wandering without water in the brutal heat of
the day, the elements push him to the brink of death. Dazed from thirst, he
finds the bones of a sheep and tries ravenously to break them open, hoping
to find some blood with which to wet his parched mouth. The effort serves
only to weaken him further, and he begins to speculate that living in slavery
is better than dying of thirst in the desert. He scrambles up a hill to get his
bearings, and, from on high, he sees the glitter of water some distance
away. By night, he travels toward it.19
Suddenly, in the night, he hears the bark of a dog, followed by two or
three gunshots. He would later learn that the dog belonged to a shepherd,
who had fired into the air to scare off what was assumed to be a thief. Khan

16
Āshtiyānı̄, ʿIbratnāma, 51–53. 17
Āshtiyānı̄, ʿIbratnāma, 52–54.
18
Āshtiyānı̄, ʿIbratnāma, 55–56. 19
Āshtiyānı̄, ʿIbratnāma, 56–57.
 Mı̄rzā Mahmūd’s Story 95
˙
Muhammad later came upon the shepherd in his hunt for Mı̄rzā Mahmūd,
˙ ˙
and the shepherd described the runaway’s footprints: They must belong to
a fugitive, the shepherd shrewdly reasoned, since if they had been the
footprints of a Turkmen they would not have passed through Egyptian
thorn (khar-i mughı̄lān).20
Meanwhile, Mı̄rzā Mahmūd had made his way down to the river that he
˙
had seen from the hilltop. Thanking God that he had not died of thirst, he
went down to the waterside. Just as he is about to drink, he hears a voice
behind him, warning him that the water is bitter and salty. He turns to see
Khan Muhammad, who greets him with brutal blows from a horse-whip
˙
(qamchı̈). Khan Muhammad drags him into the water and bids him to
˙
drink it; Mı̄rzā Mahmūd vomits from a single sip. His master then gives
˙
him some bread and clean water, and brings him to a Turkmen tent where
he is given a yogurt drink (dūgh). Khan Muhammad assures him that if he
˙
had not consumed the bread, fresh water, and yogurt, the foul river-water
would have killed him.21
When they return together to Panjdih, Khan Muhammad beats Mı̄rzā
˙
Mahmūd once more and shackles him so tightly that escape is impossible.
˙
In his manacled state, Mı̄rzā Mahmūd assumes, at the very least, that he
˙
would no longer be made to work the hand-mill (dastās) or graze the
camels. But he is made to graze the camels anyway, though he is no longer
much use in this line of work. Khan Muhammad’s wife, noticing his poor
˙
work with the camels, insists that Mı̄rzā Mahmūd be starved, so that he
˙
can suffer the same fate as the camels he looks after. He goes hungry until,
in a stroke of luck, he finds two small melons (kharbūza) and
a watermelon (hindivāna) along with two discs of bread out in the
fields.22
One day, Mı̄rzā Mahmūd hears a local Turkmen crier (jarchi) announce
˙
to the people of Panjdih that a raid was being organized in the direction of
Mashhad, and that any who wished to partake should begin preparing
provisions and readying their horses (literally, keeping the horses “raw”
[khāmı̄]). Khan Muhammad is not up for raiding himself, but he has a fine
˙
horse and he plans to lend it to a raider in exchange for one-half of any
spoils the raider earns. Mı̄rzā Mahmūd is assigned to look after the horse in
˙
the meantime. The thought that the animal might be used to drag another
human being into slavery is more than he can take; he fetches his arsenic.
The horse dies that very night from the poisoned barley Mı̄rzā Mahmūd
˙
20
Āshtiyānı̄, ʿIbratnāma, 57–58. 21
Āshtiyānı̄, ʿIbratnāma, 58.
22
Āshtiyānı̄, ʿIbratnāma, 59.
96 From Despair to Liberation

puts in its feedbag, and Khan Muhammad seethes with rage, cursing any-
˙
one who comes near.23
The raid, as it turns out, ends in disaster for the two hundred horsemen
who gathered to take part in it. They encounter strong resistance on the
way, and ninety-six of them perish in the resulting battle. Another eighty-
two are taken prisoner and dragged to Mashhad, where, according
to Mı̄rzā Mahmūd, they are all beheaded, and their heads are sent to
˙
Tehran (majmūʿa-i ānhā-rā sar burı̄da ravāna-i dār al-khilāfa namūdand).
A number of Turkmens flee to the mountains, saving themselves, and after
eight or nine days they return to Panjdih, describing their ordeal to their
townsmen.24
Mı̄rzā Mahmūd still has some arsenic left, and it is not long before he
˙
comes up with yet another plan to torment his cruel master. One day,
a Turkmen trading caravan passes through Panjdih, and Khan Muhammad
˙
plans to make some money by paying a camel-driver to convey a load of
grain to Herat using his camels. Mı̄rzā Mahmūd mixes arsenic with their
˙
doughy fodder (khamı̄r), and when the camel-driver comes to collect the
camels they are already dead. Khan Muhammad loved his camels more
˙
than any son or brother, Mı̄rzā Mahmūd recalls, and he is bitterly sad-
˙
dened by their deaths.25
Sometime thereafter, Mı̄rzā Mahmūd falls terribly ill for a period of
˙
several weeks, nearly dying from his mysterious sickness.26 When his
condition finally improves, he learns something intriguing from the mem-
bers of a Bukharan caravan passing through the village: In Bukhara, slaves
are neither chained nor injured by their masters; those with skills in
calligraphy, composition, and crafts are respected. Mı̄rzā Mahmūd wishes
˙
intensely to make his way to Bukhara, although he cannot help but recall
an ominous verse from Rumi: “If you’re going to Bukhara then you’re
insane / worthy of the dungeon and the chains.” Poetic advice notwith-
standing, he begins to gather information on the Iranian merchants who
trade and live in Bukhara, and he gradually develops a plan of action.27
The plan that materializes is a brilliant one. Mı̄rzā Mahmūd forges
˙
a letter to Khan Muhammad from three known merchants trafficking
˙
goods between Bukhara, Panjdih, and parts south: Āqā Muhammad
˙
Kāzim, Āqā Mı̄r Taqı̄, and Qāsim Bay. The letter says, among other things,
that the merchants would give Khan Muhammad 250 tillas for Mı̄rzā
˙ ˙

23
Āshtiyānı̄, ʿIbratnāma, 63–64. 24 Āshtiyānı̄, ʿIbratnāma, 65–66.
25
Āshtiyānı̄, ʿIbratnāma, 66. 26 Āshtiyānı̄, ʿIbratnāma, 66–72.
27
Āshtiyānı̄, ʿIbratnāma, 73–74.
 Mı̄rzā Mahmūd’s Story 97
˙
Mahmūd, as the elites of Bukhara had use for a highly literate slave like
˙
him. To make the forged letter more realistic, Mı̄rzā Mahmūd steals some
˙
soap and carves a seal for each of the three merchants. All that remains is to
make sure the letter is delivered to Khan Muhammad by someone osten-
˙
sibly unrelated to Mı̄rzā Mahmūd, so as to avoid any suspicion of his role
˙
in the ruse. The perfect man for the job was a Turkmen named Niyāz Qul
whose father was a surgeon. Mı̄rzā Mahmūd tasks him with delivering the
˙
letter to Khan Muhammad at the bazaar, and he tells the young man to
˙
explain that he had received the letter from a Salar Turkmen who was
traveling from Bukhara to Herat. In return, Mı̄rzā Mahmūd agrees to knit
˙
Niyāz Qul a pair of socks for the winter.28
The ruse works. Khan Muhammad receives the letter and is ultimately
˙
convinced of its veracity. He reads it in front of Mı̄rzā Mahmūd and
˙
exclaims, “I’ve struck it rich!” (ganjı̄ yāfta!). Instead of his usual nickname
of “Demon” (ghol), Khan Muhammad begins calling Mı̄rzā Mahmūd the
˙ ˙
respectful “Āqā Mı̄rzā.” Before long, the news arrives that “Āqā Mı̄rzā” is
to leave on a caravan for Bukhara. He is introduced to a caravan-leader
named Qara Bay, who chains him by the neck to a camel, and when they
leave Panjdih on the long road to Bukhara, Mı̄rzā Mahmūd is forced to
˙
march shackled and on foot during the night. Khan Muhammad sends
˙
along a reply to the merchants’ fabricated letter, saying that he was sending
along his slave, and that he himself would follow in some ten days’ time,
but that he would not sell Mı̄rzā Mahmūd for any less than 300 tillas.29
˙ ˙
The road to Bukhara brings fresh tortures for Mı̄rzā Mahmūd. He is put
˙
to work driving the camels, collecting firewood, and cooking for the other
members of the caravan. Qara Bay beats him repeatedly, and Mı̄rzā
Mahmūd sees no rest. Passing through the hinterlands, he observes other
˙
Iranian slaves: He sees many of them occupied with working the land
(dehqānı̄) in places like Lubāb and Ghanjū. When the caravan comes to
Qaraqul, news spreads that a literate slave has arrived, and Mı̄rzā Mahmūd
˙
is brought before a potential buyer. He is taken to a majlis with a number of
Turkmens arranged around a felt carpet, and he is invited to drink with
them. But he is warned, mysteriously, not to point his feet toward the
verses of the Qur’ān that are in their presence, nor to drink before
them. Mı̄rzā Mahmūd looks around in confusion, seeing no verses
˙
28
Āshtiyānı̄, ʿIbratnāma, 76–77. Mı̄rzā Mahmūd observes with surprise that scarcely any of
˙
the Turkmens in the village knew how to make a decent pair of socks. He had previously
taught one of them the craft in exchange for a handsome payment: nearly a maund of bread
and some quantity of dried beef jerky (qāq).
29
Āshtiyānı̄, ʿIbratnāma, 80.
98 From Despair to Liberation

anywhere in sight, and he remarks that he can see nothing of the Qur’ān
around them. His hosts reply that a group of “qizilbāsh feltmakers”
(namadmālān-i qizilbāsh)30 had added Qur’ānic verse inscriptions to the
margins of the felt carpet before them. Mı̄rzā Mahmūd glances at the felt
˙
and observes that it is only poetry, and not scripture, that is woven there,
and he teases his hosts for the error. Hoping to save face, one of the
inquisitors engages him in a debate on the subject of ritual ablutions,
and Mı̄rzā Mahmūd – to the discomfort of the Turkmens around him –
˙
quickly proves his superior erudition on matters of Islamic ethics.31
The debate soon turns to the relative merits of Shiʿites and Sunnis, and
whether it is licit for Sunnis to take Shiʿites into slavery. Mı̄rzā Mahmūd
˙
proposes to his hosts that they are contravening their own religion by
holding Shiʿite slaves, but they dismiss the idea. Man-selling, they say, was
a custom from ancient times (ādamfurūshı̄ az qadı̄m rasm būda). Why else,
after all, did the sons of Ya’qūb sell their own brother Yusūf? (And Mı̄rzā
Mahmūd, they scoff, is no more honorable than Yusūf.) Why else, more-
˙
over, do Shiʿites themselves sell the Qur’ān, when they are no dearer than
the holy book? At this moment, it dawns on Mı̄rzā Mahmūd that he is
˙
talking to Sarı̈qs, who, he writes, have no other trade besides thievery
(duzdı̄). He jokingly observes aloud to them that they themselves are
mentioned by name in the Qur’ān: al-sāriq, in the scripture, means
“thief.” At this taunt, the party grows openly hostile, and Mı̄rzā
Mahmūd hastens away.32
˙
Mı̄rzā Mahmūd accompanies the caravan to the city of Bukhara without
˙
further incident, and upon arrival there Qara Bay takes Mı̄rzā Mahmūd to
˙
the man he was perhaps least eager to see: Āqā Muhammad Kāzim, whose
˙
name and seal he had forged. Naturally, Āqā Muhammad is flabbergasted
˙
to meet his supposed purchase, having no knowledge of any alleged
correspondence with the slave’s owner. His predictable objection
leaves Mı̄rzā Mahmūd histrionically heartbroken, and Āqā Muhammad
˙ ˙
reassures the poor slave that he needs only to have patience; the merchant
agrees to liberate him from Khan Muhammad. Āqā Muhammad’s plan is
˙ ˙
that a man named Shāhrukh Khan would buy Mı̄rzā Mahmūd and send
33 ˙
him off to Mashhad.
Mı̄rzā Mahmūd is also permitted to stay at Āqā Muhammad’s residence
˙ ˙
while awaiting his master Khan Muhammad’s arrival in Bukhara, and the
˙

30
The phrase “qizilbāsh” was long used as a generic term for Shiʿite Iranians in Central Asia.
31
Āshtiyānı̄, ʿIbratnāma, 83–84. 32 Āshtiyānı̄, ʿIbratnāma, 84–85.
33
Āshtiyānı̄, ʿIbratnāma, 86.
 Mı̄rzā Mahmūd’s Story 99
˙
merchant allows him to venture out on an errand, giving him some money
to fetch ice from the bazaar. On his way there, Mı̄rzā Mahmūd finds
˙
a blackened coin on the ground. The occasion provides the perfect catalyst
for a display of his keen entrepreneurship. At the bazaar, Mı̄rzā Mahmūd
˙
finds a man selling dyed eggs, and he uses the found coin to buy a red egg
(bayza-i qirmizı̄) from him. Returning home with the ice in his hands and
the egg under his arm, he borrows a pen-knife from Āqā Muhammad’s
˙
retainer, Qāsim, and sets to work making fine engravings (naqash o hakākı̄)
in the egg. When the work is done, he returns to the bazaar, showing off
the egg to great acclaim. One onlooker buys the egg for 14 pul – fourteen
times what Mı̄rzā Mahmūd had paid for it in its plain form. He uses the
˙
money to buy fourteen more eggs, which he engraves and sells for 8 or 10
pul each. From his earnings, he keeps no more than 20 pul and gives the
rest to Qāsim, both in payment for the use of his pen-knife and, perhaps, in
the expectation that he would prove to be a valuable ally later on.34
Mı̄rzā Mahmūd takes his remaining earnings to a bathhouse, where he
˙
meets another potential ally: an elderly barber and bath attendant who
reveals that he is likewise a Shiʿite, but that he has been practicing taqı̄ya
(pretending to be a Sunni) for many years in Bukhara. He reassures Mı̄rzā
Mahmūd that he never fails to help his Iranian brothers who are enslaved
˙
in Bukhara, and Qāsim later confirms the old man’s claims. Within just
a few days, Mı̄rzā Mahmūd has already begun to establish a network of
˙
friends and admirers.35
The next day, Mı̄rzā Mahmūd returns to the bazaar and finds a paper-
˙
dealer’s shop selling fine Samarqandi paper. He engages the shopkeepers in
conversation and again quickly proves his erudition. They ask him for
a lesson in Qur’ānic study and nahv (reading and grammar in Arabic), and
˙
he corrects their mistakes as they practice recitation. Ashamed of their
blunders, they ask for the chance to evaluate a sample of Mı̄rzā Mahmūd’s
˙
calligraphic skills, and in nastaliq he pens a verse from Saʿadi: “Skill is the
greater fault in the eyes of the opponent / Saʿadi is a rose, but in the
enemies’ eyes a thorn.” He then produces lines in a variety of classical
calligraphic forms: naskh, thuluth, shekasta, and taʿliq. Students gather
around to observe him, and they all notice his poor state of dress: Mı̄rzā
Mahmūd goes barefoot. “If this man was not a Shiʿite,” one student
˙
remarks, “I would pay all the money in the world for him” (agar ı̄n mard
shı̄ʿa nabūd man u-ra be har qı̄matı̄ ke bāvad mı̄-kharı̄dam). The student
reflects that it is for the best that he remains in slavery: If Mı̄rzā Mahmūd is
˙
34
Āshtiyānı̄, ʿIbratnāma, 88. 35
Āshtiyānı̄, ʿIbratnāma, 88–89.
100 From Despair to Liberation

ever to be emancipated, according to the young man’s logic, he will never

become a Sunni (hargiz sunni nakhwāhad shod).36
Mı̄rzā Mahmūd continues the very next day with his energetic net-
˙
working. He meets the Shiʿite barber at the bazaar, where the latter
introduces him to a man named Hajji Muhammad Sālih, who quickly
˙ ˙ ˙
proposes to be Mı̄rzā Mahmūd’s patron. The man offers him some
˙
money as a stipend. Mı̄rzā Mahmūd replies that it would be wrong of
˙
him to accept charity, but that he would be happy to go to work as an
37
employee.
With new friends and new prospects of employment, Mı̄rzā Mahmūd
˙
returns to Āqā Muhammad’s residence, where, unfortunately, a complica-
˙
tion had arisen: his master, Khan Muhammad, had arrived from Panjdih,
˙
eager to collect his expected payment. Khan Muhammad quickly learns
˙
from Āqā Muhammad that he will get nothing like the price he had
˙
anticipated. This was not to say, of course, that there was no demand
for Mı̄rzā Mahmūd: Khan Muhammad learns that his slave had indeed
˙ ˙
stirred up some interest in Bukhara. A slave-dealer (ghulām-jallāb) named
ʿAzı̄mbay is summoned to appraise Mı̄rzā Mahmūd, and – likely in collu-
˙ ˙
sion with Āqā Muhammad – he estimates the slave’s value at the low sum
˙
of 20 tumans. At this, Khan Muhammad is speechless. Without another
˙
word, he returns to the caravanserai that served as his lodging place.
Two or three days later, a Jewish man named Musa negotiates with
Khan Muhammad and Āqā Muhammad to buy Mı̄rzā Mahmūd for 22
˙ ˙ ˙
tumans. Thus begins a curious series of transactions. Mı̄rzā Mahmūd
˙
lives at Musa’s hujra for no more than two days before Hajji Muhammad,
˙
his would-be patron, buys him from Musa. Hajji Muhammad explains
˙
himself by saying that it is not acceptable for a Muslim to be the servant of
a Jew (rāzı̄ nashod ki musulmān khidmatkār-i yahūdı̄ bāshad). Two days
˙
later, he turns around and sells Mı̄rzā Mahmūd to a man named Hājjı̄
˙ ˙
Rahı̄m Herātı̄, who buys Mı̄rzā Mahmūd for 20 tumans, and then
˙ ˙
promptly re-sells him to Āqā Muhammad, who promises once again
˙
that eventually Mı̄rzā Mahmūd would become a free man.38 In the
˙
meantime Mı̄rzā Mahmūd goes to work as an accountant for an acquain-
˙
tance of Āqā Muhammad.
˙
Soon enough, Khan Muhammad appears once again at the merchant
˙
Āqā Muhammad’s residence, accompanied by two or three Bukharans. He
˙
grabs Mı̄rzā Mahmūd’s collar, announcing that he still owns him,
˙
36
Āshtiyānı̄, ʿIbratnāma, 91. 37 Āshtiyānı̄, ʿIbratnāma, 92.
38
Āshtiyānı̄, ʿIbratnāma, 93–94.
 Mı̄rzā Mahmūd’s Story 101
˙
regardless of the present circumstances. At that moment, all of the
injuries Mı̄rzā Mahmūd had suffered under that cruel master rush into
˙
his mind, and he grabs Khan Muhammad’s beard and gives it a yank,
˙
punching him in the face with his other hand and knocking out the two
or three decaying teeth which remained in his mouth. Khan Muhammad,
˙
helpless, can do nothing more than complain to Āqā Muhammad, who
˙
declares that the Turkmen no longer has any business with his former slave,
as his ownership had been terminated when he agreed to sell Mı̄rzā
Mahmūd some days previously. Taking stock of the intensity of Khan
˙
Muhammad’s agitation, however, the merchant must have wondered if
˙
swindling this man out of his property would really be worth the effort:
Āqā Muhammad, in the end, shows some willingness to negotiate with
˙
Khan Muhammad for the eventual return of Mı̄rzā Mahmūd to Panjdih.39
˙ ˙
Upon hearing this, Mı̄rzā Mahmūd flees the merchant’s residence, and
˙
he eventually makes his way to the home of a man named Mı̄r ʿĀsad, who
was rumored to have freed several slaves and to have aided them in their
journeys to elsewhere. Arriving at his door, however, Mı̄rzā Mahmūd
˙
learns that the man is out shopping at the bazaar, and he is told to return
later. Terrified of being found by Khan Muhammad, Āqā Muhammad, or
˙ ˙
any of their helpers, Mı̄rzā Mahmūd takes refuge for some time among the
˙
worshippers in a nearby mosque. He later returns to Mı̄r ʿĀsad’s house
once more, only to find that he still had not returned home. With nowhere
else to go, Mı̄rzā Mahmūd heads back to the mosque – “there is,” he writes,
˙
“no better hiding-place than that.”40
Mı̄rzā Mahmūd pretends to sleep at the mosque until the crowd
˙
gathers once more for their prayers. As he prays with them, Mı̄rzā
Mahmūd wonders what he will do if Mı̄r ʿĀsad has not returned by
˙
nightfall. On his way back to the man’s residence, he runs into one of
Āqā Muhammad’s retainers – evidently a fellow slave, and a man with
˙
whom Mı̄rzā Mahmūd had become friends. He learns from this man that
˙
Khan Muhammad, enraged, had taken his case to the hākim of Bukhara.
˙ ˙
The retainer’s best suggestion for Mı̄rzā Mahmūd is that he take refuge
˙
with Mı̄r ʿĀsad. They go to his residence together, and they wait for him
for some time, but he does not appear. Unable to wait indefinitely at the
stranger’s house, Mı̄rzā Mahmūd realizes – surely with profound sor-
˙
row – that he has nowhere else to go, and no better option than to return
to Āqā Muhammad and hope for the best.41
˙
39
Āshtiyānı̄, ʿIbratnāma, 95–97. 40
Āshtiyānı̄, ʿIbratnāma, 98–99.
41
Āshtiyānı̄, ʿIbratnāma, 99.
102 From Despair to Liberation

When he arrives at Āqā Muhammad’s residence, he finds Khan

˙
Muhammad sitting there, eyeing him with rage “like a shot bear” (misl-i
˙
khirs-i tı̄r khwārda). Mı̄rzā Mahmūd attempts to explain his disappearance
˙
by claiming that he had needed to attend the wedding of a friend. Khan
Muhammad insists that Mı̄rzā Mahmūd must return with him to Panjdih
˙ ˙
that same day. Mı̄rzā Mahmūd replies simply that there is a problem with
˙
this plan. “What is the issue?” Khan Muhammad asks. Mı̄rzā Mahmūd
˙ ˙
replies: “That I must either drown you or drown myself in the waters of the
Jayhūn” (bāyad tū-rā dar āb-i daryā-yi jayhūn ghurq o halāk kunām yā
khodam-rā). Āqā Muhammad, meanwhile, understandably wanting noth-
˙
ing more to do with the conflict, declares simply that Mı̄rzā Mahmūd is
˙
Khan Muhammad’s slave, and that he can do with him whatever he
42 ˙
wishes.
With Mı̄rzā Mahmūd’s sad fate seemingly sealed, the news comes that
˙
Shāhrukh Khan – the man who was initially expected to purchase Mı̄rzā
Mahmūd and, eventually, to free him – has returned to Bukhara. As quickly
˙
as the news arrives, Mı̄rzā Mahmūd is spirited off to Shāhrukh Khan’s
˙
residence by his friend Qāsim, unbeknown to Khan Muhammad and Āqā
˙
Muhammad, and Shāhrukh interviews him about his background and his
˙
past work for the Iranian military. Mı̄rzā Mahmūd stays there for two or
˙
three days. Eventually, Āqā Muhammad arrives to investigate, and he and
˙
Shāhrukh Khan hatch a plan together to convince Khan Muhammad
˙
that Mı̄rzā Mahmūd had fled to Kabul.
˙
For once, however, Khan Muhammad is not to be fooled, and he
˙
arrives, furious, at Shāhrukh Khan’s door, demanding the release of his
slave and threatening to take the matter to the Amı̄r of Bukhara himself.
Shāhrukh Khan replies that Mı̄rzā Mahmūd is a sayyid, and that he never
˙
should have been bought or sold in the first place; moreover, if Khan
Muhammad insists on bringing the case before the Amı̄r, Shāhrukh
˙
would describe Mı̄rzā Mahmūd’s high birth and talents before the court
˙
and observe that such a gifted slave belongs only in the service of His
Majesty. In that way, Khan Muhammad could expect to end up with no
˙
slave and no payment.43
Khan Muhammad sees that his case had become hopeless, but Āqā
˙
Muhammad, perhaps mindful of the Turkmen’s unpredictable temper,
˙
negotiates with Shāhrukh Khan to offer a settlement of 30 tillas in
exchange for the slave. He counsels Khan Muhammad that it would be
˙
wise to take the offer. Khan Muhammad replies that he had paid 14
˙
42
Āshtiyānı̄, ʿIbratnāma, 99–100. 43
Āshtiyānı̄, ʿIbratnāma, 103.
 Mı̄rzā Mahmūd’s Story 103
˙
tillas when he bought Mı̄rzā Mahmūd; that the slave had outstanding debts
˙ ˙
to him totaling 18 tillas; that on two of the occasions when he had run
˙
away, Khan Muhammad had paid a 3 tilla reward leading to his capture;
˙ ˙
that he had covered the 3–4 tilla customs toll for transporting his slave to
˙
Bukhara, along with another 3 tillas to rent the camel that conveyed him
˙
there. Why, then, after all this, should he sell Mı̄rzā Mahmūd for 30 tillas?
˙ ˙
Āqā Muhammad answers bluntly: because, he says, there would be no
˙
higher offer. Khan Muhammad agrees to the deal, announcing that, hen-
˙
ceforth, he intended to quit the business of slave-dealing (man baʿd tark-i
ası̄rfurūshı̄-rā khwāham), repenting of the fact that he had ever gotten
involved in it and vowing never again to buy or sell a slave (tawba kardam
ke dı̄gar ası̄r nikharam va nifurūsham).44
Shāhrukh Khan immediately shows Mı̄rzā Mahmūd his hospitality, as
˙
does his brother, Muhammad Hassan Khan, and Mı̄rzā Mahmūd
˙ ˙ ˙
impresses them in the days to come such that he is assigned to oversee
their household’s kitchen and stables. He is provided with a stipend
equivalent to three Iranian tumans per month. In his spare time, he is
given the freedom to work in other occupations, some of which prove
quite lucrative: In just five months, Mı̄rzā Mahmūd reports earning 70
˙
tumans from his labors as well as six or seven suits of clothing.
Meanwhile, Mı̄rzā Mahmūd’s erroneous reputation as a sayyid reasserts
˙
itself thanks to Āqā Muhammad Khan, who spreads rumors of his noble
˙
birth and urges him to maintain the charade, which, he says, could come
to benefit him in Bukhara. Mı̄rzā Mahmūd soon finds himself dubbed
˙
“Āqā Sayyid Taqı̄,” an honorific he does little to shake off, despite his
45
guilty conscience.
Some two months later, Mı̄rzā Mahmūd finds himself on his way to
˙
Samarqand, tagging along with his master’s retinue. His trip to the
ancient city offers him an opportunity to reflect on the region more
generally, revealing some of his thoughts on nature and society in
Turkestan. He observes that the climate in Samarqand is vastly superior
to that of Bukhara, and that it is generally free from the ravages of
guinea worm, a problem that in Bukhara had reached epidemic
proportions. Mı̄rzā Mahmūd describes Bukhara’s guinea worm crisis
˙
with a palpable horror – and in sickening detail – and reveals another of
the jobs he worked while living in that city: He had been tasked many
times with drawing guinea worms from the bodies of unfortunate vic-
tims. He is understandably relieved to find that the ailment hardly exists

44
Āshtiyānı̄, ʿIbratnāma, 104. 45
Āshtiyānı̄, ʿIbratnāma, 105–106.
104 From Despair to Liberation

in Samarqand, which he attributes to the good quality of its air and

water.46
Less appealing in his eyes, however, is Samarqand’s social climate: He
describes Samarqand as a city of hedonists (ʿayyāsh), strewn with multiple
opium dens (kūknārkhana), in each of which could be found a “simple
youth with curly hair” (javānı̄-i sāda ba gı̄sūvān-i mujaʿad) serving tea and
passing around the opium pipes. Smoking cannabis mixed with tobacco
(chars) and opium-eating (taryāk-khurdan), he claims, are also exceedingly
common in the city, and as a result the men of Samarqand are generally
lazy, timid, sickly, lacking in energy, and yellowish in pallor.47
With unrest growing in the region on the eve of the Russian conquest of
the khanate of Kokand, Mı̄rzā Mahmūd soon finds himself headed to the
˙
capital city of Kokand itself as his master, Shāhrukh Khan, is appointed to
oversee the Bukharan armory (tupkhana). Shāhrukh Khan is welcomed by
many of the city’s elites, but he and his retinue are soon on the move again
in the face of hostilities from the forces of a Kazakh leader named Mullā
Qulı̄ Khan (ʿAlimqul).48 It seems that Mı̄rzā Mahmūd did not see much of
˙
the city of Kokand, as he offers a scant description of it, noting only the
presence of esteemed burial sites along with his impression that the city’s
men are yellowish and lean, while its women are ravishing (dilrubā) and
lively (rūhparvar). With this, Mı̄rzā Mahmūd heads back on the road again,
˙ ˙
journeying with his master and his retinue to the mountains near Osh as
violence engulfs the region. Mı̄rzā Mahmūd here offers valuable reportage
˙
on local struggles that destabilized Kokand on the eve of the Russian
conquests, as the population of the capital city was split in their allegiances
between Khudāyār, the incumbent ruler, and Mullā Qulı̄ Khan.49
Mı̄rzā Mahmūd stays at Shāhrukh Khan’s side even as the latter begins
˙
to slander his own ruler, the Amı̄r of Bukhara (out of jealousy, Mı̄rzā
Mahmūd alleges), calling him a traitor. The Amı̄r does not take kindly to
˙
these slights, and rumors soon surface that the Amı̄r plans to place
Shāhrukh Khan under arrest. This revelation must have come as
a traumatic one for Mı̄rzā Mahmūd, who under Shāhrukh Khan’s steward-
˙
ship was considered a retainer of the court (nökar-i dı̄vān) and had even
been confirmed as chur-aqasi, a title, in Mı̄rzā Mahmūd’s telling, higher
˙
46
Āshtiyānı̄, ʿIbratnāma, 109. 47 Āshtiyānı̄, ʿIbratnāma, 110–111.
48
I am grateful to Alexander Morrison for identifying this figure for me. Amazingly,
ʿAlimqul’s memoirs are available to us in English translation: Timur Beisembiev,
The Life of Alimqul: A Native Chronicle of Nineteenth Century Central Asia (New York:
Routledge, 2015).
49
Āshtiyānı̄, ʿIbratnāma, 111–119.
 Mı̄rzā Mahmūd’s Story 105
˙
than that of yuzbāshi.50 When Mı̄rzā Mahmūd goes to Shāhrukh Khan’s
˙
residence one day to discuss the frightening rumors going around, he finds
that he is too late: His master’s place is swarming with men tasked by the
Bukharan Amı̄r with emptying it. Mı̄rzā Mahmūd’s own goods were
˙
expropriated, his master’s horses were gathered in the yard, and all of
the women and servants of his mobile estate had been stripped “as naked as
captives” and carried off, weeping.
Mı̄rzā Mahmūd is taken to a government dungeon (mahbūs-i hukūmatı̄)
˙ ˙ ˙
in the citadel of Sar Teppe. The citadel is home to multiple prisons, each
heinous in its own characteristic way. One is called the “Damphouse”
(narm-khana), so called because of its intolerable humidity. Another is
called the “Tickhouse” (kana-khana), where prisoners would suffer ticks
“as big as camels” that would suck their blood so vigorously that they
would be dead within just two days. Prominent prisoners, such as
Shāhrukh Khan and his brother, were held in one of two or three small,
private yards. The grave of Siyāvush Dāmād Afrāsiyāb was also said to be in
the dungeon.51
Within three days, the Amı̄r of Bukhara ordered that all of the goods
and prisoners should be sent onward to Bukhara, and that all of Shāhrukh
Khan’s horses should be sold for profit. Mı̄rzā Mahmūd gains at this point
˙
a preview of his most likely fate, as two Iranian slaves that Shāhrukh Khan
had bought and freed were resold to buyers elsewhere. Some employees of
the Amı̄r then demand money from Mı̄rzā Mahmūd, offering him his
˙
freedom in return, but he assures them that all of his worldly holdings
had already been taken; Mı̄rzā Mahmūd figures that the hope of extorting
˙
money from him is all that delays his inevitable fate of being sold.
Soon enough, for reasons that are unclear to him at the time, Mı̄rzā
Mahmūd is told that he is to have a personal audience with the Amı̄r. He
˙
finds himself being led to a caravanserai, within which he is taken to
a room illuminated with dozens of candles. He realizes that he has been
taken to some secret cell, and asks his captors where they have taken him.
One laughs, and replies that he is in the “house of pain” (manzil-i taʿb) –
another of the Amı̄r’s dungeons. In this dungeon he sees just one other
man: a Jew, who shares a water-pipe with him and says that he was
imprisoned after a quarrel over silk with two men who later requested
his detainment. Mı̄rzā Mahmūd requests that the guards move him to
˙
50
It seems unlikely that his position was really so lofty as this, however: Mı̄rzā Mahmūd
˙
reveals that his main job at this point of his captivity was to supervise his master’s horses
and stables, a job of considerable responsibility but hardly equal to that of a yuzbāshi.
51
Āshtiyānı̄, ʿIbratnāma, 122.
106 From Despair to Liberation

a more hospitable place, and they consent, shifting him to a room where
some of Shāhrukh Khan’s men were being kept.52
Mı̄rzā Mahmūd is not imprisoned there for long before he learns that
˙
the time for his audience with the Amı̄r has come. He is brought out
before the Amı̄r and some of his men, and he realizes upon being brought
into the light that his place of detainment was in fact part of a complex of
stables. The Amı̄r’s men demand money from Mı̄rzā Mahmūd, but he
˙
replies once more that whatever he had was already taken from him
“except for this depressed body and oppressed heart” (savā-yi ı̄n jism-i
afsarda va jān-i sitam rası̄da). This answer does not satisfy the Amı̄r, who
orders that Mı̄rzā Mahmūd be stretched between two posts and branded
˙
(dāgh kardan va ʿuqābayn kashı̄dan). Before the hot brand can touch the
slave’s flesh, however, the Amı̄r changes his mind, and instead of pro-
ceeding with the torture he tells Mı̄rzā Mahmūd of his intentions to take
˙
ownership of him. Mı̄rzā Mahmūd, perhaps dazed from his ordeal,
˙
replies that he would never consent to this, and asks to be sold to some
53
other owner instead.
The Amı̄r, angered, has Mı̄rzā Mahmūd turned over to his vizier, who
˙
receives him kindly and, offering to purchase him with a payment of 30
tillas to the state treasury, plans to put Mı̄rzā Mahmūd to work tutoring his
˙ ˙
son. Mı̄rzā Mahmūd goes to work, and he soon proves his talents as
˙
a scribe as well, so that the vizier gives him over to work as an accountant
for his brother, a merchant at the silk bazaar. Mı̄rzā Mahmūd likewise
˙
performs admirably in his new position, earning the esteem of the silk
merchants and settling into a comfortable new routine. He later learns that
his former master, Shāhrukh Khan, has not been as lucky: On the Amı̄r’s
orders, both he and his brother were beheaded, their bodies buried
dishonorably.54
Some days later, an international scandal is precipitated as the Amı̄r
orders the imprisonment of some English travelers and the confiscation of
their goods. A spy of Armenian descent named Mı̄rzā Yaʿqūb Khan is
dispatched to investigate this and other matters, posing as a merchant.
This secret agent soon learns of Mı̄rzā Mahmūd and endeavors to inter-
˙
view him as a possible source and ally. Mı̄rzā Mahmūd passes on the news
˙
of Shāhrukh Khan’s brutal murder, along with details concerning the
military and finances of Bukhara. According to Mı̄rzā Mahmūd, Mı̄rzā
˙
Yaʿqūb Khan records this information in a letter to officials in France.

52
Āshtiyānı̄, ʿIbratnāma, 124. 53
Āshtiyānı̄, ʿIbratnāma, 125.
54
Āshtiyānı̄, ʿIbratnāma, 128.
 Mı̄rzā Mahmūd’s Story 107
˙
The spy manages to extract covert information from several others too
before arousing the suspicion of some individuals loyal to the Amı̄r, who
inform the vizier of his activities. Mı̄rzā Yaʿqūb is imprisoned in a small,
dark room, and his possessions are scoured for evidence.
(Fortunately, Mı̄rzā Mahmūd reports, the Amı̄r’s henchmen were unable
˙
to read his letter to France.) Mı̄rzā Yaʿqūb claims to his interrogators that
those he had met with privately in the khanate, including Mı̄rzā Mahmūd,
˙
were simply old acquaintances. But this is not enough to secure Mı̄rzā
Mahmūd’s safety: He is beaten so brutally that he fears for his life, and he
˙
learns that the vizier has ordered his execution as a punishment for talking
with the spy.55
Mı̄rzā Mahmūd’s life is saved only through the intervention of Iranian
˙
friends and acquaintances, who convince the vizier to sell him instead of
having him killed. A slave-dealer (barda-furūsh) is summoned, but
because Mı̄rzā Mahmūd is still badly bruised from his beating, the vizier
˙
decides to imprison him until he heals enough to be sold. Thus Mı̄rzā
Mahmūd finds himself in a dungeon for the third time. This time, he is held
˙
in a small room “darker than night and more dreadful than the heart of the
aggrieved” (tārı̄ktar az shab-i dı̄jūr va mūhishtar az dil-i ranjūr).
˙
Once more, Mı̄rzā Mahmūd manages to improve his situation by using
˙
his wits. Each day, after evening namāz, the vizier would pass by Mı̄rzā
Mahmūd’s cell, and Mı̄rzā Mahmūd would perceive the light of his lantern
˙ ˙
moving past the cell door. One evening, as he passes by, evidently unaware
that Mı̄rzā Mahmūd knew he was near, the prisoner shouts out prayers and
˙
lamentations to the effect that God knows of his innocence and that he has
committed no betrayal. The vizier comes closer to his cell, listens for a time
to Mı̄rzā Mahmūd’s cries, and then leaves. A half-hour later, the cell door
˙
opens; the vizier’s servants bring Mı̄rzā Mahmūd a candle and a carpet,
˙
telling him that their master’s heart ached for his predicament and that he
56
had been forgiven.
Mı̄rzā Mahmūd is freed from the dungeon, interrogated once more
˙
about his connections with Mı̄rzā Yaʿqūb Khan, and for some time held
in Ghijduvan, and afterwards in a smaller settlement some distance from
Bukhara. In the meantime, the Amı̄r of Bukhara was facing one of the
greatest crises in the khanate’s history: The armies of the Russian Empire,
having subjugated the town of Kokand, were now headed for Tashkent, the
Kokand khanate’s most important city. In the midst of the long siege that
followed, Mı̄rzā Mahmūd claims that his master, the vizier, occupied
˙
55
Āshtiyānı̄, ʿIbratnāma, 129–133. 56
Āshtiyānı̄, ʿIbratnāma, 134–135.
108 From Despair to Liberation

himself with a lavish circumcision ceremony for his son, for which he paid
the outrageous sum of 4,000 tumans. Mı̄rzā Mahmūd explains that the
˙
manumission of slaves often served as a key element in circumcision
ceremonies among Sunni elites (chūn qaʿida-i ahl-i tisnan dar chūnı̄n
jashn azād kardan-i usārā hast), and that the vizier had intended to liberate
twelve Iranian slaves during the course of the event. The Amı̄r of Bukhara,
however, evidently desperate for revenue, forbids him on the grounds that
the slaves could be sold and the proceeds given over to the treasury.57
Mı̄rzā Mahmūd, noting the vizier’s disappointment over the Amı̄r’s
˙
greed, attempts to convince the vizier to free him and allow him to return
to his homeland. It is no use, however, and soon enough Mı̄rzā Mahmūd
˙
receives an unusual new assignment: The Amı̄r arranges to have a room in
his residence elaborately painted, and he orders Mı̄rzā Mahmūd to join
˙
eight artists from Bukhara for the task. Among them, Mı̄rzā Mahmūd
˙
recalls, three – Nāzim and Ahmad Kalla, and ʿAbd al-Qādir Khwāja –
˙ ˙
were quite famous in the khanate. The walls of the residence were to be
decorated with human and animal figures as well as floral motifs.
Here, Mı̄rzā Mahmūd provides a window onto the workings of decorative
˙
artists in Bukhara. Observing that drawing human likenesses was prohib-
ited in Bukhara (suratkashı̄-ra mamnūʿ o mażmūm būd), Mı̄rzā Mahmūd
˙ ˙
relates that his fellow artists were unskilled at it and relied on old images by
Iranian artists that they had carried with them under their arms. They
traced these images through the process of pouncing and then outlined the
images in black with a brush before adding color. Mı̄rzā Mahmūd allows
˙
that they had considerable skill in drawing arabesque designs (islı̄mı̄) as
well as flowers and other elements from nature, though he cannot help but
add that their skill in these areas was still inferior to his own. In his own
telling, they soon recognize this fact as well, referring to him as “master of
masters” and seeking his guidance on their own work.58
The Amı̄r likewise praises Mı̄rzā Mahmūd’s artistry, and rewards him
˙
generously. At one point, the Amı̄r asks him about his circumstances,
wondering whether Mı̄rzā Mahmūd considered himself a slave or a free
˙
man while under the vizier’s dominion. Mı̄rzā Mahmūd cleverly – if
˙
evasively – replies that “the free people of the world all wish to be slaves
to Your Excellency” (āzādān-i jahān hama-hama ārzū-yi bandagı̄-i janābli
dārand). The Amı̄r, pleased with his answer, thanks Mı̄rzā Mahmūd. “I will
˙
give you a family and make you a man of the house [i.e. one of my own
servants],” he tells him. “I will give you a wife from among my slave-girls”

