ISBN 978-92-3-103467-1 Contents CONTENTS

Contents

PREFACE 11

DESCRIPTION OF THE PROJECT 14

MEMBERS OF THE INTERNATIONAL SCIENTIFIC COMMITTEE

(in alphabetical order) 17

MEMBERS OF THE INTERNATIONAL SCIENTIFIC COMMITTEE (since

1993) 18

MEMBERS OF THE READING COMMITTEE 19

LIST OF CONTRIBUTORS 20

INTRODUCTION 22

MUHAMMAD SEYFEYDINOVICH ASIMOV

(ABIMI) IN PIAM MEMORIAM 26

1 CENTRAL ASIA UNDER THE UMAYYADS 27

THE APPEARANCE OF THE ARABS IN CENTRAL ASIA UNDER THE
UMAYYADS AND THE ESTABLISHMENT OF ISLAM 28
Central Asia on the eve of the Arab incursions . . . . . 28
The appearance of the Arabs . . . . . . . . . . . . . . . . 29
CENTRAL ASIA UNDER THE EARLY c ABBASIDS .. 30
The course of the c Abbasid revolution and its significance . 30
The aftermath of the c Abbasid revolution and the fall of Abu Muslim . 36
The consolidation of c Abbasid power . . . . . . . . . . . . . . . . 39
Political, social and sectarian dissent in the early c Abbasid period . 41
The achievement of a degree of stability under al-Ma'mun 45

2
ISBN 978-92-3-103467-1 Contents CONTENTS

2 SECTARIAN AND NATIONAL MOVEMENTS 47

The Kharijite movement . . . . . . . . . . . . . . 48
Currents of Shicism: the Kaysaniyya and the Hashimiyya 51
Heterodox Muslim and neo-Mazdakite movements: al-Muqannac, Babak, and
others . . . . . . . . . . . . 53
The later development of Shfism:
the Twelvers, the Zaydis and the Ismacilis 57
The beginnings of the disintegration of the cAbbasid caliphate in the east . 63

3 THE STATES OF THE OGHUZ 66

The Oghuz .. 66
The Kimek . 74
The Kipchak . 77

4 THE SAMANID STATE 83

The creation of the Samanid state . 84
The system of government . . . . 86
The nature of political authority under the Samanids . 87
Agriculture 88
Mining .. . 89
Crafts .. . 90
Domestic and external trade . 90
Material culture 92
Intellectual life 93
Central Asia and the Ismacili movement 100
The ethnic composition of the Samanid state and the creation of an Eastern
Persian-Tajik ethnic identity 101

5 THE GHAZNAVIDS 102

The prehistory of the Ghaznavids . 103
The Establishment of Sebtiktegin in Ghazna 104
The succession of Mahmud . . . . . . . 106
The zenith of the empire under Mahmud 107
Mascud and the Seljuqs . . . . . . . . . 113
The nature and structure of the Ghaznavid state 117
Cultural and intellectual life . 121
Agriculture and trade . . . . 122

3
ISBN 978-92-3-103467-1 Contents CONTENTS

6 THE KARAKHANIDS 125

Sources for Karakhanid history . 125
The conquest of Transoxania by the Karakhanids: the division into appanages 126
The two Karakhanid Khanates; the policy of Ibrahim b. Nasr Tamghach Khan 133
Karakhanids, Seljuqs and Kara Khitay . . . . . . . . . . . . . . 138
The Karakhanids and the Khwarazm Shah Muhammad b. Tekish 141
lqtiics and the structure of the state 143
Towns and trade . . . . . . . . . . . . . . . . . . . . 147

7 THE SELJUQS AND THE KHWARAZM SHAHS 150

THE ORIGINS OF THE SELJUQS AND THE ESTABLISHMENT OF SELJUQ
POWER IN THE ISLAMIC LANDS UP TO 1055 . . . . . . . 151
THE CONSOLIDATION OF THE SELJUQ SULTANATE IN IRAN . 161
Historical survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
The structure of the Seljuq state in the east . . . . . . . . . . . . . . . 165
THE EASTERN SELJUQ SULTANATE (1118-57) AND THE RISE AND
FLORESCENCE OF THE KHWARAZM SHAHS OF ANUSHTEGIN'S
LINE UP TO THE APPEARANCE OF THE MONGOLS (1097- 1219) 167
Historical survey . . . . . . . . . . . . 167
The structure of the eastern Seljuq state 178
The structure of the Khwarazmian state 179

8 THE GHURIDS 182

The region of Ghur and the beginnings of Islamization 182
The rise of the Ghurids as an independent power . 185
The apogee of the Ghurid sultanate . . . 186
The Bamiyan amirate . . . . . . . . . . . . . . . 189
The Ghurid sultanate as a world power . . . . . . 190
The political and social organization of the Ghurid state . 191
Cultural developments . . . . . . . . . . . . . . . . . . 194

9 THE UIGHURS, THE KYRGYZ AND THE TANGUT (EIGHTH TO THE

THIRTEENTH CENTURY) 196
THE UIGHURS IN MONGOLIA AND THE KYRGYZ 196
THE UIGHUR KINGDOM OF KOCHO . . . . . . 206
THE TANGUT HSI HSIA KINGDOM (982-1227) 212

4
ISBN 978-92-3-103467-1 Contents CONTENTS

10 THE WESTERN HIMALAYAN STATES 221

The Trakhan dynasty of Gilgit . . 222
The Maglot ruling family of Nager 225
The Ayash ruling family of Hunza 225
The Kator royal family of Chitral . 225
Baltistan . . . . . . . . . . . . . . 225
Relations with Tibet, Kashgharia and the trans-Pamir regions 227
Relations with Kashmir . . . . . . . . . . 227
Long-term socio-economic developments 228
Socio-religious developments . . . . . . . 231

11 THE KITAN AND THE KARA KHITAY 232

The rise of the Kitan . . . 232
The rise of the Kara Khitay . . . . . . . . . 239

12 THE MONGOLS AND THEIR STATE IN THE TWELFTH TO THE THIR-

TEENTH CENTURY 248
The socio-economic and political situation . . . . . . . . . . . . . . . . . . . 249
Chinggis Khan and the founding of the Mongol state . . . . . . . . . . . . . 251
Chinggis Khan's campaigns of conquest: The foundation of the Mongol empire 257
The Mongol empire during the reign of Chinggis Khanr's successors 259
The Yuan empire of the Mongols and its fall . . . . . . . . . . 264

13 CENTRAL ASIA UNDER THE RULE OF SUCCESSORS 265

14 THE DELHI SULTANATE 273

THE ESTABLISHMENT OF THE SULTANATE IN THE THIRTEENTH
CENTURY AND THE NATURE OF THE NEW STRUCTURES IN
INDIA . . . . . . . . . . 274
Background . . . . . . . . . . . . 27 4
Qutb al-Din Aybak (1206- 10) . . 275
Shams al-Din Iltutmish (1210-36) 275
Nasir al-Din Mahmud (1246-66) and Ghiyath al-Din Balban (Ulugh Khan)
(1266-87) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
The end of Turkish supremacy . . . . . . . . . . . . . . . . . . . . . . . 277
Jalal al-Din Rruz Khalji (1290- 6) and c Ala' al-Din Khalji (1296- 1316) . 277
Income levels among the ruling and scholarly elites 281

5
ISBN 978-92-3-103467-1 Contents CONTENTS

Agrarian conditions in the fourteenth century 282

The political structure of the state . . . . 283
Social and economic developments . . . 283
THE DELHI SULTANATE, 1316-1526 284
The Tughluqids (1320- 1412) 284
The Sayyids (1414-51) 291
The Lodis (1451- 1526) . . . 293

15 THE REGIONS OF SIND, BALUCHISTAN, MULTAN AND KASHMIR:

THE HISTORICAL, SOCIAL AND ECONOMIC SETTING 297
THE RULERS OF SIND, BALUCHISTAN AND MULTAN (750- 1500) . . . 298
The c Abbasid period and the Fatimid interlude (mid-eighth to the end of the
tenth century) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
The Period of the Ghaznavid and Ghurid Sultanates (eleventh and twelfth cen-
turies) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
The era of the local independent states . . . . . . . . . . . . . . . . . . 304
KASHMIR UNDER THE SULTANS OF THE SHAH MIR DYNASTY 310

16 CENTRAL ASIA TIMUR 1370 323

Formation of Timur's empire . . 325
The Sarbadar movement . . . . 326
Balkh and Samarkand under Timur . 327
Reorganization of Timur's army . . 330
Timur's military campaigns . . . . . 332
Socio-economic conditions under Timur 339
Urban development, crafts and trade 342
Relations with west European rulers 345
The succession struggle . . . . . . . 346

17 THE TIMURID STATES IN THE FIFTEENTH AND SIXTEENTH CEN-

TURIES 350
The aftermath of Timur's death . . . . . . . . . . 350
The rule of Ulugh Beg . . . . . . . . . . . . . . 351
The power struggle following Shah Rukh's death 353
The realm divides . . . . . . . . . 354
Agriculture, livestock and hunting 356
Landownership and taxation . . . 357

6
ISBN 978-92-3-103467-1 Contents CONTENTS

Towns, handicrafts and trade . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Revolts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
The Dasht-i Kipchak at the end of the fifteenth century and the beginning of
the sixteenth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

18 POPULAR MOVEMENTS, RELIGIOUS TRENDS AND SUFI INFLU-

ENCE ON THE MASSES IN THE POST-c ABBASID PERIOD 367
Religious groups of the Middle East in early Islamic times 367
The rise of Sufism and the Sufi orders in Central Asia 370
Socio-religious and politico-religious movements 376
Sufi orders in India . . . 379
The Mahdawi movement . . . . . . . . . . . . . 382

19 SOCIO-ECONOMIC DEVELOPMENT: FOOD AND CLOTHING IN EAST-

ERN IRAN AND CENTRAL ASIA 383
THE EASTERN ISLAMIC LANDS, FROM IRAN TO THE FRONTIERS
WITH CHINA 384
Food and diet 384
Dress .... 388
MONGOLIA 390
Shelter, crafts and dress 390
Food and diet . . . . . 393

20 COINAGE AND THE MONETARY SYSTEM 395

CENTRAL ASIA . . . . . . . . . . . . . . . . . 395
Coinage and the circulation of money from the eighth to the tenth century . . 396
Coinage and the circulation of money from the eleventh century to the begin-
ning of the thirteenth . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
Coinage and the circulation of money under the Mongols (thirteenth and four-
teenth centuries) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
Coinage and the circulation of money in Transoxania under Timur and the
Timurids (late fourteenth and fifteenth centuries) . . 412
AFGHANISTAN, PAKISTAN AND NORTHERN INDIA 417

CONCLUSION 425

MAPS 426

7
ISBN 978-92-3-103467-1 Contents CONTENTS

BIBLIOGRAPHY AND REFERENCES 439

GLOSSARY 464

INDEX 466

History of civilizations of Central Asia 474

8
ISBN 978-92-3-103467-1 Contents History of Civilizations of Central Asia

History of Civilizations of Central Asia

The age of achievement:

A.O. 750 to the end of the fifteenth century

Volume IV

Part One
The historical, social and economic setting

Editors: M. S. Asimov and C. E. Bosworth

Multiple History Series

UNESCO Publishing

9
ISBN 978-92-3-103467-1 Contents COPYRIGHT

The authors are responsible for the choice and the presentation of the facts contained in
this book and for the opinions expressed therein, which are not necessarily those of
UNESCO and do not commit the Organization.

The designations employed and the presentation of material throughout this publication
do not imply the expression of any opinion whatsoever on the part of UNESCO
concerning the legal status of any country, territory, city or area or of its authorities, or
concerning the delimitation of its frontiers or boundaries.

Compiled by I. Iskender-Mochiri
English text edited by Jana Gough
Composed by Editions du Moufion, 94270 Le Kremlin-Bicetre (France)
Printed by Imprimerie Darantiere, 21800 Quetigny (France)

Published in 1998 by the United Nations Educational, Scientific and Cultural

Organization 7 place de Fontenoy, 75352 Paris 07 SP

ISBN 92-3-103467-7

©UNESCO 1998

10
ISBN 978-92-3-103467-1 Contents PREFACE

PREFACE
Federico Mayor
Director-General of UNESCO

ONE of the purposes of UNESCO, as proclaimed in its Constitution, is 'to develop and
to increase the means of communication between ... peoples and to employ these means
for the purposes of mutual understanding and a truer and more perfect knowledge of each
other's lives'. The History of the Scientijic and Cultural Development of Mankind, pub-
lished in 1968, was a major early response on the part of UNESCO to the task of enabling
the peoples of the world to have a keener sense of their collective destiny by highlighting
their individual contributions to the history of humanity. This universal history - itself now
undergoing a fundamental revision - has been followed by a number of regional projects,
including the General History of Africa and the planned volumes on Latin America, the
Caribbean and on aspects of Islamic culture. The History of Civilizations of Central Asia
is an integral part of this wider enterprise.
It is appropriate that the second of UNESCO's regional histories should be concerned
with Central Asia. For, like Africa, Central Asia is a region whose cultural heritage has
tended to be excluded from the main focus of historical attention. Yet from time immemo-
rial the area has served as the generator of population movements within the Eurasian
land-mass. The history of the ancient and medieval worlds, in particular, was shaped to an
important extent by the succession of peoples that arose out of the steppe, desert, oases and
mountain ranges of this vast area extending from the Caspian Sea to the high plateaux of
Mongolia. From the Cimmerians mentioned in Homer's Odyssey,the Scythians described
by Herodotus, the Hsiung-nu whose incursions led the emperors of China to build the Great
Wall, the sixth-century Turks who extended their empire to the boundaries of Byzantium,
the Kitan who gave their name to ancient Cathay, through to the Mongols who erupted
into world history in the thirteenth century under Genghis Khan, the nomadic horsemen
of Central Asia helped to define the limits and test the mettle of the great civilizations of
Europe and Asia.

11
ISBN 978-92-3-103467-1 Contents PREFACE

Nor is it sufficient to identify the peoples of Central Asia simply with nomadic cultures.
This is to ignore the complex symbiosis within Central Asia itself between nomadism
and settlement, between pastoralists and agriculturalists. It is to overlook above all the
burgeoning of the great cities of Central Asia such as Samarkand, Bukhara and Khiva,
which established themselves in the late Middle Ages as outstanding centres of intellectual
inquiry and artistic creation. The seminal writings of the philosopher-scientist Avicenna
(a native of Bukhara) and the timeless masterpieces of Timurid architecture epitomize the
flowering of medieval culture in the steppes and deserts of Central Asia.
The civilizations of Central Asia did not, of course, develop in a vacuum. The impact
of Islam was pervasive and fundamental. The great civilizations on the periphery of the
Eurasian continent likewise exerted an important influence on these lands. For some 1,500
years this arid inland sea - far removed from the earth's true oceans - was crucial as the
route along which merchandise (notably silk) and ideas flowed between China, India, Iran
and Europe. The influence of Iran - although the core of its civilization lies in South-West
Asia - was particularly strong, to the extent that it is sometimes difficult to establish a clear
boundary between the civilization of the Iranian motherland and that of the outlying lands
of Central Asia.
To the rich variety of peoples of Central Asia was thus added a multiplicity of external
influences. For century after century, the region experienced the influx of foreign art and
ideas, colliding and merging with the indigenous patterns of Central Asia. Migrations and
the recurrent shock of military invasion, mingling and displacing peoples and cultures,
combined to maintain the vast region in flux.
The systole and diastole of population movements down the ages add to the difficulty of
delimiting a region whose topology alone does not prescribe clear boundaries. Thus, when,
at the nineteenth session of its General Conference, UNESCO decided to embark on a His-
tory of Civilizations of Central Asia the first problem to be resolved was to define the scope
of the region concerned. Subsequently, at a UNESCO meeting held in 1978, it was agreed
that the study on Central Asia should deal with the civilizations of Afghanistan, north-
eastern Iran, Pakistan, northern India, western China, Mongolia and the former Soviet
Central Asian republics. The appellation 'Central Asia', as employed in this History, refers
to this area, which corresponds to a clearly discernible cultural and historical reality.
UNESCO's call to specialists, and particularly to scholars native to the region, to partic-
ipate in the undertaking met with a wide and generous response. The project was deemed
by academics to be an excellent opportunity to draw back the curtain that had veiled Cen-
tral Asia for so long. However, none were in any doubt as to the huge dimensions of the
task.

12
ISBN 978-92-3-103467-1 Contents PREFACE

An ad hoe International Scientific Committee was formed in 1980 to plan and prepare
the work, which it was agreed should cover, in six volumes, the history of Central Asia
from earliest times to the present day. The Committee's initial task was to decide where
pre-eminence should be given in the very wide canvas before it. In due course, a proper
balance was struck and teams of editors and authors were selected.
The preparation of the History of Civilizations of Central Asia is now well advanced.
The best resources of research and archaeology have been used to make the work as thor-
ough as possible, and countless annals consulted in major centres throughout the region. It
is my sincere wish that this, the fourth volume, and those that follow will bring instruction
and pleasure to readers all over the world.
It remains for me to thank the President, Rapporteur and members of the International
Scientific Committee, and the editors, authors and teams of specialists who have collabo-
rated to shed new light on Central Asia with this detailed account of its vital and stirring
past. I am sure it will prove a notable contribution to the study and mutual appreciation of
the cultures that are the common heritage of mankind.

13
ISBN 978-92-3-103467-1 Contents DESCRIPTION OF THE PROJECT

DESCRIPTION OF THE PROJECT

M. S. Asimov

THE General Conference of UNESCO, at its nineteenth session (Nairobi, October, Novem-
ber 1976), adopted the resolution which authorized the Director-General to undertake,
among other activities aimed at promoting appreciation and respect for cultural identity,
a new project on the preparation of a History of Civilizations of Central Asia. zThis project
was a natural consequence of a pilot project on the study of Central Asia which was
approved during the fourteenth session of the UNESCO General Conference in Novem-
ber 1966.
The purpose of this pilot project, as it was formulated in the UNESCO programme, was
to make better known the civilizations of the peoples living in the regions of Central Asia
through studies of their archaeology, history, languages and literature. At its initial stage,
the participating Member States included Afghanistan, India, Iran, Pakistan and the former
Soviet Union. Later, Mongolia and China joined the UNESCO Central Asian project, thus
enlarging the area to cover the cultures of Mongolia and the western regions of China.
In this work, Central Asia should be understood as a cultural entity developed in the
course of the long history of civilizations of peoples of the region and the above delimita-
tion should not be taken as rigid boundaries either now or in the future.
In the absence of any existing survey of such large scope which could have served as
a model, UNESCO has had to proceed by stages in this difficult task of presenting an
integrated narrative of complex historical events from earliest times to the present day.
The first stage was designed to obtain better knowledge of the civilizations of Central
Asia by encouraging archaeological and historical research and the study of literature and
the history of science. A new project was therefore launched to promote studies in five
major domains: the archaeology and the history of the Kushan empire, the history of the
arts of Central Asia, the contribution of the peoples of Central Asia to the development of
science, the history of ideas and philosophy, and the literatures of Central Asia.
An International Association for the Study of Cultures of Central Asia (IASCCA), a
non-governmental scholarly organization, was founded on the initiative of the Tajik scholar

14
ISBN 978-92-3-103467-1 Contents DESCRIPTION OF THE PROJECT

B. Gafurov in 1973, assembling scholars of the area for the co-ordination of interdis-
ciplinary studies of their own cultures and the promotion of regional and international
co-operation.
Created under the auspices of UNESCO, the new Association became, from the very
beginning of its activity, the principal consultative body of UNESCO in the implementation
of its programme on the study of Central Asian cultures and the preparation of a History
of Civilizations of Central Asia.
The second stage concentrated on the modern aspects of Central Asian civilizations and
the eastward extension of the geographical boundaries of research in the new programme.
A series of international scholarly conferences and symposia were organized in the coun-
tries of the area to promote studies on Central Asian cultures.
Two meetings of experts, held in 1978 and 1979 at UNESCO Headquarters, concluded
that the project launched in 1967 for the study of cultures of Central Asia had led to con-
siderable progress in research and contributed to strengthening existing institutions in the
countries of the region. The experts consequently advised the Secretariat on the methodol-
ogy and the preparation of the History. On the basis of its recommendations it was decided
that this publication should consist of six volumes covering chronologically the whole his-
tory of Central Asian civilizations ranging from their very inception up to the present.
Furthermore, the experts recommended that the experience acquired by UNESCO during
the preparation of the History of the Scientijic and Cultural Development of Mankind and
of the General History of Africa should also be taken into account by those responsible
for the drafting of the History. As to its presentation, they supported the opinion expressed
by the UNESCO Secretariat that the publication, while being a scholarly work, should be
accessible to a general readership.
Since history constitutes an uninterrupted sequence of events, it was decided not to give
undue emphasis to any specific date. Events preceding or subsequent to those indicated
here are dealt with in each volume whenever their inclusion is justified by the requirements
of scholarship.
The third and final stage consisted of setting up in August 1980 an International Scien-
tific Committee of nineteen members, who sit in a personal capacity, to take reponsibility
for the preparation of the History. The Committee thus created included two scholars from
each of the seven Central Asian countries - Afghanistan, China, India, Islamic Republic of
Iran, Pakistan, Mongolia and the former USSR - and five experts from other countries -
Hungary, Japan, Turkey, the United Kingdom and the United States of America.
The Committee's first session was held at UNESCO Headquarters in December 1980.
Real work on the preparation of the publication of the History of Civilizations of Central

15
ISBN 978-92-3-103467-1 Contents DESCRIPTION OF THE PROJECT

Asia started, in fact, in 1981. It was decided that scholars selected by virtue of their quali-
fications and achievements relating to Central Asian history and culture should ensure the
objective presentation, and also the high scientific and intellectual standard, of this History.
Members of the International Scientific Committee decided that the new project should
correspond to the noble aims and principles of UNESCO and thereby should contribute
to the promotion of mutual understanding and peace between nations. The Committee
followed the recommendation of the experts delineating for the purpose of this work the
geographical area of Central Asia to reflect the common historical and cultural experience.
The first session of the International Committee decided most of the principal matters
concerning the implementation of this complex project, beginning with the drafting of
plans and defining the objectives and methods of work of the Committee itself.
The Bureau of the International Scientific Committee consists of a president, four vice-
presidents and a rapporteur. The Bureau's task is to supervise the execution of the project
between the sessions of the International Scientific Committee. The reading committee,
consisting of four members, was created in 1986 to revise and finalize the manuscripts
after editing Volumes I and II. Another reading committee was constituted in 1989 for
Volumes III and IV.
The authors and editors are scholars from the present twelve countries of Central Asia
and experts from other regions. Thus, this work is the result of the regional and of the inter-
national collaboration of scholars within the framework of the programme of the United
Nations Educational, Scientific and Cultural Organization (UNESCO).
The International Scientific Committee and myself express particular gratitude to Mrs
Irene Iskender-Mochiri for her arduous and selfless work in preparing the volumes for the
press.
It is our sincere hope that the publication of the fourth volume of the History of Civiliza-
tions of Central Asia will be a further step towards the promotion of the cultural identity
of the peoples of Central Asia, strengthening their common cultural heritage and, conse-
quently, will foster a better understanding among the peoples of the world.

16
ISBN 978-92-3-103467-1 Contents MEMBERS OF THE COMMITTEE

MEMBERS OF THE INTERNATIONAL

SCIENTIFIC COMMITTEE
(in alphabetical order)

Dr F. R. ALLCHIN (United Kingdom) Professor Lrn CuNKUAN (People's

Republic of China)
t Professor M. S. AsIMov (Tajikistan)
President Dr L. I. MIROSHNIKOV (Russian Federation)
Editor of Volume IV (Parts One and Two)
Professor S. NATSAGDORJ (Mongolia)
Dr N. A. BALOCH (Pakistan)
t Professor B. N. PURI (India)
Professor M. BASTANI PARIZI (Islamic Co-editor of Volume II
Republic of Iran)
Professor M. H. Z. SAFI (Afghanistan)
Professor Sh. BIRA (Mongolia) Professor A. SAYILI (Turkey)

Professor A. H. DANI (Pakistan) Dr R. SHABAN! SAMGHABADI (Islamic

Editor of Volume I Republic of Iran)
Co-editor of Volume III
t Professor K. ENOKI (Japan)
Professor D. SINOR (United States of
Professor G. F. ETEMADI(Afghanistan)
America)
Co-editor of Volume II
t Professor B. K. THAPAR (India)
Professor J. HARMATTA (Hungary)
Editor of Volume II Professor ZHANG GuANG-DA (People's
Republic of China)
Co-editor of Volume III

17
ISBN 978-92-3-103467-1 Contents MEMBERS OF THE COMMITTEE

MEMBERS OF THE INTERNATIONAL

SCIENTIFIC COMMITTEE (since 1993)

Professor C. ADLE Professor H.-P. FRANCFORT

(Islamic Republic of Iran) (France)
President and Editor of Volume V
Professor I. HABIB
Professor D. ALIMOVA (India)
(Uzbekistan) Editor of Volume V

Professor M. ANNANEPESOV Dr L. MIROSHNIKOV

(Turkmenistan) (Russian Federation)

t Professor M. S. AsIMov Professor M. SHUKUROV

(Tajikistan) (Tajikistan)
President and Editor of Volume IV
Professor D. SINOR
(Parts One and Two)
(United States of America)
Professor K. BAIPAKOV
Dr A. TABYSHALIEVA
(Kazakhstan)
(Kyrgyz Republic)
Co-editor of Volume V
Professor I. ToGAN
Professor Sh. BIRA
(Turkey)
(Mongolia)
Professor H. UMEMURA
Professor A. H. DANI
(Japan)
(Pakistan)
Editor of Volume I Professor Wu YuNGUI
(People's Republic of China)
Professor R. F ARHADI
(Afghanistan)

18
ISBN 978-92-3-103467-1 Contents MEMBERS OF THE COMMITTEE

MEMBERS OF THE READING COMMITTEE

Professor R. N. FRYE Professor D. SINOR

(United States of America) (United States of America)

Professor J. HARMATTA
(Hungary)

19
ISBN 978-92-3-103467-1 Contents LIST OF CONTRIBUTORS

LIST OF CONTRIBUTORS

S. G. AGAJANOV 191041 St. Petersburg

Vargy street, 2, n ° 52 Russian Federation
Moscow
C. E. BOSWORTH
Russian Federation
The University of Manchester
B. AKHMEDOV Department of Middle Eastern Studies
Aybek street, 34 Manchester M13 9PL
Tashkent 700019 United Kingdom
Uzbekistan
F. DAFTARY

K. Z. AsHRAFYAN 77 Hamilton Terrace

Institute of Oriental Studies LondonNW8
UL Rozhdestvenka 12 United Kingdom
Moscow 103777
A. H. DANI
Russian Federation
Dani House
N. A. BALOCH no. 17, st no. 10
12 Sind University (old campus) Shalimar 8/3 Islamabad
Hyderabad, Sind Pakistan
Pakistan
E. A. DAVIDOVICH

Sh. BIRA UL Veshnjakovskaja

International Association for Mongol Dom 6, korp 3, kv 124
Studies Moscow 111402
Central P.O. Box 941 Russian Federation
Ulan Bator-11
R.N. FRYE
Mongolia
Tower Hill Road
0. G. BoLSHAKov Brimfield, MA 01010
Institute of Oriental Studies United States of America
Dvortsovaya N aberezhnaya, 18

20
ISBN 978-92-3-103467-1 Contents LIST OF CONTRIBUTORS

GENG SHIMIN Tajik Academy of Sciences

2-Department of Minority Languages Prospekt Lenin 33
Central Institute for Nationalities Dushanbe 7340025
27, Baishiqiao Road Tajikistan
Beijing
t K. A. NIZAMI
People's Republic of China
A. Q. RAFIQI
N. KASAI
73 Balgarden
College of Theology and Islamic Education
Srinagar, Kashmir
University of Tehran
India
Tehran
Islamic Republic of Iran RIAZUL ISLAM

C-1, staff Town,

Y. I. KYCHANOV
University of Karachi
Institute of Oriental Studies
Karachi 75270
Dvortsovaya N aberezhnaya, 18
Pakistan
191041 St. Petersburg
Russian Federation A. SEVIM

Dil, Tarih-Gografya Faktiltesi

R. G. MUKMINOVA
Tarih Boltimti
Uzbek Academy of Sciences
Ankara
Tashkent
Turkey
Uzbekistan
D. SINOR
S. NATSAGDORJ
Distinguished Professor Emeritus
Mongolian Academy of Sciences
Department of Central Eurasian Studies
Ulan Bator
Goodbody Hall 157
Mongolia
Bloomington, Indiana 47405-2401
N. N. NEGMATOV United States of America
Institute of History

21
ISBN 978-92-3-103467-1 Contents INTRODUCTION

INTRODUCTION
C. E. Bosworth

Volume IV of this History covers some eight centuries, A.D. 750-1500, centuries during
which the new faith of Islam arose, first in Arabia, and then gradually spread eastwards
and northwards, affecting, over a long period of time, much of Central Asia, the southern
fringes of Siberia and the eastern regions of China, competing in these areas with older-
established faiths. These were also centuries in which nomadic and military empires arose
in the heart of Asia and then impinged on the history of adjacent, well-established civiliza-
tions and cultures - China, India, Islamic Western Asia and Christian eastern and central
Europe - to an unparalleled extent.
Previous ethnic and expansionist movements originating in Inner Asia, such as the
Kushans in northern India, and the Huns and Avars in post-Roman eastern and central
Europe, had been limited in their impact. But the expansion of hitherto little-known peo-
ples of what is now the Mongolian region and the lands around Lake Baikal - first the
Turks, then the Kitans and finally the Mongols - was of a wider, international significance.
In the case of the Mongols, it affected two continents, Asia and Europe. The boundaries
of T'ang China held firm against the first Eastern Turk empire, although with the disap-
pearance of that Chinese dynasty in the tenth century, the way was open for the Kitans
to establish themselves in northern China. They were initially regarded by the Chinese as
yet another Barbarian dynasty, but speedily, as usually happened in such cases, became
Sinicized. They were also to have an impact on the Central Asian steppelands (in what
became known as Turkistan and in Moghulistan to its north) as the Kara Khitay dynasty of
the Islamic sources of the twelfth and early thirteenth centuries.
The westwards migrations, from the later tenth century onwards, of Turkic tribal nomads
were to have long-term effects on the ethnic and social composition of the 'northern tier'
of the Middle East, i.e. the lands running westwards from Afghanistan through Iran to
Anatolia. Regions which had been Iranian in population and language since the time of
the lndo-Aryan migrations eastwards into Central Asia in the second millennium B.c. now
felt the impact of Turkicization. The process mainly affected Transoxania and Khwarazm,

22
ISBN 978-92-3-103467-1 Contents INTRODUCTION

although even here, in the cities and towns and in the villages, especially of the valleys
and mountainous uplands running eastwards to the Pamirs and the T'ien Shan mountains,
the Iranian Tajiks preserved much of their language and culture. Hence the New Persian
language of the Tajiks, together with some modem remnants of the Middle Persian Sog-
dian language, provide the cultural and ethnic raison d'etre for the Republic of Tajikistan,
although there remain Tajiks outside the boundaries of this modem republic. Substantial
lands further west, such as Azerbaijan, parts of Fars and, above all, Anatolia, likewise
became Turkicized, although the history of these regions lies outside the scope of this
History.
Turkish and, to a much lesser extent, Mongol military expansion south of the Oxus river
into what is now Afghanistan and northern India was also to have lasting political effects
on these regions. The establishment of the line of Ghaznavid sultans, who were originally
of Turkish slave origin, in eastern Afghanistan was the catalyst for several centuries of
expansion of Turkish and Afghan power, lured by the riches of the northern Indian plains.
Hence military dynasties of Turkish ethnic origin came to rule as far north in the subconti-
nent as Kashmir, as far east as Bengal and Assam, and as far south as the Deccan. But since
this was a question of the imposition of military aristocracies over alien populations, the
ethnic and linguistic composition of Afghanistan and India was not substantially affected
save in those districts of northern Afghanistan which adjoined the upper Oxus.
The changes brought about by the movements of Turkish peoples were accelerated by
the Mongol invasions of the thirteenth century, which had an immediate and cataclysmic
effect on the economic and social organization of the lands which they overran. In the
case of China, the discontinuity was speedily repaired as the Mongol Yuan dynasty and
their followers were absorbed into the traditional local society and into the enveloping
religious patterns and culture of China. Inner Asia itself benefited from the religious and
cultural ferment stirred up by transcontinental movements of peoples, soldiers, administra-
tors, merchants, men of religion, and so on. The region had always been content to absorb
influences from the surrounding higher civilizations. Within the confederations of the early
Turks and then of the Mongols, various faiths of both East and West found an honoured
place at the side of the indigenous animism of the shamans. Buddhist monks could travel
westwards into Transoxania, northern and eastern Afghanistan and northern India in the
earlier period, while much of our knowledge of the Mongols in their homeland and north-
ern China stems from the accounts of Western friars and envoys like William of Rubruck
and travellers like Marco Polo. In the long run, however, it was Lamaist Buddhism which
established itself in the Mongolian region and in Tibet, and Islam among the Turkish peo-
ples of Transoxania, southern Siberia and East Turkistan or Xinjiang. Earlier faiths such as

23
ISBN 978-92-3-103467-1 Contents INTRODUCTION

Manichaeism and Nestorian Christianity, which had flourished in some parts of the region,
disappeared well before the end of our period, i.e. before c. 1500.
In Transoxania and the northern regions of the Islamic Middle East, the effects of the
Mongol invasions were sharply felt in such fields as social organization and land tenure,
and the pastoralization of many parts of Iran and Anatolia was accentuated. It should be
remembered, however, that much of the Islamic world, including the Arabian peninsula,
most of Syria and all of Egypt and North Africa, was unaffected by the Mongols. In the
demonology of modern Arab writers attempting to explain the increasing rigidity oflslamic
religion and civilization from this time onwards, the effects of the Turkish and Mongol
invasions loom large; but in fact, the natural resilience of the Middle Eastern and Central
Asian cultures soon reasserted itself, and in respect of demographic and economic decline,
the effects of the Black Death were probably more lasting. It was in eastern Europe, above
all in Russia, that the constituting of the Turco-Mongol Golden Horde in the Kipchak and
south Russian steppes was to have a major, enduring influence on the course of the region's
history and, in particular, on the development of the Grand Duchy of Muscovy, the nucleus
of the future Russian empire.
A special feature of the form of Islam popular in Central Asia was the prominence of
Sufi mysticism, for much of the work of conversion among the Turkish tribes coming into
the Islamic world had been accomplished by dervish missionaries. Adherence to certain
of the Sufi brotherhoods, such as the Yasawiyya and, at a later date, the Naqshbandiyya,
became especially characteristic of Central Asian Islam. The popularity of the latter tarfqa
(Sufi order) spread to northern India, so that the Naqshbandiyya eventually took their place,
at the side of the Qadiriyya, the Chishtiyya and the Suhrawardiyya, as the dominant Sufi
orders in the subcontinent.
The rich religious and intellectual background of Central Asian scholars and literary
men helped the region to make a characteristic and significant contribution to world civ-
ilization; these intellectual and cultural achievements will be examined in Part Two of
Volume IV.
Some word of explanation is necessary concerning the vicissitudes of the editing of Part
One of Volume IV. The original Editor was Professor Muhammad S. Asimov of the Tajik
SSR, now Tajikistan, with Professor Yar Muhammad Khan of Pakistan as Co-Editor. Pro-
fessor Yar Muhammad Khan died at an early stage and I myself was invited to act as Editor.
Some of the contributors for the chapters had already been commissioned, but several oth-
ers had to be chosen subsequently. Unfortunately, Professor Asimov was assassinated in
Dushanbe in an apparently unmotivated killing in July 1996 and the actual editing of the
chapters and the checking of translations, several from Russian and other languages, has

24
ISBN 978-92-3-103467-1 Contents INTRODUCTION

devolved on myself. This I could never have done without the unstinting help and co-
operation of Mme Irene Iskender-Mochiri of UNESCO, who has acted as Secretary and
Co-ordinator of the whole project; the present book owes a great debt to her energy and
dedication.

25
ISBN 978-92-3-103467-1 Contents MUHAMMAD SEYFEYDINOVICH

MUHAMMAD SEYFEYDINOVICH ASIMOV

(ASIMI) IN PIAM MEMORIAM

R. N. Frye

On 29 July 1996 Professor Asimov was shot in the head by an unknown assailant as he
left his home in Dushanbe, Tajikistan. He died instantly, leaving a gap in the ranks of his
country's intellectual leaders, as well as in the International Scientific Committee which
initiated the project for the History of Civilizations of Central Asia (Asimov was one of the
editors of Volume IV). He was also a Co-editor of Volume IV of the History of Humanity. In
short, Asimov was UNESCO's most active contact in Central Asia. He made many friends
in Paris and contributed greatly to the success of these two major projects.
I first met Muhammad Asimov in Dushanbe in 1965, just after he became President
of the Academy of Sciences of Tajikistan, and talked to him many times afterwards. His
kindly mien and willingness to help others, even an American, was unusual at that time
of stress between the great powers. After the departure of Bobojan Gafurov to Moscow to
head the Institute of Oriental Studies of the USSR, Asimov became the leader of cultural
and intellectual pursuits in Tajikistan, and the representative of Tajikistan in all-Soviet and
international meetings.
Asimov was born on 25 August 1920 in Khujand and studied physics at the University
of Samarkand. But his interest and studies took him into fields of philosophy and history
and other facets of culture. He had a distinguished career, the details of which cannot
be elaborated here, except to mention a few items. He was Minister of Education of the
Government of Tajikistan in 1962, and, as already noted, President of the Academy of
Sciences until his retirement in 1991. He received the Nehru Prize for his contribution
to friendship among peoples in 1980. In 1990 he was the founder of Payvand, a cultural
organization for scholarly relations between Persian-speaking peoples in which he was
active until his death.
Asimov was a kindly man who refrained from the attacks on colleagues that are charac-
teristic of some scholars. He was active in intellectual pursuits until his death and he was
a great help to those working on the projects of UNESCO. He will be sorely missed by his
friends and associates.

26
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE UMAYYADS

CENTRAL ASIA UNDER THE UMAYYADS

AND THE EARLY cABBASIDS*

C. E. Bosworth and 0. G. Bolshakov

Contents

THE APPEARANCE OF THE ARABS IN CENTRAL ASIA UNDER THE UMAYYADS

AND THE ESTABLISHMENT OF ISLAM 28
Central Asia on the eve of the Arab incursions . 28
The appearance of the Arabs . . . . . . . . . . 29
CENTRAL ASIA UNDER THE EARLY c ABBASIDS 30
The course of the c Abbasid revolution and its significance . 30
The aftermath of the c Abbasid revolution and the fall of Abii Muslim . 36
The consolidation of c Abbasid power . . . . . . . . . . . . . . . . 39
Political, social and sectarian dissent in the early c Abbasid period . 41
The achievement of a degree of stability under al-Ma'miin .... 45

* See Map 1, pp. 426-7.

27
ISBN 978-92-3-103467-1 Contents Central Asia Arab incursions ...

Part One

THE APPEARANCE OF THE ARABS IN CENTRAL

ASIA UNDER THE UMAYYADS AND THE
ESTABLISHMENT OF ISLAM
(C. E. Bosworth)

Central Asia on the eve of the Arab incursions

Central Asia in the early seventh century was, ethnically, still largely an Iranian land whose
people used various Middle Iranian languages. In Transoxania there was a network of Sog-
dian city-states whose people used the Sogdian language, but there was possibly some
knowledge in the main towns at least of the Middle Persian Parthian language, because of
the strong cultural influence of the adjacent, powerful Sasanian empire. However, Sogdian
survived for at least two or three more centuries, especially in the countryside and in moun-
tainous areas, with such modern descendants as Yaghnobi. In Bactria, the provinces along
the upper Oxus river, now part of the Republic of Tajikistan and northern Afghanistan,
political control was exercised by epigoni of the Hephthalites. North of the Hindu Kush,
such a leader of the Hephthalites (in Arabic sources, Hayatila) as Tarkhan Nizak was to put
up a strenuous though ultimately unsuccessful resistance to the incoming Muslim Arabs. In
Khwarazm, an ancient Iranian civilization still flourished under the indigenous dynasty of
Afrighid Khwarazm Shahs, whose names, but not their chronology, are known to us from
the native scholar al-Biriini's Kitab al-Athar al-baqiya [Chronology of Ancient Nations],
known to modern scholars as the Chronology and written in c. 1000-1003. Along the north-
ern fringes of Transoxania, and in the deserts surrounding the oasis region of Khwarazm,
were Turkish tribes of the south-western group, such as Kaduk, Kimek, Kipchak and
Oghuz, and these were probably already infiltrating into the settled, agricultural lands of
Transoxania and Ferghana.
From the religious point of view, no single faith was dominant throughout the region.
In East Turkistan, the Tarim basin and its fringes, the Indo-Iranian culture of such cen-
tres as Khotan and Kocho was still vital, although soon to yield to the Uighur Turks, and

28
ISBN 978-92-3-103467-1 Contents The appearance of the Arabs

this culture was still dominated by Buddhism. Likewise, Buddhism was strong in Bactria;
Balkh, and its famed monastery of Nawa Vihara (Arabized as Naw Bahar), was a major
centre of the faith. But Buddhist influence in Sogdia had been waning for some time and
when the Chinese Buddhist pilgrim Hstian-tsang arrived at Samarkand c. 630, he found
Buddhism there in full decline, and Zoroastrianism, backed by the military and cultural
prestige of the Sasanian empire, in the ascendant. Christianity was also strong, however,
with Nestorians, Jacobites and Melkites all represented in Transoxania and Khwarazm.
There was a Christian bishop at Mervin 334 and probably one in Samarkand by the sixth
century. Manichaeism and other dualist faiths were represented, with the followers of Mani
finding a particularly favourable reception among the Uighurs in East Turkistan; and neo-
Mazdakites are mentioned also in Samarkand.
Watered by such rivers as the Zarafshan, the Amu Darya ( Oxus) and the Syr Darya
(Jaxartes) and their tributaries, the regions of Transoxania and Khwarazm were fertile,
flourishing agricultural areas. The adventurous merchants of their cities carried on long-
distance trade through Inner Eurasia, so that we know of the existence of Sogdian trad-
ing colonies as far east as northern China and Khwarazmian ones as far west as southern
Russia.

The appearance of the Arabs

Having overthrown the Sasanian empire, the Arabs first crossed the Oxus in 653-4 during
the caliphate of cuthman (644-56), but such vital crossing-points as Amul-i Shatt and
Tirmidh (Termez) were not secured until some time later; only then was it strategically wise
for the Arab commanders to commit large bodies of troops for raids across the river. Hence
it was not until 674, under the first Umayyad caliph Mucawiya I (661-80), that his general
cubayd Allah b. Ziyad crossed the Oxus and defeated the forces of the Bukhar Khudat,
the local Sogdian ruler of Bukhara. Civil warfare and an anti-caliph who set himself up
in rivalry to the Umayyads held back Arab progress; the Sogdian city-states, meanwhile,
sent fruitless embassies to Peking to induce the Chinese emperor, who claimed a vague
suzerainty over Central Asia, to intervene.
It was the Arab general Qutayba b. Muslim al-Bahili, governor of Khurasan and the
East from 705 to 715, who first established a firm Arab hold in the lands beyond the Oxus.
He conquered Bukhara and Paykand in 706-9; he made Tarkhan and then Ghurak, the
rulers of Samarkand, his vassals and he built mosques and introduced the practices of
Islam into these cities; he repelled invasions in 707 and 712 by the Kaghan of the Eastern
Turks, whose help had been called in by the alarmed Sogdian princes; he fought and in 710

29
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

killed the Hephthalite leader Tarkhan Nizak in Tukharistan, the district of Bactria south of
the middle Oxus; he campaigned in the middle Syr Darya lands; and he twice invaded
Khwarazm in 712, killing the local shah and inflicting considerable damage on the fabric
of local Khwarazmian culture, without however securing any significant foothold for Islam.
After Qutayba's recall and death in 715, however, Arab fortunes suffered sharp reverses
over the following two decades or so. In 728 the Kaghan of the Ttirgesh (Western Turks)
inflicted a crushing defeat on Arab troops who had invaded Ferghana (the so-called 'Day of
Thirst') and allied with the Sogdian princes, so that by that year, of their former possessions
across the Oxus the Arabs held only Samarkand and Dabusiyya. The Arab and Muslim
position was not re-established until the appointment to Khurasan of another governor
of genius, Nasr b. Sayyar al-Kinan! (738-48). He alleviated the discontent of the local
peoples who had converted to Islam but were still forced to pay the jizya (poll tax) to the
Arab treasury; he conciliated the rebel in Bactria and Transoxania, al-Harith b. Surayj;
and he penetrated into Ferghana again. But his successes were negatived by the growing
danger to Umayyad control in the eastern provinces of the caliphate, stemming from the
propagandist, missionary movement (dacwa) of Abu Muslim al-Khurasani and other pro-
c Abbasid leaders from among the Arab settlers in Khurasan. Nasr b. Sayyar was forced to
abandon the eastern provinces of the caliphate by 748 and, retreating westwards, he was
killed by the advancing c Abbasid army (see Part Two below).