57
Āshtiyānı̄, ʿIbratnāma, 139–140. 58
Āshtiyānı̄, ʿIbratnāma, 140–142.
 Mı̄rzā Mahmūd’s Story 109
˙
(az chūrı̄hā-yi andarun zan be tū mı̄-deham). The offer would surely have
thrilled many slaves, but Mı̄rzā Mahmūd realizes that if he marries
˙
a woman from the household of the Amı̄r then it will become all the
more unlikely that he can ever be freed. He obsequiously demurs and,
delicately, asks the Amı̄r for his freedom instead. His evident ingratitude
angers the Amı̄r, who remarks on how inappropriate it would be if a Shiʿite
was sent back to live among his fellow “infidels” after having passed into
the “abode of Islam.” Mı̄rzā Mahmūd reports (perhaps dubiously) that he
˙
offered a tart reply to this insult, and the enraged Amı̄r storms off,
leaving Mı̄rzā Mahmūd’s fellow artists wondering over his audacity,
˙
which seems to them like a death-wish.59
Mı̄rzā Mahmūd continues to prove his worth as an artist, however, and
˙
he suffers no punishment for his impudence. But he cannot find a means to
free himself, even after he manages to receive a letter of manumission
sealed by a chief judge in Bukhara. In the months to come he petitions the
Amı̄r repeatedly for his freedom. He does this so often that the Amı̄r,
finally losing patience, tells him that if he brings one more petition then he
would end the conversation by killing him (ay ghulām agar yakdafaʿa dı̄gar
ʿariża-i tū be hużūr-i man bı̄-āyad bedūn javāb o savāl koshta khwāhı̄
˙
shod).60
Distraught as the prospect of gaining his freedom through official
channels slips away, Mı̄rzā Mahmūd nevertheless continues to put his
˙
many talents to work in Bukhara. One evening at the vizier’s house, after
hearing a recitation by a poet favored among Bukharan elites, Mı̄rzā
Mahmūd is asked teasingly by the Amı̄r if he likes the poet’s work. He
˙
replies that he does not, much to the amusement of the gathering, and
proceeds to offer insightful critiques of its failings. The poet too appreci-
ates his corrections, and soon enough Mı̄rzā Mahmūd finds himself in the
˙
role of a minor court poet, occasionally writing verses that he would
submit to the Amı̄r or have others recite for him. Mı̄rzā Mahmūd com-
˙
plains bitterly about the paltry remuneration he receives for his work – no
more than a few ceremonial robes. (We, however, as observers, are more
likely to be struck simply by his progression from a field-slave distributing
camel fodder in a Turkmen village to a court-poet among Bukharan
royalty.61)
Meanwhile, Mı̄rzā Mahmūd uses some of his savings to help fellow
˙
slaves on their journeys home. He offers 60 tumans toward the price of

59
Āshtiyānı̄, ʿIbratnāma, 143–144. 60
Āshtiyānı̄, ʿIbratnāma, 143–147.
61
Āshtiyānı̄, ʿIbratnāma, 147–150.
110 From Despair to Liberation

freedom for a slave-woman from Mashhad, and he assists with the travel
expenses for three other manumitted slaves on their way to Herat. He also
finds work as a calligrapher, working on several copies of the Qur’ān.
During this period of steady work but limited prospects for freedom, he
hangs his hopes on the only possible means of emancipation he could still
imagine: If the Russians conquered Bukhara, he reasons, he could poten-
tially find himself a free man amid the chaos.62
In the months to come, this once-distant possibility starts to seem increas-
ingly likely. After the fall of Tashkent, Mı̄rzā Mahmūd observes the Russians
˙
advance into Bukharan territory, while the Amı̄r of Bukhara declines oppor-
tunities to ease tensions and to forestall a conquest that seems increasingly
inevitable. Desperate to increase his fighting force, the Amı̄r issues an order
that all Iranian slaves in his domains must be brought to Bukhara, where any
of them willing to enter military service would be purchased by the Amı̄r and
outfitted for war. Within three weeks, 4–5,000 slaves had gathered at
various caravanserais, and from among them 1,000 entered the armed
forces. These new troops fought the Russians at Ura-Teppe, but their assis-
tance could not secure the city, which fell to the Tsar’s army.63
With the Amı̄r increasingly panic-stricken before the Russian
onslaught, Mı̄rzā Mahmūd’s thoughts turn again to the possibility that
˙
he might gain his freedom simply by requesting it, figuring that his poetic
and artistic talents would be little use to the Bukharan elites amid the
present chaos. He contacts the vizier, who replies that both the Amı̄r and
the city of Bukhara were indeed in dire straits. Finally, he agrees to
grant Mı̄rzā Mahmūd his freedom.64
˙
The vizier gives Mı̄rzā Mahmūd a letter to help secure his free passage
˙
back to Iran. The letter states that Mı̄rzā Mahmūd is the vizier’s slave and
˙
that no Turkmens should block his way at the river-crossings or otherwise
interfere with him on the way to Mashhad. Mı̄rzā Mahmūd receives letters
˙
from Mullā ʿAbd al-Karı̄m and some other well-known merchants as well,
and he readies his belongings and prepares a horse for the journey home.
He pays 70 tumans to a guide who promises to convey him safely to Iran,
and two days later he joins a Turkmen caravan bound for Merv. The party
suffers an attack on the way, but Mı̄rzā Mahmūd and some of his fellow
˙
travelers finally arrive in the town, where Mı̄rzā Mahmūd finds lodging
˙
with a merchant who had made his fortune shipping goods between
Mashhad and Bukhara.65

62
Āshtiyānı̄, ʿIbratnāma, 150. 63 Āshtiyānı̄, ʿIbratnāma, 161–166.
64
Āshtiyānı̄, ʿIbratnāma, 169–170. 65 Āshtiyānı̄, ʿIbratnāma, 172.
 Reflecting on Mı̄rzā Mahmūd’s Ordeal 111
˙
A few days later, Mı̄rzā Mahmūd leaves the merchant’s house in the
˙
company of four soldiers, all headed for Sarakhs. They lose the road,
however, and wander for a few days, trying to find their way. At night,
they resist the urge to make fires for fear of Turkmen thieves. Finally, they
see Sarakhs before them in the distance. In celebration, they make tea and
share a water-pipe.66
Mı̄rzā Mahmūd stays in Sarakhs for forty-five days, likely fearing to
˙
continue onward because of the presence of raiders from Merv said to be
decamped not far outside the city, at the shrine of ʿUlaq Bābā. This saint,
also known (according to Mı̄rzā Mahmūd) as Luqman Sarakhsı̄, was held
˙
in such esteem by the Turkmens that runaway captives who managed to
reach his shrine were permitted to remain free, and that if a Turkmen
hunting party chased prey into the vicinity of it, the hunters would give up
the chase. Finally, Mı̄rzā Mahmūd joins a large party for the three-day
˙
journey to Mashhad, where he finds lodging at an Uzbek caravanserai. He
remains in Mashhad for twelve days before heading home to Tehran.
Home at last, he sees his mother and his brother for the first time in
a decade.67

REFLECTING ON MĪ RZĀ MAḤM Ū D’S ORDEAL

Mı̄rzā Mahmūd’s narrative is, above all, a story of how one slave managed,
˙
through the force of his own ingenuity, to influence his fate against extra-
ordinary odds. Nevertheless, it would be naı̈ve to see Mı̄rzā Mahmūd’s
˙
story in purely optimistic terms, stressing nothing more than his triumph
over adversity. Notwithstanding the gains in quality-of-life that Mı̄rzā
Mahmūd gradually achieved, desperation was the driving force behind
˙
his audacity and his creativity. If we are to speak of his “agency,” we must
remember that he was his masters’ agent in most things. Meanwhile, in
matters where he seized some personal initiative, danger was ever-present,
and his efforts to improve his circumstances sometimes left him brutally
beaten. More than once, he reports being beaten so badly that he expected
to die from his injuries. Good luck was as much a factor in Mı̄rzā
Mahmūd’s eventual success as his own considerable genius: He was
˙
lucky that he was not beaten to death; he was lucky that Khan
Muhammad did not conclude with certainty that Mı̄rzā Mahmūd had
˙ ˙
poisoned him and his family; he was lucky to find his way while lost in

66
Āshtiyānı̄, ʿIbratnāma, 174. 67
Āshtiyānı̄, ʿIbratnāma, 175–179.
112 From Despair to Liberation

the desert; and he was lucky that the unusual trick he deployed to get
himself to Bukhara worked as well as it did.
Mı̄rzā Mahmūd deserves much credit, however, for making expert use
˙
of the freedoms available to him. In Panjdih, these freedoms were much
more limited than in Bukhara, but even while shackled, Mı̄rzā Mahmūd
˙
evidently had access to a rural postal system – a concession that Khan
Muhammad allowed, we may assume, because he hoped his literate slave
˙
would use it to negotiate for his own ransom. Mı̄rzā Mahmūd also appears
˙
to have enjoyed free social exchanges with his fellow slaves as well as with
various townsmen and visitors, and he made key allies. He draws on the
postal system and his village alliances to effect his escape, as well as to
torment his master. In Bukhara, he enjoys much more extensive freedoms:
Here, he even has the time and the liberty to earn a living above and
beyond the support offered by his various masters. At times, he also has
the freedom to travel around the city on his own. He improves his position
much more quickly here than he had among the Turkmens. He reveals
himself to be an entrepreneur, a skilled artist, and a valuable companion to
Bukharan elites.
The range of different jobs and roles Mı̄rzā Mahmūd occupies during
˙
his time as a slave is revealing on two counts. First, it reveals how the
author’s many talents could, given the right setting, serve to define the
nature of his captivity. Second, it reveals the many sorts of jobs that could
be delegated to slaves in Central Asia. Mı̄rzā Mahmūd worked as
˙
a herdsman, a domestic servant, a water-bearer, a stable-keeper, a doctor,
a painter, a calligrapher, a craftsman, an Arabic instructor, an accountant,
and a poet. Single-handedly, he reveals the social mobility that could be
available to talented, fortunate slaves in an urban environment like
Bukhara. He also reveals the veritable dead end of slavery in the hinter-
lands: For all his talents, there was no work available to him in the village
of Panjdih other than the meanest physical labor. Given that most slaves in
Bukhara and Khwarazm worked in rural settings that were more similar to
Panjdih than to the capital cities, we can assume that his experiences in
Panjdih were closer to the norm. The nature of Mı̄rzā Mahmūd’s survival
˙
strategies, moreover, differed in either location: In Bukhara, he exercised
productive talents; in Panjdih, where those talents were useless to him, he
exercised trickery. Instead of craftwork and poetry, he feigns madness and
plots his escape. His talents define the nature of his captivity, but the nature
of his captivity defines which talents he can use.
Mı̄rzā Mahmūd frequently meets with other Iranian slaves, and we find
˙
them working as messengers, domestic servants, soldiers, and agriculturalists.
 Reflecting on Mı̄rzā Mahmūd’s Ordeal 113
˙
We never meet another slave who had attained such high stations as he
himself achieved in Bukhara, however. The scarcity of educated slaves in
Bukhara factored into his easy mobility, but also played a part in the perpe-
tuation of his captivity: His kind were always in demand, such that letting
a slave with his talents get away was unthinkable. His skills were a blessing
and a curse.
Along with information on slaves’ jobs and conditions of labor, Mı̄rzā
Mahmūd also reveals much about the inner workings of the slave trade. He
˙
shows clearly how the slave trade followed caravan networks from cities
like Mashhad to Bukhara. We learn that the trade in Iranian slaves was
sometimes conducted – at least in part – by Iranian merchants, who
financed caravans and loaded them with diverse shipments of goods for
sale. It was by these same caravan routes that manumitted slaves were
conveyed back to their homelands. Mı̄rzā Mahmūd provides evidence of
˙
merchant-specialists who may have dealt predominantly in slaves, but
upon his arrival in Bukhara, when he is bought and sold repeatedly by
various individuals in quick succession, only one of these individuals is
identified as a “slave-dealer.” This series of sales also hints at a trait slaves
shared with any other commodity: They were sometimes bought and sold
purely in the interests of market speculation. It was easy to find buyers
for Mı̄rzā Mahmūd, both because his initial liaison in town was a well-
˙
connected merchant and because his literacy made him a relatively safe
investment.
To what extent can we believe Mı̄rzā Mahmūd’s narrative? Even if we
˙
allow that some of the episodes Mı̄rzā Mahmūd recounts seem far-fetched,
˙
the remarkable level of detail accompanying his most memorable achieve-
ments lends the narrative an overall atmosphere of earnest documentation.
When he claims to have accompanied Shāhrukh Khan across the region,
for example, he provides extremely specific details concerning shrines and
other major sites that they pass by. When he claims to have been impri-
soned in multiple dungeons, he takes the time to recount conversations
and particulars that are not directly related to the development of his
narrative. Throughout the text, in other words, Mı̄rzā Mahmūd provides
˙
details that one would be unlikely to include merely as fabrications. It is
also worth noting that not all of the tactics Mı̄rzā Mahmūd describes
˙
having used to better his situation cast his wits in a positive light; especially
while in Panjdih, his ploys were more likely to fail miserably than to
succeed.
Nevertheless, Mı̄rzā Mahmūd’s skilled prose alone suggests an author
˙
who may be acutely aware of his audience and their expectations. Such an
114 From Despair to Liberation

erudite man would not have been unaware of genre conventions that were
then prevalent in the realm of “adventure” literature. One particular area
in which embellishment seems likely is in the invariably witty, audacious
repartee that Mı̄rzā Mahmūd delivers at every opportunity. Did he really
˙
insult the Sunni tradition while standing amid a hostile (and probably
armed) group of Sunni Turkmens? Did he really mock the religiosity of
Bukharans in the presence of the Bukharan Amı̄r himself? It is possible, but
seems unlikely. We can be sure, meanwhile, that such dialogue would have
earned appreciative laughs from readers in Iran.
As we shall see in the chapters to come, the best indication of Mı̄rzā
Mahmūd’s reliability is in the numerous instances in which later travelers –
˙
who would never have read his unpublished account – confirm and corro-
borate his claims about the workings of the Central Asian slave
trade. Mı̄rzā Mahmūd himself is remarkable in every respect, but his
˙
experiences, for all their inherent drama, are not altogether unique. He
was not the only Iranian slave to have gained a high station in Bukhara; to
have earned a good living for himself through his talents; to have served
the Amı̄r. And he is certainly not the only slave to have suffered miserably
in the desert, to labor in the hot sun for months on end, and to be beaten
and abused by a cruel master. His account is valuable in part because his
ordeal shares so much with that of other slaves: He provides something
like a voice for those who suffered a similar fate.
 4

The Slaves’ World

Jobs, Roles, and Families

Mῑrzā Mahmūd’s narrative reveals the remarkable range of occupations

˙
a slave could hold in different parts of Central Asia, but slaves in the region
played many other parts as well. In this chapter, I will show something of
the diversity of slaves’ jobs and roles, along with the various environments
in which they lived and performed their work. Some types of work were
designated especially for slaves, while other types of work appear to have
been off-limits to them. Slaves’ origins and capabilities often played a part
in defining the kind of work they did: Russian slaves, for example, were
much more likely to be found doing certain types of labor, while other
types were largely the purview of Iranian slaves. In light of slaves’ varied
labor roles, I will argue that a limited and contingent degree of social
mobility could be available to seemingly any slave, though elite political
posts were reserved exclusively for men. I will also explore slaves’ mar-
riages, examining a variety of possibilities and outcomes related to family
life among slaves.
Considering the diverse roles and environments occupied by slaves
reveals that they were ubiquitous in many parts of the region. They con-
stituted a fundamental aspect of Central Asian societies down to the late
nineteenth century.

LEGAL FRAMEWORKS

The Hanafi Muslim legal lexicon of Central Asia in the nineteenth century
recognized at least six distinct types of slave, distinguished by the means of
their acquisition and by certain circumstances of their use. Each category

115
116 The Slaves’ World: Jobs, Roles, and Families

carried with it particular legal obligations on the part of a slave’s

owner. The Sharh-i mukhtasar-i viqāya-i turkı̄, published in 1901 in
˙
Tashkent, describes these categories, all of which are familiar (though
sometimes under different names) in the broader context of Hanafi law
on slavery:1
(1) Zarkharı̄d. This is the category of slave that, ostensibly, enjoyed the
lowest number of protections. The word literally means “bought for gold,”
and presumably applies to most of those slaves bought in Central Asian
bazaars and caravanserais. Such slaves could be given by their masters as
gifts; pawned; left as an inheritance; rented out for use by others; or made
to constitute part of a dowry.
(2) Khanazād. This category encompasses the children of slaves; the
word means, roughly, “house-born.”2 These slaves could not under any
circumstances be sold at the market. Vambery observed that the law, in this
case, had pervaded social mores as well: “The sale of a khanezad is
regarded as a disgraceful action, and one who commits such an act is
branded as a thief and a robber.”3
(3) Mukātab.4 Slaves in this category could not be sold, given as gifts,
pawned, left as inheritance, or made part of a dowry, because the terms of
their servitude included a contract (kitāba) stipulating that they would be
able to buy their own freedom after a certain period. To accomplish this,
the slave would be allowed to save whatever proceeds he or she could
manage to earn during the period of servitude. (In the context of ancient
Roman slavery, such proceeds were called the peculium.) Vambery
explains that these proceeds often came directly from the slave’s owner:
many slaves, he writes, “receive after a certain time either monthly wages,
or else a share of the produce of the land or cattle committed to their

1
Faiziev, Buxoro feodal jamiyatida qullardan foidalanishga doir hujjatlar, 34–35. This author
adds a seventh category to the six described here, which he calls the shartli qul (conditional
slave). This slave was, according to Faiziev, distinguished by undergoing a fixed period of
service, at the end of which he was freed – presumably without having to buy his manumis-
sion (see Faiziev, 35). We might add a further category to the list if we offset those slaves
given special permission by their owner to transact business on their behalf. Naturally, we
might expect all categories of slave that appear in the traditional Hanafi canon of fiqh to
appear in use in Central Asia in the nineteenth century.
2
Other terms describing the same category include qin and khālis qul (cf. Buxoro feodal
jamiyatida qullardan foidalanishga doir hujjatlar, 34).
3
Vambery, Sketches of Central Asia, 226.
4
This and several other categories in evidence in our Central Asian sources can easily be
found in the traditional Hanafi fiqh canon. See, for example, ‘Alı̄ ibn Abı̄ Bakr Marghı̄nānı̄,
The Hedaya, trans. Charles Hamilton (London: Allen & Co., 1870), second edition, 267;
637.
 Legal Frameworks 117

care.”5 Petrushevsky, writing about slavery in Iran, adds that a female slave
under such a contract (the mukātaba) would traditionally be exempted
from the necessity to submit to sexual intercourse with her master.6
Interestingly, Vambery seems to conflate all Central Asian slaves other
than the khanazād with this category: “As the Iranian is generally more
active and frugal than his Turanian neighbour, the slaves in Turkestan, in
a remarkably short time, get together a little capital. This is employed by
most of them in ransoming themselves from slavery, which they have the
right to do after seven years’ service.”7
(4) Mudabbar.8 Slaves in this category, as with the mukātab, could not
be sold, given as gifts, pawned, left as an inheritance, or made part of
a dowry. This is because the terms of their contract included a tadbı̄r
(pledge) from their master that they would be freed upon their master’s
death.
(5) Umm walad.9 Slaves in this category (which means “mother of the
child”) had borne children by their master. Such a slave likewise could not
be sold, given as a gift, pawned, left as an inheritance, or made part of
a dowry; like the mudabbar, her manumission was guaranteed upon the
owner’s death. We can find at least one famous example of an umm walad
in a proximate context: the Bukharan Amı̄r Muzaffar’s own mother had
been a slave, purchased at the public market.10
(6) Istı̄lād.11 A slave in this category has been impregnated by her
owner. Upon the child’s birth, she will become an umm walad. A slave in
the category of istı̄lād, like an umm walad, cannot be sold, given as a gift,
pawned, left as an inheritance, or made part of a dowry, and is automati-
cally manumitted upon the death of her owner.
These, at least in theory, were the Hanafi Muslim legal categories that
governed the slave system in Central Asia. It must be remembered that the
existence of such legal categories as those above reveals only legal norms,
and not necessarily observable phenomena. The social ideals and customs
indicated by these norms are not often clearly evident in other kinds of
sources – eyewitness reports and autobiographies, for example – that
concern slaves’ experiences. However, the existence of dozens of legal

5
Vambery, Sketches of Central Asia, 223.
6
I. P. Petrushevsky, Islam in Iran (Albany, NY: State University of New York Press, 1985),
157.
7
Vambery, Sketches of Central Asia, 226. Emphasis added.
8 9
See Hedaya, 488; 267; 637. Hedaya, 32; 120; 267; 637.
10
Faiziev, Buxoro feodal jamiyatida qullardan foidalanishga doir hujjatlar, 27.
11
Hedaya, 274; 488; 514.
118 The Slaves’ World: Jobs, Roles, and Families

documents on slavery – such as those compiled by T. Faiziev and discussed

earlier – provides strong evidence that contestations pertaining to slaves
were raised and adjudicated in the region’s Islamic courts. Extant bills of
sale and deeds of manumission signed by Hanafi Muslim judges, mean-
while, suggest that the formal Islamic legal system was used to define who
should be considered a slave.

SLAVES’ ROLES AND OCCUPATIONS

One of the earliest European sources on slavery in Khiva, published in

1815, focused exclusively on the agricultural labor performed by its slaves,
and noted differences between the mechanics of labor in urban and rural
contexts:
The usual work of slaves is as follows: 1) to fertilize and cultivate the available
lands, gardens, and orchards. Land is plowed by a small two-ox plow; in the
cities it is plowed with pickaxes. 2) To sow and harvest grain; in the cities, to
plant vegetables in the gardens, clean the canals and channels, and prepare
them for [the] irrigation [waters]; 3) daily, to grind flour in the hand-mill
(ruchnyi zhernov) and pound grains in the “leg” (nozhnyi) mortar; 4) to carry
out [the necessary work involving] the wagons and agricultural implements; in
a word, to bear all domestic burdens. Thus we have the exercise of the captives,
which continuously occupies them, such that they have not the slightest leisure
time during the day, save for a few minutes to take their nourishment and a few
hours of calm sleep. The slightest mistake or pause in their labor is punished
severely.12

As we will see, the “usual work” of slaves extended far beyond the
agricultural sphere, but the field-labor and grain-processing of slaves
appears to have played a very significant role in Khwarazmian agricul-
ture. Major Blankenagel’, who traveled to Khiva in 1793–1794, pre-
sented the field-labor done by slaves as a central element in the cycle of
production and trade: “The fields are in large part cultivated by slaves.
[Any] one of these [slaves] succeeds so much in this that he can provide
for more than a large family through his field-cultivation, and not
a small fraction of that [surplus] produce is sold off. In this way, regard-
less of the small number of people put to work cultivating the land when
compared to the total number of residents, [the Khwarazmians] have
sufficiently abundant fruits and produce that the surplus can be sold to

12
“Nevol’niki v Khive,” Vestnik Evropy 80:7 (1815), 245.
 Slaves’ Roles and Occupations 119

neighboring Kirghiz [i.e. Kazakhs] and Turkmens.”13 He elsewhere

hints that these fieldworkers were generally Iranian, noting that
a “large part” of the 20,000 Persian slaves he estimated to be in
Khwarazm “tend the fields.”14 Some twenty years later, Murav’ev
would make a similar observation, claiming that Iranian slaves brought
into the region by the Turkmens were “generally the ploughmen.”15
Many slaves worked the land on the large estates and gardens owned by
Uzbek elites. Zalesov reports that some of these wealthy Uzbeks owned
“lots of lands and gardens,” which were “worked almost exclusively by
Persian slaves.”16 The situation was similar in Bukhara, of which
Meyendorff writes: “There is not, I think, a single Bokharian citizen who
is in easy circumstances that does not possess a garden and a villa outside
the town, in which he spends the hot days of the summer. Owners of land
let their property, or else have it worked by slaves.”17 An individual land-
owner could keep a great number of Iranian slaves for this sort of business:
“The wealthy Bokharians,” Meyendorff writes, “possess generally about
40 slaves, but some of the most distinguished, as, for instance, the Koosh-
beghi, have about a hundred, for they require a large retinue, and have,
besides, many gardens and much land, demanding a greater number of
hands to labour.”18 Some Iranian slaves interviewed by Russian border
officials reported that their duties included grain-planting and other agri-
cultural labor, though they generally offered no other information about
the nature and conditions of their work.19
Muhammad Rahim Khan, ruler of Khwarazm, evidently put some of his
˙ ˙
slaves to work on the irrigation canals, and Murav’ev’s description of the
khan’s labor-force hints at the diverse backgrounds of the workers
involved in this aspect of agriculture: “[The khan’s] estates are full of
canals, and are carefully cultivated by slaves, and by Sarts and
Karakalpaks, who have settled and built villages on them, and whose
contribution to the kettle-tax has in consequence been remitted.”20
Lieutenant Gladyshev, visiting Khwarazm nearly a hundred years prior,

13
Blankenagel’, Zamechaniia Maiora Blankenagelia vposledstvii poezdki ego iz Orenburga
v Khivu, v 1793–94 godakh (St. Pb., 1858). 13.
14
Blankenagel’, Zamechaniia Maiora Blankenagelia, 17.
15
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 144.
16
Zalesov, “Pis’mo iz Khivy,” Voennyi sbornik 1 (1859), 288.
17
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara (Madras:
Spectator Press, 1840), 62.
18
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara, 62.
19
See, for example, TsGAKaz 4.1.3646, ff. 74a–76b; TsGAKaz 4.1.2821, 6a–b.
20
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 139.
120 The Slaves’ World: Jobs, Roles, and Families

likewise reported on diverse laborers cleaning the canals, though he iden-

tified them exclusively as slaves, witnessing “Russians, Kalmyks, and for-
eigners numbering three thousand people, of whom he became aware
because in the spring they were put to work cleaning the canals around the
city.”21 It was not only among the Khwarazmians and Bukharans that
slaves were employed in agriculture; we find mention of slaves working
the land for Turkmens too, as Mı̄rzā Mahmūd occasionally had done,
˙
though evidence for agricultural slavery among Turkmens is not very
22
extensive.
We rarely find reports of Russians working in agriculture, though
Witkiewicz met one Russian in Bukhara who had previously worked in
the Khwarazmian khan’s garden. This man, named Igor’, seems not to
have liked the work – or the khan: “[Igor’] escaped from Khiva, where he
had lived with the khan, and he says that he repeatedly planned to use the
hoe with which he worked the khan’s garden to bash the khan’s brains in
[lit. ‘chop the khan’s forehead’], but he would rather not give himself up to
the torture which would inevitably await him thereafter, and so he fled to
Bukhara.”23
Herding and livestock-raising were sometimes undertaken by slaves,
though the scholar Semeniuk cautions that this would have been a rare
occupation for them among pastoral nomads, who had the greatest volume
of herds, and his claim is supported by the relative scarcity of sources that
mention slaves engaged in herding.24 Nevertheless, some slaves mention in
their own autobiographical accounts having served Kazakh masters by
herding livestock.25 Mı̄rzā Mahmūd, likewise, cared for his Turkmen
˙
master’s camels. The Scottish explorer Alexander Burnes met an Iranian
slave who had been put to work herding sheep;26 and the English govern-
ess Lucy Atkinson, while adventuring in the region with her architect
husband, met a Russian enslaved by Kazakhs who had been engaged as
a cowherd among them.27
The military was another sphere occupied by both Iranian and Russian
slaves, though the lower-ranking soldiers were, it seems, more likely to be

21
Ia. V. Khanykov, Poezdka iz Orska v Khivu i obratno, sovershennaia v 1740–1741 godakh
Gladyshevym i Muravinym (St. Pb, 1851), 18.
22
See Turgan-Mirza-Baranovskii, Russkie v Akhal Teke. 1879 g. (St. Pb, 1881), 71.
23
Vitkevich, Zapiski o Bukharskom khanstve (Moscow: Nauka, 1983), 116.
24
Semeniuk, “Likvidatsiia rabstva v Kazakhstane,” 168–169
25
See, for example, TsGAKaz 4.1.2821, 2a–3a; 6a–b.
26
Burnes, Travels into Bokhara, Vol. 1, 256.
27
Lucy Atkinson, Recollections of Tartar Steppes and Their Inhabitants (London: J. Murray,
1863), 289–290.
 Slaves’ Roles and Occupations 121

Iranian (some Iranians held higher military posts, however). The general
predominance of Iranians in the military appears to be a common trait of
both Khwarazm and Bukhara. While travel-reports rarely distinguish among
Central Asian soldiers except in terms of rank, our efforts to ascertain the
military roles of slaves are aided greatly by Central Asian narrative chroni-
cles, such as the Khwarazmian chronicle Firdaws al-iqbāl by Shir
Muhammad Mı̄rāb Munis and Muhammad Riza Mı̄rāb Agāhı̄.28 Slave
˙ ˙ ¯
soldiers are mentioned under several names in Munis and Agāhı̄’s history.
They sometimes appear simply as “qul” (apparently the most neutral and
general term for a slave in Khwarazm).29 We also find reference to
a “mamlūk” engaged in battle,30 and slaves in the personal retinue of the
khaqan are referred to as mamālı̄k-i khāssa.31 Several slave soldiers grouped
˙˙
together are referred to as qullar nökari (“slave retinue”).32 Another group is
referred to as consisting of qullar sipāhi, a term that seems to have denoted
more specifically an organized military detachment.33 A military detach-
ment of slaves is also called, at one point, qullar dastasi.34 We also find slaves
working as envoys, dispatched by powerful amı̄rs – in this capacity, the slave
is referred to as toghma.35 Finally, Khwarazmian khans frequently made use
of a private corps of slave bodyguards; these individuals were called altun-
jilaw (“golden bridle”), and Yuri Bregel notes that they could on occasion
hold posts in the khan’s court.36 It was perhaps this sort of access that
allowed for one of the most significant acts of rebellion in the annals of
Khwarazmian slavery: the assassination, in 1727, of Shı̄r Ghāzı̄ Khan by his
own slaves. (The Firdaws al-iqbāl notes an interesting chronogram of this
khan’s death-date: “Dād az ghulāmān” – “Help against the slaves!”)37
Our European eyewitnesses also confirm the prominent role of Iranian
slaves in the Khwarazmian and Bukharan militaries. Zalesov, approaching

28
Shir Muhammad Mirab Munis and Muhammad Riza Mirab Agahi, Firdaws al-iqbāl, trans.
Yuri Bregel (Leiden: Brill, 1999). In the course of the work, Munis cites a slave in his own
family’s employ as a source: “When [I], this humble [author], was a child, [we] had
a decrepit Mashhad-born [slave] inherited from my [great-]grandfather Ishim Biy. He
would tell: ‘When Mashhad was raided by Shir Ghazi Khan, more than five thousand
men and women were taken into captivity. I was one of them’” (Firdaws al-iqbāl, 58).
29
Firdaws al-iqbāl, 427, 464, 491. 30 Firdaws al-iqbāl, 307. 31 Firdaws al-iqbāl, 317.
32
Firdaws al-iqbāl, 473. 33 Firdaws al-iqbāl, 426. 34 Firdaws al-iqbāl, 317.
35
Firdaws al-iqbāl, 465, 468. However, one further reference to toghma (p. 477) leaves it
open to speculation whether this term could apply also to slaves working in other
capacities, or to slaves generally.
36
Firdaws al-iqbāl, 584n381. Bregel also speculates that they may have been identical with
the qullar dastasi (slave military detachments) mentioned earlier (Firdaws al-iqbāl,
633n847).
37
Firdaws al-iqbāl, 61.
122 The Slaves’ World: Jobs, Roles, and Families

the town of Khiva after an arduous journey through terrible heat and
a veritable fog of dust, arrived at the city walls to find it ranged with
Iranian guardsmen: “At the entrance to Khiva, we saw for the first time
the khan’s regular army, arranged in a row on either side of the gate. Its
guards were up to 50 men, being without exception Persian captives.”38
The army was not exclusively Iranian, however: Murav’ev reports that
Khwarazmian artillerymen were generally Russian slaves, while
Witkiewicz met a converted Russian slave who served as high-ranking
Bukharan artillery officer, though he notes that the head of the Bukharan
artillery – the töpchibashi – was an Uzbek.39 P. M. Stremoukhov, who had
served as director of the Asiatic Department for the Russian foreign
ministry (among other positions), claims that the Bukharan infantry was
recruited almost entirely from the Iranian slave population, of whom there
were more than 10,000 in the armed forces.40 The Russian officer
P. Shubinskii similarly observed that “a considerable portion of the
Bukharan military” were slaves.41
According to Witkiewicz, who met with a yuzbāshi of Bukhara in the
1830s, the Bukharans occasionally made an active effort to recruit fugi-
tives and slaves into the military, though their efforts were, around the time
of Witkiewicz’s journey, meeting with mixed success:
He [the yuzbāshi] said that they are recruiting an army made up of runaway Tatars
and Russian captives, and would soon punish the troublemakers; that after Kurban
Bayram, the Khan himself planned to ride on Shahrisabz. We note that the Khan
actually did recruit into his army a man from among 10 of our Tatar fugitives
(in this case we speak of soldiers), keeping him by means of deception and without
pay; and besides this he unveiled a firman by which all captives in private hands
were invited to run away from their masters and come to the Ark, to the palace,
where they would promptly be recorded as sepoys, as soldiers. This challenge was
taken up by some 25 people, who were held in a most pitiable condition.42

Russian slaves often served the Bukharan and Khwarazmian militaries

through craftsmanship and blacksmithing as well, the latter work being,
according to Murav’ev, nearly monopolized by the Russians in Khiva.43
In Bukhara too Russians were often tradesmen: Burnes met a Russian slave

38
Zalesov, “Pis’mo iz Khivy,” 277.
39
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 153; Vitkevich,
Zapiski o Bukharskom khanstve, 120.
40
Stremoukhov, “Poezdka v Bukharu,” Russkii vestnik 6 (1875), 651.
41
Shubinskii, “Ocherki Bukhary,” Istoricheskii vestnik 7 (1892), 125.
42
Vitkevich, Zapiski o Bukharskom khanstve, 108.
43
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 144.
 Slaves’ Roles and Occupations 123

employed as a carpenter,44 and Witkiewicz, who interviewed a number of

Russians in the city, met several others working in this trade. One of these
craftsmen, named Ivan, may have been a freedman who chose to settle in
the khanate. He had previously lived in Astrakhan and had been taken
captive on the Caspian Sea. He reported living in Khiva for three years, and
then fleeing to Bukhara, where he worked as a carpenter “for the khan,”
making gun carriages for the Bukharan army.45 Another Russian slave had
likewise escaped his master and come to the attention of the khan, being
conscripted into the Bukharan army as well as serving as a shoemaker and
carpenter; at the time Witkiewicz met him, this man intended to build
carriages under Ivan’s supervision.46 A Pole named Mikhal’skii, who
Witkiewicz interviewed, was also working as a craftsman in Bukhara,
having arrived there by a very roundabout route. He had been taken
prisoner in 1812, but somehow escaped to Orenburg, where he evidently
served as a guard along the border from 1816 to 1817. “Here they wanted
to punish him,” Witkiewicz writes, “for the fact that his gun discharged
during a hunt and badly injured his hand; they suspected that he wanted to
make himself ineligible for duty.” Rather than face his punishment, he fled
into the steppe, where he was captured by Kazakhs and sold to the kushbegi
of Bukhara. When Witkiewicz met Mikhal’skii, he was sixty years old, and
was serving as “a shoemaker, carpenter, mechanic, and whatever else.” He
had previously set up a cobbler’s shop in Bukhara along with a Tatar
acquaintance, “but it was a catastrophe: they couldn’t sew a decent pair
of galoshes.” Sometime after this, he cast a pair of guns for the khan. He
was a convert to Islam and had married a Bukharan wife, with whom he
had three children.47
Being a slave of a Bukharan kushbegi could come with notable benefits.
At the time of his visit in 1820, Meyendorff reported that the “whole of the
administration” of the domain was in the hands of the kushbegi’s family
and slaves, a state of affairs Meyendorff considered less than ideal for the
development of civic morale: “Thus we find Bokhara presents a repetition
of the comedy eventually played by every despotic country, having a Prime
Minister possessing unlimited powers, which he either can exercise
44
Burnes, Travels into Bokhara, Vol. 1, 294.
45
Vitkevich, Zapiski o Bukharskom khanstve, 115–116.
46
Vitkevich, Zapiski o Bukharskom khanstve, 115.
47
Vitkevich, Zapiski o Bukharskom khanstve, 115. Demaisons appears to have met this man
too; he writes of him: “This Pole, after many years working for the kushbegi, was quite
fluent in Uzbek and Persian. Like the rest of the slaves in Bukhara, he had to work all day
for his master, who often left him without even any bread” (Demezon, Zapiski
o Bukharskom khanstve, 39).
124 The Slaves’ World: Jobs, Roles, and Families

himself or by his subordinates, who do not possess the noble feeling which
we call patriotism.”48 We may indeed question the “patriotism” of even
highly promoted slaves, but generalized patriotism was surely less impor-
tant than specific loyalty within the administration, a quality which the
appointment of family and personal slaves was probably intended to
foster. As elsewhere in the Muslim world, it was not uncommon to find
slaves occupying administrative posts, as well as undertaking other court
duties. Murav’ev notes some such cases in Khiva,49 while, in Bukhara,
Shubinskii writes that nearly every minor court official (as well as every
palace servant) was a slave.50
The spheres of Islamic jurisprudence and prayer-leading, meanwhile,
serve as examples of trades in which slaves were given little, if any, place.
A child of slaves could take on these kinds of leadership roles, but the ranks
of Islamic jurists, imams, and other high positions within religious institu-
tions seem to have been closed off to converted slaves. One powerful case-
study in the alienation of slaves and former-slaves from the higher echelons
of the region’s religious life can be seen in the social bifurcation of Khwaja-
Eli, a Khwarazmian town known for consisting in large part of sayyids,
who were exempt from most obligations to the khan. M. Alikhanov-
Avarskii visited the town around 1873, and observed its social
composition:
In order to maintain their privileged position, Khwajas [sayyids] give their daugh-
ters in marriage to, and themselves marry, only the descendants of Muhammad, and
carefully guard their city from any foreign elements. In spite of this, there are
several hundred ordinary mortals in Khwaja-Eli – the descendants of slaves who
were released or ransomed themselves, being [therefore] descendants of Persian
captives sold here. These latter ones, of course, do not enjoy the privileges of the
city; they live in a special quarter [of town] and annually contribute about 1,000
rubles to the khan’s treasury.51

Does this mean that those among the manumitted converts to Sunnism and
their descendants converted back to Shiʿism after gaining their freedom?
This is one possibility, but another is simply that the ranks of the religious
elite were closed to them; they may have been regarded with suspicion as
“foreign elements” or – still more likely – held in lower regard due to their
non-sayyid ancestry.

48
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara, 51.
49
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 50; 130; 133.
50
Shubinskii, “Ocherki Bukhary,” 125.
51
M. Alikhanov-Avarskii, Pokhod v Khivu (Kavkazskikh otriadov), 196.
 Slaves’ Roles and Occupations 125

Slaves could, however, sometimes be found working as servants for

religious institutions. They could even be permanently “attached” to
religious institutions by formal endowment contracts. In one waqf deed
dating to 1693, for example, it is stipulated that two families of slaves were
to remain constantly in residence at a large Bukharan mausoleum complex,
and that their responsibilities to the shrine were to pass from generation to
generation. Likewise stipulated in the document is the right of each sub-
sequent generation of slaves to draw their compensation from funds
established by the waqf.52
The fact that slaves and former-slaves are rarely, if ever, seen among the
class of Islamic jurists and imams does not mean that slaves never served as
educators, broadly considered. On the contrary, the education of children
among nomads was not infrequently entrusted to slaves, and it is not
unreasonable to presume that this included some measure of religious
education as well. One Tatar slave among the Kazakhs reports that he
spent his time among them teaching his owner’s children to read and
write.53 Russians also served in this capacity; one Russian slave described
having been treated well by his Teke Turkmen mistress, who tasked him
with educating her children.54 In other contexts, historians have consid-
ered the ways in which slaves’ educating children of a different background
can result in cultural transmission, and it is tempting to assume that
Kazakh children who received some education from Tatars, or Turkmen
children who received some education from Russians, would glean from
them more than just literacy. In the diverse cultural environment of Central
Asia, though, it is difficult to isolate those aspects of culture that may have
had “slave-mentors” as their vector. Even so, tentative hypotheses along
these lines are possible in certain, specific areas: given the prominence of
Russian slaves in Central Asian ironworking and military technology in the
nineteenth century, for example, we might expect these fields to show
a significant Russian imprint.
Slaves could also be used to provide musical entertainment, and we
may assume that these slave-musicians were typically Iranian, as we have
no evidence of Russian slaves performing such a role. It is likely that this
use of slaves was a phenomenon particular to the elite, though we have
strikingly little evidence either way. We lack any sources from Central
Asia which would allow us to reconstruct an environment akin to that of

52
Vakf Subkhan-Kuli-Khana Bukharskogo 1693 g., 213; 230–231.
53
TsGAKaz 4.1.197, 81a–b.
54
A. Rzhevuskii, “Ot Tiflisa do Dengil’-tepe,” Voennyi sbornik 9 (1884), 173–174.
126 The Slaves’ World: Jobs, Roles, and Families

the famous “singing girls” of the medieval Middle East – an environment

in which slaves (and particularly slave-girls) provided the best-known
form of spectator entertainment. Twentieth-century accounts of
so-called bachcha bāzı̄ (“boy play”) in Afghanistan, which often involves
effeminately clothed boys dancing and singing for small audiences, may
describe a practice that existed also in nineteenth-century Central Asia,
but there are few reliable reports of slave-entertainers of any age or
gender that we can date to that period. Meyendorff offers one of the
few accounts along these lines, providing his evidence mostly as
a template for expressing a more generalized moral outrage of the kind
so common in the era’s travel literature: “I once asked a young Bokharian
of good family,” he writes, “of what his amusements consisted; he said
that he gave midday dinner parties accompanied by the music of his
slaves; further, that he sometimes attended the chase; and, lastly, that
he kept Jawanis or boy favourites. The calm and unhesitating way he told
me this astonished me, and proved to me how well acquainted they are
here with the most horrible of all vices.”55
In any event, slaves’ work was often defined by what they were capable
of, and we may assume that a slave who was a competent musician or
dancer would have been put to work doing what he or she did best. That
personal talents and perceived competencies could define a slave’s sphere
of action is well-evidenced by the predominance of Russians among mili-
tary engineers, and by that of Iranians working in the royal court (the
stereotype of Iranians as savvy bureaucrats and court attachés was long-
held throughout the Muslim world). But it would be wrong to assume that
most slaves performed only one role, or that they occupied only one sphere
of action (agriculture, say, or the military). For example, any slave, given
their owner’s consent, could perform work beyond what was assigned to
them if they had the time and the inclination, and they could even legally
keep the additional income they earned if their owner allowed them to do
so. A slave put to work predominantly as a planter could work as
a shepherd on the side, and vice-versa. A slave could also be delegated as
his or her owner’s agent in matters of business, acting in the owner’s place
and enacting the rights of a free person only as a proxy (such a go-between
in the Islamic legal context is typically called a vakı̄l). In Hanafi legal
tradition, slaves could even be charged to speak on behalf of their owners
in court cases, though I have not yet seen any direct evidence of this
occurring in early modern Central Asia.

55
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara, 61–62.
 Slaves’ Roles and Occupations 127

Slaves could also serve as their masters’ agents in ventures whose

legality was more dubious, and indeed they would have been ideal candi-
dates for illicit work (at least from their masters’ perspective). Major
Blankenagel’ recalled the experiences of a Russian slave who was profit-
ably employed in illegal labor:
A Russian named Maksim, who had been slave at Khiva . . . told me that he was
briefly acquainted with a slave named Ivan belonging to a Bukharan kazi, who told
him that, with his owner’s consent, he secretly extracted the metals from gold and
silver ores and got the proceeds from this, and that in half a year he had accrued the
amount – totaling 30 silver coins – [necessary] to buy himself [from his owner, i.e.
to purchase his own freedom] . . . Maksim had often helped Ivan in his work,
sorting the crushed ore and alloying the metals . . . Maksim had also been in the
mines which Khivan Sarts used to work in olden times, where he had found many
[more] of those inexhaustible lumps of ore similar to the ones he had seen [while
working] with Ivan. Though it was strictly prohibited to work in the Khivan mines,
greed compels some to secretly extract the ore and take it back to Bukhara, where
they sell it as a commodity much in demand. Maksim was at Khiva for 20 years, and
he had amassed many skills for enriching himself by means of the land. I tried to
convince him to come back to Russia with me, but he did not agree, fearing that he
would not make it back to Khiva, where he had a wife and children in slavery. All
the same, I promised him both a dual passport and the funds, allotted by Her
Imperial Majesty, [necessarily] to buy these slaves, and he agreed to my proposal.56

Maksim’s experiences underscore the degree to which cultivating skills

could enhance a slave’s circumstances. His account is also remarkable as an
example – a unique one, as far as I have seen – of cooperation and
sociability among slaves from different domains: here, we find
a Bukharan slave teaching a trade to a Khwarazmian slave, with the two
then working side-by-side. The illegal nature of the work in question, as
well as the slaves’ shared Russian background, likely facilitated this osten-
sibly rare kind of partnership.
Slaves’ talents could also directly affect their sale-price, and Meyendorff
claimed that competency in a craft could even double their market value:
“The price of a well-built man is about 40 or 50 Tellas (600 to 800 Francs).
Should he be acquainted with any profession, such as turning, making
shoes, or the work of a blacksmith, then his value increases to about 100
Tellas (16,000 Francs).”57 Beyond this, facility in a trade most certainly
played a part in life beyond manumission. Those former slaves who chose

56
Blankenagel’, Zamechaniia Maiora Blankenagelia, 14–15.
57
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara, 61. Thus,
professional skills could also make it more difficult for a slave to purchase their own
128 The Slaves’ World: Jobs, Roles, and Families

to remain in the region had to find their own way – and often did – using
either the skills that they had before their capture, or those they acquired
during their captivity. From this we can hypothesize that those manu-
mitted slaves who voluntarily chose to stay in Khwarazm, Bukhara, or
the steppe after their release were more likely to be skilled laborers, or at
least laborers who had found a niche, such as the Russian slave Ivan,
mentioned earlier, who continued working in the Khwarazmian mines
after buying his own freedom, or the Persian slaves who stayed on as
court functionaries after spending much of their captivity at work in
a similar capacity.

THE ROLES AND POSITION OF ENSLAVED WOMEN

The literature on enslaved women in Central Asia is quite meager when

compared to that concerning male slaves. Elsewhere in the Near East,
enslaved women were extensively involved in agricultural labor, so that
we might expect to find a similar trend in Central Asia, but mentions of
enslaved women working in the fields are scarce in the sources I have seen.
Nevertheless, some important features concerning enslaved women’s
spheres of action emerge from the scattered evidence available, and
a number of women left their own accounts of their captivity in the form
of interviews at the Russian border.
The most commonly discussed subject in our sources concerning
enslaved women in the region, particularly in travel literature and eye-
witness reportage, is marriage: enslaved girls and women were purchased
to be married to male slaves, and they were sometimes married by free
men. They could also serve as concubines, particularly for elites. That
enslaved women in the Muslim world could occupy intimate and central
roles within their owners’ household, even to the extent of bearing and
raising their owners’ children, has inspired some historians to hypothesize
that they were liable to enjoy better treatment and a greater permanency of
position than their male counterparts.58 In particular, the legal status of
umm walad (“mother of the child”), by which an enslaved woman who
bears her owner’s child is free upon the owner’s death, has often been cited

freedom if these skills were demonstrated before the slave was sold, since the price of
buying one’s own freedom was never less than one’s sale-price.
58
See, for example, Anthony A. Lee, “Enslaved African Women in Nineteenth-Century Iran:
The Life of Fezzeh Khanom of Shiraz,” Iranian Studies 45:3 (2012), 426–427; 437.
 The Roles and Position of Enslaved Women 129

as a key advantage, the likes of which naturally was not available to male
slaves. Permanency of position did not necessarily equate to better treat-
ment, however, and our sources on the Central Asian context often serve
to emphasize elements of social instability and misfortune that were
unique to enslaved women, and from which even bearing their owners’
children served as scant protection. While marriage served to incorporate
them into the household and – especially among nomadic Kazakhs – into
the local community, the death of a husband could spell disaster for an
enslaved woman and reveal the tenuous nature of the bonds her marriage
had forged with the larger society around her. The following two exam-
ples, drawn from the testimony of escaped, formerly enslaved women
themselves, will illustrate this point.
On March 30, 1800, a woman who had fled the steppe arrived at the
Russian border and was interviewed by border officials. She told these
officials that she did not know her own age or precisely where she had
originally come from. She had been with the Kazakh Baybakt clan since her
early childhood, living with a man whose name is recorded as Tulagan
Khudaynazarov. When she had come of age, she learned from this man that
she was a Kalmyk, and that Tulagan had been among a group of Kazakhs
that had taken her and some other Kalmyks captive near the town of
Ural’sk. This group of captives was then divided up, and Tulagan became
her owner. According to the woman, Tulagan eventually came to value her
above his own Kazakh wife. She bore him three sons and three daughters;
two of the sons and two of the daughters died in childhood. In 1790, ten
years before she arrived at the Russian border, Tulagan died, and the
woman and her children fell upon hard times. Five years later, in poverty,
she sold her only surviving son to another member of the Baybakt clan,
named Baymirza, and her daughter was given in marriage to a Kazakh from
the Alachin clan, among whom the woman and her daughter both went to
live. There, they were both pressed into servitude (upotrebili v prislugu)
and suffered abuses whose harshness they could not bear. Finding an
opportune moment, the woman fled to the Russian border at the Nizhni-
Ural’sk Line, from which she was transferred to the Orenburg Border
Commission offices. The officials there, as per usual, asked her if she
wished to be baptized. She declined, and was permitted to reside among
the Muslims living at Orenburg.59
On the very same day, another former slave from the steppe was
processed by border officials, and her story, though different in most

59
TsGAKaz 4.1.198 ff. 36a–b.
130 The Slaves’ World: Jobs, Roles, and Families

details, bears some striking resemblances to the previous one. Her name
was recorded as Chiben, and she was forty-four years old at the time of her
interview. She was a Kalmyk, born and raised near the Chinese border, and
she had been taken captive along with her mother and several other
Kalmyks of both genders by a number of men from the Kazakh Middle
Horde. She was taken, along with her mother and sister, to their captors’
ulus and, soon after, she was traded to a Kazakh from the Little Horde for
nine horses, a fleece-and-fur coat, and a gun. Her mother and sister
remained with the Middle Horde.
When Chiben came of age, she became the concubine (nalozhnitsa) of
her new owner, bearing him two sons and a daughter. Her daughter had
been married off some five years previously – in 1795 – to a Kazakh of the
Jabaltay (?) clan. In the winter of 1799–1800, Chiben’s owner died, and
Chiben (who is referred to at this point in the document as the man’s wife)
remained among the Kazakhs along with her sons. She claimed that, after
the death of her owner, the others in the tribe began to inflict “intolerable
oppression” upon her and her children. She fled to the Nizhni-Ural’sk Line
along with her youngest son, who was then thirteen years old. Her other
son, fifteen years old, remained among the Kazakhs. She too declined to be
baptized, and asked that she might be given leave to visit the daughter
mentioned previously, who had married into another Kazakh clan.60
We find striking common elements in these narratives: Both of these
women were Kalmyks, and both had been taken captive at a young age;
both evidently grew up with the owners whose children they would
eventually bear; both saw their daughters married off to different
Kazakh clans; and, most strikingly, both reported suffering abuse and
mistreatment starting from the time their owners died. It is not clear
whether either woman was given the traditional rights of an umm
walad – that is, the right of manumission after their owners’ death –
though the fact that the first woman was able to sell her own son might
hint at her free status. She was nevertheless forced into servitude once
more when she transitioned between clans. Regardless of their legal stand-
ing in these two communities, it is clear that their long period of concu-
binage or marriage within the clan had not resulted in their full integration
on equal footing with other clan members. Their wellbeing was a function
of their owner’s presence, and in their owner’s absence they faced aliena-
tion, or worse, despite the fact that they had each grown up almost entirely
within the clan community. Their slave status left permanent traces.

60
TsGAKaz 4.1.198 ff. 36b–37a.
 The Roles and Position of Enslaved Women 131

It is conceivable that these women exaggerated the alleged abuses they

suffered in order to play on the sympathies of border officials, but this
would hardly have been necessary: There is no evidence that the treatment
of enslaved people by their owners had any bearing on Russian officials’
reception or treatment of them, either as a matter of policy or as a matter
of record. The fact that both women came to the border only after the
death of their owners suggests, at the very least, that they faced some sort
of unwanted fate within their own communities in widowhood. It is likely
that both were beyond the age of possible remarriage, and that the clan was
unable or unwilling to support them in lieu of their spouses.61
As the following Russian border report shows, however, it was not only
women who could face an undesirable future in the steppe after the death
of a spouse. Manas Sadykov, said to be a Kalmyk,62 was seventy years old
when he was received by border officials on the June 19, 1800. He was
originally from a village near the border with China, and was taken captive
in his early youth by Middle Horde Kazakhs of the Atagay clan. They
changed his name to N___ Sakay, and when he came of age he was married
to a Kazakh woman named Janalia. They had a son and a daughter
together, whose names are recorded as Kunabay and Jaima; the latter
was given in marriage to a Kalmyk former slave who lived with the clan.
When Manas’ wife died, however, he no longer wished to live with the
clan, and he arrived at the Russian border along with his Kalmyk son-in-
law, though his own son, Kunabay, remained in the orda. He requested to
be baptized, after which he was evidently settled in Stavropol. It is possible
that his advanced age made it difficult for him to do any manual labor for
the clan. But ultimately there is not much reason to doubt his own expla-
nation of his motivations: When he lost his wife, he lost his will to stay
with the clan.63
Even the slave-concubines of a khan could face an uncertain future after
their royal owner’s death. G. N. Zelenin, in his remembrances of the
Russian mission to Khiva in 1842, as recorded by I. N. Zakhar’in, indicates
as much in a lurid tale concerning the sexual services that had been offered
to Russian officers during the embassy. We may justifiably cast suspicion
61
More personal factors might also have played a part, of course: In the case of the first
woman, for example, it is possible that her owner’s first wife’s jealousy over their relation-
ship could have had some impact on the way she was treated after his death.
62
If in fact this slave was a Kalmyk, Manas was certainly not the name he was born with:
Manas is, after all, the legendary Kyrgyz hero (identified as a “Noghai” in earlier versions)
who, according to the epic of the same name, battled Kalmyks and slaughtered in great
numbers. It is possible that his Kazakh owners renamed him ironically.
63
TsGAKaz 4.1.198 f. 54a.
132 The Slaves’ World: Jobs, Roles, and Families

upon Zelenin’s story, given its resemblance to the popular Orientalist

erotica of the same period, but it is nevertheless a unique account, not
altogether impossible, and it is worth reproducing here. The story begins
with a Tatar that Zelenin met in Khiva:
It turned out that he was a runaway soldier of ours, a Tatar and a native of
Orenburg. Serving in the Orenburg garrison, he had often met with Zelenin on
the streets of the city and remembered his face, being that of an officer to whom
he had needed to doff his cap. He now had his own house in Khiva, and he already
had a family there. Inviting his ‘countrymen’ over [to his house], he treated them
to tea and asked that they come back often. But Zelenin was afraid to visit him
again, as the neighboring Khivans cast very unfriendly glares upon this ‘Russian,’
and the Tatar’s home was on the outskirts of the city, along the ravine. This
[runaway] soldier was later [seen] in the residence of the embassy, and he offered
its occupants – all of them young and single people – quite intimate services. But
such a provider’s place was already occupied. The garden, in which a reserve
‘palace’ of the Khan was situated (and where our embassy was housed), was
overseen by a special gardener from among the Persian captives, a man who
knows his trade well. And here, his wife, also a Persian captive, offered the officers
[of the embassy] her confidential services, costing only a poluimperial: in the
Khan’s garden, at night, accompanying this lady gardener, a female figure would
appear in a veil, a silk robe, and shalvar [loose-fitting pants], wrapped in muslin
and in a Persian scarf. Upon demand, several such figures could appear in the
garden at once. In strict secrecy the Persian lady [gardener] would make assur-
ances that the mysterious strangers were the former wives of the dead Khan,
whose material circumstances under the new Khan were nothing to envy . . . It is
quite possible that the lady gardener exaggerated the qualities and circumstances
by which she came to recommend her ladies, and that they may have been simply
Persian slave-girls, maidservants from the former Khan’s harem; but their cos-
tumes were always very luxurious, and simple slave-girls could not have owned
such things.64

Whether we choose to believe the claims of the Iranian gardener (or, for
that matter, of Zelenin himself), the tale offers another possible answer to
the question of what happened to enslaved wives and concubines once
their owners died. It seems intuitive to presume that enslaved women, even
once manumitted, did not share the same potential for social mobility as
their male counterparts, and indeed evidence of their mobility – beyond
their place in an individual household – is largely lacking. There are few
reports of enslaved women taking up crafts or trades, and if the lack of
reportage is at all indicative of real social circumstances, then it is possible

64
Zakhar’in, “Posol’stvu v Khivu v 1842 godu,” Istoricheskii vestnik 11 (1894), 442.
 The Roles and Position of Enslaved Women 133

that enslaved women were also less likely to secure the earnings necessary
to buy their own freedom.65
In this light, it is not surprising that, of the several Russian slaves
interviewed by Witkiewicz in Bukhara, the individual in the saddest cir-
cumstances was the lone woman among them. She was “some major’s
daughter,” then fifty years old, who had been captured when she was
fifteen. She had been manumitted at some point, but her present occupa-
tion when Witkiewicz met her was, according to him, nothing more than
“following after and imposing herself on all [Russian] newcomers.”66
Murav’ev too reports on the fate of enslaved women after their owner’s
death, though he seems to conflate the fate of enslaved women with that of
women generally. “No one, not even the nearest relation,” he writes, “is
permitted to enter the female apartments, and the women are condemned
to a life of the strictest solitude and most dreadful ennui. They are slaves,
and on a man’s death, the son has a right to sell them at pleasure.”67
According to Witkiewicz, at least some manumitted Iranian women in
Bukhara were able to remarry, presumably of their own volition: “Most
of the captives in Bukhara are Persian, and a significant portion of the
current population of Bukhara is descended from the mixing of Tajiks and
Uzbeks with freed Persians. One must, however, note that the Uzbeks will
take for themselves the daughter of a Tajik or even a Persian, but they will
not give their own daughters [in marriage] to them.”68
The death of an elite personage could leave a great many enslaved
women in a precarious position. One khan of Khwarazm, writes the
American journalist Januarius Aloysius MacGahan, had four wives and
“about a hundred slave women; he seems to have some from each of the
races that are found in his dominions. The exact number I did not ascer-
tain; the Khan himself one could not ask, as it is considered extremely
unpolite, in Central Asia, to make any mention to a man of his wife, or
wives.”69 Vambery reports the same number of “legitimate” wives for

65
There is some evidence, however, that in rare cases women were able to purchase them-
selves from their owners. One such case is confirmed by a manumission contract preserved
in V. L. Viatkin’s collection of sixteenth-century legal documents from Samarkand, in
which an Afghan woman bought her own freedom at the age of twenty-five (Kaziiskie
dokumenty XVI veka [Tashkent, 1937], doc. no. 16, p. 49).
66
Vitkevich, Zapiski o Bukharskom khanstve, 117.
67
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 160.
68
Vitkevich, Zapiski o Bukharskom khanstve, 115.
69
MacGahan, Campaigning on the Oxus, and the Fall of Khiva (London: Sampson Low,
Marston, Low, and Searle, 1874), 279. According to Joseph Wolff, the Bukharan Amı̄r
Nasrullah likewise had four wives – “but,” Wolff writes, “it is said that his wives hate him,
134 The Slaves’ World: Jobs, Roles, and Families

Muzaffar al-Dı̄n, Amı̄r of Bokhara, who, he writes, “has (for it is a custom

of his religion) four legitimate wives and about twenty others, the former
natives of Bokhara, the latter slaves, and, as I was told seriously, only
employed to tend upon the children, of whom there are sixteen, ten girls
(but I beg pardon, princesses), and six boys (Tore).”70
As little as we know about the fate of manumitted women, it is no
clearer from our sources what sort of work women tended to do while still
in slavery. While we know they bore children for their owners, it is not
even entirely clear that enslaved women would have been primarily
responsible for raising these children. The lack of exposés from inside
the harem should not surprise us, despite the comparatively vast literature
on slave-women – or “odalisques” – from elsewhere in the Muslim world
(particularly the Ottoman Empire). It must be remembered that most of
our literature on the subject is entirely fictional, little more than the
product of European fantasies, and consists of speculation rather than
eyewitness reportage. Precious few foreign men ever gained admittance
to the female quarters of an elite Ottoman residence, and the same may be
assumed of male travelers in Central Asia. Among the few travelers who
purported to describe the functioning of a Central Asian harem, we may at
least note the report of Vambery, who offers the dubious claim that most
Uzbeks had little taste for Iranian slaves as concubines because they did not
find Iranian women sufficiently beautiful; he likewise notes, more plausi-
bly, that relative poverty prevented most Uzbeks from indulging exten-
sively in polygamy with slave-women:
As for the female slaves, they do not by any means enjoy the position which is
occupied by the daughters of Circassia and Georgia in the harems of Turkey and
Persia. On the contrary, their position is rather to be compared with that of the
negresses in those countries. It is very easy to explain why. In the first place, the
daughters of Turkestan correspond better to the ideas of beauty entertained by
Oezbegs and Tadjiks than the Iranian women, who with their olive complexions
and large noses, would never bear off the apple of Paris from the fair, full-cheeked
Oezbeg women. In the second place, in consequence of their poverty the

and that they are Persian slaves” (Wolff, Narrative of a Mission to Bukhara in the Years
1843–1845, to Ascertain the Fate of Colonel Stoddart and Captain Conolly, Vol. 1
[London: John W. Parker, 1845], 327).
70
Vambery, Travels in Central Asia (New York: Harper & Brothers, 1865), 226. Elsewhere,
Vambery complicates the notion that slave-wives were not “legitimate,” writing that “in
Bokhara, where the Oezbeg aristocratic is of little moment by the side of the predominant
Persian element, the sovereigns often take slaves for their lawful wives. Such was the
mother of the present Emir, such is one of his wives, both of them of Iranian origin”
(Sketches of Central Asia [London: W.H. Allen & Co., 1868], 221).
 The Roles and Position of Enslaved Women 135

inhabitants of Central Asia do not indulge in polygamy to such an extent as the

Mohammedans of the West. Besides this, the Oezbeg has generally too much
aristocratic pride to share his bed and board with a slave, whom he has bought
for money. In Bokhara it is true that we find instances to the contrary, but that is
only among the high functionaries of state, and even they only take such women as
have been brought as children into the country. In the middle classes such mis-
alliances are very rare phenomena. Besides, marriage is much easier here than in
other Mohammedan countries. Hence female slaves are kept only as articles of
luxury in the harems of the great, or as domestic servants.71

Vambery’s claims about the rarity of Iranian wives and concubines con-
trasts with most other accounts, but it cannot be dismissed out-of-hand:
there is simply too little information on the labors of Central Asian slave-
women to decisively confirm what sort of domestic roles they played and
what place they tended to have in the household. It is plausible, at least,
that many enslaved women served as maidservants, as Vambery suggests.
Concerning Khwarazm, Zelenin recounts that “with [slave] girls, their
dealings were much simpler: the beautiful ones among them filled the
harems of Khivan nobles and wealthy merchants, and the ugly occupied
[the same place, but] as slave-girls and servants.”72
A still smaller body of evidence is available concerning the role of
enslaved women among the Turkmens. Eyewitness and secondhand
reportage on the subject is, moreover, sometimes contradictory, and it
is thoroughly permeated with the tone of contempt and derision com-
mon to nearly all nineteenth-century writings on the Turkmens, whether
it be English, Russian, or Iranian. General agreement exists, at least, that
slave women often became concubines among the Turkmens, as
Grodekov claims for the Teke: “The women are the chief prize, becom-
ing, on their arrival at the aoul, the concubines of their captors; their
children being brought up in slavery.”73 Rzhevuskii commented on these
arrangements in the larger context of Turkmen marriage and social
organization:
Turkmens can have several wives, and, if they have sufficient funds, they set out
several tents, one for each wife, arranged near one-another. In conditions of
poverty, they all live under the same roof. Divorce is completely dependent on
the husband[’s will]. Aside from their wives, Turkmens often associate with their
female slaves, as a consequence of which all the Yomuds are divided into the kul and

71
Vambery, Sketches of Central Asia, 218–219.
72
Zakhar’in, “Posol’stvu v Khivu v 1842 godu,” 445.
73
Grodekoff, Colonel Grodekoff ’s Ride from Samarcand to Herat (London: W.H. Allen &
Co., 1880), 134.
136 The Slaves’ World: Jobs, Roles, and Families

the iz. To the first category belong all who are born of slave women; to the second,
of Yomud mothers. The kul’s inheritance is half that of the iz.74

Whether Turkmens could actually marry their enslaved women remained

a matter of debate, however. Some claimed that this practice was strictly
forbidden, as in the following – highly suspect – report on the Teke:
Custom prohibits the Teke from marrying slave-women, and this is one of the
most prominent motifs of Teke poetry. The beautiful, delicate Persians strongly
tempt them, but their law does not permit the Tekes to enrich their tribe with
cross-breeds, and for that reason, in those cases where a love-struck gentleman
exercises the right of the victor by impregnating his captive and she gives birth,
the baby is either killed or sold into slavery. Generally speaking, the family
environment in Akhal-Teke is fairly wholesome, although (or perhaps because?)
husbands often beat their wives.75

Suffice to say this is the only extant source that claims the Turkmens killed
their own children, and I believe we may dismiss the notion as a silly
fantasy. At the opposite extreme is the British traveler John Wood, who
claimed not only that Turkmens could marry their slaves, but even that
their free-born wives became slaves upon marriage. “The custom of man-
stealing,” Wood writes, “appears to have smothered every better feeling,
and the practice of trafficking in human beings extends even into their
domestic arrangements; for their wives are as much articles of property as
their slaves, and are bought and sold with the same callous indifference.”76
Suffice to say that efforts to better understand the predicament of enslaved
women among the Turkmens must await the revelation of better sources
than these, if such sources exist.
The relative lack of eyewitness information on enslaved women mostly
indicates an imbalance of access, and perhaps an imbalance of interest on
the part of witnesses, and should not be thought to suggest a scarcity of
women among the ranks of the enslaved. Among the Kazakhs, Semeniuk
alleges that women and children formed the predominant enslaved demo-
graphic, and that slaves were therefore predominantly found occupying
women’s and children’s roles in the household.77 This makes some intui-
tive sense given that women and children were less likely to attack their
own captors (a constant danger) or to successfully flee from them, and also

74
Rzhevuskii, “Ot Tiflisa do Dengil’-tepe,” 292.
75
“Zavoevanie Akhal-Tekinskogo oazisa,” Istoricheskii vestnik 7 (1881), 823.
76
Wood, Journey to the Source of the River Oxus, 133.
77
Semeniuk, “Likvidatsiia rabstva v Kazakhstane,” 175. Stremoukhov made a similar claim
concerning the traffic in women in Bukhara (“Poezdka v Bukharu,” 690).
 Marriage among Slaves 137

because slave-women could carry a price similar to that of men on the open
market: they were, in other words, easier to obtain and no less valuable
than men.78 But there are few eyewitness reports by observers who actually
witnessed enslaved women and children at work in Kazakh households
and communities, and few of the formerly enslaved women interviewed at
the Russian border described the sort of work they did. Moreover, as far as
I have seen, women are in the minority among those who escaped captivity
in the steppe. There are a few possible reasons for this. First, women may
have been less inclined to attempt the dangerous and arduous journey to
the nearest Russian fortification. Second, given that many – perhaps even
most – enslaved women in the steppe bore children by their owners or by
others, their children may have served as an additional incentive not to flee
the clan.

MARRIAGE AMONG SLAVES

Oftentimes these children were the product of marriages that were

arranged between enslaved men and women. Integrating slaves into the
social fabric of the community, however temporarily, was, according to
some eyewitnesses, the main reason why male slaves were permitted to
marry at all. “Desiring as much as possible to tie the slaves to their new
fatherland,” writes Zalesov of Khwarazm, “the Uzbeks usually try to marry
them off, finding them brides who, for the most part, are also Persian
captives from their master’s harem, or else dilapidated beauties from the
Khan’s harem.”79 Because the children of male slaves were likewise slaves,
the owners also benefited from their slaves’ marriages, as these unions
could serve to produce more human property for them. It is likely for this
reason – rather than pure generosity – that we find reports of some owners
purchasing women specifically for the purposes of marrying them to their
male slaves. Burnes, for example, interviewed an Iranian slave from
Mashhad who, he writes, “gave us a favourable account of his treatment
by his master, who intended to purchase a wife for him; but he had no hope
of his liberty.”80 In Khwarazm and Bukhara, it seems that the purchased
wives were generally Iranian, though they could be married to male slaves
78
Some reports on Persian–Turkmen conflicts exist in which it appears that only women
and children were taken captive – this seems especially to have been a common strategy
inflicted by Persian troops on the Turkmens. See, for example, Russko-Turkmenskie
otnosheniia v XVIII–XIX vv. (Ashgabat, 1963): doc. no. 299 (419–423); doc. no. 373
(453–454); doc. no. 373 (505–506).
79
Zalesov, “Pis’mo iz Khivy,” 286. 80 Burnes, Travels into Bokhara, Vol. 1, 256.
138 The Slaves’ World: Jobs, Roles, and Families

of any background. Meyendorff, among others, observed that enslaved

Iranian women were used for the purposes of increasing the general labor
pool: “[W]e find among the slaves in Bokhara Hezurehs, Chitrars,
Siapuchs, and even Georgians; their number never diminishes, for they
are given Persian women in marriage, it being the interest of their masters
to keep them up.”81
Among the Kazakhs, there is evidence of a remarkably different
dynamic: while I have found no indication of slaves in Khwarazm and
Bukhara marrying free, non-Iranian women (free Uzbeks, for example), it
was apparently not unusual for slaves in the steppe to take free Kazakh
brides. Particular customs governed this phenomenon. In at least some
parts of the steppe, if a free female member of the orda married a slave,
their children would remain technically free, but they would likewise serve
the family that their enslaved father served.82 Different customs applied to
enslaved women who married free Kazakh men: These women would
evidently convert and be manumitted as part of the terms of their marriage,
but their original owner (if this person was not the woman’s new husband)
could decide even then whether to send the woman to her husband or,
rejecting the marriage, keep her in servitude.83
Among the Kazakhs, it seems that marriage could be a very effective
means of integrating male slaves into the community. The most vivid
indication of this fact is the high incidence of married slaves who, having
been surrendered by their owners and manumitted by decree at the Russian
border, chose to return to their homes among the Kazakhs rather than
opting to be repatriated elsewhere. A few exemplary individuals, all of
whom were interviewed in September of 1852 and all of whom are
referred to as “Afghans” in records of border officials, will serve to illus-
trate this and other common features of slave marriages among the
Kazakhs:
(1) An Afghan, forty-five years old, was surrendered to the Russian
authorities by his owner and interviewed by Russian border officials. He
did not know the details of how he had been taken captive, only that he had
initially been sold by Turkmens to a Kazakh named Khuday-Berdi for sixty
tillas, and that, when he was four years old, he was sold to his present
˙
owner at the same price. He lived with his present owner from that time
on. During this period he received his own tent (otdel’naia kibitka),
81
Meyendorff, Journey of the Russian Mission from Orenbourg to Bokhara, 61–62.
82
S. Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka i Bukeevskoe
khanstvo (Almaty: Arys, 2009), 281.
83
Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka, 281.
 Marriage among Slaves 139

married a Kazakh woman, and raised three sons and two daughters with
her. Upon receiving his Russian-decreed freedom at Orenburg, he evi-
dently returned home again to his family among the Kazakhs.84
(2) An Afghan, thirty-four years old and originally from Herat, told the
Russian authorities that he had been captured by Turkmens nineteen years
previously while he was out watching after cattle. These Turkmens carried
him along with them for the next two years before finally trading him to
a Kazakh for some quantity of sheep. He lived with this Kazakh from that
time on. He married a Kazakh woman, though he does not report having
any children with her. Upon being surrendered by his owner and receiving
his Russian-ordained manumission, he too seems to have simply returned
home again to his wife among the Kazakhs, choosing not to return to
Herat.85
(3) An Afghan, forty years old, whose name was recorded as Baydullah,
was also originally from Herat. He was taken captive by Turkmens at ten
years old, and he remained among the Turkmens for the next five years,
after which they sold him to his present Kazakh owner, with whom he lived
for more than two decades before being surrendered to the Russian autho-
rities. He had a Kazakh wife and a daughter, and opted not to be returned
to his original home in Herat.86
(4) An Afghan whose name is recorded as Tursun87 was twenty-two
years old when he was surrendered by his Kazakh owner to the Russian
authorities. He reports that he came from among the Afghans (iz afgant-
sev), but does not specify a town or village. This is likely because he was
captured early in his youth, since he claimed not to know the details of his
original captivity or who sold him to his present owner. He was married
to a Kazakh woman, with whom he had a son. He too reported possessing
his own tent, and he did not wish for border officials to return him to his
original home (of which he seems to have had no knowledge in any
case).88
According to the Soviet historian Zimanov, slaves among the Kazakhs
who gained the confidence of their owners were sometimes given not only
their own tents but even their own livestock. Those who attained this level
of autonomy were generally withdrawn from the pool of slaves who were
sold and resold as commodities, though their owners retained this right at
all times.89 It is not clear whether this custom had strong parallels among
84
TsGAKaz 4.1.3646 ff. 47b–48a. 85 TsGAKaz 4.1.3646 ff. 48a–b.
86
TsGAKaz 4.1.3646 ff. 48b–49a.
87
He was most likely given this name by the Kazakhs. 88 TsGAKaz 4.1.3646 f. 49b.
89
Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka, 278.
140 The Slaves’ World: Jobs, Roles, and Families

the slaves of Khwarazm and Bukhara, though Alexander Burnes reports

having met an enslaved Russian in Bukhara who had not only a wife and
child, but even his own slaves.90 In the context of the khanates too there
can be little doubt about the effectiveness of marriage in developing a male
slave’s local loyalties – if not to his owner, then at least to his new family.
When Joseph Wolff, the missionary and adventurer, met with a Bukharan
official concerning, among other things, the release of Russian and Persian
slaves, the official was eager to show him a selection of those Russians who
had no desire to return to Russia; their disinclination was couched in terms
that should not surprise us: some claimed to be deserters, but others
replied, “We are married here, and have wives and children.”91

LIMITED SPACES FOR SLAVES’ AUTONOMY

Evidence of autonomy and independent initiatives among the enslaved are

not abundant in our sources, which generally lack the depth of detail we
find in Mı̄rzā Mahmūd’s narrative. Nevertheless, the evidence that exists –
˙
slaves entrusted with their own households in the steppe, for example, as
well as some of the examples I will offer here – are helpful in reconstruct-
ing the possible range of slaves’ experiences. On the more dramatic end of
the spectrum, we have at least one account, offered by Murav’ev, of
a woman who – like Mı̄rzā Mahmūd – was able to convince her master
˙
to re-sell her in order to enhance her living conditions. Despite her
master’s eventual consent, the circumstances reported by Murav’ev, who
traveled with the woman and her owner in a caravan, are tragic:
On the leading camel sat Fatima, by birth a Kurd, and formerly Said’s father’s
concubine. She had been 12 years his slave, and now, wishing to improve her lot,
begged her master to sell her in Khiva. On his refusal to part with her, she
threatened to commit suicide; so he gave in. It is incredible what this woman
endured on the road. Clothed only in rags, she led the caravan day and night,
hardly slept or ate, and, when we halted, attended to the camels, cooked her
master’s food, &c . . . Said himself became disrespectful, as the following incident
will show: He had taken his female slave Fatima to all the villages and bazars about,
but could nowhere get his price for her. This poor woman lived in the same room
with all the rest of the party, but when they went out she used to be so ill-treated by
the other people in the fortress, that I had frequently to send Petrovitch to drive
them off. On one of these occasions the ruffians behaved so badly to her, that

90
Burnes, Travels into Bokhara, Vol. 1, 294.
91
Wolff, Narrative of a Mission to Bokhara, Vol. 2, 42.
 Limited Spaces for Slaves’ Autonomy 141

Fatima fled, and declared that she would certainly take her own life if she were not
sold soon. When Said returned I represented the state of things to him, and begged
him to have it altered, and to sell Fatima, whose presence was a scandal to us all. He
listened to me quietly, then rose up and said, “Farewell Murad Beg! My service is at
an end, for since this is to be your treatment of me, I leave you. Fatima is my slave,
and I shall sell her whenever and to whomsoever I please.” With those words he
went away, but I called him back, and he came, probably supposing I would ask his
pardon for my interference . . . I made the matter up with him, and next day he sold
Fatima.92

Vambery offers another memorable (but, like all of his reportage, poten-
tially fictional) story concerning the influence enslaved women could have
over their owners. He claims that two of the traveling-companions who
accompanied him for a short while on the road to Khiva were a Turkmen
and his wife, the latter having convinced her husband to travel there. She
had been captured some time previously in a surprise raid, during which
her husband had been badly wounded. She undertook the journey,
Vambery tells us, to ascertain whether her former husband was still alive,
what person or persons had bought her children, and what had become of
her 12-year-old daughter. “The poor woman,” Vambery writes, “by extra-
ordinary fidelity and laboriousness, had so enchained her new master, that
he consented to accompany her on her sorrowful journey of enquiry. I was
always asking him what he would do if her former husband were forth-
coming, but his mind on that point was made up – the law guaranteed him
his possession.”93
Such rare anecdotes aside, there can be no doubt that the most common
and widely documented means by which both enslaved men and women
demonstrated their autonomy was by fleeing from their owners. Along
with evidence from the many slaves who fled to the Russian border (the
subject of the following chapter), stories concerning escaped slaves also
appear in many travelers’ reports, such as the narrative which emerged in
a colonial court case witnessed by J. A. MacGahan. A Russian colonel
presided over the court, into which an elderly Khivan woman entered,
“leading a lubberly-looking young man about fourteen, and bowing almost
to the earth at every step, and [she] addressed the Colonel, whom she took
for General Kaufmann, as the ‘Yarim-Padshah,’ or half-emperor, which
title the Colonel accepted with grave composure.” She wore a tall white
turban and a “dirty-looking tunic,” and she presented the colonel with an

92
Murav’ev, Muraviev’s Journey to Khiva through the Turcoman Country, 31n; 56.
93
Vambery, Travels in Central Asia, 76.
142 The Slaves’ World: Jobs, Roles, and Families

offering of bread and apricots before making her case. She claimed that her
son, the “gawky boy” next to her, had been robbed of his wife by “a vile
thieving dog of a Persian slave,” who had also stolen her donkey and used it
to carry the boy’s wife away. “So he is three times a thief,” the colonel
replied. “He stole the donkey, the girl, and himself.” The colonel then
asked the woman about the identity of the stolen bride. “She is a Persian
girl,” the woman replied. “I bought her from a Turcoman who had just
brought her from Astrabad, and I paid fifty tillahs for her. The dog of
a slave must have bewitched her, for as soon as she saw him she flew into
his arms, weeping and crying, and said, ‘he was her old playmate.’ That
was nonsense, and I beat her for it soundly. The marriage was to be
celebrated in a few days; but as soon as the Russians came, the vile hussy
persuaded the slave to run away with her, and I believe they are as good as
married.” In the end, MacGahan writes, the escaped slaves were never
returned to the woman and her son, and neither was the donkey.94
Collusion among slaves was a common means of escape both in settled
lands and in the steppe, with many of those who came to the Russian
border arriving in the company of others.
Beyond fleeing from their owners, there were other, less drastic – albeit
also less well-evidenced – means by which the enslaved endeavored to shape
their own experiences. For example, those up for sale or newly purchased
could try to influence their appeal to certain buyers by modulating their
behavior.95 Some documents suggest that a slave showing defects that became
obvious only after his or her sale could be returned in exchange for a refund,
or for a new, more satisfactory slave. A terse Bukharan document in Persian,
published in facsimile by S. K. Ismailova, reveals this sort of exchange; its
author writes: “[To] Muhammadshah Bay: a female slave given by us to
˙
Qarategin has turned out to be insane [divāna buda ast]. On receiving her,
you should punish her and keep her for yourself. In her place send a female
slave [chūrı̄] brought from Nurāt.”96 That such a “refund” could be imposed
through legal means is evidenced by another Bukharan document, published
in facsimile by Faiziev, in which the buyer of a female slave realized only too
late that his purchase had a serious skin-disease and took his complaint before
the court in hopes of getting his money back.97 It is easy to imagine slaves

94
MacGahan, Campaigning on the Oxus, 199–201.
95
See, for example, Zakhar’in, “Posol’stvu v Khivu v 1842 godu,” 285.
96
Ismailova, “Dokumenty o rabstva v bukharskom khanstve v XIX– nachale XX v.” Izvestiia
Akademii nauk Tadzhikskoi SSR, otdelenie obshchestvennykh nauk 2/72 (1973), 27.
97
Faiziev, Buxoro feodal jamiyatida qullardan foidalanishga doir hujjatlar (XIX asr), doc.
no. 7, 102–103; 25.
 Limited Spaces for Slaves’ Autonomy 143

making use of such legal processes to change owners, perhaps by faking

insanity or incompetence, or otherwise by altering their behavior and mana-
ging their own conduct.
Mı̄rzā Mahmūd’s narrative, meanwhile, points toward the different
˙
ways in which slaves could modulate their own skills and talents in the
presence of their owners in order to resist – or change the nature of –
certain types of work or work-regimes. Abilities that might be put to
uncomfortable use were likely hidden; skilled artisans may have been apt
to conceal their craft if they feared it would increase their owners’ demand
for their labor without any further benefit to themselves. Women faced
with concubinage could endeavor to make themselves less desirable to
their owners, or to make themselves more desirable if it might result in
better treatment.
In this chapter, we have seen something of the diversity of slave labor,
slaves’ roles, and slaves’ family life in Central Asia. Slaves worked the land;
they processed the land’s produce; they worked as blacksmiths, artillery-
men, guardsmen, herdsmen, maidservants, harem attendants, court atten-
dants, shrine attendants, concubines, well-diggers, entertainers,
carpenters, leather-workers, textile-workers, teachers, soldiers, officers,
and administrators. They could be appointed as their owners’ agents in
commerce. They were wives and mothers. Even if an individual owned no
slaves, he or she would have been immersed in a world partially made and
thoroughly inhabited by slaves.
Unlike slaves who were specifically acquired to tend to a monoculture,
such as slaves on colonial-era Caribbean sugar plantations, each individual
Central Asian slave would likely have been put to work in a variety of tasks,
occupying multiple social spheres and spaces throughout his or her life-
time. Enslaved men had a particularly broad range of mobility, as they were
able to work more extensively outside the home and in a greater variety of
trades. The work a slave performed could be defined by his or her cap-
abilities, be it metalworking or leading a military regiment. Generally
speaking, though, greater mobility should not be mistaken for greater
autonomy: Work was delegated to slaves by their owners, and slaves’
abilities to earn their own autonomous revenue or define their own labor
regimen depended entirely upon their masters’ will.
Having observed in this chapter some of the many roles and spheres of
action occupied by Central Asian slaves, the next chapter will provide
a closer look at one of the most dramatic means by which slaves asserted
their autonomy: by fleeing from their masters.
 5

From Slaves to Serfs

Manumission along the Kazakh Frontier

Throughout the eighteenth and nineteenth centuries, thousands of slaves,

predominantly Iranians and Kalmyks, issued from the steppe and arrived
at the Russian border. At first, these slaves were exclusively refugees who
had escaped their owners. Later on, following the imperial decree that all
steppe peoples must give up their slaves to Russian officials, those slaves
who arrived at the border usually came with their owners in tow. Two
difficult decisions faced the officials during this period: they had to decide
who was really a slave, a process which involved defining slavery itself, and
they had to decide what was to be done with those slaves who, having been
freed either by escape or by decree, were now at the disposal of the Russian
Empire. The results of these negotiations were often surprising, and the
border authorities’ internal records preserve the story of how some slaves
became serfs, and how some slave-owners became foster parents. This
chapter considers such decisions as these and the logic behind them,
while challenging the longstanding consensus among historians that the
Russian project of manumitting slaves from the steppe was motivated by
ideals of abolitionism.
By the mid-eighteenth century, a steady stream of escaped slaves and
captives from among the Kazakhs were arriving at Russian garrisons along
the steppe frontier. What these escapees expected from the Russians is not
clear; perhaps some expected safe passage to settle within the Russian
Empire, or help in returning to their original homes elsewhere in the
region. Others may have had no preconceptions at all, simply preferring
to take their chances with the Russians rather than remain in captivity.
Having taken the risk of fleeing into the steppe, either alone or in small
groups, many of these escapees would have been profoundly disappointed