Part Two

CENTRAL ASIA UNDER THE EARLY cABBASIDS

( 0. G. Bolshakov)

The course of the cAbbasid revolution and its

significance
In the 740s the Umayyad caliphate was in the throes of a deep internal crisis. To the dis-
content of the subject population, who did not enjoy the same rights as the Arabs and were
unable to acquire them even by adopting Islam, was added the growing dissatisfaction of
the Arabs themselves with Umayyad rule. That dissatisfaction shattered the unity which

30
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

had secured their dominion over a vast territory almost twice as large as the Roman empire
at the zenith of its power.
The various strands in the anti-Umayyad movements in Muslim society itself may be
subsumed under two main groupings: the egalitarian Kharijites and the Shfite charismatic
tendency. The Kharijites fought to re-establish the original state of equality between mem-
bers of the Muslim community; they also opposed social disparities and the inherited power
of the caliphs who, they believed, should be chosen by the community from among suit-
able candidates irrespective of their origins. All of these views were summarized in a slo-
gan calling for a return to the Qur' an and the way of the Prophet. The Shfites, however,
wished hereditary power to be vested in the family of the Prophet, meaning the descen-
dants of cAli, who, in their belief, embodied the divine grace inherited from Muhammad
and transmitted from one divinely chosen head of the community, the imam, to the next.
The Umayyads managed to quell the uncoordinated rebellions of the Kharijites and
Shfites so long as the bulk of the Arab population remained aware of its common inter-
ests, but the gradual build-up of resentment at the actions of individual caliphs and their
governors, the rivalry between individual tribes and the memories of old conflicts divided
them into a multitude of hostile groups unrelated to either social or religious doctrines.
The detonator of the explosion which destroyed the Umayyad caliphate was provided by
Khurasan, where all of these contradictions could be seen at their most acute.
In the year 744 the caliph al-Walid II was killed, ushering in a period of internecine
strife. There were three caliphs in the space of seven months. Provincial governors were
unable to keep up with political changes and tried to take advantage of the situation; the
Kharijite movement was everywhere on the increase. The caliph Marwan II, who came to
power at the end of 744, succeeded in pacifying the heart of the empire, Syria and Iraq.
In Kufa the Shicites swore an oath of allegiance to cAbd Allah b. Mucawiya (the great-
grandson of the Prophet's cousin) as caliph; after a hard struggle he was expelled from Iraq
but found support in Iran.
In Khurasan a pre-existing enmity between two groupings of Arabs increased against
the background of this political instability: the northern Arab tribal grouping of the Mudar
(Tamim, Qays and Kinana) and the southern Arab tribes (Yemenis), who were joined by
the northern Arab group of RabJ:Ca. This was an ancient rivalry involving an ever-growing
number of grievances.
The governor of Khurasan at that time was Nasr b. Sayyar, a member of a small tribe,
the Layth, from the Mudar group. Although this old warrior was a skilled politician,
he was unable to reconcile the opposing sides, partly because he himself had played a
part in several armed conflicts in over half a century of activity. In the summer of 744

31
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

Nasr b. Sayyar, fearing that the internecine strife might spread to Khurasan, incarcerated
the leader of the southern Arab grouping, Judayc b. cAli al-Kirmani. Judayc managed to
escape from prison in the citadel of Merv and took refuge in the settlement of his tribe,
protected by 3,000 loyal troops. Although Nasr did not pursue him, the Merv oasis was
pervaded by an atmosphere of armed conflict that the least spark could ignite into open
war.
This situation was exploited by the cAbbasids, the descendants of the Prophet's uncle
cAbbas, to spread their propaganda. The cAbbasid movement had sprung up in the 720s,
when the cAlid imam cAbd Allah b. Muhammad b. al-Hanafiyya bequeathed his secret
organization before his death to cAbbas's great-grandson Muhammad. He allegedly passed
on to him a 'green scroll' said to contain a secret meaning entrusted to cAli by the Prophet
together with the right to the imamate. Great care was taken to ensure that this organi-
zation, based in Kufa, was kept secret: only the most trusted individuals met the imam,
usually in Mecca during the pilgrimage, when such meetings could not arouse suspicion.
Propaganda was conducted on behalf of an unnamed imam 'from the Prophet's family
pleasing to Allah'. This anonymity widened potential support, as such a description of
the imam corresponded to the expectations of the Shfites. Many propagandists sent to
Khurasan died, but no one was able to betray the name of the imam as they did not know
for whom they were canvassing support. Moreover, the Imam Muhammad restrained his
supporters from premature action, awaiting a favourable moment. The Imam Muhammad
died in 743 before that moment arrived. He was succeeded by his son Ibrahim.
At the height of the disturbances caused by the murder of al-Walid II, the head of
the cAbbasid organization Abu Salama al-Khallal appeared in Khurasan, with instructions
from the new imam. After spending four months in Khurasan, Abu Salama returned safely
to the imam with money that had been collected in the region. As Ibrahim took a lik-
ing to the intelligent slave accompanying him, Abu Salama made a gift of him to the
imam. Ibrahim freed the slave, making him his confidant and giving him the name cAbd
al-Rahman and the kunya (patronymic) of Abu Muslim. In 745 Abu Muslim arrived in
Khurasan and Merv as the imam's plenipotentiary representative.
In the meantime, the situation in Khurasan had become even more involved and tense.
In the spring of 745 the leader of the Kharijites of Khurasan, al-Harith b. Surayj, who
had been pardoned by the caliph, returned from the 'land of the Turks' with a detachment
of battle-hardened troops. Nasr b. Sayyar attempted to win him over with rich gifts and
promises of high office, but the inflexible Kharijite responded that he required nothing
and would support whoever promised to follow the Qur' an and the way of the Prophet.

32
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

Al-Harith's stance drew many supporters to him and he became a potent political force in
Merv.
At the end of March, al-Harith attacked Nasr b. Sayyar with the support of Judayc al-
Kirmani, and together they managed to expel Nasr from Merv to Nishapur. The victors
immediately began to settle accounts with each other, and before a month had gone by,
al-Harith was killed in a battle with Judayc. After establishing a firm hold in Merv, Judayc
decided to deal a final blow to Nasr. In a battle near Merv al-Rudh, Judayc was defeated,
but Nasr lacked the forces to defeat his rival decisively. The stalemate near Merv al-Rudh
continued into the winter. Judayc's son cAli sent him a detachment of 1,000 men with sup-
plies and clothing, but supporters of Nasr attacked him on the way and looted the baggage
train. On learning of this, the inhabitants of Merv rebelled against cAli. Judayc had no other
option but to seek a reconciliation with Nasr. At the beginning of 746 the rivals returned to
Merv.
At that time Abu Muslim was in western Khurasan. The Imam Ibrahim decided that
the long-awaited moment had finally arrived (particularly since all of the caliph's forces
were engaged in crushing rebellions in Iraq and southern Iran) and ordered Abu Muslim
to prepare a rising in the Merv oasis. Another tragedy developed there: the son of al-
Harith killed Judayc al-Kirmani at Nasr's instigation. This further exacerbated relationships
between the northern and the southern Arab tribes (the latter were led by cAli, the son of
Judayc al-Kirmani).
The head of the cAbbasid organization in Merv, Sulayman b. Kathir, gave Abu Muslim a
hostile reception, but a majority submitted without question to the order of the imam. At the
end of April 747 he dispatched messages throughout Khurasan calling for the start of the
rebellion. On 25 Ramadan (9 June) 747, two black banners (the colour of the cAbbasids)
sent by the imam were raised in the settlement of Safizanj and the rebels clothed themselves
in black. By the time of the feast at the end of the Ramadan fast, 4,000 men had assembled
under the banners of Abu Muslim. Calls for the overthrow of the Umayyad tyrants, and the
transfer of power to a caliph from the Prophet's family who would follow the Qur'an and
the way of the Prophet, drew a great variety of malcontents and opposition groups to Abu
Muslim: from Shfites and Kharijites to local dihqans (landowners) and to slaves who had
been promised their freedom. The slaves, however, were dispatched to a special camp and
not issued with arms.
Nasr b. Sayyar did not at first realize which was the greater of the dangers facing
him. Instead of nipping the rebellion in the bud, he continued his struggle against cAli
al-Kirmani and the new leader of the Kharijites, Shayban, who had appeared in Merv. A
small cavalry detachment which was dispatched against Abu Muslim was defeated and its

33
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

commander taken prisoner. Abu Muslim rapidly seized the initiative; capturing a village,
he blocked Nasr's path to Nishapur, and his emissaries stirred up rebellion in Merv al-
Rudh, Amul and Zamma, Talaqan and Nasa. But he did not attack Nasr, proposing instead
that he join the movement and promising to hand over its leadership. When Nasr rejected
this proposal, Abu Muslim enlisted the support of cAli al-Kirmanl. Nasr turned to the sec-
tion of the Arab population that had not so far been involved in internecine strife, appealing
to them to defend Islam and the Arabs against heathens, slaves and the Arab rabble. This
call was well received; Nasr was joined by Shayban and there were desertions from Abu
Muslim's camp. But Nasr still did not attack Abu Muslim, who neutralized Nasr's efforts
by demonstrating his piety in every possible way. As a result, he managed to split Nasr and
Shayban and to attract new supporters.
Finding himself in a hopeless situation, Nasr made a desperate appeal to the caliph,
Marwan II, for help but the governor oflraq, Ibn Hubayra, who was hostile to Nasr, inter-
cepted all his messages. Abu Muslim now decided to attack. He moved his forces up to
Merv and, after waiting until a battle had begun between Nasr and cAli, entered the town
without any opposition on 14 or 15 February 748. Abandoning his family to their fate,
Nasr fled from Merv to Nishapur with a small escort and began to assemble his forces
there. Shayban also refused to swear allegiance to Abu Muslim and departed for Sarakhs.
Abu Muslim dispatched the army of Qahtaba b. Humayd against them. Qahtaba routed
Shayban and then defeated a 2,000-strong force under the son of Nasr b. Sayyar near Tus.
The road to Nishapur was open and Nasr was obliged to retreat westwards to Simnan.
It was only then that the caliph realized the danger threatening him. On his orders, a
Syrian army was summoned from southern Persia, where it had crushed a rebellion led
by cAbd Allah b. Mucawiya. But instead of blocking the road from Khurasan to Iraq, its
commander marched further north to Gurgan. Qahtaba surrounded this force and defeated
it one month later. Nasr fled further west and died in Sava as Qahtaba entered Rayy. At the
same time, another of Abu Muslim's commanders, Abu Dawud, took control of Balkh and
the whole of Tukharistan after several battles.
Qahtaba remained in Rayy for five months. The governor of Iraq, Ibn Hubayra, used this
time to assemble troops. In the spring of 749 Qahtaba's path was blocked by large forces
of the Syrian army which had been stationed in Isfahan and Nihavand. Qahtaba and his
main force defeated the Isfahan contingent, while his son besieged the army at Nihavand.
On 26 June 749 the Nihavand garrison surrendered after a four-months' siege. Ibn Hubayra
assembled some 53,000 troops and pitched camp at Jalula, awaiting Qahtaba's attack; at the
same time, an army from Syria advanced through northern Kurdistan (Shahrazur), threat-
ening Qahtaba from the rear. His defeat seemed certain, but by means of a diversionary

34
ISBN 978-92-3-103467-1 Contents cAbbasid revolution and its significance

movement he succeeded in luring lbn Hubayra out of the fortified encampment and placing
him in an unfavourable position. lbn Hubayra was thus unable to organize any resistance
to the c Abbasid army on its advance to Kufa, in spite of the fact that Qahtaba was killed in
a chance skirmish, leaving the army without its talented commander.
The Umayyads' only consolation was that they had discovered the name of the c Abbasid
imam and were able to arrest him. But the Imam lbrahim's arrest came too late to change
the situation. The c Abbasid army was unstoppable in its advance on Kufa, which it entered
on 29 August 749. Ibrahim was killed on the order of the caliph, but had managed to pass
on the message that in the event of his death the imamate would pass to his brother,Abu
'1-c Abbas cAbd Allah. Abu '1-c Abbas (who was later proclaimed caliph, with the title al-
Saffah) arrived secretly in Kufa with a group of relatives and an escort. For six weeks Abu
Salama concealed the imam's abode, intending to come to an agreement with one of the
c Alids and pass on to them the fruits of victory. When all those who had put themselves
forward as claimants refused to take power, and the imam's arrival became known to some
of the people from Khurasan, Abu Salama organized a ceremony at which allegiance was
sworn to the new caliph on 28 November 749. In his first speech he promised to establish
peace and justice and, as a first indication of the advent of a new era, he increased the
troops' pay.
Marwan II made one further attempt to halt the advance of the c Abbasid army, assuming
personal command of a large force which went to meet it. In a decisive battle on the banks
of the Greater Zab, Marwan was utterly defeated. He retreated to Syria but found no support
there either. Continuing to retreat before the c Abbasid army, Marwan eventually reached
Upper Egypt, where he was killed at Busir (July-August 750).
The establishment of c Abbasid rule disappointed many of those who had participated in
the movement which had brought them to power. The Shfites and Kharijites soon realized
that their slogans had been exploited by the c Abbasids to conceal their true aims. The
universal prosperity which had been expected did not materialize, and the promised justice
and reconciliation within the community took the form of repression and executions. The
new rulers were especially harsh in dealing with the Umayyads. All the men of the family
unlucky enough to fall into the victors' hands were killed, and even the dead did not escape
punishment: the remains of almost all of them were disinterred.

35
ISBN 978-92-3-103467-1 Contents Aftermath of the cAbbasid revolution ...

The aftermath of the cAbbasid revolution and the fall

of Abu Muslim
After the victors' enemies came the turn of their comrades-in-arms: those who had shown
disrespect at some time or who simply proved embarrassing by virtue of the fact that
the new rulers were indebted to them for their accession to power. Abu Muslim began
to take reprisals against them immediately after his victory. The first casualties were cAli:
al-Kirmani and his sons; at a word from Abu '1-c Abbas, an assassin was dispatched to deal
with Abu Salama; Sulayman b. Kathir also met a violent end.
The Shfites, who had been recent allies in the struggle against the Umayyads, responded
by organizing uprisings. In the spring of 751 in Bukhara, one of the strongholds of Arab
power in Transoxania, a rebellion was mounted by the Shicite Shadk (or Shurayk) b.
Shaykh, who declared that he had not followed the family of Muhammad in order to shed
blood and break the law. He was supported by some 30,000 men. Ziyad b. Salih, who was
sent to crush the uprising, could not have dealt with the rebels without the aid of the local
ruler, the Bukhar Khudat, and the Bukharan dihqans.
The position of the local Iranian nobility was ambiguous. Some, like the Bukhar Khudat,
became faithful supporters of the new dynasty which had promised to give them equal
rights with the Arabs if they accepted Islam; others exploited the internecine strife in the
Arab camp to restore their independence. These were assisted by the intervention of the
Chinese, whose policy towards the West was active in those years. In 748, when Abu Mus-
lim's forces were pursuing Nasr b. Sayyar, Chinese forces seized and destroyed the town
of Suyab, the headquarters of the Kaghan of the 'yellow' Ttirgesh. At the same time, the
commander-in-chief of the Western Regions, Kao-hsien-chih, was subjugating minor prin-
cipalities in the Pamirs and the upper reaches of the Indus. Two years later he appeared
in the eastern part of Transoxania, having been called to the assistance of the Ikhshid of
Ferghana against the ruler of Chach (later, Tashkent). The ruler of Chach was taken pris-
oner, sent to the imperial court and executed. His son turned to the Arabs for assistance.
In response, Abu Muslim sent Ziyad b. Salih to him while he himself established his base
in Samarkand. The Arab army advanced to Talas or Taraz (modern Jambul) where it was
besieged, but on receiving reinforcements from Abu Muslim, went over to active opera-
tions. In July 7 51 it encountered a Chinese army of 30,000 men. The outcome of the battle
was decided by the rising of the Karluk in the rear of Kao-hsien-chih. Attacked on two
sides, the Chinese force was annihilated and its commander, surrounded by bodyguards,
fought his way with difficulty through the mass of fugitives. The Arabs found themselves
in possession of a substantial booty and a large number of prisoners, among whom were

36
ISBN 978-92-3-103467-1 Contents Aftermath of the cAbbasid revolution ...

skilled silk-weavers and paper-makers. The weavers were sent to the caliph's textile work-
shops in Kufa, while the paper-makers remained in Samarkand to establish a paper-making
industry which subsequently played a major role in the development of book production in
the Muslim world.
At the same time, Abu Muslim subdued Sogdiana by fire and the sword, while the
governor of Balkh, Khalid b. Ibrahim, invaded Khuttal (whose ruler fled to Ferghana and
thence to China) and then marched north to Kish, where he defeated and killed its ruler.
Abu Muslim became the absolute ruler of Khurasan and Transoxania, having at his dis-
posal a loyal army such as not even the caliph controlled. He had become a danger. The
caliph instigated a rising by Ziyad b. Salih, who had become the governor of Bukhara and
Sogdiana, but a majority of commanders remained loyal to Abu Muslim. Ziyad fled but
died at the hands of a dihqan with whom he had sought shelter. Vexed by this failure, the
caliph's brother, Abu Jacfar, determined to kill Abu Muslim at his next audience.
An opportunity to dispose of the 'custodian of the dynasty', as the cAbbasids referred to
Abu Muslim, presented itself in the year 754, when Abu Jacfar al-Mansur became caliph
on the death of Abu '1-c Abbas al-Saffah and fate willed that Abu Muslim should again
demonstrate his loyalty by crushing a revolt organized by the uncle of the new caliph. Al-
Mansur ordered the surrender of the booty acquired in the course of the operation and when
Abu Muslim arrived to seek an explanation, the caliph ordered him to be killed.
The troops accompanying Abu Muslim accepted the news of his execution after receiv-
ing 1,000 dirhams each. There was a different reaction in Khurasan, however. The Zoroas-
trian Sunbadh rose to avenge Abu Muslim's death. The rebellion encompassed all of north-
ern Persia from Nishapur to Rayy. After 70 days, it was brutally repressed and women and
children were killed as well as men, a circumstance which indicates that the rebellion was
a popular one and not simply an army revolt. Such large forces had to be employed in
crushing it that the customary summer campaign against Byzantium in Anatolia was not
conducted that year.
The reaction of the population of Khurasan to the murder of Abu Muslim demonstrated
the complex nature of the movement that he had led. Originating in Arab-Muslim circles,
it had won the support of substantial numbers of people of differing social status, reli-
gious conviction and ethnic attachment. Whether they had accepted Islam superficially or
through inner conviction, they infused it with the ideas that were dominant in their native
environment. In the doctrines of the radical Shicites (ghulat) concerning the imamate, there
is already an idea foreign to the original ethos of Islam, i.e. the incarnation of a divine
emanation and its possible transfer by inheritance. In some sectarian groups, reverence for
the imam came to resemble the worship of a human being as a god, something which is

37
ISBN 978-92-3-103467-1 Contents Aftermath of the cAbbasid revolution ...

fundamentally incompatible with Islam. Thus the Rawandiyya (to which Abu Muslim is
thought to have belonged) were so convinced of the divinity of al-Mansur that the comman-
der of his guard appeared to them to be the Archangel Jibra'il (Gabriel). Such behaviour
so discredited the caliph in the eyes of orthodox Muslims that when members of the sect
surrounded his palace at Qasr lbn Hubayra in Iraq in order to worship him, he ordered their
dispersal and, in the event of resistance, their slaughter.
Such ideas were often combined with a belief in the transmigration of souls. This could
mean the incarnation not just of the divine spirit but of the spirit of any revered individual.
Thus the idea of the imamate became separated from its essential ingredient: the inheri-
tance of the right to rule the Muslim community within the family of cAli. Abu Muslim
also became the object of this type of worship. The Rizamiyya, who worshipped Abu 'l-
e Abbas, considered that Abu Muslim had become the imam after his death, disagreeing
only on the identity of Abu Muslim's successor. Some refused to accept his death and
awaited his return (the sect of the Abu Muslimiyya), and this encouraged the appearance
of leaders of popular movements following Abu Muslim.
Some medieval sources and, after them, some modem researchers, have linked the name
of Abu Muslim with the movement of the 'wearers of white' (Arabic, al-mubayyida; Per-
sian, safid-jamagan), a neo-Mazdakite sect whose distinctive feature was white clothes (or
a white banner) as a symbol of purity or else in opposition to the colour black espoused
by the cAbbasids. Their doctrine contained a call for equality of social status and property
within the community and also a belief in reincarnation. However, as we shall attempt to
demonstrate below, there is no convincing evidence of their link with the Abu Muslimiyya.
Certainly, the fact that medieval authors lumped a number of beliefs and groups together
under the title of al-mubayyida on the basis of their external appearance, without having
any idea as to the content of their doctrine, does not inspire confidence in this identification.
On all the evidence, Abu Muslim was an orthodox Muslim (or at least appeared so
to his entourage) but, as we have seen, the first person who sought to avenge him was a
Zoroastrian priest. Another member of his movement, Is'haq the Turk, who had organized
a revolt at roughly the same time, described himself as a follower of Zoroaster, who would
soon appear to the world and establish justice. In 757 Abu Muslim's commander, together
with his successor Khalid b. Ibrahim, were killed by followers ofls'haq. The next governor,
cAbd al-Jabbar, who had punished a group of Shicites for trying to enlist support for the
descendants of cAli, himself raised a revolt one year later and formed an alliance with the
'wearers of white', who were led by Baraz-banda.

38
ISBN 978-92-3-103467-1 Contents The consolidation of cAbbasid power

The consolidation of cAbbasid power

In spite of isolated rebellions and revolts, c Abbasid power was consolidated in Khurasan
and Transoxania to such an extent that Arab troops (from Ferghana?) are stated in the Chi-
nese sources to have helped to crush a rebellion in China during those years (757-8). The
Ttirgesh, who tried to win Transoxania from the Arabs for nearly twenty years, weakened
their position through constant internecine strife and were supplanted by the Karluk, who
in 766 seized control of the Ttirgesh pasturelands in Semirechye.
One of the most important factors in the stabilization of the situation in Transoxania
was the new attitude of the local dihqans towards the Muslim authorities. The advent of
the c Abbasids was more than the replacement of one dynasty by another: it brought about
major alterations in the social and military structure of the caliphate. In the first place, it
meant an erosion of the Arabs' dominant social position and the introduction of equal polit-
ical rights for all Muslims, Arab and non-Arab alike. Some Iranians were even convinced
that the c Abbasids intended to eradicate the Arabs in Khurasan; this seemed the only pos-
sible explanation for the order allegedly given to Abu Muslim by the Imam Ibrahim to kill
all the Arabs there. (There is no evidence that this order was in fact given.) All that can be
said is that in Syria and Jazira (Upper Mesopotamia), the army from Khurasan treated the
local Arabs, who provided support for the Umayyads, as enemies. According to a Christian
historian of the end of the eighth century, the 'Persians' slaughtered the Arabs like lambs.
The change in the situation was reflected in different ways in those regions which had
been a part of the caliphate for a century and in those which had only recently been incor-
porated into it, such as Transoxania. In Iran, much of the local nobility and many local
officials had already adopted Islam and had been absorbed into the new state prior to the
accession of the c Abbasids. They immediately provided support for the new dynasty. In
Transoxania, as we have seen, the first response to the overthrow of the Umayyads was the
restoration of independence. The defeat of the Chinese, on whose assistance many local
rulers of Transoxania counted, and the harsh reprisals carried out against unruly dihqans,
obliged the leading dihqans to seek a reconciliation with the new dynasty and to become
its loyal vassals.
By this time, many leading positions in the government structure were already occu-
pied by people from Khurasan by right of precedence as 'sons of the c Abbasid revolution'
and also because of their experience of work in a bureaucracy, which the Transoxanian
landowners lacked, managing their small domains in a patriarchal manner. Arabs continued
to occupy high posts in the army, whereas Iranians or Tajiks gained the upper hand in the
civil administration. From the earliest years of Abu '1-c Abbas's reign, the most important

39
ISBN 978-92-3-103467-1 Contents The consolidation of cAbbasid power

department, the dfwan al-kharaj (concerned with taxation and land tenure), was headed by
Khalid b. Barmak, the son of the former chief priest of a Buddhist temple in Balkh. He also
became, in effect, the first vizier in the history of the caliphate. This high administrative
post was possibly influenced by the Sasanian administrative tradition and may have con-
stituted a revival of the institution of the vizier (buzurg farmandar) or it may, on the other
hand, have been an indigenous development within the Arab ministerial tradition. With the
evolution of the post of vizier under the caliphate, genuine state budgets began to be drawn
up for the first time, and offices sprang up for various departments with extensive staffs
of officials who engaged in correspondence with the provinces and prepared estimates and
accounts. An influential stratum of officialdom, the Irano-Islamic class of secretaries (Ara-
bic, kuttab, Persian dabfran), was formed which considered itself as the main support of
the state. Their knowledge of the complex system of the kharaj (land tax), which took
account not only of the quality of the land but of the produce of the crops sown, made the
officials of the dfwan al-kharaj the guardians of knowledge which was inaccessible to the
uninitiated and was passed on by inheritance.
The choice of site for the new capital was an indication of the c Abbasids' break with
the Umayyad tradition of looking towards Syria and the culture of the Mediterranean. Al-
Mansur (7 54-7 5) inspected several sites, all in Iraq, before settling on the little village
of Baghdad on the western bank of the Tigris 30 km upstream from the former Sasanian
capital of Ctesiphon. The foundations of the new residence, which received the official
appellation of Madinat al-Salam (City of Peace), were laid in the year 762. Baghdad at
once became an international city. The 30,000-strong army of al-Mansur which lodged in
the city contained detachments from every part of Iran and, in particular, from Khurasan.
The builders of Baghdad, some of whom remained in the city after it was built, represented
all the countries of the Near East. The local population, who spoke Aramaic and some Per-
sian, was mixed with Arabs from Kufa, Basra and Wasit. Some districts which were called
after different areas of Transoxania accommodated troops from those areas. A new Mus-
lim culture gradually took shape in this ethnic cauldron, thereafter solidifying in different
language traditions.
When compared with the luxury with which the c Abbasids and their large retinue sur-
rounded themselves, the way of life of the Umayyads seemed almost ascetic; the expen-
diture of the caliph's court was equivalent to the entire tax revenue from a large region.
The finance department diligently sought means of increasing the income from taxes. The
cadasters were reviewed and taxation was increased in a number of areas of the Near East.
There are no indications, however, that similar measures were adopted in Khurasan and
Transoxania during the initial period of c Abbasid rule, although the collection of taxes

40
ISBN 978-92-3-103467-1 Contents Dissent in the early cAbbasid period

must have given rise to problems. In most regions of Khurasan and Transoxania, the
amount of taxation for districts was determined by agreements which specified only the
overall amount of the tribute, the levying of individual taxation being the duty of the local
landowners. As more inhabitants of an area adopted Islam, the potential for the collec-
tion of the jizya was reduced, and the extent of the commercial duties levied on Muslim
traders and craftsmen was restricted. Payment of the fraction which could not be collected
from them was imposed on non-Muslims. Additional requisitions could be added to those
amounts at the request of the authorities. This policy created a divided society, not between
Arabs and non-Arabs as before but between Muslims and non-Muslims. Another reason
for the new wave of tension was the rupture of the stereotype of social thinking that had
maintained the stability of social relations. The Islamic world had changed in a single gen-
eration; everything was now new, unfamiliar and unstable. The seeds of popular rebellions
with the most extreme slogans were easily sown in such a situation.

Political, social and sectarian dissent in the early

cAbbasid period
This dissent was early and most clearly demonstrated in the four-year peasant war con-
ducted by Hashim b. Hakim, nicknamed al-Muqannac (The Veiled One), who came from
a family connected with Abu Muslim's movement. He himself had taken part in the rebel-
lion of cAbd al-Jab bar supporting the view of Abu Muslim as the imam. After the rebellion
was defeated, al-Muqannac found himself in a Baghdad prison, from which he escaped to
his homeland, reaching a village near Merv. There he began to preach that he thenceforth
embodied the divine spirit which had been incarnate in Abraham, Jesus, Muhammad and
Abu Muslim. His fellow countrymen reacted with the usual scepticism to such revelations
from someone who was well known to them, refusing to take him seriously. Al-Muqannac
found more receptive ears for his propaganda, however, in Transoxania, where there was
much discontent, for the reasons indicated above. The presence of a colony of Manichaeans
in Samarkand may have provided a breeding-ground for neo-Mazdakite ideas, and the doc-
trine of transmigration may well have found a response also in the southern regions of
Transoxania, where Buddhist temples had stood in the recent past.
By the spring of 776 the agitators dispatched by al-Muqannac to spread the word had
raised a rebellion in the region of Kish (modern Kitab and Shahr-i Sabz). Hashim then
moved to the region, establishing himself in an inaccessible mountain fortress somewhere
in the upper reaches of the Kashka Darya. At that time the 'wearers of white' had seized
two small towns in the Bukhara oasis. Al-Muqannac had supporters in Samarkand, which

41
ISBN 978-92-3-103467-1 Contents Dissent in the early cAbbasid period

also joined the rebels. Jibra'll b. Yahya, who had been sent by the caliph al-Mahdi (755-85)
to crush the rebellion, spent four months trying unsuccessfully to deal with the 'wearers of
white', and then moved on to Samarkand, which he succeeded in occupying. This partial
success did not basically alter the position of the rebels; in the year 777 they occupied the
entire valley of the Kashka Darya with the exception of a few towns, extending their power
southwards to Termez, where they inflicted a significant defeat on government forces and
occupied Samarkand with the support of the Turkish Karluk.
The governor of Khurasan was unable to be of much assistance to the local authorities
as he was occupied in crushing the revolt of Yusuf al-Barm in Fushanj. Only after the
latter was defeated did the new governor, Mucadh b. Muslim, manage, in the spring of
778, to reach Transoxania with a large army. By the end of the year the rebels had been
pushed back to the region of Kish, but the onset of winter brought military operations
to a halt. The effectiveness of action by the government forces was hampered by rivalry
between the governor and the commander of the army, al-Harashi, who did not appreciate
the governor's interference. Mucadh eventually retired from the fray and al-Harashi was
left to deal with al-Muqannac alone. He tried to storm the latter's fortress but was beaten
off. He then threw all his forces against Nevaket (the site of Kamay-tepe 40 km south-
west of Shahr-i Sabz), which was defended by the brother of al-Muqannac. Only when the
defenders had thrown themselves upon his mercy was al-Harashi able to proceed to a siege
of the fortress, which dragged on for nearly a year. Exhausted by the lengthy siege, the
supporters of al-Muqannac defending the lower part of the fortress entered into negotiations
with al-Harashi, surrendered it and were pardoned. Al-Muqannac remained in the citadel
with his immediate retinue, and when he saw that the position was hopeless, committed
suicide, after first poisoning his wife and killing his favourite slave. In order to preserve
the image of a prophet who had risen to heaven, al-Muqannac ordered those who remained
alive to bum his body, but they did not do this properly: on discovering the charred remains,
the victors cut off the head and sent it to the caliph.
Muslim historians always recall that al-Muqannac declared himself to be an incarnation
of the divinity but they never mention the social aspects of his teachings. They ascribe to
him the abolition of property and the introduction of promiscuity, but there is no infor-
mation on the division of the property of the rich or the persecution of the dihqans. His
movement was obviously not aimed at the landowning classes. It was anti-Islamic in ten-
dency, bringing together all of the forces in Transoxania that were discontented with the
new dynasty in a final attempt to remove it and return to the old way of life. This may
explain why, out of eleven names of supporters of al-Muqannac, only two were Muslim
names. The rejection of al-Muqannac's ideas by most of the towns confirms the hypothesis

42
ISBN 978-92-3-103467-1 Contents Dissent in the early cAbbasid period

that the bulk of the rebels were peasants and inhabitants of remote mountain regions in the
upper reaches of the Kashka Darya where the old beliefs still persisted.
The fierce, four-year war had a harmful effect on the economic situation in Transoxania.
Samarkand, which had changed hands three times, must have suffered no less than during
its worst period in the 730s. This is probably one reason why debased silver dirhams asso-
ciated with the names of the governors Musayyab b. Zuhayr (Musayyabi) and Ghitrif b.
cAta' (Ghitrifi) started to be minted in Transoxania (see below, Chapter 20).
The rebellion of al-Muqannac marked a clear divide between two periods in the his-
tory of Transoxania. Its defeat signified the definitive triumph of Islam. Major changes had
taken place in the country in the thirty years between the revolt of Abu Muslim and the
defeat of al-Muqannac. The dihqans had lost political power. As early as 760 the Ikhshid
of Sogdiana had stopped minting his own money, at roughly the same time as the Bukhar
Khudat. The Ikhshid of Sogdiana disappeared from the historical stage, whereas the Bukhar
Khudat retained some semblance of power. After the execution of the Bukhar Khudat for
his support of the 'wearers of white', however, his descendants became common landown-
ers. Minor dihqans became ordinary subjects.
In the towns that had been city-states headed by dihqans, pride of place was occupied
by the Muslim military-administrative elite which bought up the land and palaces of the
dihqans; this process can be quite clearly traced in Bukhara. Small towns where the dihqans
had lived were, with the loss of the latter's political role, transformed into villages, whereas
tradesmen and craftsmen gathered in the large towns to be near the wealthy customers who
received salaries from the new authorities.
By that time it was either Arabic that was most widely used there or else the New Persian
language brought by the Iranians, who had already become Muslim and been integrated
into the new state system. The use of Sogdian was becoming an anachronism; it remained
a language of local usage only, of no use in Baghdad to those who hoped to make a name
for themselves at the court of the caliph.
The suppression of the rebellion of al-Muqannac enabled the cAbbasids to start extend-
ing their dominion beyond the Syr Darya, an area under the sway of the Kaduk, whose
control extended as far as Kashghar at that time. At the beginning of his reign, al-Mahdi
sent troops to Ferghana, but the areas beyond the river were cut off from the caliphate by a
revolt and the Kaduk invaded the region of Samarkand. Possibly al-Mahdi demanded after
the revolt that the rulers of those regions accept his authority. But that account is suspect,
since it is difficult to believe that the king of Tibet declared his obedience to the caliph
as the Arab author asserts. In the year 791 the Uighurs defeated the Kaduk near Besh-
bal'ik, and Ghitrif b. cAta' seized the opportunity to send troops to expel the Kaduk from

43
ISBN 978-92-3-103467-1 Contents Dissent in the early cAbbasid period

Ferghana. Fadl b. Yahya al-Barmaki, who replaced him, secured from the Afshin (local
ruler) of Ferghana recognition of his status as a vassal of the caliph. At the same time he
sent troops against the Kabul Shah. His forces carried out a raid, seizing a great deal of
booty, and returned to Sistan. Fadl b. Yahya, himself an Iranian, managed to find a com-
mon language with the local dihqans and raised a large militia during his governorship
which he called the cAbbasid militia. A part of this militia was sent to serve in Baghdad.
Al-Muqannac' s rebellion obliged the governors of Khurasan to introduce order in the
spheres of tax assessment and water use. Fadl b. Sulayman al-Tusi (783-7) lowered the
level of the kharaj, which had been raised under the previous governor al-Musayyab, and
he eliminated misuse of water by the elite, who had seized extra shares of water in the Merv
oasis in order to water their fields. The use of water in other areas was also regulated. Fadl
b. Sulayman' s governorship also witnessed the start of the work, which continued until the
year 830, of building a wall around the entire Bukhara oasis.
This period of relative well-being for Khurasan came to an end with the arrival of the
new caliphal governor, cAll b. clsa b. Mahan, who mercilessly looted the country entrusted
to his care. The complaints sent to Harlin al-Rashid (786-809) remained unanswered, as
the caliph's closest counsellors ha d been bribed with gifts from cAll b. clsa. Yahya al-
Barmaki attempted to open the ruler's eyes to cAll's abuses, but the caliph was blinded
by the richness of the gifts that he received from the governor. Rebellions broke out in
the regions under the authority of cAH b. clsa in response to his abuses. In 797 in Sistan
there was a revolt of the Kharijites led by a local landowner, Hamza b. Adharak or cAbd
Allah, who traced his origins back to the legendary Iranian King Tahmasp. The son of
cAll b. clsa who was sent to deal with him was defeated. Other attempts to suppress the
rebellion also proved unsuccessful. Only the capital of Sistan, Zaranj, remained in the
hands of the caliph's governor: the rest of the region was in the hands of the Kharijites, who
did not recognize the authority of the caliph and therefore paid him no taxes (see below,
Chapter 2).
At the same time (in 799) Abu '1-Hasib Wuhayb b. cAbd Allah raised a revolt in
Nasa (near present-day Ashgabat). The rebels took possession of a considerable area of
Khurasan and even reached Merv, where the residence of cAll b. clsa was located. It was not
until 802 that cAll b. clsa managed to defeat and kill Abu '1-Hasib Wuhayb. An even more
dangerous rebellion occurred in the year 805. It was led by Raff b. al-Layth, allegedly the
grandson of Nasr b. Sayyar. At first, it involved no more than a romantic scandal; Raff
decided to marry a married woman illegally, was arrested, fled and, in order to be able to

44
ISBN 978-92-3-103467-1 Contents Stability under al-Ma'mun

marry, raised a rebellion among the c ayyars 1 of Samarkand. The townspeople, who were
incensed by c Ali's oppression, willingly supported him. With their assistance he defeated
the forces under the command of c Ali's son which had been sent against him. The people
of Samarkand were supported by all of Transoxania.
Harlin al-Rashid eventually came to accept the complaints levelled against c Ali b. Isa
and dispatched a new governor, who arrested him and launched a war against Raff. Having
witnessed the failure of all attempts to put down the rebellion, the caliph decided to lead
the struggle against it in person, but died in Tus on the way to Merv in 809.

The achievement of a degree of stability under

al-Ma'mun
On Harlin al-Rashid's death, the caliphate was divided de facto into a western half, ruled
over by the caliph al-Amin (809-13), and an eastern half whose governor was his brother
al-Ma'mlin. The latter adopted the simplest solution, lowering the kharaj by one-quarter
and opening talks with Raff. On receiving a pardon and having been appointed governor
of Transoxania, Raff surrendered and the rebellion petered out after the reduction in the
kharaj. Many landowners from Khurasan took part in the suppression of the rebellion,
leading their own military detachments. Among them were four brothers of the Samanid
family from the region of Balkh who, ten years later, were rewarded for their loyal service
to al-Ma'mlin (by then, caliph: 813-33) by being awarded lifelong control of what were
to become the appanages of Ferghana, Chach and Herat, thus establishing the Samanid
dynasty in Transoxania.
Under the rule of al-Ma'mlin (in 813 the forces of al-Ma'mlin overthrew and killed
al-Amin, but al-Ma'mlin remained in Merv for a further five years), the gradual decentral-
ization of the caliphate began. Al-Ma'mlin was the son of Harlin al-Rashid by an Iranian
slave girl, and therefore close to the Iranian elite which was in control of the administration
and the army. His residence in Merv further strengthened his close relationship with the
area. It was to members of the Khurasanian elite that he entrusted the administration of
extensive provinces. Thus in 821 Tahir b. al-Husayn was appointed governor of all Persia.
On his death, he was succeeded by his son and the Tahirid governors controlled the region
for some fifty years.
The empire of the c Abbasids, with its diverse ethnic and religious composition, main-
tained its unity primarily by force of arms, thanks to a huge army whose maintenance

1
Members of armed young fraternities among the townspeople, found in several cities of the eastern
Iranian lands and regions as far west as Syria at this and in subsequent periods (see below, Chapter 18)

45
ISBN 978-92-3-103467-1 Contents Stability under al-Ma'mun

absorbed the greater part of the budget. Under the cAbbasids, the all-Arab forces which
had directly established the authority of the conquering people were replaced by a profes-
sional army and by the militia of the Persian dihqiins supporting the dynasty who gradually
supplanted the Arab tribal levies, thereafter recruited only for major campaigns. Lacking
the support of any ethnic or social group (only the senior officials were genuinely loyal
to the empire, but they served the state machine rather than the caliph), the caliphs were
obliged to look for an armed force which would release them from reliance on overly
independent military commanders or the forces from Khurasan. This force was a guard
of professional slave soldiers (ghilmiin, mamiilfk, plurals of ghuliim, mamluk). Slaves had
been employed as bodyguards even under the first caliphs, but it was only at the end of
al-Ma'miin's rule that they became the nucleus of the caliph's army, when his successor
al-Muctasim (833-42) purchased 3,000 Turkish ghilmiin. Surrounded by the ghilmiin, later
caliphs eventually became playthings in the hands of their own slaves.
The incorporation of Khurasan and Transoxania into the caliphate assisted their inte-
gration into the wider Islamic world, increasing commercial and cultural exchange. The
intensive development of the Transoxanian towns and the growth of their populations from
the end of the eighth century onwards was a consequence of this. The formation of the
new Iranian-Muslim culture had, however, some negative effects. In the area of written
culture a considerable proportion of the literary heritage in Middle Persian, Khwarazmian
and Sogdian was lost as was, in the visual arts, a rich tradition of monumental painting.

46
ISBN 978-92-3-103467-1 Contents SECTARIAN AND NATIONAL MOVEMENTS

SECTARIAN AND NATIONAL

MOVEMENTS IN IRAN, KHURASAN AND
TRANSOXANIA DURING UMAYYAD AND
EARLY cABBASID TIMES*
F. Daftary

Contents

The Kharijite movement 48

Currents of Shflsm: the Kaysaniyya and the Hashimiyya 51

Heterodox Muslim and neo-Mazdakite movements: al-Muqannac, Babak, and others 53

The later development of Shicism:

the Twelvers, the Zayd!s and the Ismacilis 57

The beginnings of the disintegration of the cAbbasid caliphate in the east . 63

The protest movements touched upon above in Chapter 1 (Part Two) will now be consid-
ered in greater detail. Islam heralded a new social order, under whose banner all believers
belonging to different races or classes would theoretically enjoy equality. However, during
the earliest centuries of Islam when the Islamic empire was really an 'Arab kingdom', the
Iranians, Central Asians and other non-Arab peoples who had converted to Islam in grow-
ing numbers as mawalf, or 'clients' of an Arab lord or clan, had in practice acquired an
inferior socio-economic and racial status compared to Arab Muslims, though the mawalf
themselves fared better than the empire's non-Muslim subjects, the ahl al-dhimma ('people
of the book'). The mawalf, for instance, paid special taxes, often similar to the jizya (poll

* See Map 1.

47
ISBN 978-92-3-103467-1 Contents The Kharijite movement

tax) and the kharaj (land tax) levied on the Zoroastrians and other non-Muslim subjects,
taxes which were never paid by the Arab Muslims.
From an early date, the stage was thus set for prolonged antagonism between the Arab
rulers and their Iranian and other non-Arab subjects in the eastern Islamic lands. Indeed,
the superficially Islamized peoples of the Iranian lands - especially in the remote eastern
provinces of Khurasan and Transoxania (called by the early Arab geographers and his-
torians, Ma wara' al-nahr, or 'the land beyond the river [ Oxus]'), situated far from the
caliphal centres of power in Syria and Iraq - did not submit readily to Arab rule or even
to Islam for quite some time. Different religio-political currents of thought and sectar-
ian movements, often leading to popular insurrections, persisted until the early cAbbasid
and later times in Central Asia, Khurasan and other regions of the Iranian lands. They all
expressed opposition to the established caliphate, while many of the region's movements
manifested anti-Arab or even anti-Islamic sentiment, rooted in Zoroastrianism, Mazdakism
and other Iranian traditions.

The Kharijite movement

It was under such circumstances that Kharijism found some early following in Iran. More
importantly, the Iranian lands lent support to the cAlid cause and to Shicism, which had
itself originated as an Arab party opposed to the established caliphate. All the major
branches of the Shfites, namely the Kaysaniyya, Imamiyya, Zaydiyya and Ismaciliyya,
had acquired communities of followers by the ninth century in different parts of the Iranian
lands. Other sectarian groups, engaging in armed revolts, were specific to Transoxania,
Khurasan, Azerbaijan and a few other areas of Iran. The doctrines of these rather obscure
groups, designated generically as the Khurramiyya, were based on syncretistic currents of
thought which aimed to amalgamate indigenous Iranian religious traditions with aspects
of Islamic teaching, while these Khurrami groups basically remained anti-Arab and anti-
Muslim. The activities of many of these insurrectional groups, frequently supported by
the peasantry, were also rooted in specific socio-economic grievances of the villagers and
the smallholders against the dihqans (large landowners) who had assimilated themselves
more readily into the new Arabo-Islamic system, and often acted as the provincial caliphal
agents as well. From the second half of the ninth century, when the cAbbasids began to lose
their firm central control over the outlying lands of the caliphate, Iranian 'national' senti-
ment (if this rather modem concept may be applied in a medieval Islamic context) found
more successful channels of expression in the activities of certain dynasties, starting with

48
ISBN 978-92-3-103467-1 Contents The Kharijite movement

the Saffarids, which successfully challenged the hegemony of the cAbbasids and began to
reassert Iranian identity and culture, especially in the Samanid period (see below).
There is little reliable information on most of these sectarian and revolutionary move-
ments, mainly because very few contemporary sources, including the genuine literatures
of the sectarians themselves, have survived from this early formative period in Islam. The
later Muslim authors, including the historians and the early heresiographers such as al-
Ashcari (d. 935), al-Baghdad! (d. 1037) and Ibn Hazm (d. 1064), who wrote about several
aspects of these religio-political movements, were mostly Sunnis defending the legitimacy
of the historical caliphate and the orthodoxy of Sunni Islam. As a result, they treated all of
these opposition movements as heterodoxies or heresies. On the other hand, al-Nawbakhti
and al-Qummi, the earliest Shfite heresiographers who wrote during the final decades of
the ninth century and were better informed than the Sunni authors about the internal divi-
sions of Shfism, belonged to the Imami branch and as such were inclined to misrepresent
or refute the claims of the other Shicite groups. Indeed, these groups and movements have
to be studied mainly on the basis of hostile and ill-informed sources, which freely attribute
strange extremist ideas and antinomian practices to the sectarians. As a result, some of the
teachings of these sectarians, especially the doctrines of the Khurrami groups, may never
be clarified. There are also disagreements among contemporary scholars regarding the pre-
cise social composition and economic bases of some of these sectarian movements, though
much progress has been made in recent times. It is with these reservations in mind that we
shall now take a closer look at the major sectarian movements of the Iranian lands, espe-
cially Khurasan, and Transoxania during the late Umayyad and early cAbbasid periods.
Kharijism, the first schismatic movement in Islam, originated in Iraq in connection with
the prolonged conflict between cAli b. Abi Talib (656-61) and Mucawiya (661-80). The
Kharijite rebels formed a separate Muslim community and stressed Islamic egalitarian-
ism in their doctrinal position, holding that any Muslim believer who was morally and
religiously meritorious, including even a black slave, could be elected as the imam, or
leader, of the Muslim community. The egalitarianism of the Kharijites proved particularly
appealing to the Persian and other mawalf. Indeed, some of the anti-Arab sentiment of the
Iranians found expression in their revolutionary movement, which was also opposed to the
caliphates of the Umayyads and cAbbasids. The fundamentalist Kharijite insistence on the
correct Islamic conduct, however, led to a pronounced factionalism within the Kharijite
community itself, resulting in numerous Kharijite branches and sub-sects.
From early Umayyad times, Iraqi Kharijites began to seek refuge in Persia, spread-
ing their doctrines in different regions of the Iranian lands, especially in Sistan where

49
ISBN 978-92-3-103467-1 Contents The Kharijite movement

Kharijism remained the main sectarian movement for quite some time. 1 By the second
civil war, Kharijism had become firmly established in Persia, where different Kharijite
communities embarked on a prolonged programme of anti-caliphal insurrectional activi-
ties.
Initially, Kharijism in the Iranian lands was primarily related to the activities of the most
radical Azariqa branch of the movement. The Azariqa, who held that the killing of the
women and children of non-Kharijite Muslims was licit, had established several commu-
nities in Fars and Kerman as early as 686. Later, the Kharijite movement was reorganized
in Iran by Ibn Ajarrad, who may have been from Balkh. Little is known about the activities
of Ibn Ajarrad, who founded the Ajarida branch of Kharijism. Heresiographers name some
fifteen sub-sects of the Ajarida, which were specific to Iran and were more moderate in
their views and policies than the Azariqa. 2 The various Ajarida sub-sects were particularly
active, from 724, in Sistan and other eastern regions, where this form of Kharijism acquired
some indigenous foundations. The Thacaliba, one of the major sub-sects of the Ajarida,
contributed to the revolutionary turmoil of Khurasan during the late Umayyad period, also
lending temporary support there to Abu Muslim. 3 Subsequently, several Thacaliba splinter
groups survived for some time in and around Juzjan.
In the meantime, Sistan had continued to be the main Kharijite stronghold of the Ajarida
in eastern Persia. It was in Sistan that in 795 the major Kharijite revolt of Hamza b. Adharak
(or cAbd Allah) al-Khariji unfolded. 4 Hamza, the descendant of a noble Persian dibqan
and the founder of the Hamziyya sub-sect of the Ajarida, started his rebellious activities in
Zarang, the capital of Sistan. Responding to the financial grievances of the Sistani villagers,
Hamza successfully urged them not to pay the kharaj and other taxes due to the cAbbasid
caliph; he also had a number of caliphal tax-collectors killed in the region. Hamza mobi-
lized his followers into a large army and conducted anti-c Abbasid raids for some thirty
years until his death in 828.
Hamza al-Khariji was succeeded by others in the leadership of his movement. The
Hamziyya and other Ajarida sub-sects continued their rebellious activities in eastern Per-
sia until around the middle of the ninth century, when Yacqub b. Layth and his successor
in the Saffarid dynasty broke the military power of the Kharijite rebels and ended their

1
For an excellent survey of early Kharijism in Persia and the relevant sources, see Madelung, 1988, pp.
54-76.
2
See, for instance, al-AshcarI, 1929- 30, Vol. 1, pp. 93-100; al-Baghdad!, 1328/1910, pp. 72-82; al-
ShahrastanI, 1984, pp. 108-14; 1986, pp. 394-406.
3
Al-TabarI, 1879- 1901 , Vol. II, pp. 1989-97.
4
The fullest account of this Kharijite revolt is contained in the anonymous Tarr:kh-i Srstan [History of
Sistan], 1976, pp. 123-43. See also Bosworth, 1968, pp. 87-104.

50
ISBN 978-92-3-103467-1 Contents Currents of Shicism

importance as a sectarian movement. Nevertheless, scattered Kharijite communities sur-

vived for about a century longer in Sistan, Khurasan and other eastern Iranian lands.
The Ibadiyya - representing the moderate branch of Kharijism, which eventually found
its largest popular following in eastern Arabia and among the Berbers of North Africa -
also acquired some support in the east, mainly in Khurasan, during the late Umayyad and
early cAbbasid periods. However, these early Ibadi groups of the Iranian lands were ori-
ented, unlike the Ajarida, towards Iraq; and they do not seem to have been involved in any
rebelliousactivities.

Currents of Shicism: the Kaysaniyya and the

Hashimiyya
Of the various religio-political opposition movements in Islam, Shfism produced the most
lasting impact on the sectarian topography of the Persian lands, although the supremacy of
Sunni orthodoxy remained effectively unchallenged there through the cAbbasid and later
times. Shfism originated as an Arab party (shf a) opposed to the established caliphate. It
upheld the rights of the Prophet's cousin and son-in-law, cAli b. Abi Talib, and then mainly
those of the cAlid members of the Prophet's family (ahl al-bayt), to the leadership of the
Muslim community. Representing a unified and exclusively Arab party during its first half
century, Shicism entered a new phase of its formative period with the Kufan revolt of al-
Mukhtar, which was launched in 685 on behalf of cAli's son Muhammad b. al-Hanafiyya.
Al-Mukhtar's proclamation of this cAlid as the Mahdi (the divinely guided saviour imam
who would establish justice on earth and deliver the oppressed from injustice) proved very
appealing to the discontented mawalf, who were drawn in increasing numbers into the
Shfite movement. The mawalf, especially the early ghulat (extremists) among them, intro-
duced many ideas rooted in their Irano-Zoroastrian and other non-Islamic traditions into
Shfism, which left a lasting imprint on the movement's doctrinal development.
The Shicite movement of al-Mukhtar, which survived him and the demise of his revolt in
687, became generally known as the Kaysaniyya, accounting for the overwhelming major-
ity of the Shicites until the success of the cAbbasid revolution. When Ibn al-Hanafiyya
died in 700, the Kaysaniyya split into several groups, each having its own imam and devel-
oping its own doctrines. The bulk of the Kaysaniyya now acknowledged the imamate of
Abu Hashim cAbd Allah, the eldest son of Ibn al-Hanafiyya. This Kaysani majoritarian
sub-sect, known as the Hashimiyya, was the earliest Shicite group whose teachings and
revolutionary stance were disseminated in Persia, especially in Khurasan where it found
adherents among the province's mawalf as well as Arab settlers.

51
ISBN 978-92-3-103467-1 Contents Currents of Shicism

On Abu Hashim's death, the Hashimiyya themselves split into several groups. Two
of these groups had a major impact on the Iranian lands. One of the main factions of
the Hashimiyya recognized the imamate of cAbd Allah b. Mucawiya, a great-grandson of
cAli's brother, Jacfar b. Abi Talib. Ibn Mucawiya acquired many followers in the western
and southern parts of Persia after the collapse of his Kufan revolt in 744. Receiving broad
popular support from the Persian mawalf, Kharijites and other groups, Ibn Mucawiya estab-
lished himself at Istakhr, from where he ruled for a few years over Fars and other parts of
Persia. Ibn Mucawiya was eventually defeated by the Umayyads in 748; he then sought
refuge in Khurasan and was killed in Abu Muslim's prison. 5
The sectarian followers of Ibn Mucawiya, known as the Harbiyya and later as the
Janahiyya, expounded many extremist and gnostic ideas, which have been attributed mainly
to one cAbd Allah b. al-Harb. The heresiographers, indeed, ascribe a prominent role to this
enigmatic personality for introducing some key doctrines into Kaysani thought, including
the pre-existence of souls as shadows (azilla), the transmigration of souls (tanasukh al-
arwah) and a cyclical history of eras (adwar) and aeons (akwar). Some of the ideas of the
Harbiyya-Janahiyya were adopted by other early Shfite ghulat groups, and they were also
expounded by some of the Khurramiyya groups. 6 It is indeed possible that the Harbiyya-
Janahiyya supporters of Ibn Mucawiya in western Iran may have been partially recruited
from among the local neo- Mazdakites, who provided the backbone of the Khurramiyya
movement in the Iranian lands.
In the meantime, the main faction of the Hashimiyya had recognized the cAbbasid
Muhammad b. cAli, the great-grandson of the Prophet's uncle al-c Abbas, as Abu Hashim's
successor to the imamate. They held that Abu Hashim had personally bequeathed his rights
to the imamate to this cAbbasid relative. In this way, the cAbbasids inherited the party and
the propaganda organization of the Hashimiyya, which became the main instrument of the
cAbbasid movement, and eventually of the overthrow of the Umayyads. The Hashimiyya-
cAbbasiyya party, too, influenced the syncretic doctrines of the Iranian Khurramiyya, while
the murder of Abu Muslim in 755 sparked off a long period of insurrectional activity by a
host of Khurrami groups in Transoxania, Khurasan and other Iranian lands.

5 Al-TabarI, 1879- 1901 , Vol. II, pp. 1879-87.

6
Al-NawbakhtI, 1931 , pp. 29 et seq.; al-QummI, 1963, pp. 39 et seq.; al-Nashi', 1971 , pp. 37 et seq.;
al-RazI, 1988, p. 298; Halm, 1982, pp. 64 et seq., 240-74.

52
ISBN 978-92-3-103467-1 Contents Heterodox Muslim ...

Heterodox Muslim and neo-Mazdakite movements:

al-Muqannac, Babak, and others
A few obscure sectarian movements, with possible Khurramiyya connections, sprang up
in Khurasan in the milieu of the early c Abbasid dacwa (missionary movement) during the
final decades of the Umayyad period. Around 729 a dtFf (propagandist) named c Ammar
b. Yazid and nicknamed Khidash was sent to Nishapur and Merv to head a new c Abbasid
dacwa organization in Khurasan. Khidash expounded extremist doctrines and was eventu-
ally repudiated by the c Abbasid imam, Muhammad b. c Ali; he was arrested and executed
in 736. However, Khidash had acquired followers of his own, known as the Khidash-iyya,
who held that the imamate had passed to him from the time of his repudiation by the
c Abbasids; they also denied Khidash's death. Some heresiographers report that Khidash
taught the doctrines of the Khurramiyya and also permitted promiscuity; and they, in fact,
identify the Khidashiyya with the Khurramiyya of Khurasan. 7 The matter is unclear, but
it is possible that some of the Khidashiyya may have been recruited from among the neo-
Mazdakites of Khurasan.
There also appeared at this time the movement of Bihafarid the Magian, who was
a native of Zuzan and had a Zoroastrian background. Setting himself up, possibly as a
new prophet, at Khwaf, to the south of Nishapur, Bihafarld rejected many of the practices
of his contemporary Zoroastrians and preached syncretistic doctrines based on a type of
'reformed' Zoroastrianism and on certain aspects of Islam. He also revived Persian, and the
sources attribute a book to him written in that language. His ideas and social programmes
proved attractive to the peasantry, who rallied to his side, enabling Bihafarld to launch a
revolt around 747 in northern Khurasan. Bihafarld's innovative ideas soon became intol-
erable to the leaders of the traditional Zoroastrian establishment, who complained about
his heresy to Abu Muslim. They emphasized that Bihafarld was destroying both Zoroastri-
anism and Islam. Abu Muslim had Bihafarld captured in the mountains of Badhghis and
brought to Nishapur, where he and many of his followers, known as the Bihafarldiyya,
were put to death in 749. However, the Bihafarldiyya, who continued to expect Bihafarld's
return, survived in Khurasan until at least the end of the tenth century. 8
It was Abu Muslim al-Khurasani himself who had the greatest influence on a num-
ber of sects and their rebellions in the Persian lands which can be designated specifically
as Khurramiyya or Khurramdiniyya. It did not take the c Abbasids long after their victory to

7
Al-Nashi', 1971 , pp. 32-5; al-TabarI, 1879- 1901, Vol. II, pp. 1588-9; Sharon, 1983, pp. 165-73, 181-6.
8
Ibn al-Nadim, 1973, p. 407; GardizI, 1347/1968, pp. 119-20; see Yusofi, 'Behafarld', in: Elr, 1990, Vol.
IV, pp. 88-90, containing full bibliography.

53
ISBN 978-92-3-103467-1 Contents Heterodox Muslim ...

disclaim all connections with their Shicite and extremist Kaysani (Hashimiyya-
cAbbasiyya) antecedents. Indeed, soon after establishing their own caliphate in 749, the
cAbbasids became upholders of Sunni orthodoxy, persecuting the Shicites and their cAlid
leaders. They also turned against those diFfs and revolutionary commanders who had
brought them to power, including especially Abu Muslim, the founder of the Khurasanian
army and the chief architect of the cAbbasid victory. The treacherous murder of Abu Mus-
lim in 755, on the orders of the caliph al-Mansur, provided a unique impetus for the religio-
political activities of a number of syncretic Khurrami sects.
Many aspects of these Khurrami sects and their rebellious activities, which unfolded
during early cAbbasid times in many parts of the Persian lands and in Transoxania, remain
shrouded in obscurity. However, modern scholarship has generally corroborated theme-
dieval Muslim authors' identification of the Khurramiyya of early Islamic times with the
neo- Mazdakites - these were the remnants of the earlier Mazdakiyya who had supported
the socio-religious revolutionary movement of Mazdak for reforming Zoroastrianism in
Sasanian Iran during the reign of Kavad (488-531).
By early cAbbasid times, there were still many Zoroastrian and neo-Mazdakite commu-
nities scattered throughout many parts of Central Asia and the Iranian lands, especially in
the inaccessible mountain regions and the countryside of Khurasan, Tabaristan and Azer-
baijan. A common feature of these dissident religious groups, comprised mainly of the
peasantry and the lower social strata, was their anti-Arab feeling or Iranian 'national' senti-
ment. Thus they provided a suitable recruiting ground for all types of popular protest move-
ments; and they were particularly recruited into the conglomeration of religio-political
sects known as the Khurramiyya. The Khurrami groups were also receptive to syncretis-
tic influences; and, in Islamic Iran and Central Asia, they were especially influenced by
certain extremist and messianic doctrines taught by the Shfite ghulat belonging to the
Harbiyya- Janahiyya and Hashimiyya- cAbbasiyya parties of the Kaysaniyya. As a result,
Islamic teachings of an extremist nature came to be fused with Iranian dualistic traditions
and anti-Arab motifs, giving the Khurramiyya sectarian movement its distinctive (Irano-
Islamic) syncretic nature. The protests of the Khurrami groups, which resisted assimilation
into Sunni Islam, were also rooted in conflicts of class interests and in economic difficul-
ties. The sectarians had particular grievances against the existing tax system, especially the
assessment and collection of land taxes, as well as the local landowning class of dihqans
who had assimilated more readily into the emerging Arabo-Islamic socio-economic system
of the caliphate and often shared many of the privileges of the ruling class.
The widest allegiance among the (neo-Mazdakite) Khurrami communities of the Iranian
lands and Transoxama was gained by Abu Muslim. He acquired followers of his own,

54
ISBN 978-92-3-103467-1 Contents Heterodox Muslim ...

known as the Abu Muslimiyya or Muslimiyya, who split into several groups over time.
Abu Muslim evidently gained numerous neo-Mazdakite adherents during his lifetime; and
many heresiographers indeed identify the Khurramiyya with the Abu Muslimiyya, who
recognized Abu Muslim as their imam, prophet, or even an incarnation of the divine spirit.
As the symbol of Iranian self-assertion against Arab domination, Abu Muslim became the
figurehead of the Khurramiyya and his murder led to extended Khurrami revolts. 9
Khurasan was the first region of Khurrami revolts after Abu Muslim's murder; these
revolts frequently involved the idea of avenging Abu Muslim's death. Some of the Abu
Muslimiyya-Khurramiyya there now denied that their leader was dead and began to expect
his return to establish justice in the world. Others affirmed his death and held that the
imamate had now passed from Abu Muslim to his daughter Fatima. Later, Fatima' s son
Mutahhar came to be recognized as the imam and Mahdi by some of the Khurramiyya. In
755 the Zoroastrian Sunbadh (Sindbad), a former associate of Abu Muslim, launched the
first of these popular Khurrami revolts against the cAbbasids, as reported by many Muslim
historians. Sunbadh led an army of Khurrami rebels from his base at Nishapur to Rayy,
where his following increased substantially. He also received some support in Qumis and
the Tabaristan highlands. This rebellion was suppressed after seventy days by an cAbbasid
army, but the Sunbadhiyya movement survived for some time.
According to the later Seljuq author Nizam al-Mulk, the Sunbadhiyya comprised Maz-
dakites, Zoroastrians and Shfites. He also implausibly reports that Sunbadh aimed to
destroy the Kacba. The sources attribute various anti-Islamic and anti-Arab motives to
Sunbadh, who evidently predicted the end of the Arab empire, also holding that AbuMuslim
would return together with Mazdak and the Mahdi. Sunbadh's revolt and movement, based
on religious syncretism and the anti-Arab sentiment of the Iranians and receiving the pop-
ular support of the peasantry, set the basic pattern for the activities of other Khurrami
groups.
From early on, Khurrami rebellious activities and syncretic doctrines spread from
Khurasan to Transoxania. Is'haq the Turk, who may have been one of Abu Muslim's
dt{zs operating among the Central Asian Turks, was the leader of the first of such sec-
tarian movements in Transoxania which, like that of Sunbadh, bore the twin label of Abu
Muslimiyya and Khurramiyya. He, too, predicted the return of both Abu Muslim and
Zoroaster, and used religious syncretism to unify disparate anti-c Abbasid groups. Sub-
sequently, Is'haq's movement acquired a militant character in Central Asia under the lead-
ership of one Barazbanda.

9
The classic treatment of the KhurramI sects and revolts remains that of the late Sadighi, 1938; see also
Amoretti, 1975, especially pp. 481-2, 494-519; Daniel, 1979, pp. 125-56; Madelung, 1988, pp. 1-2, 63-5.

55
ISBN 978-92-3-103467-1 Contents Heterodox Muslim ...

Around the year 766 another anti-c Abbasid revolt of a sectarian nature, with obscure
religious motives, started on the eastern fringes of Khurasan. Led by one U stadhsis (U stadh
Sis), who may have claimed prophethood, the revolt received its main support from the vil-
lagers. From its initial base in the mountainous district of Badhghis (now in north-western
Afghanistan), where Ustadhsis was joined by some of the Bihafaridiyya, the insurrection
spread rapidly to the regions of Herat and Sistan, receiving further reinforcement from the
Sistan Kharijites. This revolt was repressed after a few years by the veteran cAbbasid gen-
eral Khazim b. Khuzayma, who killed some 70,000 of the rebels. Ustadhsis himself was
captured in the mountains of Badhghis and sent to Baghdad, where he was executed on
al-Mansur's order.
The most famous of these early anti-c Abbasid movements of the Khurrramiyya in
Khurasan and Transoxania was that of al-Muqannac, whose followers were commonly
designated as the 'wearers of white' (see above, Chapter 1, Part Two). The fullest account
of al-Muqannac and his movement was given by Narshakhi, the renowned local historian
of Bukhara. Suffice it to say that all the doctrines attributed to al-Muqannac by the here-
siographers and other Muslim authors (of course, these are universally hostile to him) are
generally anti-Islamic. According to al-Biriini, al-Muqannac even enjoined his followers
to observe the laws and institutions of Mazdak. The movement of al-Muqannac survived
in Transoxania after the suppression of his revolt in 779, and the Mubayyida continued to
await the return of al-Muqannac until the twelfth century.
The Khurramiyya movement had adherents in other parts of the Iranian lands, outside
Khurasan and Transoxania. In 778 the neo-Mazdakite Muham-mira, or 'wearers of red', of
Gurgan revolted, in alliance with the local Khurrami supporters of Abu Muslim, claiming
that Abu Muslim was still alive. Led by a grandson of Abu Muslim, they advanced as far
as Rayy before the rebellion was suppressed by an army dispatched by the governor of
Tabaristan. Later, in the reign of Harlin al-Rashid (786-809), the Khurramiyya launched
insurrections in Isfahan and other localities in central Persia.
The activities of the Khurramiyya reached their peak in the movement of Babak al-
Khurrami, whose protracted rebellion based in north-western Iran seriously threatened the
stability of the cAbbasid caliphate. As the leader of the Khurramiyya of Azerbaijan, suc-
ceeding Jawidan b. Shahrak, Babak consolidated his position in the mountainous district
of Badhdh, which served as his headquarters. Babak then mobilized his largely rural Khur-
rami following into a formidable fighting force and started his revolt around the year 816.
This revolt, lasting for more than twenty years, soon spread from Azerbaijan to the west-
ern and central parts of Iran. Numerous cAbbasid campaigns against Babak proved futile,
until success was attained by the general Afshin, appointed for this purpose in 835 as

56
ISBN 978-92-3-103467-1 Contents The later development of Shicism

governor of Azerbaijan by the caliph al-Muctasim (833-42). In 837 Afshin finally seized
Babak's fortress of Badhdh and repressed the rebellion. Babak himself was captured soon
afterwards and sent to Samarra, where he was executed with extreme cruelty in 838. Lit-
tle reliable information is available on Babak's specific teachings, which were allegedly
anti-Arab and anti-Islamic. Some of the sources even report that Babak, too, was expected
to restore the religion of Mazdak. 10 Scattered groups of the Babakiyya survived, awaiting
Babak's return, until after the tenth century.
Babak's rebellion was followed, in 839, by that of the Qarinid ruler of Tabaristan,
Mazyar, a recent convert to Islam. Muslim sources accuse Mazyar of reverting to Zoroas-
trianism and of conspiring with Babak against Islam, while al-Baghdadi states that his
rebel followers, the Mazyariyya, constituted a major branch of the Khurramiyya. How-
ever, Mazyar's anti-c Abbasid rebellion developed out of his financial conflicts with cAbd
Allah b. Tahir, the Tahirid governor of the east, although in his rebellious activities Mazyar
relied increasingly on the support of the local Zoroastrian and neo-Mazdakite peasantry.
Mazyar was defeated by the Tahirids and was then executed at Samarra in 840. Soon after-
wards, Afshin, too, was accused of anti-Islamic and treacherous activities and was put to
death on the order of al-Muctasim. The Muslim sources unjustifiably depict Babak, Mazyar
and Afshin as the joint protagonists of a grand anti-Arab conspiracy.

The later development of Shicism:

the Twelvers, the Zaydls and the lsmacilis
Even after the failure of the major Khurrami revolts of early cAbbasid times, scattered
Khurrami communities engaging in lesser and sporadic insurrections survived until later
cAbbasid times in various parts of Iran, especially in Azerbaijan, Tabaristan and Khurasan.
It is possible that, during the ninth century, some of the Khurramiyya joined the revolu-
tionary movement of the Ismacilis, particularly in Khurasan and Transoxania. Despite the
claims of Nizam al-Mulk and other Sunni authors hostile towards the Ismacilis, however,
Ismacilism should not be viewed as a continuation of the neo-Mazdakite Khurramiyya,
although the two movements shared a common enmity towards the cAbbasids. Needless
to say, as Shicite Muslims, the Ismaciliyya could not subscribe to the anti-Arab, and more
importantly, anti-Islamic teachings of the Khurramiyya.
Indeed, Shicism provided another type (like Kharijism, an Islamic type) of opposition to
the established caliphate. By 760 the remnants of the radical Kaysaniyya, who had earlier

10
Sadighi, 1938, pp. 229-80; Yusofi, 'Babak _KorramI', in El r , 1989, Vol. Ill, pp. 299-306, and full bibli-
ography.

57
ISBN 978-92-3-103467-1 Contents The later development of Shicism

been mainly aborted in the cAbbasid dacwa (see above, Chapter 1), had either disintegrated
or had joined the Imamiyya branch of Shfism, which had earlier been greatly overshad-
owed by the Kaysaniyya movement. The Imamiyya, who traced the imamate through a
Husaynid Fatimid line of imams in the progeny of al-Husayn b. cAli, began to acquire
prominence under the leadership of the Imam Jacfar al-Sadiq (d. 765), who firmly estab-
lished Imam! Shfism as a distinctive religious community on a quiescent basis.
Prescribing taqiyya (the precautionary dissimulation of religious beliefs), Jacfar al-
Sadiq further taught that the sinless and infallible Shicite imam did not have to rise against
the unjust rulers of the time, as believed by the early Kufan Shica and the contemporary
Kaysaniyya and Zaydiyya, even though the caliphate too belonged by divine right to the
Shfite imam. Refrainment from all anti-regime activity became the hallmark of the politi-
cally moderate Imamiyya, later designated as the Ithna cAshariyya or the Twelvers.
The legitimist Imamiyya branch of Shicism, with its anti-revolutionary quietism, had
already spread from its original Kufan stronghold to the garrison town of Qum, in central
Persia, in the time of Jacfar al-Sadiq, marking the initiation of the Imamiyya sectarian
movement in the Iranian world. An Arab clan of the Kufan Ashacira, or colonists, had
settled in Qum in the later Umayyad period, and by the end of the eighth century the local
descendants of these Ashacira had become ardent Imam! Shfites. Thus Imam! Shfism
was introduced to Persia by the Arab Ashacira, who dominated the religious scene in Qum
for some three centuries. Madelung has skilfully described the subsequent development of
early Imam! Shicism in the Iranian world. 11 Qum remained solidly Imam! and became the
chief centre of Imam! traditionalism in the ninth century. Later, the theological school of
Qum played an important role in the development ofTwelver Shfism. Qum also influenced
the development of Imam! (Twelver) communities in other parts of central Persia during
the ninth century, notably at Rayy, which remained the second most important Imam! city
there until the Mongol times.
In Khurasan, Imam! Shfism spread during the ninth century. An Imam! community
already existed in Tus when cAli al-Rida, the eighth imam of the Twelver Shica, died and
was buried there in 818. The cAbbasid caliph al-Ma'miin (813-33) had appointed cAIU
al-Ridi as his heir apparent as part of his conciliatory policies towards the Shfites and the
cAlids. Moreover, Nishapur became one of the earliest centres of Imam! thought in the
eastern Iranian lands, due mainly to the activities of al-Fadl b. Shadhan, a learned Imam!
traditionalist, jurist and theologian who died around 873. In Transoxania, the Imamiyya
were present from the later ninth century; and by the early tenth century, Imam! thought

11
Madelung, 1988, pp. 77 et seq.

58
ISBN 978-92-3-103467-1 Contents The later development of Shicism

of a somewhat independent nature was propagated in Central Asia, from Samarkand, by

Muhammad b. Masciid al-c Ayyashi.
The Ispahbadiyya Bawandids of Tabaristan were the first Iranian dynasty to adhere to
Imam! Shicism from the middle of the eleventh century. However, the Iranian Imamiyya
found greater protectors in the Buyids, who were originally Zaydis but in later times per-
haps leaned towards Imam! Shicism. By the end of the Buyids' tutelage of the cAbbasids
in the eleventh century, small Imam! (Ithna cAshari) communities of minority status were
widely dispersed throughout the Iranian lands and Central Asia. Mainstream Imam! Shfism
achieved its greatest success in the predominantly Sunni Iranian lands when, in 1501, it was
imposed as the official creed on Safavid Iran, while Sunni orthodoxy continued to prevail
in Central Asia.
The Zaydiyya, another major branch of Shfism, appeared as a sectarian movement
in the Iranian world during early cAbbasid times, though its impact there proved to be
somewhat marginal. Zaydi Shfism arose from an anti-Umayyad revolt that Zayd b. cAH, an
uncle of Jacfar al-Sadiq, had staged at Kufa in 740. The supporters of this abortive revolt,
the earliest Zaydiyya, retained the politically militant but religiously moderate stance of
the early Kufan Shica. Thus the Zaydiyya, by contrast to the Imamiyya, developed into a
revolutionary movement and the pretenders to the Zaydi imamate were expected to rise,
sword in hand, against the illegitimate rulers of the time.
The earliest activities of the Zaydiyya in the eastern lands, including the insurrection
of Zayd's son Yahya (d. 743) in Khurasan, did not lead to any lasting sectarian results.
The later spread of Zaydi Shfism in northern Iran was greatly helped by the emigration
of a number of cAlids to the coastal region along the southern shores of the Caspian Sea,
where they sought refuge from cAbbasid persecution both in the coastal lowlands and in
the mountains. In early cAbbasid times, Tabaristan ( Mazandaran), the most populous of
the Caspian provinces, was inhabited mainly by the Daylamites, who had not yet converted
to Islam. And it was in Ruyan and other areas of western Tabaristan that Zaydi Shfism,
based on cAlid rule and Daylamite aspirations for autonomy, began to spread from around
the middle of the ninth century. Many of this region's cAlid rulers in time came to be
acknowledged as imams and dif fs by the Caspian Zaydi community, which developed
independently of the Zaydi community of Yemen, another major stronghold of Zaydism. 12
In 864 the people of western Tabaristan revolted against the fiscal exactions of the
Tahirid governors of the eastern lands, and they invited the Hasanid al-Hasan b. Zayd
(d. 884) from Rayy to become their leader. Al-Hasan soon seized all of Tabaristan and

12
More than any other modern scholar, Madelung has investigated the Caspian ZaydI community in a
number of studies; see Madelung, 1965, pp. 153-220; 1988, pp. 86-92; Elr, 1980, pp. 181 et seq.

59
ISBN 978-92-3-103467-1 Contents The later development of Shicism

established Zaydi cAlid rule in the Caspian provinces, adopting the title of al-dacf ila'l-
haqq (He Who Summons to the Truth). The subsequent attempts of the cAbbasids and
the Tahirids to regain Tabaristan were repelled by al-Hasan with the local help of the Day-
lamites. However, al-Hasan's brother and successor, Muhammad b. Zayd, was killed in 900
in a battle with the Sunni Samanids, who temporarily extended their rule over the region.
In 914 Zaydi cAlid rule was restored in Tabaristan by the Husaynid al-Hasan b. cAli
al-Utriish, known as al-Nasir li '1-Haqq. Al-Nasir converted to Zaydism large numbers of
people who had not yet even embraced Islam; and, with their support, he reconquered
Tabaristan from the Samanids. Al-Nasir came to form a distinct community of the Caspian
Zaydiyya, known as the Nasiriyya. These were separate from the Qasimiyya adherents of
the school of the Medinan Zaydi imam al-Qasim b. Ibrahim (d. 860), whose teachings had
earlier been transmitted to northern Iran. The division of the Caspian Zaydi community into
the Gilite Nasiriyya and the Daylamite Qasimiyya proved permanent, also splitting cAlid
rule into two branches there. The Iranian Zaydiyya had their golden age under the Buyids,
who patronized the Zaydi cAlids of the Caspian provinces. It was also under the Buyids
that Rayy became an important centre of Zaydi learning. Zaydism does not seem to have
had any lasting success in Central Asia, while in Khurasan it acquired some temporary
support among the cAlids of Bayhaq. Indeed, the Caspian provinces remained the main
Iranian stronghold of Zaydi Shfism. By the early Safavid decades, the surviving Zaydi
communities of that region, too, had all gone over to Twelver Shfism.
Ismacilism, another major and revolutionary branch of the Shica, had a greater and
more far-reaching impact on the Iranian lands than the Zaydiyya movement, though its
success there was ultimately checked by Sunni orthodoxy assisted by the arrival of the
all-conquering Mongols. The Ismaciliyya, retrieving much of the revolutionary zeal of the
earlier Kaysaniyya and Khurramiyya, split off from the rest of the Imamiyya on the ques-
tion of the Imam Jacfar al-Sadiq's succession. Led by a line of imams descended from
al-Sadiq's eldest son Ismacil, the Ismacili dacwa was organized as a secret and revolution-
ary Shfite movement bent on uprooting the cAbbasids.
The central leadership of the early Ismacili movement soon came to be based for a
while in Khuzistan, in south-western Iran, from where dacfs were dispatched to various
localities. The efforts of these central leaders to transform the original Ismacili splinter
groups into a greatly expanded and united movement began to bear fruit around 873. It
was at that time that numerous Ismacili dacfs began to appear in many regions of the Arab
and Iranian worlds; and their converts soon attracted the attention of the cAbbasids and
Muslim society at large as the Carmathians or Qaramita, named after Hamdan Qarmat, the
chief local leader of the movement in southern Iraq. However, the name Qarmati came to

60
ISBN 978-92-3-103467-1 Contents The later development of Shicism

be applied indiscriminately also to the Ismacili communities outside Iraq. At that time, the
Ismacili dacwa was preached in the name of the absent Muhammad b. Ismacn b. Jacfar
al-Sadiq, the seventh Ismacili imam, whose return as the eschatological Mahdi was eagerly
awaited.
The Ismacili dacwa was extended during the 870s to the Iranian lands. And there, the
dacwa was initially established in Jibal or western Iran. Khalaf al-Hallaj, the first diff of
Jibal, set up his headquarters at Rayy, from where the dacwa spread to Qum, Kashan and
other areas of central Iran under Khalaf's successors. Meanwhile, the dacwa had become
active in Fars and southern Iran under the supervision of Hamdan Qarmat and his chief
assistant, cAbdan. The dacwa was officially taken to Khurasan during the first decade of
the tenth century, although earlier it had been introduced there on the personal initiative of
Ghiyath, one of the chief dtffs of Jibal. Abu cAbd Allah al-Khadim, the first chief dtff of
Khurasan, established his regional headquarters at Nishapur. The third dt{f of Khurasan,
al-Husayn b. cAli al-Marwazi, who had earlier been a prominent commander in the service
of the Samanids, transferred the regional seat of the dacwa to Merv al-Rudh, also spread-
ing Ismacilism to Talaqan, Herat, Gharchistan, Ghur and other eastern areas. Al-Marwazl's
successor, Muhammad b. Ahmad al-Nasafi (al-Nakhshabl), a native of the Central Asian
district of Nakhshab, settled in Bukhara and spread the dacwa throughout Transoxania, also
penetrating briefly the inner circles of the Samanid court. Al-Nasafi, a brilliant philoso-
pher, was also responsible for introducing a form of Neoplatonism into Qarmafi-Ismacili
thought. 13
In the Iranian lands, the Ismacili dacwa was originally addressed to the rural population,
and the first dt{fs in Jibal concentrated their efforts on the villagers around Rayy. By
contrast to their positive response to the neo-Mazdakite Khurramiyya movement, however,
the peasantry of the Iranian lands was not attracted in large numbers to the Shfite Islamic
message of the Ismacilis during the ninth century. The early realization of the movement's
failure to acquire a large popular following which could be led in open revolt against the
local authorities, as had been the case in the Arab lands where villagers and tribesmen had
converted to Ismacilism in large numbers, led to a new dacwa policy for the Iranian world.
According to this policy, implemented especially in Khurasan and Transoxania, the dtf fs
henceforth directed their efforts towards the elite and the ruling classes. This policy, too,
failed to have any lasting success, although Abu Hatim al-Razi (d. 934), the fifth dacf of
Jibal, did manage temporarily to win the allegiance of several amirs and rulers of Jibal and
the Caspian region; and, in Khurasan and Transoxania, numerous notables were converted,

13
For the early history of the Ismacili movement in Iran and Central Asia, see Stern, 1960, pp. 56-90;
Daftary, 1990, pp. 118 et seq., 131, 162-3, 165-9, 180, 192-4.

61
ISBN 978-92-3-103467-1 Contents The later development of Shicism

including the commander al-Marwazi, who himself became a chief local ditz there. The
brief success of this policy in Central Asia reached its climax in the conversion of the
Samanid amir, Nasr II (914-43), and his vizier through the efforts of the diff al-Nasafi.
This success could not be tolerated, however, by the Sunni cuzama' (religious scholars)
and their Turkish military allies in the Samanid state. They reacted by deposing Nasr II,
under whose son and successor, Nuh I (943-54), al-Nasafi and his chief associates were
executed in 943 and their followers massacred.
Meanwhile, the unified Ismacili movement had experienced a major schism in 899.
It was at that time that the movement's central leader, cAbd Allah (cUbayd Allah) al-
Mahdi, the future founder of the Fatimid caliphate, openly claimed the Ismacili imamate
for himself and his predecessors, the same central leaders who had organized and led the
movement after Muhammad b. Ismacil. cAbd Allah also explained that the movement had
hitherto been spread on the basis of Muhammad b. Ismacil' s role as Mahdi merely to protect
the true identity of the central leaders who were continuously sought by the cAbbasids.
The declarations of cAbd Allah split the movement into two factions. One faction, later
designated as Fatimid Ismacilis, accepted cAbd Allah's claims, upholding continuity in the
Ismacili imamate. A dissident faction, based in Bahrain and southern Iraq and lacking a
united leadership, refused to recognize cAbd Allah and his predecessors, as well as his
successors on the Fatimid throne, as imams; they retained their original belief in the role
of Muhammad b. Ismacil as Mahdi. Henceforth, the term Qaramita came to be generally
applied to these dissident sectarians, who never recognized the Fatimid caliphs as their
imams. 14 Within the Iranian lands, the Ismacilis of Jibal mainly joined the Qarmati faction,
which had adherents also in Azerbaijan and western Persia. In Khurasan and Transoxania,
both wings came to be represented, though the Qarmatis predominated until the middle of
the eleventh century. The dacfs al-Razi and al-Nasafi, who engaged in a complex scholarly
discourse, were Qarmatis. These dacfs of the Iranian lands, and Abu Yacqub al-Sijistani
who later led the dacwa in Khurasan and his native Sistan, played an important part in
developing the Ismacili-Qarmati thought of this early period, which left a lasting influence
on the intellectual activities of the later Ismacilis.
By the final decades of the eleventh century, the Qarmati communities of the Iranian
lands had either disintegrated or joined the Fatimid Ismacili dacwa. In 1094 the Persian
Ismacilis became the main supporters of the Nizariyya branch of Ismacilism, severing
all ties with the Mustacliyya branch, which continued to be led by the Fatimid caliphs.
The Nizari Ismacilis of the Iranian lands were soon organized by Hasan-i Sabbah into a
revolutionary force with numerous inaccessible mountain strongholds, reminiscent of the

14
See Daftary, 1993; ' Carmathians', in: El r , 1990, Vol. IV, pp. 823 et seq.

62
ISBN 978-92-3-103467-1 Contents The beginnings of the disintegration ...

strategy adopted by some of the earlier Khurrami groups. Being opposed to the alien rule of
the Seljuq Turks, the Iranian Nizaris launched an armed revolt against the Seljuq sultanate
and succeeded in asserting their control over various parts of Iran, especially in Daylaman
and Kuhistan in south-eastern Khurasan, until they too, like the cAbbasids, became victims
of the Mongol invasions and irrevocably lost their political power in 1256.

The beginnings of the disintegration of the cAbbasid

caliphate in the east
Unlike the Hashimiyya- cAbbasiyya sectarian movement which succeeded in supplant-
ing Umayyad rule, none of the other religio-political movements of the eighth and ninth
centuries discussed here could successfully challenge the hegemony of the cAbbasids in
the eastern lands of the caliphate. Moreover, none of the early anti-c Abbasid insurrections
resulted in the separation of any territory from the caliphal domains for any extended period
of time. As a result, the territorial integrity of the cAbbasid caliphate remained intact until
after the middle of the ninth century. In the meantime, important developments were tak-
ing place, both at the centre of caliphal power in Iraq and in the provincial peripheries,
which eventually brought about the fragmentation of the cAbbasid caliphate. It was under
such circumstances that independent dynasties, devoid of any specific religious affiliation,
starting with the Saffarids, appeared in the eastern Iranian lands, also initiating the revival
of Iranian sentiment and culture.
For almost seven decades after the establishment of the cAbbasid dynasty, Iran was
governed by various eastern governors appointed from Iraq. These governors remained
unswervingly loyal to the caliph at Baghdad, citing his name on coins and in the khutba
(Friday worship oration) and sending him taxes and tributes. Starting with the appoint-
ment in 821 of Tahir b. al-Husayn, four generations of the Tahirid family ruled for some
fifty years from Khurasan as governors of the lands east of Iraq. Many Tahirids also held
office in Iraq itself. Contrary to the views of some modern scholars, however, the Tahirids
cannot be regarded as the first autonomous dynasty of the Iranian world in their time. As
Bosworth has explained in many of his studies, 15 the Tahirids, too, remained loyal ser-
vants of the cAbbasids, respecting the constitutional rights of the caliphate. They were also
highly Arabized in culture and outlook, like many other landowning aristocratic Persians
who had fully assimilated into the Arabo-Islamic culture of the period. Nevertheless, it
may be admitted that the hereditary rule of the Tahirids, who were of Persian dihqan ori-
gins and tolerated the Persian language in their entourage, did at least indirectly encourage

15
For instance, Bosworth, 1975.

63
ISBN 978-92-3-103467-1 Contents The beginnings of the disintegration ...

the resurgence of Persian language and culture in their entourage. It was an altogether dif-
ferent matter with the Saffarids, the next dynasty to appear on the political scene in the
eastern Iranian world.
As a result of the problems created by the Turkish slave soldiers and their commanders
who had come to play an increasingly important role in the central affairs of the caliphate,
especially during the anarchy of the Samarra period, caliphal control over the outlying
provinces had become seriously weakened by the middle of the ninth century. This allowed
new political powers, based on military force, to assert themselves on the fringes of the
caliphate. It was also at this time that Zaydi cAlid rule was established in Tabaristan, and
the Ismacilis and the Zanj (black slaves) launched their insurrectional activities in Iraq
itself. But the Saffarids, based in Sistan, were the first of such major military powers to
appear in the Iranian world, establishing a dynasty and separating vast territories from the
cAbbasid domains. The disintegration of the cAbbasid caliphate and the rise of independent
dynasties, which revived Iranian 'national' sentiment, had now begun.
Yacqub b. Layth (867-79), known as al-Saffar (The Coppersmith), who founded the
Saffarid dynasty, was of plebeian origins and lacked specific religious convictions, though
he was accused of Kharijite leanings; the later author Nizam al-Mulk depicts him also (on
dubious grounds) as a crypto-Ismacili. He had gradually risen to a leading position in the
cayyar16 bands of Sistan, which drove out the Tahirid amir. In 861 Yacqub himself was
proclaimed amir of Sistan. He thereupon proceeded to consolidate his position within the
province before conducting a number of military campaigns in what is now Afghanistan
and against the Kharijites. Subsequently, Yacqub directed his attention against the caliphal
territories in Iran. In 873 he entered Nishapur and ended Tahirid rule in Khurasan; then
he seized Fars in 875 and came close to taking Baghdad itself. Yacqub died in 879 in
Khuzistan. Saffarid power reached its zenith under Yacqub's brother and successor, cAmr
(879-900). cAmr was eventually defeated, in 900, in Transoxania by the Samanids and sent
to Iraq where he was executed. Henceforth, the authority and the territories of the Saffarids
diminished rapidly, eventually becoming largely restricted to Sistan.
Another development of great importance that occurred during the final decades of the
ninth century was the revival of New Persian literature and culture, initiated through the
efforts of Yacqub b. Layth and his brother cAmr, who had court poets composing Persian
verse for the first time since the Arab invasion of Iran. Soon, the plebeian Saffarids were
also equipped with a royal Iranian genealogy. The early Saffarids, indeed, pioneered the
renaissance of a specifically Irano-Islamic culture based on the 'national' aspirations of

16
See Chapter 1, note 1.

64
ISBN 978-92-3-103467-1 Contents The beginnings of the disintegration ...

the Islamized Iranians, who had continued to be aware of their Iranian identity and culture
during the centuries of Arab domination. 17
The cAbbasids survived as the spiritual heads of the Islamic world, over which they no
longer exercised any political control. The rise of the Buyids in western Iran and in Iraq,
and their subsequent internal and dynastic strife, permitted the formation of a number of
Turkish dynasties in the east; dynasties like the Ghaznavids, the Karakhanids and, most
significantly, the Seljuqs, who now established their own rule over the Iranian lands. When
the Seljuqs entered Baghdad in 1055, ostensibly to liberate the cAbbasid caliph from the
Shfite Buyids' tutelage, a new Turkish period had started in the Islamic history of the
Iranian world. The Seljuqs became the new champions of Sunni orthodoxy and sought
caliphal approval in order to legitimize their own rule. Thus the cAbbasids were once again
permitted to survive.
The appearance of Turkish dynasties in the eleventh century also checked the rapid
resurgence of Persian culture. This renaissance had, however, become irrevocable by that
time. Nasir-i Khusraw, the renowned Ismacili philosopher and dtfz of Khurasan and
Badakhshan during the late eleventh century, composed all his works in Persian. He is
also ranked among the foremost Persian poets. Moreover, the highly Islamized Iranian
Nizaris of the Alamut period, from early in the 1090s, adopted Persian as the language
of their religious writings, an unprecedented choice for a medieval Shfite community.
Indeed, the antecedents of the anti-Seljuq revolt of the Iranian Nizaris can be traced not
only to the Shfite and anti-c Abbasid movement of the earlier Ismacilis but also to the
Iranian 'national' elements fostered by the Saffarids and other Iranian dynasties. The Turk-
ish rulers themselves were soon influenced by aspects of Persian culture; thus the learned
vizier Nizam al-Mulk composed his Siyasat-nama [Book of Statecraft] for the Great Seljuq
sultan Malik Shah in Persian. The Seljuqs were superseded by other dynasties in the Iranian
world, whereas the cAbbasid caliphate enjoyed a revival of power and survived in Bagh-
dad until 1258, mainly due to the importance of the caliph's moral authority for Sunni
Muslims. Yet in the end, the cAbbasids of Baghdad, too, succumbed to the all-conquering
pagan Mongols.