144
 From Slaves to Serfs 145

by their fate upon arriving at the Russian border: Throughout the eight-
eenth century, and even well into the nineteenth, the Russian modus
operandi was generally to return escaped slaves to their Kazakh owners.
This was inspired in part by a desire to maintain good relations with local
Kazakh rulers and elites, but it may also have been the legacy of a sinister
Russian policy in the steppe: During the Bashkir Uprising of 1704–1711,
the Tsarist authorities had evidently encouraged their Kazakh allies to
plunder and enslave Bashkirs.1 Having provided for the perpetuation
and even the increase of slavery among the Kazakhs, the Russians were
certainly not inclined to give quarter to those lucky enough to escape it,
while offending their allies in the process. There was one exception,
however, and the exception was made law by a decree issued from the
Bureau of Foreign Affairs in February of 1752: escaped slaves who agreed
to be baptized would thereafter be officially welcomed into the Russian
Empire.2 Three decades later, in October of 1781, an imperial edict
established that all captives fleeing Kazakh lands would be free to settle
in Russia or to be returned to their homelands if they so desired.3
By the mid-nineteenth century, the Orenburg Border Commission had
decreed the end of slavery in the steppe, and even began employing Kazakh
“native informants” to help free the region’s slave population (a subject to
which we will return in the following chapter). As border authorities
would discover throughout this period, however, it was not always an
easy matter to negotiate the terrain of manumission. Internal documents
produced by the Orenburg Border Commission and affiliated offices pro-
vide remarkably intimate reportage in this respect. We find in these docu-
ments that Russian notions of slavery and abolition did not always map
cleanly onto the Kazakh context, and that official manumission policies
were not always well-understood even by the border officials tasked with
enforcing them. Abolishing Kazakh slavery would entail defining Kazakh
slavery, first of all – and then there was still the matter of what should be
done with all the newly manumitted men, women and children who were
now at the tsar’s disposal.
Among those escaping slavery in the steppe, it seems that many – and
especially Kalmyks – chose to convert to Christianity at the prompting of

1
Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka, 283.
2
Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka, 283. Evidence indicates
that some – especially those professing the hardship of their prior circumstances – were
allowed to remain in Russia even without being baptized, though these allowances were not
evidently supported by official policy. TsGAKaz 4.1.198 ff. 36a–b.
3
TsGAKaz 4.1.198 f. 38a.
146 From Slaves to Serfs

border officials in order to be resettled within the Russian Empire. These

new Russians were typically resettled in communities of other refugees and
immigrants, such as those that could be found in the towns of Stavropol,
Sarapul, and Karakul.4 Those choosing to be baptized sometimes reported
suffering bad treatment under their Kazakh owners, as did the escaped
slave whose name was recorded as Muhammad Akhmetov. A Persian from
˙
Sabzavar, Akhmetov had been captured by Turkmens while on a trading
mission to Astarabad. He had been sold to a Kazakh for 100 sheep, and he
remained with this owner for the next five years. Akhmetov reports that his
owner inflicted various tortures upon him and kept him in a state of
starvation. For this reason, he fled to the Lower Ural’sk line “with the
intention of becoming a Christian” and making his home in the town of
Ural’sk.5
Others who chose to be baptized claimed to be attempting to join
relatives who had immigrated previously. One such refugee reported hav-
ing been captured some twenty years prior along with his mother and
brother, both of whom had fled to Russia almost immediately. He had been
too young to join them in their flight, and so he remained in service to
a Kazakh woman for another two decades before making his escape.
Seizing an opportune moment, he stole a horse and fled to where he
hoped to make contact with his brother, who, he had learned, was serving
in an Ural Cossack army regiment based in Kazan. He was unable to find
his brother, but wished to be resettled in Stavropol nevertheless.6
In another document, we find the report that seven Iranian slaves who
had fled the steppe all shared an inclination to be baptized and repatriated as
citizens of the Russian Empire. These Iranians hailed from various cities –
Astarabad, Mashhad, and Sabzavar among them – and were evidently
unrelated. But all claimed to have relatives in Ural’sk, a coincidence we
may justifiably regard as suspect. In any event, they were duly baptized, and
their names were “Christianized”: Muhammad Khan became Konstantin;
˙
Muhammad ʿAlı̄ became Aleksei. They were then permitted to reside in
˙
Ural’sk “or some other place” under Russian dominion.7
Escaped slaves typically arrived at the border with no resources of their
own, and with no safe, practical means of getting where they were going,
whether it be to somewhere within the Russian Empire or to their original
homeland of Iran or Afghanistan. Border authorities often absorbed the
expenses – or, in later periods, charged Kazakh elites who had failed to

4
TsGAKaz 4.1.198 ff. 54a–b; 137a; 173a–b. 5 TsGAKaz 4.1.198 f. 19a.
6
TsGAKaz 4.1.198 f. 104a. 7 TsGAKaz 4.1.198 f. 15a.
 From Slaves to Serfs 147

ensure that the slaves were turned over – and also made arrangements for
necessary lodging and transport.8 The authorities could not allow just
anyone to take advantage of these accommodations, however, and
Russian officials were well aware of the possibility that an individual
arriving fresh from the wilds, with an exotic tale to tell, might not be
who he or she claimed to be. In cases of doubt, examinations were
administered. One escapee, who arrived at the border in the Spring of
1862 and whose name was recorded as Nādir Sagarkulin, claimed to be an
Iranian from Tehran, where he had lived with his mother, three brothers,
his wife, and two children (all of whose names were recorded by border
officials). Two years previously, he had been appointed to a military
detachment tasked with raiding a Turkmen encampment, but he was
taken captive in the campaign and sold to a Bukharan, with whom he
lived for no more than half a year. This Bukharan sold him to a Kokandian
merchant, and he went to live with that owner in Tashkent and in
Turkestan. This Kokandian treated him cruelly, and he fled to Fort
Perovskii in order to ask the Russian officials to send him back home, via
Astrakhan, as he had no other means of making the journey. Before agree-
ing to this, however, the border authorities had him examined by a titular
medical counselor of the Syr Darya Line, who reported that Sagarkulin
appeared to be the age he claimed to be. He also offered his opinion on the
matter that probably inspired this and other examinations: that there
“were no signs of corporal punishment on his body by which one could
suspect him of being a criminal fleeing Russia[n authorities].”9 Another
test was also deemed necessary: Sagarkulin was brought before some
Bukharan merchants, who traded at the Fort Perovskii bazaar, in order to
confirm that he was indeed an Iranian.10 Having confirmed to the modest
extent possible that Nādir Sagarkulin was more likely an Iranian escaped
slave than a disguised fugitive from Russia, the border officials consented
to transport him back home to Iran via Astrakhan.
The next step for Sagarkulin, and for any other refugee who was to be
conveyed back home, was the arrangement of transport. For this, border
officials relied on the region’s caravans, and in some cases the safety of
the passengers was ensured by the appointment of a police escort or

8
TsGAKaz 4.1.3641 ff. 137a–b; TsGAKaz 4.1.3573 ff. 41a–b; TsGAKaz 4.1.195 f. 10b;
TsGAKaz 4.1.3573 f. 68a.
9
TsGAKaz 4.1.195 ff. 10a–11a.
10
TsGAKaz 4.1.195 f. 11a. These same merchants also verified the origins of at least one
other escaped Persian slave, whose name is recorded as Fayzullah Muhammad Kasymov
(TsGAKaz 4.1.195 f. 5a).
148 From Slaves to Serfs

others who might watch over them.11 Iranian slaves headed back to Iran
would sometimes be added to the ranks of caravans bound for Bukhara,
and the caravan drivers themselves were made to sign a document con-
firming that they would guarantee the delivery of these temporary wards
of the state.
One Bukharan caravan driver – Rahı̄mbay Atambaev – was employed
˙
repeatedly for these purposes.12 In the course of one of his missions,
Atambaev became involved in a curious episode that reveals the lengths
to which the Commission would go in order to ensure the safe repatria-
tion of former slaves. Having been tasked with delivering an Uzbek
refugee whose name is recorded as Muslimberdi, Rahı̄mbay informed
˙
this Uzbek that his caravan, which was set to depart from the Orenburg
trading grounds, would be traveling through the territory of the Kazakh
clan among which Muslimberdi had been kept as a slave. Because his
former master still lived among them, it was resolved that Muslimberdi
should instead leave on another caravan that would bypass this danger-
ous territory. The Uzbek could rest assured that a suitable caravan would
be available to him, since another Bukharan caravan was being driven by
Kazakhs from a clan hostile to that of his former owner. They would most
certainly avoid their rival clan’s territory. The officials of the Orenburg
Customs House agreed to put Muslimberdi on this caravan.
The caravan was not imminently scheduled to depart, however, and
Muslimberdi had no choice but to wait in Orenburg. The problem was that
Muslimberdi had no resources of any kind, and no way to care for himself
in the meantime. A Customs House official named Burtsov agreed to put
him to work, offering him clothing, lodgings, and the promise that he
would send him off with some money when the time came. Muslimberdi
ended up remaining in Burtsov’s employ for the next six months, evidently
opting to remain in Orenburg rather than to return home on a caravan.
Nevertheless, he turned up at the border authorities’ offices again some-
time later, claiming that Burtsov had refused to give him the clothing and
money that he had been promised. Burtsov responded with the claim that
he was in no way obligated to pay him, and, moreover, that Muslimberdi
had since moved on to work for an Afghan prince living in Orenburg who
paid him only in food, but who ultimately turned him away. It was only,
Burtsov alleged, because of his present unemployment that Muslimberdi
had come to the border authorities seeking money in “reparation.” At this

11
TsGAKaz 4.1.206 f. 34a; 4.1.198 f. 15a.
12
TsGAKaz 4.1.3573 f. 67a; TsGAKaz 4.1.1490 ff. 17a–18a.
 From Slaves to Serfs 149

point it appears to be Burtsov who requested that Muslimberdi be put on

a Bukharan caravan and sent back home.13
Notwithstanding complications such as these, repatriating former
slaves to their original homelands appears usually to have been
a relatively straightforward matter. Those who chose to be baptized
and resettled in the Russian Empire, however, occasioned a more intri-
cate protocol. Details on the fate that awaited these refugees are pre-
served in a work by Nikolai Blinov,14 and what Blinov reveals is quite
astonishing. Having escaped slavery in the steppe and having placed
themselves at the mercy of the Russian Empire, these former slaves
were converted not only into Christians, but also into serfs. For those
historians who consider serfdom a type of slavery, it could be argued that
these refugees were thus promptly re-enslaved by the very state that
purported to free them.
Blinov’s discussion concerns just one of the many towns in which
escapees were often resettled: Sarapul, a settlement on the right bank of
the Kama River. He writes that Sarapul served as a point of resettlement
for Persians, Khivans, Afghans, Bukharans, and “white Arabs” (belye
arapy), and that these immigrants had, by the late nineteenth century,
integrated with the population such that one often met with burghers
and peasants “with hair and beards as black as pitch” – hallmarks of
“Asian” descent – among the light-haired Russian population. These,
Blinov tells us, are escaped slaves who had converted to Christianity and
had found safe passage into the Russian Empire thanks to the government’s
concern for the fate of foreign migrants.15
Blinov dates the Russian Empire’s decision to receive converted refu-
gees to the early 1760s, the dawn of Catherine the Great’s reign, citing
a decree that specified that “people of the various Asian nations” who were
received at the border should be asked whether they wished to settle in the
Empire and, if so, where they wished to settle. Some would then enter into
the employ of private individuals (as Muslimberdi had done); the rest were
conveyed to Moscow, where they would receive new clothing and provi-
sions while awaiting resettlement. The ultimate goal was for these indivi-
duals to adopt “Russian customs” and to acquire basic proficiency in the
Russian language so that they would be well-adapted to their new home-
land. The first group of such refugees to arrive in Moscow consisted of
seventy-two men and women, none of whom spoke enough Russian to

13
TsGAKaz 4.1.1490 ff. 17a–18a.
14 15
Blinov, Sarapul’: istoricheskii ocherk’ (Sarapul, 1887). Blinov, Sarapul’, 13.
150 From Slaves to Serfs

express precisely where in the Russian Empire they wished to settle. It was
thus decided that they should be sent eastward again, to the vicinity of
Kazan and Orenburg. Here they would find settlements, surrounded by
suitable farmland, in which they could meet “the local peasants who,
through proximity to the Bashkirs and Tatars, might know their language.”
The Kalmyk refugees would be separated from the Persians. Accordingly,
fifty-four Kalmyks were sent to Sarapul, and eighteen Persians were sent to
Karakul.16
The refugees were sent off in a convoy of twenty carts. They were
overseen by two Russian officers, one of whom served as a translator,
along with three soldiers tasked with ensuring that no harm came to
them. They were ordered to travel as quickly as possible, no less than
fifty versts per day, and a stipend of five kopeks per day was allotted to each
refugee for food. If any among them became ill and died during the
journey, they were to be buried in the nearest churchyard. Upon their
arrival at Sarapul and Karakul, they would be received by local officials
who were appointed as their stewards. These stewards were tasked with
dividing them into groups of two, three, or four individuals – or however
the steward saw fit – and arranging for them to receive training in the
Russian language. These small groups would be given over to reliable local
peasants who would train them in all things customary for a Russian
peasant, including the mechanics of tilling and plowing, as well as
Christian observances. The refugees were to attend church on Sundays
and on holidays, though no abuse or injury should be inflicted upon them if
they neglected to do so. During their first year of residence, each individual
would continue to receive an allowance of five kopeks per day.17
After one year, those refugees who had proven themselves capable
would receive an allotment of land, along with two horses, one cow,
three sheep, three shirts, a coat, a hat, and a pair of gloves. The amount
of land they were to receive was commensurate with that worked by other
peasants in the area, adjusted on a case-by-case basis according to the needs
of each family. They would also be given three rubles each from the
treasury for the purchase of any necessary instruments, along with
a monthly allotment of rye and bread until such time when they were
able to produce their own bread. In all their possessions they were to be
made satisfied vis-à-vis the other peasants in the area. If any of these new
settlers should wish to marry a local, or vice versa, the marriage was not to
be prevented. If anyone, moreover, was to impress any kind of unfreedom

16 17
Blinov, Sarapul’, 14–15. Blinov, Sarapul’, 15.
 From Slaves to Serfs 151

upon these new residents, the “firm gaze” of local officials would fall upon
the offender. Provided with these rights and amenities, the refugees would
be able to enter the “condition of peasanthood” (krest’ianskoe sostoianie).
After six years’ residence, these new inhabitants would begin to pay the
notorious dues (obrok) demanded of all Russian peasants.18
According to Blinov, the system outlined here had not been developed
specifically to accommodate escaped slaves from the steppe. In fact, the
protocol for converting these new citizens into peasants had been
adapted from an identical system that had been developed for Polish
immigrants and outlined in a royal manifest from May of 1763.
Evidently, the system was deemed effective for steppe refugees as well,
as their numbers steadily increased in the decades to come: by 1777, 184
men and 124 women, 308 people in all, had been resettled in Sarapul
alone.
Their religious conversion, meanwhile, may or may not have been
considered effective, judging by the experiences of a Russian official who
passed through Sarapul in 1778. This official found himself petitioned to
re-baptize the local settlers. While many of these ones had presumably
been baptized already prior to their resettlement, they had “forgotten the
Christian faith” (zabyli khrest’ianskuiu veru) and had been eating carrion
and other impermissible foods. The consistory obligingly cleansed one
such offender through prayer and baptized another, teaching the latter
how to pray.19
While Blinov characterizes the conversion of former slaves into pea-
sants as an example of the Russian state’s “concern” for migrants, it is clear
that there were less charitable motives involved. Above all, these new
peasants were being enlisted in the ongoing effort to colonize the
Russian frontier, as the Russian Empire was engaged in bringing the
borderlands under cultivation and populating them with Russian subjects.
The process of resettling converted Central Asians had, moreover, been
going on for decades. In 1738, for example, the head of Russia’s Kalmyk
Commission had resettled over 2,000 “Kalmyk new Christians” to the fort
of Stavropol, north of Samara. Here, as in Sarapul, the settlers were trained
in “the plow and the Russian way” and encouraged to engage with the local
churches.20 Between 1719 and 1744, the peasant population nearly tripled

18
Blinov, Sarapul’, 16–17. 19 Blinov, Sarapul’, 17.
20
Willard Sunderland, Taming the Wild Field: Colonization and Empire on the Russian
Steppe (Ithaca, NY: Cornell University Press, 2004), 49.
152 From Slaves to Serfs

in Saratov province, swelled by the influx of migrants resettled there by the

state.21
The resettlement of escaped slaves was a novel development in this
effort, directly linking colonization with manumission. It would not be
until well into the nineteenth century, however, that Russian-sponsored
manumission would be strongly linked to antislavery and abolitionism. For
proof of this, we need only consider the fact that, until 1825, Russian
citizens and officials were permitted – and even encouraged – to purchase
Kazakh children themselves, just as long as they agreed to free them when
they reached twenty-five years of age.22
In the early 1850s, in concert with the rising tide of antislavery senti-
ment throughout the Empire, orders were given that all steppe peoples
under Russian rule must surrender their slaves to the nearest officials to be
manumitted. These laws would greatly increase the volume of slaves turn-
ing up at Russian border offices. As we shall see, however, these notionally
abolitionist orders did not often result in functionally abolitionist policies.
As the border officials would soon discover, it was not always clear what
should be done with these newly surrendered slaves – or even if they
should be defined as slaves.
In the 1850s and 1860s, slaves surrendered by their owners to Russian
officials would have had three options: returning to their original home-
land; settling in the Russian Empire; or, if the slave was over twenty-two
years of age, he or she could choose to return to life among the Kazakhs.
In all cases, slaves were transferred only after being formally manumitted
by border officials. Since many of the slaves in question had been taken
captive or purchased at a young age, their owner’s home was often the only
home they had ever known. This helps to explain the high proportion of
slaves who, once manumitted, simply chose to return to the steppe. In such
cases, border officials usually interviewed both the owners who had sur-
rendered their slaves and the slaves themselves, presumably to observe the
level of correspondence between their respective tales. In these interviews,
we often find claims that the slave did not really live as a slave (ne v vide
raba), even though he or she had been purchased with money or with
livestock.
Such was the case with one Kazakh slave-owner of the Altyn clan, whose
name is recorded as Batygul, and who admitted that she had purchased two
“Asian” slaves, whose names are recorded as Gamadan (Hamadan) and

21
Blinov, Sarapul’, 19.
22
Zimanov, Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka, 283.
 From Slaves to Serfs 153

Ak-Dzhulay. She owned these children for fourteen and sixteen years,
respectively. She had purchased Gamadan for thirty Bukharan coins, and
Ak-Dzhulay for twenty-five. Batygul avowed that they lived with her under
humane conditions (po chelovechestvu), without any shame and in no kind
of slavery. As evidence of this, she adds that she even arranged their
marriages. She placed them at the disposal of the authorities, as was
demanded of her, claiming that neither of them wished to be returned to
their original homeland, and implying that they would both prefer to
remain with her.23
An interview with Gamadan appears in this document immediately
following his owner’s statement. He declares his age as twenty-six, and
mentions that he does not know his patronymic. He is illiterate, a Muslim,
originally from among the Hazara. He had been taken captive at a young
age by unknown assailants (khishniki) and sold to Bukhara. Some fourteen
years previously, he was purchased by Batygul for thirty Bukharan coins,
and he had since lived with her as a son (v vide syna). Gamadan declares
that he does not wish to return to his original homeland among the Hazara,
since it was “an unknown place”; he prefers to remain with Batygul.24
Ak-Dzhulay’s statement follows this. He describes himself as twenty-
one years old, illiterate, a Muslim, and likewise originally from among the
Hazara, though he adds that he does not know which clan was his. He was
evidently very young when he was captured and sold to Bukhara. Sixteen
years previously he was purchased in Bukhara for twenty-five coins by
Batygul, who took him as her adopted son (vzial v usynovlenii). He adds
that his original homeland is unknown to him, as is his original family,
since he had been alienated from them for so long, and for this reason he
did not want to return to the Hazara. He expresses the wish to remain with
Batygul, under whom he claims to suffer no manner of oppression, and to
fulfill all his obligations among the Kazakhs.25
We find a similar situation in the case of an Arab slave whose name was
recorded as Derbis [Darvish] Muhammad. This slave had also been living
˙
among the Altyn clan, and his owner reports having purchased him in
Bukhara for twenty-five Bukharan coins some fourteen years previously,
when Derbis Muhammad was just eight years old. The owner, whose name
˙
was recorded as Taylanbay-Dzhan Aristanov, claims to have adopted the
boy as his son, and mentions that he had even arranged his marriage. He
declares that Derbis Muhammad has no desire to return to his original
˙
23 24
TsGAKaz 4.1.3641 f. 38a. TsGAKaz 4.1.3641 ff. 38a–b.
25
TsGAKaz 4.1.3641 f. 38b.
154 From Slaves to Serfs

homeland, but that he agreed to turn the young man over in accordance
with the will of his superiors, which had only recently been announced to
him. Placing his slave at the disposal of the authorities, Aristanov adds that
if Derbis Muhammad would be permitted by the border officials to con-
˙
tinue living with him, he would allow the young man to do so. A statement
by Derbis Muhammad has also been preserved, and in it the young man
˙
confirms every detail provided by his owner: that he is twenty-two years
old, an Arab, sold in Bukhara to Aristanov at the age of eight for twenty-
five Bukharan coins; that he lives with Aristanov as his adopted son; and
that he has no wish to return to his original home, of which, he says, he
knows nothing. He expresses his “unashamed desire to live always among
the Kazakhs” and, moreover, with this same Aristanov.26
It appears that a great many adult slaves who had been surrendered by
their owners chose to remain in the steppe. Some may have experienced
moments of indecision, weighing the chance at a new but uncertain life
against their familiar captivity. One slave of Iranian origin, whose name
was recorded as Kulgilbay, originally agreed to convert to Christianity, but
then changed his mind and renounced that intention, choosing instead to
continue living with his former owner among the Kazakhs.27 Most others
chose similarly. In one group of eight slaves – seven men and one woman –
that arrived at the border, only the woman, whose name is recorded as
Summanaz, expressed a wish to be returned to her original homeland.
Summanaz was duly taken away from her Kazakh owner and placed in the
care of the border office. Meanwhile, the others became the subject of
a fatigued request by the border official tasked with processing them. This
official penned an inquiry to his superiors, asking if, in order to save time
and avoid “burdensome correspondence,” he might simply provide those
“Asians” who wished to remain among the Kazakhs with their freedom and
forego the process of submitting any more “complicated reports” about
them – promising, however, to submit full reports for those who wished to
be sent back to their original homes.28
Other border officials seemed to take the process more seriously.
In October of 1851, another group of Iranian slaves was divided according
to their wishes: one wished to return home to Iran, and the rest to return
among the Kazakhs. The official receiving this group wrote to his superior
asking how he should proceed, as the process for repatriating surrendered
slaves was, at that time, still in its infancy. The superior officer

26 27
TsGAKaz 4.1.3641 f. 37a. TsGAKaz 4.1.198 f. 49a.
28
TsGAKaz 4.1.198 ff.56a–b.
 From Slaves to Serfs 155

unhesitatingly sent the slave who wished to return home back to Iran via
Astrakhan, arranging for border offices to absorb any costs of transport.
Regarding the rest, however, the official implored his comrades to certify
that returning among the Kazakhs “was indeed their true and proper
desire,” and he took the opportunity to “remind the [Border]
Commission of its responsibility” to ensure that former slaves returning
to their former owners be treated equitably from then on in terms of their
“personal rights” (lichnaia prava) and responsibilities among the Kazakhs.
He asked to be appraised of the results once these Persians had returned to
the steppe.29
In the case of adult former slaves who chose to return to their former
owners, it must be asked: what had Russian-sponsored manumission
really accomplished for them? First of all, we have no way of knowing
whether manumission at the hands of the Russians was considered to be
a valid form of manumission among Kazakhs. We also have no way of
knowing whether this type of manumission really prevented the “freed-
men” from being sold off or traded in the future. Nor do we know if the
process had any real impact on the social standing, treatment, and obli-
gations of these former slaves once they returned to the steppe. Finally,
and perhaps most significantly, we do not know if most Russian officials
considered these ambiguities at all significant to what they were ulti-
mately trying to accomplish – a point to which we will soon return.
It may be tempting to presume that the widespread desire among slaves
to return to their owners indicates that they were generally well-treated,
but we have reason to hesitate here: Russian officials could offer these
slaves transportation elsewhere, but they could not offer any alternative
livelihoods beyond the unfamiliar – and quite possibly unappealing –
existence of the newly minted Russian peasant. Since many of these
manumitted slaves had been taken from their relatives in childhood, the
communities in which they had been enslaved formed the closest thing
they had to a family network. Moreover, those enslaved men who had
married free women among the Kazakhs may not have been able or willing
to ask their wives to abandon their homes and join them in their bid for
a more total freedom. The prospect of freedom and a new beginning may,
in other words, have struck many of these individuals as a profoundly
lonely option. In this light, their “manumission” and return to the steppe
may serve only to emphasize the fundamental tragedy of their
circumstances.

29
TsGAKaz 4.1.3573 f. 68a.
156 From Slaves to Serfs

Such were the options presented to adult slaves, but border policy
recognized minors – those under twenty-two – as a different case that
called for different protocols. Many children were brought before border
officials, and most often their owners, claimed to have “adopted them,”
noting also that these child slaves knew nothing of their original home-
lands. Three examples will suffice to give a sense of these interviews, which
resemble the depositions concerning adult slaves in all but the age of the
slaves in question.
In one document, a woman whose name was recorded as Sar Kulova,
the wife of one Churtegen Khalimakov, reports that her husband had
purchased a seven-year-old boy named Azat in Bukhara the previous
winter, paying twenty-five Bukharan coins for him. Kulova avows that
the boy is their adopted son, and that Azat has no desire to return to his
homeland, “all the more so since [that place] is completely alien to him,”
given his young age. She presents Azat to the border authorities with the
request that they not simply return him to his homeland, where no one
knows anything of him, and where there is no one to be a mother to him to
the extent that she herself is.30
In the same month, we find a Kazakh of the Altyn clan whose name is
recorded as Mamirbay surrendering to the Commission two boys, Pivan
and Khudaybergan, whom he had purchased in Bukhara earlier that year,
the former for twenty-five Bukharan coins and the latter for nineteen.
Hearing only now, he says, that the authorities were no longer permitting
them to be kept, he surrendered them to the Kazakh Sultan who was
working as an agent of the border authorities. He notes, however, that
they are quite young – Pivan being no more than 14, Khudaybergan no
more than 15 – and have expressed a “persistent desire” to live with him.
A statement from Pivan follows in the document, and the boy confirms that
he is fourteen years old. He says that he is a Muslim, from among the
Arabs, and that he was taken to Bukhara two years previously by
Turkmens. In the present year he had been purchased by Mamirbay.
Following this we find a more concise interview with Khudaybergan,
who reveals only that he is fifteen years old and was purchased by
Mamirbay for nineteen Bukharan coins in Bukhara, where he was born,
though he does not remember his mother and father. Both boys express
that they do not wish to be returned to their original homeland, “as we do
not know our parents or any relations there.”31 In Pivan’s case, we may
wonder if this last claim is true: in his own telling he was twelve years old

30 31
TsGAKaz 4.1.3641 f. 36b. TsGAKaz 4.1.3641 f. 36a.
 From Slaves to Serfs 157

when he was taken to Bukhara by Turkmens. It is possible that these

Turkmens were his original captors and, if so, he would likely have
known his biological family. We do not know these details, of course,
but neither can we rule out the possibility that some child slaves
expressed a desire to remain with their owners only under duress. Their
owners may even have been in the room while these children were being
interviewed.
In the case of an enslaved boy named Amanzhul, it appears more likely
that he was truly unfamiliar with his biological family, though at least one
detail in his story remains suspicious. His owner, whose name is recorded
as Sansizbay, reports in his interview that he had bought Amanzhul four
years previously, in Bukhara, for just fourteen Bukharan coins. The boy
was at that time “no more than six years old,” and Sansizbay claims to keep
him as an adopted son, and not as a slave (ne v vide raba). He says that
Amanzhul has no knowledge of his original homeland and prefers to stay
with him. A brief statement from Amanzhul follows, in which the boy
declares that he is ten years old, a Muslim, and illiterate; and that he is of
“Central Asian” origins, though he cannot specify more than this (pri-
khozhu iz sredneaziatskikh vladenii, no iz kakogo imenno ne znaiu). He
does not know how he ended up in Bukhara or whom he lived with there,
though he knows that he was eventually purchased by Sansizbay for four-
teen Bukharan coins. He adds that he does not wish to be sent back to his
original homeland, as he knows nothing of that place, nor is he able learn
anything more about it. One detail may call his narrative into question,
however. The Kazakh name “Amanzhul” was apparently given to him by
his present owner, as the boy claims that he does not even know what name
he had gone by previously.32 It was common for a slave to receive a new
name from his or her owner, but the idea that a ten-year-old boy would not
know what name he had gone by at four years old ought to make us wonder
to what extent the boy chose to exaggerate the obscurity of his origins –
and why he might have been inclined to do so.
What was to be done with children such as these, who had been
“adopted” by the men who had purchased them? Were they really slaves?
It was clear to the Border Commission, first of all, that a child’s profession
of loyalty to his or her owner should be regarded differently from that of
an adult slave. For one thing, their prospects in the absence of any sort of
guardianship were significantly worse than those of adult slaves – a fact
that would not have been lost on the children themselves. Children,

32
TsGAKaz 4.1.3641 ff. 37b–38a.
158 From Slaves to Serfs

moreover, would be more vulnerable to the influence of threats, lies, or

other forms of manipulation on the part of their owners. This is not to
suggest that their owners were being dishonest in characterizing these boys
and girls as their own adopted children. In fact, since they faced no
punishment for voluntarily turning these children over to the Russians,
and since – as we shall soon see – border officials’ decision on what to do
with these children did not rest entirely on whether or not their owners
considered them to be slaves, it is most likely that we are witnessing
a cultural disconnect in which Russian definitions of slavery were lost in
translation for those Kazakhs drawn into the new system of enforced
manumission.
In fact, Russian manumission policies – as well as official definitions of
who should be considered a slave – were sometimes unclear even to border
officials themselves. They were likewise unclear to some native informants
tasked with rounding up and exposing the slave-owners. One internal
document concerning a local Kazakh biy preserves an unusual ethno-
graphic preamble, in which an official evidently felt compelled to explain
to his superiors the ambiguity of notions of “slavery” in the Kazakh
context. In the old days, the official explains, some Kazakhs would freely
send their children into others’ care. Other Kazakhs – in times of famine,
for example – would sell their children. This was, the official observes,
forbidden by Islamic law, but in such instances they were not being pur-
chased as slaves, but rather in place of biological children, since their
purchasers typically had no children of their own. Still others bought
maidens (devits) to marry, and this transaction was carried out not against
anyone’s will, but by agreement from both sides (ne chrez [sic] nevol’no
a po soglasno oboikh storon). Both Kazakhs and Karakalpaks engaged in
this custom, according to the officer, who then offers his inquiry: Would
those Kazakhs and Karakalpaks still be subject to the new manumission
laws if, having been purchased before the passing of these laws, they
currently lived as wives and children according to common consent?33
The officer’s inquiry was not an abstract consideration; he needed an
answer immediately, as he had just received a report from a Kazakh biy,
operating as a native informant, concerning a young man who was in
precisely the ambiguous position outlined here: perhaps a slave, perhaps
not. Roughly fourteen years previously, in 1847, “at the time of the
hunger” (vo vremia goloda), a Kazakh whose name is recorded as
Batyrbay Baydelov purchased the young man from a Karakalpak for one

33
TsGAKaz 383.1.89 f. 14a.
 From Slaves to Serfs 159

camel and one horse. The young man, named Dawlat, was at that time just
two years old. He was the Karakalpak’s own son. Dawlat confirmed under
the biy’s questioning that he had been sold in early childhood, and he
declared that he had no desire to return among the Karakalpaks. He was,
moreover, satisfied with his present owner because, he told the biy,
“Batyrbay Baydelov does not have children of his own and considers me
his son, and he does not use me for labor” (ne upotrebliaet na raboty).34
As it turns out, the border offices already had a policy concerning cases
such as these – though it is revealing that the policy was apparently not
known or understood by all of the officers tasked with implementing it.
The policy had been established by a pronouncement of the Orenburg
Governor General on December 1, 1860, which reinforced, first of all, the
fact that those immigrants who had freed themselves from slavery under
Kazakhs subject to Orenburg’s authority could choose to swear allegiance
to Russia, or to the Kazakhs, or to be returned to their original homeland
elsewhere.35 As for slaves under twenty-two years old, however, their fate
was not entirely their own to decide. Instead, border officials would have
them turned over to “trustworthy hands” (blagonadezhnye ruki) until they
reached the age of full maturity (sovershennyi vozrast). At that point they
would achieve “full freedom” (pol’naia svoboda), and they would be able
to make their own decision about where to go next.36 These children
would, in other words, be placed in foster care.
Regardless of age, an individual turned over to border officials was
granted his or her freedom as a standard matter of bureaucratic process.
When adult slaves chose to return among the Kazakhs, the documentation
of their manumission would often include the detail that their former
owners were henceforth forbidden from enslaving them. Children who
had allegedly been “adopted” by their owners were likewise officially
manumitted by the Russians. Women who had been purchased as brides
were likely also manumitted, though I have yet to see a manumission
document pertaining to such a woman.
Combining the above information on official policies with the nature of
the information collected from slaves at the border allows us to ascertain
something of the Orenburg Border Commission’s working definition of
slavery. First, it is useful to observe what was absent from slaves’ inter-
views. It is clear, for example, that border officials appear remarkably
unconcerned with labor as a condition of slavery; it seems that slaves

34 35
TsGAKaz 383.1.89 ff. 14a–b. TsGAKaz 383.1.184 ff. 11b–12a.
36
TsGAKaz 383.1.184 f. 12a.
160 From Slaves to Serfs

were only rarely asked about the nature of the labor that had been
expected of them, or about the nature of their upkeep and compensation,
if any. Likewise, their owners, having surrendered them, were rarely asked
how they had used them and what kind of work they had assigned to them.
If such questions came up over the course of these interviews, neither the
questions nor the answers were typically documented.
By contrast, all surrendered slaves were asked when and where they had
been purchased; from whom they had been purchased; who had purchased
them; and at what price they had been sold. The fact of having been bought
and sold emerges in these documents as the central defining feature of their
slavery. Their owners were made to supply precisely the same information
concerning purchase and sale: when, where, from whom, and at what
price. This may be explained in part as both a means of checking
a slave’s autobiography against the testimony of his or her owner and,
more simply, as a way to add a key validating detail to their categorization
as slaves. Above all, the contours of these slaves’ biographies are defined by
periods of ownership.
A slave, then, was an individual who had been bought and sold.
Purchase and sale are the two ever-present factors in these interviews,
which served also as documentation of manumission. Purchasing an indi-
vidual marked one, in the eyes of the border authorities, as being subject to
punishment unless the purchased individual was surrendered to the state.
This fact remained constant regardless of the circumstances of sale. It also
remained constant regardless of what relationship the owner claimed to
have with their purchase. Owners who “adopted” their purchases,
arranged their marriages, and so on, were still made to surrender them to
border offices.
But where did border officials find “trustworthy hands” to receive those
manumitted slaves who were not yet of age? The fate of a child slave
named Ashirbay reveals the surprising details of this process. Ashirbay
was ten years old when he was brought to the border along with three
other child slaves: Nasir, nine years old, Mazhik [?], fifteen, and Azim,
eighteen. They were inspected by the Bukharan merchants at Fort
Perovskii to verify their origins, and none showed any signs of abuse at
the hands of his owner. The official charged with processing them logged
a request that the two youngest boys – Ashirbay and Nasir – be given over
into fosterage until they had reached the age of maturity.37 A third boy,
Mazhik, was eventually processed into fosterage as well.38 Azim appears to

37 38
TsGAKaz 381.1.184 ff. 18a–b. TsGAKaz 381.1.184 f. 29a.
 From Slaves to Serfs 161

have been allowed to remain with his Kazakh owner under the conditions
that the latter not attempt to re-enslave the boy, despite the fact that, at
eighteen, Azim would technically have been eligible for fosterage.39
In the case of Ashirbay, we find that the “trustworthy hands” chosen for
his fosterage were those of his former owner, the Kazakh who had pur-
chased him as a slave. The child had expressed a wish to remain in the
steppe, and while Russian border policy dictated that he be placed some-
where to receive his “education” until he came of age, there was nothing in
the policy that stipulated where, precisely, a child was to be sent or who
qualified as a “trustworthy” foster parent. We may assume that the Russian
state lacked adequate foster homes in the region to receive the influx of
freed slaves. Otherwise, the decision to send Ashirbay back to his owner
seems difficult to comprehend in light of policies that seem designed
specifically to prevent such a circumstance. The official in charge specified
for the record that the former owner to whom Ashirbay would be returned
was not to receive him as a slave. Ashirbay had, at least on paper, been
manumitted. Thanks to the Russian state, his owner had become his foster
father.
Here, we may find some explanation for a seeming lapse of logic in
border officials’ usual interview regimen. Why, after all, had officials both-
ered documenting so often whether or not a slave lived with his or her owner
as an “adopted” child, only to manumit the child anyway? On first glance,
the claim of adoption seems like a desperate and vain effort on the part of
the owners to keep their slaves or avoid punishment, and on the part of the
slaves to placate their owners or to avoid an uncertain fate. Even in those
cases where the bond between owner and slave may truly have been close,
such as that between parent and child, the owner was still made to surrender
any child he or she had purchased. So why would border officials bother
reporting these supposed “familial” bonds? As we see in the case of Ashirbay,
it was likely these professed bonds – along with the medical examination
that uncovered no evidence of his abuse – that allowed his former owner to
foster him after his manumission.
We have now observed three possible outcomes in Russia’s border-
manumission system. Some slaves – a small number, probably – were
given access back to their homelands; a larger number were resettled and
turned into serfs; and perhaps a still larger number were sent back to the
steppe. Should we think of these as “abolitionist” policies? Was Russia an
“abolitionist” empire in the steppe?

39
TsGAKaz 381.1.184 ff. 27a–b.
162 From Slaves to Serfs

In the case of those slaves who made it back home, we can perhaps
answer in the affirmative. In the case of those who were turned into serfs,
the answer depends on whether one considers serfdom to be freedom.
The third case, though – the case of slaves sent back to the steppe – is
especially revealing. If so many slaves who had been surrendered to
Russian border authorities were simply returned to the care of their
former owners, then what was the purpose of manumitting these ones
in the first place? Why did the Russian state bother establishing a network
of native informants, for example, only to remit “freed” slaves back into
the steppe?
Along with providing new agricultural laborers for the settling of the
borderlands, the purpose of this system of manumission was not to ensure
the freedom of the enslaved – and in most cases it did no such thing – but to
impress upon the slave-owner that he was now visible to the Russian state.
It was about jurisdiction. Equally, Russian-enforced manumission was the
tool of an expanding bureaucracy: The information provided in border
authorities’ interviews served to document and make legible individuals –
both slaves and their owners – who may otherwise have been practically
invisible to the Empire; it served, in other words, as a sort of census.
Finally, this process served to make Russian power more visible to all
the people of the steppe. It spread a powerful message: The reach of
Russian colonial power could now extend into their homes and divest
them of their property. It could turn their slaves into Russian serfs and – at
least on Russian paper – it could turn slave-owners into foster fathers.
It could even send slaves back to Iran, “legally” dismantling what many
Kazakhs may have considered to be a relationship between parent and
child.
 6

The Khan as Russian Agent

Native Informants and Abolition

In matters of intelligence-gathering as well as enforcement, the Russian

Empire came to rely extensively on informants drawn from local Kazakh
and Turkmen populations along the steppe and Iranian frontiers. These
informants became instrumental in Russian efforts to militate against
slave-raiding when Russian victims were involved. However, while
Russian authorities tried to suppress the seizure of Russians throughout
the early and mid-nineteenth century, they simultaneously encouraged the
Turkmens’ ongoing slave raids into northern Iran. As often as not, in any
case, whether along the Iranian frontier or in the Kazakh steppe, the
informants did not play the roles intended for them: acutely aware of
their importance to Russian officials, many Kazakh and Turkmen “agents”
took advantage of their positions by manipulating their Russian patrons
for their own benefit or for the benefit of their communities. Far from
being mere colonial puppets, many of these informants came to use their
newfound status not only for personal interests, but also as a form of
resistance against their own colonial patrons. Meanwhile, the impact of
the informant system on captive-taking in the region was, as we shall see,
decidedly mixed.