17
See Spuler, 1952, pp. 225 et seq.; Stern, 1971 , especially pp. 535-9; Bosworth, 1978- 79, pp. 59-75,
reprinted in 1982, article VIII; 1994; see also Rypka et al., 1968, pp. 126-71.

65
ISBN 978-92-3-103467-1 Contents The Oghuz

THE STATES OF THE OGHUZ,

THE KIMEK AND THE K'f PCHAK*
S. G. Agajanov

Contents

The Oghuz . 66
The Kimek 74
The Kipchak . 77

The Oghuz
During the ninth and tenth centuries, the nomadic Turkic Oghuz tribes formed a principal-
ity on the middle and lower reaches of the Syr Darya (J axartes ), in the Aral Sea region and
the area of the northern Caspian. There are a number of obscure points in the history of the
formation of the Oghuz people and principality in western Central Asia and Kazakhstan.
The late S. P. Tolstov considered the home of the Oghuz to be the deserts and steppes of the
Aral Sea region. In his view, they had lived there in ancient times before migrating from
western to eastern Central Asia. 1 In spite of its originality, however, this viewpoint did not
gain general acceptance.
Research in recent decades points to the conclusion that the Oghuz in western Central
Asia originally came from the eastern T'ien Shan region. Oghuz historical tales relate that
the headquarters of their supreme ruler or leader was at one time situated on the shores of
Lake Issyk-ktil. According to different versions of this legend, there was strife among the
Oghuz caused by the hostile relations between their ruler and his son, Oghuz Khan. In his

* See Maps 1 and 2.

1
Tolstov, 1948.

66
ISBN 978-92-3-103467-1 Contents The Oghuz

account of this old legend, the Persian historian Rashid al-Din, who lived at the end of the
thirteenth and the beginning of the fourteenth century, wrote that after a lengthy struggle,
Oghuz Khan seized his father's lands in the district of Talas. Oghuz Khan then managed to
subdue a large number of lands and regions, including Almallk and Alatagh. The Oghuz
subsequently launched major campaigns to the west and appeared on the borders of Tran-
soxania and Khwarazm, from where they penetrated to the lower reaches of the Itil (Volga).
The new Oghuz capital, the town of Yengi-kent, was built in the regions which they had
conquered, on the lower Syr Darya. 2
The historical traditions of the Oghuz, which go back to the early Middle Ages, include
a number of legendary episodes and superimpositions from a later period. On the whole,
however, they clearly reflect the main thrust of their migrations from east to west. In the
course of these migrations, which took the form of predatory incursions and wars, they
reached the borders of Europe. The movements of the Oghuz covered a vast area from
Semirechye and the western T'ien Shan to the Aral Sea and the northern Caspian. 3
Study of the medieval sources by scholars shows that the migrations of the Oghuz began
as early as the eighth century. In 742, following the decline and fall of the Kaghanate
of the Western Turks, considerable changes occurred in the history of Central Asia. A
confederation was formed of Turkic tribes - the Basmil, the Uighurs and the Kaduk -
whose leaders shared political power. The leader of the Basmil bestowed upon himself the
ancient Turkic title of Kaghan, while the leaders of the Uighurs and the Kaduk acquired
the title of Yabghu. In the year 744, however, the Uighurs and the Kaduk combined to
defeat the forces of the Basmil. The head of the Uighur tribes then became the supreme
Kaghan, while the leader of the Kaduk retained the title of Yabghu. More than twenty
years later, in 766, the Kaduk left eastern Central Asia and conquered Semirechye. 4 The
Oghuz tribes of western Central Asia also played a part in these tumultuous events. The
Kaduk conquest of the western T'ien Shan led to a conflict with the Oghuz of the Issyk-ktil
region. In the course of this struggle for power, many of the Oghuz apparently moved to
the southwestern regions of western Central Asia. To judge by some of the archaeological
evidence, the Oghuz crossed the middle reaches of the Syr Darya and the foothills of the
Karatau in this migration. 5 From there they gradually began to penetrate to the Aral steppes
and the northern shores of the Caspian. In the first half of the ninth century there are
references to the presence of the Oghuz on the boundaries of Khurasan, where they carried
out armed raids. Local governors of the Tahirid family were obliged to construct fortified

2
Kononov, 1958, pp. 39, 41; Mirkhwand, 1841 ; Rashid al-Din, pp. 410-12; Ne§ri, 1949, pp. 9, 10.
3
Agajanov, 1969, pp. 122-9.
4
Pritsak, 1953, pp. 307-401.
5
Levina, 1972.

67
ISBN 978-92-3-103467-1 Contents The Oghuz

outposts (ribats) in the region. cAbd Allah b. Tahir (828-44) had similar fortifications built
in Dihistan and Farawa in the area of present-day Turkmenistan. 6 By that time, Oghuz
leaders had already achieved political hegemony in the Aral Sea region, and at the end of
the ninth century, the borders of their domains stretched, according to Byzantine sources,
as far as the River Ural.
According to the Byzantine emperor Constantine Porphyrogenitus, the Oghuz formed
an alliance with the Khazars in c. 893-8 and defeated the Pechenegs, who lived between
the Volga and the Ural rivers. Most of the Pechenegs were driven from the areas in which
they had long been established; only a few of them wished to remain there and settled with
the people known as the Ghuzz (Oghuz). The fierce encounter between the Oghuz and the
Pechenegs described by Constantine was the culmination of a struggle which had started at
an earlier date. The leaders of the Oghuz, who were present on the borders of Transoxania
during the second half of the eighth century, initially combined with certain of the Karluk
and Kimek and attacked the Pecheneg-Kangars. The Pechenegs apparently headed a tribal
alliance which also included the Pechina, the Nukarda and the Bajgird. Unfortunately,
the sources provide practically no information about these peoples, most of whom were
probably Turkic, while others may have been Finno-Ugrian. One of the above-mentioned
tribes, the Baj gird, may be identified with elements of the modem Bashkirs. The legends of
the Bashkir tribes, the Buryazn, the Tangaur and the Usergan, recount that they originally
lived in the valley of the Syr Darya and in the Aral Sea region. Some ethnographers believe
that they left those regions in the eighth and ninth centuries and settled in the foothills of
the southern Urals and the Ural region. 7
The tenth-century Arab geographer and traveller, al-Mascudi, tells us that the Pech-
enegs, the Pechina, the Nukarda and the Bajgird occupied the steppe around the Aral Sea.
The Oghuz, together with some of the Karluk and the Kimek, engaged in a bloody struggle
against the Pecheneg confederation. The Pechenegs, Pechina, Nukarda and Bajgird, unable
to withstand the pressure of their enemies, left the Aral Sea region and moved down into
Asia Minor. Around 932-3 these four Turkic tribes entered the confines of the Byzantine
empire. 8 Judging by all the evidence, the events described by al-Mascudi must have taken
place between the middle and the end of the ninth century. 9
The wars between the Oghuz and the Pecheneg-Kangars with their allies were long
and hard-fought. Turkmen folk-tales tell of the savage battles between the Oghuz and the
It-Bejene people, whom scholars believe to be the Pechenegs. The leaders of the Oghuz
6
Al-Baladhuri, 1932, p. 320.
7
Kuseev and Shitova, 1963, p. 15.
8
Al-Mascudi, 1894, p. 180.
9
Agajanov, 1969, p. 129.

68
ISBN 978-92-3-103467-1 Contents The Oghuz

tribe of the Sahr or Salur played an active part in this struggle. Before the clash with the
Pechenegs, they occupied the foothills of Kazykurt and the upper reaches of the Badam
river along the middle reaches of the Syr Darya. In the struggle of the Sahr with the It-
Bejene people, success went now to one side and now to the other, with attacks being
launched by both. It eventually ended with the victory of the Oghuz, who occupied the
lower reaches of the Syr Darya and the Aral Sea region. Those of the Pechenegs who
remained after their defeat at the end of the ninth century were assimilated by the Sahr and
other Oghuz tribes.
In their fierce struggle against the Pechenegs, the Oghuz leaders depended for assistance
on some of the neighbouring Turkic tribes. Most of these tribes were Karluk from western
Semirechye and Kimek from central Kazakhstan. The Kimek tribes allied with the Oghuz
acquired some pastureland in the steppes of the Aral Sea region and the northern Caspian.
Apparently, these groups were in the course of time gradually assimilated into the Oghuz
people.
The next act in the bitter warfare between the Oghuz and the Pecheneg confederation
was played out at the end of the ninth century and the beginning of the tenth. As indicated
above, the Oghuz leaders inflicted a decisive defeat on the Pechenegs and seized their
lands on the lower reaches of the Volga and the Ural rivers. In this dogged fighting, they
enlisted the support of the Khazar Kaghanate, which had hostile relations with the Pech-
enegs. These historical events, which were accompanied by the mass migration of nomadic
tribes, ended with the arrival of the Oghuz on the borders of Europe and Asia.
The tenacious struggle against the Pecheneg confederation, waged over a period of
many years by the Oghuz, enabled them to consolidate their position around the most
powerful tribal leaders. For a long time, political power was held by the leaders of the Sahr,
one of the largest Oghuz tribes. The formation of the Oghuz principality in the steppes
probably began as early as the eighth century, after the destruction of the Kaghanate of the
Western Turks. However, the process can scarcely have been completed by the year 766.
The first reliable references to the Oghuz state are provided by Arab sources at the end of
the ninth century and the beginning of the tenth; according to al-Yacqubi, the Oghuz had
their own separate dominion (mamlaka) side-by-side with other Turkic tribes. 10 Similar
accounts may be found in the writings of Ibn al-FaqTh, who lists the Oghuz among the
'kingly' peoples. 11 Al-Yacqubi wrote his historical and geographic work in the year 891
and Ibn al-Faqih completed his work in 903. It is therefore possible that the rise of the

10
Al-Yacqubi, 1892, p. 295.
11
Ibn al-Faqih, p. 168.

69
ISBN 978-92-3-103467-1 Contents The Oghuz

Oghuz principality, with its capital in Yengi-kent on the lower reaches of the Syr Darya,
also dates from roughly this period.
The formation of the Oghuz principality occurred at the same time as the development
of an essentially new ethnic grouping. The core of the Oghuz confederation was originally
constituted by the so-called Ok Tughra Oghuz. 12 According to historical traditions, there
were initially 10,000 families in all. 13 Subsequently, however, their numbers were swelled
by other newly arrived and local ethnic components. In addition to the Pecheneg-Kangars
and other steppe tribes, they also incorporated Alans and As'i who had settled on the steppes
from the Aral Sea to the eastern shores of the Caspian. These last were of Inda-European
origin; they had adopted Turkic ways as a result of mixing with the Pechenegs. 14
The Oghuz also incorporated Turkic elements from Semirechye. These were chiefly
Khalaj, groups of Kaduk and other western Turkic peoples. They included elements of the
Yughra and the Charuk who had previously inhabited the valley of the Chu and the Talas.
The Yaruklugh, who are referred to as one of the Oghuz tribes in the eleventh century, were
probably descendants of the Charuk. 15
Individual components of the eastern Turkic peoples of the steppes also played a part in
the formation of the Oghuz. Chief among these were the Imek-Kimek tribes, the Bayundur,
the Imur and the Kay, most of whom inhabited the area between the rivers Ob and Irtysh.
They were probably subdivisions of those tribes which had become the allies of the Oghuz
leaders in their struggle against the Pecheneg-Kangar grouping. Finding themselves far to
the west after the victory over the Pechenegs, they were eventually incorporated into the
Oghuz.
The Oghuz were made up of a number of tribes composed of a large number of clans.
Mahmud al-Kashghari, the author of an eleventh-century Arabic-Turkish dictionary, asserts
that they had originally consisted of twenty-four tribes. Later authors, however, including

12
The precise significance of this expression remains unclear, but three main lexical components can be
identified. The first is the widely known term ok which literally means 'arrow' (or, in a broader sense,
'branch', 'clan', 'tribe'). The second component, tughra, occurred in the language of the Oghuz in the
eleventh century, meaning 'the seal of the khan, the monogram or decree of the khan'. The third compo-
nent is a common designation, probably with an ethnic content, for the entire grouping or tribal grouping.
It is worth pointing out that the monogram of the Seljuq dynasty (of Oghuz origin), which was also known
as tughra, depicted a bow and arrows. The symbolic content of the monogram was based on the notion of
the bow as a symbol of kingly power and of the arrows as a sign of dependency and submission of the
junior to the power of the senior. Hence it may be supposed that Ok Tughra Oghuz was the designation of a
grouping of tribes and clans whose distinguishing symbol was the image of an arrow. In other words, it was
an ethnopohtical union subject to leaders or rulers who enjoyed (or aspired to) the supreme prerogatives of
kingly power.
13
Rashid al-Din, pp. 410-12.
14
Al-Biruni, 1966, pp. 95, 96.
15
Al-KashgharI, 1917-19, pp. 304-7.

70
ISBN 978-92-3-103467-1 Contents The Oghuz

Marwazi, speak only of twelve Oghuz tribes. 16 The divergence between the sources may
be explained by the division of the Oghuz into two exogamous groups, the Buzuk and the
Uchuk, which were incorporated in the right and left wings respectively of their forces.
Each group consisted of twenty-four tribes and was in tum divided into two equal parts.
The Buzuk belonged to the right wing, the 'elders', whereas the Uchuk belonged to the
'younger' tribes of the left wing. The elders enjoyed great privileges, particularly in con-
nection with the election of the Khan, or supreme ruler.
The tribal and clan divisions of the Oghuz were known as boy, kok and oba. The term
boy usually denoted the tribe whereas oba and kok were applied to clan divisions. Oghuz
clans and tribes formed part of larger tribal groupings referred to as il. In medieval Arabic-
Turkish glossaries, this term normally has the meaning of 'people' or 'political grouping'.
According to al-Kashgari, in the eleventh century the Oghuz consisted of the following
tribes: Kin'ik, Bayundur, Yiva, Sahr, Afshar, Bektili, Bukduz, Bayat, Yazg'ir, Imur, Karab-
ulak, Alka-bulak, Igd'ir, Uregir, Tut'irka, Ula-yondulug, Ttiger, Jebni, Bejene, Yavuldar and
Yaruklug. 17 Various of these tribal names are still to be found among the Turkmen people
today.
From the ninth to the eleventh century, Oghuz tribes inhabited the vast area of western
Central Asia and what is now Kazakhstan. Their encampments were scattered along the
Irgiz, the Ural, the Emba, the Uil and on the right bank of Lake Aralsor. Isolated groups
reached the southern shore of Lake Balkhash to the east, where they held the impregnable
fortress of Gorguz. The Oghuz also inhabited the Aral Sea region, the valley of the Syr
Darya, the foothills of the Karatau and the Chu valley, but they did not always constitute
a majority of the population. Their numbers were greatest in the Aral Sea region, on the
eastern and northern shores of the Caspian and the lower reaches of the Syr Darya. They
bordered on settled agricultural regions in Khwarazm, Transoxama and Khurasan. In the
tenth century, the steppe stretching westward from the Khazar lands and the deserts of the
northern Caspian, the central, Zaunguzsk and south-eastern Karakum and the Kyzyl Kum,
were known as the Oghuz desert. Their camps and fortresses stretched as far west as the
southern Urals and the lower Volga. Moving along the Urals and as far as the left bank of
the Volga, the Oghuz tribes bordered on the lands of the Bashkirs and the Burtas.
The political and economic centre of the Oghuz principality was Yengikent, the 'new
settlement'. The town lay not far from the developed, cultivated oases of Khwarazm and
Transoxania, and also lay on the path of major caravan routes through the Kimek steppes

16
MarwazI, 1942, p. 29.
17
Al-KashgharI, 1917-1 9, pp. 56-7.

71
ISBN 978-92-3-103467-1 Contents The Oghuz

to the valley of the Sarysu, Kengir and Ishim, and the towns on the middle reaches of the
Syr Darya and the southern Urals.
Like most other nomadic groupings, that of the Oghuz was not monolithic. Al-Idrisi
testifies to the presence of several 'princes' or 'kings' among the Oghuz. 18 At the time, they
had an overall supreme ruler who bore the title of Yabghu. The power of the Yabghu was
hereditary, although he was formally considered to be 'elected' to the kingship. According
to the account given by the tenth-century Arab traveller lbn Fadlan, the Oghuz chose their
rulers and decided other matters at popular assemblies. Such assemblies were nevertheless
rarely held and it was the council of the nobility (kankash) which played the chief role
in everyday life. The Oghuz rulers were chosen from the leading paternal lines (urug)
according to the unwritten rules of customary law (tore). This was based on the privileged
access to power of the oldest member of the clan. 19
The Yabghus had their own heirs who bore the title of Inal. Tutors (atabegs) were
appointed for the education of these heirs. The wives of the Oghuz rulers bore the honorary
title of Khatun and played an important role in court life. The Yabghu had co-rulers or sub-
stitutes with the title of KOlerkin who wielded great authority. They participated in the
resolution of complex disputes, acting as supreme arbiters. The commander-in-chief of the
troops played an important role in Oghuz society. This commander (su-bashi) had a mili-
tary council, and the Oghuz so-bas hi" frequently meddled in tribal politics, at times openly
opposing the Yabghu. Between the ninth and the eleventh century, the Oghuz principal-
ity had only the most primitive administrative machinery. There were officials, including
tax-collectors who collected tribute from the nomadic and settled populations. The Khan's
collectors had their own mounted detachments which employed military force to collect
taxes in the event of a refusal to pay. Most of the Oghuz were simple nomads, referred to
as er, meaning 'person', 'man', 'warrior'. An er was a full member of society and not a
slave (kul) or bondsman (ki"rnak). An important role in the social and political life of the
Oghuz was played by the Khans and Iliks who governed the tribal units. They possessed
their own guards, who consisted of young, privileged slaves (ghulam, oglan). The next
rung on the social ladder was occupied by the Beg, whose power was transmitted on a
hereditary basis. There were various categories of Begs among the Oghuz, the most impor-
tant of whom were the Ulug Begs and the Begler Begs. The Ulug Begs controlled the clan
and tribal associations, whereas the Begler Begs commanded the right and left wings of
the army. 20

18
Al-IdrisI, pp. 108-9.
19
Tarr:kh-i Al-i Seljaq, p. 14.
20
Agajanov, 1969, pp. 108-16.

72
ISBN 978-92-3-103467-1 Contents The Oghuz

The political grouping of the Syr Darya Yabghus played a noteworthy part in the polit-
ical and military history of Inner Eurasia. The Oghuz conducted frequent campaigns and
raids against neighbouring regions, such as Khwarazm, Transoxania, Khurasan and Volga
Bulgharia, but there were also peaceful trading relations. The nomadic Oghuz acquired
herds of horses and other livestock in the neighbouring agricultural lands. They also sold
and exchanged skins, wool, pelts and other goods in the local markets, buying in return
chiefly grain, handicrafts and other products.
The appearance of the Oghuz in the historical arena of western Central Asia and eastern
Europe changed the balance of forces in western Inner Eurasia. In 965 a military treaty was
concluded by the Oghuz Yabghu and the Russian prince Svyatoslav against the Khazars;
a consequence of this military-political alliance was the destruction of the Khazar state,
which had been a rival of ancient Rus. 21
It would appear that Khwarazm also played a part in these events, providing military
assistance to the Khazar Kaghanate. Medieval Arab historians refer to the successes of the
Khazars and their Khwarazmian allies. The Khwarazmian armies managed to drive out the
'Turks' who had invaded the Khazar lands. However, a conflict soon broke out between
the allies, allegedly because of the Khazars' refusal to accept Islam. Miskawayh speaks
of an attack launched on the land of the Khazars by the army of the Rus. Judging by the
fact that the same date is given in the Arab and ancient Russian manuscripts, the account
must refer to the struggle of the Oghuz in alliance with a Russian force against the Khazar
Kaghanate. 22
The defeat of the Khazars in 965 was preceded by a lengthy struggle with the Oghuz.
Clashes between the Oghuz and the Khazars began on the western approaches to the
Mang'ishlak peninsula. The Khazars barred the access of the Oghuz to the rich, open Don
and Black Sea steppes. The international trading routes controlled by them, which ran from
western Central Asia and the Volga region to eastern Europe, also held a great attraction
for the Oghuz. All these factors determined the Yabghu's policy of establishing an alliance
with ancient Rus against the Khazar Kaghanate.
The defeat of the Khazar Kaghanate contributed to the growing military influence of
the Oghuz, whose political importance in eastern Europe was greatly strengthened after
the defeat of Volga Bulgharia. In 985 Prince Vladimir conducted a joint campaign with the
Oghuz against the Bulghars; according to ancient Russian chronicles, the prince's force

21
Tolstov, 1948; Dunlop, 1954.
22
Miskawayh, 1921 , p. 209; Ibn al-Athir, 1870, p. 418.

73
ISBN 978-92-3-103467-1 Contents The Kimek

sailed along the Kama in boats while their allies (torki) rode along the bank on their
horses. 23
At the end of the tenth century and the beginning of the eleventh, the Oghuz political
grouping started to decline. The main reason for this was the rebellion among the bulk of
the nomadic population, which began during the reign of c Ali: Khan (who came to power
around the middle or at the beginning of the second half of the tenth century) and was in
reaction to oppression by officials and the attempted introduction of a system of regular
taxation.
Discontent with the fiscal policy of the Yabghu was exploited by leaders of the tribes
in the Seljuq group, who migrated in the middle of the tenth century from the middle to
the lower reaches of the Syr Darya in the area of Jand and who led the rebellion against
the Oghuz rulers of Yengi-kent. They suffered a major defeat, however, and moved away
from the Aral Sea region to Transoxania and thence to Khwarazm and Khurasan. The
defeat of the Seljuq leaders at the head of these popular 'disturbances' helped to bolster
the power of Shah Malik, the son and successor of c Ali: Khan. In 1041 the Oghuz Yabghu
assembled a large army and seized neighbouring Khwarazm. Two years later, however, he
fell into the hands of the Seljuq leaders who had occupied northern Khurasan and western
Iran. The long struggle against the movement of rebellion and the clashes and war with
the Seljuqs undermined the Oghuz from within: in the middle of the eleventh century, the
Oghuz principality collapsed under the assault of the Kipchak, who had invaded the Aral
Sea and the northern Caspian regions. 24

The Kimek
These Kipchak, who put an end to the Syr Darya Yabghus, were a western branch of
the Kimek tribal confederation. The Kipchak cannot be completely identified with the
Kimek (Yemek), however, as their common origin was already in the distant past. Dur-
ing the eighth century or at the beginning of the ninth, the Kipchak entered the Kimek
tribal grouping. 25 The Kimek federation was originally composed of seven tribes: the Imur,
Imak, Tatar, Bayundur, Kipchak, Nilkar and Ajlad. Originally from the steppes of eastern
Central Asia, they migrated to the territory of present-day Kazakhstan. The migration of
the bulk of these tribes (or their constituent clans) took place after the defeat of the Uighur
Kaghanate in Mongolia, in the year 840, by the Kyrgyz of the Yenisei. Apparently some of

23
Povest' vremennykh let, 1950, p. 257.
24
Agajanov, 1969, pp. 151-8.
25
Kumekov, 1972, pp. 35-48.

74
ISBN 978-92-3-103467-1 Contents The Kimek

them, in particular the Imur and the Bayundur, joined the Kimek tribal confederation while
others were absorbed into the Oghuz.
The main area in which the Kimek initially resettled was the Irtysh river steppes. Up to
the middle of the eighth century, they lived with the Turkic tribes of the southern Altai and
the Tarbagatay to the south and the Kyrgyz of the Yenisei to the east. At some time during
the second half of the eighth century or at the beginning of the ninth, the Kimek clans
and several tribes moved to north-eastern Semirechye and the foothills of the Dzhungar
range, while at the same time, the Kipchak tribes of the Irtysh migrated southwards and
westwards. 26
From the ninth to the eleventh century, the Kimek were more densely concentrated in
the basin of the middle Irtysh and in north-eastern Semirechye. Individual Kimek groups
and a large proportion of the Kipchak occupied the steppes of central Kazakhstan and the
northern Lake Balkhash region, extending as far west as the Aral Sea region and the south-
ern Urals. On the middle reaches of the Syr Darya, they roamed the area of Sawran and the
town of Turkistan, while their eastern borders stretched to the Tarbagatay mountains and
the Dzhungarian Alatau.
Up to the middle of the seventh century, the Kimek, along with other steppe peoples, had
been part of the Kaghanate of the Western Turks. After its collapse in 656, they gradually
developed into an independent tribal confederation. This process received considerable
impetus during the ninth century from the fall of the Uighur Kaghanate. The head of the
Kimek, who had previously held the modest title of shad tutuk, was subsequently called
the Kaghan. According to Arab and Persian writers of the ninth to the twelfth century, the
Kaghan enjoyed considerable power, 27 appointing the leaders of tribes, referred to in the
sources as mulak (kings). According to al-Idrisi, power in the clan of the Kimek rulers was
transmitted on a hereditary basis. The supreme ruler, the Kaghan, had eleven 'stewards'
whose duties were also transmitted from father to son, 28 and he and his court nobles resided
in a capital situated in the valley of the Irtysh. Al-ldrisi records that the Kaghan possessed
his own officials, a hajib (palace chamberlain) and a vizier, the head of the administration,
and that the Kimek nobility wore costly garments made of red and yellow silk. 29
The Kimek principality, formed at the end of the ninth century and the beginning of
the tenth, was divided into a number of domains like the later ulus. The rulers of these
appanages, who were the descendants of hereditary tribal rulers, received their lands from

26
References to the resettlement of the Kimek and Ki'pchak tribes may be found in the work oflbn al-FaqTh,
in the Hudad az-c alam, and in the writings of GardizI and al-IdrisI; the latter provides the corresponding maps.
27
Hudad az-cazam, 1930, p. 68.
28
Ibid., p. 18.
29
Al-IdrisI, pp. 68 et seq.

75
ISBN 978-92-3-103467-1 Contents The Kimek

the Kaghan in return for military service. Each domain supplied mounted and infantry
detachments for campaigns or in the event of a sudden attack. Although the leaders of these
groupings were subject to the Kaghan, the most powerful of them were semi-independent
'kings'. Like the supreme ruler, they also had fortress-residences which were usually
located in elevated positions. There they maintained contingents of their forces and stored
their treasures and food supplies. 30
Some Kimek groups moved for the winter to the steppes between the Ural and the
Emba and spent the summer as nomads in the area of the Irtysh, especially when there
were severe winters in what is now eastern and central Kazakhstan. Some of the Oghuz
also moved to pastures in the Kimek country near Lake Mankur, probably in the foothills
of the Alatau. Like the Kipchak and the Oghuz, the Kimek bred horses, sheep, goats, oxen,
cows and camels. Sheep, in particular, played an important part in their economy. Al-
ldrisi, describing the life of the nomads, writes that they 'used fat instead of vegetable oil
and tallow for lighting'. The horses of the steppe-dwellers were noted for their hardiness
and their ability to adapt to the harsh conditions of the arid zone. Gardizi refers to the
huge herds of horses raised by the Kimek, and Al-ldrisi notes that the nomads preferred
horsemeat to beef or mutton and made koumiss (a drink of fermented mares' milk). 31
The Kimek also possessed cattle, i.e. cows and oxen; these tended to be owned by semi-
sedentary elements, although oxen were also used as draught animals. The steppe-dwellers
usually harnessed them to carts on which they placed their yurts (wooden-framed tents
covered with felt). Like the Oghuz and the Kipchak, the Kimek hunted furry animals such
as the fox, marten and beaver, and further took the pelts of sable, ermine and predators
like tigers and snow leopards. The fur and hides of wild animals and the meat and skins
of domestic livestock were sold or exchanged at points adjacent to the settled lands in the
south.
The Kimek engaged to some extent in agriculture, but those involved must have been
semi-sedentary and settled groups. Al-ldrisi, describing this section of the Turkic popula-
tion, writes that although they were 'nomadic, they till the soil, sow and harvest'. 32 They
settled on the land for various reasons, but chiefly on account of cattle plague in hard win-
ters, outbreaks of epizootic disease or when the cattle were driven off by hostile tribes.
The semi-sedentary groups of the Kimek and other steppe peoples mainly sowed millet,
although al-ldrisi reports that they also grew wheat, barley and even rice. 33 According to
Fakhr al-Din Mubarakshah, the settled Turkic population also tended vegetable gardens

30
Al-IdrisI, pp. 68 et seq.
31
Bartol'd, 1973, Vol. 8, pp. 28, 45; al-IdrisI, pp. 69 et seq.
32
Al-IdrisI, p. 69.
33
Ibid., pp. 66, 69.

76
ISBN 978-92-3-103467-1 Contents The KYpchak

and even vineyards, 34 yet agriculture must have been underdeveloped and could scarcely
have met their own needs.
In addition to raising cattle, hunting and agriculture, the Kimek, like the Oghuz and
the Kipchak, practised certain crafts. The skins of domestic animals were used to make
various types of footwear, vessels, quivers, bow covers and horses' harness. The women
made felt from wool, wove clothes and produced felt in large pieces and pile-less rugs. Furs
were also used for clothes. The manufacture of weapons such as bows, quivers, arrows and
spears occupied an important place among household crafts. According to al-Idrisi again,
the Kimek and the Oghuz were skilled in iron-working and knew how to make beautiful,
refined ornaments from gold and silver; the Kimek ruler wore a golden crown and clothes
sewn with golden thread. 35
In the ninth and tenth centuries, the Kimek state was one of the strongest nomadic pow-
ers in Central Asia, but it gradually began to decline when the system of semi- independent
domains sapped the authority of the Kaghans, so that the Kimek tribal entity collapsed,
unable to withstand the pressure of neighbouring nomadic peoples.

The K"lpchak
The history of this period is rich in major events that have not been sufficiently investi-
gated. Minorsky, Bartol' d and others have linked these events with the external policy of
the Liao empire in northern China, that of the probably Mongol Kitan. Drawing on the
accounts of Marwazi and other medieval writers, they take the view that Kitan expansion
was responsible for provoking a chain reaction of migration among the peoples of Inner
Asia. This wave of migration began in eastern Central Asia and in the Far East with the
movement of the Kun tribe, which had been attacked by the Kay people. The displaced
Kun invaded the lands of the Sari tribe, which some scholars believe to have been a group
of the Kipchak. Apparently, this wave of migrations affected the Kimek tribes of the Irtysh
and north-eastern Semirechye and further drew in the Kipchak, who had pressed against
the neighbouring Oghuz in the Aral Sea region, the northern Caspian and on the lower
Volga, so that the Kipchak invaded the basin of the lower Syr Darya and seized the region
between the Volga and the Ural rivers.
Medieval sources contain scant information on the early history of the Kipchak tribes.
The ethnonym is first encountered in the seventh century, when the people were living on
the upper reaches of the Irtysh and the adjacent steppes of what is now eastern Kazakhstan

34
Fakhr al-Din Mubarakshah, 1927, p. 41.
35
Al-IdrisI, pp. 68 et seq.

77
ISBN 978-92-3-103467-1 Contents The KYpchak

and the southern Altai. In the tenth century, part of the Kipchak population was to be found
in areas of central Kazakhstan and to the north of the Aral Sea on the north-eastern edges
of the Kara Kum, reaching western Siberia in the east. 36
Initially, the Kipchak were subject to the Kimek Kaghans, but gradually formed an
independent ethno-political grouping. By the middle of the eleventh century, hegemony in
the steppes of western Central Asia, Kazakhstan and the lower Volga region had passed
to the Kipchak after they defeated the Oghuz and seized their lands. Their advance was
also directed towards the borders of Khwarazm and the lower Oxus region. According
to the Ghaznavid historian Bayhaqi, the Kipchak, Yigrak and Kujat settled in the steppes
bordering on Khwarazm. 37 Judging by several historical reports, the Yigrak (Igrak) were a
part of the Oghuz who had gradually become integrated with the Kipchak. The origin of the
Kujat is unclear, although they may have been one of the eastern Turkic tribes incorporated
in the Kipchak. In the middle of the eleventh century, during the Khwarazmian struggle for
political independence from the Ghaznavids, the Kipchak, Yigrak and Kujat were drawn
into the service of the Khwarazm Shahs as auxiliaries, with the Kipchak leaders acquiring
particular prominence in the Khwarazmian army. 38
Towards the end of the eleventh century, the Kipchak appear on the map of al-Kashgari
as inhabitants of the Aral Sea and the northern Caspian regions, occupying this region
alternately with the Oghuz tribes whom they had subjugated. The south-eastern borders of
their territory extended as far as the neighbourhood of Taraz (present-day Jambul), where
the fortress of Kenjak Sengir was built. To the east, the camps and pastures of different
groups of Kipchak extended as far as the Irtysh and the western slopes of the Altai. 39
The collapse of the Oghuz and Kimek states in the middle of the eleventh century con-
tributed to the consolidation of the tribes within the Kipchak federation, whereas previ-
ously, small, uncoordinated groups of Kipchak occupied a vast expanse from the Irtysh to
the Volga. They then moved far to the west, taking control of the south Russian and Black
Sea steppes, so that in eastern sources this entire region became known as the Dasht-i
Kipchak, and in ancient Russian sources as Polovetskoye Pole. 40 This westward and south-
ward migration of the Kipchak from the east was accompanied by the formation of a new
ethno-political grouping. In addition to groups of Oghuz, other Turkic elements entered
the Kipchak federation, whose integration with the Kipchak was assisted by the similarity
and cognate status of their languages and by the similarity of their economy and social

36
Istoriya Kazakhskoy SSR. S drevneyshykb vremen do nasbykh dney, 1979, Vol. 2, pp. 50 et seq.
37
Abu '1-Fadl BayhaqI, 1969, pp. 153, 827.
38
Ibid., pp. 59-70.
39
Istoriya Kazakhskoy SSR. S drevneyshykh vremen do nashykh dney, 1979, Vol. 2, p. 52.
40
Polnoe sobranie russkikh letopisey, 1962, Vol. 1, p. 152.

78
ISBN 978-92-3-103467-1 Contents The KYpchak

structure. Thus the Kipchak absorbed large groups of Yemek (Kimek) who had migrated
from the region of the Irtysh and Semirechye as far as the borders of Europe. According
to a twelfth-century Persian encyclopaedist, the town of Saksin in the region between the
Volga and the Ural rivers was subject to attacks by the Kipchak and the Yemek, 41 and
reports mention that some of the latter had penetrated as far as the Kama river at this
period. 42 These groups were probably gradually assimilated by the Turkic population of
Volga Bulgharia.
The Kipchak federation also included the Bayundur, who were of Oghuz or Kimek
origin; and the Baya'ut, the Kangh and the tribe of the Urani, some of whom served in the
guard of the Khwarazm Shahs, were also to be found in it. 43 The Baya'ut were probably
of Kimek origin, but the Kangh tribe was a component of the Pecheneg-Kangars. Other
steppe tribes and peoples also lived in the Dasht-i Kipchak, including some of the ancient
Bashkirs and the so-called Uighurs, living to the north and east of the Aral Sea. Islamic
historians of the thirteenth and fourteenth centuries name a whole series of Kipchak tribes.
From the eleventh to the beginning of the thirteenth century, processes of ethnic assim-
ilation and consolidation of considerable complexity took place throughout the Dasht-
i Kipchak. The western Kipchak tribes absorbed people of Oghuz, Pecheneg, ancient
Bashkir, Bulghar and other origins; the eastern Kipchak merged with the Oghuz-Kimek,
Karluk, Kara Khitay and others, but after a comparatively lengthy period of ethnic devel-
opment, they all acquired the common ethnonym of 'Kipchak'.
There was no uniform administrative and political system in the Dasht-i Kipchak. The
largest Kipchak groupings were in the region of the Dniepr, the Dniestr, the Ural and the
lower Volga; such confederations or hordes were most frequently referred to by the Old
Turkic term il, whereas the ethno-terntorial groupings were known as ulus. In this same
period, there were also a number of groupings of varying sizes in the eastern Dasht-i
Kipchak, most of which were led by a Khan, and also in the Aral Sea and the northern
Caspian regions along the lower and middle reaches of the Syr Darya and in what is now
eastern Kazakhstan. The pasturelands of each of these groupings were strictly delimited;
migratory movements were conducted in accordance with established custom and with the
permission of the principal leaders and elders of the tribe. The Kipchak tribes were led by
their Begs, and the tribal groupings by the hereditary families of Khans, who had their own
military forces, servants and slaves.

41
Agajanov, 1969, p. 162.
42
Polnoe sobranie russkikh letopisey, 1962, Vol. 1, p. 389.
43
Al-NasawI, 1895, p. 44; JuwainI, 1958, p. 465.

79
ISBN 978-92-3-103467-1 Contents The KYpchak

The appearance of the Kipchak on the historical stage altered the balance of political
forces in the Kazakh steppes, western Central Asia and eastern Europe. Their southward
movement further affected the agricultural regions of Khwarazm, Gurgan and Khurasan. At
the beginning of the fifth decade of the eleventh century, Khwarazm was taken over by the
Seljuq leaders, who defended here the last Oghuz Yabghu, Shah Malik of Jand. This victory
of 1041 helped to shatter the power of the Oghuz, and the neighbouring Kipchak did not
fail to take advantage of this; they seized the Aral Sea region, and by way of Mang'ishlak
and the U styurt plateau surged into eastern Europe. The K'ipchak now occupied a vast
expanse of steppelands in what is now western and southern Kazakhstan, apparently by the
mid-eleventh century, when we find references in the contemporary sources to important
K'ipchak rulers living on the lower reaches of the Syr Darya and in the Aral Sea regions with
their centre at Jand. According to the historian Ibn al-Athir, Seljuq forces from Khurasan
conducted a campaign against them in 1065, when Sultan Alp Arslan brought a large army
to the region of the Kara Kum which had been seized by the K'ipchak and the Turkmens
who had joined them. Fearful of attack, the supreme Khan of the K'ipchak concluded a
peace treaty and submitted to the Seljuq ruler.
The French Altaicist and Sinologist Paul Pelliot advanced the hypothesis that a large
proportion of the eastern K'ipchak were ruled by the Ilbari dynasty in the twelfth century
and at the beginning of the thirteenth. 44 One member of this ruling family was Ulugh
Khan who, in his youth, had been a slave in northern India but subsequently became sultan
of Delhi in the thirteenth century. The historian Jiizjani writes that, before being taken
prisoner, he considered himself as 'the Khan of the Ilbari and the Shah of the Yemek' .45
According to the same account, the father of Ulugh Khan was 'in Turkistan, chief among
the tribe of the Ilbari and bore the title of Khan'; his grandfather had been a member of the
clan of Abar Khan of the Ilbari, who had ruled 'over 10,000 families' .46
In the twelfth century and at the beginning of the thirteenth, the K'ipchak Khans success-
fully competed with the Khwarazm Shahs, who had built up a powerful empire in western
Asia (see below, Chapter 7). There were struggles between them for control of the towns
along the lower and middle Syr Darya through which passed the trade routes from east-
ern Europe to Central Asia. The first shah of the line of Anushtegin, Sultan Qutb al-Din
Muhammad (ruled as a Seljuq vassal 1097-1127), was more concerned with the protection
of the northern frontiers of his domains than with decisive operations against the warring
nomads of the Dasht-i K'ipchak, apparently because of threats from the expanding and

44
Pelliot, 1920.
45
JUzjanI, 1881- 97, p. 1294.
46
Ibid., p. 800.

80
ISBN 978-92-3-103467-1 Contents The KYpchak

vigorous Kipchak. Ats'iz b. Muhammad, who replaced Qutb al-Din Muhammad as shah in
1127, conducted a military campaign against Mang'ishlak, when the Khwarazmian forces
defeated the K'ipchak and their Oghuz allies and occupied the fortress of Jand. In 1133,
moreover, Ats'iz inflicted a decisive defeat on the K'ipchak Khan, who 'enjoyed the greatest
respect among the infidels' .47 This led to serious disturbances among the nomadic tribes in
the area. This K'ipchak confederation was apparently headed by a dynasty from the Urani
tribe. Sources from the end of the twelfth century and the beginning of the thirteenth men-
tion a whole series of K'ipchak nobles from the Urani tribe who played an important role in
political events on the borders of Khwarazm. Among these are the names of Alp Kara Urani
and Kad'ir Buku Khan - the latter's daughter was married to the Khwarazm Shah Tekish.
This princess (Terken Khatun) had a brother called Gayir Khan Inalchik, one of the pre-
tenders to the title of supreme ruler of the western part of the Dasht-i K'ipchak. Inalchik,
whose sobriquet was Alp Derek, contended for power with Kad'ir Buku Khan (also known
by the name of K'iran or Ikran), who considered himself to be the head of the K'ipchak of
what is now southern and western Kazakhstan. The Khwarazm Shahs took every oppor-
tunity to stir up internecine strife among the K'ipchak leaders in order to weaken them:
they exploited the fierce struggle of the K'ipchak leaders for control of the town of Signak
(the main centre of the K'ipchak domains in the Syr Darya steppes) and the fierce disputes
between Kad'i'r Buku Khan and his nephew Alp Derek.
In 1195 Tekish launched a campaign in the direction of Signak against Kad'i'r Buku
Khan. However, an event occurred during the battle which was unforeseen by the Khwarazm
Shah and which decided the fate of his military campaign. The rulers of Khwarazm had
long been accustomed to recruiting K'ipchak for military service, and many nomad lead-
ers held high posts and titles in their state. As has been mentioned above, K'ipchak from
the Urani tribe formed a guard for the ruler of Khwarazm. During the campaign of 1195,
however, the K'ipchak troops in the Khwarazm Shah's army were reluctant to fight against
their kinsmen, and at a decisive moment in the ensuing battle, they betrayed Tekish and
went over to the side of Kad'i'r Buku Khan. The Khwarazm Shah was defeated, but even-
tually managed to subdue S'ignak, using the forces of Gayir Khan Inalchik, who had been
appointed governor of the town of Utrar. However, the anarchic nomads continued to chafe
against control from Khwarazm. In 1215 they rose in rebellion in the neighbourhood of
Jand and S'ignak, but were severely defeated; the K'ipchak who survived the defeat were
obliged to withdraw northwards into the steppes of what is now central Kazakhstan.
The campaigns of the Khwarazm Shahs against the K'ipchak tribes of the Syr Darya
and the Aral Sea region weakened K'ipchak power and led to their dispersal across Central

47
Bartol'd, 1898, p.37.

81
ISBN 978-92-3-103467-1 Contents The KYpchak

Eurasia, contributing to the political fragmentation of the Dasht-i Kipchak. The Kipchak
were therefore unable to show effective resistance during the Chinggisid invasions and at
the beginning of the thirteenth century, the Dasht-i Kipchak was conquered and incorpo-
rated into the vast Mongol empire.