KAZAKH INFORMANTS ALONG THE STEPPE FRONTIER

From its inception at the end of the eighteenth century, the Orenburg
Border Commission involved itself in the resolution of disputes among
neighboring Kazakhs, providing both an alternative source of legal author-
ity and an alternative framework of legal enforcement for many in the

163
164 The Khan as Russian Agent

steppe. A diverse range of disputes came before the Commission, concern-

ing everything from the theft of livestock to cases of assault. In some cases,
the Commission served as merely one of several Russian-officiated courts
of appeal available to complainants. In one fairly typical case from the turn
of the nineteenth century, for example, a Kazakh named Akutin, who
claimed to have been assaulted, first brought his case before the Ural
military chancery and then before the Border Commission; a record of
his complaint is preserved both in Arabic-script Turkic and in Russian
translation. According to the records of the case, Akutin crossed the Ural
River into the steppe along with his wife in order to trade in fish. He did so
without permission from the chief (nachal’nik) of his settlement (krepost’)
at Topolinsk. A Kazakh demanded payment of debts from him while he
was across the river, and he refused to pay. The Kazakh in question then
pursued him back to Topolinsk, where they quarreled and the Kazakh
made two wounds in Akutin’s head and took two of his horses. The Ural
military chancery evidently sent Akutin onward to Orenburg in order to
seek the return of his confiscated property.1
Given its dual role in enforcement and adjudication, it is natural that
border authorities also involved themselves early on in cases of kidnapping
and enslavement. As early as 1822, a decree was issued prohibiting slave-
holding among the Elder Horde, with a penalty of 150 rubles extracted
from offenders.2 Lacking a comprehensive network for patrol and surveil-
lance across the steppe, the Commission relied from the beginning on the
agency and cooperation of locals and steppe peoples to define and report
crises as well as to bring forward information leading to the capture of
suspected offenders. Modest and ad hoc in its origins, this novel system of
delegating and receiving native informants in the steppe would become
increasingly systematic and complex over the course of the nineteenth
century. Similar strategies would be implemented to the south, along the
Khurasan frontier. By the middle of the century, native informants would
form the backbone of the Russian Empire’s accelerating efforts toward
pacifying the steppe and eliminating the phenomenon of captive-taking –
at least insofar as it impacted Russian citizens.
As early as 1800, Kazakh informants were already submitting formal
reports to the military governor of Orenburg in order to alert him to the
presence of captives among the Kazakhs. The primary concern appears to
have been the release of Russian subjects, but some Kazakh and Kalmyk
captives were also freed by these means. One report from January, 1800,

1 2
TsGAKaz 4.1.195 f. 48a. Semeniuk, “Likvidatsiia rabstva v Kazakhstane,” 236.
 Kazakh Informants along the Steppe Frontier 165

recognizes the “diligence and laudable deeds” of a Kazakh sultan who

delivered to the Commission a number of captives who had previously
been taken by Kazakh “thieves,” including two Saratov “peasants” and
nine officials from the Ural Office. “Along with these,” the document adds
(leaving little ambiguity as to the captives’ relative order of priority) that
twenty-one Kazakhs of both genders were also delivered. One Kazakh,
named Imangul, was held “under guard” at Ural’sk, and the sultan asked
that he be released. In recognition of the sultan’s service, the border
authorities honored the request and set Imangul free.3 It is not clear how
the sultan managed to secure more than two dozen captives for the border
authorities, and we cannot rule out the likelihood that they were taken in
a raid, perhaps accomplished specifically for the purposes of securing
a reward.
Other missions were undertaken with a higher level of cooperation
from the Russian side. On one occasion, a Kazakh khan of the Junior
Horde contacted the Commission in order to alert officials that he knew
the whereabouts of some Kazakhs who had been implicated in the kidnap-
ping of a young Russian boy. The alleged kidnappers, the khan claimed,
had taken refuge in a dense expanse of reeds on the seashore, and he
offered the Russians his services: “I, for my part,” the khan testified,
“will not fail to point out which reeds the nomads are hiding in.”
The Russian officials, with the khan’s assistance, evidently undertook the
delicate operation of seizing the kidnappers and freeing the boy.4
The use of Russian funds rather than troops or favors was often neces-
sary to affect the release of captives. In one case recorded by border
officials in 1799–1800, some Kazakhs took the son of a merchant named
Poliakov, along with fifty rubles in plunder. Somehow, a Kazakh sultan
ended up in possession of the boy (named Vasilii) and pledged to the
Russians that he held him in safekeeping. He promised to turn the boy
over, but, evidently, only in exchange for payment.5 While the documen-
tary record of the exchange does not explicitly accuse the sultan of com-
plicity or participation in the boy’s capture, it is clear that the payment is
tantamount to ransom. In such a case, the informant was clearly
a beneficiary of the crime itself.
Engaging with Kazakh informants to secure captives could sometimes
be a relatively straightforward affair, as in the previous examples, but other
cases suggest a world of possible complications, and point toward the

3 4
TsGAKaz 4.1.195 f. 34a. TsGAKaz 4.1.195 ff. 151a–152a.
5
TsGAKaz 4.1.195 f. 204a.
166 The Khan as Russian Agent

innumerable avenues by which Russian colonial forces could find them-

selves manipulated by their would-be colonial subjects. One of the most
detailed narratives along these lines is provided by Witkiewicz, who was
personally involved in the freeing of a Russian captive during his mission to
the Bukhara in the mid-1830s. After twenty-five horses were stolen from
a caravan in which Witkiewicz was traveling, patrols were sent forth from
the caravan to search alongside the Syr Darya, and this search turned up
a “famous” thief named Kuldzhan Karakchi of the Diurtkara clan. When
the alleged thief was brought back to the caravan, he gave assurances that
he had only made himself vulnerable to capture in order to alert the
Russians that a Cossack captive named Ivan Stepanov was in residence
with his clan. Stepanov had been captured along with his wife while
making hay, and the captors were said to be Dzhegalbay fugitives who
had been living among the Diurtkara. The stolen horses were also among
the Diurtkara, and Kuldzhan Karakchi promised to return all of them
along with Stepanov. The merchants of the caravan wanted to detain
Karakchi and deliver him to Khwarazm for punishment, but Witkiewicz
forbade it, since the suspect was technically a Russian subject.6
Witkiewicz ordered Karakchi’s release and demanded that he bring
forth the captive and the stolen horses. Karakchi returned to his clan
along with a caravan guide, a Kazakh from among the Chumekey, who
would be tasked with transporting the captive and the horses back to the
caravan. Once back at the alleged thief ’s aul, Witkiewicz writes, “a
comedy played out”: the people of the aul tied Karakchi up, acting as if
they suspected him of giving false testimony, and insisted that the stolen
horses were not among them. Those responsible for the theft, they
claimed, could be found at some headland or peninsula on the Aral Sea.
There, at Aran, north of the mouth of the Syr Darya, lived some 500 tents
of alleged “robbers” known as the Karkru-Aranchi, consisting of forty
clans who subsisted through trapping saiga and goats, growing grain, and
thieving.7
Stepanov, however, unlike the stolen horses, was found among the
Diurtkara. A sultan-governor from Orsk sent a Cossack agent to collabo-
rate with Witkiewicz in order to secure his release. The man had refused to
travel to the aul with Kuldzhan Karakchi, fearing the latter would kill him,
but he wished to make an attempt at buying Stepanov’s freedom.
Witkiewicz forbade it, but the agent insisted on being given the necessary

6
Vitkevich, Zapiski o Bukharskom khanstve, 118.
7
Vitkevich, Zapiski o Bukharskom khanstve, 118.
 Kazakh Informants along the Steppe Frontier 167

funds. These funds were promptly divided among the Diurtkara, who
declined to release Stepanov. In the meantime, they had sold his wife –
who had given birth in the steppe – to some Khivans. In all, the initial
efforts by Witkiewicz and his fellow-travelers to free Stepanov while on
their way to Bukhara were a disaster.8
Later, on the way back from Bukhara, he would have another opportu-
nity to rescue Stepanov. When the caravan was passing roughly 100 versts
from where the Diurtkara were residing, Witkiewicz sent them a guide as
an envoy. The guide returned with a Kazakh named Bulush Bay, who was
identified as the son-in-law of the man with whom Stepanov was living,
a Diurtkara Kazakh named Kulbay. Witkiewicz “informed him of his
obligation to render service” to the Russian Empire, promising him also
a reward if he brought forth the captive. Bulush Bay returned among the
Diurtkara, but came back to the caravan empty-handed seven days later,
requesting an official document which pledged that he would not be
prosecuted after bringing the captive. Witkiewicz provided him with the
requested papers along with some gifts and sent him back to the Diurtkara.
Four days later, Bulush Bay returned empty-handed once again – this time,
badly beaten and bloody. He returned the gifts Witkiewicz had given him,
not wanting to receive payment for an unfulfilled mission. He had, it
turned out, quarreled with his father-in-law, who resented the prospect
of giving up the captive with no visible ransom on the table, especially in
light of the ransom that had been brought earlier by the Cossack agent.
In the midst of their heated discussions, a man named Bek Mirza, who had
sold Stepanov’s wife to Khiva, managed to convince many others in the
clan not to give up their captive. Kuldzhan Karakchi argued likewise, even
offering to buy the captive himself, with the aim of re-selling him to some
Khivans.9
On the verge of losing hope, Witkiewicz sent Bulush Bay back to the
clan once again, this time accompanied by two others. These three
returned to the caravan twelve days later, reporting that they had met
with a man named Tlyaulii, of the Diurtkara. Tlyaulii, identified as
a “robber,” had previously fought with Bek Mirza and was presently on
the run from some Khivans. It was, therefore, not difficult to convince him
to serve the Russian Empire. Tlyaulii sent Kuldzhan Karakchi (who had
evidently changed his tune) along with the caravan envoys and some others
to Kulbay to negotiate the release of Stepanov. This group managed to

8
Vitkevich, Zapiski o Bukharskom khanstve, 118.
9
Vitkevich, Zapiski o Bukharskom khanstve, 118–119.
168 The Khan as Russian Agent

intimidate Kulbay into giving up the captive, and Stepanov was promptly
sent to join the caravan.10
Along with the captive came a letter from Tlyaulii. In it, Tlyaulii
evidently blamed the oppressive tactics of the Russian Empire for discord
in the steppe. He requested that Witkiewicz address Russian commanders
in order to mediate and enact a peaceful settlement between the clans of
the Diurtkara, Dzhagalbay, Tamyn, and Tabyn. The ultimate goal would be
to enact a separation between the Russians and these clans, in order to
ensure a relationship “free of harassment” between Russians and nomads.
As a gesture of goodwill, Tlyaulii also offered to facilitate the capture of
Bek Mirza’s band of robbers.11
As for the liberated man, Stepanov, it seems that the hardships of his
captivity and the loss of his wife had taken a tragic toll on him. He was,
Witkiewicz writes, “a sick man, weak and foolish. It was impossible to
learn anything from him.”12
In Witkiewicz’s story, as well as in previous examples, the initiative of
local, non-Russian actors guides all efforts and determines the outcome in
securing the release of captives. The relationship between Russian colonial
forces and steppe natives was hardly one in which the “Great Power” was
able to simply delegate mercenaries to do its bidding. Until the mid-
nineteenth century, at least, captors or slave-owners were free to negotiate
ransom settlements with Russian officials, and agents in the employ of the
Russians could negotiate the terms of their own remuneration. As we see in
the case of Stepanov, captors could boldly decline to free a captive after
receiving ransom, though such a betrayal would have been counterpro-
ductive for those hoping to make a steady business out of Russian ransom
payments. Go-betweens, meanwhile, could decline to deliver the freed-
men, choosing instead to pocket their pay and sell off their new cargo.
These agents could – and did – conspire together with captors to extract
larger sums in ransom;13 perhaps they were sometimes paid off or

10
Vitkevich, Zapiski o Bukharskom khanstve, 119. 11 Ibid.
12
“Kazak Stepanov—chelovek bol’noi, slabyi i glupyi; iz nego nel’zia bylo vyvedat’ nichego.”
Vitkevich, Zapiski o Bukharskom khanstve, 119.
13
Vambery writes of meeting in his travels “one of two sailors from the new station at
Ashourada; the other had died in captivity about a year before. They had fallen into the
hands of the Karatchis some years previously, in one of their night expeditions. Their
government offered to ransom them, but the Turkomans demanded an exorbitant sum (five
hundred ducats for one); and as during the negotiation Tcherkes Bay, the brother of
Kotshak Khan was sent by the Russians to Siberia, where he died, the liberation of the
unfortunate Christians became matter of still greater difficulty; and now the survivor will
soon succumb under the hardships of his captivity, as his comrade has done before him.”
 Kazakh Informants along the Steppe Frontier 169

otherwise encouraged by their contacts to deliver false information to

Russian officials. In short, those residents of the steppe who served as
points of contact between Russian officials and captors or slave-owners
had a practically unlimited ability to guide negotiations by controlling the
nature of the information passed between the two camps. Witkiewicz, for
one, reports only what was related to him by his interlocutors; he had to
choose whether to believe them.
The initiative of these interlocutors – whether they be khans-turned-
informants or caravan guides-turned-emissaries – could also serve ends
beyond mere personal gain. With Russian influence becoming ever more
visible in the steppe, go-betweens could use their privileged positions for
the purposes of resistance. This could be accomplished both by foiling
Russian ambitions on a case-by-case basis or, as in the case of Tlyaulii, by
positioning oneself to submit a request. Though there is no evidence that
Tlyaulii’s request was taken seriously by Russian officials, the nature of his
appeal is striking in itself. It shows a keen awareness that the incursions of
Russian influence had interfered with relations among steppe clans.
Witkiewicz makes no mention of Tlyaulii being offered any form of
compensation other than the chance to submit his appeal.
By the mid-nineteenth century, the ad hoc process by which Russian
officials endeavored to free captives and slaves from the steppe was
replaced with a more rigorous, deliberate, and wide-ranging system.
By 1851–1852, the Orenburg Border Commission had begun circulating,
among Kazakhs in the western reaches of the steppe, decrees in Arabic-
script Turkic ordering the release of slaves and captives. Some of the
earliest decrees that I have seen specify that Iranian Shiʿites – called simply
“Qizilbash” in these documents, which doubtless indicated Iranian slaves –
were living in the area (shu örinda qizilbāsh vilāyatlik yasarlar bar), and
that it is these particular slaves that must be surrendered.14 One decree
from December of 1851 mentions “Persian [captives] and other captives”
bought from Khiva or from “Khivan subjects,” with the captives in ques-
tion being “transferred from one owner to another” or given as dowries.
The decree calls upon all Kazakhs under Orenburg’s authority, without

He adds, in a note: “When I afterwards drew the attention of the Russians to the
occurrence, they laboured to excuse themselves, saying that they did not desire to accus-
tom the Turkomans to such large ransoms, for that with any encouragement these bold
robbers would devote themselves night and day to their profitable depredations”
(Vambery, Travels in Central Asia, 79).
14
TsGAKaz 4.1.3573, unnumbered folio.
170 The Khan as Russian Agent

exception, to turn over these captives promptly. Henceforth, the decree

explains, if any slaves or captives were to be found living with any Kazakhs,
regardless of rank, that individual would be exposed and brought to
justice, and they would be subject to “a monetary penalty of 150 silver
rubles for each captive and 6 months’ imprisonment or, depending on the
gravity of the circumstances surrounding the acquisition of the captive,
deprivation of property rights, corporal punishment, and banishment to
Siberia.”15
The Border Commission evidently printed and circulated at least 1,000
copies of this decree by January of 1852.16 Further copies of this or similar
decrees – in groups of 100 and of 230 pressings – were issued in the months
to come.17 No information is available on the reception of these docu-
ments in the steppe. Since the majority of Kazakhs were illiterate at the
time, we may assume that the decrees were meant to be read aloud.
We know, at least, that two or three issuances of the general emancipation
order were not considered sufficient: The Border Commission would issue
still more decrees of a very similar nature over the course of the next two
decades. We also know that the Kazakh informants who chose to coop-
erate early on in the process of liberating slaves in the steppe included
powerful figures like Sultan Arslan. The Sultan received reports from his
own informants on slaveholders under his authority, ensured the release of
captives, and then delivered them to the Border Commission along with
testimony that often included a chain-of-transmission revealing how he
learned of the captive’s presence among his subjects.18
In general we can only speculate about the various individual motiva-
tions that would have inspired Kazakh elites to cooperate with the
Russians in enforcing manumission decrees. The contents of one Russian-
authored decree in Turkic, however, dating from 1860, can help to clarify
at least one motivation: Here, Russian officials threaten punishment not
only for those Kazakhs who held slaves and captives, but also for those
leaders who claimed the offenders as subjects. Bearing the signatures and
seals of five Kazakh bays (elites), this decree, like the others, declares that
“if any Kazakh is in possession of a slave, he must turn this slave over to the
governor, and after this he must not buy any [other] slave” (gar kim birär
qazaqning qolinda qul bolsa, ul qulni hākim husurina täbshurur wa mun-
˙
dan song qul sätub älmas). It concludes with the announcement that if any

15
TsGAKaz 4.1.3573 ff. 112a–113a. See also Semeniuk, 237.
16
TsGAKaz 4.1.3573 f. 132a. 17 TsGAKaz 4.1.3573 ff. 147a; 188a.
18
TsGAKaz 3573 f. 193a.
 Kazakh Informants along the Steppe Frontier 171

Kazakh is to “open his hand” (qol ächilsa), i.e. to release his or her slave,
then this Kazakh as well as the bay overseeing him or her (ul qazaq ham
aning ustidan qarab turguvchi bay) would not be “subject to merciless
inquiry” (hich bir rahimsiz tekshirdilar).19 Elites, in other words, could
˙
be held accountable for the slaveholding of their subjects.
Another document, dating from June of 1852, hints at more direct
threats against noncompliant steppe elites. Having been informed of the
captivity of an Iranian whose name is recorded as Asan Yakubov, Russian
officials informed the sultan of Asan’s owner’s orda that the captive was
henceforth to live and work where he pleased, and that no others should
take him into slavery. The sultan was tasked with the execution of these
orders, which were evidently passed to him in secret by way of one of his
retainers. (This individual was referred to as a “clerk” in the document; the
official who drafted it noted that he disapproved of the instructions, all the
more so since the “clerk” was merely a dependent of the sultan and there-
fore provided no additional influence over the outcome). The secret orders
included an explicit warning: If anything transpired that was inconsistent
with the above demands, Russian officials would be informed of it.20
In reality, slaveholding Kazakhs whom other Kazakhs brought before
Russian officials were not always dealt the punishments specified in these
dramatic Russian decrees. One group of offenders, who were charged with
owning four slaves among them, was evidently pardoned entirely; accord-
ing to the official report, it was not thought necessary to hold them to any
liability for failing to surrender their slaves, since they simply “did not
know of the prohibitive order” (o zapretitel’nom rasporiazhenii) against
slave-owning.21
Captors and slave-owners, meanwhile, could stand to benefit from
surrendering their contraband at border offices, at least after 1852.
In January of that year, a Russian border official proposed that offenders
who surrendered their captives should receive compensation (voznograzh-
deniia) for their loss.22 It appears that this policy was implemented in at
least some cases; one slave owner, for example, was issued one ruble and
fifty kopeks in compensation for each of the thirty-six sheep he had
formerly paid to a fellow Kazakh for the Iranian slave he surrendered at the
border.23 This kind of compensatory system may hint at the disappointing
impact of previous emancipation decrees. Compensating slave-owners for

19
TsGAKaz 383.1.89, unnumbered folio. 20 TsGAKaz 4.1.3641 f. 22a.
21
TsGAKaz 383.1.184 f. 19a. 22 TsGAKaz 4.1.3641 f. 140a.
23
TsGAKaz 4.1.3573 f. 133a.
172 The Khan as Russian Agent

their loss, moreover, is not altogether different from paying ransom,

and the new system surely risked offering a further incentive for captive-
taking. Nevertheless, authorization was given for systematized compensa-
tion, to be implemented in June of 1852.24 It is not clear that the system
was fully implemented, however, and may have been stalled by officials’
misgivings: By September of 1852, the official who had recommended the
compensation system eight months earlier submitted a report noting that
he still had not received a decision from his superiors on the question of
compensating compliant slaveholders. He reiterated his recommendation,
urging them to come to a final decision on the matter.25
Slaveholders and informants were not the only parties who sometimes
stood to gain compensation during the manumission process. The testi-
mony of an Iranian slave whose name was recorded as Yul-Muhammad
Asanov indicates that the freedmen themselves could sometimes earn
reparations for their suffering. Asanov relates in his interview with
Russian officials that he was first taken captive at the age of seven by
Turkmens who soon sold him to a Kazakh of the Alchin clan. He spent
eight years with this owner before being sold to a Kazakh of the
Dzhagalbay clan. He fled this owner, but he was captured and then sold
to a Kazakh of the Chagatay clan whose name is recorded as Bay Murzabek
Yamanbaev. He lived as Yamanbaev’s slave for the next twenty-three years
until, in 1854, he was freed. “Upon being freed from 24 years of slavery,”
Asanov testifies, “Murzabek Yamanbaev, instead of giving me any remu-
neration, cut my head with a knife and hit me in the face with a whip,
knocking out one of my teeth. Hoping for nothing more than what is my
share, I have repeatedly appealed to the Russian officials. But for their part,
down to the present time, I have received neither my recompense nor any
information [concerning my case].” Upon being freed, Asanov reports
having been given only “meager rags” to wear, and he could not find
a way to provide for himself “even by the most strenuous effort.”
In these dire circumstances, he turned to Russian officials to seek compen-
sation from his former master and, as he phrased it in his interview, to
“dispense gracious satisfaction to me for my years of slavery.”26 Following
his testimony, the wheels of justice turned at a leisurely pace. Asanov first
came before officials on August 2, 1857. By April of 1858, his former
owner had evidently died, and it was his heirs who were at that time
ordered to “render compensation” (okazat’ posobie) to Asanov for his

24
TsGAKaz 4.1.3573 f. 187a. 25 TsGAKaz 4.1.3641 f. 140a.
26
TsGAKaz 4.1.3730 ff. 16a–18a.
 Kazakh Informants along the Steppe Frontier 173

twenty-three years of service. Finally, in November of that same year,

it was confirmed that he had received some compensation from these
heirs.27
Taken altogether, the influx of slaves and slaveholders arriving at the
border in the 1850s and 1860s demonstrates that Russian abolition
decrees and the policies and practices that came with them had at least
some impact on the steppe’s slave population. Even at mid-century, how-
ever, progress was decidedly gradual: Further abolition decrees were dis-
seminated in 1860 and again in 1861; the last known decree was circulated
in the steppe as late as 1869.28 By that time, the influx of Iranian slaves
from Khurasan had slowed considerably. While the conquest of the
Khwarazmian domains in 1873 did not, as we have seen, end the slave
trade, a major waystation and market for slaves was thrown into turmoil.
The resale market in Khwarazm that had long been available to Kazakh
captors and dealers surely contracted. Meanwhile, with the strengthening
of Russian border defenses and the weakening and increasing dependency
of neighboring Kazakh clans, fewer Russian subjects were falling into
Kazakh hands. It was in this context that Russian-sponsored emancipation,
however modest in its impacts, further depleted the steppe slave popula-
tion from within. The final quarter of the nineteenth century was tanta-
mount to dusk for the slave trade in the Kazakh steppe.
The sun would not set on slavery itself, however. As I have shown in
previous chapters, many slaves who were “manumitted” at the Russian
border were simply returned to their former owners, either as “foster
children” or as laborers. Some slaves had free Kazakh wives and had
integrated into free society to a degree unimaginable for slaves in other
regions of the world. Many others were not so lucky.
The success of Russian emancipation efforts relied entirely on Kazakh
informants. The Empire was at the mercy of these informants simply to
identify slaveholders and slaves amid the general population. Other agents
were often needed to transport offenders to the border along with their
slaves. The pool of local go-betweens willing to take on these roles was
remarkably diverse: Khans and sultans, fugitives and thieves all played
a part. Each had his or her own motives, and each was seemingly able to
negotiate individual terms of compensation. Rich opportunities for
manipulating these negotiations must have presented themselves to many
an agent. In the end, however, whatever the costs, it appears that the
Russian Empire, by engaging informant networks, achieved certain

27 28
TsGAKaz 4.1.3730 f. 28a. Semeniuk, 238.
174 The Khan as Russian Agent

substantial aims: it reduced the threat of captivity for Russian subjects in or

near the steppe; it engaged in a novel means of extending Russian govern-
ance over subjects and neighboring populations; and it initiated new forms
of dependency and service among the Kazakhs.

INFORMANTS AND LOCAL AGENTS AMONG THE TURKMENS

Meanwhile, a parallel strategy was being enacted to the south, along the
shores of the Caspian. In 1828, the Treaty of Turkmenchay stipulated that
Iranian ships would cede the right to navigate the Caspian and its coasts
freely; henceforth, no Iranian military vessels would patrol the Sea. These
rights were transferred exclusively to the Russian Empire. Russia’s new-
found supremacy on the sea came with certain responsibilities for the
benefit of Iran, however: The Russian military would take over the task
of policing the Caspian shore, along which Turkmen raiding parties had
long engaged in systematic attacks on Iranian villages for the purpose of
taking captives for ransom or sale. In this effort, the Turkmens had been
aided by the challenges of patrolling the terrain: The thick reed-beds
offered them excellent cover; their narrow, flat-bottomed boats presented
a low profile as they cruised along the coastline; and, if caught on their way
to launching an attack, empty-handed Turkmens could always claim that
they were merely fishing.29
For more than a decade before the 1828 treaty, the Russian Empire had
been using the Turkmens as a “proxy” weapon against Iran. Yomut
Turkmens who supported an 1813 rebellion against the shah in
Astarabad are believed to have been acting as Russian agents. Kiyat
Khan, leader of the Jafarbay clan of Turkmens, likewise enjoyed Russian
support even as the governor of Mazandaran led military expeditions
against him in 1826.30 British officials stationed in Tehran, observing the
Turkmens’ continual borderland raids, were well aware of Russia’s strat-
egy; in 1837, the minister Sir John McNeil wrote that the Yomut Turkmens
“in all probability will not submit to Iran so long as the connection they
have formed with Russia will afford them sufficient protection.”31 That
same year, Kiyat Khan, who had been leading kidnapping raids along the

29
Cf. C. A. Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),”
Istoricheskii vestnik 12 (1900), 1037.
30
Mohammad Ali Kazembeyki, Society, Politics, and Economics in Māzandarān, Iran,
1848–1914 (London: Routledge, 2003), 45.
31
Kazembeyki, Society, Politics, and Economics in Māzandarān, Iran, 1848–1914, 45.
 Informants and Local Agents among the Turkmens 175

Caspian coast, was able to escape an Iranian punitive expedition by taking

cover with the Russian military, which spirited him to safety and even sent
navy forces to head off the Iranian mission. Mohammad Ali Kazembeyki
sums up the situation: “Those of [the Turkmens] who navigated with
a Russian passport, remained out of Iranian judicial jurisdiction, and
even when they committed piracy, Russian protection prevented the
local authorities in Māzandarān and Astarābād prosecuting them.”32
Meanwhile, the Russian military monopoly over the Caspian Sea made it
impossible for Iranian forces to pursue Turkmens beyond the beaches.
The basics of this dynamic – Turkmens raiding northern Iran and then
taking cover in Russian-patrolled territory – persisted for decades, such
that, in 1860, the British consul C. F. MacKenzie was reporting much the
same news as had been reported by McNeil over twenty years prior:
“No point of the coast up to Ashraf,” he wrote, “is free from [Turkmen]
attacks . . . The Russians are perfectly aware of this and laugh at the
insufficient efforts, supineness, and apathy of the Persian chiefs.”33
By the late 1850s, however, a problem had emerged for Russian military
personnel stationed in the region. It seems that the same Turkmens who
captured Iranians sometimes also captured Russian soldiers and officers.
The services of informants – agents who might tip off the Russians to
coming raids, or identify chief raiders – were thus badly needed, and the
Russians sought them among Turkmen elites. As they had done with the
Kazakhs of the steppe, Russian military officials attempted to build new
patronage networks among the Turkmens. Backed by the ever-present
threat of military force, some Turkmen groups along the Caspian surren-
dered to a system by which their khans would have to be confirmed in their
status by Russian officials at the seaside military station. According to the
officer C. A. Gunaropulo, the khans who consented to working with the
Empire could receive considerable benefits: They drew a salary and
received gifts along with it. But their position could be precarious.
In those cases in which perpetrators of raids and robberies were not caught
in the act, and their captives were not returned, the responsibility would
fall on the nearest aul and its headmen. The community would have either
to extradite the criminals, if they were known to them, or to make every
effort to find them. Otherwise, the chief elder of the aul would be faced
with stern punishment. According to Gunaropulo, the man regarded as the
chief of the community could carry exclusive responsibility in these cases.

32
Kazembeyki, Society, Politics, and Economics in Māzandarān, Iran, 1848–1914, 46–47.
33
Kazembeyki, Society, Politics, and Economics in Māzandarān, Iran, 1848–1914, 49.
176 The Khan as Russian Agent

Aware of the benefits of cultivating good relations with the Russian mili-
tary station, they would sometimes make an effort to stop raids and
robbers themselves.34
The story of the notorious slave-raider Ergeld Khan serves as
a remarkable case study in Russian patronage along the Caspian shore.
After years of Russian military patrols and interventions had interfered in
his main line of work, Ergeld Khan was in a desperate situation.
Nevertheless, after his long tenure as the chief raider in his community
(“bravely serving his tribe,” as Gunaropulo phrases it), he maintained
considerable influence over his comrades and high esteem among them.
The elders of his aul, observing the benefits of maintaining peaceful rela-
tions with the Russian military station, resolved to propose Ergeld as
khan – pending Russian approval, of course.35 It was a highly intelligent
maneuver on the part of the elders: With this gesture, they could relieve
themselves of the accusation that their community was harboring a felon
“at large”; they could alleviate the destabilizing pressure that the presence
of a destitute and dissatisfied warrior and his followers could have on the
aul; they could maintain good relations with the station and reap its
rewards; and they could ensure all of this by offering the Russians some-
thing of immense value: an ally in the fight against slave-raiding who knew
the business very intimately indeed.
The benefits of working with Ergeld Khan were not lost on the Russians,
or on Ergeld himself. Lengthy negotiations followed, during which it was
resolved that Ergeld Khan would be well-compensated for his cooperation.
Gunaropulo had the opportunity to meet him during the khan’s visit to the
station, and the Russian officer barely conceals his admiration in describing
him: “He was quite a remarkable individual, both in his outward appearance
and in the details of his life, which was full of sundry adventures. He was
a tall man of majestic bearing, expressive features, and a penetrating gaze.
He was covered in [old] wounds. This man, with his appealing appearance,
did not give the impression of being a robber. He was no more than 40 years
old, and his name inspired terror in the Persian coastal dwellings, but he
rarely resorted to the dagger – only [doing so] to save his own life.”36 In the
end, Ergeld took the title of “khan-intermediary” (khan-posrednik).
It was not long before Ergeld Khan would have a chance to demonstrate
his usefulness to his new patrons. An Iranian captive was taken by

34
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1039–1040.
35
Ibid.
36
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1040.
 Informants and Local Agents among the Turkmens 177

Turkmens, and Ergeld Khan informed station officers that the captive was
presently located in an aul far from the coast, but that the relatives of his
captors could be found in the aul of Gasan-Guli (sic). Russian station
officers had previously shown relatively little interest in freeing Iranian
captives, but they made an exception in order to enlist Ergeld Khan’s
services. Gunaropulo joined Ergeld in the efforts to secure the captive’s
release, and he provides a detailed narrative of his adventure with the
Khan. As with Witkiewicz’s narrations of the effort to free a captive from
the steppe, Gunaropulo’s account offers a rich view of the politics of
manumission on the “frontlines.”
The proximity of the captors’ relatives was considered a useful lead:
“Among the Turkmens,” Gunaropulo writes, “there is the custom of aven-
ging any insult to their relatives, but also of answering for their actions.”37
Gunaropulo set off in a schooner along with the station chief, Ergeld Khan,
an interpreter, and some others. They were accompanied by ten armed
sailors divided between two Turkmen boats – brought, Gunaropulo writes,
for the sake of “presentation.” They anchored the schooner about four
miles from shore, and continued onward in the two Turkmen boats,
entering the mouth of the Gasan Guli River. On either side of the river
there were “countless tents,” and the arrival of the boats evidently inspired
“great wonder and surprise” among the Turkmens. “It seemed as if the
entire population of the aul poured forth to meet us,” writes Gunaropulo,
“and the banks of the river resembled two colorful, rippling ribbons, from
which could be heard a vague hum [of voices], which accompanied us the
whole time as we moved along the river.”38
To avoid the throngs of people, they sailed onward and made landing
slightly downriver. A few minutes later, elders from the village came to find
out the reason for their arrival. Gunaropulo suggested to the khan that he
should take charge in the negotiations. A servant of the khan brought out
a samovar, a chest, some carpets, some dried fruit, and other items, and the
Turkmen elders sat in a circle and had tea. The khan invited Gunaropulo to
join them, ordering his servant to provide the officer with a pillow, of the
kind provided to esteemed guests at Turkmen gatherings. Gunaropulo
asked the interpreter to convey to him the precise contents of the ensuing
discussions. After some pleasantries, the khan explained to the elders that
the Russian station chief was very unhappy with them because they knew
the location of the Iranian captive but had made no efforts to secure his

37 38
Ibid. Ibid.
178 The Khan as Russian Agent

release. In light of the privileges available to them and the good relations
they had thus far maintained with the Russians, the khan asked that they
not rupture that “friendship” by impeding the common effort to free the
captive.39
The elders swore that they had taken no part in what had befallen the
Iranian captive, and that they did not know if he was really located in the
aul mentioned by the khan. They insisted on the veracity of their testimony
as Ergeld reiterated his suspicions. Gunaropulo allows that the elders may
well have spoken honestly, but the khan had been provided with “over-
whelming evidence” by his trustworthy scouts (lazutchiki), and he felt
certain of the captive’s whereabouts. Ergeld was unwavering in the matter;
indicating to Gunaropulo and the interpreter that they should follow him,
he stepped away from the gathering and proposed that it was necessary to
invite the elders onto the schooner, and to hold one of them in captivity as
collateral for the return of the Iranian captive. When Gunaropulo invited
the elders to join him onboard the schooner, where they should receive
a personal audience from the station chief, they bowed deeply and indi-
cated their consent.40
Their distance from the village had not, in the meantime, provided
much relief from the throngs of onlookers. “We were surrounded by
a dense, colorful ring,” Gunaropulo writes, “which, gradually closing in,
would have crushed us had the requests and admonitions of the khan and
the elders not mitigated the crowd’s intensity.”41 The elders’ march toward
the boats provoked alarm among the crowd, and several Turkmens with
“fearful, questioning eyes” ran over and grabbed Gunaropulo by his
sleeves and the tails of his coat, crying out “Sibir! Sibir!” (“Siberia,
Siberia!”) – they evidently recognized the possibility that their elders
could be imprisoned, and feared that they would be sent to the notorious
prison camps of Siberia. A dangerous moment followed:
At this time, a sailor who was overzealous in the performance of his duty . . . shoved
a Turkmen woman off of me so violently that she fell. Words cannot express the
brutal expressions which then appeared on the faces of most of these Turkmens, of
both sexes. Fearing a formidable demonstration on their part of the deep indigna-
tion which, although somewhat weakened, evidently still gripped them, I made
a show in that same moment of striking the sailor, though really I merely knocked
his cap from his head. A single moment, and there was a metamorphosis: the grim,
embittered faces brightened, and we [managed to] enter the boat, albeit not
39
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1040–1041.
40
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1041.
41
Ibid.
 Informants and Local Agents among the Turkmens 179

without some effort in delicately reducing, as much as possible, the impediments

presented to us by the Turkmens who were holding onto the elders, while assuring
them that the elders would return to the village that same day.42

As they sailed downriver with the elders, “angry Turkmens dotted the
shoreline” shouting at the boats, running after them, shaking their fists,
and spitting in their direction. Arriving at the schooner, the station chief
greeted the elders and they made their introductions. After this, the
Russian official pointed to one of the elders who had been recommended
to him by Ergeld as a relative of the captors who were sought, and he
announced that he was to be placed under arrest until the return of the
Iranian captive. The elders – unlike the others from their aul – evidently
did not expect this outcome. They expressed their displeasure, but had no
choice other than to submit. The schooner weighed anchor with its hos-
tage onboard.43
As the schooner approached the military station, the Turkmen elder
caught sight of something which filled him with palpable dread: a three-
masted transport vessel, decommissioned from service and standing on
dead anchors. The primary purpose of the vessel was to store gunpowder,
but it served also as a dungeon for the “lower sorts and guilty Turkmens”
(dlia nizhnikh chinov i provinivshikhsia turkmen). This fact, Gunaropulo
writes, was well-known among Turkmens. It is not clear whether this
dungeon is where they detained the elder, who was innocent of all but
being related to a suspected slave-raider. A few days later, his fellow elders
from the aul came to plead with the station chief to release their associate,
vowing to see the Iranian captive returned.
The Russians’ tactics here were inarguably brutish, inhumane, and quite
possibly counterproductive in the long term: It is unlikely that they could,
in the future, expect anything but the harshest welcome from the aul
whose elders they had offended. It seems that the people of that commu-
nity already knew very well the dangers involved in negotiating with the
Russians. Their outrage upon seeing their elders heading downriver was
well-founded, and it hardly seemed to surprise Gunaropulo. The presence
of Ergeld Khan may or may not have impressed the elders, but it certainly
did not serve to ease the minds of the general populace. Which is not to say
that Ergeld’s participation had no value: From the beginning, the entire
operation appears to have been under his direction. His own network of
informants identified the likely whereabouts of the Iranian captive; he

42
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1041–1042.
43
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1042.
180 The Khan as Russian Agent

pointed out their relatives’ community; he undertook negotiations with

the elders entirely on his own; and he suggested the hostage trick. He was
no mere informant for the Russians in this enterprise; he was the mission’s
commander.
The mission, moreover, was a success: Promptly after pleading with
Russian officials to free their compatriot, the elders managed to deliver the
Iranian captive to the station.
Such complicated missions for the freeing of hostages were not always
necessary. Most often, it appears that the Russians of the Caspian station
simply engaged in the ransom economy, paying a redemption fee for Russian
captives. According to Gunaropulo, the station had a special fund for ransom
payments, formed out of the voluntary contributions of officers serving there.
Each officer contributed three “Dutch coins” (gollandskii chervonets)
per year.44 The informal nature of this arrangement – and the fact that the
ransom fund was not subsidized by higher authorities – suggests that ransom-
ing captives was not part of the mission that the Empire had intended for the
Caspian station. Most likely, the “ransom collection” arose as a practical
response to the complexities involved in the more general mission of “pacify-
ing” the Turkmens. Suffice to say, however, it is not clear how perpetuating
the ransom economy might have aided in reducing the deprivations wrought
by Turkmens. Instead, it proves the bargaining power of the Turkmens and
the relative weakness of the Russians, their powerful station and equipment
notwithstanding.
Sometimes ransom exchanges went terribly wrong for the Russians.
In early 1860, two Russian sailors, named Potakeev and Ivanov, were
captured by Turkmens along the Iranian coastline. These two captives
became the joint property of three families of Turkmens, and these families
were jointly involved in negotiating the terms of his release. Several times,
Gunaropulo tells us, the negotiations were concluded to the satisfaction of
all parties, and the agreed-upon ransom money was brought to the
families, only to be rejected: Each time, the families would dramatically
raise the ransom price. The process, overall, was incredibly dilatory:
The captives remained with their Turkmen captors for seven years, “per-
forming the most burdensome of labors,” before another strategy was
attempted. Once again, it was a Turkmen “khan-intermediary” who led
the way. Ana-San Khan [sic] proposed a novel approach: The khan sug-
gested that a trustworthy Turkmen should be sent to the aul where the
captives were residing with the ostensible mission of purchasing an

44
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1043.
 Informants and Local Agents among the Turkmens 181

arghumak horse – a Central Asian breed famous for its speed and hardi-
ness. Once in possession of the horse, the Turkmen agent would seize an
opportune moment to give it over to the Russian captives, so that they
might ride it to freedom. Incredibly, the plan was a success: The sailors
were freed, and the horse on which they escaped became joint property of
the station guards.45
The “khan-intermediaries” in the employ of the Russian Empire could,
no doubt, serve as very effective agents. It is clear that they served the
Russians as a survival strategy, both for themselves and for their commu-
nities. Threatened with the loss of proceeds from raiding and ransoming,
and naturally alarmed by the proximity of the Russian navy, these inter-
mediaries did what was necessary to maintain a measure of independence
as well as to secure some revenue. In the short term, these aims appear to
have been met: Compensations were offered, and “indirect rule” remained
the order of the day for communities led by “intermediary” khans. But in
the long term, some Turkmen leaders found their aims tragically frustrated
by their Russian patrons.
In a letter dating from no later than 1839, Kiyat Khan, the Turkmen ruler
who had enjoyed the protection of the Empire for decades as he led raids in
Iranian territory, related his sorrow and disappointment to a top Russian
official in the Caucasus. In 1812, the khan writes, a Russian general sent an
envoy to the Teke, Yomut, and Gökleng Turkmens, and members of these
tribes chose Kiyat Khan by common agreement to conduct necessary political
errands in Russia. “Now,” he writes, “I am fading and nearing the end of my
life; it is a pity that not one of my desires has been fulfilled.” He made
“repeated requests” to leading Russian officials concerning various subjects,
but never received any satisfaction. Even so, the khan pledges that he has not
lost his zeal for diligently serving the Russian state. The greatest disappoint-
ment had come quite recently: He had received the news that a Russian
plenary minister had indicated to the Iranian court that the khan’s territories
along the Atrek and Gurgen Rivers would be the rightful territory of Iran.
The minister evidently paid the khan a visit thereafter: “Yesterday the min-
ister came to our post, and he very easily learned the business of centuries!
We entreat Your Excellency to ask [him] what graves are in the aforesaid
domains: [are they] Turkmen or Persian? Your Excellency will likely not agree
with the perspective according to which we should be deprived of land left to
us by our forefathers.” The Khan concludes his complaint with a request that
he himself may have known to be outlandish:

45
Gunaropulo, “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera),” 1043.
182 The Khan as Russian Agent

I have an abiding desire which, to my misfortune, has not hitherto been achieved,
though I disclosed all to the chief superintendent [Rtishchev]; I disclose it now to
Your Excellency, so that the weight of it shall not lie upon my chest. Here it is:
I have wished that His Majesty the Emperor should give me an army for the
conquest of Astarabad and the subjugation of its local inhabitants to my will. But,
to my boundless regret, this goal has not been reached. Nearing the end of my life, it
is a pity to die without having achieved my desire.46

As disappointed as the khan may have been that he never had a chance to
conquer Astarabad, his bold appeal should not overshadow the deeper
tragedy in his letter. Caught, like his fellow Turkmen elites, between
Russia, Khwarazm, and Iran, the khan worked with the Russians for
some three decades only to find, at the end of his life, thanks to Russian
negotiations, that his Turkmen forefathers’ graves were on Iranian soil.

46
Russko-turkmenskie otnosheniia v XVII–XIX v. (Ashgabat, 1963), doc. no. 244.
 7

The Conquest of Khiva and the Myth of Russian

Abolitionism in Central Asia

The Russian conquest of Khiva in the spring of 1873 came with high
expectations among the abolitionists of rival empires. “The first steps
have already been taken,” the British officer Herbert Wood would soon
afterwards write, “for the diffusion of light in these dark places of the
earth, for Humanity owes to Russia the cessation of brigandage and
slavery, which from the earliest historical times have been the scourges of
the Oxus countries.”1 When the expected emancipation was accom-
plished, in the form of a decree drafted by the khan at General
Kaufman’s behest, few were likely surprised. But perhaps they should
have been. In this chapter, I will argue that the Russian-sponsored eman-
cipation of Khwarazm’s slaves was far from inevitable; that it had not been
planned in advance by Kaufman or his superiors; that abolitionism, con-
trary to popular belief, played no significant part in motivating the Russian
conquest of the region; and that credit for the slaves’ mass emancipation –
one of the most revolutionary acts in the history of Central Asia – should
rightfully go to the slaves themselves.
Throughout the 1860s and early 1870s, right down to the eve of Khiva’s
conquest, a declining but significant trade in Iranian slaves had continued,
notwithstanding the near-total cessation in raids among steppe Kazakhs
and the concurrent Russian and Qajar efforts to subjugate Turkmens to the
south. In fact, imperial efforts in Khurasan sometimes served only to
provide more captives for Central Asian markets. In 1860, for example,
a Qajar campaign brought an estimated 13,000 infantry, 10,000 cavalry,
and 33 cannons to Merv, only to be routed by Teke adversaries; so many

1
Wood, The Shores of Lake Aral, xii.

183
184 Conquest of Khiva and Myth of Russian Abolitionism

captives were taken and sold that, according to Kuropatkin, the price of
a single Iranian slave in the markets of Bukhara and Khiva fell to seven
rubles and fifty kopeks.2 M. Alikhanov-Avarskii ventured to propose the
scope of the disaster: “a Persian detachment of Prince Sultan Murad, on its
way to Merv, left no less than 20,000 of their soldiers in the hands of local
Teke, and the majority of those were sold to Khiva.”3
Miserable conditions in northern Iran in the early 1870s may also have
contributed to the trade: Grodekov observes that the “terrible famine that
raged throughout Persia” in 1871 resulted in the Iranians being “so wea-
kened by want of food that they offered the Tekes no resistance whatever.”
He met with a man who had been held captive by Tekes for many years,
and the man related to him the ease with which these weakened victims
could be taken: “[he] told me that the poorer nomads used to ride on
donkeys to Serakhs, armed only with cudgels, and drive back the villagers
in droves to slavery at Merv. He himself was captured in this manner
during one of the Alamans.”4
As we have seen in a previous chapter, the subjugation of Bukhara
by June of 1868 had not brought about the end of the slave trade in
those quarters either – nor did such a goal appear to be a matter of urgency
for its Russian conquerors. It was not until 1873, some five years later, that
the Russians would write abolition into a treaty with the Bukharan Amı̄r,
and even then the ruler was allowed a deferment of ten years before full
emancipation would have to take effect. In the eighteen-point treaty in
question, in which ten of the points related to commercial relations
between Russia and Bukhara, the abolition of the slave trade occupied
the seventeenth point:5
In deference to the Emperor of all the Russias, and for the greater glory of His
Imperial Majesty, His Eminence the Amir Sayed Muzaffar has resolved that hence-
forth and for ever the shameful trade in men which is so contrary to the laws of
humanity, shall be abolished within the limits of Bukhara. In conformity with this
resolution, Sayed Muzaffar shall immediately send to all his beks the strictest
orders to that effect. Besides the order abolishing the slave trade, commands shall
be sent to all the frontier towns of Bukhara to which slaves are brought for sale

2
A. N. Kuropatkin, Zavoevanie turkmenii: pokhod v akhal-teke v 1880–81 gg., 96. A. Maslov
offers a similar summary of the ill-fated campaign, estimating the Qajar forces at 20,000
and the resulting price of a slave at an even more shocking 1.5 rubles each: See Maslov,
Rossiia v srednei azii, 386.
3
M. Alikhanov-Avarskii, Pokhod v khivu (Kavkazkikh otriadov) 1873, 280. Emphasis added.
4
Grodekoff, 129.
5
Demetrius Charles Boulger, England and Russia in Central Asia, Vol. 1 (London: W.H.
Allen, 1879), 337.
 Conquest of Khiva and Myth of Russian Abolitionism 185

from neighbouring countries, to the effect that in case slaves should be brought to
such places, notwithstanding the orders of the Amir, the same should be taken from
their owners and immediately liberated.