82
ISBN 978-92-3-103467-1 Contents THE SAMANID STATE

THE SAMANID STATE*

N. N. Negmatov

Contents

The creation of the Samanid state . 84

The system of government . . . . 86
The nature of political authority under the Samanids . 87
Agriculture 88
Mining . 89
Crafts . 90
Domestic and external trade . 90
Material culture 92
Intellectual life 93
Central Asia and the Ismacili movement 100
The ethnic composition of the Samanid state and the creation of an Eastern Persian-
Tajik ethnic identity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

By the tenth century, the majority of the various Iranian peoples of Khurasan, Transoxania
and Khwarazm - Persians, Bactrians, Sogdians, Khwarazmians and others - were using
the New Persian (Parsi-Dari) language as their spoken and written form of communication,
although such Middle Iranian languages as Khwarazmian and Sogdian were still in use in
certain regions - in the case of the former, for some four centuries subsequently. The terms
Tazfklg (Middle Persian) and Tazf (New Persian), originally coined in western Persia to
denote the conquering Arabs, now came in Khurasan and Transoxania to be applied to all
the Muslims there (at this time, the majority were Persians), in distinction to the largely
pagan Turks of the adjacent steppelands. Very soon it became used for the Persians as

* See Map 3.

83
ISBN 978-92-3-103467-1 Contents The creation of the Samanid state

against the incoming Turkish tribal or military ruling class. From the thirteenth century
onwards, the later form Tazfk developed and became standard in Central Asia.
Under the Samanids, i.e. in the later ninth and the tenth centuries, there was a period of
expansion and florescence for the cities and towns of Khurasan and Transoxania - Nisha-
pur, Merv, Balkh, Bukhara, Samarkand, Khujand, Bunjikat, Hulbuk, and so on - around
which the cultural, social and economic life of the whole region was concentrated.

The creation of the Samanid state 1

Almost contemporaneously with the Tahirids (see above, Chapter 2), the representatives
of another local dynasty, that of the Samanids, began to assert themselves in Transoxania.
An ancestor of the Samanids, Saman-Khuda, was the founder and ruler of the estate of
Saman, which is identified in a number of sources with various settlements of the same
name in the provinces of Balkh, Samarkand and Tirmidh (Termez). Saman-Khuda enjoyed
the patronage of the governor of Khurasan, Asad b. cAbd Allah al-Qasri (725-7, 735-8),
and converted to Islam, while Saman-Khuda's son Asad was later converted to Islam in
Merv by the future caliph al-Ma'mun (813-33). The sons of Asad - Nuh, Ahmad, Yahya
and Ilyas - took part in putting down the uprisings of Rafic b. al-Layth (805-10), served
the caliph al-Ma'man and, as a reward for their services to him, were appointed rulers of
a number of rich provinces: Nuh of Samarkand, Ahmad of Ferghana, Yahya of Chach and
U srushana, and Ilyas of Herat.
The Samanid brothers, while initially subject to the Tahirids, were largely autonomous
rulers in their own territories, minted bronze coins in their own names, and mustered mili-
tias and mounted campaigns against surrounding provinces. The energetic Nuh b. Asad did
much for the advancement of the Samanids and the consolidation of Transoxania, subju-
gating the province of Isfijab on the frontier with the Turkic nomads in 839-40. However,
the basis of the future unitary Samanid state was laid by Ahmad b. Asad, the most capable
of the brothers. Ahmad had seven sons: Nasr, Yacqub, Yahya, Asad, Ismacn, Is'haq and
Hamid. After the death of Niih b. Asad in 841/842, the governorship of Samarkand passed
into the hands of Ahmad's son Nasr, and after the death of Yahya b. Asad, the governance
of Chach and Usrushana was transferred to Yacqub, another of Ahmad's sons. Thus, in the
middle of the ninth century, the first partial unification of Transoxania took place, with its
main provinces under the authority of the house of Ahmad.
The process of uniting the state was completed with the entry into the political arena
of Nasr's brother, the energetic, intelligent and far-sighted Ismacn b. Ahmad, who became

1
For the chronology of the Samanids, see Bosworth, 1996, pp. 170-1.

84
ISBN 978-92-3-103467-1 Contents The creation of the Samanid state

the virtual founder of the Samanid state. The activity of Ismacil (b. 849) began in Bukhara,
where he was invited by his elder brother Nasr (865-92) to be governor after the fall of
Tahirid power in 874. The more powerful role played by Ismacil and his appanage of
Bukhara, the military clash between the brothers in 888 and the victory of Ismacil over
Nasr greatly strengthened Ismacil's de facto power in Transoxania. In the subsequent years
of his rule, Ismacil (892-907) displayed great ability in setting up an orderly system of cen-
tral and local state administration (dfwans), organizing the army and ensuring the internal
and external security of the country, and creating opportunities for economic and commer-
cial development and for the resurgence of local scientific and literary traditions. Ismacil's
military victories over the Saffarids in 898 and 900 made possible the incorporation into
the Samanid state of Khurasan, Sistan, Tukharistan and Kabulistan, while in the Syr Darya
basin he managed to create a fortified border from Isfijab to Taraz and eastern Ferghana. A
little later, at the high point of Samanid power, the southern frontiers of the state extended
as far as the Sulayman mountains, Ghazna, Kandahar and the Persian Gulf. But at the
same time, the peripheral provinces retained a substantial degree of autonomy, paid no
regular tribute and, indeed, many of the local ruling dynasties in those provinces were not
replaced. Some degree of religious freedom also continued. Not all had yet fully converted
to Islam in the upper Oxus river valley, where the populations of Badakhshan, Wakhan and
Shughnan continued to profess their ancient religions.
Ismacil was succeeded by Ahmad b. Ismacil (907-14), portrayed in the sources as a
devout Muslim. He reinstated Arabic as the language of administration in place of Persian
and favoured officials who knew Arabic, thereby displeasing the court circles and palace
guards. He was soon assassinated by plotters; the youthful Nasr II b. Ahmad (914-43)
ascended the throne, and the task of ruling the state was entrusted to the enlightened viziers
cAbd Allah Jayhani (914-18) and Abu '1-Fadl Balcami (918-38), who restored order, put
down rebellions and created the essential conditions for social and cultural development.
The reign of Nasr II, like that of Ismacli, was a period of florescence for the Samanid
state. The rule of subsequent amirs was distinguished by a gradual consolidation of cen-
tripetal forces in the state, more frequent rebellions and court disputes and an increase in
the influence of the Turkish palace guards. In the reign of Niih II (976-97), there were
increasingly frequent attempts at invasion by the Karakhanids (see below, Chapter 6), and
in the south-east the rise of the Ghaznavids began (see below, Chapter 5). In the reign of the
later Samanids, the Karakhanids invaded Transoxania and the Samanid lands were divided
up between the Karakhanids and the Ghaznavids. The energetic Ismacil b. Nuh al-Muntasir

85
ISBN 978-92-3-103467-1 Contents The system of government

made unsuccessful attempts to re-establish the Samanid state in 999-1005 before he was
killed. 2
The documentary sources are unanimous in confirming that the Samanids were able
to create a strong, partly centralized state (but see below, Chapter 6), which, for almost
a century, at the time of its florescence, shielded Transoxania and Khurasan from large-
scale external attack or incursions from the steppes and created important preconditions
for the development of agriculture, craftsmanship and trade, learning, literature and art. In
this connection it is worth noting the words of a contemporary of the later Samanids, the
geographer and traveller al-Maqdisi, who comments that 'the character of the Samanids,
their appearance and their respect for learning and for men of learning make them the
best of rulers'. Among the members of the Samanid dynasty, and also their viziers, were
some who valued learning, poetry and art, and a number of them wrote poetry and works
of geography and history. For example, the fine poetry of the Samanid Ism~fll b. Nuh al-
Muntasir is well known, as are the geographic writings of the Samanid vizier Abu cAbd
Allah Muhammad b. Ahmad b, Nasr J ayhani and the historical works of the vizier Abu
cAli Muhammad b. Muhammad Balcami (d. 974).

The system of government

In this system, the head of state was the amir, and the provinces were governed by his
appointed governors. Together with the royal court (dargah), ten centralized military and
civilian bureaux (dfwans) were created. The main dfwan (the dfwan of the vizier) con-
trolled all administrative, political and economic institutions. Financial affairs were con-
ducted by the dfwan of the mustawf f, diplomatic relations and important state papers were
the responsibility of the camfd al-mulk (or the dfwanal-rasa'il, i.e. of correspondence), the
royal guard and military affairs generally came under the dzwan of the sahib al-shurta,
and the state mail and the clandestine surveillance of local rulers and officials were dealt
with by the dfwan of the sahib al-barfd which was answerable solely to the central author-
ity. Markets, weights and measures and trade generally, and, later, public morals, were
controlled by the dzwan of the muhtasib. Control over revenue and expenditure and over
other state matters was exercised by the dfwan of the mushrif (inspector). There were also
dfwans for juridical affairs, state lands and waqf(pious endowment) lands. The local organs
of all the dzwans, apart from the postal administration, were responsible both to the central
authority and to the local provincial rulers.

2
Barthold, 1968 .

86
ISBN 978-92-3-103467-1 Contents The nature of political authority ...

The very names of the dfwans indicate the functions of a clearly structured state appa-
ratus, one of whose major functions was the collecting of taxes from the population of the
agricultural oases, artisans and merchants. Available information indicates that the over-
all budget of the Samanid state amounted to some 45 million dirhams, of which about 20
million dirhams were spent on maintaining the army and state officials.

The nature of political authority under the Samanids

The Samanid state was a provincial successor-state to the cAbbasid caliphate centred on
Iraq. As was almost universal in the Islamic world at this time, society was hierarchical,
with the caliph-imams being, in theory at least, the delegators of all authority, so that the
Samanid amirs were their lieutenants. In practice, the amirs enjoyed virtual independence,
but were careful to pay lip-service to the caliphal ideal. This ideal was one of a hierarchical,
socially static society under a ruler governing with the ultimate authority of God and the
divinely ordained sharf a and exercising power in the Sunni Islamic tradition. Within this
system, the indigenous Iranian landowning classes (dihqans) retained much of their ancient
local authority and social influence, especially in outlying provinces of the Samanid ami-
rate such as Usrushana, Ilaq and the upper Oxus principalities. Thus fixed land assets were
in the hands of the Samanid dynasty, major landowners, state officials and the Muslim reli-
gious leaders (the waqflands). Minor landowners and peasants had the use of the so-called
kharaj lands. The taxes they paid to the state treasury depended on the quality of the land
and the crops grown. The waqflands, and the lands of many landowners, were exempted
from taxes, while some properties (the cushr lands) only had to pay a tithe. The principal
form of land use was the share-cropping system, under which the land was divided into
small plots and let out to tenants in exchange for a specified proportion of the harvest.
Vast masses of peasants were almost or entirely landless, with a largely subsistence
economy, and their situation may have been a contributory cause for some of the outbreaks
of unrest in the period. In 907 in Ghur and Gharchistan, some 10,000 peasants and artisans
rose up under the leadership of one Abu Bilal. In 961 the citizens of Bukhara revolted.
Perceptible elements of some of these movements were various religious heresies; one of
these, for example, was the Carmathian or Ismacili movement, which gained strength in
the first half of the tenth century (see above, Chapter 2).
The struggle of centrifugal forces against the central government, together with discord
and endless intrigue in the amir's court and the dfwans, eventually led to the decline of the
Samanid state and enfeebled it politically and militarily. The economic weakening of the
state and its financial difficulties exacerbated this decline still further, so that at the end of

87
ISBN 978-92-3-103467-1 Contents Agriculture

the tenth century the Samanid state was unable to withstand the external invasions of the
Karakhanids.

Agriculture
The unification of Khurasan and Transoxania around a single centre, the preservation of
the country's de facto independence, the ending of incursions by nomads, and a number of
other measures taken by the Samanids, created the essential preconditions for the devel-
opment of the economic and cultural life of the country. Agriculture, mining, the craft
industries and trade developed further. 3 Under the Samanids, Transoxania and Khurasan
were shielded from external invasion by the might of the state. This circumstance, together
with the other measures taken, promoted the successful development of Transoxania. At
the period we are considering, the main provinces of Transoxania, and above all the Zaraf-
shan valley, were, in economic terms, the most advanced provinces of the east.
Transoxania and Khurasan were primarily agrarian lands. Agriculture was based in the
main on the use of water from seasonally filled wadis, springs and artificially constructed
surface and underground canals. In Transoxania there were numerous working irrigation
systems that had been created earlier, particularly in the Zarafshan and Ferghana valleys
and in a number of other regions. For example, in the Bukharan oasis, the Samjan and
Shahrud canals were operating, as was the irrigation system of Varagsar in the central
Zarafshan valley. There were also canals in the regions of the main town of Usrushana,
Bunjikat, and in other provinces. During the ninth and tenth centuries a number of new
canals and hydrotechnical installations were constructed. For example, in one of the gorges
of the Bast Tagh mountain range, in the Nuratau peak, a large stone dam was built (today
called Khan Bandi) and a water reservoir created. New canals were also dug from the Hari
Rud and Helmand rivers.
Non-irrigated land (i.e. land where dry-farming was possible) was cultivated as well as
irrigated land. In the irrigated lands of Sogdiana-in Ferghana, Chach, Usrushana, Khuttal,
Chaghaniyan, Khurasan, Gharchistan and Sistan - the peasants cultivated wheat, barley,
rice, millet, legumes, oil-seed and other crops, while cotton was an enormously important
agricultural crop in a number of regions (the Zarafshan and Ferghana valleys, the Merv
oasis, and so on). In the cultivated oases, horticulture was highly developed (apricots,
peaches, apples, pears, quinces, cherries, plums, pomegranates, figs, walnuts, almonds),
as were viticulture (with dozens of varieties of grape, the Herat grape being especially cel-
ebrated) and the cultivation of vegetables and melons. The melons of Bukhara and Merv

3
Negmatov, 1977, pp. 66-87.

88
ISBN 978-92-3-103467-1 Contents Mining

were particularly well known and were regarded as the best in the world at that time. Date
palms were grown in Sistan and sugar cane in the province of Balkh. This catalogue of
agricultural crops indicates how specific was the method of cultivation used, being primar-
ily based on artificial irrigation. Geographers of the ninth and tenth centuries give many
enthusiastic accounts of the fertility of the oases and their 'unbroken carpets of greenery'.
The most fertile lands were considered to be the province of Balkh and the region between
Herat and Merv al-Rudh. In Transoxania and Khurasan and their mountain and steppe
regions, animal husbandry was practised, particularly the rearing of sheep and horses.

Mining
In various regions of Central Asia many different kinds of minerals were mined and
processed. Evidence for this comes both from written sources and from archaeological
data. In Badakhshan, Darvaz, Rushan and Shughnan, rubies, lapis lazuli and silver were
mined; in Tukharistan, lead, sulphur and other metals and minerals; in the upper Zarafshan
valley, iron, gold, silver and vitriol; in Usrushana, large quantities of iron; and in Asbara
(Isfara), coal was reportedly to be found. Many minerals were mined in Ferghana: iron,
tin, silver, mercury, copper, lead, tar, asbestos, turquoise, sal ammoniac and, apparently,
petroleum oil. Ilaq (the Ahangaran valley) was known as a major centre for the processing
of silver and lead ore. In Ilaq, and in the Kashka Darya basin, salt was mined. Minerals
were processed in Khurasan: turquoise (in the district of Rivand, near Nishapur), marble
(in the district of Bayhaq), fine stone for craft working (in the Tus region), gold and iron (in
Gharchistan), iron (in the Nishapur district), copper (in the Merv district), vitriol, sulphur,
lead, arsenic (in the Balkh district), jet, clay for pottery, and so on. The mountains of Jurjan
produced gold, silver, iron, copper and various kinds of vitriol; silver came from Parwan
and Panjshir, and marble from Bayhaq.
Archaeological and geological research gives some idea of the development, scale and
technology of mining and of how deposits were prospected and explored. By the stan-
dards of the time, ore-workings were extremely successful. Numerous mine-workings of
the Middle Ages have been excavated and studied. They include such enormous mines as
the group of Koni Mansura, Kukhi Sim, Konjol, Kansai, Tarazkan and others in the Kara-
maz mountains, Koni Gut in Ferghana, and so on. The mining industry employed both free
men and serfs and slaves. Thus in Koni Gut, the iron fetters of slave miners have been
found. Mining occupied an important place in the economy of Transoxania and Khurasan
in the ninth and tenth centuries. It is possible, by analogy with the state of development of
the mining industry, to judge the general level of development of the economy of the age.

89
ISBN 978-92-3-103467-1 Contents Domestic and external trade

Crafts
The growth in agricultural output and mining was the basis for an advance in the craft
industries, which were economically very important for the population of Transoxania and
Khurasan. One of the main developments took place in textiles. Several centres of this
craft are known: the settlements of Zandana (near Bukhara), Wadhar (near Samarkand)
and Darzangi (in the Surkhan Darya valley). These centres produced fine cotton fabrics and
their wares were widely known far beyond the boundaries of Transoxania. Both fine and
coarse cotton fabrics were produced in large quantities in Samarkand, Dabusiyya, Bukhara,
Iskijkat, Nasaf (Karshi), Kish (Shahr-i Sabz), Benakat and other locations. Woollen cloth
and garments were manufactured in many places, including towns near the nomadic steppes:
in Dizak, Urgench, Arbinjan and Chach. These places, particularly Chach, were centres of
leather-working, manufacturing leather goods and other items for which there was a market
among the nomads.
The paper produced in Samarkand, which was famous in Transoxania and throughout
all the countries of the Middle and Near East, gradually displaced papyrus and parchment
in the central Arab lands. Glass was also produced in Samarkand and was highly prized
in China. In the towns and population centres of Ferghana and Usrushana, and also in
Khujand, armour, weapons, agricultural implements and metal dishes were produced.
The processing of agricultural products was widely developed. For example, in the Hari
Rud valley, Balkh, Sistan and other provinces there were many watermills in operation.
One river alone, the Balkh river, provided the motive power for seventy water mills. Wind
energy was effectively used: it powered windmills and also, as reported in the Tarfkh-i
Sfstan [History of Sistan], raised water from the wells for irrigation.
The production of ceramics flourished; potters made exquisite and high-quality glazed
wares decorated with a variety of ornamentation, and it became fashionable to inscribe
the rims of ceremonial ware with passages from the Qur' an, quotations from the works of
celebrated poets or simply messages of goodwill. Potters also made cheap ceramics on a
large scale intended for the poor, while metal and glassware were also produced.

Domestic and external trade

The development of agriculture, mining and the craft industries, together with urban expan-
sion, led, in tum, to an increase in the exchange of goods between town and country on the
one hand, and, on the other, between rural and agrarian regions and the nomadic steppe.
Bread, dried fruits, cloth, weapons and utensils were exported from agrarian regions and

90
ISBN 978-92-3-103467-1 Contents Domestic and external trade

towns into the surrounding nomadic steppe. In exchange, the nomads brought livestock,
wool, dried skins and slaves to the markets of Transoxania and Khurasan.
A very clear picture of the specialization of certain towns and localities in the production
of goods, for both domestic and foreign markets, is given by al-Maqdisl. Bukhara exported
fine fabrics, cloth for use as floor coverings, various fabrics, carpets, prayer mats, brass
lamps, saddle-girths, grease, sheepskins, oil for anointing the head, meat and melons. From
Karminiya came napkins; from Dabusiyya and Wadhar, Wadhari cloth; from Arbinjan, red
woollen winter cloaks, prayer mats, skins, strong hemp, sulphur and pewter vessels; from
Samarkand, brocade, which was exported to the Turkish peoples, silver cloth (sfmgfn), red
cloth (mamarajil), linen cloth (sinizi), much silk and silk cloth, tents, paper, glass, large
brass cauldrons, elegant goblets, stirrups, bits, straps, hazelnuts and walnuts; from Dizak,
high-grade wool and woollen clothing; from Ferghana, Turkish slaves (for resale), white
cloth, armaments, swords, copper, iron; from Chach, cotton, which went to the Turkish
peoples, cloaks, tents, shoulder ornaments, high saddles made of horse leather, processed
skins, fine bows, quivers, needles, scissors and grain; from Termez, soap and fragrant resin;
from Balkh, skins, soap, sesame, rice, walnuts, almonds, raisins, honey, vitriol, sulphur,
lead, women's shawls and striped garments; from Merv, brocade, embroidered fabrics,
cloaks, silk and cotton shawls, sesame oil, cheese, copper, etc.; from Nishapur, yam, cloth,
cotton and woollen garments, turbans, shawls, cloaks, needles, knives, iron, etc.; from Nasa
and Abiward, cotton and silk garments, silk, sesame oil, fox and sable furs; and from Herat,
large quantities of raisins, pistachios, honey and other delicacies.
In the ninth and tenth centuries, the caravan trade with China, India, Iran, the Cau-
casus and the countries of Western Asia and eastern Europe acquired importance. The
major caravan routes of Asia passed through Khurasan and Transoxania. Local merchants
embarked on enterprising commercial expeditions and probably reached China, India and
Inner Asia. Al-Maqdisi reports that he saw people travelling with caravans from Sogdi-
ana and Khurasan to Tibet and China. The busiest route was from the countries of the
Levant to China, the so-called 'Great Silk Route'. It passed through Baghdad, Ramadan,
Nishapur, Merv, Amul (Charju), Bukhara, Samarkand, Usrushana, Chach, Taraz (Jambul),
Balasaghun, the southern shore of Lake Issyk-ktil and thence into Mongolia and northern
China.
The route to eastern Europe, described by Ibn Fadlan, was likewise of importance. It
passed through Merv, Bukhara, Khwarazm, Zanjan, the River Emba, Yik (the Urals) and
Cheremshan to the city of Bulghar on the middle Volga. The caravans took the same route
into the Khazar kingdom as far as the city of Itil on the lower Volga. There was also trade
with the ancient Russian princedoms through the trans-shipment points of the cities of

91
ISBN 978-92-3-103467-1 Contents Material culture

Itil and Bulghar, to which rice, dried fruits, cotton, woollen and silken fabrics and silver
dirhams were exported. From Rus, Bulghar and Khazaria came furs, wax, honey, skins,
cattle and Slav and Turkish slaves. The large number of Samanid coins found in various
regions as far away as northern Russia and the Baltic is evidence of the busy trading links
with the ancient Rus and their western neighbours.
Caravanserais were set up along the roads, with storerooms and accommodation quar-
ters, wells and other facilities. Caravans were both small and large; the largest caravan of
the caliph al-Muqtadir (908-32) comprised 3,000 pack animals and 5,000 men. Caravans
had personnel to service them (drivers and armed guards). Artisans, scholars and diplo-
matic envoys often travelled with them.
In the period that we are considering, the range of goods changed. Whereas luxury items
such as jewellery made of gold, silver and precious stones, costly weapons, bronze and
silverware, glass, high-quality fabrics, and so on, had earlier been the main goods carried
in the caravan trade, now there were increasing quantities of utilitarian items, foodstuffs,
raw materials for the craft industry and craftwares.

Material culture
Under the Samanids, there was a widespread development of urbanization and of architec-
ture and the decorative arts. The standard attained in these areas by architects and master
artisan-decorators is shown above all by the Samanid mausoleum in Bukhara, which was
probably built in the early tenth century. The mausoleum is constructed of baked brick
and is in the form of a cube topped by a dome. The external comers have three-quarter
columns and in the middle of each of their four fa<;ades is an entrance bay with a pointed
arch. Inside the mausoleum, the simplicity and massiveness of the forms are combined with
the decorative treatment of brick-tiled wall surfaces. The decoration is austere and simple.
The architectural style of the mausoleum, while expressing the traditions of the past, at the
same time shows clear signs of something new that is characteristic of provincial, dynastic
architecture.
Another outstanding building of the subsequent phase of monumental architecture is
the cArab Ata mausoleum in the village of Tim near Kattakurgan. This single-chamber
structure, square in plan, has a splendid decorative portal. Similar to it is the small mau-
soleum of the last Samanid, Ismacn b. Nuh al-Muntasir, near the town of Kerki. It is square
in plan and was built c. 1005 of baked brick, topped by a dome based on a hexagon of
niches and squinches. The centralized-plan structure of the mausoleum, emphasized by
the small octagonal comer columns, relates it to the mausoleum of the Samanids, but its

92
ISBN 978-92-3-103467-1 Contents Intellectual life

pfshtaq (fore-porch) and incised bricks give it an affinity with the later monuments of the
eleventh century. The c Arab Ata and al-Muntasir mausoleums show the evolution of this
form from the centralized-plan structure to a portal structure. These Samanid mausoleums
also exemplify an important process in the history of architecture: the development of the
use of baked brick in monumental architecture.
Documentary sources refer to the construction in towns of central mosques, mausoleums,
minarets and covered bazaars. The main building material was raw brick (adobe). It is to
this period that such adobe buildings as the Kirk kiz caravanserai in Termez, two Kiz-
kala buildings in ancient Merv, and others belong. The structures in Bukhara were, for the
most part, timber-framed adobe buildings. Remarkable examples of architectural detail and
decoration have survived in the regions of the upper Zarafshan, such as the celebrated mon-
uments of carved wood, including the mihrab from the settlement of Iskodar, and columns
from the settlements of Obburdon, Kurut, Rarz, Fatmev and others, all richly ornamented
with vegetation and including stylized animal motifs. Among the monuments of applied
art of the time is the exquisite carved stucco of the palace of the princes of Khuttal in
Hulbuk (the small town of Khishtep in the settlement of Kurbansaid in the southern region
of modem Tajikistan), with painted relief ornamentation ranging from simple geometric
rosettes to highly complex designs and wickerwork patterns forming large stucco panels.
One fine monument using this decorative technique is the tenth-century alabaster panel
from Afrasiyab (Samarkand), which shows traces of a rich and ancient tradition. It is com-
posed of trefoils, sinuous leafed stems, circles filled with small squares, six-pointed stars,
and so on.
The degree of sophistication attained in the applied arts is demonstrated by the Afrasiyab
glazed ceramics of the ninth and tenth centuries which were the peak of Central Asian artis-
tic achievement in ceramic technique and decoration. A white or red engobe covering was
initially applied to this type of ware as a ground; it was then painted white, red, black or
other colours and a transparent glaze was applied on top, giving the ware a luminous sheen.
The surfaces of dishes, bowls and other vessels were ornamented with various geometric
designs and plant motifs, as well as inscriptions conveying greetings. Animals and birds
were sometimes also depicted.

Intellectual life
In the ninth and tenth centuries, intellectual life in Transoxania and Khurasan attained a
high level. It was inevitable that the local Samanid dynasty, seeking support among its lit-
erate classes, should cultivate and promote local cultural traditions, literacy and literature.

93
ISBN 978-92-3-103467-1 Contents Intellectual life

Poetry in Persian made rapid strides and is best exemplified in the work of Abu cAbd Allah
Rudaki, the father of Tajik-Persian poetry, of Abu '1-Qasim Firdawsi, the greatest poet of
the age and author of the famous Shah-nama [Book of Kings], and of a pleiad of other mag-
nificent poets. Poetry was only one manifestation of this culture; equally interesting was
the development of scholarship and its various branches (mathematics, astronomy, geogra-
phy, chemistry, medicine, history and philology), with such outstanding exponents as Ibn
Sina (Avicenna), Abu Rayhan al-Biruni and others (see below; for a detailed discussion of
these arts and sciences, see Volume IV, Part Two).
The main towns - Bukhara, Samarkand, Balkh, Merv, Nishapur, Khujand, Bunjikat,
Hulbuk, Termez and others - became the major cultural centres of the time. Scholars, poets,
artists and other men of culture from many Muslim countries gathered in the Samanid
capital of Bukhara, where a fertile soil was created for the burgeoning of creative thought,
so that it became one of the outstanding cultural centres of the East. In Bukhara, a rich
library was assembled known by the name of Siwan al-hikma (Storehouse of Wisdom),
containing books on various branches of learning, including the most esoteric; and the
rarest and best works of scholarship were to be found in the Bukhara book bazaar.
Although a generally hostile and contemptuous attitude towards the Arabs was one of
the main features of the so-called Shucubiyya literary movement in the early cAbbasid
days (eighth and ninth centuries), by the tenth century the movement no longer involved
opposition by the Iranian peoples to the Arabs and to Arab culture per se. Anti-caliphal
and even anti-Islamic motives were still discernible in the Shucubiyya of the tenth century,
but the Tajik-Persian poets and people of culture extolled the history and culture of their
own people without rejecting the cultural achievements of the Arabs.

SCHOLARSHIP

In the ninth and tenth centuries, scholarship made great strides and there was a host of
scholars in the various branches of knowledge who gradually began to write in Persian.
In the fields of history, literary studies and geography, there were Abu Bakr Narshakhi (d.
959), the author of the Tarzkh-i Bukhara [History of Bukhara]; the Samanid vizier Abu
cAll Muhammad Balcami, the author of a Persian epitomized translation, with numerous
interpolations, of the work of al-Tabar!; the great connoisseur and scholar of poetry and
history Abu '1-Mu' ayyad Balkhi, the author of books in Persian such as cAja'ib al-buldan
[Marvels of the Lands], a Shah-nama and a Garshasp-nama [Epic of Garshasp]; and the
anonymous geographer who wrote the Hudud az_cazam [The Limits of the World].
Islamic religious and legal scholarship was particularly flourishing at this time in the
north-eastern provinces of the Iranian world and enjoyed the enthusiastic patronage of the

94
ISBN 978-92-3-103467-1 Contents Intellectual life

Samanid amirs; Ahmad b. Ismacil is said to have been assassinated in 914 by his Turkish
slave soldiers because of his undue frequentation of the cuzama' (religious scholars) and
religious lawyers. It was not without significance for the strength of Sunni Muslim learning
in these provinces that five out of the six authors of the canonical collections of hadfth
(Muslim tradition), the Sunan, should be from the north-eastern Iranian world or have
connections with it.
In Transoxania and Khurasan, mathematics and astronomy of global significance devel-
oped out of a synthesis between the traditional local expertise in irrigation and other tech-
nology, and the mathematics of ancient Greece and India - this was the foundation of
all subsequent mathematical advances in Asia and Europe. Particularly prominent among
these scholars was the eminent mathematician, astronomer and geographer Muhammad
b. Musa al-Khwarazmi (780-c. 850), whose name lives to this day in modern mathe-
matical terminology in the word 'logarithm' (the medieval distortion al-garizm from 'al-
Khwarazml'). The name of the science of algebra (in Arabic, al-jabr) also comes from the
first word of the title of al-Khwarazmi's work on algebra. He discovered how to solve lin-
ear and quadratic equations. The mathematician Ahmad b. cAbd Allah al-Marwazi, known
as al-Habash al-Khasib (d. c. 870), already used tangents and cotangents and their tables
in his calculations. The astronomer Abu Macshar Balkhi (d. 886) wrote some forty works.
The astronomer and mathematician Abu cAbbas Ahmad al-Farghani (ninth century), in
his work Usal cilm al-nujam [Principles of Astronomy], expounded the knowledge of
his time and described instruments and the sundial. The astronomer and mathematician
Abu Mahmud Khujandi (tenth century) invented the sextant, which is used as an astro-
nomical instrument for accurately determining the positions of the planets and the fixed
stars appearing in the vicinity of the planets. The mathematician Abu '1-Wafa al-Buzjanu
(940-98) solved a number of geodesic and geometric problems, gave a systematic account
of trigonometry and, together with al-Battani, was the founder of trigonometry. In the
tenth and eleventh centuries, the mathematician Abu 'I-Hasan al-Nasawi and many oth-
ers worked in the region.
Medicine also developed. A number of scholars were active in that field in the ninth and
tenth centuries, the most outstanding of whom was Zakariyya Razi. Another able physician
was Akhawayni Bukhari. He was engaged in medical practice, particularly in the field of
mental disorders, and was one of the first to use Persian to write his medical treatises
in which he expounded on human anatomy and physiology, described illnesses and their
cures, and gave the dosage of medicines and as many as ten original prescriptions. Another
learned physician was Hakim Maysari, the author of many books including a medical work

95
ISBN 978-92-3-103467-1 Contents Intellectual life

in Persian verse, the Danish-nama [Book of Knowledge]. The well-known pharmacopeia

of Muwaffaq is also written in Persian.

ABO CALI IBN SINA AND ABO RAYHAN AL-BIRONI

In this period, the two greatest medieval polymaths of the East, Abu c Ali: Ibn Sina and
Abu Rayhan al-Biruni from Khwarazm, began their work which has left an indelible mark
on world scholarship. Abu c Ali: Ibn Sina (known to the West as Avicenna) was born c.
980 in the village of Afshana near Bukhara. His childhood and early student years were
spent in his native village, after which the family moved to Bukhara, where he studied all
the sciences of that time and, while still a young man, began his scholarly and medical
activity. After the fall of the Samanid dynasty, he was sought in vain by the Ghaznavid
Sultan Mahmud. Ibn Sina lived in Urgench and Isfahan before becoming a court physician
and vizier to the Kakuyid ruler of Ramadan. He died and was buried in Ramadan on 18
June 1037.
Ibn Sina was the greatest scholar of the Islamic Middle Ages - a philosopher, nat-
ural scientist, physician, mathematician, poet and original thinker. He gave voice to the
developing tendencies of his time and, in an age when religious orthodoxy was dominant,
endeavoured to revive interest in the study of nature, to give an impetus to creative and
analytical thought and to review critically what had been achieved by scholarship in ear-
lier days, systematizing and developing it further. In his research, Ibn Sina placed great
emphasis on experimentation, practical experience and the objective observation of facts.
He wrote many works, the most important of which were al-Qanun fi 'l-tibb [The Medical
Canon], a medical encyclopedia in five parts, and Kitab al-Shifa' [The Book of Healing],
a philosophical work in eighteen parts, which were written in Arabic. He became widely
celebrated in the East and in Europe, and many of his works were translated into sev-
eral European languages. His al-Qanun fi 'l-tibb was used for many centuries for teaching
and treatment in both East and West. It was translated into Latin (as early as the twelfth
century), classical Hebrew, Persian, Urdu and other languages; the Latin translation went
through thirty editions. In the al-Qanun fi 'l-tibb, Ibn Sina expounds a general theory of
medicine, anatomy, physiology, surgery, diagnosis, therapeutics, drugs and prophylactic
medicines, and discourses on acute and chronic illness. In Persian, he wrote a short philo-
sophical encyclopedia entitled the Danish-nama [The Book of Knowledge] in which he
touches on questions of philosophy, logic, mathematics and astronomy. In the Kitab al-
Shifa' he gives an exposition of questions of logic, natural science, metaphysics, mathe-
matics and other sciences of his time. These books by Ibn Sina may be regarded as among
the best works of enlightened medieval thought of the tenth and eleventh centuries. He was

96
ISBN 978-92-3-103467-1 Contents Intellectual life

also a poet, writing in Persian and mainly using the rubitf (quatrain). In these verses, he
openly expresses something like free-thinking.
Another outstanding encyclopedist, Abu Rayhan al-Biriini (973-1048), was the author
of many works of major importance in which he examines problems of mathematics,
astronomy, physics, botany, geography, geodesy, general geology, mineralogy, history,
chronology and ethnography. One particularly well-known work of his is al-Athar al-
baqiya [Memorials of Past Generations], which among many other topics describes in
detail the calendar systems of the peoples of the Middle and Near East, including the peo-
ples of Sogdiana and Khwarazm. His great work on India is an important description and
criticism of the higher Indian learning and sciences of the early Middle Ages. Part of his
treatise Kitab al-Tajhfm il-awa'il sinac at al-tanjfm [The Key to Astronomy], with its popu-
lar exposition of the fundamentals of mathematics, astronomy and astrology, has survived.
He also wrote al-Qaniin al-Mascadr [The Canon of (Sultan) Mascud], a lengthy treatise
on mathematical and descriptive geography; and Kitab al-Jamahir [Book of Data for the
Recognition of Gems] (on mineralogy), in which he gives detailed information on over
fifty minerals, ores, metals and alloys, and on glass, enamel, porcelain and so on, and in
which he also catalogues mineral deposits and gives other data. Mention must also be made
of his book Istikhraj al-awtar [Chords] on the topography of Central Asia; and his Rules
for Determining the Specijic Gravities of Minerals, among others.

LITERATURE
In the ninth and tenth centuries, there was an enormous growth in literature, principally
poetry. It was during the Samanid period that Persian literature began to appear in Transox-
ania and was officially recognized. The development of an Islamic New Persian literature
thus began in Transoxania and Khurasan rather than further west in Fars. The work of a
large number of poets dates from this period, of whom the best known are Rudaki (d. 941 ),
Daqiqi (d. 977) and Firdawsi (d. c. 1020).
Together with the nascent literature in Persian, literature in Arabic also continued to
enjoy high prestige and royal favour under the Samanids. The most active part in the cre-
ation of this literature, both original and in translation, was taken by the Central Asians
and Khurasanians. For example, Thacalibi wrote his anthology, the Yatfmat al-dahr [The
Unique Pearl], in Arabic; in its fourth section, it gives a detailed account of the poets of the
Samanid period living in Bukhara and Khurasan, and also those of Khwarazm, who wrote
in Arabic.
The recognized founder of Persian-Tajik classical poetry, and a man of great culture,
was Rudaki, who was born in the village of Panjrudak in what is today the Panjikent region

97
ISBN 978-92-3-103467-1 Contents Intellectual life

of modern Tajikistan. From his early years, Rudaki's poetic gift, his fine voice and his
skilled playing on the chang (a harp-like musical instrument) made him popular. He was
invited to the Samanid court, where he spent almost the whole of the rest of his life. Some-
what fewer than 2,000 lines of his output have survived, but they demonstrate his mastery
in all the poetic genres of the age. He perfected the basic verse forms of medieval Persian
poetry: mathnawf, qasfda, ghazal and rubacf. Another noted poet was Shahid Balkhi, born
near Balkh in the village of Jakhudanak. Little is known about his life, but at one time he
was one of the best court poets of the Samanid Nasr II, and one of the leading scholars of
the age. A close student of Rudaki, he died in 936, before Rudaki, and the latter wrote a
touching elegy on the death of his favourite student.
Daqiqi began his literary activity at the court of the ruler of the principality of
Chaghaniyan and was later invited to Bukhara. Under the Samanids, particular attention
was paid in literary circles to making compilations of ancient legends embodying the heroic
traditions of their past. It was thus that the prose Shah-nama of Abu Mansur was created.
Nuh II entrusted Daqiqi with the task of putting the Shah-nama into verse, but the poet
was killed in 997 before completing his work. The part of Daqiqi' s poem describing the
struggle between Gushtasp and Arjasp was included by Firdawsi in his own Shah-nama. In
addition to a fragment of his Shah-nama, the surviving work of Daqiqi includes qasfdas,
ghazals, qif as and bayts.
The greatest poet of that age, however, was Firdawsi. He was born in Khurasan, in a
suburb of Tus, between 934 and 941 into a middle-ranking aristocratic family. His youth
coincided with the period of growth of the Samanid state. The idea of stimulating a resur-
gence of the Iranian national and heroic spirit in culture induced him to write his epic Shah-
nama, which he completed in 994, by which time the Samanid state was in full decline. The
second version of the poem (completed in 1010) was presented to the Ghaz-navid Sultan
Mahmud; but the poet's masterpiece was not appreciated in his lifetime.
In his poem, Firdawsi revealed to the people their heroic past and apparently sought
to arouse their sense of patriotism. He made the basis of his work the epic of the Iranian
peoples (the ancient Saka tribes, Sogdians, Khwarazmians and others), including the tales
about such epic heroes as Rustam, Siyawush and Isfandiyar, and about national uprisings
led by the blacksmith Kawa and Mazdak. Firdawsi drew much inspiration from the monu-
ments of material culture in Khurasan and Transoxania. In particular, a whole epic cycle of
exploits connected with the name of Rustam - the struggle of the heroes against monsters
- is depicted in the murals of ancient Panjikent. 4 Another scene in temple painting, also

4
Belenitskiy, 1960.

98
ISBN 978-92-3-103467-1 Contents Intellectual life

in Panjikent, is devoted to mourning for the mythological hero Siyawush. 5 In the ruined
palace of the Afshins of Usrushana (near Shahristan), a large carved wooden panel has
been discovered depicting scenes of the struggle of the forces of good (the prince and the
blacksmith Kawa) with the wicked and murderous King Zahak. The Rustam epic of the
Sogdian texts, the Sogdian tales of Siyawush, Faridiin and the blacksmith Kawa, and these
paintings at Panjikent and Bunjikat, undeniably indicate the Central Asian provenance of
many of the themes of the Shah-nama.
Firdawsi also masterfully reworked the written sources (the Sasanian historical chroni-
cle, the Khwaday-namak [Book of Kings], Mansur Balkhi's Shah-nama, and others), draw-
ing on the treasures of material culture and on popular oral legends. He recounts events in
his own way and celebrates the ideas of Iranian national feeling and love of his homeland.
From beginning to end, the poem is written in a heroic metre (mutaqarib), in an austere and
monumental poetic style, and is distinguished by its epic grandeur and unique emotional
tone. It contains between 50,000 and 60,000 couplets (bayts). Firdawsi's Shah-nama has
immortalized the author's name and is a major contribution to the treasure-house of world
culture; it had an enormous influence on the subsequent development of Persian literature
and there have been frequent translations of his poetry into many languages of both East
and West.

MUSIC

In the ninth and tenth centuries, the musical culture of the eastern Iranian peoples, the ori-
gins of which went back to antiquity, continued to develop. They had a long tradition of
virtuoso musicianship and a developed musical theory. In the period that we are consider-
ing, a series of patterns of musical art in various genres was created. Classical professional
music in the oral tradition and settings of poems to music continued to be produced and
these were subsequently combined in the duvazda maqam (twelve maqams, or modes),
and then reworked in the shash maqam (six maqams). Poetry and song were not only inti-
mately interwoven but were regarded as branches of a single art. Many poets were at the
same time well-known musicians, singers and musical theorists. Popular creative art had a
marked influence on the development of professional poetry and music.
As noted above, the poet Riidaki was also a gifted musician and a well-known singer.
He had a fine voice and was a performer on several instruments (cad, chang, barbat and
rad). It was he who was responsible for reworking two classical modes and melodies, the

5
D'yakonov, 1951.

99
ISBN 978-92-3-103467-1 Contents Central Asia and the lsmacili movement

tarona and the ufar, and tradition has it that he himself perfected the design and nature of
the rad.
Among the musicians of the age were the poet and musicologist Abu Hafs Sughdi, who
invented the musical instruments called shah-rad and maszkar, the poet and composer Abu
Salih (ninth century); the poet Abu Tayyib Tahir al-Khurasani, an outstanding performer of
one of the maqams called khusrawan; the well-known chang-player Lukari Changzan; the
musicologist Abu 'I-Abbas Bakhtiyar; clsa Barbati; one of the best-known female singers,
Sitt Zarin; the flautist Zilzil Razi; and the flautist and tunbar-player cAllbegi. The founda-
tions of musical theory were also laid, and Ibn Sina is even credited with having invented
the instrument known as the shahnay (surnay).
This eastern Iranian music became a firmly established part of all aspects of social life
and was an inseparable element in weddings and feasts, festivals and battles, anniversaries
and formal ceremonies, religious and funerary rites. In particular, the historian Narshakhi
refers to songs of mourning, and we find information on celebratory, military and funer-
ary music in Firdawsl's Shah-nama and in the poetry of Rudaki and Daqiql. There were
a number of musical modes: the dastan-i khusrawan, surad (rah, khafif), tarona and oth-
ers, and a large number of melodies: khusrawanf, cushshaq, rast, bada, irak, zarafkanda,
busalik, sipaha'f, nawa, basta, chawasht, etc. There were also a large number of musical
instruments: the stringed ones mentioned above, and others; wind instruments like the nay,
shahnay (surnay), shahrad, shaypar, karnay (baq), etc.; and percussion instruments like
the daff (da'ira), tanbarak and tabl (shandaf). In addition to those that have been enu-
merated, one must also mention the duruya, zfr, chagana, santar, kapun, urghunan, anka,
etc.

Central Asia and the lsmacili movement

In the ninth and tenth centuries, the Sunni form of orthodox Islam became firmly estab-
lished in the eastern lands. Yet even though Islam was already generally accepted as the
state religion and had spread to almost all parts of Central Asia, there continued to exist, at
the same time, vestigial communities of Zoroastrians, Manichaeans and Christians.
In the first half of the tenth century, the propaganda of Is macilism achieved a foothold in
Central Asia, as has been described in Chapter 2 above. The movement was initially led by
the Samanid commander Husayn b. cAll Marwazl. In an attempt to turn it to his advantage,
he quickly provoked an uprising but was defeated and captured. Leadership next passed to
the Ismacili propagandist Muhammad b. Ahmad Nakhshabi, whose activities were at first
highly successful. Many eminent dignitaries converted to Ismacilism and the movement

100
ISBN 978-92-3-103467-1 Contents The ethnic composition ...

gathered strength, particularly during the reign of Nasr II, who may also have accepted
its teachings. This incensed the orthodox Sunni Muslim religious classes, who conspired
against Nasr II with representatives of the Turkish palace guard. Even though the plot was
discovered, Nasr II was obliged to renounce the throne in favour of his son Nuh I (943-54).
Nuh organized the extermination of Ismacili sympathizers and the execution of the leaders
of the movement of Muhammad Nakhshabi. The doctrine survived in Transoxania only
as a clandestine tendency, although in Bukhara a secret organization of supporters of the
heresy may have been active.

The ethnic composition of the Samanid state and the

creation of an Eastern Persian-Tajik ethnic identity
In early Islamic times, there were a number of Iranian peoples in Khurasan and Tran-
soxania. The Zarafshan valley, the Kashka Darya oasis and U srushana were occupied by
Sogdians; the upper Oxus basin and its tributaries by the Bactrian-Tukharian population;
the basin of the lower Oxus by the Khwarazmians; the Ferghana valley by the Iranian
Ferghanans; the southwestern oases of Central Asia by the Iranian Khurasanians; and the
Pamir mountains and their foothills, and the mountains surrounding the Ferghana valley,
by the remnants of Saka and other early Iranian peoples. All these peoples were ethnically
related and spoke languages and dialects of the Middle Iranian and New Persian language
groups; they were the basis for the emergence and gradual consolidation of what became
an Eastern Persian-Tajik ethnic identity.

101
ISBN 978-92-3-103467-1 Contents THE GHAZNAVIDS

THE GHAZNAVIDS*
C. E. Bosworth

Contents

The prehistory of the Ghaznavids . 103

The Establishment of Sebiiktegin in Ghazna 104
The succession of Mahmud . . . . . . . 106
The zenith of the empire under Mahmud 107
Mascud and the Seljuqs . . . . . . . . . 113
The nature and structure of the Ghaznavid state 117
Cultural and intellectual life . 121
Agriculture and trade . . . . 122

The establishment of the Ghaznavid amirate in what is now Afghanistan in the last quar-
ter of the tenth century A.D. represents the culmination of a process which had begun in
the Samanid amirate whereby the military bases of the state had been transformed from
a reliance on the indigenous, Iranian landed classes, the dihqans, to a substantial depen-
dence on Turkish slave troops. Until the decay of their power in the second half of the tenth
century, the Samanid amirs had kept a firm hold on the direction of affairs, and the flour-
ishing state of their lands in Transoxania and Khurasan had won them admiring comments
from such Arab geographers who had travelled through their territories as lbn Hawqal and
al-Maqdisi (see above, Chapter 4). However, the decline of the amirs' personal authority
and the growth of centrifugal forces in the state, aggravated by the personal ambitions of
the great military commanders, had plunged the Samanid amirate into increasing crisis and
chaos; it was these difficulties, and the mutual rivalries of leading figures, which allowed
the formation of the Ghaznavid amirate. For roughly half a century, it was to be the most

* See Map 4.

102
ISBN 978-92-3-103467-1 Contents The prehistory of the Ghaznavids

powerful state known in the eastern Islamic lands since the weakening of the cAbbasid
caliphate.

The prehistory of the Ghaznavids

Prominent among the disputing Turkish generals at the Samanid court in the middle years
of the tenth century was the commander-in-chief (Persian, ispahsalar; Arabic, hajib al-
hujjab) of the Samanid army in Khurasan, Alptegin (appointed to this office by Amir cAbd
al-Malik in 961), who worked with the Persian vizier Abu cAli Muhammad Balcami to
secure an ascendancy in the state for their own personal interests. But their attempt, on
cAbd al-Malik's death in 961, to impose their own candidate on the throne in Bukhara, the
dead man's young and pliant son, failed, and under the new amir, cAbd al-Malik's brother
Mansur I, a rival group of Turkish generals headed by Fa'iq Khassa was now supreme at
court. Threatened by the new regime and squeezed out of power, Alptegin prudently with-
drew to the far eastern fringes of the Samanid empire with his personal force of Turkish
professional, military ghulams (slave soldiers) and a group of Iranian ghazfs (fighters for
the faith). The sources state that his aim was to carry on holy war against the infidels there
and thus earn divine merit. In reality, Alptegin was seeking safety for his own person, and
was very probably influenced by the example of a predecessor of his, another Turkish gen-
eral of the Samanids, Karategin Isfijabi, who before his death in 929 had built up round
himself a petty principality in what is now south-eastern Afghanistan centred on Bust and
the region of al-Rukhkhaj or Arachosia, nominally still subject to the Samanids but in prac-
tice autonomous. This principality had continued there under succeeding Turkish ghulam
leaders. 1
Alptegin's destination was the region in eastern Afghanistan of Zabulistan, centred
on the small and obscure town of Ghazna or Ghazni, where he now found himself; the
Samanid authorities in Bukhara had to make the best of the situation and to send Alptegin
an investiture patent as local governor in a region where their control had in any case been
very shadowy or even nonexistent. In fact, Ghazna seems to have been held by a local fam-
ily, whose generic name may have been that of Lawik (although the reading as a personal
name, Anuk, has been less plausibly suggested), doubtfully Muslim and closely linked
with the indigenous rulers in Kabul of the Hindushahi family. Alptegin's son Abu Is'haq
Ibrahim, who succeeded his father briefly in 963, had in 964-5 to flee to Bukhara when
the Lawik dispossessed by Alptegin returned. A decade or so later, the people of Ghazna,
chafing under the tyranny of one of the Turkish commanders who had by then come to

1
On the early Ghaznavids, see Barthold, 1968, pp. 249-51; Nazim, 193 1, pp. 24-5; Boswarth, 1963, p.37.

103
ISBN 978-92-3-103467-1 Contents The Establishment of Sebuktegin in Ghazna

power there, Bori or Bori Tegin, again invited Lawik back. After this temporary hiatus,
however, the town remained firmly in Turkish hands, although the neighbouring town of
Gardiz - which seems to have had its own local ruling family, that of the Abu Mansur Afiah
mentioned at the time of the first Saffarid Yacqub b. Layth a century before - did not fall to
the Turks until c. 974-5, since Bilgetegin, Abu Is'haq lbrahim's successor in Ghazna, was
killed at its siege. 2

The establishment of Sebuktegin in Ghazna

Thus the rule of the line of Turkish generals, all originally in the Samanid service, gradu-
ally became firmly established in the eastern part of modem Afghanistan. On the evidence
of the few surviving coins of the period, however, they still recognized the Samanid amirs
in Bukhara as their suzerains. This was the position when Sebtiktegin (this seems to be
the most probable form for this name; Turkish sevuklsebuk tegin or 'beloved prince') in
977 took over from the deposed Bori, beginning a twenty-year rule in Ghazna. Sebtiktegin
had been one of the most trusted personal slaves of Alptegin, accompanying him on his
withdrawal to Ghazna in 962. All that we know of his antecedents stems from a collec-
tion of aphorisms on statecraft and kingly power allegedly left by him to his son Mahmud,
the Pand-nama [Epistle Containing Pieces of Advice], in which it is stated that he came
from the Turks of Barskhan, on the shores of the Issyk-ktil in the region later known as
Semirechye (now in Kyrgyzstan). Obsequious genealogists later fabricated for Sebtiktegin
a genealogy stretching back to the Sasanian emperors of Persia, but in fact he probably
came from one of the component tribes of the Karluk Turkic group. Regarding his sub-
sequent career, the elaborate account in the Seljuq vizier Nizam al-Mulk's Siyasat-nama
[Book of Statecraft] of Sebtiktegin's rise to fame under Alptegin's patronage because of his
outstanding personal qualities can hardly be taken at face value. 3 Be this as it may, Sebtik-
tegin now began an uninterrupted period of power in Ghazna (977-97), still acknowledging
the Samanids as his nominal overlords: placing their names before his own on the coins
which he minted and being content, it would appear from the inscription on his extant tomb
in Ghazna, with the title expressing his subordinate status, al-hajib al-ajall (Most Noble
Commander). Yet in practice, he was securely laying the foundation of an independent
Ghaznavid state which his son Mahmud was to erect into a mighty, supranational empire.
On arriving in Ghazna, Alptegin's ghulams had established on the surrounding agricultural
lands a series of territorial revenue assignments (iqtac s) for their support. Sebtiktegin now

2
Nazim, 1931 , pp. 25-7, 175-6; Bosworth, 1963, pp. 37-9; 1965, pp. 16-22; 1975, pp. 164-5.
3
Nazim, 193 1, pp. 28-9; 1933, pp. 605-28; Bosworth, 1963, pp. 39-41; 1975, pp. 164-5.

104
ISBN 978-92-3-103467-1 Contents The Establishment of Sebuktegin in Ghazna

reformed this system, insisting on control from the central dfwan in Ghazna and ensuring
that all soldiers had adequate stipends. 4 The Turks there were nevertheless still an iso-
lated pocket in a hostile environment, with powers to their east like the Hindushahis of
Wayhind and other north Indian rulers, whose attitude was bound to be hostile, for east-
ern Afghanistan had always tended both historically and culturally to be part of the Indian
world. 5
Accordingly, Sebtiktegin may have made a conscious decision that a policy of expan-
sion would preserve the dynamic of his Turkish followers and ensure the survival and
future florescence of his petty lordship. He early moved against the existing line of Turkish
rulers in Bust, overthrowing their leader Baytuz, and also adding Qusdar (in north-eastern
Baluchistan) to his dominions (c. 977-8). More significant for the future direction of Ghaz-
navid expansion were clashes with the Hindushahis, who held the Kabul river basin and
the Panjab plains. Retaliatory attacks on Ghazna by the Rajah Jaypal (c. 986-7) led to
Sebtiktegin's victory over his forces and the extension of Ghaznavid power into the region
of Lamghan and the Kabul river valley as far as Peshawar. There is nothing to show that
Islamic religious motives were uppermost here, but Islam must have been implanted in
these regions, and the tradition of winter plunder raids from the mountain rim of eastern
Afghanistan down to the Indian plains now took shape. 6
The firmness of Sebtiktegin's power in Ghazna and his expansionist policies enabled
him in the later years of his reign to intervene in the politics of the Samanid state which had
originally nurtured him. The power of the amirs was now in steep and irreversible decline,
so that in 992 the capital Bukhara had been temporarily occupied by the invading Turkish
Karakhanids, and the perpetuation of the amir's authority was threatened by the alliance of
two of the most powerful Turkish generals in the state, Abu c Ali Simjuri and Fa' iq Khassa.
Against this last threat, Amir Nuh II b. Mansur summoned to his aid Sebtiktegin and his son
Mahmud. The two swore allegiance to Nuh at Kish and appeared in Khurasan with their
army. In a battle near Herat the royal forces secured a complete victory over the two rebel
generals, who fled westwards to the northern Persian territories of the Samanids' Buyid
rivals in 994. For these services, Sebtiktegin received the honorific title (laqab) of nasir
al-dfn wa 'l-dawla (Helper of Religion and the State) and Mahmud that of sayf al-dawla
(Sword of the State) plus command of the Samanid army in Khurasan; and an attempted
revanche by Abu c Ali and Fa' iq was defeated in the following year. Fresh incursions into
Transoxania by the Karakhanids in 996 rendered the Samanid amir even more dependent
4
Bosworth, 1963, pp. 41-2, 124-5.
5 Ibid., pp. 41-3.
6
Nazim, 1927, pp. 485-95; 1931 , pp. 29-30; Habib, 195 1, pp. 14-16; Ray, 1931 - 6, pp. 55-83; Bosworth,
1975, pp. 165-8.

105
ISBN 978-92-3-103467-1 Contents The succession of Mahmud

on Sebtiktegin and Mahmud. They were able to compel Nuh to nominate a vizier favourable
to their interests, and then to negotiate a peace treaty with the Karakhanids which left the
latter in control of the Syr Darya valley and the Ghaznavids with the whole of Khurasan,
which was never again to be controlled by the Samanids. 7

The succession of Mahmud

Sebtiktegin died in 997 and Ghaznavid control of Khurasan was thrown in jeopardy.Mahmud
had to hurry eastwards to Ghazna in order to wrest power from his younger half-brother
Ismacil (997-8) whom Sebtiktegin - perhaps influenced by the fact that Ismacil's mother
was a daughter of Alptegin - had appointed his successor there. Mahmud's superior mil-
itary skill soon made him master in Ghazna (998); the new Samanid amir, Mansur II b.
Nuh, invested him with what is now Afghanistan and eastern Khurasan and he was able
to recover western Khurasan from the Turkish general Begtuzun. Thus by 999 Mahmud
was in complete control of the whole of Afghanistan and the former Samanid territories
south of the Oxus (Amu Darya) and now regarded himself as an independent sovereign.
He secured legitimation of his new power from the cAbbasid caliph in Baghdad, al-Qadir
(991-1031), who bestowed on him the titles of walfamfral-mu'minfn (Friend of the Com-
mander of the Faithful) and yamfn al-dawla (Right Hand of the State) the latter being the
one by which Mahmud became best known and which was at times applied to the Ghaz-
navid dynasty as a whole (thus the historian Jiizjani refers to them in his Tabaqat-i Nasirf
as al-Sebuktigfniyya al-Yamfniyya al-Mahmudiyya ).Thus was inaugurated, also, the tra-
dition whereby the Ghaznavid sultans always buttressed their power by caliphal approval,
were assiduous in sending gifts from Indian plunder to Baghdad, and carefully cultivated
an image of defenders of Sunni orthodoxy against the caliph's opponents and rivals such as
the Shfite Buyids oflraq and western Iran and the Ismacili Fatimids of Egypt and Syria. 8
An agreement was reached at this point with the Karakhanid Ilig Nasr b. cAll mak-
ing the Oxus the boundary between the two empires, for the shrunken Samanid amirate
came to an inglorious end when the Ilig occupied Bukhara definitively in 999. This was a
historical event whose significance cannot have been apparent at the time. After 1017 the
north-eastern lands of Islam were wholly in the hands of two Turkish sovereign powers,
ending the rule there of indigenous Iranian dynasties. The region became open to a steady
flow of Turkish immigration from the Inner Asian steppes. Thus the process began of con-
verting what had been in pre-Islamic times 'l'Iran exterieur' into a majority Turkish ethnic
7
Barthold, 1968, pp 261-8; Nazim, 1931 , pp. 30-2; Bosworth, 1963, p. 44; l962b, pp. 217-18; 1975,pp.
168-9.
8
Nazim, 193 1, pp. 36-45; Bosworth, 1963, pp. 45-6; l962a , pp. 62-3; 1975, pp. 168-9.

106
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

and linguistic region by the sixteenth century; while the gradual influx over subsequent
centuries of Turkish pastoral nomads and their herds from beyond the Jaxartes (Syr Darya)
was to have a decisive effect on the pattern of land utilization and agricultural economy in
Transoxania and the northern tier of the Middle Eastern lands.

The zenith of the empire under Mahmud

Mahmud's thirty-two year reign (998-1030)-lengthy by contemporary standards -enabled
him to build up, by ceaseless campaigning, a vast military empire. From a nucleus in
Afghanistan and Khurasan, this empire stretched by his death from the fringes of Azer-
baijan and Kurdistan in the west to the Ganges valley of northern India in the east, and
from Khwarazm and the upper Oxus principalities in the north to the Indian Ocean shores
of Sind and Makran in the south. Not since the heyday of the c Abbasid caliphs had one
man ruled so much territory, and that from the insignificant town of Ghazna. For some two
centuries, until the decline of Ghurid power in India at the beginning of the thirteenth cen-
tury, Ghazna was to be a place of international significance. Only then did it relapse into
its former obscurity, so that, visiting it three centuries later, Babur, the founder of Mughal
rule in India, was to muse, 'Ghazna is a very humble place; strange indeed it is that rulers
in whose hands were Hindustan and Khurasanat should have chosen it for their capital.' 9
As noted above, the pattern of expansionism had been set by Sebtiktegin, but his son
Mahmud was now far better endowed with both military and financial resources than his
father had been. Mahmud took over the Samanid forces in Khurasan and his successes
as a war leader ensured a steady stream of free soldiers and volunteers to supplement
his core of Turkish ghulams. Above all, the fiscal resources of the province of Khurasan,
with its rich agricultural oases and its urban centres for commerce and industry, provided
a steady income from taxation for the maintenance of the highly expensive Ghaznavid
standing army, a financial injection which the much sparser economies of the plateaux and
mountains of the Hindu Kush-Pamirs region could never have supplied.
The protection of Khurasan and the Oxus frontier was thus a prime concern ofMahmud' s,
for, despite the agreement with the Karakhanids and the sultan's marriage in 1000 to
a daughter of the Ilig (possibly the Mahd-i Chigil of certain sources), the Khans cov-
eted Khurasan for themselves. While Mahmud was absent at Multan in India in 1006,
Karakhanid armies swept down on Balkh and on Nishapur, where the local inhabitants,
exasperated at the rapacity of Ghaznavid tax-collectors, actually welcomed the invaders.
Returning swiftly, Mahmud restored the situation, but a second invasion under the Ilig and

9
Babur-nama, cited in Bosworth, 1963, p. 36.

107
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

his kinsman Kad'ir Khan Yusuf of Khotan came in the following year. This was stemmed
by a brilliant victory of Mahmud's near Balkh in 1008, in which the Ghaznavids' war ele-
phants struck terror into the Karakhanid ranks, unfamiliar with these awe-inspiring beasts.
This defeat quelled Karakhanid designs on Khurasan; being a tribal confederation rather
than a unitary state as was the Ghaznavid empire, the Khans were never again able to
present a united front against Mahmiid. 10 The sultan, for his part, negotiated marriage
links (thus his son Mascud, the future sultan, married a daughter of the Great Khan Arslan
Khan Mansur, brother of the Ilig Nasr) and skilfully exploited those dissensions within the
Karakhanid family, which were later to lead to a division of territories into a Western and
an Eastern Khanate.
The main Karakhanid threat to Mahmud's position was now to come from his imme-
diate neighbour to the north, cAli b. Bughra Khan Hasan or Harlin, called cAli Te gin, of
Bukhara and Samarkand - until his death in 1034, he was the most strenuous opponent of
Ghaznavid ambitions in Central Asia. In 1025 Mahmud invaded Transoxania with the aim
of destroying cAli Tegin and he made an alliance with the latter's rival, Kad'ir Khan Yusuf
(now ruling in Khotan and Kashghar), sealed, as usual, by marriage links. Although cAli
Tegin was temporarily driven out of Samarkand, he returned in 1026 when Mahmud left
Transoxania in order to prepare for the Somnath expeditions (see below), and the sultan
made no further efforts in this quarter. Barthold is probably correct in stating that Mahmud
preferred to leave cAli Tegin in Transoxania as a counterbalance to Kad'ir Khan Yiisuf. 11
Significant, however, of the access of prestige which Mahmud's campaigns in Transoxania
and Khwarazm brought him within Inner Asia is the historian Gardizi's mention under the
year 1026 of embassies from the distant Kita (sc. the Kitan or Liao of northern China) and
the Uighurs of Kocho in East Turkistan (what is now Xinjiang), coming to the sultan to
seek marriage alliances for their rulers; such ties with pagans were courteously but firmly
rejected by Mahmud. 12
Mahmud's activist policy in Transoxania during these years had been facilitated by his
acquisition of an important bridgehead across the Oxus, possession of which enabled him
to turn the flank of the Karakhanids and exert pressure on cAli Te gin. The ancient Iranian
kingdom of Khwarazm had been ruled until 995 by the old-established line of Afrighids of
Kath, but control subsequently passed to a new line of Khwarazm Shahs, the Ma'munids
of Gurganj. Although the shahs had been nominally subject to the Samanids, in practice
they had been independent, especially as Khwarazm formed an isolated salient of settled,
10
Barthold, 1968, pp. 272-4; Nazim, 1931 , pp. 47-52; Pritsak, 1953-4, pp. 28-9; Bosworth, 1963, p. 116;
1975, pp. 170-1.
11
Barthold, 1968, pp. 274, 279-86; Nazim, 1931 , pp. 52-3; Pritsak, 1953-4, pp. 29-33; 1950, pp. 216-24.
12
Barthold, 1968, p. 286; Nazim, 1931 , p. 56; Minorsky, 1942, pp. 19-20, 76-80.

108
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

irrigated land within the surrounding deserts and steppes. The province's agricultural rich-
ness and its historic commerce with Inner Asia and southern Russia attracted Mahmiid's
greed. He married his sister (known as Hurra-yi Khuttalf or Kalj() to Ma'miin b. Ma'miin
Khwarazm Shah and, by deliberately provoking local Khwarazmian feeling and pursuing
an unscrupulous diplomacy which led to his brother-in-law's murder, secured in 1017 a
pretext for intervention in Khwarazm. A Ghaznavid invasion took place; the Ma'munids
were overthrown, thereby putting an end to the last independent Iranian line in Central
Asia; and Khwarazm was incorporated into the Ghaznavid empire under the governorship
of Altuntash, a former ghulam of Sebtiktegin, who now ruled there with the traditional title
of Khwarazm Shah, the first of a series of Turks to bear that designation. 13
Within what is now Afghanistan and Baluchistan, the heartlands of the Ghaznavid
empire, various local rulers - some of whom had been loosely dependent on the Samanids
- were brought into the Ghaznavid orbit. A local prince was allowed to remain in Qusdar
as a vassal after a Ghaznavid show of strength there in 1010-11, as also in Makran, the
coastal strip of Baluchistan, where a Ghaznavid force intervened in 1026 in a succession
dispute. 14 Existing local lines seem also to have been left in the trans-Oxus principali-
ties of Chaghaniyan and Khuttal. These had strategic value as bridgeheads for Ghaznavid
campaigns against the Karakhanids and were also the first line of defence for northern
Afghanistan against predatory peoples like the Kumijis of the Buttaman mountains (in
what is now Tajikistan) (perhaps remnants of an element of the Hephthalite confederation),
whose raids were to be encouraged in the 1030s by the Karakhanid prince Bori Tegin (the
later Tamghach Khan Ibrahim). 15 We know that in Mahmud's time, an amir survived in
Chaghaniyan from the old-established Muhtajid family, and in Mascud b. Mahmud's reign
the local amir (of unspecified family) was the sultan's son-in-law. 16
Within Afghanistan proper, there was no strategic need to maintain buffer-states like
these. The ruler of Gharchistan in northern Afghanistan, the Shir, was deposed in 1012
and his principality was incorporated into the Ghaznavid empire, and the neighbouring
one of Guzgan likewise in 1010-11 when its rulers, the Farighunids, apparently failed
in the male line. 17 The remote region of Ghur in central Afghanistan remained, however,
substantially a pagan enclave outside Ghaznavid control. Mahmud sent expeditions in 1011
and 1020 and with difficulty secured the submission of some local chiefs, including the

13
Sachau, 1873,pp. 292-311; Barthold, 1968, pp. 275-9; Nazim, 1931 , pp. 56-62, 177-8; Tolstov, 1953,
pp. 290-1; Bosworth, 1963, p. 237; 1975, pp. 174-5.
14
Nazim, 1931 , p. 74; Bosworth, 1975, p. 173.
15
Bosworth, 1963, pp. 73, 239; Bosworth and Clauson, 1965, pp. 8-9; Bosworth, 1975, pp. 173-4.
16
Bosworth, 1963, p. 237.
17
Nazim, 193 1, pp. 60-2, 177-8; Bosworth, 1975, pp. 171-2.

109
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

lord of Ahangaran, Muhammad b. Suri of the Shansabani family (who were to be the
driving force behind the remarkable rise to power in the next century of the Ghaznavids'
supplanters, the Ghurids). Ghur was never properly subdued, however, and the implantation
there of Islam was to be a slow process (see below, Chapter 8). 18
Sistan, in south-western Afghanistan, was a region with strong traditions of its own
independence going back to the Saffarids, when it had been the centre of a vast if transient
military empire. Despite the collapse of this empire, scions of the Saffarids had survived in
Sistan as petty rulers under generally nominal Samanid suzerainty. Amir Khalaf b. Ahmad
had become Sebtiktegin's neighbour after the latter's annexation of Bust, and he feared
for the integrity of his own territories. He temporarily seized Bust while Sebtiktegin was
involved with Jaypal, seized Pushang while Mahmud was disputing with Ismacn over the
succession and intrigued with the Karakhanids. Hence Mahmud sent forces into Sistan in
999 and 1003, on the latter occasion deposing Khalaf and annexing his territories. Never-
theless, local feeling in Sistan was always strongly anti-Ghaznavid, and when the Seljuq
Turks appeared on the fringes of Sistan in the 1040s, in Mawdud b. Mas'ud's sultanate, the
people there threw off the Ghaznavid yoke and raised to power a line of local chiefs as the
Maliks of Nimruz. 19
The story of Mahmud's Indian campaigns does not directly concern the history of Cen-
tral Asia, but so important an aspect of Ghaznavid policy requires some discussion. Sebtik-
tegin's clashes with Jaypal have been mentioned above. Mahmud likewise felt that the
Hindushahis were a major obstacle to any expansion into northern India. He defeated
J aypal in 1001, and his son Anand pal in 1009, and then a coalition of the princes of
Kashmir and other regions of northern India, alarmed at the threat from the Turushkas
(as the [Ghaznavids'] Turks appear in Indian sources), under the leadership of Anandpal's
son Trilochanpal (1004). Further coalitions, including the Rajahs of Kalinjar and Kanawj,
failed to stem the Ghaznavid onslaught. Trilochanpal died in 1021, and with the death of
his son Bhimpal in 1026, the line of Hindushahi Rajahs came to an end; surviving mem-
bers of the family took refuge in Kashmir (which Mahmud made a disastrous attempt at
raiding in 1021, but which was not to be seriously penetrated by Islam until the fourteenth
century). 20
Other expeditions were mounted by Mahmud into what is now eastern Panjab and into
the Ganges plains (modem Uttar Pradesh) and central India (modem Madhya Pradesh).
Thus in 1004 the Rajah of Bhatinda, to the south of Lahore, was attacked and an immense

18
Nazim, 193 1, pp. 70-3.
19
Ibid., 1931, pp. 69-70, 187-9; Bosworth, 1975, pp. 172-3; 1994, pp. 321-8, 368 et seq.
20
Nazim, 193 1, pp. 86-96; Ray, 193 1-6, Vol. I, pp. 84-101; Habib, 195 1, pp. 23-4, 27-33, 36-7, 41-7.

110
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

booty, including 120 elephants, taken. In 1009 the ruler of Narayanpur, in north-eastern
Rajasthan, was humbled, agreeing in his peace treaty with the sultan to supply a contin-
gent of 2,000 Indian soldiers for the Ghaznavid army. The rich and fertile Doab between
the Ganges and the Yamuna was the seat of many wealthy shrines and temples. Thanesar,
where the Hindus venerated an idol named Chakraswami (Lord of the Wheel), was plun-
dered in 1014 and the idol carried off to Ghazna. The expedition in the winter of 1018-19
captured rich booty from Mathura or Muttra, the reputed birthplace of the deified Indian
hero Krishna, and then pushed on to confront one of the leading Hindu princes of north-
ern India, the Pratihara Rajah of Kanawj, gaining from the whole expedition what was
reckoned by the historian cutbi at 3 million dirhams, 55,000 slaves and 350 elephants. 21
Over the ensuing years, expeditions were sent against a coalition of princes under Ganda
of Kalinjar, including also the rulers of Kanawj and Gwalior (1019-20,1022-3), penetrat-
ing into central India. But the culmination of Mahmiid's Indian campaigns was his attack
on Somnath on the southwestern coast of the Kathiawar peninsula (1025-6), where lay a
temple with the lingam of the moon-god Mahadeva, endowed with fabulous riches. The
raid involved an arduous march from Multan across the Thar desert and an equally diffi-
cult return one through Cutch, harassed by the Jhats, to Sind, with 30,000 of the regular
army plus volunteers. The immense plunder gained from the despoiled temples, said to
total over 10 million dinars and brought back to Ghazna, gave ample recompense for the
hardship endured. Above all other raids of the sultan, the attack on Somnath caught the
imagination of the Islamic world. Rich gifts were sent to Baghdad and the caliph awarded
the sultan - his reputation as 'hammer of the infidels' now much inflated - further honorific
titles. Over the centuries, a rich accretion of stories and legends was to attach itself to the
historical core of the episode. 22
Although later generations oflndian Muslims were to venerate Mahmud as the founding
father of Islam in India, the Ghaznavid expeditions should in reality be seen as essentially
plunder raids, as has been emphasized by, for example, Mohammad Habib. 23 Their aim
was to exact tribute from the Indian princes in the shape of gold, elephants, slaves and -
quite often - troop contingents for the Ghaznavid army, while the treasures from despoiled
temples were taken to Ghazna and either converted into negotiable form or else used to
adorn and beautify public buildings such as palaces, gardens and mosques. Such gains
were especially welcomed by pious Muslims as mal-i halal (lawful wealth) as opposed to
the taxation collected from Muslim subjects, frequently by oppression and violence. No

21
Nazim, 193 1, pp. 99-104, 106-10, 197-203; Ray, 193 1-6, Vol. I, pp. 84-98; Habib, 195 1, pp. 38-42.
22
Nazim, 1931 , pp. 110-22, 209-24; Habib, 1951 , pp. 48, 51-9.
23
See Habib, 195 1, pp. 76-7, 81-4

111
ISBN 978-92-3-103467-1 Contents The zenith of the empire under Mahmud

conditions of adherence to Islam were imposed on the Indians, nor could any such condi-
tions be enforced, for as soon as the Ghaznavid forces withdrew homewards, any forced
converts would have apostatized. Only in western Panjab, where Lahore became the
concentration-point for ghazfs, can Islam have become gradually implanted at this time,
adjacent as this region was to existing Muslim communities of the middle and lower Indus
valley established there since the Arab conquest of Sind at the beginning of the eighth cen-
tury. Only at Lahore, too, was there an attempt in the latter years of Mahmud' s reign and in
the early ones of Mas cud's to set up a civil administration, with the aim of making western
Panjab something like a regular province of the empire for fiscal purposes. This attempt
foundered because of the unpacified state of northern India and the sultan's inability to
control bellicose and volatile military elements and ghazfs in the Muslim garrisons. 24
Over the three centuries since Muhammad b. al-Qasim al-Thaqafi's conquest, the exist-
ing Islam of the Indus valley Arab communities had acquired what was, in the eyes of the
sternly orthodox Mahmud, a heretical tinge. During the tenth century, the Muslims of Sind
and Multan had come to recognize the spiritual and moral supremacy not of the cAbbasids
but of their Shicite Fatimid rivals (see above, Chapter 2). Hence although the local ruler
in Multan, Abu '1-Fat'h Dawud, had been on friendly terms with the Ghaznavids, in two
campaigns of 1006 and 1010 the sultan attacked Multan, massacred the Ismacilis there
(called in contemporary phraseology Qaramita or Carmathians) and deposed Abu '1-Fat'h.
In cutbi's words, 'He was unable, in the interests of religion, to endure that he [Abu '1-
Fat'h] should remain in power, seeing the vileness of his evildoing and the abomination of
his affair.' 25 The fact that the remaining people of the prosperous trading city of Multan had
to pay a heavy fine to save it from being plundered by the Ghaznavid army, however, indi-
cates that the enforcement of orthodoxy could have its profitable side. Even so, Ismacilism
survived there and, probably after Mascud's deposition and death in 1041, the Ismacilis of
Multan once more rose against Ghaznavid control, under Abu '1-Fat'h Dawud's son, and
the new sultan, Mawdud, had to send a further punitive expedition. 26
Such campaigns as these formed part of the image that Mahmud carefully built up
round himself as the hero of Sunni orthodoxy and the scourge of heretic Muslims and of
infidels like the Hindus or the pagans of Ghur and Kafiristan (modern Nuristan). Although
earlier in his career as a commander in Khurasan, Mahmud had, like his father, given sup-
port to the leaders of the pietistic Karramiyya sect, which was strong in Nishapur, 27 the

24
Habib, 1951 , pp. 73, 76-87; Bosworth, 1963, pp. 75-6, 77-8, 114-16; 1962a, pp. 54-6.
25
Al-Tarikh al-Yaminii, cited in Bosworth, 1963, p. 52.
26
Nazim, 1931, pp. 96-9; Ray, 1931- 6, Vol. I, pp. 1-54; Stern, 1949, pp. 298-307; Bosworth, 1963, pp.
52, 76; 1962a,pp.54-5; 1977,pp.30-1.
27
Barthold, 1968, pp. 289-90; Bosworth, 1963, pp. 185-9; Bulliet, 1972, pp. 42-3, 159-60, 203-5.

112
ISBN 978-92-3-103467-1 Contents Mascud and the Seljuqs

Ghaznavids assimilated themselves to the norm for most eastern Iranian and Turkish peo-
ples: adherence to orthodox Sunnism and the Hanafi law school. This involved allegiance to
the moral and spiritual heads of Sunni Islam, the cAbbasid caliphs, by now living a reduced
existence in Baghdad as pensioners of the Buyids and with their court overshadowed cul-
turally and intellectually by the splendour of the Fatimids in Cairo. Mahmud needed the
moral backing of the cAbbasids when first he took over Khurasan and supplanted his lawful
suzerains, the Samanids, just as his son Mascud was to require it for the succession struggle
with his brother Muhammad on their father's death in 1030. Hence gifts from plunder and
announcements of victories were sent regularly to Baghdad, a harmless envoy from the
Fatimids to the Ghaznavid court was summarily executed and accusations of 'Carmathian'
sympathies were used to justify the removal of the sultan's internal enemies; in return, the
sultans received from Baghdad grandiloquent titles and other insignia of royal power. 28
The excuse of an anti-Shfite crusade became the justification for Mahmud's last great
campaign, directed against the Buyid amirate of northern Iran. The weak ruler there, Majd
al-Dawla, unable to control his Daylamite soldiery, injudiciously appealed to the sultan for
help. Until then, Mahmud had been circumspect in his dealings with the Buyids, still the
dominant power in Iraq and in Iran west of Khurasan. On this pretext, in 1029 Mahmud
marched against Rayy, deposed the amir and sacked the city, carrying off rich booty from
what was the main commercial and industrial centre of northern Iran. Massacres of hetero-
dox elements, described as Batiniyya (sc. Ismacilis), Muctazilites, Mazdakites, and so on,
gave the sultan religious backing for his aggression. Possession of Rayy, the strategic key
to northern Iran, opened up the possibility of a drive towards the west and crusades against
the Byzantines and Fatimids. Amir Mascud was dispatched with an army against local
Daylamite and Kurdish rulers in western Iran, although these operations were brought to
an end by the sultan's death in 1030. The Ghaznavids thus came to control most of north-
ern Iran, either through direct conquest or through vassals like the Ziyarids of Gurgan and
Tabaristan in the Caspian coastlands. Their rule was to last a mere seven years, however,
for the growing power of the Seljuqs and their Turkmen followers made it impossible for
the sultans to retain their Iranian and Central Asian provinces (see below). 29

Mascod and the Seljuqs

On Mahmud's death, rule in Ghazna passed briefly to his son Muhammad in accordance
with the dead man's wishes, but after a few months, Mascud arrived back from Rayy with

28
Bosworth, 1963, pp. 51-4; 1962a, pp. 59-66.
29
Nazim, 193 1, pp. 80-5; Bosworth, 1963, pp. 53; l 962a, pp. 67-75.

113
ISBN 978-92-3-103467-1 Contents Mascud and the Seljuqs

the army of the west and Muhammad's support melted away. In an empire built up solely
through Mahmiid's personal skill as a war leader, the military and civilian notables in
Ghazna speedily recognized that the more experienced Masciid was better fitted to main-
tain the momentum of his father's conquests. Unfortunately, Mascud had not inherited all
Mahmud's capabilities; though personally brave in the field, his judgement was less sound
and his advisers were to complain of his arbitrary decisions and unwillingness to listen to
good counsel. In the early years of his reign, he conducted a vendetta against the great men
of state who had been dominant in Mahmud's reign and who had in many cases initially
supported the ephemeral sultanate of Muhammad, the Mahmudiyan or Pidariyan ('adher-
ents of the father'), as Bayhaqi calls them; instead, there began the ascendancy of the
Mascadiyan or Naw-khwastagan ('upstarts'), who tended to act as the sultan's yes-men.
But it may be that we are unduly influenced by hindsight and by the fact that Mascud's
reign (1030-41) ended in failure and his death. Differences in character between him and
his father were important but not decisive. The crucial point was that Masciid inherited a
vastly over-stretched empire, one which was rapidly threatened by a new factor that even-
tually overwhelmed the Ghaznavids in the west: the irruption of the Turkmen hordes into
Khwarazm and northern Iran. 30
At first, Mascud continued his father's policies in both India and the Iranian lands.
He tried to round off the recent Iranian conquests by the acquisition of Kerman from the
Buyids in 1003, having dreams of outflanking the Buyids via Makran and Oman and lib-
erating the caliph in Baghdad from their tutelage, but Ghaznavid financial exactions in
Kerman favoured the return of the Buyids the following year. 31 In 1035 the sultan led a
punitive expedition against the Ziyarid ruler Abu Kalijar because of his arrears of trib-
ute; but the violent methods of the Ghaznavid army in collecting taxation at the capital
of Tabaristan, Amul, caused a revulsion against Ghaznavid rule which echoed throughout
the eastern Islamic world. 32 In India, Mahmud's gains were retained and Mascud person-
ally led fresh campaigns, such as that of 1037 against the 'Virgin Fortress' of Hansi to
the north-west of Delhi; he also managed to restore order in Panjab after the comman-
der of the army of India based in Lahore, Ahmad Inaltegin, rebelled in 1033. But the
difficulties of controlling unruly ghazf elements led to frequent disturbances in the Ghaz-
navid territories in India. Moreover, because of Mahmiid's spectacular successes there, and
the importance of an inflow of tribute and plunder in maintaining the fabric of the state,
India and its problems tended to dominate the central councils of the empire, and from the

30
Barthold, 1968, p. 293; Bosworth, 1963, pp. 227-34.
31
Bosworth, l962a , pp. 74-5.
32
Bosworth, 1963, pp. 84, 90-1.

114
ISBN 978-92-3-103467-1 Contents Mascud and the Seljuqs

middle years of Masciid's sultanate onwards, this became a source of weakness. Threats
to the territorial integrity of the empire on its northern frontiers, from the Karakhanids
and the Seljuqs, were not properly faced and the defences there neglected. Thus despite
promptings from his more perspicacious ministers, Masciid's attentions remained divided,
with the result that the situation in the west became out of hand: the Turkmens enlarged
the foothold gained there in Mahmud's closing years and could not now be dislodged. 33
Masciid initially cultivated good relations with the Karakhanid Kad'ir Khan Yusuf and
his son and successor, Bughra Khan Sulayman, and himself married one ofYusuf's daugh-
ters in 1034. cAli Tegin remained the common enemy and in 1032 Mascud sent an army
against him into Transoxania under the Khwarazm Shah Altuntash. Bukhara was captured,
but a battle at Dabusiyya against cAli Tegin and his Seljuq allies was indecisive and led
to the death of the wise and experienced Khwarazm Shah. The latter was succeeded as
governor by his son Harlin who, mindful of Mascud's earlier attempt to have his father
Altuntash, as one of the Mahmudiyan, murdered, adopted a hostile attitude to Mascud.
Harlin allied himself with cAli Tegin for a joint attack on Ghaznavid territories in northern
Afghanistan in 1034 and also gave help to the Seljuqs who had settled on the fringes of
Khwarazm. Mascud' s position was saved by the deaths of cAli Tegin in 1034 and of Harlin
(the latter was assassinated at the sultan's instigation in 1035). But cAli Te gin's sons were
still active along the Oxus frontier; and after 1038, Bori Tegin (see above) began harry-
ing Khuttal and Wakhsh in alliance with the Kumijis and the Turkmens, soon taking over
almost all Transoxania. 34 Khwarazm had meanwhile slipped irrevocably from Ghaznavid
control. The sultan could only join with a group of the Oghuz hostile to the Seljuqs, those
under the leadership of Shah Malik, who held the ancient Turkic title of Yabghu and who
controlled the towns of Jand and Yengi-kent at the debouchment of the Syr Darya into
the Aral Sea (see Chapter 7, Part One, below). Shah Malik did in fact secure control of
Khwarazm and proclaim Ghaznavid suzerainty there once more in 1041, but by that time
Mascud was dead. Three years later, the Seljuqs expelled Shah Malik himself and became
universally victorious in Khurasan and Khwarazm. 35
As mentioned above, the role of the Seljuqs was decisive in the downfall of Ghaz-
navid power in the west. Bands of Oghuz Turks from the steppes of south-western Siberia
(the modem Kazakhstan) had been infiltrating southwards into the settled lands since the
last decades of the tenth century, at times aiding the last Samanids and at others their
Karakhanid supplanters. Support from these Oghuz - since the start of the eleventh century
33
Bosworth, 1963, p. 235.
34
Barthold, 1968, pp. 293-6; Bosworth, 1963, pp. 235-40.
35
Sachau, 1873, pp. 310-12; Barthold, 1968, pp. 296-9, 302; Tolstov, 1953, pp. 291-2; Pritsak, 1953,
pp. 405-10; Bosworth, 1963, pp. 238-9; 1968a, pp. 51-2.

115
ISBN 978-92-3-103467-1 Contents Mascud and the Seljuqs

at least superficially Islamized - was decisive in enabling cAli Tegin to retain power in the
Bukhara region from c. 1025 onwards. Dislodged by the combined operations of Mahmud
of Ghazna and his ally Kad'ir Khan Yusuf, some 4,000 Oghuz families, former followers
of Arslan Isra'il b. Seljuq, sought permission from the sultan to settle on pastures in north-
ern Khurasan, promising to guard the frontiers there against further nomadic incursions.
This proved a delusion and Ghaznavid forces had to disperse bands of plundering Oghuz
throughout northern Persia. Others of the Oghuz, led by the Seljuqs Toghrll, Chaghr'i, Musa
and Ibrahim Inal, remained north of the Oxus, involved in the politics of Khwarazm until
they too were compelled to move southwards into Ghaznavid Khurasan in 1035. There,
their herds devastated the agriculture of the oases, and by their depredations they disrupted
long-distance commerce also. Ghaznavid forces sent against them failed to achieve per-
manent success and by 1037-8 such leading towns as Rayy, Merv and Nishapur opened
their gates to the Seljuqs, despairing of ever receiving adequate protection from the sultan.
Mascud was at last deflected from his Indian preoccupations and marched westwards; but
his heavily armed and ponderous, conventional-type Islamic army was defeated by some
16,000 Turkmens at Dandanqan in the desert between Merv and Sarakhs in 1040. 36
The result of this battle, one of the most decisive in the history of the eastern Islamic
world, was that the Seljuqs, now proclaiming their allegiance to the Baghdad caliph as their
sole suzerain, were able within the next 20 years to take over the whole of Iran and make
it the nucleus of the Great Seljuq empire. The Ghaznavids lost all their western provinces;
the frontier was stabilized in c. 1059 on a line roughly bisecting modern Afghanistan from
north to south. 37 In future, the orientation of the Ghaznavid empire was to be towards India
(for which the despairing Sultan Mascud had set out after the Dandanqan debacle) and the
exploitation of its riches. Thus the history of the remaining 140 or so years of the sultanate
concern primarily eastern Afghanistan and India rather than Central Asia; in the last years
of its existence, the sultans ruled from Lahore and not Ghazna.
Mawdud b. Mascud (1041-8) was the last ruler to endeavour to concert operations with
the Karakhanids against the Seljuqs. 38 The upper Oxus territories remained the subject of
Ghaznavid-Seljuq disputes during the reign of Ibrahim b. Mascud (1059-99), but Bahram
Shah (1118-52) ruled in a loose vassal status to the Great Seljuqs, who were however
unable to save the last Ghaznavids from the rising power of the Ghurids from the modern
Ghorat province of central Afghanistan (see below, Chapter 8). 39 In retrospect, it appears
that the vast empire assembled by Mahmud could no longer be held together by one man
36
Barthold, 1968, pp. 297, 302-3; Nazim, 1931 , pp. 62-6; Bosworth, 1963, p. 241-68; 1968a, pp. 15-23.
37
Bosworth, 1977, pp. 50-2.
38
Bosworth, 1968a, pp. 52-3; 1977, pp. 25-7.
39
39. Bosworth, 1968a, pp. 158-9; 1977, pp. 52-5, 91-101.