The ambitious language of the treaty was surely undermined by the ten-
year grace period permitted, and indeed the slave trade in Bukhara con-
tinued. Nevertheless, the rumor had evidently spread both among Central
Asian slaves and among foreign journalists that the era of slavery was
drawing to a close and that Russian troops were consummate liberators,
breaking slaves’ shackles wherever they went. So it was that the slaves of
Khwarazm began to revolt against their masters and free themselves before
the Tsar’s invading armies had even arrived at the capital.6 The closer the
Russian detachments came to the city, the greater the number of slaves
seizing the occasion to liberate themselves in the “conviction and the
hope . . . that the fall of Khiva would also bring them their freedom . . .
[A]t first they appeared in small parties, but the bands steadily increased as
the [Russians] advanced.”7 Many joined the Russians, marching beside
them.
The possibility of slaves fleeing or rebelling at the approach of the
Russian forces had been anticipated by many slave-owners well in advance
of the Khivan campaign.8 Alikhanov-Avarskii witnessed that Iranians,
freeing themselves, “flee from their masters and appear at our camp in
masses, despite the fact that the Khivans, as a precaution, have in recent
times been keeping them chained and under close supervision.”9 By all
accounts, slaves met their presumed liberators with dramatic displays of
ecstatic joy and relief. Grodekov described how “[t]he people loudly
greeted the troops, and especially boisterous were the noisy greetings of
the Persian slaves. They had begun to free themselves immediately upon
the troops’ entry into the capital of the Khanate, and in the crowd one
could see many slaves jostling forward, showing the troops arms and legs
bloodied from tight shackles.”10 MacGahan would later witness a similar
spectacle, writing how he and the group of Russian soldiers he accompa-
nied “came upon a crowd of Persian slaves, who received us with shouts,

6
E. Zheliabuzhskii, Ocherki i zavoevanie khivy (Moscow, 1875), 116.
7
Schmidt, 121. See also V. N. Trotskii, Opisanie khivinskago pokhoda 1873 goda (St Pb,
1898), 229.
8
While traveling to Khiva, David Ker interviewed a Russian citizen who predicted that “as
soon as our brothers get within cannon shot of the place, the Persians will up and chop the
Khivans to bits” (Ker, On the Road to Khiva [London: Henry S. King and Co.], 101).
9
Alikhanov-Avarskii, 217.
10
Grodekov, Khivinskii pokhod 1873 goda deistviia kavkazskikh otriadov, 314.
186 Conquest of Khiva and Myth of Russian Abolitionism

cries, and tears of joy. They were wild with excitement. They had heard
that wherever the Russians went slavery disappeared, and they did not
doubt that it would be the case here. Some had already liberated them-
selves; and I saw several engaged in cutting the chains of three or four
miserable beings, shouting the while, and laughing and crying all at once in
the wildest manner.”11
Foreign observers, too, would have had good reason for suspecting
that abolition was on the Tsar’s list of priorities. After all, the Tsar’s
public announcement and explanation for the campaign, which appeared
in the newspaper Russkii invalid and was translated for the New York
Times, makes no fewer than four references to slavery and captivity in
describing its incentives. For example, the circular alleges that the khan
had previously been prompted to “immediately set at liberty all Russian
prisoners,” but that he had declined to do so, and so they remained
“slaves in the possession of the Khan and his high functionaries.”12
It also notes “several merchants and others” who were killed or “dragged
into slavery” on the road between Orsk and Kazalinsk.13 Slavery was
a topic of concern and an incentive for the campaign at the highest levels –
or so it was made to seem.
It is perhaps for this reason that the khan’s final, desperate effort to
stave off invasion came in the form of selective manumission. The khan
ordered twenty-one Russian slaves gathered together and sent off to
General Kaufman along with a petition (ʿarznāma) asking for peace, all
to be delivered by a high-ranking emissary (identified in a Khivan source as
Murtaża Biy Khwāja, hākı̄m of Khwāja Eli). Kaufman neither accepted the
˙
petition nor permitted the emissary to return home, keeping Murtaża Biy
with his retinue on the way to Khiva. The twenty-one former slaves were
duly received, but their manumission did nothing to slow Kaufman’s
advance.14 These freedmen, of whom eleven were Cossacks, had all been
captured in 1869 and 1870 by Kazakhs and sold to Khiva; it was evidently
claimed by the khan that they were the only Russian subjects still held as
slaves there, “with the exception,” MacGahan writes, “of one old man
taken in Perovsky’s disastrous expedition” of 1839–40, who had con-
verted to Islam, married, and preferred not to leave.15 It is unlikely that

11
MacGahan, 233; see also Alikhanov-Avarskii, 217.
12
“Russia and Khiva: The Empire’s Manifesto Prefatory to the Expedition,” New York
Times, April 13, 1873.
13
Ibid.
14
MS IVAN Uz No. 12581, ff. 21b–22a. See also MacGahan, Campaigning on the Oxus, 20.
15
MacGahan, 20. More Russian slaves were to be freed during the ensuing conquest.
 Conquest of Khiva and Myth of Russian Abolitionism 187

the khan believed these concessions alone would be sufficient to convince

Kaufman to turn his troops back. He probably hoped the gesture might
show his willingness to negotiate given the opportunity. Indeed, the
previous year he had sent two envoys – one to the Grand Duke Mikhail
in the Caucasus and another to Orenburg – in an apparent effort to arrive
at a diplomatic resolution.16 But the opportunity was not to come.
As Kaufman entered the capital on March 28, 1873, a large-scale slave
uprising had begun in earnest. Emboldened by rumors of the city’s
imminent downfall, slaves broke from their owners, plundered them,
and in some cases, according to a later Khivan source called the
Āzādnāma, began tormenting them.17 Scenes like this could be witnessed
not only in the capital but also in towns and villages throughout
Khwarazm, and it was reported that both owners and their (now former)
slaves seized the occasion to engage in acts of cruelty against one
another.18 Before long, groups of Khivans began approaching
Kaufman, pleading that he do something to help put down the uprising.
In what may have been the first act of governance on the part of the
Russian Empire in Khwarazm, Kaufman ordered an inquiry into the
crimes of rebelling former slaves, in the course of which two were
found guilty and hung from the gallows.19
MacGahan witnessed the aftermath. “I saw their dead bodies hanging
from the beams in one of the bazaars,” he writes, “where they remained for
several days. I may mention that many of Kaufmann’s officers strongly
condemned this act, thinking that the Persians had too much reason for
taking some vengeance on the masters, to be thus severely treated.
The punishment, however, had the double effect of cowing the Persians
and of encouraging the masters to punish them severely for the use they
had made of their liberty. Several poor fellows came into our camp and
showed us gashes in the soles of their feet or in the calves of their legs, in
which was strewed cut horsehair.”20 This evidently quelled the uprising for
a time, but it was not long before the unrest was resumed with renewed
ferocity.21

16
I am grateful to Alexander Morrison for this observation.
17
MS IVAN Uz No. 12581, f. 24a. Hassan Muhammad Amı̄n Oghlı̈ claims that the leader of
˙ ˙
the slave uprising was a man who was called Ibrahı̄m Basmachi, known as the first
lithographic publisher (using stone plates: tash matbʿaachilik) in Khwarazm. For his
˙
efforts in the uprising, he was allegedly dubbed “Ibrahı̄m Sultān”).
18 ˙
Zheliabuzhskii, Ocherki i zavoevanie khivy, 116; Schmidt, The Russian Expedition to
Khiva in 1873, 121; MacGahan, Campaigning on the Oxus, 310.
19
MS IVAN Uz No. 12581, ff. 24a–25b. 20 MacGahan, Campaigning on the Oxus, 311.
21
MS IVAN Uz No. 12581, ff. 24b–25a.
188 Conquest of Khiva and Myth of Russian Abolitionism

Kaufman’s decision to demand the liberation of the slaves was directly

inspired by the uprising undertaken by the slaves themselves. Granted,
some Russian officers had supported individual acts of manumission on
an ad hoc basis during the campaign: those slaves who had escaped and
joined themselves to General Verevkin’s ranks, for example, were wel-
comed by that officer, who declared them to be free. In early June,
Verevkin corresponded both with officials in Astrakhan and with
Kaufman concerning the liberation of slaves, and the latter replied that
he would propose the subject of general abolition to the new council that
he had formed among Khivan elites to negotiate the surrender and future
governance of Khwarazm. As Schuyler notes, however, echoing the senti-
ments of both Central Asian and foreign commentators, “the most impel-
ling reason for the emancipation was, that slaves who had run away from
their masters had begun to rob, pillage, and murder; and the masters,
fearing to be deprived of their slaves, were imprisoning and torturing
those who remained to prevent them from running away.”22 So it was
that, shortly after Muhammad Rahim Khan returned to the conquered
˙ ˙
capital from his temporary hiding place among the Yomuts, he was
compelled by Kaufman to draft an edict of abolition and general manu-
mission, to be circulated throughout the Khanate.23 Khwarazm’s eman-
cipation proclamation read:24

22
Schuyler, Turkistan, 353; see also MS IVAN Uz No. 12581, f. 25a; MacGahan,
Campaigning on the Oxus, 311.
23
Kaufman, reflecting on these events in an interview with MacGahan, offered a mocking
appraisal of the Khan’s political abilities and intelligence, which, ironically, may actually
reveal the naiveté of the general himself: “Having decided to emancipate the slaves,
[Kaufman] wrote the Khan a letter one day, informing him of his decision, and requesting
him to issue a proclamation to that effect. The last part of the letter contained advice and
counsel as to the best means of carrying out the measure, and among other things,
requested the Khan to make arrangements with the governors of the different provinces
to have the proclamation read all over the Khanate the same day, in order not to give the
Uzbegs an opportunity for maltreating the Persians. The Khan, however, having read the
first part of the letter, immediately, without stopping to finish it, wrote out a proclamation,
and ordered it to be proclaimed through the streets next day by a herald, and then went to
Kaufmann, with childish eagerness, to tell him what he had done, and show him how
prompt he was to obey his wishes. ‘But,’ said Kaufmann, ‘did you not read the last part of
my letter ?’ ‘No,’ said the Khan, ‘I did not know it was necessary.’ ‘Why, yes,’ said
Kaufmann, ‘with us the last part of a letter is often the best. In it I advised you not to
issue your proclamation for a few days yet’” (ibid., 278–279). It is much more likely that
the Khan’s “mistake” was a deliberate and none-too-subtle attempt at resistance, intended
to undermine Kaufman’s efforts.
24
Zheliabuzhskii, Ocherki i zavoevanie khivy, 117; see also Trotskii, Opisanie khivinskago
pokhoda 1873 goda, 246.
 Conquest of Khiva and Myth of Russian Abolitionism 189

I, Sayyid Muhammad Rahim Bahadur Khan, out of deep respect for the Russian
Emperor, command all of my subjects promptly to provide total freedom [pol’naia
svoboda] to all the slaves of my Khanate. Henceforth slavery is to be abolished in
my Khanate for all time. May this humanitarian act [chelovekoliubivoe delo] serve
as a pledge of eternal friendship between all of my glorious people and the great
Russian people.
I command that my will be carried out precisely, under threat of strict punish-
ment. All former slaves, henceforth free, must be considered to have rights equal to
those of my other subjects, and subject to the same penalties as they are, and tried
for disturbing the peace in the country and for disorderliness – thus I urge all of
them to maintain order.
All former slaves are granted the right to live anywhere in my Khanate, or to
leave it for wherever they desire. For those who wish to leave the Khanate, special
measures will be taken. Female slaves [zhenshini-rabyni] are freed on equal footing
with men. Disputes between married women and their husbands will be sorted out
according to sharia law [po sharigatu].

This edict was to be followed, several weeks later, in mid-August, by

a treaty between the Khan and the Russian Empire which included the
provision that all slaves were officially freed, and that reducing people to
slavery and trading in human beings were absolutely prohibited.25
The June decree was announced publicly in towns and villages through-
out Khwarazm.26 Soon, slaves could be seen celebrating all across the
Khanate, shouting out joyfully in the streets. Freedmen were witnessed
affixing Russian imperial emblems and colors to their clothing in a show of
gratitude.27 In the weeks to come, these freedmen – whose numbers have
been estimated at anywhere between 27,000 and 58,500 – would have to
choose whether to remain in Khwarazm or to attempt to return to their
original homeland. For nearly all of them, that homeland was Iran, and the
journey there would involve crossing dangerous ground. There was no
guarantee that homecoming parties venturing south would be safe from
Turkmen raids. Indeed, those who were most eager to return home in the
days that followed the news of their liberation met a miserable end:
“[F]ollowing the announcement of freedom,” Alikhanov-Avarskii writes,
25
MS IVAN Uz No. 12581, ff. 26a–b; see also MacGahan, Campaigning on the Oxus, 419;
S. V. Zhukovskii, Snosheniia Rossii s Bukharoi i Khivoi za poslednee trekhsotletie
(Petrograd, 1915), 179–83; Das Staatsarchiv: Sammlung der officiellen Actenstücke zur
Geschichte der Gegenwart (Leipzig: von Duncker & Humbolt), 134–137. The Russian text
of the decree has rabstvo i torg liud’mi for “slavery and trading in human beings.”
26
MS IVAN Uz No. 12581, f. 25b; MacGahan, Campaigning on the Oxus, 311.
27
MS IVAN Uz No. 12581, f. 25b. Hassan Muhammad Amı̄n Oghlı̈ writes that even some
˙ ˙
famous Khivan personages – among whom he lists Murād ʿAlı̄ Mahram and Muhammad
˙
ʿAlı̄ Mahram – did the same.
190 Conquest of Khiva and Myth of Russian Abolitionism

“Persians began to gather into large parties for the return to their home-
land. Two such parties, of 700–800 souls each, have already left the oasis
and moved into the desert where, they say, they were slaughtered without
exception by Turkmens fleeing Khiva . . . In light of such rumors, the
unfortunate Persians no longer hurry home, but instead collect near our
detachment – their numbers already adding to some 1,500 men, women,
and children – in order to follow along the Caspian shore during the return
march of the Caucasians . . . ”28
Although those who wished to remain in Khwarazm rather than risking
a journey elsewhere were guaranteed equal rights with other freedmen
under the provisions of the liberation decree, Emil Schmidt claims that, at
first, few showed any desire to stay where they were.29 The author of the
Āzādnāma, Hassan Muhammad Amı̄n Oghlı̈, provides quite a different
˙ ˙
assessment, however, estimating that a substantial proportion of the freed-
men, totaling no less than 22,500 Iranians and “Azeri Turks,” wished to
become Khwarazmian citizens. By this author’s estimation, that left some
36,000 who were prepared to journey to Iran and elsewhere.
Grodekov recalled seeing the fourth group of freedmen to leave
Khwarazm following behind Russian troops on the way to the Mangyshlaq.
Each was given a daily allowance of twenty kopeks, though “it was noted that
they were not making good use of the allotment given to them, and instead of
stocking up on provisions for the voyage through the desert, they bought
only some goodies which were made available to them in Kungrad . . . Little
benefit was had from the Persians in the armed forces. They were for the most
part weak and sick, depleted by being overworked by masters who, for the
slightest offense, laid shackles upon them and punished them cruelly. There
were, of course, healthy and strong individuals among the Persians, but these
too did nothing without prodding. They were sometimes caught stealing.”30
It is not clear why Grodekov would have expected military or tactical
assistance from these refugees, but his remarks help to reveal something of
their poor condition during the long march home. This group would suffer
still more on the road. They – as well as their Russian escorts – were
tormented by thirst on their journey through the desert. Grodekov recalls
that “[o]n coming to a well, six Persians climbed down into it and did not

28
Alikhanov-Avarskii, Pokhod v khivu, 280. Schmidt observed that “[t]hose who returned
alone without availing themselves of the Russian convoy and aid have had much to suffer,
and many of these unprotected helpless creatures were murdered by the Turkomans on the
road.” (Schmidt, The Russian Expedition to Khiva in 1873.)
29
Schmidt, The Russian Expedition to Khiva in 1873, 122.
30
Grodekov, Khivinskii pokhod 1873 goda, 326.
 Conquest of Khiva and Myth of Russian Abolitionism 191

want to leave, even when ropes were lowered to bring them back up. One of
the ropes broke while hauling a Persian up, but he survived.”31
When they arrived at Krasnovodsk, on the Caspian Sea, these Iranians
were offered provisions totaling two pounds of flour and one-eighth of
a pound of butter per day for each adult, and half this amount for each
child. Grodekov recalls that a number of them expressed a desire to
accompany the detachment all the way back to Russia. “But,” he writes,
“when the first steamship arrived to take the Persians back to their home-
land, all of them abandoned these intentions, some from a desire to see
their fatherland and others for fear of being taken into the army as soldiers,
as was claimed by a rumor someone spread among them. Although the
officers tried to explain to them that the White Tsar had no need of the
services of such people, dozens of whom could be sold to Khiva or Bukhara
by a single Turkmen, they were thoroughly convinced by the [rumor] they
had heard, and not one of them went off to Russia.”32 MacGahan confirms
that this group arrived safely at their destination.33 Another group of some
500 freedmen, however, which had set off at the same time but had opted
to travel along the Atrek River, was pillaged and massacred by Teke
Turkmens.34
Among the thousands of freedmen who had gathered in certain
Khwarazmian villages and bazaars to await their journey home, there
were many who could not immediately afford the expenses required for
the long journey. According to Hassan Muhammad Amı̄n Oghlı̈, these ones
˙ ˙
could expect aid neither from their former masters nor from Iran (nor,
presumably, from the Russian military). Still, they hoped to manage the
return journey eventually, and as time wore on their patience grew thin.
In the villages of Katta Bāgh and Tāshhawż, some of them began taking out
˙
their frustration by attacking their former masters, while others resorted to
banditry and other forms of aggression.35
The Āzādnāma recounts the tragic events that followed. At this tense
time, a man named ʿAwz Keldi Khan from among the Yomuts came to the
freedmen in Katta Bāgh, telling them that he could lead them, safe and
sound, through the Turkmen deserts and back to their homeland in Iran.
The former slaves made a show of gratitude for his offer, and some went to

31
Grodekov, Khivinskii pokhod 1873 goda, 330.
32
Grodekov, Khivinskii pokhod 1873 goda, 342.
33
MacGahan, Campaigning on the Oxus, 311–312.
34
MS IVAN Uz No. 12581, f. 27b. This may be the group to which Schuyler refers on p. 354
of his Turkistan.
35
MS IVAN Uz No. 12581, ff. 27b–28a.
192 Conquest of Khiva and Myth of Russian Abolitionism

their comrades in Tāshhawż in order to spread the news. Many agreed to

˙
follow him. Several days were spent working out the preparations.
The plans gained the favorable attention of the local headman, ʿAlı̄
Muhammad Khan, and on his orders a large group departed for Iran via
˙
Kohne Urgench, under the leadership of ʿAwz Keldi Khan. All the while,
the Āzādnāma relates, this self-proclaimed guide had been surveying the
freedmen’s worldly possessions and, in a clandestine manner (yashirin
ravishda), had passed along the details of his plan to his fellow Yomuts,
telling them that the former slaves would have many possessions among
them (qullarning qolida māl-i dunyā köbdur).36
With this information in hand, a massive host of some 17,000 armed
Yomut horsemen gathered while ʿAwz Keldi Khan led a trusting cohort of
several thousand former slaves into the desert. First, they stopped in
Kohne Urgench, where they stocked up on supplies and provisions for
the arduous road to come. For several days they traveled deeper into the
desert, their water supply gradually depleting. Finally, they came upon
a well and stopped to drink. It was here that, all of a sudden, they found
themselves being fired upon. The 17,000 mounted Turkmens emerged and
engaged them in a grueling siege, which lasted the next twelve hours.
At the end of it, some 5,000 former slaves lay dead. All of their possessions
were pillaged.37
Schmidt also reports on the misery that awaited the former slaves of
Khwarazm following the departure of the Russian troops, although he
gives a lower casualty estimate. “No sooner had the last Russian troops
turned their back upon the Khanate,” Schmidt writes, “than the news
reached Petro-Alexandrovsk that the Turkomans had attacked
a numerous party of former slaves (about 1,700 in number) near Kunya-
Urgenj, who were on their way home; of these wretched people
a considerable number had been sabred on the spot and the rest captured
by the Turkomans.”38 A courier on his way from Petro-Alexandrovsk
reported seeing “the bodies of hundreds of Persians who had been mas-
sacred after the Russians left the country”; another messenger, headed to
Mangyshlaq, likewise reported “hundreds of corpses” of Iranian former-
slaves by the roadside.39
These acts of cruelty on the part of the Turkmens were unprecedented:
Central Asian tribes had never before been known to slaughter Iranians en

36
MS IVAN Uz No. 12581, ff. 28a–29b. 37 MS IVAN Uz No. 12581, ff. 30a–31a.
38
Schmidt, The Russian Expedition to Khiva in 1873, 169.
39
Schuyler, Turkistan, Vol. 2, 354.
 Conquest of Khiva and Myth of Russian Abolitionism 193

masse, a strategy that, among other things, would have undermined the
business of slave-dealing. For all the reports of dehumanizing treatment
inflicted upon slaves while in Turkmen captivity, reports of outright mur-
der before 1873 are strikingly rare, and can usually be explained by
rational means: Those captives who were too slow or ill to manage the
journey north from Khurasan might be killed, for example, or those who
fought back. But it is difficult to ascertain any practical reason for the mass
murder of thousands of former slaves. Perhaps some wanted to send the
message that abolition for the sake of the slaves themselves would only
lead to further misery – a desperate attempt to reinforce the market for
slaves that abolition undermined. Some may have hoped to send a message
to the slaves themselves that nothing good would come of fleeing their
masters, perhaps in the hope of “stabilizing” the slave system. But these
practical-minded explanations are ultimately unconvincing.
It is more appropriate to reflect on these massacres both as acts of
resistance and as expressions of collective desperation. In the weeks after
the conquest of Khiva, Kaufman had wasted little time in proving his
cruelty to the Turkmens. Soon after the return of the khan to the capital,
Kaufman imposed upon the Yomuts a war indemnity totaling the impos-
sible sum of 600,000 rubles, of which half was demanded to be paid by
the Yomuts of Khwarazm within two weeks.40 The Yomut elders did not
resist, despite the absurdity of the demand in light of their meager assets,
which mostly consisted of livestock: They agreed to pay, but asked for
more time to collect the funds. Instead, Kaufman arrested twelve elders,
holding seven of them as hostages. The general did not wait even the
allotted two weeks before sending a detachment to the Yomuts in western
Khwarazm, ordering them to “proceed to the work of slaughter” if there
was no evidence that the required indemnity was forthcoming.41
The Cossacks massacred the Yomuts, who were not expecting the assault
and were not prepared to fight back. The other Turkmen tribes of
Khwarazm were then ordered to pay the remaining 310,500 rubles of
the fine. They could raise only 92,000 rubles among them, but
Kaufman – already facing domestic criticism for the slaughter of the
Yomuts – softened his approach, taking twenty-six Turkmen hostages

40
Zheliabuzhskii, Ocherki i zavoevanie khivy, 119; Becker, Russia’s Protectorates in Central
Asia, 59.
41
Gali Oda Tealakh, “The Russian Advance in Central Asia and the British Response,
1834–1884” (Ph.D. Dissertation, Durham University, 1991), 209. A witness claims to
have heard the leader of this detachment, Golovachev, ordering his men to “kill all of
them” (i.e. the Yomuts), without respect to sex or age (Schuyler, Turkistan, Vol. 2, 359).
194 Conquest of Khiva and Myth of Russian Abolitionism

but sparing their lives.42 All the Turkmens of the region, whether subjects
of Russia or self-independent tribes, must have been horrified by these
events, which appear calculated to ensure – rather than to prevent –
ongoing conflict with the Tsar’s army.
After these acts of terrorism on the part of the Russian military,
allowing the safe passage of the freed slaves through Turkmen country
would have been a tacit admission of Turkmen weakness, or even of
Turkmen subjection to Russia. Their territories had never been a safe
space for Iranians before; if that should change simply because Russian
officers gave Iranians permission to return home, it would be as if the
Russians – or still worse, the Iranians themselves – had dictated the terms
by which one could pass through the Turkmen-dominated deserts.
The Turkmens could, at the very least, assert their dominion over the
shrinking territory in which they could still move and act freely.
Inasmuch as their massacre of the freedmen served as a desperate asser-
tion of their independence from the forces closing in from the north, we
must observe a terrible irony: The imperial drive which had freed these
slaves had likewise sealed their fate.
The Turkmens now had to face an uncertain future. Raiding and slave-
dealing had been a major economic force in Turkmen communities for
which viable alternatives for securing a livelihood were few. Largely alie-
nated from the agricultural lands of the oases, which were mostly owned
by Uzbeks (in Khwarazm and Bukhara) and Iranians (in northern Iran),
Turkmen tribes took advantage of every available means of securing food
and resources, engaging in fishing, hunting, herding, and artisanship.
Raiding and slave-dealing, however, not only provided for influxes of
capital and trade-goods on a grander scale than these other occupations,
but also served as the foundation for valued customs and social phenom-
ena. The raids provided the supreme template for the proving of masculi-
nity: A small-scale herder or fisherman could moonlight as a warrior.
Young men, sharing the experience with their elders, could come of age
in combat. The world-renowned horse culture of the Turkmens had also
become closely linked to the raiding expeditions, and it was in these
contexts that the horsemanship and gallantry praised in the Turkmens’
vast oral literature, music, and poetry found its most vivid contemporary
expression. The raids also served as a catalyst for the redistribution of
goods, and the social functions that this engenders: the enactment of
hierarchy, the proving of generosity, and the apportioning of prestige.

42
Becker, Russia’s Protectorates in Central Asia, 58.
 The Russian Conquest of the Turkmen Deserts 195

The wealth generated by raids and slave-trading also provided for the
creation of trade networks. The luxury goods and animals looted in raids,
as well as the captives and the revenue generated from selling them,
allowed for liaisons with neighboring Turkmens as well as with Kazakhs,
Uzbeks, and others. The importance, moreover, of the goods, materials,
animals, enslaved laborers, and luxury wares gained by raiding can hardly
be overstated in light of the poverty and chronic instability that character-
ized tribal life in the deserts in Central Asia. In an environment where
herdsmen could lose the majority of their livelihood in a single season of
brutal weather, the small measure of stability provided by supplemental
revenue must have been precious indeed.
In light of all this, one can begin to comprehend the desperation and
outrage that must have gripped many Turkmen communities as they
realized that the markets and infrastructure for their trade in slaves was
collapsing with the Russian advance. What’s more, they could see the very
cargo they had gone to such lengths to secure literally walking back to
freedom before their eyes. And for those who believed that slavery among
Sunnis was a charitable fate for Shiʿites, as it often brought about the
auspicious occasion of conversion, the slaves’ return to Iran must have
been all the more galling. The Turkmens’ work was undone, and their own
futures were uncertain. At the very least, they could endeavor to steal what
little goods the freedmen carried with them on the way back home. Having
dehumanized their Iranian victims for decades already, it is no great
surprise that the Turkmens’ anger should have found its outlet in astonish-
ing acts of inhumanity.

THE RUSSIAN CONQUEST OF THE TURKMEN DESERTS

The conquest of Khiva is sometimes portrayed as the campaign that ended

the slave trade in Central Asia. It was nothing of the sort. The slave trade
continued, as we have seen, bolstered mostly by the ongoing tensions to
the south between Turkmen tribes and the Iranians. Nor was it intended to
be anything of the sort: Russian operations against the Turkmens in the
1850s and 1860s, more often than not, were justified as a means of
ensuring the safety only of Russian citizens and soldiers; and the establish-
ment of Russian military outposts on the Caspian Sea must be understood
in light of the larger imperial project of establishing a permanent presence
south of Khwarazm. Before the 1870s and 1880s, however, this presence
was peripheral. As we have seen, the activities of the Russians’ Caspian
196 Conquest of Khiva and Myth of Russian Abolitionism

bases, in consultation with its Astarabad consulate, had some impact in

reducing the incidence of raids on Iranian towns and villages in the region,
and the establishment of patronage networks and alliances between the
Russian military and Turkmen elites was a profound and novel develop-
ment. But the slave trade was also reduced by many factors that had little
or nothing to do with Russian interventions.
In the eighteenth and nineteenth centuries, the geography of slave-
raiding among the Turkmens was generally distinguished by several
spheres: The Salar Turkmens raided from their base at Sarakhs and took
captives in their campaigns for the control of that town and its environs;
the Sarı̈qs did the same in vicinity of Marv; the Ersaris took captives in and
around Herat; the Chawdurs conducted raids along the Russian frontier;
the Yomuts raided the Caspian coast and the villages in the vicinity of
Astarabad; and the Tekes, based at Akhal, likewise raided the Iranian
frontier, though some authors claim that their raiding activities were less
frequent than those of other tribes before the 1860s.43 By the time Khiva
fell in 1873, all of these vectors in the slave trade had been interrupted to
some extent, or even erased entirely. The Salars suffered a massive defeat in
ʿAbbās Mı̄rzā’s Sarakhs campaign of 1833, and would never again regain
their former notoriety for taking captives in battle or raiding local settle-
ments. The Chawdurs were reported by a Caspian flotilla commander to
have ceased their raids on Russian subjects completely after the construc-
tion of the Novo-Petrovsk fort on the Mangyshlaq peninsula in 1846.44
Relatively little is known about the Ersaris in the mid- and late nineteenth
century, but there is no evidence that they conducted significant raids in the
vicinity of Herat after the city fell decisively into the hands of the Afghans
after 1863.
As for the Sarı̈qs, their ascendancy – as well as their ability to take
captives – eroded along with their hold on the oasis of Merv. In 1855,
Khwarazmian forces suffered a major defeat near Sarakhs at the hands of
the Tekes, who proceeded to drive the Sarı̈qs from the vicinity of Merv as
they migrated into the region. The Khwarazmians, who had long employed
Turkmens of various tribes in their struggle for Merv, would never retake the
oasis, nor would they have any further need of hiring Turkmens as mercen-
aries (in which capacity they were also often captive-takers); the Sarı̈qs, who
during their troubled residency there had allegedly become one of the

43
Marvin, Marv, the Queen of the World, 179.
44
Russko-turkmenskie otnosheniia v XVIII–XIX vv., 469, doc. no. 341; see also Vambery,
Sketches of Central Asia, 212.
 The Russian Conquest of the Turkmen Deserts 197

principal Turkmen tribes engaged in slave-raiding,45 abandoned Merv

entirely by 1858. “Thus,” William Wood writes, “both of the major players
at Merv in the first half of the nineteenth century suddenly found themselves
on the outside looking in.”46 Meanwhile, from 1855 to 1867, Khwarazm
was rocked by uprisings among Yomuts and other Turkmen groups within
the Khanate, undermining whatever slave trade might have persisted
between those groups and Khwarazmian markets.47
The Tekes consolidated their control over Merv and held their ground
against Qajar campaigns; they would maintain that control until 1884,
when the Russians occupied the city. In the intervening years, they took
advantage of the persistent demand for slaves in the markets to the north
to become the preeminent slave dealers of the region. In 1859, they
formed an alliance with the Salars and Sarı̈qs, and dominated a consider-
able expanse of territory under two simultaneous rulers (Kushı̄d Khan in
Merv and Nūr Verdi Khan in Akhal). But the market for their slave
trading was curtailed by the fall of Khiva. “The march to Khiva,” writes
Grodekov, “and its results – the suppression of slavery in Khiva and
Bokhara – struck a heavy blow at the ascendency of the Tekes.
It became useless for them to steal the Persians for slaves, since they
had no market for the disposal of them. For two years after the fall of
Khiva the Alamans entirely ceased; and although they recommenced
again in 1876, they have never yet reached the proportions they attained
before the closing of the slave markets.”48
In the cases concerning the decline of slave-raiding among Turkmen
groups, Russian influence was either nonexistent, as in the fall of the Salars
at Sarakhs, the Sarı̈qs at Merv, and the ascendancy of the Afghans in Herat,
or indirect, as in the interruption of markets (i.e. in Khwarazm) or access to
victims (i.e. along the Russian frontier). In the case of the Yomuts along
the shores of the Caspian, however, we find a more direct engagement,
involving – as we have seen – skirmishes, intelligence-gathering opera-
tions, the hiring of native informants, and other tactics of war. (It is,
naturally, for this reason that the decline of raiding among the Yomuts is
better-documented than among any other Central Asian population except
perhaps the Kazakhs.) While some Yomut elites chose to serve – or at least
manipulate – the Russians as informants and ostensible allies, a great many
others responded to Russian encroachments with energetic and sustained
45
Marvin, Marv, the Queen of the World, 179.
46
Wood, “The Sariq Turkmens of Merv and the Khanate of Khiva in the Early Nineteenth
Century,” 253.
47
See Bregel, Khorezmskie turkmeny, 197–228. 48 Grodekov, 132–133.
198 Conquest of Khiva and Myth of Russian Abolitionism

resistance. Caught between the forces of the tsar and those of the shah,
many directed their opposition toward both, and submitted to neither.
The violence was generally reciprocal. In 1850, in response to increas-
ing Russian influence and control in their territory, a group of Yomuts
launched an abortive campaign against the tsar’s Ashur Ada sea-station.
They arrived on Easter night in forty large boats. “Leaping ashore,”
Gunaropulo writes, “the Turkmens rushed to the brightly-lit church,
being the only place clearly visible against the background of dark night.
The people, praying in the garden adjacent to the church, saw the armed
Turkmens approaching and, with a desperate cry, rushed into the church.
At that moment the Turkmens heard gunfire, as the guard team had quickly
assembled, and they fled, leaving four of their own dead but still taking
three women with them.”49 According to Gunaropulo, the next few years
saw a decline of Russian influence over the Yomuts. In 1854, however,
station officials resolved upon a violent reassertion of Russian force: They
made a landing assault on the Yomut village of Hassan-Quli, burning
twenty-eight boats and taking several prominent men as captives (“for
whom the Turkmens paid dearly in ransom”).50
Yomut relations with the Qajars were still more combustible. In 1858,
after the Qajar governor of Astarabad, Jaʿfar Qulı̄ Khan Ilkhan, forcibly
expelled Turkmens from the vicinity of Kara Kala with a force of some
10,000 men, taking some back to Astarabad as captives, the Turkmens
began plundering the countryside and taking captives of their own in
reprisal.51 The raids continued for the next three years, undoubtedly
both as a means for the Yomuts to generate revenue as well as a means of
asserting that they could not be evicted from their homes without reper-
cussions. In September, 1863, while likely still reeling from their disastrous
defeat by the Tekes at Merv two years earlier, the Qajars led an ambitious
series of attacks against the Yomuts of Astarabad and the Atrek. This
campaign too came with significant losses: At first their advance was
beaten back, and one Qajar detachment had to retreat from Astarabad
after being stricken with a typhoid outbreak that killed up to twenty
soldiers per day. But the military governor of Bujnurd, Haydar Qulı̄
˙
Khan, led a successful raid on the upper Atrek. Sixty-three Yomut captives
were taken, along with 3,000 camels and 20,000 sheep. Fifteen Turkmens
who were killed during the attack were beheaded, and their severed heads
49
Gunaropulo, “V Turkmenskoi stepi,” 1043.
50
ibid. “Since that time,” Gunaropulo writes, “the pirates have been subdued, and have never
since dared to attack the station.”
51
Russko-turkmenskie otnosheniia v XVIII–XIX vv., 453–54, doc. no. 328.
 The Russian Conquest of the Turkmen Deserts 199

sent to Haydar Qulı̄ Khan’s camp. This “success” inspired the Qajar
˙
minister of war to demand that the coastal Turkmens return all of the
Iranian captives they had taken in recent years. Turkmen elders readily
agreed to this, though naturally many of these captives would already have
been sold into slavery in Khwarazm or Bukhara.52 By the next year, the
Yomuts had evidently given up just thirty captives, and in November of
1864 they suffered another attack at the hands of the Qajars. An Iranian
detachment ventured along the Atrek River, capturing fifteen elderly
Turkmen women, as well as some 1,000 camels and 12,000 sheep.
According to Russian internal documents, the Qajars were demanding
not only the surrender of captives, but also that the Turkmens submit to
becoming Iranian citizens.53 In fact, given the clear impossibility of meet-
ing the Qajar government’s demands – that is, of surrendering captives
who had already been sold and, in some cases, may already have been
dwelling as far away as the Kazakh steppe – we may conclude that these
demands were little more than a pretext for the ongoing efforts to sub-
jugate the Turkmens.
A Turkmen uprising against the brutal Astarabad governor came in
1867, followed by a Qajar counter-assault, which pushed many Yomuts
north of the Atrek. Turkmens who were left behind – including women and
children – suffered torture at the hands of the Iranians and were mas-
sacred. Some who had fled returned home to become Qajar subjects.54
While north of the Atrek, the fleeing Yomuts seem to have been careful not
to draw the Russians into the conflict. But it was during this same period, in
the mid-1860s, that we find a distinct shift in Russia’s diplomatic concep-
tion of the Turkmens: We find in Tsarist political discourse a heightened
preoccupation with putting a stop to their “predatory practices.” These
“practices” among the Turkmens, sporadic and ongoing as they were, had
not substantially increased in the mid-1860s. So what had changed?
As Mehmet Saray observes, the shift in diplomatic language corresponded
precisely with the dawn of Russia’s conquest of the region; it served as
a rationale and provided an ideological basis for the conquests of Turkmen
country that would follow in the coming two decades.55
With Russia’s recognition in 1869 of the Atrek River as the northern
boundary of Iran, the Turkmens occupying the borderlands found them-
selves facing both new challenges and new possibilities for resistance.

52
Russko-turkmenskie otnosheniia v XVIII–XIX vv., 503–504, doc. no. 373.
53
Russko-turkmenskie otnosheniia v XVIII–XIX vv., 505–506, doc. no. 374.
54
Saray, “Russo-Turkmen Relations up to 1874,” 26. 55 ibid., 27.
200 Conquest of Khiva and Myth of Russian Abolitionism

V. I. Markozov, commander of the newly constructed Krasnovodsk garri-

son, recognized the inorganic nature of the new border and the curious
manner in which it divided Turkmen territory:56
Man’s nature and way of life are completely identical on the right and left banks of
the Atrek. And both here and there . . . live the same Yomut Turkmens, who, as
mentioned, migrate now on one side of the river and now on the other, depending
on the time of year . . . Thus, the Atrek does not constitute an ethnographic
boundary. As a habitat boundary, the Atrek differs little from the Kara-Suu, and
still less from the Gurgen, and in any case it has not the integrity of the Danube or
anything of that sort because it is everywhere easy to ford, except, of course, in
periods of flooding. Moreover, Persia never had any sort of rights to the strip of
land between the Atrek and the Kara-Suu, because she never genuinely possessed it.
In all that expanse, one could see Persians only in the form of slaves among the
Turkmen population. There were almost no tents whose owners did not have at
least one Persian, held in shackles.

The commander’s chief concern, however, was not to preserve the territor-
ial integrity of the Turkmens, but to patrol and subjugate them more easily.57
He notes that his men had often engaged in shootouts with Turkmens who
lived for part of the year on one side of the border and part of the year –
beyond Russian jurisdiction – on the other. The inconvenience of policing
such a population, he implies, likewise had negative consequences for the
Iranians themselves, whose citizens the Russian troops sometimes had the
occasion to liberate. “Consequently,” he writes, “when the Krasnovodsk
detachment approached the Turkmen predators’ nest [priblizilsia k gnezdu
turkmenskikh khishchnikov] and began to subdue their thieving impulses
and to cross the Atrek, [the soldiers] many times liberated Persians who had
been languishing in pits on iron chains, and sent them off to Astarabad
through the medium of our Astarabad consulate or the Ashur Ada sea

56
Markozov, Krasnovodskii otriad, 24.
57
The appearance of the Krasnovodsk garrison on Turkmen territory (the land where it stood
was called Kizil-Suu by the Turkmens) produced multiple responses among the Yomuts and
Tekes. The latter opposed it unreservedly. The Yomuts, however, were internally divided.
The Atabay branch of the Yomuts, living to the north and south of the Atrek, opposed
Russian occupation in any form. The Jafarbay branch, who fished the Caspian and traded
with the Russians, showed no opposition to it whatsoever. The new Russian incursions
were certainly the inspiration for a new wave of sporadic attacks, including an abortive
assault on a Russian encampment at Mikhailovsk Bay in October of 1870. Meanwhile, the
Tekes reportedly cautioned Yomuts of the region to avoid any sort of partnership with the
Russians. The Yomuts quickly learned, however, that resisting Russian demands could be
futile: In 1872, when Markozov’s exorbitant demand for 3,500 camels for his troops was
rebuffed by the Yomuts, the camels were simply taken by force. See Saray, “Russo-Turkmen
Relations up to 1874,” 29–33.
 The Russian Conquest of the Turkmen Deserts 201

station, which itself owes its existence to the need to protect Persia from
seaward incursions by the Turkmen, for which ends it was taken over by
Russia in the Turkmenchay treaty.”58 Markozov’s suggestion that the sea
station was created mostly to protect Iranian citizens is dishonest, even
preposterous, but his evidently sincere exasperation with the challenges
posed by the Atrek border is revealing.59
The irony of the fact that Russian and Qajar missions of “pacification”
were themselves significant motivators of Turkmen violence and resistance –
and thus of ongoing captive-taking – was surely lost on Markozov.
An awareness of this cause-and-effect is absent among our Russian eye-
witnesses in general, for whom Turkmen raids seem to have been nothing
more than the reflexive and age-old “custom” of a barbarous people. In any
case, Markozov was not alone in complaining that the militant subjugation
of the Turkmens along the Russo-Iranian frontier had become, by the late
1860s, a disproportionately Russian endeavor. And he was surely correct in
observing that this endeavor was complicated by the placement of the
border itself. Regarding the Turkmens living south of the Atrek, Grodekov
wondered with good reason whether they could really be considered Qajar
subjects: “‘How are we Persian subjects,’ the Turkmens say, ‘when we carry
their people into captivity, and they ours?’”60
The Turkmens along the Atrek meanwhile found ingenious ways to take
advantage of the curious division of their territory. Their continued raids
on their “fellow” Qajar subjects, combined with the Qajars’ inability to
effectively police the territory between Astarabad and the Atrek, appar-
ently induced a kind of agreement whereby the Qajars would cease their
punitive assaults entirely if the Turkmens would restrict their raiding
forays to the lands north of the river. The appeal of this alleged agreement
would have rested, for the Turkmens, on the presumption that the Russians
would not cross the river into Iranian territory to avenge or take back their

58
“Russo-Turkmen Relations up to 1874,” 24.
59
It is worth noting that the evident incentive behind Markozov’s memoir was his effort to
exculpate himself for his own failure to reach Khiva with his detachment in 1873; and that
one of the reasons for that failure was apparently the refusal of local Turkmens to supply
his troops with camels. I am grateful to Alexander Morrison for these observations.
60
Grodekov, Khivinskii pokhod 1873 goda, 59. Nevertheless, the Shah had been laying claim
to the Turkmens of the Atrek region and the Caspian coast for many years by this time, and
by the eve of the 1869 border agreement the Qajars were still claiming dominion as far
north as Krasnovodsk. A. M. Gorchakov, the Russian foreign minister and imperial
chancellor, scoffed that Iran “had no better claim to the country of the Turkmens than
the King of Italy to the throne of Jerusalem” (Saray, “Russo-Turkmen Relations up to
1874,” 28–29).
202 Conquest of Khiva and Myth of Russian Abolitionism

own subjects. Indeed, protocols were in place that would make such forays
inconvenient: The Russian embassy in Tehran and the consulate in
Astarabad had to be informed of any Russian missions south of the
Atrek, and the Qajars themselves would likely object to them.61 Thus an
appealing tactic arose for the resistant Turkmens of the borderland: They
could use Iran as a safe harbor while continuing to undermine the paral-
yzed Russian troops to the north or, in the event that the Russians actually
ventured south of the Atrek, they could potentially benefit from the
resulting tensions as their two adversaries, the Qajars and Russians,
squabbled over border violations and the details of who should risk their
troops fighting the Turkmens. This tactic, described by both Grodekov and
Markozov, could also involve an additional, incidental benefit for the
Qajars: the inevitable injuries that the Russians might inflict upon the
Turkmens who attacked them.
In July of 1871, Markozov read a letter from the chief of the Ashur Ada
marine station giving notice of a joint attack on Russian troops by Tekes,
“Khivans,” and Atabays (a branch of the Yomuts). These Atabays were
reportedly supported by the Qajar governor of Astarabad, who hoped to
distract them from conducting raids in the Iranian provinces and, inciden-
tally, to ensure “that they would suffer greatly by making an attack on
[Russian] troops.”62 For their part, the Turkmens, from Grodekov’s perspec-
tive, seemed to be attempting to incite Markozov’s forces into crossing the
Atrek, as Turkmen raiding parties repeatedly stole regimental camels and
occasionally captured soldiers sent out into the desert to gather fuel.
Meanwhile, some Turkmen elders complained to Markozov that people in
their community were continually finding their camels stolen and mutilated –
they suspected this was meant as punishment by other Turkmens who were
understandably upset by their devotion to the Russians. The Russian con-
sulate at Astarabad urged the town’s Qajar governor to send border guards to
the Atrek shores, but, unsurprisingly, the governor declined: Any loss suf-
fered by Russians and their allies was a Russian – not Qajar – problem, and if
the Russians wished to “pacify” Turkmens along the Atrek, it would hence-
forth be their own initiative. The Russians, led by Markozov, campaigned
southward to the Atrek in February, 1872, killing several Turkmens.63
Notwithstanding incidents like this, the majority of Yomuts appear to
have resolved upon nonviolent relations with the fervently militant

61
Saray, “Russo-Turkmen Relations up to 1874,” 59.
62
Markozov, Krasnovodskii otriad, 26.
63
Grodekov, Khivinskii pokhod 1873 goda, 60–62.
 The Russian Conquest of the Turkmen Deserts 203

Russians well before the fall of Khiva. Between 1869 and 1873, hostilities
on the part of the Yomuts were relatively rare, and for the most part limited
to Atabay camel-stealing near the Atrek. After the failure of Khwarazmian
Yomut resistance during the Khivan campaign, many Yomuts had chosen to
subject themselves to the Tsar, and the wave of submissions only increased
after Khwarazm’s conquest. As Markozov, on Kaufman’s orders, surveyed
Yomut territories in the western part of the Khanate in the summer of
1873, he and his men were met with no hostility whatsoever; on the
contrary, the Turkmens “not only professed their entire submission but
showed it in deeds.”64
Though peace prevailed, it was not peace that Kaufman was looking for,
but a pretext for further campaigns of conquest. This is the most likely
explanation for his “punitive” massacre of the noncombative Yomuts of
Khwarazm in July of 1873, which doubtless had the effect of teaching
Turkmens in the Khanate and beyond that making peace with the Russians
was a futile exercise. The hostilities were not long in coming: Later that
same month, a group of Yomuts attacked a Russian camp, only to be driven
back. At least 500 – and perhaps as many as 1,300 – Turkmens died in
battle.65 Further Russian raids came in the winter, when Kaufman had
Lieutenant Colonel N. A. Ivanov lead a harsh attack on two Yomut clans
implicated in the slaughter of the freedmen.66 The remaining years of the
1870s would be marked by troubled attempts to affect the staged annexa-
tion of the Turkmen country. In 1877, Kizil Arvat fell to General Lomakin,
though he was soon compelled to retreat.67 In that same year, Chikishlar
on the Caspian coast was established as a base for further campaigns
against the Turkmens, but soon became a beacon for their raids.68
Further Russian campaigns also ended in failure, and the Tekes, undeterred
and evidently unintimidated, showed their resistance with a successful raid
on Krasnovodsk in April of 1879 and further attacks on both Chikishlar
and Krasnovodsk after that. That same year, an ambitious charge by a large
Russian army on the Teke stronghold of Geok Teppe was repulsed.

64
Tealakh, “The Russian Advance in Central Asia and the British Response, 1834–1884,”
207.
65
Grodekov, Khivinskii pokhod, 293–300; MacGahan, Campaigning on the Oxus,
3385–3386; 390; Saray, “Russo-Turkmen Relations up to 1874,” 40. The next day, in
pursuit of the fleeing Turkmens, the Russians torched Turkmen houses and grain reserves,
“and the cavalry, which was in advance, cut down every person, man, woman, and child”
(Schuyler, Turkistan, Vol. 2, 361–362; Saray, “Russo-Turkmen Relations up to 1874,” 40).
66
Becker, Russia’s Protectorates in Central Asia, 64.
67
Kuropatkin, Zavoevanie turkmenii, 96.
68
Teren’tev, Istoriia zavoevaniia srednei azii, Vol. 3, 3–4.
204 Conquest of Khiva and Myth of Russian Abolitionism

Russia’s humiliating inability to put down the hostilities that the Tsar’s
generals had chosen to instigate led to the installation of a new general to
lead the charge against the Turkmens. M. D. Skobelev, a veteran of the
Khiva and Kokand campaigns as well as the Yomut massacre, was installed
and instructed by the Tsar to make an example of the Turkmens; the Tsar
demanded “no retreat from the plan once adopted, no dangerous back-
ward step that might be taken as evidence of our weakness in the eyes of
Europe and Asia, that might encourage our foe.”69 Skobelev eagerly
adopted the language of punishment and vengeance when rallying his
men, but his initial efforts against the Tekes in 1880 ended in failure. He
nevertheless invited the Tekes to submit, making preposterous demands by
way of peace terms. These demands included the payment of a 1,000,000
ruble war indemnity, the full Russian occupation of major Teke towns and
villages, the surrender of many Teke elders and chieftains as Russian
captives, and the release of all remaining Russian captives.70
The Turkmens, unsurprisingly, declined these terms of surrender, and
in January of 1881 Skobelev led a siege of Geok Teppe that would be
remembered as a massacre of astonishing brutality. The number of vic-
tims has been estimated at up to 20,000 Akhal Teke Turkmens.71
Ashgabat was occupied by the Russians two months later, following
which the Tekes were offered – and generally accepted – Russia’s “pro-
tection” within the newly created Trans-Caspian district. The Russians
then pressed onward toward Merv. When the city was surrendered with-
out a battle in 1884, it marked the final stage in the Russian conquest of
Turkmen territories. The legendary independence of the Turkmens –
cherished among the tribes themselves and elegized by foreign observers –
became a thing of the past.

CONCLUSION

To what extent was the eradication of slavery a factor in Russia’s conquest

of Central Asia? Were the Tsar’s troops really the vanguard of abolitionism
that many observers assumed? A closer look at the evidence suggests that
they were not. Time and again, it was the liberation of Russian slaves, and

69
Tealakh, “The Russian Advance in Central Asia and the British Response, 1834–1884,”
220.
70
Grodekov, Voina v turkmenii, Vol. 2, 70–72.
71
Slavomı́r Horák, “The Battle Gökdepe in the Turkmen post-Soviet Historical Discourse,”
Central Asian Survey 34/2 (2015), 153.
 Conclusion 205

not the extermination of slavery in general, that served as a rhetorical

rallying point in the war effort. In the Tsar’s written address to the Russian
people, published on the eve of the Khivan campaign and offering a list of
justifications for it, slavery is, as noted earlier, mentioned no less than four
times – but no mention is made of the Iranian slaves, of which there were
no fewer than 30,000. It was, rather, the Russian slaves alone that were
targeted for liberation. The disastrous Perovskii campaign of 1839 had
likewise been, among other things, a mission to liberate Russian slaves, and
it was exclusively Russian slaves that the khan turned over as a concession
at that time. Once again, on the eve of the Khivan conquest, a different
khan tried the same old tactic, releasing Russian slaves in hopes of pacify-
ing the Tsar. The release of these twenty-one slaves as a concession when
the total slave population of the realm numbered anywhere from 30,000 to
60,000 would seem nearly comical, and could perhaps have been taken as
a sarcastic taunt, had the liberation of all slaves actually been one of the
Empire’s demands. But, after all, there had never been any demand that the
khan liberate non-Russian slaves in the icy diplomatic exchanges leading
up to the conquest.
MacGahan’s claim, moreover, that “the emancipation of the slaves has
always followed the occupation of any place in Central Asia by the
Russians”72 was simply untrue. The Russian conquest of Bukhara in the
late 1860s had resulted in a number of concessions rendered in ambitious
treaties, but an order of general emancipation was not among them.
Kaufman reportedly voiced his displeasure to the Bukharan ruler over
the continuance of the trade, and claimed, in 1870, to be convinced that
it had ceased; but it had not. Schuyler alleges that, two years later,
a Russian official from the ministry of finance offered Kaufman an impas-
sioned report calling his attention to the flourishing of the slave trade, but
that “no notice was taken” of the information provided.73 If in fact this
report produced any effect at all, it must have been a delayed effect: It was
only in the autumn of 1873 – some five years after Bukhara’s conquest –
that, by Russian demand, the universal emancipation of slavery became
a law in the territory. But as we have seen, the enforcement of that law
involved the bizarre concession that the Bukharan Amı̄r could take an
entire decade to enact that emancipation, during which time all slaves
would stay with their present masters. Attempts to escape to freedom were

72
MacGahan, Campaigning on the Oxus, 233.
73
Schuyler, Turkistan, Vol. 2, 310. I have been unable to find a copy of this report or its
contents.
206 Conquest of Khiva and Myth of Russian Abolitionism

punishable by death. In 1874, Kaufman alerted the Amı̄r to evidence of an

ongoing slave trade, but the ruler denied that any such trade existed. It did
indeed, however, and one Russian agent even witnessed a slave market
operating in Bukhara as late as 1878. Nevertheless, Kaufman declined to
press the issue, and in 1876 he instructed an official in Zarafshan okrug “in
the majority of cases to decline interference in the slaves’ affairs.”74 These
may seem like strange instructions coming from the “emancipator” of
Khiva, but they make perfect sense once we abandon false notions of
Russian abolitionism and realize, as Seymour Becker observes, that “slav-
ery was, after all, an internal affair of no vital practical interest to
Russia.”75
By the time Merv fell to the Russians, moreover, captive-taking and
slave-dealing had already been on the decline in the region for decades, for
reasons that usually had little to do with Russian influence. In fact, Russian
presence sometimes increased the incidence of captive-taking.
The ongoing raids perpetrated by Tekes and Yomuts in the 1860s and
1870s were a direct result of the violence inflicted upon them by the
Tsar’s armies and the staged annexation of their territories. The Russian
imperial narrative of these events, employed by officers like Markozov and
Skobelev, reversed the relationship between cause and effect, creating
a cyclical rationale of conquest: Russian officers deliberately provoked
and inflamed Turkmen hostilities, and then observed that further punish-
ment was necessary to “pacify” these tribes.76
Kaufman’s agreement with the Khwarazmian khan to liberate the Iranian
slaves was notable not because it was the final strike against Iranian slavery
in Central Asia, but because it was the first. We must wonder why the slaves
of Khwarazm were liberated by decree within weeks of the city’s conquest,
while the slaves of Bukhara waited nearly half a decade for an “emancipa-
tion” decree stipulating that they would be freed only after another ten years
of bondage. The slave uprising of Khwarazm provides the answer. By casting
off their chains and rising up against their masters, the slaves revealed the
fundamental weakness of the slave-owning elites. While the slaves hailed the
arrival of Russian troops and the new governance that they would bring,

74
Becker, Russia’s Protectorates in Central Asia, 67–68. 75 ibid., 68.
76
Granted, the annexation of Merv likely helped to reduce the taking of Iranian captives
among the Tekes and their allies: Before General Komarov agreed to meet a deputation
suing for peace outside the town, he insisted that all the slaves in Merv be liberated
(A. Maslov, Rossiia v srednei azii, 380). But there is no reason to believe that this result
was a significant motivation in Merv’s occupation, nor in the brutal campaigns of conquest
leading up to it.
 Conclusion 207

they simultaneously proved their own ability to make the region ungovern-
able. They forced the Russians to take sides: With newspapers around the
world reporting on the events, the Tsar’s men could side with the slave-
owning classes, attempt to crush the uprising, and see that the slaves were
shackled and chained once more, or they could take their chances by
liberating the slaves.
It was not a foregone conclusion that the Russians would side with the
slaves. On the contrary, the public hanging of two slaves for their partici-
pation in the uprising was, as we have seen, perhaps the first act of Russian
governance in Khwarazm. The two bodies, left to rot in the heat of the
summer sun, were displayed for all to see, offering a grim “official” com-
ment on the uprising and certainly creating confusion among those slaves
who had anticipated having Kaufman as their ally. If the uprising had
ended with the execution of these two slaves, alleviating for Kaufman
the challenges of governing a realm in chaos, perhaps – as in Bukhara –
the emancipation decree would not have been issued for months or even
years to come. But the uprising began again with a fresh intensity, the slaves
proving that not even the threat of death could stop them. They had
already defied their masters; if the Russians wished only to become their
new masters, then the slaves would defy them too.
Kaufman, then, was left with no choice but to facilitate the movement
of liberation that was already in progress. He had no other reasonable
option: If he had turned his troops on the rioting slaves, the Russians could
have found themselves facing 30–60,000 new enemies, a great many of
whom had already demonstrated their boldness, their determination, and
their anger. The Khwarazmian military, itself consisting largely of Iranian
slaves, had just been routed and could hardly be relied upon to put down
the uprising for the convenience of their new occupiers (or, for that matter,
on behalf of their fallen khan). The slaves, inspired by the anticipation of
Russian solidarity, acted together to make it a foregone conclusion, even if
their confidence in its inevitability had been unwarranted.
The decision to abolish (however gradually) slavery in Bukhara, a step
which was undertaken just a few months later, must also be credited in
large part to the slaves of Khwarazm. By forcing Kaufman’s hand in
abolishing slavery in their territory, these slaves made it inevitable that
the same privilege should be extended to the slaves of the neighboring
domain. Continuing to countenance Bukharan slavery would have been
supremely awkward for Russian administrators who were about to reduce
Bukhara to “protectorate” status. Having gained a reputation as emanci-
pators, Russian officials had to consolidate their position in Central Asia
208 Conquest of Khiva and Myth of Russian Abolitionism

with the eyes of the world upon them. Still more urgently, the possibility of
a slave uprising in Bukhara, patterned on events next-door, could not have
escaped their attention.
However, the Russian press was all too happy to give Kaufman full
credit for the emancipation, and he has been credited for it ever since.
On the 25th anniversary of the conquest, the popular Russian journal Niva
offered a typical rhapsody: “The immediate consequences of the campaign
were: the liberation of 40,000 captives and slaves, the complete pacifica-
tion [polnoe umirotvorenie] of the country, the cessation of license, loot-
ing, and brigandage in the Khivan oasis and neighboring territory . . . This
achievement is inscribed in golden letters in the annals of Russian history,
and the memory of it will be cherished by the Russian people forever.”77
The “golden letters” would read “Kaufman,” though the General may very
well have been more surprised than pleased to find himself the emancipa-
tor of 30–60,000 Muslims. This had never been his intention until his hand
was forced.
Ultimately, then, visions of the conquest of Khwarazm as a Russian
abolitionist enterprise are a fiction. There is no evidence that the liberation
of tens of thousands of Iranian slaves played a significant role – or any role
whatsoever – in inspiring the campaign, nor any good evidence that
Kaufman had planned their liberation at any time before the slave uprising
began. In the five previous years following the conquest of Bukhara, the
Russians left no legal infrastructure or treaties in place to guarantee the
manumission of even a single Bukharan slave, let alone the complete
abolition of slavery, and there is little reason to believe that anything
more ambitious was planned for the slaves of Khwarazm. On the contrary,
it was the slaves who created the conditions for their own liberation. Their
courageous uprising rendered it more convenient for the Russians to
facilitate their freedom than to force them back into bondage. For
Kaufman, the erstwhile executioner of rebelling slaves, emancipation
became the only rational option. Despite his role in drafting the emancipa-
tion decree, then, it cannot be said that he freed the slaves; they had not
waited for a Russian invitation to cast off their chains. The slaves of
Khwarazm had freed themselves.
In this book, I have attempted to show that slavery was a fundamental
aspect of Central Asian society down to the late nineteenth century; that the
decentralized slave trade involved the circulation of slaves and their sellers
throughout the region, linking the metropole and periphery as well as the

77
“Dvadtsatipiatiletie pokoreniia Khivy,” Niva 24 (1898), 477–478.
 Conclusion 209

nomadic and the sedentary; that the slave trade was a key issue in diplomacy
between Central Asia, Russia, and Iran, as well as a factor which motivated
imperial invasions; that slaves could exert an influence on the nature of their
captivity by using or concealing their talents, or by manipulating the Islamic
legal system to their own advantage; that escaped slaves “liberated” at the
Russian border were often simply transferred into another corvée labor
system as part of the Russian effort to develop settler colonies in the border-
lands; that there was never any coherent Russian “abolitionist” enterprise in
Central Asia; that Turkmen “raiders” and “native informants” were used as
proxy fighters on the Iranian frontier; and that the slaves of Khwarazm, by
freeing themselves through courageous rebellion, provoked the abolition of
slavery throughout the region.
This book approaches slavery in Central Asia from multiple angles, in
other words, but it hardly exhausts the subject. Future research – both my
own and others’ – will, I hope, make use of the many documents on slavery
that, no doubt, have yet to be uncovered in the region’s archives. Much
remains to be said, moreover, about the parallels between Central Asian
slavery and forms of slavery elsewhere in Eurasia and in the wider world.
(A global-comparative approach to Central Asian slavery might start by
pointing out a conspicuously parallel “cotton connection”: The collapse of
the slave trade in the Civil War-era American South helped to crush the cotton
economy there, thus providing a market for Central Asian cotton, but devel-
oping the cotton economy in Central Asia would allegedly require the
collapse of its slave trade as part of the Russian “pacification” of the
region.78) Several recent books, meanwhile, offer intriguing avenues for
comparative studies focused more narrowly on Russia and Central Asia.
First, while I have attempted to show the surprising degree of autonomy
that was possible for “native informants” tasked with policing the slave trade
for the Russian Empire, recent work by Ian Campbell79 has shown how
Russia’s Kazakh “informants” demonstrated this kind of agency in other
spheres as well, oftentimes carving out advantages not only for themselves,
but also for their communities. A similar agency is seen in the legal sphere,
and new work by Paolo Sartori80 demonstrates the extent to which Muslim
jurists were permitted to carve out and maintain spheres of influence even as
other aspects of governance were being subsumed rigorously under Russian

78
I am grateful to James Millward for this observation.
79
Campbell, Knowledge and the Ends of Empire: Kazak Intermediaries and Russian Rule on
the Steppe, 1731–1917 (Ithaca, NY: Cornell University Press, 2017).
80
Sartori, Visions of Justice: Sharı̄ʿa and Cultural Change in Russian Central Asia (Leiden:
Brill, 2017).
210 Conquest of Khiva and Myth of Russian Abolitionism

imperial jurisdiction – a dynamic demonstrated by the extant legal documents

on slavery as well, which prove that the manumission, sale, and purchase of
slaves was still being supervised by Islamic court judges well into the 1880s.
Recent research by Christine Nölle-Karimi and Mohammad Ali Kazembeyki,
meanwhile, has shed light on the intricate political dynamics of northern Iran
and Afghanistan in the nineteenth century.81 Finally, a recent book by Erika
Monahan82 on Siberian merchants has expanded our knowledge of the trade
networks crisscrossing early modern Central Asia, further illuminating the
decentralized nature of Eurasian overland commerce.
When it comes to further research on Central Asian slavery, the impact
of slavery’s decline on the region’s agricultural and mercantile economies
is a particularly promising topic, which, unfortunately, I was not ready to
address here. It seems likely, for one thing, that the slave trade’s collapse
served to make some Turkmens more dependent on labor opportunities in
the industries developed under Russian imperial governance, such as
cotton agriculture. A related question concerns the agricultural workforce:
Who replaced the Iranian slaves as laborers in the fields, and what kinds of
social changes did this transition of manpower involve?
One question haunts me above all others: What happened to Central
Asia’s slaves – emancipated or otherwise – after the mid-1870s?
In 1877, four years after the conquest of Khiva, a petition arrived before
Muhammad Rahı̄m Khan, who was by this time reduced to the Tsar’s
˙ ˙
puppet ruler, bearing a complaint from a maker of fur coats. The author
of the petition was named Nūr Muhammad Makhdūm, and his complaint
˙
was that he and his fellow furriers at a Khiva bazaar were being taxed at
a rate they could no longer afford. Times had changed. “Business at our
bazaar is stagnant,” the furrier writes. “In times past, everyone had 5 or 10
slaves. The slaves received coats. Now the slaves are free and they have
gone off, and no one will take the coats.”83 Slavery, at least in Khiva,
appeared to be a thing of the past.84

81
Kazembeyki, Society, Politics, and Economics in Māzandarān, Iran, 1848–1914; Nölle-
Karimi, The Pearl in its Midst: Herat and the Mapping of Khurasan (15th–19th Centuries)
(Austrian Academy of Sciences Press, 2014).
82
Monahan, The Merchants of Siberia: Trade in Early Modern Eurasia (Ithaca, NY: Cornell
University Press, 2016).
83
“Brunlar har kimning besh-on dāna dughma bolur érdi dughmalargha fustun alur érdilar
emdi dughmalar āzād bolup ketip turur fustunni hichkim almay turur.” MS IVAN Uz
No. 12581, f. 11.
84
According to some, slavery would persist in Bukhara even into the Soviet period: see
M. A. Abduraimov, Ocherki agrarnykh otnoshenii v bukharskom khanstve v XVI–pervoi
polovine XIX veka, t. 2. Tashkent: Fan, 1970, 236.
 Conclusion 211

Or did the slave system simply take other forms, while becoming less
visible to foreign observers? Large agricultural estates still needed hands to
work them, and the will to secure cheap labor no doubt still existed among
the owners of those estates. While the flow of captives into the khanates
had been interrupted, more localized systems of corvée labor – such as
debt-based servitude, indentured servitude as a means of private dispute
resolution, and indenturing oneself as a last resort in circumstances of
extreme poverty – would presumably have continued unabated. After all,
there was nothing in the Khwarazmian or Bukharan abolition declarations
that would militate against these forms of bondage. Travelers and obser-
vers in the region in the final decades of the pre-Soviet era, convinced of
the success of Russian-sponsored abolition, would likely have overlooked
more subtle signs of unfree servitude. Unfortunately, available sources do
not provide much basis for speculation on this front.
It is clear from our sources, at least, that the centuries-long crisis of Iranian
captives being channeled into Khwarazm had finally come to an end. But
what became of the tens of thousands of former slaves who chose to remain in
the khanate? This demographic must have constituted a very large proportion
of the total number of freedmen. While reports of parties massacred on their
way to Iran are many, reports and eyewitness accounts of freedmen returning
home safely are scarce, scattered, and lacking in detail. The most notable
instances of repatriation were those accomplished via Russian steamships that
departed for Iran from Krasnovodsk, but as far as I can tell, the fortunate
former slaves boarding these ships numbered no more than a few thousand at
most. Others likely made their way to Russia eventually, as had some Iranian
former slaves of previous generations, though there is scant information on
such migrations. A lack of sources undermines efforts to reconstruct the freed
slaves’ fates; the tide of foreign observers that arrived with the Russian
regiments dissipated long before the trajectory of all the freed slaves could
be observed. Thus Schuyler writes, “[t]here were estimated to be 30,000
slaves in the Khanate, but it is supposed that not more than 5,000 of these
were actually freed before the departure of the Russians.”85 By the phrase
“actually freed,” Schuyler is hinting at a distinction between decree and
effect, which was likewise pursued by MacGahan, who wrote that “Those
who remained in Khiva, though emancipated, are not, I fancy, much better off
than before. Some Russian officers seemed to think that three-fourths of the
Persians would remain slaves still, and were of the opinion that General
Kaufmann did not act vigorously enough in this matter. However that be,

85
Schuyler, Turkistan, Vol. 2, 354.
212 Conquest of Khiva and Myth of Russian Abolitionism

there can be little doubt that the theoretical abolition of slavery will ulti-
mately result in its practical abolition.”86 These observations are, of course,
little more than guesswork.
Fortunately, Central Asian sources offer some hints as to the fate of the
freedmen who remained in Khwarazm. The evidence they provide suggests
a diverse spectrum of outcomes, ranging from degrees of eventual assim-
ilation into Khwarazmian society to the establishment of communities that
were quite separate and distinctive. Many former slaves settled in the
capital, where they and their descendants could be found living in their
own neighborhood (mahalla) well into the Soviet period. Two distinctive
mosques catered to this demographic in Khiva.87 Others chose to settle in
Tāshhawż and other small towns and qishlaqs where they had gathered
˙
soon after the declaration of abolition. According to the Azādnāma, by the
early Soviet period these communities had come to resemble the Uzbek
communities of Khwarazm both in their language (with Uzbek predomi-
nating over Persian, which had allegedly been forgotten) and in their
customs (urf-adatlar).88
Today in Uzbekistan, members of an eroniy (“Iranian”) diaspora com-
munity can be found throughout the country. Many attend a distinctive
Shiʿite mosque in Samarqand’s Panjob neighborhood. Many others live in
Tashkent, and overall the population of eroniy citizens in the country
numbers in the thousands. Some are descended from Iranian merchants
and others from migrant laborers. Many, however, can surely trace their
histories down a darker path. As we have seen, their enslaved forebears were
ubiquitous in Central Asia for a matter of centuries. In certain regions, such
as Khwarazm, slaves occupied such a diverse range of jobs and roles, in such
large numbers, that they can be considered a fundamental element of
society. The impact of slavery on Central Asian history has gone largely
unrecognized, but it can hardly be overestimated.

86
MacGahan, Campaigning on the Oxus, 312. 87 MS IVAN Uz No. 12581, f. 27a.
88
MS IVAN Uz No. 12581, ff. 31a–b. The fact of assimilation is especially striking when
contrasted with the distinctiveness of a proximate Iranian diaspora community, then living
in Ghazi-Abad rayon, consisting at that time of some 100 households of “Farsiyanlar” who
had retained Persian and maintained Iranian customs. These “Farsiyanlar” had never been
enslaved, having been relocated to Khwarazm from Aq-Darband in the time of Allāh Qulı̄
Khan and offered enough land so that they eventually commanded their own mounted
force of retainers (atligh nökar) and gained tax exemptions owing to their elite status (see
MS IVAN Uz No. 12581, ff. 31b–32a). One can well imagine that the Iranian former
slaves, starting their free lives in Khwarazm at the bottom of the social ladder, may have
felt greater pressure to assimilate in order to find work and gain a social foothold.
 Bibliography

ARCHIVAL COLLECTIONS

Tsentralnyi Gosudarstvennyi Arkhiv Respubliki Kazakhstan (TsGARKaz)/Central

State Archive of the Republic of Kazakhstan. Almaty, Kazakhstan.

PUBLISHED DOCUMENT COLLECTIONS

Agadzhanov, S. G., ed. Russko-turkmenskie otnosheniia v XVIII–XIX vv.: do prisoe-

dineniia Turkmenii k Rossii: sbornik arkhivnykh dokumentov. Ashkhabad: Izd-vo
Akademii nauk Turkmenskoi SSR, 1963.
Faiziev, T. Buxoro feodal jamiyatida qullardan foydalanishga doir hujjatlar
(XIX asr). Tashkent: Fan, 1990.
Fitrat, Bedil and B. S. Sergeev, eds. Kaziiskie dokumenty XVI veka. Tekst, perevod,
ukazatek’ vstrechaiushchikhsia iuridicheskikh terminov i primechaniia. Tashkent:
Izd. Komiteta Nauk UzSSR, 1937.
Ismailova, S. K. “Dokumenty o rabstva v bukharskom khanstve v XIX– nachale XX
v.” Izvestiia Akademii nauk Tadzhikskoi SSR, otdelenie obshchestvennykh nauk
2/72 (1973), 20–30.
Materiali po istorii turkmen i turkmenii, t. II. XVI–XIX vv. Iranskie, bukharskie,
i khivinskie istochniki. Moscow: 1938–39.
Roloff, Gustav, ed. Das Staatsarchiv: Sammlung der officiellen Actenstücke zur
Geschichte der Gegenwart. Leipzig: von Duncker & Humbolt. 1861–71.
Serebrennikov, A. G., ed. Sbornik Materialov dlia istorii zavoevaniia
Turkestanskogo Kraia (1839–1844), 4 vols. Tashkent: Tip. Sht. Turkestanskogo
V.O., 1908–12.

PRIMARY SOURCES

Abbott, James. Narrative of a Journey from Heraut to Khiva, Moscow, and

St. Petersburgh, during the Late Russian Invasion of Khiva, 2 vols. London:
W.H. Allen and Co., 1843.

213
214 Bibliography

Alikhanov-Avarskii, M. Pokhod v Khivu (Kavkazkikh otriadov) 1873.

St. Petersburg: Step’ i oazis, 1899.
Amı̄n Oghlı̈, Hassan Muhammad. Āzādnāma. MS IVAN Uz No. 12581.
Anonymous, ˙“Iz istorii ˙ Kazakhstana XVIII v.” Krasnyi arkhiv 2/87 (1938),
129–154.
Anonymous. “Nevol’niki v Khive,” Vestnik Evropy 80:7 (1815), 244–247.
Anonymous. “Khiva za sto let nazad.” Syn otechestva 1 (1842), 33–38.
Anonymous. “Dvadtsatipiatiletie pokoreniia Khivy,” Niva 24 (1898),
477–478.
Anonymous. “Zavoevanie Akhal-Tekinskogo oazisa.” Istoricheskii vestnik 7
(1881), 577–596.
Anonymous. “Russia and Khiva: The Empire’s Manifesto Prefatory to the
Expedition,” New York Times, April 13, 1873.
Āshtı̄yānı̄, Mı̄rzā Mahmūd Taqı̄. ʿIbratnāma: Khatirati az dawran-i pas az jangha-yi
herat va merv, ed.˙ Husayn ʿImadı̄ Āshtı̄yānı̄. Tehran: Nashr-i markaz, 1382/
2003.
Atkinson, Lucy. Recollections of Tartar Steppes and Their Inhabitants. London:
J. Murray, 1863.
Beisembiev, Timur. The Life of Alimqul: A Native Chronicle of Nineteenth Century
Central Asia. New York: Routledge, 2015.
Blaramberg, I. F. Vospominaniia. Moscow: Glavnaia redaktsiia vostochnoi litera-
tury; TSentral’naia Aziia istochnikakh i materialakh XIX–nachala XX veka,
1978.
Blinov, N. Sarapul’: istoricheskii ocherk’. Sarapul, 1887.
Boulger, Demetrius Charles. England and Russia in Central Asia, Vol. 1. London:
W.H. Allen, 1879.
Brydges, Harford Jones. The Dynasty of the Kajars. London: John Bohn,
1833.
Burnes, Alexander. Travels into Bokhara, 2 vols. London: John Murray, 1839.
Dal’, V. I. “Pis’mo k druziam iz Khivinskoi Ekspeditsii.” Russkii Arkhiv 3 & 4
(1867), 402–431, 606–639.
de Levchine, A. Description des hordes et des steppes des Kirghiz-Kazaks ou Kirghiz-
Kaissaks. Paris: Impr. royale, 1840.
de Meyendorff, Georges. A Journey from Orenburg to Bukhara in the Year 1820.
Calcutta: Foreign Department Press, 1870.
DuBois, W. E. B. The Suppression of the African Slave Trade to the United States of
America, 1638–1870. New York: Longmans, Green, and Co., 1904.
Ferrier, J. P. Caravan Journeys and Wanderings in Persia, Afghanistan, Turkistan, and
Beloochistan. London: J. Murray, 1857.
Ghafūr, Muhammad ‘Alı̄ Khan. Rūznāma-yi safar-i Khvārazm. Tehran: Daftar-i
Mutāla‘āt-i˙ Siyāsı̄ va Bayn al-Milalı̄, Vizārat-i Umūr-i Khārijah, 1994.
˙
Gladyshev and Muravin. Poezdka iz Orska v Khivu i obratno, sovershennaia
v 1740–41 godakh poruchikom Gladyshevym i geodezistom Muravinym.
St. Petersburg, 1851.
Gordon, Edward. The Roof of the World: Being a Narrative of a Journey over the
High Plateau of Tibet to the Russian Frontier and the Oxus Sources on the Pamir.
Edinburgh: Edmonston and Douglas, 1876.
 Bibliography 215

Grodekov, Governor-General N. I. Khivinskii pokhod 1873 goda deistviia kavkazs-

kikh otriadov. St. Petersburg: Tip. V.S. Balashova, 1883.
Grodekoff, N. I. Colonel Grodekoff ’s ride from Samarcand to Herat, through Balkh
and the Uzbek States of Afghan Turkestan. Translated by Charles Marvin.
London: W.H. Allen, 1880.
Gunaropulo, C. A. “V Turkmenskoi stepi. (Iz zapisok chernomorskogo ofitsera.)”
Istoricheskii vestnik 11/12, 555–584, 1033–1051.
Hidāyat, Rizā Quli Khan. Sifāratnāma-yi Khwarazm (Relation de l’Ambassade au
Kharezm [Khiva] de Riza Qouly Khan. Texte Persan), ed. Charles Schefer. Paris:
Ernest Leroux, 1876.
Iskandar Beg Munshı̄, Tārı̄kh-i ʿālamārā-yi ʿAbbāsı̄. Translated by Roger Savory
as History of Shah ʿAbbas the Great, Vol. 2. Boulder, CO: Caravan Books,
1978.
Ivanin, M. Opisanie Zimnego Pokhoda v Khivu v 1839–40 g. St. Petersburg,
1874.
Ivanin, M. and Golosov, D. “Pokhod v Khivu v 1839 godu otriada russkikh voisk,
pod nachal’stvom General-Ad’iutanta Perovskago.” Voennyi Sbornik (1863)
1, 3–72; 2, 309–358; 3, 3–71.
Jenkinson, Anthony. Early Voyages and Travels to Russia and Persia by Anthony
Jenkinson and Other Englishmen, Vol. 1, ed. E. Delmar Morgan and C. H. Coote.
London: Hakluyt Society, 1886.
Karpiuk, S. G., ed. Puteshestviia po Vostoku v opokhu Ekateriny II. Moscow:
Vostochnaia literatura, 1995.
Kemp, P. M., ed. and trans. Russian Travelers to India and Persia (1624–1798):
Kotov, Yefremov, Danibegov. Delhi: Jiwan Prakashan, 1959.
Ker, David. On the Road to Khiva. London: Henry S. King and Co., 1874.
Khalfin, N. A., ed. Zapiski o Bukharskom khanstve. Moscow, 1983.
Khan, ‘Abbās Qulı̄, Safarnāma-yi Bukhārā, ed. Hussain Zamānı̄. Tehran:
˙
Pizhūhishgāh-i ‘Ulūm-i Insānı̄ va Mutāla‘āt-i Farhangı̄ vā bastah bih Vizārat-i
˙
Farhang va Āmūzish-i ‘Ālı̄, 1373/1995.
Khanykov, Ia.V. Poezdka iz Orska v Khivu i obratno, sovershennaia v 1740–1741
godakh Gladyshevym i Muravinym. St. Petersburg, 1851.
Khoroshkhin, A. Sbornik statei: Kasaiushchikhsia do Turkestanskago kraia.
St. Petersburg, 1876.
Kostenko, L. F. Puteshestvie v Bukharu russkoi missii v 1870 godu. St. Petersburg,
1871.
Kosyrev, E. M. “Pokhod v Khivu v 1839 godu (Iz zapisok uchastnika).” Istoricheskii
vestnik 8 (1898), 538–545.
Kraft, I. I. “Unichtozhenie rabstva v kirgizskoi stepi.” Iz kirgizskoi stariny.
Orenburg, 1900.
Kraft, I. I. Sudebnaia chast’ v Turkestanskom krae i stepnykh oblastiakh. Orenburg,
1898.
Kuropatkin, A. N. Zavoevanie turkmenii: pokhod v akhal-teke v 1880–81 gg.
St. Petersburg, 1899.
Lal, Mohan. Travels in the Panjab, Afghanistan, and Turkistan, to Balk, Bokhara,
and Herat; and a Visit to Great Britain and Germany. London: W.H. Allen,
1846.
216 Bibliography

Leskov, Nikolai. The Enchanted Wanderer and Other Stories, trans. Richard Pevear
and Larissa Volokhonsky. New York, NY: Alfred A. Knopf, 2013.
MacGahan, J. A. Campaigning on the Oxus, and the Fall of Khiva. London:
Sampson Low, 1874.
Major Blankenagel’. Zamechaniia maiora Blankenagelia, vposledstvie poezdki ego
iz Orenburga v Khivu v 1793–94 godakh. St. Petersburg, 1858.
Marghı̄nānı̄, ‘Alı̄ ibn Abı̄ Bakr. The Hedaya, trans. Charles Hamilton. London:
Allen & Co., 1870 (second edition).
Markozov, V. Krasnovodskii otriad. Ego zhizn’ i sluzhba so dnia vysadki na vos-
tochnyi bereg Kaspiiskogo moria po 1873 g. St. Petersburg: 1898.
Maslov, A. “Rossiia v Srednei Azii. (Ocherk nashikh noveishikh priobretenii).”
Istoricheskii vestnik 5 (1885).
Meyendorff, Baron Georg von. A Journey from Orenburg to Bukhara in the Year
1820. Calcutta, 1870.
Mῑrpanja, Ismaʿı̄l Sarhang. Khātirāt-i asārat: ruznāma-yi safar-i Khwarazm va
khiva, ed. Safā’ ad-Dı̄n Tabarrā’iyān. Tehran: Muʿssassa-yi Pazhuhish va
Mutālaʿat-i Farhangı̄, 1370/1991.
Munis, Shı̄r Muhammad Mı̄rāb and Rizā Mı̄rāb Āgāhı̄, Muhammad. Firdaws
al- iqbāl, trans.˙ Yuri Bregel. Leiden: Brill, 1999. ˙
Murav’ev, N. N. Muraviev’s Journey to Khiva through the Turcoman Country,
1819–20. Calcutta: Foreign Department Press, 1871.
Petrusevich, N. G. “Turkmeny mezhdu starym’ ruslom’ Amu-Daryi (Uzboem’)
i severnymi okrainami Persii.” Zapiski kavkazskago otdela imperatorskago russ-
kago geograficheskago obshchestva, Vol. 11. Tbilisi, 1880.
Rzhevuskii, A. “Ot Tiflisa do Dengil’-tepe.” Voennyi sbornik 9 (1884).
Schmidt, Emil. The Russian Expedition to Khiva in 1873. Calcutta: Foreign
Department Press, 1876.
Schuyler, Eugene. Turkistan: Notes of a Journey in Russian Turkistan, Khokand,
Bukhara, and Kuldja. New York, NY: Scribner, Armstrong, 1876.
Shubinskii, P. “Ocherki bukhary,” Istoricheskii vestnik 7 (1892).
Stremoukhov, N. P. “Poezdka v Bukharu.” Russkii vestnik 6 (1876).
Tugan-Mirza-Baranovskii, V. A. Russkie v Akhal-Teke. St. Petersburg, 1881.
Vambery, Arminius. Travels in Central Asia. New York, NY: Harper & Bros.,
1865.
Vambery, Arminius. Sketches of Central Asia. London: W.H. Allen & Co.,
1868.
Vambery, Arminius. History of Bokhara from the Earliest Period down to the
Present. London: Henry S. King, 1873.
Wolff, Joseph. Narrative of a Mission to Bukhara in the Years 1843–1845, Vol. 2.
London: J.W. Parker, 1845.
Wood, Herbert. The Shores of Lake Aral. London: Smith, Elder & Co., 1876.
Yate, C. E. Khurasan and Sistan. London: Blackwood and Sons, 1900.
Zakhar’in, I. N. Graf V. A. Perovskii i ego zimnii pokhod v Khivu. St. Petersburg:
Tip. P. P. Soikina, 1899 and 1901.
Zalesov, N. “Pis’mo iz Khivy,” Voennyi sbornik 1 (1859).
Zalesov, N. “Posol’stvu v Khivu v 1842 godu.” Istoricheskii vestnik 11 (1894).
Zheliabuzhskii, E. Ocherki i zavoevanie khivy. Moscow, 1875.
 Bibliography 217