116
ISBN 978-92-3-103467-1 Contents Nature and structure of the Ghaznavid state

once the incursions of the Seljuqs had reached a certain level of intensity. The central
administration in Ghazna (on which, see below) was unable to preserve its communications
with distant provinces like northern Persia and Khwarazm once these regions came under
pressure or threw off allegiance to the sultan. In fact, one might conclude that the loss of
the western provinces to the Seljuqs reduced the Ghaznavid empire to more manageable
proportions, thus enabling it to survive right down to its extinction in 1186 by the Ghurids:
a respectable span of life for an Islamic state.

The nature and structure of the Ghaznavid state

The Ghaznavids display the phenomenon of the rapid transformation of a line of bar-
barian, originally Turkish slaves into monarchs within the Irano-Islamic tradition who
presided as authoritarian rulers over a multi-ethnic realm comprising Iranians or Tajik:s,
Turks, Afghans, Indians and others. In this age before the evolution of the nation-state, the
possession of subjects who, with their various ethnic backgrounds, could contribute differ-
ing expertise and skills to the functioning of the state was regarded as a source of strength
rather than of weakness. The sultans could never forget their Turkish ethnicity, since the
essential core of their military support was likewise Turkish. Thus they had to stay attuned
to the needs and aspirations of those troops and, above all, to act as successful war leaders
and suppliers of plunder; when Mascud's nerve seemed to fail after Dandanqan, the army
abandoned him. But for the administration and financing of their empire, the Ghaznavids
early recognized the need for the services of their Iranian subjects, above all, for the secre-
tarial class, whose secular traditions and practices went back beyond the Islamic caliphate
and ultimately to the Sasanians. To this heritage had been added the Islamic element, with
a stress on the ruler's divine backing but also on his duty to act in consonance with the laws
of God as exemplified in the sharf a; yet in practice, rulers behaved largely as despots.
All these elements came together within what might be called the Ghaznavid 'power-
state', in which the sultan and his servants, both military and civilian, stood over and
against the mass of subjects (the division which was later to be termed, among the Ottomans,
that of caskerfs and recaya). There were no national or patriotic feelings which could act
as a cement for society, since the vision of the subject, were he peasant, trader or artisan,
was confined to his own locality and to the protection of his own immediate interests. It
was the duty of the subjects to pay taxes to the state; in Bayhaqi's words, 'It is vital that
they should be in complete fear and trembling of the king and the army, and give [them]
complete obedience.' 40 The sultan and his servants, on the other hand, had an implied duty

40
Tarr:kh-i Mascadr:, cited in Bosworth, 1963, p. 50.

117
ISBN 978-92-3-103467-1 Contents Nature and structure of the Ghaznavid state

to protect the subjects militarily and thereby enable them to carry on their avocations; but
the two spheres of responsibility were never to mix. According to Bayhaqi again, Mahmud
severely censured the people of Balkh for resisting the Karakhanid invaders in 1006, thus
causing the destruction in the town of a market belonging to the sultan which had brought
in much revenue; they should simply have submitted to the more powerful incomers. 41
Hence Mahmud's son Mascud was only following the same line of argument when at the
end of his life he resolved to abandon Ghazna for India, instructing his officials to make the
best terms they could with the Seljuqs whom he expected to occupy the capital and replace
his dynasty there. 42
Both the central and the provincial administration were run by Tajiks, with the vizier
at the head of the hierarchy. The sultans disposed of the services of some of the leading
litterateurs and officials of the age, such as Sebtiktegin's chief secretary Abu '1-Fat'h Busti,
and the vizier under both Mahmud and Mascud, Ahmad b. Hasan Maymandi. There was a
fivefold division in the central bureaucracy, with separate dfwans for the vizier, the chief
secretary, the secretary for the army, the head of intelligence and the postal service and
the chief steward of the household. 43 Financial demands were the overriding considera-
tion. The maintenance of a powerful standing army and the mounting of frequent military
campaigns were very expensive, as was the sultans' extravagant lifestyle in their palaces
and gardens and their lavish spending on public buildings. The bureaucracy, and with it the
wages bill, grew with imperial expansion. 44 The vizier was thus under constant pressure
to increase the flow of revenue and to find new sources of taxation. Failure here meant
dismissal, torture to disgorge personal gains and often execution; of the six viziers serving
Mahmud, Muhammad and Mascud, three died violent deaths and one suffered prolonged
imprisonment. The vizier had also to keep control over provincial governors and officials,
who might be tempted, through distance from the capital, to withhold taxation and rebel;
this control was exercised through a network of couriers and spies, the barfd and ishraf
system, thus contributing to the atmosphere of fear and suspicion in the empire. 45
There is ample evidence in the sources of the harshness of provincial tax-collectors, but
the sufferings of the subject population were even worse when the soldiers of the army
were allotted assignments of revenue (barats) which they then collected personally. We
know that the people of Rayy, who had originally welcomed the Ghaznavids as liberators
from the excesses of the Buyid troops there, speedily turned against an oppressive military

41
Barthold, 1968, p. 291; Bosworth, 1963, pp. 48-51.
42
Bosworth, 1963, p. 51.
43
Nazim, 193 1, pp. 130-47; Bosworth, 1963, pp. 55-97.
44
Bosworth, 1963, pp. 65-7.
45
Nazim, 193 1, pp. 132-7, 144-6; Bosworth, 1963, pp. 70-3, 93-7.

118
ISBN 978-92-3-103467-1 Contents Nature and structure of the Ghaznavid state

governor and were later reluctant to put up any serious resistance to the Seljuqs; and the
notorious exactions of the vizier Isfara'ini and the governor Abu '1-Fadl Suri in Khurasan
contributed powerfully to the population's disenchantment with the Ghaznavids and the
capitulation of their towns to the Seljuqs without a blow. 46
The dfwan of the chief secretary dealt with correspondence with provincial officials and
with external rulers. We possess in sources such as Bayhaqi's history and c Aqili's Athar
al-wuzara' [Famous Past Deeds of the Viziers or Past Traces of the Viziers] the texts, in
florid Arabic and Persian, of several letters to the Karakhanids and the c Abbasid caliphs,
including announcements of victories (jat'h-namas). 47 Of especial importance in a mili-
tary state like that of the Ghaznavids was the dfwan of the carid (secretary for the army),
who organized mustering, the provision of materiel and the commissariat, and pay. There
were regular army parades on ceremonial occasions such as the reception of diplomatic
envoys or the celebration of the Islamic festivals and the Iranian ones of Nawriiz, Sada and
Mihrgan, these parades being often held on a plain outside the capital. 48
The core of the army was an elite force of Turkish ghulams who guarded the sultan's
palaces and person and hence were known as ghulaman-i saray (palace guards). What
we know from the literary sources of their rich uniforms and bejewelled weapons has
received striking confirmation from the remains of mural paintings in the audience hall
of the complex of Ghaznavid palaces at Lashkar-i Bazar near Bust. 49 From these palace
ghulams - recorded as amounting to 4,000 at a review in 1037 - were drawn the holders
of household and ceremonial offices such as the sultan's armour-bearer. Outside this inner
group, however, a wider force of Turkish slave troops formed the backbone of the army; the
Ghaznavids were thus continuing a feature of military organization begun by the c Abbasids
in the early ninth century and adopted by most of their successor-states, including the
Samanids. Turkish troops like these were valued above all for their hardiness, stemming
from their harsh early life in the steppes, their skill as mounted archers and the single-
minded loyalty which in theory (though not always in practice) they gave to their master.
The sources tell us little about the tribal origins of the Ghaznavids' Turks, who came
mostly via the slave markets of Transoxania or as gifts from the Karakhanid rulers, but
specifically mentioned are soldiers from the Karluk, Yaghma, Kay, Tukhsi and Chigil
and the men of Khotan (?Uighurs), while the frontier auxiliaries recruited by the Ghaz-
navid governors in Khwarazm apparently included men from the Kipchak and Kangh. 50

46
Bosworth, 1963, pp. 79-91; cf. Bulliet, 1972, pp. 69-70.
47
Nazim, 1931 , pp. 142-4; Bosworth, 1963, pp. 91-3.
48
Nazim, 193 1, pp. 137-8; Bosworth, 1963, pp. 122-6.
49
Bosworth, 1963, p. 104; Schlumberger et al., 1978, Part lA, pp. 101-8, Part IB, pp. 121-4.
50
Bosworth, 1963, pp. 98-105, 109-10.

119
ISBN 978-92-3-103467-1 Contents Nature and structure of the Ghaznavid state

Other nationalities within the Ghaznavid forces included free troops from the local Arabs
and Kurds, valued as dashing cavalrymen and skirmishers; Daylamite infantrymen, who
fought with their characteristic weapons of the spear and javelin; and Indians. These last
were probably in large part slaves, valued for their loyalty and at times used as a counter-
balance to the Turks; conversion to Islam does not seem to have been necessarily required
of them. 51 The elephants which were deployed as beasts of war in the Ghaznavid armies
have been mentioned above; it was the sultans who reintroduced these animals, taken as
tribute from the Indian princes, into military usage in the Persian lands, where they had last
been thus employed by the Sasanians. 52
The ideology of the sultans was strongly orthodox and Sunni, and, as already described,
they strove to build a contemporary image as defenders of Islam against heterodoxy and
infidelity. We have little evidence of any deep personal faith on the part of Mahmud or
Mascud, and they certainly enjoyed wine-drinking parties to the full. They nevertheless
recognized the importance of the official religious institution of the cuzama' (learned men)
as part of the fabric of state and often used scholars as diplomatic envoys, with an especial
penchant for the Hanafis, characterized by Mahmud at one point as the madhhab-i rast
(righteous law-school). Hence the services of the leading Nishapur Hanafi family of schol-
ars and lawyers, the Tabanis, were often called upon. In 995 Mahmud, as commander in
Khurasan, had invited Abu Salih Tabani to become head of the Hanafi lawyers there and to
teach in a madrasa (Islamic college); Abu Salih's nephew Abu Sadiq was later appointed
by Mahmud as chief judge in Khuttal, and 1037 sent by Mascud to head a successful
embassy to the Karakhanids, with the promise of the judgeship of Nishapur as a reward.
Another prominent Hanafi family in Khurasan was that of the Sacidis; Mahmud appointed
the judge Abu '1-c Ala Sacid as tutor to his sons Muhammad and Mascud, and this scholar
was to play a leading role in reducing the power of the Karramiyya in 1012. 53
There was clearly considerable royal patronage of the Sunni revival in the eastern
Islamic lands as part of the general movement which had grown up in reaction to the
bid for political power in the tenth century of Shfism. Already in the opening years of
the eleventh century, it is recorded that in Khuttal there were over twenty madrasas, these
colleges being one of the chief instruments for the education and training of an orthodox
Sunni religious and official class. 54 Virtually all members of the ruling strata of Islamic
society were susceptible to the appeal of a holy man or mystic, while people were often
aware that there were many charlatans in the ranks of the Sufis, and this seems to have

51
Ibid., pp. 107-12.
52
Ibid., pp. 115-18.
53
Ibid., pp. 175-8; Bulliet, 1972, pp. 63-4, 201-4.
54
Bosworth, 1963, pp. 175, 176-7.

120
ISBN 978-92-3-103467-1 Contents Cultural and intellectual life

been substantially the attitude of the sultans. It is difficult to distinguish fact from pious
fiction in the biography of the Khurasanian Shaykh Abu Sacid of Mayhana by his descen-
dant; hence we cannot know for sure whether the sultan really paid off the remaining debt
on Abu Sacid's khanaqah (convent) when the latter died in 1049. One would expect the
Ghaznavids to have lent more support to the moderate Sufi groups of the time, such as that
around the Nishapur scholar Abu '1-Qasim al-Qushayri, than to an extravagant thaumaturge
like Abu Sacid. 55 (See further below, Chapter 18.)

Cultural and intellectual life

As well as being aware of the importance of the Islamic religion in buttressing their author-
ity, the Ghaznavids were concerned to conform to the norm of traditional Islamic rulers
who made their courts centres of culture and learning, with the monarch as the recipient of
laudatory poetry and as the dispenser in tum of patronage and largesse. Displaying some-
thing of the admiration that barbarians often showed for higher things, Mahmud seems to
have chosen the ideal of the Samanid court, which had nurtured such luminaries as Rudaki
and Daqiqi (see above, Chapter 4 ), for his own circle at Ghazna, even if we discount the
later literary biographer Dawlat Shah's claim that there were 400 poets in attendance on
Mahmud, led by the laureate cunsuri and hymning the sultan's praises as a Maecenas and
as the scourge of the kafirs (pagans). The courts of Mahmud and Mascud certainly nurtured
a fine school of lyric poetry in New Persian, with such notable figures as cunsuri, Far-
rukhi and Manuchihri (there were poets of similar calibre at the courts of later sultans like
Mascud III (1099-1115) and Bahram Shah). 56 There are indications that Turkish poetry
was also known there. 57 But modem scholars have emphasized that such encouragement
did not necessarily arise from a disinterested love of learning. A well-known anecdote by
the twelfth-century writer Nizami cAriidi Samarqandi describes how Mahmud peremp-
torily demanded of the Ma'munid Khwarazm Shah that he dispatch to his own court in
Ghazna the leading literary and scientific figures in Gurganj (the physician and philosopher
Ibn Sina was so little enchanted at the prospect that he fled westwards, eventually reaching
the court of the Daylamite prince Ibn Kakuya at Isfahan). Indeed, the greatest intellect of
the age, the polymath Abu Rayhan al-Biruni (973-1048), while taking advantage of the
facilities offered to this Khwarazmian scholar by service under the Ghaznavids, including
the opportunity to visit India and gather material for his magnum opus, the Tahqfq ma li

55
Ibid., pp. 187-94; Meier, 1976, pp. 325-6.
56
Browne, 1906, pp. 90-129, 317-22, 324-6; Rypka et al., 1968, pp. 172-7, 196-7.
57
Koprtiltizade, 1934, pp. 26-32; Gandje'i', 1986, pp. 68-9.

121
ISBN 978-92-3-103467-1 Contents Agriculture and trade

'l-Hind [Inquiry Into What is to be Found in India], never seems to have been especially
close to Mahmud or to have enjoyed any direct official encouragement from him. 58
As observed above, a substantial proportion of the state revenue went on the sultans'
palaces and on entertainment there. To some degree, this was necessary to the functioning
of the state, which ensured the loyalty of the army and the bureaucracy by the distribution
of favours, offices and largesse and by the visible and conspicuous image of the sultan as
the munificent and successful head of state. Such sources as Bayhaqi describe at length
the splendours of court life, with rich clothing and robes of honour provided by the royal
workshops (karkhanahs) for the embroidery of tiraz decoration, and with the traditional
medieval Islamic separation of the open, public, court life of the ruler from his private life
in the harem with its eunuch attendants. The numerous palaces and gardens at Ghazna and
at provincial centres like Herat and Balkh have failed to survive in the harsh Afghan envi-
ronment, although the remaining ruins at Lashkar-i Bazar (see above) show the magnificent
scale on which such palaces were conceived. 59

Agriculture and trade

Given the fact that the sources for Ghaznavid history are all products of an elite, court cul-
ture, we can only piece together odd fragments of information on the life and social habits
of the mass of the population, the peasants, traders and artisans. We can, however, assume
that demands for taxation (the kharaj, or land tax) and a multitude of local tolls and dues
(mukus), pressed hard on them; and we know from Bayhaqi that forced labour, corvees
(mard-bzgarl), was exacted for the construction of palaces and the driving of game on
royal hunts - a practice strongly entrenched in Iranian and Afghan life and lasting almost
to the present day. 60 Agriculture was concentrated on the oases and was essentially small-
scale and designed for subsistence within the rural area concerned or for supplying towns
like Herat, Merv and Nishapur which could not grow enough food for themselves. Only
certain highly specialized foodstuffs like truffles and the edible earth of Khurasan are men-
tioned as being exported as far as Egypt and the Turkish lands. 61 Within the Ghaznavid
realm, comprising the plateaux and mountain regions of Afghanistan and Khurasan, the
only significant permanent sources of running water were rivers like the Oxus, Murghab
and Helmand, on which the geographers mention the existence of water mills. Most irri-
gation came from subterranean qanats or karfzs, requiring large injections of capital for

58
Bosworth, 1963, pp. 131-4.
59
Ibid., pp. 135-41.
60
Bosworth, 1963, pp. 79-91, 141.
61
Bosworth, 1968b, pp. 131-2.

122
ISBN 978-92-3-103467-1 Contents Agriculture and trade

construction and maintenance which only the more wealthy landholders could provide. The
sultans themselves are recorded as responsible for hydraulic constructions in the region of
Ghazna itself. Only a few favoured upland areas allowed dry-farming; the lush pastures of
the upper Oxus valley and its tributaries were, however, famed for horse breeding. 62
Industrial production was the small-scale activity of artisans and craftsmen, and was
mainly for local consumption. Only within Khurasan, where virtually all the towns pro-
duced textiles or carpets, were certain celebrated local fabrics, such as the cattabz and
saqlatunf silk brocades of Nishapur, the white cottons of Herat and the gold-threaded mul-
ham cloth of Merv, exported outside the province. 63 It is only for such towns as these
that we know anything about municipal organization, involving the presence of a class of
notables and leading families, who produced scholars, preachers and judges for the official
religious hierarchy (such as the above-mentioned Tabanis and Sacidis), and the existence
of such municipal officers as the ra'fs al-balad (the head of the local community vis-a-vis
the central government), with functions analogous to those of the later Iranian office of
kalantar. In Nishapur, the important family of Mlkalls held this office of ra'fs or zacfm for
much of the later Samanid period and the early Ghaznavid one. 64
Families like this were also involved in the caravan trade which linked Khurasan and
Afghanistan with Transoxania and the steppes on the one hand, and with Baghdad and
Iraq on the other, and whose disruption by the Oghuz invaders predisposed the towns of
Khurasan to come to terms with the Seljuqs. One aspect of this long-distance trade was of
course the traffic in slaves, for Turkish slaves had since the ninth century regularly been
transported across Khurasan from the Transoxanian slave markets en route for Iraq and
the Islamic heartlands; likewise, we know that slaves came directly across the upper Oxus
lands through the intermediacy of the Karakhanids. 65 Finally, the inflow of plunder from
India to Ghazna involved a traffic in Indian slaves and the conveyance of bullion, trophies
of war, and so on. Some of this last, including captured idols and similar spoils, was appar-
ently incorporated into the fabric of the sultan's new buildings in the capital. Other items
required the services of a staff of assayers and valuers in Ghazna in order to turn these into
a negotiable form or into the precious metals required for the mining of the high-quality
gold and silver coinage which was a feature of Ghaznavid monetary practice and which
must have stimulated economic activity within the whole eastern Islamic world. 66

62
Bosworth, 1963, pp. 113, 139, 152-7, 161-2.
63
Ibid., pp. 150-2; 1968b, pp. 133, 135.
64
Bosworth, 1963, pp. 171-5; Bulliet, 1972, pp. 66-8.
65
Bosworth, 1963,p.101; 1970,pp.4-6, 16-17.
66
Ibid., pp. 78-9, 140.

123
ISBN 978-92-3-103467-1 Contents Agriculture and trade

In conclusion, one might observe that the establishment of the Ghaznavid sultanate rep-
resents the first major breakthrough of Turkish power in eastern Islam against the indige-
nous Iranian and other peoples. Although the impressive empire built up by the early rulers
could not be sustained, the Ghaznavids' destruction or weakening oflocal dynasties and of
the landed classes by the imposition of rule by a central bureaucracy in Ghazna did much
to prepare the way for the coming of the Great Seljuqs, the Khwarazm Shahs of Ats'iz's
line or Anushteginids, and so on. The pattern of the despotic power-state introduced by the
Ghaznavids became the norm for many of the subsequent pre-modern Islamic dynasties. 67

67
Bosworth, 1970, pp. 14, 18.

124
ISBN 978-92-3-103467-1 Contents Sources for Karakhanid history

THE KARAKHANIDS*
E. A. Davidovich

Contents

Sources for Karakhanid history . 125

The conquest of Transoxania by the Karakhanids: the division into appanages 126
The two Karakhanid Khanates; the policy of Ibrahim b. Nasr Tamghach Khan 133
Karakhanids, Seljuqs and Kara Khitay . . . . . . . . . . . . . . 138
The Karakhanids and the Khwarazm Shah Muhammad b. Tekish 141
Iqtilcs and the structure of the state 143
Towns and trade . . . . . . . . . . 147

Sources for Karakhanid history

The history of the Karakhanids is one of the least studied periods in the history of Central
Asia and East Turkistan. The most important information is derived from written sources
of a later date. These sources provide a very detailed account of relations between the
Karakhanids and the neighbouring states of the Ghaznavids, the Seljuqs, the Khwarazm
Shahs and the Kara Khitay. Information about events within the Karakhanid state is sparse
and frequently contradictory, and has quite often proved to be completely incorrect.
Barthold conducted a critical analysis of the principal written sources in the 1920s, and his
works on the history of the Karakhanids still provide the fullest and most reliable account.
Other historians have since managed to extract no more than isolated details from the
written sources.

* See Maps 3 and 4.

125
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

The second historical source is Karakhanid coinage. Barthold and other historians real-
ized that Karakhanid coinage represented a major source for the political history, supple-
menting the written sources and correcting their at times contradictory testimonies. They
also understood, however, that Karakhanid coinage represented the most difficult branch
of Islamic numismatics and that its attribution demanded a great deal of preparatory work.
Barthold wrote despondently that it was often impossible to determine even such a simple
point as the number of persons referred to in the inscriptions on coins. 1 None the less,
he made careful use of the Karakhanid coins published in his day, and this enabled him
to discover a great deal of information, although it did not prevent him from making the
occasional mistake.
The content of the written sources and the progress made in the study of the coins did
not permit a reliable list of Karakhanid rulers to be constructed at that time; this is reflected
in the reference books used by all orientalists, such as handbooks for the chronology and
the genealogy of Muslim dynasties. The section on the Karakhanids in Lane-Poole's book,
even with Barthold's additions, 2 is very brief and fragmentary and contains errors. A sim-
ilar section in the work by Zambaur should simply not be consulted. 3 Bosworth based his
section on the Karakhanids in his book on the work of Pritsak. 4 Later numismatists and
historians (Vasmer, Pritsak, Davidovich, Kochnev and others) have published and studied
an enormous number of Karakhanid coins, developing methods for their attribution, sup-
plementing and correcting the information in the written sources for political history and
providing fresh data for the chronology and genealogy not only of dynastic heads but also
of a number of appanage-holders. Karakhanid coinage has also provided information for
the study of a number of questions in the areas of social and economic history.

The conquest of Transoxania by the Karakhanids: the

division into appanages
The confederation of Turkish tribes present in Kashghar and Semirechye in the ninth and
tenth centuries was ruled by a dynasty referred to in the literature as the Ilek Khans or
Karakhanids. 5 The accounts provided in the sources regarding the composition of this con-
federation and the origin of the dynasty itself are contradictory and have given rise to
1
Barthold, 1928, p. 274.
2
Lane-Poole, 1899, pp. 110-12 (there is no genealogical table).
3
Zambaur, 1927, pp. 206-7 (on the errors in the section on the Karakhanids, see Davidovich, 1957,
pp. 115-17).
4
Bosworth, 1965, pp. 111-14.
5 Both designations represent titles. The title Kara Kaghan was the most important Turkish title up till the

end of the dynasty.

126
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

several hypotheses. 6 It is, however, probable that the dynasty came from the Yaghma or
Chigil tribes.
Clashes between the Samanids and the Karakhanids began to occur in the ninth century.
The Samanids even advanced some small distance to the east into the lands of the Turkish
peoples. In 840 they took Isfijab and built walls around it to protect its inhabitants' crops
from the raids of the nomads. Isfijab was not merely a military outpost, however; brisk
trade with the nomads was also conducted there and the town contained many bazaars
and caravanserais. The traders from Bukhara, Samarkand and other large towns of Tran-
soxania constructed separate caravanserais for themselves. Significantly, however, Isfijab
remained a largely independent possession of the local Turkish dynasty, which owed only
three obligations to the Samanids: military service, the presentation of symbolic gifts and
the inscription of the name of the Samanid amir as suzerain on their coinage. The names
of several members of this Turkish family who ruled Isfijab in the tenth century are known
from the legends on coins and from the manuscript sources. At a later date, in 893, the
Samanid Ism~fll b. Ahmad took Taraz, long a possession of the Karakhanids, and con-
verted the Christian church there into a mosque; Taraz was another major trading post for
exchanges with the Turkish nomads.
In the mid-tenth century, the Karakhanids themselves adopted Islam and declared it to
be the religion of their tribal society. They began to take Muslim names and, later, Muslim
honorifics (alqab; pl. of laqab). But the regnal titles conveying the real or formal position
of their holders in the dynastic hierarchy were Turkish: Khan and Kaghan (Kara Khan and
Kara Kaghan), Ilek (Ilig), Tegin, etc. The names of animals were a regular element in the
Turkish titles of the Karakhanids: thus Arslan (lion), Bughra (camel), Toghan (falcon), Bori
(wolf), Toghrul or Toghril (a bird of prey), etc. The Karakhanids later also began to use the
Arab titles sultan and sultan al-salatfn (sultan of sultans). The titles of the members of the
dynasty changed with their changing position, normally upwards, in the dynastic hierarchy.
In the final decade of the tenth century, the Karakhanids began a systematic struggle
against the Samanids for control of Transoxania. As Muslims who had already had contacts
with the Islamic culture of Transoxania, they knew that one of the principal emblems of
power, providing material proof of the control of a town, a region or a state, was the minting
of coins in one's own name. That is why the coins that made their appearance with the
first military successes of the Karakhanids represent a most important indicator for their
political history.
The conquest of Transoxania was initiated by two cousins, c Ali b. Musa (the head of the
dynasty; title: Kara Khan or Arslan Khan) and Hasan b. Sulayman (title: Bughra Khan).
6
Pritsak, 1953, pp. 21-2.

127
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

Table 1. Genealogy of the Karakhanids mentioned in the present chapter

cAl"d~

~_Kita~-- ~~' ) Hmn(H

(J.iHBJm}
,.....,.-.~ .... ----

I_ A:hru:I

H~~
fEfumT~-11)

:l~S-: H~d'$ ~ tl!ir:' K.air..iklt.1tid, y1u! (~fo t dtt-

d~vil~o ir'l~i!Ji ~O ~L - ) rfi'i~llltio.ntdi Oi1i oi mt ~ds.

Pritsak proposed that the families of these two cousins be referred to as the c Alids and the
Hasanids, 7 and this nomenclature is most convenient for a consideration of the subsequent
history of the Karakhanids, as the relations between the two families determined the events
of the time (see Table 1).
The following account of the conquest is provided by the written and numismatic
sources. The first campaign was led by Hasan b. Sulayman Bughra Khan. The Karakhanids
took Isfijab in 380/990, Ferghana in 381/991-2 and Ilaq, Samarkand and the Samanid
7
Pritsak, 1953, p. 26.

128
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

capital, Bukhara, in 382/992. These military successes were celebrated by a political ges-
ture: the minting of coins in the name of Bughra Khan. 8 Having fallen ill in Bukhara,
he travelled to Samarkand and from there set out for Kashghar but died on the way in
382/992. The Samanid ruler returned to Bukhara and took control of the central regions
of Transoxania, whereas the Karakhanids retained a part of the north-eastern and eastern
regions. The initiative then passed to the family of c Ali b. Musa. One of his sons, Nasr
b. c Ali, played a particularly active role; in 386/996 he conquered the region of Chach, in
387/997 Samarkand and in 389/999, having encountered no resistance, he also took the
capital, Bukhara. 9
There were two reasons for the speedy and effortless victory of the Karakhanids. The
first was that the members of the military and bureaucratic structures of the Samanid state,
who wielded a great deal of power, fought among themselves and also, at times, against
the head of the dynasty. The vassal rulers of Khwarazm, Khurasan and Tukharistan had
become virtually independent and took part in the internecine strife, sometimes on the side
of the Samanids and sometimes against them. A new state had thus emerged, the state of
the Ghaznavids. In this complex situation, the Samanid ruler was unable to concentrate all
his forces on the struggle against the Karakhanids. Second, the Karakhanids were Muslims,
and their arrival merely represented a change of rulers at the apex of government at a time
when many people were unhappy with the Samanids and entertained hopes of fresh privi-
leges and advantages under the Karakhanids. Certain leading representatives of the military
and bureaucratic class assisted the Karakhanids, and the dihqans (major landowners) also
took their side. The populace, on the other hand, looked on the change of dynasties with
indifference.
There seems no doubt that the Karakhanids rewarded generously those who had assisted
them. A typical case is that of Begtuzun, who had risen to the highest office under the
Samanids and had settled the fate of the throne on more than one occasion. When the
Karakhanids reached Bukhara in 999, Begtuzun rallied to their cause and, as their vassal,
went on to govern several towns including Kish (coinage 399-402/1008-12) and Khu-
jand (coinage 415/1024-5). 10 The position of the Ilaq dihqans was strengthened under the
Karakhanids. From the year 992 onwards, the Karakhanid coinage for the region refers to

8
On the early coinage of the Karakhanids, see Ishankhanov and Kochnev, 1979. (Since many of the dates
given in this chapter stem from coin legends, the original hijri forms of these dates are given in this chapter
- Editors.)
9
On the unsuccessful struggle of the last Samanid against the Karakhanids, see above, Chapter 4.
°
1
Kochnev, 1982, pp. 152-7.

129
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

the dihqans, the direct rulers of Ilaq, 11 as vassals. That was a privilege they had not enjoyed
under the Samanids.
The Karakhanid state exhibited several prominent features during its first period (until
c. 1040). First, there was the idea of the integrity of the state, which found expression in
the recognition of the head of the dynasty and was reflected in the references to him in
coin legends as suzerain. Second, there was the division of the state into appanages, which
lacked stable borders because of the internecine strife. A third feature was the hierarchical
structure of political power, which was reflected in the differing 'value' of titles. Lastly,
there were the common economic rights - to one and the same appanage sometimes nom-
inal, and sometimes real - of several members of the dynasty.
Four sons of c Ali b. Musa (Ahmad, Nasr, Mansur and Muhammad) held their own
independent appanages within the Karakhanid state; two of them (Ahmad and Mansur)
became, in tum, head of the dynasty after the death of their father in 388/998. The first
to do so was Ahmad b. c Ali (not Nasr b. c Ali, as believed by many historians). Ahmad,
to whom the Muslim written sources make practically no reference, adopted his father's
title (Kara Kaghan) as well as the title Toghan Khan, but Nasir al-Haqq Khan is found
more often on the coinage. His own appanage was located in Semirechye, but he also
held Chach. The chief town of the appanage and the capital of the Karakhanid state at
the time was Balasaghun (thus in Muslim sources), otherwise referred to as Quz Urdu
and Ulush Urdu (Mahmud Kashghari). The dynastic head lived not in the town but in
his nearby army encampment (ordo, urdu); the nomadic traditions and way of life of the
Karakhanids were still very strong at the time. Palaeographic studies provide us with two
possible interpretations of the legends on the coins struck in the capital: Qara Urdu and
Quz Urdu. Pritsak made a special study of the question of Turkish colour symbolism and
concluded that the headquarters of the dynastic head near Balasaghun was indeed called
Qara Urdu. 12 Kashghar at that time, and subsequently, was in the hands of the Hasanids.
From 395/1004-5 Yusuf Kadi'r Khan (the son of Hasan Bughra Khan) regularly struck
coins there in his own name with the title malik al-mashriq (King of the East).
The most influential and the best-known figure was the c Alid Nasr b. c Ali (the con-
queror of Transoxania). Although his titles were modest (initially Tegin and later Ilek), he
held a vast, wealthy and prestigious appanage that comprised the central areas of Tran-
soxania (Samarkand, Bukhara, etc.), Ferghana and also, at certain periods, other areas and
towns. The principal town in his appanage was U zgend in Ferghana. Nasr b. c Ali was in
practice an independent ruler but formally recognized his brother, Ahmad b. c Ali, as head

11
Davidovich, 1978, pp. 80-100.
12
Pritsak, 1955, p. 15.

130
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

of the dynasty. They both appear on most of the coins from Nasr' s appanage as suzerain
and vassal (with the emphasis on Nasr's independence, however). There are instances of
joint economic ownership, including that of the wealthy town and region of Khujand: the
brothers shared the income from this domain. Ahmad and Nasr also transferred control of
individual towns and regions in their enormous appanages to other individuals (some of
whom were not members of the dynasty) as their vassals.
After the death of Nasr b. cAli in 403/1012-13, his appanage was broken up. A large
part fell to the head of the dynasty, but his other brothers were also stirred to action. Mansur
b. cAli, who, according to the coinage, had assumed the august title of Arslan Khan during
Ahmad's lifetime, seized the capital Balasaghun and many other towns. This act evidently
led to a state of war between them, and they were reconciled only as a result of the media-
tion of the Khwarazm Shah Ma'mun. Nevertheless, in defiance of every tradition governing
the hierarchy of titles, the title of Khan was for a certain time borne by three Karakhanids:
the cAlids, Ahmad and Mansur, and the Hasanid, Yusuf b. Hasan.
Among the external political events of the period during which Ahmad b. cAli was
the head of the dynasty, mention should be made of the war against the 'infidel' Turkish
peoples to the north-east and east of the frontiers of the Karakhanid state, and also of
relations with the Ghaznavids to the south-west and the south. Ahmad's father had fallen
in combat against the infidel Turks in 388/998, and Ahmad himself fought against them on
at least two occasions. Following the first clash, he obtained the title of ghazf (fighter for
the faith). 13 At a later date (not before 403/1012-13), the pagan Turks invaded the domains
of the Karakhanids and almost reached the capital Balasaghun, but many ghazfvolunteers,
including some from neighbouring Muslim states, responded to the appeal of Ahmad b.
cAli, who repelled the invaders and gained a brilliant victory, seizing both prisoners and
vast spoils. News of this major campaign spread throughout the Muslim world, hence the
detailed and exaggerated accounts found in the sources (the dates for this campaign vary).
Relations with the Ghaznavids were not stable, and the brothers Ahmad and Nasr con-
ducted independent foreign policies. Nasr and Mahmud of Ghazna at first agreed that the
Oxus (Amu Darya) should be considered the frontier between their two states. However,
Nasr attempted on two occasions to expand his appanage southwards at the expense of
Ghaznavid territory. Ahmad, on the other hand, relied on an alliance with Mahmud of
Ghazna whenever relations with his brother worsened.
After the death of Ahmad b. cAli in 408/1017-18, his brother Mansur b. cAli became
the nominal head of the dynasty, with the title of Arslan Khan; later (after 415/1024-5),
supremacy passed to the Hasanids. It is important to emphasize that there was no

13
On a silver coin from Isfijab 398/1007-8, see Kochnev, 1988, pp. 197-9.

131
ISBN 978-92-3-103467-1 Contents The conquest of Transoxania ...

precise delimitation of territory between the c Alids and the Hasanids during the first period
(regardless of which family representative was head of the dynasty). Some Hasanids held
appanages in Transoxania and declared themselves to be vassals of the c Alids. Another
significant feature of the time was the further development of the hierarchical structure
of power and joint economic ownership, i.e. the sharing of the income from the same
appanage between several members of the dynasty. This process can be closely traced
in the coinage of the Karakhanids. We may consider by way of example the coins of
Akhsik:ath from 409/1018-19 to 415/1024-5. The head of the dynasty at that time was
Mansur b. c Ali Arslan Khan, who is referred to as suzerain on these coins. The ruler of the
town during these years was his brother Muhammad b. c Ali Ilek, who at one time ceded
some of his rights and income to one of his nephews, Ahmad b. Mansur (coinage of the
years 409-10/1018-20), at another time to a second nephew, Muhammad b. Nasr (coinage
of the years 413-15/1022-5), and at a third time to both together (coinage of the year
412/1021-2). The year 415/1024-5 appears to have been most eventful: the two previous
joint owners, Muhammad b. c Ali and his nephew Muhammad b. Nasr, were joined by the
latter's son, c Abbas. The head of the dynasty, the c Alid Mansur b. c Ali Arslan Khan, died
in the same year: the Hasanid Toghan Khan II(= Muhammad b. Hasan) 14 then appears on
the coinage as suzerain, while the c Alids Muhammad b. Nasr and his son c Abbas figure as
joint owners of Akhsikath.
Even before Muhammad b. Hasan Toghan Khan II became the head of the Karakhanid
dynasty, another member of the same family, c Ali Tegin (= c Ali b. Hasan), played an
extremely active role in the central region of Transoxania. The written sources and the
coinage both provide a great deal of information about him. Having been imprisoned by
the head of the dynasty, the c Alid Mansur b. c Ali Arslan Khan, c Ali Tegin escaped from
captivity (not later than 411/1020-1) to seize control of Bukhara and several other towns
and regions. After the death of Mansur b. c Ali, he extended his domains still further. Peace
did not reign among the Hasanids at that time. c Ali Tegin was allied to his brother Muham-
mad Toghan Khan II (the head of the dynasty), in opposition to their brother, Yusuf Kad'ir
Khan (the long-established ruler of Kashghar), who sought to become the head of the
Karakhanids. Yusuf Kad'ir Khan formally achieved his ambition, but central Transoxania
remained in the hands of c Ali Tegin until his death in 426/1034-5 and was then passed on
to his sons.
Mahmud of Ghazna also intervened in the internal strife between the Karakhanids.
According to the sources, complaints about c Ali Tegin from the region's inhabitants

14
The written sources identify Toghan Khan with various Karakhanids. Kochnev, 1979, pp. 125-31, con-
vincingly shows on the basis of coin inscriptions that he was, in fact, Muhammad b. Hasan.

132
ISBN 978-92-3-103467-1 Contents The two Karakhanid Khanates ...

provided the pretext for Mahmud's campaign against Transoxania. Mahmud was him-
self displeased, as cAli Tegin allegedly did not allow his envoys passage to Yusuf Kadi'r
Khan in East Turkistan. The real reason uniting Mahmud and the Karakhanid Kadi'r Khan,
however, was the threat posed by the strengthening position of cAli Tegin. The allies met in
the year 416/1025 to the south of Samarkand, exchanged gifts of great value and decided
to join forces in order to wrest Transoxania fromc Ali's grasp. Mahmud routed his Turkmen
allies, and cAli then abandoned Bukhara and Samarkand: his baggage train was pillaged
en route. Although Mahmud of Ghazna did not consolidate these military successes and
returned to his capital, he had achieved a great deal. Henceforth, none of the Karakhanids
represented a threat; their forces counterbalanced each other, although the balance of forces
in the Karakhanid state continued to be a matter of concern to the Ghaznavids even after
Mahmud's death (see above, Chapter 5).

The two Karakhanid Khanates; the policy of Ibrahim b.