SECONDARY SOURCES

Alam, Muzaffar. “Trade, State Policy and Regional Change: Aspects of Mughal-
Uzbek Commercial Relations, c. 1550–1750.” Journal of the Economic and
Social History of the Orient 37:3 (1994), 202–227.
Amanat, Abbas and Arash, Khazeni. “The Steppe Roads of Central Asia and the
Persian Captivity Narrative of Mahmud Mirza Taqi Ashtiyani.” In Nile Green,
ed., Writing Travel in Central Asian History, 113–133. Bloomington, IN: Indiana
University Press, 2014.
Ayalon, David. The Mamluk Military Society: Collected Studies. London:
Varorium, 1979.
Becker, Seymour. Russia’s Protectorates in Central Asia: Bukhara and Khiva,
1865–1924. New York, NY: Routledge, 2004 (1969).
Beckwith, Christopher I. Empires of the Silk Road: A History of Central Eurasia
from the Bronze Age to the Present. Princeton, NJ: Princeton University Press,
2009.
Bregel, Yuri. Khorezmskie turkmeny v XIX veke. Moscow, 1961.
Bregel, Yuri. “The New Uzbek States: Bukhara Khiva and Qoqand: c. 1750–1886.”
In Nicola Di Cosmo, Allen J. Frank and Peter B. Golden, eds., The Cambridge
History of Inner Asia: The Chingghisid Age, 392–393. Cambridge: Cambridge
University Press, 2009.
Burton, Audrey. The Bukharans: A Dynastic, Diplomatic, and Commercial History
1550–1702. New York, NY: St. Martin’s Press, 1997.
Campbell, Ian. Knowledge and the Ends of Empire: Kazak Intermediaries and
Russian Rule on the Steppe, 1731–1917. Ithaca, NY: Cornell University Press,
2017.
Chatterjee, Indrani and Richard M. Eaton, eds. Slavery & South Asian History.
Bloomington, IN: Indiana University Press, 2006.
Clarence-Smith and Gervase, William. Islam and the Abolition of Slavery. Oxford:
Oxford University Press, 2006.
Crossley, Pamela K. “Slavery in Early-Modern China.” In David Eltis and
Stanley Engerman, eds., The Cambridge World History of Slavery, Vol. 3,
186–216. Cambridge: Cambridge University Press, 2011.
Dale, Stephen. Indian Merchants and Eurasian Trade, 1600–1750. Cambridge:
Cambridge University Press, 1994.
Davies, Brian L. Warfare, State and Society on the Black Sea Steppe, 1500–1700.
New York, NY: Routledge, 2007.
Drescher, Seymour. Abolition: A History of Slavery and Anti-Slavery. Cambridge:
Cambridge University Press, 2009.
Erdem, Y. Hakan. “Slavery and Social Life in Nineteenth-Century Turco-Egyptian
Khartoum.” In Terence Walz and Kenneth M. Cuno, eds., Race and Slavery in the
Middle East: Histories of Trans-Saharan Africans in 19th-Century Egypt, Sudan,
and the Ottoman Mediterranean, 125–146. Cairo: American University in Cairo
Press, 2010.
Farah, M. D. “Autocratic Abolitionists: Tsarist Russian Anti-Slavery Campaigns.”
In William Mulligan, ed., A Global History of Anti-Slavery Politics in the
Nineteenth Century, 97–117. New York, NY: Palgrave, 2013.
218 Bibliography

Fisher, Alan W. “Muscovy and the Black Sea Slave Trade.” Canadian-American
Slavic Studies 6:4 (1972), 575–594.
Gao, Yuan. “Captivity and Empire: Russian Captivity Narratives in Fact and
Fiction.” M.A. thesis, Nazarbayev University, 2016.
Golden, Peter B. “The Terminology of Slavery and Servitude in Medieval Turkic.”
In Devin DeWeese, ed., Studies on Central Asia in Honor of Yuri Bregel, 27–56.
Bloomington, IN: Research Institute for Inner Asian Studies, 2001).
Grant, Bruce. The Captive and the Gift: Cultural Histories of Sovereignty in Russia
and the Caucasus. Ithaca, NY: Cornell University Press, 2009.
Gross, Jo-Ann and Asom Urunbaev, eds. The Letters of Khwāja ʿUbayd Allāh Ahrār
˙
and his Associates. Leiden: Brill, 2002.
Gustafson, James M. “Qajar Ambitions in the Great Game: Notes on the Embassy
of ‘Abbas Qoli Khan to the Amῑr of Bokhara, 1844.” Iranian Studies 46:4 (2013),
535–552.
Hopkins, Benjamin D. “Race, Sex and Slavery: ‘Forced Labour’ in Central Asia and
Afghanistan in the Early 19th Century.” Modern Asian Studies 42:2 (2008),
629–671.
Hopper, Matthew S. Slaves of One Master: Globalization and Slavery in Arabia in
the Age of Empire. New Haven, CT: Yale University Press, 2015.
Horák, Slavomı́r. “The Battle Gökdepe in the Turkmen post-Soviet Historical
Discourse.” Central Asian Survey 34:2 (2015), 149–161.
Kazembeyki, Mohammad Ali. Society, Politics, and Economics in Māzandarān,
Iran, 1848–1914. London: Routledge, 2003.
Khazanov, Anatoly. Nomads and the Outside World. Madison: University of
Wisconsin Press, 1994 (second edition).
Khazanov, Anatoly and Andre Wink, eds. Nomads in the Sedentary World.
New York, NY: Routledge, 2001.
Khazeni, Arash. “Across the Black Sands and the Red: Travel Writing, Nature, and
the Reclamation of the Eurasian Steppe circa 1850.” International Journal of
Middle East Studies 42 (2010), 591–614.
Kreiten, Irma. “A Colonial Experiment in Cleansing: the Russian Conquest of
Western Caucasus, 1856–65.” Journal of Genocide Research 11:2–3 (2009),
213-241.
Kurtynova-D’Herlugnan, Liubov. The Tsar’s Abolitionists: Languages of
Rationalization and Self-Description in the Russian Empire. Leiden: Brill,
2010.
Lee, Anthony A. “Enslaved African Women in Nineteenth-Century Iran: The Life
of Fezzeh Khanom of Shiraz.” Iranian Studies 45:3 (2012), 417–437.
Levey, Benjamin. “Jungar Refugees and the Making of Empire on Qing
China’s Kazakh Frontier, 1759–1773.” Ph.D. dissertation, Harvard
University, 2013.
Levi, Scott C. “Hindus beyond the Hindu Kush: Indians in the Central Asian Slave
Trade.” Journal of the Royal Asiatic Society 12:3 (2002), 277–288.
Levi, Scott C. The Indian Diaspora in Central Asia and Its Trade, 1550–1900.
Leiden: Brill, 2002.
Levi, Scott C. “Early Modern Central Asia in World History.” History Compass
10:11 (2012), 866–878.
 Bibliography 219

Levi, Scott C. “India, Russia and the Eighteenth-Century Transformation of the

Central Asian Caravan Trade.” Journal of the Economic and Social History of the
Orient 42:4 (1999), 519–526.
Levi, Scott C. and Muzaffar Alam, eds. India and Central Asia: Commerce and
Culture, 1500–1800. Oxford: Oxford University Press, 2007.
Markovitz, Claude. The Global World of Indian Merchants, 1750–1947: Traders of
Sind from Bukhara to Panama. New York, NY: Cambridge University Press, 2000.
Markovitz, Claude, Jacques Pouchepaass and Sanjay Subrahmanyam, eds. Society
and Circulation: Mobile People and Itinerant Cultures in South Asia, 1750–1950.
London: Anthem, 2003.
Marvin, Charles. Merv, Queen of the World, and the Scourge of the Man-Stealing
Turcomans. London: W.H. Allen, 1881.
McChesney, Robert D. Central Asia: Foundations of Change. Princeton, NJ:
Darwin Press, 1996.
Miller, Joseph C. Slavery and Slaving in World History: A Bibliography, 2 vols.
Armonk, NY: M.E. Sharpe, 1999.
Mirzai, Behnaz A. A History of Slavery and Emancipation in Iran. Austin, TX:
University of Texas Press, 2017.
Montana, Ismael M. The Abolition of Slavery in Ottoman Tunisia. Gainesville, FL:
University Press of Florida, 2013.
Morier, James. The Adventures of Hajji Baba of Ispahan. London: MacMillan,
1902.
Morrison, Alexander. “Twin Imperial Disasters. The Invasions of Khiva and
Afghanistan in the Russian and British Official Mind, 1839–1842.” Modern
Asian Studies 48:1 (2013), 253–300.
Morrison, Alexander. “Killing the Cotton Canard and Getting Rid of the Great
Game: Rewriting the Russian Conquest of Central Asia, 1814–1895.” Central
Asian Survey 33:2 (2014), 131–142.
Mukminova, R. K. Sotsial’naia differentsiatsiia naseleniia gorodov Uzbekistana
v XV–XVI vv. Tashkent, 1985.
Newby, Laura. “Bondage on Qing China ’s North-Western Frontier.” Modern Asian
Studies 47:3 (2013), 968–994.
Nölle-Karimi, Christine. “‘Different in All Respects’: Bukhara and Khiva as Viewed
by Kāǧār Envoys.” In Yavuz Köse, ed., Şehrâyı̂n: die Welt der Osmanen, die
˙
Osmanen in der Welt: Wahrnehmungen, Begegnungen und Abgrenzungen;
Festschrift Hans Georg Majer, 435–446. Wiesbaden: Harrassowitz, 2012.
Nölle-Karimi, Christine. The Pearl in its Midst: Herat and the Mapping of Khurasan
(15th–19th Centuries). Austrian Academy of Sciences Press, 2014.
Nölle-Karimi, Christine. “On the Edge: Eastern Khurasan in the Perception of
Qajar Officials.” Eurasian Studies 14 (2016), 135–177.
Petrushevsky, I. P. Islam in Iran. Albany, NY: State University of New York Press,
1985.
Ricks, Thomas. “Slaves and Slave-Trading in Shi’i Iran, AD 1500–1900.” Journal of
Asian and African Studies 36:4 (2001), 407–418.
Rossabi, Morris. “The ‘Decline’ of the Central Asian Caravan Trade.”
In James Tracey, ed., The Rise of Merchant Empires, 351–370. Cambridge:
Cambridge University Press, 1990.
220 Bibliography

Saray, Mehmet. “Russo-Turkmen Relations up to 1874.” Central Asian Survey 3:4

(1984), 15–48.
Sartori, Paolo. Visions of Justice: Sharı̄ʿa and Cultural Change in Russian Central
Asia. Leiden: Brill, 2017.
Shepherd, Gill. “The Comorians and the East African Slave Trade.” In James
L. Watson, ed., Asian and African Systems of Slavery, 73–99. Oxford: Basil
Blackwell, 1980.
Sela, Ron. The Legendary Biographies of Tamerlane: Islam and Heroic Apocrypha in
Central Asia. Cambridge: Cambridge University Press, 2013.
Sela, Ron. “Prescribing the Boundaries of Knowledge: Seventeenth-Century
Russian Diplomatic Missions to Central Asia.” In Nile Green, ed., Writing
Travel in Central Asian History, 69–88. Bloomington, IN: Indiana University
Press, 2014.
Semeniuk, G. I. “Likvidatsiia rabstva v Kazakhstane.” Reprinted in
Zh. O. Artykbaev, ed., Raby i tiulenguty v kazakhskoi stepi. Astana: Altyn
Kitap, 2006.
Sergeev, Evgeny. The Great Game, 1856–1907: Russo-British Relations in Central
and East Asia. Baltimore, MD: Johns Hopkins University Press, 2013.
Smolarz, Elena. “Speaking about Freedom and Dependency: Representations and
Experiences of Russian Enslaved Captives in Central Asia in the First Half of the
19th Century.” Journal of Global Slavery 2 (2017), 44–71.
Sneath, David. The Headless State: Aristocratic Orders, Kinship Society, &
Misrepresentations of Nomadic Inner Asia. New York, NY: Columbia University
Press, 2007.
Stanziani, Alessandro. Bondage: Labor and Rights in Eurasia from the Sixteenth to
the Early Twentieth Centuries. New York, NY: Berghahn, 2014.
Subtelny, Maria. Timurids in Transition: Turko-Persian Politics and Acculturation in
Medieval Iran. Leiden: Brill, 2007.
Sunderland, Willard. Taming the Wild Field: Colonization and Empire on the
Russian Steppe. Ithaca, NY: Cornell University Press, 2004.
Tealakh, Gali Oda. “The Russian Advance in Central Asia and the British Response,
1834–1884.” Ph.D. Dissertation, Durham University, 1991.
Terent’ev, M. A. Istoriia zavoevaniia Srednei Azii, 3 vols. St. Petersburg, 1906.
Toledano, Ehud R. As If Silent and Absent. New Haven, CT: Yale University Press,
2007.
Toledano, Ehud R. The Ottoman Slave Trade and its Suppression: 1840–1890.
Princeton, NJ: Princeton University Press, 1982.
Toledano, Ehud R. “Abolition and Anti-slavery in the Ottoman Empire: A Case to
Answer?” Forthcoming, 2018.
Trotskii, V. N. Opisanie khivinskago pokhoda 1873 goda. St. Petersburg, 1898.
Troutt Powell, Eve. Tell This in My Memory: Stories of Enslavement from Egypt,
Sudan, and the Ottoman Empire. Palo Alto, CA: Stanford University Press,
2013.
UNESCO / Ecole d’Architecture Paris Val de Seine EVCAU Research Team.
Inventory of Caravanserais in Central Asia. 2004.
Witzenrath, Christoph, ed. Eurasian Slavery, Ransom and Abolition in World
History, 1200–1860. Aldershot: Ashgate, 2015.
 Bibliography 221

Wood, William A. “The Sariq Turkmens of Merv and the Khanate of Khiva in the
Early Nineteenth Century.” Ph.D. dissertation, Indiana University, 1998.
Zhukovskii, S. V. Snosheniia Rossii s Bukharoi i Khivoi za poslednee trekhsotletie.
Petrograd, 1915.
Zimanov, S. Obshchestvennyi stroi kazakhov pervoi poloviny XIX veka i Bukeevskoe
khanstvo. Almaty: Arys, 2009.
 Index

ʿAbbās Mirzā, 37, 54, 196 work as court poet, 109

ʿAbbās Qulı̄ Khan, 39 work as painter, 108
ʿAbd al-Ahad, 78 Ashur Ada, 198, 202
˙
ʿAbd al-ʿAzı̄z Sultān, 36 Aslanov, Bogdan, 52
˙
ʿAbd al-Mu’min, 34, 36 Astarabad, 14, 33, 40, 81, 146, 174, 175,
ʿAbdullāh (slave from Herat), 62, 65 182, 196, 198, 199, 200, 201, 202
ʿAbdullāh Khan, 36 Astrakhan, 48, 123, 147
ʿAbdullāh Khan II, 34 Atrek River, 14, 19, 191, 198, 199, 200,
Abolitionism 201, 202, 203
In Western Europe, 183 ʿAwz Keldi Khan, 191, 192
Afghans, 61, 69, 80, 138, 196, 197
Ahrār, Khwāja ʿUbaydullāh, 35 Badakhshan, 75, 79, 80
˙
Akhal, 196 Badghis, 38
Alikhanov-Avarskii, Maksut, 185, 189 Balahisar, 75
ʿAlimqul (Mullā Qulῑ Khan), 104 Balamurghab, 93, 94
Allāh Qulı̄ Khan, 38, 50 Balkh, 34, 49, 75, 89
altun-jilaw, 121 Bashkirs, 145, 150
American South, 209 Bekovich-Cherkasskii expedition, 49
Amirabad, 37 Blankenagel’, Major, 28, 58, 59, 118, 119,
Arabs, 61, 153, 154 127
Ashgabat, 67 Blaramberg, I.F., 75, 76
Ashraf, 175 Britain, 22, 38
Āshtiyānῑ, Mῑrzā Mahmūd Taqῑ, 23, 88–114 diplomacy with Central Asia, 41
˙
failed attempts to escape from Panjdih, 90, Bukhara, viii, 13, 16, 18, 21, 22, 23, 25, 26,
91, 93, 95 27, 28, 30, 36, 38, 39, 47, 48, 49, 52,
freed from slavery, 110 54, 55, 57, 58, 59, 64, 65, 68, 70, 73,
held in dungeons, 107 74, 75, 76, 77, 78, 79, 80, 81, 82, 83,
life with Āqā Muhammad, 98–103 84, 85, 86, 88, 89, 96, 97, 98, 99, 100,
˙
life with Khan Muhammad, 89–97 101, 102, 103, 104, 105, 106, 107, 108,
˙
life with Shāhrukh Khan, 105 109, 110, 112, 113, 114, 119, 120, 121,
plot to poison his master, 93 122, 123, 124, 127, 128, 133, 134, 136,
sent to Bukhara by master, 97 137, 138, 140, 148, 153, 154, 156, 157,
taken captive by Turkmens, 89 166, 167, 184, 185, 191, 194, 199, 205,
work as calligrapher, 100 206, 207, 208, 210

222
 Index 223

abolition of slavery in, 77, 83, 184, 185, Hanafi law, 34, 84, 115–118, 126,
206 143
Russian conquest of, 55, 76, 184, 205 harem, 17, 132, 134, 137, 143
Russian missions to, 52 Haydar Qulı̄ Khan, 198
˙
slave-owning among elites, 119, 123 Hazaras, 61, 138, 153
statistics on slavery in, 26, 59 Herat, 37, 38, 62, 63, 64, 75, 91, 97, 139,
Burnes, Alexander, 32, 73, 74, 77, 120, 140 196, 197
Herātı̄, Shams al-Dı̄n, 34
Caravans, 22, 23 Hidāyat, Riżā Qulı̄ Khan, 39–40
Caribbean, slavery in the, 143 Hisar, 36
Caspian Sea, 22, 52, 78, 123, 174, 175,
195 Ilbars Khan, 37
Catherine the Great, 149 Ilkhan, Jaʿfar Qulı̄ Khan, 198
Charjui, 75, 77 Imām Qulı̄ Khan, 36, 48
Chawdurs, 36, 38, 196 India, 33, 49, 56, 57, 58
Chikishlar, 203 Iran, v, 13, 14, 15, 16, 20, 22, 23, 34,
China, 56, 131 35, 37, 38, 39, 41, 42, 43, 44, 46, 53,
Chuvash, 52 56, 82, 89, 110, 114, 117, 128, 146,
Cossacks, 18, 75, 146, 186, 193 148, 154, 162, 163, 174, 175, 181, 189,
Crimea, 78 190, 191, 194, 195, 199, 201, 202, 209,
210, 211
Danilevskii, G.I., 44, 46 conditions in the 1870s, 184
Demaisons, P.I., 74 diplomacy with Central Asia, 13, 38–47
Dost Muhammad Khan, 92 diplomatic missions to end slave trade, 47
˙
hostilities with Turkmens, 38, 174
Elder Horde, 164 raided by Turkmens under Russian
Ergeld Khan, 176–180 protection, 175
eroniy, 212 relations with Bukhara, 39
Ersaris, 196 rights over the Caspian Sea, 174
travel literature in, 39
Fath ‘Alῑ Khan, 37 treatment of captive Turkmens, 15–17
Ferghana, 36 Iskandar Beg, 33
Firdaws al-iqbāl, 16, 17, 121 Istanbul, 33, 58, 73
Forced migration, 19 Istı̄lād, 117
Fort Perovskii, 147 Ivanov, Lieutenant Colonel N.A., 203
France, 107
Jamshidis, 38
Gens, G.F., 50 Jenkinson, Anthony, 47
Geok Teppe, 203, 204 Junior Horde, 165
Georgians, 138 Juybāri shaykhs, 35
Ghafūr, Muhammad ʿAlῑ Khan, 40–47
˙
Ghazni, 75 Kabul, 33, 75, 80, 102
Gladyshev, Lieutenant Dmitry, 27, 37, 119 Kalmyks, 27, 30, 36, 49, 61, 120, 129, 130,
Gökleng, 19, 181 131, 144, 145, 150, 164
Gorchakov, Alexander, 20 Karakalpaks, 61, 64, 67, 69, 70, 119, 158,
Gribov, Anisim, 49 159
Grodekov, Governor-General N.I., 135, Karakul, 74, 146, 150
184, 185, 190, 191, 197, 201, 202, 203, Katlish, 19
204 Kaufman, General Konstantin von, 76, 77,
Gunaropulo, C.A., 175, 176–180 183, 186, 187, 188, 193, 203, 205, 206,
Gurgen River, 181 207, 208
224 Index

Kazakhs, v, vi, 13, 17, 18, 19, 21, 25, 26, 28, Kohne Urgench, 192
29, 31, 36, 52, 53, 55, 57, 60, 62, 64, 65, Kokand, 104, 107, 204
66, 67, 68, 69, 70, 86, 104, 119, 120, Kostenko, Captain L.F., 27, 28,
123, 125, 129, 130, 131, 136, 137, 138, 76, 77
139, 144, 145, 146, 148, 152, 153, 154, Kosyrev, E.M., 51
155, 156, 157, 158, 159, 161, 162, 163, Kraft, I.I., 52, 53, 66
164, 165, 166, 167, 169, 170, 171, 172, Krasnovodsk, 203, 211
173, 175, 183, 186, 195, 197, 199, 209 Krivobokova, Akulina, 23
as captives, 16 Kunduz, 29
Russian perceptions of, 19–20 Kuropatkin, A.N., 184, 203
slavery among, 120, 144–162
Kazalinsk, 186 Labab, 89
Kazan, 48, 146, 150 Lal, Mohan, 74, 75, 80
Khanazād, 116 Lobov, Major, 75
Khiva, v, 19, 21, 22, 23, 24, 27, 28, 29, 30, Lomakin, General N.P., 203
31, 32, 36, 37, 38, 39, 40, 41, 43, 44,
47, 48, 49, 50, 51, 52, 53, 55, 58, 59, MacGahan, J.A., 31, 32, 77, 133, 141, 142,
63, 64, 65, 67, 68, 70, 71, 72, 73, 74, 185, 186, 187, 188, 189, 191, 203, 205,
76, 77, 78, 83, 86, 89, 118, 119, 120, 211, 212
122, 124, 127, 131, 132, 133, 140, 141, MacKenzie, C.F., 175
167, 169, 182, 183, 185, 186, 187, 190, mamālı̄k-i khāssa, 121
˙˙
191, 192, 193, 195, 196, 197, 201, 203, mamlūk, 121
204, 206, 210, 211, 212 Mangyshlaq, 192, 196
Khivaqābād, 19 Manumission, deeds of, 84, 109, 118
Khokhlov, Ivan, 48 Marchaq, 94
Khrisanf (Orthodox priest), 27 Maris, 52
Khudāyār Khan, 104 Markozov, V.I., 199–203, 206
Khurasan, 14, 16, 19, 23, 25, 33, 34, 35, 36, Marvin, Charles, 14
37, 38, 40, 54, 57, 59, 64, 85, 164, 173, Mashhad, 14, 15, 16, 33, 34, 41, 42, 46, 67,
183, 193, 210 75, 89, 90, 91, 95, 96, 98, 110, 111,
Khwaja-Eli, 124, 186 113, 121, 137, 146
Khwarazm, viii, 13, 14, 16, 21, 22, 23, 25, Mazandaran, 22, 174, 175
26, 28, 30, 36, 38, 39, 41, 42, 43, 46, McNeil, Sir John, 174
47, 49, 50, 52, 54, 55, 56, 59, 70, 76, Merv, 14, 15, 28, 37, 38, 39, 45, 63, 88, 110,
77, 112, 119, 121, 128, 133, 135, 137, 111, 183, 184, 196, 197, 198, 204, 206
138, 140, 166, 173, 183, 185, 187, 188, Meyendorff, Baron, 16, 21, 22, 28, 31, 59,
189, 190, 192, 193, 194, 195, 197, 199, 119, 123, 124, 126, 127, 138
203, 206, 207, 208, 209, 211, 212 Middle Horde, 68, 130, 131
abolition of slavery in, 190, 206 Mı̄rpanja, Ismāʿı̄l Sarhang, 22
relations with Kazakhs, 21 Mordvins, 52
relations with Turkmens, 21, 37, 38 Morier, James Justinian, 15
Russian missions to, 52 Mudabbar, 117
Russian perceptions of, 21, 71 Muhammad Amı̄n Khan, 39, 40
˙
slave uprising of 1873, 185, 186, 187, Muhammad Rahim Khan, 188
˙ ˙
188, 206, 208 Muhammad Walῑ Khan, 46
˙
slave-owning among elites, 21, 119 Mukātab, 116
statistics on slavery in, 26, 59, 205 Murav’ev, N.N., 22, 23, 27, 28, 29, 30, 31,
Kirghiz, 61 50, 51, 58, 59, 63, 64, 65, 71, 119, 122,
Kiyat Khan, 174, 181 124, 133, 140, 141
petition to Russia, 181–182 Muravin, I., 37
Kizil Arvat, 203 Murtaża Biy Khwāja, 186
 Index 225

Muscovy Company, 48 abolitionism in, 144, 152, 161, 183, 208

Muzaffar al-Dı̄n, 134 compensating former slaves, 172–173
compensating slaveowners, 171–172
Nādir Shah, 19, 37, 71 conquest of Bukhara, 110
conquest of Khiva, 37 conversion of slaves to Christianity, 145,
Najaf ʿAlı̄ Khan, 19 146, 149, 150, 151
Nasrullāh Khan, 39 converting slaves into serfs, 152
˙
Nesselrode, Karl, 50 dealings with Ergeld Khan, 176–180
New York Times, 186 diplomacy with Central Asia, 13, 41,
Nicholas I (Tsar), 50 47–54
Nikoforov (Russian ambassador), 43–44 fugitives from, 147
Noghays, 48, 78 indemnities against Turkmens, 193–194,
Novo-Petrovsk, 196 204
Nūr Muhammad Makhdūm, 210 manumission decrees in the steppe, 171
˙
mission to free Stepanov, 165–168
Orenburg, 129, 132, 148, 150, ownership of slaves by Russians, 152
164, 187 perceived problems of diplomacy with
Orenburg Border Commission, 50, 60, 64, nomads, 20
67, 129, 144–162, 163, 169 perceptions of nomadic slave raiding, 19
Orsk, 75, 186 placing former slaves in foster care, 161
Ottoman Empire, 56, 59, 60, 134 policy on slavery in the Kazakh steppe,
145, 152, 154–155, 161–162
Panjdih, 93, 94, 95, 96, 97, 100, 101, 102, protection of Turkmens raiding Iran, 175
112, 113 punishment of slaveowners, 171
Panjob (neighborhood in Samarqand), repatriating former slaves to Iran,
212 147–149
Pars, 67, 86 return of escaped slaves to Kazakh owners,
Pazukhin, Boris and Semen, 49 145
peculium, 116 rights over the Caspian Sea, 174
Perovskii expedition of 1839–40, use of Kazakh informants, 163
51, 205 use of Turkmen informants, 174–182
Perovskii, General V.A., 50, 51 Russkii invalid, 186
Petro-Alexandrovsk, 192 Rzhevuskii, A., 135
Petrusevich, N.G., 15
Poles, 78 Saʿadi, 99
Pugachev Rebellion, 52 Sabzavar, 14, 146
Safavids, 33
Qajar Dynasty, 13, 14, 15, 16, 19, 37–38, 39, Salars, 37, 82, 93, 94, 97, 196, 197
40, 41, 42, 46, 47, 88, 183, 184, 197, Samarqand, 36, 83, 103, 212
198, 199, 201, 202, 219 Sar Teppe, 105
Qaradashli, 38 Sarakhs, 20, 37, 38, 90, 111, 184, 196, 197
Qaraqul, 89 Sarapul, 146, 149, 150, 151
Qarshi, 75, 80 Sarı̈qs, 23, 38, 89, 91, 98, 196, 197
Qongrat, 17 Schmidt, Emil, 59, 76, 185, 187, 190, 192
Quchan, 37 Schuyler, Eugene, 80–84, 188, 205, 211
qullar dastasi, 121 serfdom, 149–152
qullar nökari, 121 Shah ʿAbbās I, 33, 36
qullar sipāhi, 121 Shah Ismaʿı̄l, 34
Shahrisabz, 36
Rumi, 96 Shı̄r Ghāzı̄ Khan, 121
Russian Empire Shirgen, 89
226 Index

Shubinskii, P., 78, 122, 124 changing names of, 30

Siberia, 56, 168, 170, 178, 210 conversion of, 49, 124
Sistan, 23 cultural influence of, 125
Skobelev, General M.D., 204, 206 enslaved children, 136, 137, 138, 141,
Slave trade, 57–59, 65 152–154, 156–158, 161, 173
after 1873, 76, 79, 83, 206, 210, 211 enslaved women, 17, 29, 117, 137
among nomads, 57, 64, 66–67, 69, 85 freed after owner’s death, 117, 128
bazaars of Bukhara, 76, 80 given as gifts, 116
bazaars of Khwarazm, 71–73 in administrative roles, 26
decline of, 76 in military roles, 26, 37, 110, 122, 207
factors impacting the price of slaves, 29 insanity among, 142
fluctuation of, 23, 183 Iranian, 26, 33, 37, 38–47, 59, 75, 82, 85,
grain as a part of, 25 98, 100, 108, 113, 119, 122, 132, 134,
slaves traded for livestock, 29 138, 144, 146, 148, 154, 173, 175, 200,
tax revenue from, 22 211
Slave-raiding, 20, 26, 37, 41, 63, 67, 95, jobs after manumission, 128
141, 165, 174, 184, 194 marriage, 115, 128, 133, 141, 173
as a routine aspect of warfare, 13–19 marrying free-born brides, 138
Slavery marrying free-born husbands, 138
among Kazakhs, 139 meager rations, 30
among nomads, 70, 85, 120, 138, owned by rulers, 24, 78, 120, 132, 134
144–162 performing illegal work, 127
and agriculture, 25, 35 price of, 23, 29, 42, 63, 64, 66, 68, 82,
escape from, 95, 137, 141–142, 144, 100, 103, 127, 153, 156, 157, 184
145–147, 152, 206 ransom of, 13, 18, 22, 23, 53, 89, 165,
rural character of, 58 180
Slaves reasons for marriage among, 138
ability to work multiple jobs, 126 rented out, 116
and agriculture, 26, 118–120, 128, 132, repatriation of, 23
210 Russian, 26, 27, 47, 51, 53, 76, 77, 120
and animal husbandry, 26, 91, 95, 96, 97, sexual exploitation of, 17, 90, 117, 126,
120 129, 132, 136, 137
as backsmiths, 26 treatment of, 30–33, 64, 89, 90, 130
as blacksmiths, 122 work at shrines, 125
as bodyguards, 121 work on canals, 27, 120
as carpenters, 26, 123 Stavropol, 131, 146, 151
as craftsmen, 123 Stremoukhov, N.P., 78, 79, 122, 136
as engineers, 126 Sultan Husayn Mı̄rzā, 35
˙
as go-betweens for their masters, 126 Syn otechestva, 71
as government administrators, 124, 126
as maids, 135 Tatars, 61, 78, 132, 150
as metallurgists, 123 Taze Qongrat, 38
as miners, 127–128 Tehran, 25, 88, 96, 111, 147, 174, 202
as musicians, 26, 126 Tekes, 14, 15, 30, 36, 63, 67, 72, 73, 120,
as part of dowry, 116, 117 125, 135, 136, 181, 183, 184, 191, 196,
as shoemakers, 123 202, 203, 204
as teachers, 26, 125 Thomson, William Taylour, 44
as widows, 129–133 Timurid era, 34, 35
barred from work as imams, 124 toghma, 121
buying their own freedom, 22, 24, 26, Topolinsk, 164
117, 133 Treaty of Turkmenchay, 174
 Index 227

Tugan-Mirza-Baranovskii, V.A., 30, 72, 73, UNESCO, 85

120 Ural River, 164
Turbat, 37 Ural’sk, 129, 130, 146, 165
Turkmens, 13, 14, 15, 16, 17, 19, 21, 23, 25, Urgench, 63, 64, 67, 68, 86, 89
29, 30, 34, 36, 37, 38, 41, 45, 47, 52, Uzbeks, 33, 34, 35, 36, 37, 57, 61, 72, 79,
57, 61, 62, 63, 64, 65, 66, 67, 69, 70, 111, 119, 122, 123, 148, 212
72, 79, 82, 86, 88, 89, 91, 93, 94, 96,
97, 110, 111, 112, 114, 119, 120, vakı̄l, 126
135–136, 137, 138, 139, 146, 156, Vambery, Arminius, 18, 19, 20, 22, 23, 24,
163, 172, 174, 175, 177, 178, 179, 180, 32, 34, 59, 72, 74, 75, 80, 116, 117,
181, 183, 190, 191, 192, 193, 194, 195, 133, 134, 135, 141, 168, 169, 196
196, 197, 198, 199, 200, 201, 202, 203, Verevkin, General N.A., 188
204, 210
as captives, 15 Witkiewicz, Jan Prosper, 27, 29, 74, 77,
as informants, 174–182 120, 122, 123, 133, 166, 167, 168, 169,
compared to pirates, 19 177
culture of slave-raiding among, 194 Wolff, Joseph, 20, 26, 39, 59, 133, 140
inadequacy of sources on enslaved women Wood, Herbert, 183
among, 136 Wood, John, 136
massacre of freed slaves, 193
raiding along Caspian coast, 174 Yemreli, 38
Russian indemnities against, 193–194, Yomuts, 14, 17, 21, 31, 36, 38, 72, 136, 174,
204 181, 188, 191, 192, 193, 196, 197, 198,
Russian perceptions of, 15, 20 199, 200, 202, 203, 204, 206
Turshiz, 37
Zakhar’in, I.N., 31, 51, 131, 132, 135, 142
ʿUbaydullāh Khan, 34 Zalesov, Nikolai, 31, 32, 41, 44, 45, 46, 58,
Ukrainians, 78 71, 72, 73, 119, 121, 122, 137
ʿUlaq Bābā, shrine of, 111 Zarkharı̄d, 116
Umm walad, 117, 128, 130 Zelenin, G.N., 131, 132, 135
 Other titles in the series

Agricultural Innovation in the Early Islamic World: The Diffusion of Crops and
Farming Techniques, 700–1100, Andrew M. Watson
Muslim Tradition: Studies in Chronology, Provenance and Authorship of Early
Hadith, G. H. A. Juynboll
Social History of Timbuktu: The Role of Muslim Scholars and Notables
1400–1900, Elias N. Saad
Sex and Society in Islam: Birth Control before the Nineteenth Century, B.F.
Musallam
Towns and Townsmen of Ottoman Anatolia: Trade, Crafts and Food Production in
an Urban Setting 1520–1650, Suraiya Faroqhi
Unlawful Gain and Legitimate Profit in Islamic Law: Riba, Gharar and Islamic
Banking, Nabil A. Saleh
Men of Modest Substance: House Owners and House Property in Seventeenth-
Century Ankara and Kayseri, Suraiya Faroqhi
Roman, Provincial and Islamic Law: The Origins of the Islamic Patronate,
Patricia Crone
Economic Life in Ottoman Jerusalem, Amnon Cohen
Mannerism in Arabic Poetry: A Structural Analysis of Selected Texts (3rd Century
AH/9th Century AD–5th Century AH/11th Century AD), Stefan Sperl
The Rise and Rule of Tamerlane, Beatrice Forbes Manz
Popular Culture in Medieval Cairo, Boaz Shoshan
Early Philosophical Shiism: The Ismaili Neoplatonism of Abu Yaʿqub Al-Sijistani,
Paul E. Walker
Indian Merchants and Eurasian Trade, 1600–1750, Stephen Frederic Dale
Palestinian Peasants and Ottoman Officials: Rural Administration around
Sixteenth-Century Jerusalem, Amy Singer
Arabic Historical Thought in the Classical Period, Tarif Khalidi
Mongols and Mamluks: The Mamluk-Ilkhanid War, 1260–1281, Reuven Amitai-
Preiss
Knowledge and Social Practice in Medieval Damascus, 1190–1350,
Michael Chamberlain
The Politics of Households in Ottoman Egypt: The Rise of the Qazdağlis,
Jane Hathaway
Hierarchy and Egalitarianism in Islamic Thought, Louise Marlow
Commodity and Exchange in the Mongol Empire: A Cultural History of Islamic
Textiles, Thomas T. Allsen
State and Provincial Society in the Ottoman Empire: Mosul, 1540–1834, Dina
Rizk Khoury
The Mamluks in Egyptian Politics and Society, Thomas Philipp and
Ulrich Haarmann (eds.)
The Delhi Sultanate: A Political and Military History, Peter Jackson
European and Islamic Trade in the Early Ottoman State: The Merchants of Genoa
and Turkey, Kate Fleet
The Ottoman City between East and West: Aleppo, Izmir, and Istanbul,
Edhem Eldem, Daniel Goffman, and Bruce Masters
The Politics of Trade in Safavid Iran: Silk for Silver, 1600–1730, Rudolph P. Matthee
The Idea of Idolatry and the Emergence of Islam: From Polemic to History,
G. R. Hawting
A Monetary History of the Ottoman Empire, Şevket Pamuk
Classical Arabic Biography: The Heirs of the Prophets in the Age of Al-Ma’mun,
Michael Cooperson
Empire and Elites after the Muslim Conquest: The Transformation of Northern
Mesopotamia, Chase F. Robinson
Poverty and Charity in Medieval Islam: Mamluk Egypt, 1250–1517, Adam Sabra
Culture and Conquest in Mongol Eurasia, Thomas T. Allsen
Christians and Jews in the Ottoman Arab World: The Roots of Sectarianism,
Bruce Masters
Arabic Administration in Norman Sicily: The Royal Diwan, Jeremy Johns
Law, Society and Culture in the Maghrib, 1300–1500, David S. Powers
Revival and Reform in Islam: The Legacy of Muhammad al-Shawkani,
Bernard Haykel
Tolerance and Coercion in Islam: Interfaith Relations in the Muslim Tradition,
Yohanan Friedmann
Guns for the Sultan: Military Power and the Weapons Industry in the Ottoman
Empire, Gábor Ágoston
Marriage, Money and Divorce in Medieval Islamic Society, Yossef Rapoport
The Empire of the Qara Khitai in Eurasian History: Between China and the Islamic
World, Michal Biran
Domesticity and Power in the Early Mughal World, Ruby Lal
Power, Politics and Religion in Timurid Iran, Beatrice Forbes Manz
Postal Systems in the Pre-Modern Islamic World, Adam J. Silverstein
Kingship and Ideology in the Islamic and Mongol Worlds, Anne F. Broadbridge
Justice, Punishment and the Medieval Muslim Imagination, Christian Lange
The Shiites of Lebanon under Ottoman Rule, 1516–1788, Stefan Winter
Women and Slavery in the Late Ottoman Empire, Madeline Zilfi
The Second Ottoman Empire: Political and Social Transformation in the Early
Modern World, Baki Tezcan
The Legendary Biographies of Tamerlane: Islam and Heroic Apocrypha in Central Asia,
Ron Sela
Non-Muslims in the Early Islamic Empire: From Surrender to Coexistence,
Milka Levy-Rubin
The Origins of the Shi‘a: Identity, Ritual, and Sacred Space in Eighth-Century Kufa,
Najam Haider
Politics, Law, and Community in Islamic Thought: The Taymiyyan Moment,
Ovamir Anjum
The Power of Oratory in the Medieval Muslim World, Linda G. Jones
Animals in the Qur’an, Sarra Tlili
The Logic of Law Making in Islam: Women and Prayer in the Legal Tradition,
Behnam Sadeghi
Empire and Power in the Reign of Süleyman: Narrating the Sixteenth-Century
Ottoman World, Kaya Şahin
Law and Piety in Medieval Islam, Megan H. Reid
Women and the Transmission of Religious Knowledge in Islam, Asma Sayeed
The New Muslims of Post-Conquest Iran: Tradition, Memory, and Conversion,
Sarah Bowen Savant
The Mamluk City in the Middle East: History, Culture, and the Urban Landscape,
Nimrod Luz
Disability in the Ottoman Arab World, 1500–1800, Sara Scalenghe
The Holy City of Medina: Sacred Space in Early Islamic Arabia, Harry Munt
Muslim Midwives: The Craft of Birthing in the Premodern Middle East,
Avner Giladi
Doubt in Islamic Law: A History of Legal Maxims, Interpretation, and Islamic
Criminal Law, Intisar A. Rabb
The Second Formation of Islamic Law: The Hanafi School in the Early Modern
Ottoman Empire, Guy Burak
Sexual Violation in Islamic Law: Substance, Evidence, and Procedure, Hina Azam
Gender Hierarchy in the Qur’an: Medieval Interpretations, Modern Responses,
Karen Bauer
Intellectual Networks in Timurid Iran: Sharaf al-Din ‘Ali Yazdi and the Islamicate
Republic of Letters, Ilker Evrim Binbaş
Authority and Identity in Medieval Islamic Historiography: Persian Histories from
the Peripheries, Mimi Hanaoka
The Economics of Ottoman Justice: Settlement and Trial in the Sharia Courts,
Metin Coşgel and Boğaç Ergene
The Mystics of al-Andalus: Ibn Barrajan and Islamic Thought in the Twelfth
Century, Yousef Casewit
Muhammad’s Heirs: The Rise of Muslim Scholarly Communities, 622–950,
Jonathan E. Brockopp
The First of the Modern Ottomans: The Intellectual History of Ahmed Vasif,
Ethan Menchinger
Non-Muslim Provinces under Early Islam: Islamic Rule and Iranian Legitimacy in
Armenia and Caucasian Albania, Alison Vacca
Women and the Making of the Mongol Empire, Anne F. Broadbridge
Slavery and Empire in Central Asia, Jeff Eden
Christianity in Fifteenth-Century Iraq, Thomas A. Carlson