Nasr Tamghach Khan
The formation of two states (Khanates) was a watershed in the history of the Karakhanids.
Barthold notes in passing that Ibrahim b. Nasr established an independent state in Transox-
ania but is extremely circumspect about the date of this event, supposing that Ibrahim could
still have ruled as a vassal in Bukhara in 433/1041-2. 15 Pritsak was the first to give special,
detailed consideration to this question. In essence, his view is that following the death of
the nominal head of the dynasty, Yusuf Kadi'r Khan in 424/1032, two of his sons assumed
the highest titles: Sulayman b. Yusuf in Balasaghun and Kashghar became Arslan Khan,
and Muhammad b. Yusuf in Taraz and Isfijab took the title of Bughra Khan. Two sons
of the cAlid Nasr b. cAli broke away completely from the Hasanids: Muhammad b. Nasr
became the ruler of the whole of Ferghana, with his residence at U zgend and the title of
Arslan Khan, whereas lbraim b. Nasr established himself in the centre of Transoxania. The
process of partition culminated in the year 433/1042-3: an Eastern Khanate was formed
with its capital at Balasaghun (later, Kashghar) and a Western Khanate with its capital at
Uzgend (later Samarkand). The head of each Khanate bore the title Arslan Khan, and the
second-ranking associate Khans the title Bughra Khan. 16
This concept, while orderly in appearance, does not tally with some of the written
sources or with any of the numismatic evidence. Indeed, the independent state of the West-
ern Karakhanids was formed in an entirely different manner. Muhammad b. Nasr was never

15
Barthold, 1928, p. 304.
16
Pritsak, 1950, pp. 227-8; 1953, pp. 37, 44-5.

133
ISBN 978-92-3-103467-1 Contents The two Karakhanid Khanates ...

head of that state, nor did he bear the title Arslan Khan. Moreover, Ferghana did not con-
stitute the core nor U zgend the capital of the Western Khanate. Muhammad b. Nasr (better
known as c Ayn al-Dawla) was always the holder of an appanage and a vassal of either his
close relatives, the c Alids, or the more distant Hasanids. 17 Between the years 411/1020-1
and 447/1055-6 coins bearing his name appeared at different periods in the various towns
of Ferghana, in Khujand and in Ilaq. For example, when the head of the dynasty of all the
Karakhanids was his uncle Mansur b. c Ali Arslan Khan (up to 415/1024-5), Muhammad
b. Nasr controlled Uzgend (the main town of Ferghana) and Akhsikath, recognizing his
two uncles Mansur b. c Ali and Muhammad b. c Ali as suzerains. Even in economic terms
he was not in full possession of these towns, as he was obliged to share the revenue with
other members of the dynasty. In the year 415/1024-5, when the Hasanids became dynastic
heads, Muhammad b. Nasr recognized their suzerainty and retained control of both towns
for a certain time. In 431/1040 a celebrated battle took place between the Ghaznavids and
the Seljuqs near Dandanqan; the Seljuqs were victorious, and Khurasan passed into their
hands (see above, Chapter 5). They then informed the following Karakhanids of their vic-
tory: the two Khans of Turkistan (i.e. the brothers Sulayman b. Yusuf and Muhammad b.
Yusuf), the sons of c Ali Tegin in Transoxania and also the brothers of the c Alid family, Bori
Tegin (i.e. Ibrahim b. Nasr) and c Ayn al-Dawla (i.e. Muhammad b. Nasr). The Karakhanid
state was still formally united; the Hasanid family enjoyed clear political preponderance,
although two sons of Nasr b. c Ali, Muhammad (their vassal) and Ibrahim (who had already
engaged in a struggle against them), were well known to the world beyond and were held
in high esteem.
When the Karakhanid state split into two independent Khanates, Ferghana fell within
the bounds of the Eastern Khanate, i.e. within the sphere of influence of the Hasanids.
Both Hasanids - Sulayman b. Yusuf Arslan Khan and Muhammad b. Yusuf Bughra Khan
- appear as suzerains on the coinage of a number of towns there (Uzgend, Kuba, Marghi-
nan and Akhsikath) in the year 440/1048-9. The northern part of Ferghana (Akhsikath)
belonged to the latter and the south-east (Uzgend, Kuba, etc.) to the former, although these
boundaries were not rigid. The Hasanids also had their vassals in Ferghana. Muhammad b.
Nasr c Ayn al-Dawla was one of the vassals of Sulayman b. Yusuf Arslan Khan and held
some of the towns there in appanage (it is not known whether at certain times he controlled
such towns as Kuba and Marghinan). Muhammad b. Nasr remained an appanage-holder, a
vassal and nothing more, until the end of his life.
The political career of the other son of Nasr b. c Ali, Ibrahim b. Nasr (the Bori Tegin
of the written sources), was very different. Bori Tegin was for some time a prisoner of the

17
Davidovich, 1968, pp. 69-74.

134
ISBN 978-92-3-103467-1 Contents The two Karakhanid Khanates ...

Hasanids (the sons of c Ali Tegin in Transoxania). Escaping to join his brother in Uzgend,
he then moved south to the mountains, where he assembled an army, intending to win back
some regions from the Ghaznavids. He first laid waste the areas of Khuttal and Wakhsh
(in modern southern Tajikistan) and then took control of the area of Chaghaniyan (in mod-
ern southern Uzbekistan). Coins were minted in his name in Chaghaniyan from 430/1038-9.
It was from there that Bori Tegin began the battle for Transoxania against the sons of c Ali
Tegin. He gained several victories over them at the beginning of 431/1039, and by the
following year had seized a considerable part of central Transoxania. He marked his mil-
itary successes by a political act: in 431/1039-40 (coinage of Chaghaniyan) he replaced
his modest title Bori Tegin with the title Tamghach Bughra Khan (Kaghan). 18 Ibrahim
Tamghach Khan immediately made Samarkand his capital. Such were the origins of the
independent state, the Western Khanate: all the initiative was taken not by Muhammad b.
Nasr but by his brother, Ibrahim b. Nasr. The division of the Karakhanid state also estab-
lished the demarcation line between the two families, the c Alids and the Hasanids, ruling
respectively the Western and Eastern Khanates. The border between the two Khanates
changed repeatedly. The bone of contention was Ferghana, with its wealthy towns, mineral
resources and fertile land. Several areas along the course of the Jaxartes ( Syr Darya) also
changed hands.
Ibrahim Tamghach Khan no longer conducted an active foreign policy after forming an
independent state. However, he successfully exploited the civil strife among the Eastern
Karakhanids, the struggle between the Hasanid brothers, Sulayman and Muhammad. No
later than the year 451/1059-60 Ibrahim attached Ferghana to his domain. The conquest
of the region was duly marked by the striking of coins in his name at a number of towns
(such as Uzgend, Akhsikath and Marghinan).
Muslim historians considered Ibrahim Tamghach Khan to be a great and devout sov-
ereign. His domestic policy does indeed reveal him to have been a quite exceptional ruler.
Barthold discovered in the written sources some amusing stories about his life and deeds, 19
each of which has a rational core. These stories, together with the indirect evidence pro-
vided by the coinage of the time, show that Ibrahim did indeed concern himself with inter-
nal order in the country, the security of the population, respect for property, trade and
currency circulation. According to one tale, some robbers once wrote on the gate of the
citadel of Samarkand, 'We are like an onion, the more we are cut the bigger we grow.' The
Khan ordered to be written underneath their words, 'I stand here like a gardener; however

18
Davidovich, 1970, pp. 88-94.
19
Barthold, 1928, pp. 311-13.

135
ISBN 978-92-3-103467-1 Contents The two Karakhanid Khanates ...

much you grow I will uproot you.' 20 It is clear from another story that he did indeed instil
terror among thieves and robbers and that the ordinary people in his dominions felt safe. It
may be concluded from indirect evidence that state control of market prices existed during
lbrahim's reign. On one occasion, the butchers complained to him that the statutory price
of meat was too low and asked him to raise it, promising to contribute 1,000 dinars to the
treasury. The Khan agreed but forbade the people to buy any meat. The butchers were then
obliged to pay more money into the treasury in order to have the old, low price of meat
restored.
Normal trading always depended on the organization of currency circulation. Ibrahim
took responsibility for this aspect of economic life. During his rule, a single system of
coinage with different denominations circulated throughout the Western Karakhanid
Khanate, creating good, stable market conditions. The dirhams struck with the name and
title oflbrahim Tamghach Khan were known as mu'ayyadf. They were made of low-grade
silver, but the addition of copper was not a fraud carried out in secret. The population
knew the official standard of purity of the mu' ayyadf dirhams; their value, which tallied
with that standard, fluctuated slightly and was fixed in terms of pure gold. Greater purchas-
ing power was attached to the Bukhar Khudat dirhams, which were struck on the model
of the Sasanian coinage (see below, Chapter 20). By the beginning of the ninth century
these dirhams were divided into three groups, each with its own name, on the basis of the
quantity of silver they contained: the highest-grade coins (with over 70 per cent silver)
were known as musayyabf, the lower-grade (over 40 per cent silver) as muhammadf and
the copper coins with no silver content as ghitrlfi. Muhammadf and ghitrifl dirhams were
still in circulation under the Karakhanids. The rate for the copper ghitriji was equal to or
higher than that for pure silver, but the rate for the muhammadf was higher than that for
the ghitrlfi. This phenomenon developed in the ninth and tenth centuries and continued
under the Karakhanids, which clearly shows that they pursued the same financial and fiscal
policies as had been applied by the Arab governors and Samanids in the previous period.
In this context, it is important to note that the three types of dirham provided a satisfactory
basis for the different levels of internal trade under Ibrahim Tamghach Khan (see below,
Chapter 20).
An important component of lbrahim's financial and fiscal policy was the currency
reform that he introduced after conquering Ferghana, previously a part of the Eastern
Karakhanid Khanate. Copper-lead dirhams were issued in the towns of Ferghana and
in several other towns of the Eastern Khanate. These were coins of irregular shape and
differing weight, with serrated edges. Ibrahim banned their circulation in Ferghana and
20
Ibid., p. 312.

136
ISBN 978-92-3-103467-1 Contents The two Karakhanid Khanates ...

Mu' ayyadi dirhams began to be issued in various towns of the region. This led to very
different consequences in the two Khanates. Ferghana was incorporated into the currency
area of the Western Khanate, whereas the bulk of the banned copper-lead dirhams flowed
into the Eastern Khanate (mostly to the territory of modem Kyrgyzstan and southern
Kazakhstan). However, the level of trade there did not require such large quantities of
copper-lead dirhams and the resulting inflation led to a severe currency crisis. People no
longer wanted the devalued coins and they ended up in deposits.
Ibrahim waged a successful struggle against the appanage system, which had been the
cause of endless fratricidal strife, and the reassignment of towns and regions. He did not
set up a centralized state, but managed to reduce considerably the number of appanages
and the rights of appanage-holders. This was a great political triumph and one of the most
important factors contributing to the stability of the Western Karakhanid Khanate under
Ibrahim Tamghach Khan.
The economic successes were of even greater importance, however. We may assume
that substantial sums flowed into the coffers of the central government. This was one of the
factors underpinning the considerable building activity that took place. Both Ibrahim and
his son Nasr Shams al-Mulk engaged in major building projects. Ibrahim built a hospital
in Samarkand where he not only cared for the sick but also gave shelter to the poor. He
provided the hospital with funds for the maintenance of the doctors and auxiliary staff,
for the patients' meals, for light, for firewood for the kitchen and for the repair of the
premises. In Samarkand he also built a madrasa (Islamic school), providing the wages of
the teachers, grants for the students, books for the library, the lighting of the premises,
etc. 21 For the benefit of the caravan trade, his son built ribats (caravanserais) in the steppes
between Bukhara and Samarkand (Ribat-i Malik, 'the ribat of the king') and on the road
from Samarkand to Khujand. Nasr Shams al-Mulk also restored the mosque and minaret
in Bukhara and, close to the town, built a beautiful palace at a place that was thereafter
known as Shamsabad. Muhammad b. Nasr (lbrahim's brother) and his son, c Abbas, built
a mausoleum in Ferghana (now known as the Shah Fadila mausoleum), the interior of
which is decorated with elaborate alabaster carving and inscriptions recording the names
of both men. 22 At this period the Karakhanids still maintained their nomadic traditions,
but the extent and diversity of the civil and religious buildings constructed testify to the
fact that the culture and traditions of the settled population of Transoxania had been more
extensively and profoundly assimilated.

21
Khadr, 1967. One of the waqfs is dated Rajab 458/May-June 1066.
22
Nastich and Kochnev, 1988, pp. 68-77.

137
ISBN 978-92-3-103467-1 Contents Karakhanids, Seljuqs and Kara Khitay

Karakhanids, Seljuqs and Kara Khitay

It was clear even in the days of Ibrahim Tamghach Khan and his son, Nasr Shams al-Mulk,
that the Seljuqs had designs on Transoxania, but they achieved no decisive successes. On
the contrary, Nasr took possession of the Seljuq towns in northern Khurasan, albeit for just
a short time. But during the reign of lbrahim's grandson Ahmad b. Khidr (d. 1095), the
Seljuqs took Samarkand with the support of the town's religious classes (friction had long
existed between them and the Karakhanids), together with other domains belonging to the
Western Khanate. Even the Karakhanids of Kashghar declared their submission. For half
a century the Karakhanids became vassals of the Seljuqs. The vassal status of the Eastern
Karakhanids was of short duration. Moreover, at the beginning of the twelfth century they
invaded Transoxania, advancing into the domains of the Seljuqs themselves; for a short
time they even held the Seljuq town of Termez on the Oxus. The Western Karakhanids
were more dependent on the Seljuqs, although nothing is known of the financial aspect
of their dependence. (Did they pay tribute?) Their political dependence was considerable,
however: the Seljuqs placed on the Karakhanid throne in Samarkand whichever members
of the dynasty they required. The vassal status of the Western Karakhanids is also reflected
in the coinage, some of which bears the names of Seljuq sultans.
The following internal events are also worthy of note. The Seljuqs soon restored to the
Karakhanid Ahmad b. Khidr the throne that they had wrested from him. The long-standing
strife between the Karakhanid rulers and the cuzama' (scholars learned in the Islamic legal
and theological sciences) grew more intense during Ahmad's reign, however, and the clergy
gained the upper hand in the struggle. They accused Ahmad of heresy and in 1095 secured
his execution. Barthold correctly emphasized that 'this event must be regarded as the great-
est of the successes gained by the priesthood in alliance with the military classes over the
government and the mass of citizens' .23
In 1102 the Seljuqs placed Muhammad Arslan Khan, the great-grandson of Ibrahim
Tamghach Khan, on the throne of Samarkand and helped him to overcome another claimant.
The situation in Transoxania remained peaceful and stable (until 1130) during Muham-
mad's reign, and he himself became renowned for his building activities. The scale of civic
construction was considerable. Muhammad built a ribat, restored the walls of the Bukhara
citadel and constructed a new outer wall surrounding the entire town; he also raised three
new palaces, one in the citadel and two in the town. Having managed to maintain smooth
relations with the cuzama', he erected many religious buildings. Thus on the outskirts of
Bukhara (on the site of the palace at Shamsabad, which had by that time been destroyed),

23
Barthold, 1928, p. 318.

138
ISBN 978-92-3-103467-1 Contents Karakhanids, Seljuqs and Kara Khitay

he laid out an area for ceremonial acts of worship (namazgah); within the town he built a
magnificent new mosque, the minaret of which still stands, and also repaired an old one.
At the end of his life, Muhammad was struck down by paralysis. Conflicts with the
religious classes broke out once more and the Seljuqs again interfered in the affairs of the
Karakhanids. The renowned Seljuq Sultan Sanjar took Samarkand in 1130 and began to
dispose of the throne in a despotic fashion, replacing one Karakhanid with another.
It was at that time that the Kara Khitay (referred to as such by the Muslim sources) made
their appearance on the political scene. The Khitay had established an enormous empire as
early as the end of the tenth century, the residence of the dynasty being in northern China.
They were driven out by the nomadic Jtirchen, and surviving elements made their way
westwards. The Khitay first took over the domains of the Eastern Karakhanids and in 1137
defeated the Western Karakhanids near Khujand. They did not move into Transoxania in
the same year, however. The chief of the Kara Khitay bore the title of Gtir Khan and his
capital was located not far from Balasaghun.
In 1141 the Qatwan steppe to the north of Samarkand was the scene of a decisive battle
between the Kara Khitay and the Seljuqs in which the latter were utterly defeated. Sultan
Sanjar and the ruler of the Karakhanids both fled to Khurasan and the Kara Khitay took
control of Transoxania. They did not, however, destroy the dynasty of the Karakhanids.
We do not find the names of Kara Khitay Gtir Khans on Karakhanid coins, which indicates
that they were not interested in what was for the Muslims an important mark of political
supremacy. The financial aspect of the conquest found expression in the tribute that the
Kara Khitay exacted from the Karakhanids. The Kara Khitay did not remain in Transoxa-
nia, however, and did not themselves collect the tribute; that task was performed on their
behalf by Karakhanid officials.
The dominion of the Kara Khitay did not bring peace and tranquillity to Transoxania.
Various nomadic Turkic tribes and federations living within its bounds ( Kaduk, Turk-
mens, etc.) were highly active at the time. The Kaduk killed Ibrahim III Tamghach Khan
(536-51/1141-56), head of the Western Karakhanids, and left his body on the steppe. The
next head of the Western Karakhanids took vengeance on the Kaduk, killing their leader.
The Kara Khitay Gtir Khan demanded that all the Kaduk move to Kashghar and take up
agriculture; he clearly hoped to assist the vassal Karakhanids in this way and establish order
in Transoxania. The result was quite the opposite, however. A new war against the Kaduk
ended with the victory of cAli b. Hasan, but the victory did not prove decisive. Mascud b.
Hasan (the brother of cAli b. Hasan) mounted the throne after a long and exhausting strug-
gle of perhaps two years' duration against the nomadic pagan Turks, a struggle that was to
have ruinous consequences for Transoxania. Mascud won a decisive victory in 556/1160-1

139
ISBN 978-92-3-103467-1 Contents Karakhanids, Seljuqs and Kara Khitay

on the steppe near the famous Ribat-i Malik, which has survived to this day, where trav-
ellers and caravans carrying merchandise stopped on the 'shah's road' between Bukhara
and Samarkand. It was shortly after Mascud's accession to the throne that Muhammad
al-Katib al-Samarqandi presented him with his Sindbad-nama [Book of Sindbad]: in the
introduction and the final part of this work, Samarqandi extravagantly praises the ruler for
his victory in the fierce struggle against the nomads.
Rukn al-Dunya wa 'l-Din Kilich Mascud Tamghach Khan was a well-known figure
mentioned in many of the written sources. Al-Samarqandi dedicated two celebrated works
to him, the above-mentioned Sindbad-nama and the later Acrad al-siyasa, and the poet
Suzan! Samarqandi a number of qasfdas (odes). In 560/1164-5 Mascud restored the walls
of Bukhara, which had been destroyed. He also conducted a successful campaign against
the Kaduk to the south (in Nakhshab, Kish, Chaghaniyan and Termez) and established
order there. He suppressed a rising by one of his commanders and was successful in his
operations against the Oghuz, who had plundered Khurasan. The date at which his rule
came to an end (566/1170-1) 24 and the names of his successors in Samarkand have been
determined from coins.
Two developments affected the state of the Western Khanate in the second half of the
twelfth century and the beginning of the thirteenth. First, lasting control of the throne
of Samarkand passed to the Ferghana branch of the Karakhanids. The above-mentioned
brothers, cAH and Mascud, were members of this family, as were all subsequent rulers until
the end of the dynasty. The second change was that the region of Ferghana itself became
formally as well as de facto independent. The rulers of Ferghana, who were members of
the same family, struck coins in their own name in Uzgend, the chief town of the region,
bearing no reference to the Karakhanids of Samarkand as their suzerains.
The idea of the unity of the Western Khanate was expressed through the prestige of the
titles, those of the rulers of Samarkand being slightly more august than those of the rulers
of Ferghana. Thus, at the time when Samarkand was ruled by Mascud b. Hasan and his
son Muhammad (under their title of Tamghach Khan), Ferghana was held by their close
relatives, the brothers Mahmud b. Husayn and Ibrahim b. Husayn. Ibrahim issued coins
annually from 559/1163-4 to 574/1178-9 in his own name under the title Arslan Khan.
At that time, the title Arslan Khan (Kaghan) had a somewhat lower status than the title
Tamghach Khan in the Western Khanate. By 574/1178-9 Ibrahim b. Husayn was ruling
in Samarkand and the title ulugh sultan al-salatfn (great sultan of sultans) appeared on

24
Two manuscripts of Jamal QarshI erroneously give the year A.H. 560. The description of the Karakhanids
contained in these manuscripts also contains many other errors, which have been corrected with information
drawn from the legends on coins (see Davidovich, 1977, pp. 179-83).

140
ISBN 978-92-3-103467-1 Contents Karakhanids and the Khwarazm Shah ...

his coinage. This title, together with the throne, was passed on to his son cuthman at the
beginning of the thirteenth century. His other son, Kad'ir b. Ibrahim, had his residence
in U zgend and his title was lower in status than that of his father and brother. The last
Karakhanid in Ferghana was another member of the Ferghana family, Mahmud b. Ahmad.

The Karakhanids and the Khwarazm Shah

Muhammad b. Tekish
cAla' al-Din Muhammad b. Tekish (1200-20), the penultimate member of the Anushte-
ginids (the third dynasty of Khwarazm Shahs), consolidated and extended his father's
military successes in the south. His main task was then to defeat the Kara Khitay. The
Anushteginids were vassals of the Kara Khitay (like the Karakhanids) and paid tribute to
them. Muhammad initially wished to have the Karakhanids as his allies in the fight against
the common enemy, the 'infidel' Kara Khitay. A threat from a different quarter also awaited
the Kara Khitay. To the east there appeared the nomadic Mongol tribe of the Naiman, 25 led
by the warlike Ktichltig. The generalized conflict that brought about the downfall of both
the Karakhanids and the Kara Khitay was preceded by the following events and upheavals.
Bukhara was in practice controlled by the sadrs (leaders of the Muslim religious classes)
of the Burhan family. They were extremely rich and used their power to oppress the peo-
ple. In 1207 a certain Sanjar, the son of a seller of shields, led a revolt of the citizenry
against the sadrs and seized control of the town. This provided the pretext for a campaign
against Bukhara; cAla' al-Din Muhammad b. Tekish took the town and quelled the revolt,
but also destroyed the power of the sadrs. He did not have Sanjar executed but instead sent
him to Khwarazm. Muhammad ordered the citadel and the walls of Bukhara to be rebuilt
and he himself returned to his domains. Subsequent relations with the Karakhanids 26 were
to a considerable extent influenced by the cunning policy of cuthman, who manoeuvred
between the Kara Khitay and Muhammad b. Tekish. Initially, he took the side of the lat-
ter, and relations between them were those of allies: cuthman even held a higher title
(ulugh sultan al-salatfn) than did Muhammad (sultan). However, cuthman soon sided
with the Kara Khitay and sought the hand in marriage of the Gtir Khan's daughter. On
being refused, he reverted to his alliance with Muhammad, but this time as the latter's
vassal. In 606/1209-10 coins were struck to mark the new relationship between the two

25
Ktichltig the Naiman fled into the lands of the Kara Khitay after the defeat inflicted on him in 1208 by
Chinggis Khan.
26
The accounts provided by the written sources about events at this time are contradictory: the dates vary
and are unreliable. The legends on coins have provided the basis for major amendments and a reconstruction
of the major changes (see Davidovich, 1957, pp. 113-14; 1994).

141
ISBN 978-92-3-103467-1 Contents Karakhanids and the Khwarazm Shah ...

rulers; the names of both appear on the coins, that of Muhammad as suzerain and that of
cuthman as his vassal. cuthman was obliged to curb his ambitions: both rulers are given
the same title, sultan, on these coins. The Kara Khitay Gtir Khan, responding to cuthman's
'treason', seized control of Samarkand but spared the town and prevented it from being
pillaged. However, he was soon compelled to abandon Samarkand and return to his main
possessions, where the Naiman Ktichltig had achieved great successes, even seizing the
Gtir Khan's treasury, which was stored in Uzgend.
The Gtir Khan's reversals inspired cuthman to acknowledge cAla' al-Din Muham-
mad b. Tekish yet again as his suzerain, and this acknowledgement was duly marked in
607/1210-11 by new coins in both their names that were struck in Samarkand. Many major
events occurred in the course of that year. The Gtir Khan won a victory against Ktichltig,
but the other half of his army, under his commander Tayanku, was defeated by Muham-
mad b. Tekish, and Tayanku was taken prisoner. cAla' al-Din Muhammad's victory was
the cause of great rejoicing among the Muslims: henceforth, the ruler was referred to in
documents and on the coinage as 'The Second Iskandar' (i.e. the second Alexander) and
the equal of Sultan Sanjar.
cAla' al-Din Muhammad dispatched envoys to all the leading Karakhanids demanding
their submission. cuthman had already acknowledged Muhammad as his suzerain and the
others now followed suit. Thus Kadi'r b. Ibrahim, ruler of Ferghana and second ruler in
status among the Karakhanids, acknowledged his vassal status in the same way as his
brother cuthman, that is to say, coins bearing Muhammad's name as well as his own were
struck in Uzgend. Kadi'r curbed his ambitions to an even greater extent than cuthman; his
title on the coins is lower than that of Muhammad. The minor ruler who held Utrar was
dilatory in declaring his submission and was therefore stripped of his domains and sent to
Nasa. Muhammad then incorporated Utrar into his dominions and struck coins in his own
name in the town in 607/1210-11. This was the first indication of a change in Muhammad's
policy towards the Karakhanids.
Shortly afterwards, cuthman married the daughter of cAla' al-Din Muhammad b. Tekish
and went to live in Khwarazm for a considerable period. He then returned to Samarkand in
the company of a Khwarazmian plenipotentiary. Clearly, Muhammad did not trust cuthman
- and rightly so, for the latter was not content with his vassal status or the fact that he had
to submit to the plenipotentiary. The population of Samarkand was also unhappy with the
behaviour of the Khwarazmians. cuthman therefore once more took sides with the Kara
Khitay, and the inhabitants of Samarkand massacred all the Khwarazmians there in the
most barbaric fashion. In 1212 Muhammad b. Tekish captured Samarkand and on this occa-
sion showed no mercy. In 609/1212 he ordered the execution of the Karakhanid cuthman

142
ISBN 978-92-3-103467-1 Contents lqtacs and the structure of the state

and again sent envoys to the Karakhanids of Ferghana demanding their submission. The
last ruler of Ferghana, Mahmud b. Ahmad, postponed the end of the dynasty by accepting
vassal status and confirming it by the issue of coins in two names in 609/1212-13. This
was no more than a short-term compromise, however. On the coins of Uzgend struck in
610/1213-14 we find only the name of Muhammad b. Tekish. Starting in the same year,
coins were also regularly struck in his name in Samarkand, the capital of the new Anushte-
ginid empire, and then in other towns formerly held by the Karakhanids.
In conclusion, we may say that cAla al-Din Muhammad b. Tekish did not initially intend
to destroy the Karakhanid dynasty but merely sought allies in his struggle with the Kara
Khitay. He considered it normal that the title of the Karakhanid cuthman should be higher
than his own and laid no claim to any of the insignia of power in the Karakhanid state.
Subsequently, however, the Karakhanids were obliged to acknowledge themselves as vas-
sals of Muhammad b. Tekish: both cuthman in Samarkand and Kad'ir in U zgend confirmed
their vassal status by striking coins in two names and adopting titles with an inferior status.
In the third and final act, the Karakhanids gradually surrendered their domains - and, in
many cases, their lives - to Muhammad b. Tekish. The eastern branch thus disappeared as
a result of the struggle between the Kara Khitay, the Naiman Ktichltig and their internal
enemies in Kashghar.

Jqtacs and the structure of the state

Historians have for long drawn a contrast between the structures of the Karakhanid and
Samanid states and the significance of the institution of the iqtif (assignment of revenue
from an estate) in those states. They considered that the Samanid state was centralized
and that iqtt{s had not developed there, whereas iqtt{s and the system of appanages were
dominant under the Karakhanids.
It is known that the Samanids allocated half their budget for the wages of their troops
and officials, a fact that attracted the scholars' attention when they came to evaluate the
significance of the iqtt{; there is also direct evidence indicating that the Samanids did not
grant iqtifs. The existence on the borders of the state of hereditary vassal domains that were
virtually independent did not contradict this assessment, since such domains could not be
classed as iqtt{. However, the situation in the main part of the Samanid state remained
outside the consideration of the historians. Narshakhi, writing in his Tarzkh-i Bukhara
[History of Bukhara] (mid-tenth century), does not employ the term iqtac, although an
analysis of the information provided about the relationship between the Samanid Nasr
I (the head of the dynasty) and his brother Isma en establishes beyond any doubt that

143
ISBN 978-92-3-103467-1 Contents lqtacs and the structure of the state

Bukhara was held by the latter as an iqtif, that is to say, as a conditional reward for services
rendered in the capacity of governor with the right to levy for his own benefit a part of the
income of Bukhara and, later, the entire income from the town. 27 It is also clear from the
legends on Samanid coins that Bukhara, Akhsikath, Kuba, Nasrabad and other towns and
regions were held as iqtif s for various periods of time by members of the dynasty and
by senior military and civilian officials as rewards for their services. These grants were
neither lifelong nor hereditary, although attempts were made to move in that direction
and were resisted by the central government. 28 Given the presence of such domains and
appanages within the main body of the Samanid state, we cannot consider it centralized;
the appanage system was already developed in the tenth century, which means that it was
not simply introduced by the Karakhanids. The institution of iqtif was also quite well
developed under the Samanids, but the grants were always large ones, given in return for
service as the governor of a town or region, and so on. Middle-and lower-ranking members
of the military and official class and simple soldiers received fixed payments in cash from
the treasury. Such were the characteristics of the state structure and the institution of the
iqttf under the Samanids.
The conclusion that the iqttf was dominant under the Karakhanids did not rest on facts
but solely on an analogy with the Seljuq state. But an analogy does not in itself consti-
tute proof, and it automatically overlooks the varying ways in which the same institutions
develop in different states. Variations certainly did occur; convincing proof is furnished
by a comparison between the iqtif in the Seljuq and Ghaznavid states. Under the Seljuqs
there were large and small iqtilc s that were granted to members of the dynasty, to various
members of the military and official class and to ordinary soldiers. The Ghaznavid Sebtik-
tegin (977-97) (see above, Chapter 5) seized land allotments from his forces and reformed
the system of rewards for service: 'the central power in Ghazna was now strong enough to
resume the fiefs and substitute a system of cash payments. In general, his successors for
at least the next two or three generations maintained the system of paying the troops in
cash.' 29 The Ghaznavids also avoided distributing large iqtt{s. Nizam al-Mulk provides an
interesting account concerning the governor of Khwarazm, Altuntash, whose salary (paid
from the treasury of Mahmud of Ghazna) represented twice as much as the entire rev-
enue from Khwarazm. Altuntash wished to keep the taxes collected there in settlement of
half his salary, but the wazfr (vizier) called him to order: the taxes belonged to the ruler
and not to his officials; they should therefore be handed over to the treasury in return for
27
Davidovich, 1954, pp. 71-7.
28
Davidovich, 1954, pp. 77-117. Later publications describe further examples of conditional rewards for
services consistent with the existence of iqta's.
29
Bosworth, 1973, p. 125.

144
ISBN 978-92-3-103467-1 Contents lqtacs and the structure of the state

a receipt, and payment should then be requested for services rendered. 3°Contemporaries
were clearly aware of the danger that the system of the iqtzf posed for the integrity of the
state and its economy. This awareness and the indisputable existence of the two variants
mentioned above (Seljuq and Ghaznavid) prove that the significance of the iqtiic in the life
of society and the state was determined not only by the objective features of development
but also by a whole set of specific causes particular to each state (the state of the economy
and the treasury, the political situation and financial policy, and so on).
Consequently, the question of the iqtiic under the Karakhanids requires specific histor-
ical research based on local sources; the inscriptions on coins are the one source of this
type available at present. During the first period, up to the division of the Karakhanid state
into two Khanates, the appanage system exhibited the following features. 31 The head of
the dynasty had his own appanage, and major appanage-holders recognized the head of the
dynasty as their suzerain (the political dimension); in most cases, the latter also appeared
on their coins. The head of the dynasty enjoyed no other rights within their appanages. But
even these principal and major appanages had no fixed and stable borders; nor were they
hereditary. The head of the dynasty and the major appanage-holders transmitted some of
their rights to their vassals and sub-vassals; this gave rise to a complex, multi-tiered system
of joint economic ownership based on vassalage (the economic dimension), in which the
joint owners (usually two or three but sometimes four) divided up in specific proportions
the entire revenue from the town (or region) or items of such revenue. The lower level of
the hierarchy was occupied, and joint economic ownership held, not only by members of
the dynasty but also by longstanding local owners from the period before the Karakhanids
and by members of the military and official class who had entered their service. A trend
towards the consolidation of inheritances was none the less observable in certain regions
and towns, although a similar trend also developed under the Samanids. In other words,
the appanage system during the first period of Karakhanid rule differed from that under
the Samanids not so much in quality as in quantity, i.e. by its all-inclusiveness. This was
the result of a development process that was accelerated by the nomadic Karakhanids. The
financial nature of appanages and joint ownership under the Karakhanids during this first
period corresponds essentially to the institution that received the designation iqtiic.
Several changes occurred in the structure of the state during the second period, after the
division into two separate Khanates. Chief among these were a reduction in the number

30
Nizam al-Mulk, 1978, pp. 230-1.
31
Barthold somewhat simplifies the characteristics of the appanage system under the Karakhanids, sup-
posing that the state 'was divided into a number of appanages, the large ones being in turn subdivided into
many small ones' (Barthold, 1928, p. 268).

145
ISBN 978-92-3-103467-1 Contents lqtacs and the structure of the state

of appanages and the political rights of the appanage-holders, and attempts to consolidate
the central authority, to expand its powers and to establish a monopoly over the coinage.
Unfortunately, the economic dimension of the changes during this second period cannot be
deduced from the numismatic evidence.
The third stage, however - that starting particularly from the second half of the twelfth
century - witnessed the consolidation of the appanage system and a considerable increase
in the rights of holders. Appanages on the borders of the Western Khanate became hered-
itary and independent even in political terms. The Karakhanids who held Ferghana and
other lesser appanages such as Marghinan, Kasan, Binakat and Utrar issued on a regular
or, in the case of the minor appanages, an occasional basis, coins that make no reference to
their suzerains. The multi-tiered system of a hierarchy of dependants and joint economic
ownership, so typical of the first period, had ceased to exist.
At this third period, there were also domains that did not belong to members of the
dynasty. Thus Bukhara was held on a hereditary basis by members of a clerical line, the
Al-i Burhan, upon whom was conferred the title of sadr-i jahan (Pillar of the World) and
the office of ra'fs (headman) of Bukhara. They themselves collected the taxes, and the
Kara Khitay sent a special envoy to receive the town's tribute. The local rulers did not
issue coins in their own names (we know only of Karakhanid coins in Bukhara during this
time), but were otherwise independent. It is also impossible to refer to hereditary domains
of this type as iqtacs.
Even this scant evidence is sufficient to show that we would be committing errors of
methodology and of fact if we evaluated the institution of the iqtac under the Karakhanids
on the basis of Seljuq evidence. The inscriptions on the coins cannot tell us whether, under
the Karakhanids, small grants were made to ordinary soldiers and to minor and middle-
ranking members of the army and the civilian bureaucracy; consequently, there are simply
no data available for the purposes of comparison with the Seljuq system. On the other
hand, we may confidently conclude that there are no similarities between the Ghaznavid
and Karakhanid systems during the first period. We can also deduce from inscriptions on
the coins that the system of rewards and ownership had developed and acquired features
'in the upper echelons of power' that clearly demonstrate the inappropriateness of apply-
ing the term iqtac to it. The system of appanages in the Karakhanid state also underwent
considerable change; the process was not identical in the Western and Eastern Khanates,
however, being less developed in the latter.
In terms of the form of ownership, any holding or joint holding given in reward for
services was state property (ground-rent/-tax, urban taxes, etc.). It is essential to note
that members of the dynasty who received an income from state property on this basis

146
ISBN 978-92-3-103467-1 Contents Towns and trade

were fully aware of their precarious position. They therefore attempted to purchase land
and all kinds of income-generating premises as milk (private property). The purchase of
milk was registered in the offices of the qadf Uudge) through the issue of a wathfqa (legal
deed) and was a secure form of property protected by the law. In 1128 Qubawi completed
the translation from Arabic into Persian of Narshakhl's Tarfkh-i Bukhara, adding some
details of life in his own day, including examples of such private property acquired by the
Karakhanids. The owner of milk enjoyed full rights to dispose of his property as he wished:
by sale, donation, bequest to his heirs or assignment to a waqf (endowment for pious pur-
poses). The Karakhanids were great builders, and all buildings were maintained by means
of such endowments, so that the volume of waqf property expanded considerably during
the eleventh and twelfth centuries. The Karakhanids also endowed waqfs for the benefit of
their descendants, the most effective means of preserving property amid all the political
upheavals of the time.

Towns and trade

The eleventh and twelfth centuries were a period during which towns continued to grow
and crafts and trade continued to expand. Three aspects of the development of urban life
at the time are worthy of attention. The physical expansion of many towns, limited by the
walls of their rabads (suburbs), had come to a halt in the tenth century. In these towns
the density of the urban fabric increased under the Karakhanids, and a growing number of
premises were concerned with trade and craftwork, especially on the outskirts. The areas
of other towns did expand in the eleventh and twelfth centuries, especially those towns on
the borders of the state, on the frontier with the nomadic world.
Lastly, the relative importance of the older towns was altered in certain sub-regions.
Thus of the many towns, large and small, in Ferghana in the tenth century under the
Samanids, Akhsikath (then the region's capital) was considered the most important, fol-
lowed by Kuba and Ush; Uzgend was two-thirds the size of Ush. Under the Karakhanids,
Uzgend became the principal town and capital of both the region and the appanage; Marghi-
nan and Kasan also gained in importance. In the regions of Chach and Ilaq, the valleys of
the rivers Chirchik and Angren, an urban explosion took place under the Samanids; there
was no other region of Transoxania where so many towns were to be found in such a small
area. The main towns in Chach and Ilaq were respectively Binkath and Tunkath. Under the
Karakhanids, they were gradually overshadowed by Binakat and Naukat, but did not lose
their importance: tenth-to twelfth-century ceramics have been found at all the excavated
archaeological sites. On the middle course of the Syr Darya, new towns sprang up during

147
ISBN 978-92-3-103467-1 Contents Towns and trade

the Karakhanid period, and old ones expanded. In the latter category we find Isfijab, the
possession of a local Turkish dynasty that had managed to hold on to its position under
the Karakhanids, and also Barab-Utrar, which, in the twelfth century, was an independent
appanage held by the Karakhanids of Ferghana.
Every town had its bazaars and caravanserai. The craftsmen's booths were not to be
found only in the bazaars, however, but also built against the walls of the caravanserai
or simply set up in the street. The ribats along the roads between towns and the cara-
vanserais inside the towns built under the Karakhanids testify to the lively caravan trade in
the eleventh and twelfth centuries.
Baths for men and women were an essential amenity of urban life. Bath houses and cara-
vanserais represented good sources of income, and people were therefore ready and willing
to build them. They were often originally milk but were frequently transferred to waqf own-
ership. Qubawi tells of two magnificent 'royal' bath-houses that were built in Bukhara by
Ahmad, head of the Karakhanid dynasty and grandson of the renowned Ibrahim Tamghach
Khan. References to many private and waqfbath-houses may be found in the list of prop-
erties transferred to waqf ownership in two waqf-namas belonging to Ibrahim in which he
describes the boundaries of the hospital and madrasa that he built in Samarkand. Booths
belonging to craftsmen and traders also attached themselves to the walls of these buildings.
To the best of our knowledge, strenuous efforts were made to keep the towns clean. It
was forbidden to throw rubbish into the streets and alleyways, which were considered to
be public property. Deep wells for rubbish and sewage, covered by earthenware or wooden
lids, were provided in private courtyards as well as in public places, houses and palaces,
according to the archaeological evidence. Archaeologists have discovered ceramic water
pipes and segments of paved streets and courtyards dating from the eleventh and twelfth
centuries. Blown window glass was also in use at the time.
Materials produced in large quantities such as ceramics and glass also provide a fair idea
of the development of urban crafts. The ceramic industry under the Karakhanids developed
its own style, and there were changes in techniques and in the range of colours used to dec-
orate ceramics. Crockery completely covered with a blue glaze enjoyed great popularity.
Equally in demand were bowls and cups with an underglaze decoration in the form of
intricate geometric and wickerwork patterns in an elegant combination of light-brown and
dark-brown tones. Kufic inscriptions were still employed, but they were transformed into
an illegible form of decoration imitating Kufic script. Unglazed ceramics resembling met-
alware in their form and decoration were in great vogue. Such jugs and flasks in grey clay
are entirely covered by geometric, stylized vegetal patterns in relief, arranged in several
horizontal bands around the vessels. Wild animals are also depicted on this type of ware.

148
ISBN 978-92-3-103467-1 Contents Towns and trade

Attractive pottery became cheaper and more easily affordable for the bulk of the urban
population as a result of certain technical innovations and the standardization of forms. It
is an important sign of the times that this urban ware reached the most remote mountain
areas. Yet the influence of the ceramic traditions of these mountain areas on urban ware
is of no lesser interest. Unglazed ware thrown on the potter's wheel and embellished with
simple terracotta designs made its appearance in the towns, copying the forms and deco-
ration of the unthrown mountain ware. This indicates that the population of the towns was
swelled in the eleventh and twelfth centuries not only by lowland peasants and by settled
Turkish peoples but also by mountain people whose tastes were at once reflected by urban
craftsmen.
The glassware, produced in a variety of forms and sizes for a variety of purposes, also
merits attention. Flasks for toiletries and pharmaceuticals, bowls, stemless drinking cups
and long-stemmed goblets are among the common finds of archaeologists; blown from
transparent, coloured glass, they also ceased to be luxury items. Although engraved and
silver-encrusted bronze jugs were very expensive, simpler, almost unadorned versions have
also been found.
In the course of the eleventh and twelfth centuries, baked brick came to be used more
widely, especially in major construction projects such as palaces, mosques, madrasas,
mausoleums and bridges. Patterned brick facework and engraved terracotta were exten-
sively employed for the decoration of important buildings. Epigraphic masterpieces in
engraved terracotta - Kufic and Naskhi inscriptions in high relief framed by elegant plant
and geometric ornamentation - have survived to the present day. Extant Karakhanid monu-
ments include four mausoleums in Ferghana, the mosque of Magok-i Attari and the Kalan
minaret (of the Friday mosque) in Bukhara as well as the above-mentioned Ribat-i Malik
in the steppes between Bukhara and Samarkand.

149
ISBN 978-92-3-103467-1 Contents THE SELJUQS AND THE KHWARAZM SHAHS

THE SELJUQS AND

THE KHWARAZM SHAHS*
A. Sevim and C. E. Bosworth

Contents

THE ORIGINS OF THE SELJUQS AND THE ESTABLISHMENT OF SELJUQ

POWER IN THE ISLAMIC LANDS UP TO 1055 . . . . . . . 151
THE CONSOLIDATION OF THE SELJUQ SULTANATE IN IRAN . 161
Historical survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
The structure of the Seljuq state in the east . . . . . . . . . . . . . . . 165
THE EASTERN SELJUQ SULTANATE (1118- 57) AND THE RISE AND FLORES-
CENCE OF THE KHWARAZM SHAHS OF ANUSHTEGIN'S LINE UP TO
THE APPEARANCE OF THE MONGOLS (1097- 1219) . 167
Historical survey . . . . . . . . . . . . 167
The structure of the eastern Seljuq state 178
The structure of the Khwarazmian state 179

* See Maps 4 and 5.

150
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

Part One

THE ORIGINS OF THE SELJUQS AND THE

ESTABLISHMENT OF SELJUQ POWER IN THE
ISLAMIC LANDS UP TO 1055
(A. Sevim)

What might be called the prehistory of the Seljuqs is in many ways obscure. The Seljuqs
were a family of the Kin'ik tribe of the Oghuz Turkish people, a tribe which the lexicog-
rapher Mahmud al-Kashghari(who completed his Dfwan lughat al-turk in 1074) describes
as the leading, princely tribe of the Oghuz (for the early history of the Oghuz, see above,
Chapter 3). 1 In the early and middle decades of the tenth century the Oghuz nomadized
in the steppes to the north of the Aral Sea and the east of the Caspian Sea, reaching as far
west as the Ural and Emba rivers. They were, however, at a much lower stage of cultural
development than such peoples as the Bulghars on the middle Volga and the Khazars on
the lower Volga or the Kaduk to their east. The envoy of the c Abbasid caliph to the king
of the Bulghars, Ahmad b. Fadlan, travelled from Baghdad via Khwarazm and then across
the Oghuz steppes to the middle Volga. His Risa/a [Epistle] containing his travel account
provides us with our earliest authentic, as opposed to semi-legendary, information on the
Oghuz. He describes them as animistic in belief, following the counsel of their shamans in
some religio-cultural matters but in effect largely irreligious, and as wandering 'like stray-
ing wild asses' in the region of the Ustyurt plateau to the east of the Caspian. 2 It may be
that, at this time, they had some relationship with the Khans of the powerful Khazar state,
since certain sources state that Dukak and his son Seljuq3 served the king of the Khazars.

1
Al-KashgharI, 1939-41 , Vol. 1, pp. 55-8; Barthold, 1935, pp. 79 et seq.; Kafesoglu, 1955; Barthold,
1962, pp. 109-16; Bosworth, 1973, pp. 210-11, 219; Kafesoglu, 1988, pp. 21-2.
2
Ibn Fadlan, 1939, sections 20 et seq.; Barthold, 1962, pp. 91-9; Bosworth, 1973, pp. 215-18.
3
The conventional spelling of European scholarly usage is adopted here. Strict adherence to the phono-
logical laws of Turkish would require something like Seljuk, as in certain early sources, but by far the most
common spelling of the medieval Arabic and Persian sources is S. ljaq. See Rasonyi, 1939; Bosworth, 1973,
pp. 298-9.

151
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

It is further possible that the general southwards movement of the Oghuz at this time was
in response to pressure from other Turkish peoples, the Kipchak and Kimek, in the steppes
further to their north.
The tribal leader of the Oghuz was the Yabghu, a title dating back to Orkhon Turkish
times, and there were various other titles in the tribe denoting military office. Among the
shadowy ancestors of Seljuq himself are mentioned one Temtir-yallgh ('[the man with the]
iron bow') and Seljuq's own father Dukak (unless Temtir-yallgh and Dukak were the same
person). The Yabghu appointed Seljuq as su-bashi", or war leader (Ibn Fadlan's sahib al-
jaysh), but seems speedily to have considered him as a potential rival. Hence at some point
soon after the middle of the tenth century, Seljuq and his followers moved further up the
Jaxartes (Syr Darya) from the Yabghu's winter capital ofYengi-kent ('New Town') to Jand.
They were now near the borders of the Dar al-Islam, and soon afterwards, Islam began to
spread among the Seljuq family and their retainers, as it was at this time spreading among
the so-called 'trucial Turks', i.e. those on the northern borders of the Samanid state, semi-
Islamized and in some sort of loose treaty relationship with the Muslims. Later Seljuq
historiography then states that they immediately became ghazfs (fighters for the Islamic
faith), and that Seljuq's son Mika'il was killed in such fighting, so that his two sons Toghril
Beg Muhammad and Chaghr'i Beg Dawiid had to be brought up by their grandfather. 4
The Seljuqs' position at Jand allowed them to be recruited as auxiliary troops for
the defence of the Samanid amirate against pressure on its northern boundaries from the
Karakhanids and their Kaduk followers (see above, Chapters 4 and 6). The Samanid Niih
II b. Mansur recruited Seljuq troops under the leadership of Seljuq's son Arslan Isra'il, and
subsequently allowed them to settle near the small town of Nakhshab or Nur (modem Nur
Ata) between Bukhara and Samarkand (c. 990). Seljuq himself did not migrate southwards
and died at Jand in c. 1007. After his death, his eldest son Arslan became the Yabghu of
the Seljuqs, adopting this title as a conscious act of rivalry with the elder branch of the
Oghuz - for use of the title implied headship of the whole Oghuz people - from whom
Seljuq had split by his move from Yengi-kent. The rivalry between these two branches of
the Oghuz was to continue almost until the middle of the eleventh century, until the branch
of the Seljuqs was finally triumphant over the branch of the original Yabghu (see below).
It seems that various bands of Oghuz gave their allegiance to several leading figures in the
Seljuq family, such as Toghr'il, Chaghri and Musa, so that, in subsequent decades, when the
Turkmans (as the Oghuz, Ktipchak and related tribes of the south-western Turkish people

4
Cahen, 1949, pp. 41 et seq.; Pritsak, 1953, pp. 397-405; Barthold, 1962, pp. 99-100; Turan, 1971 ;
Bosworth, 1973, pp. 219-23; Kafesoglu, 1988, pp. 22-5.

152
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

begin to be called in the Islamic sources) 5 were overrunning Khurasan and northern Iran,
such separate bodies as the Yinaliyan, followers of Ibrahim Yinal, are mentioned.
Meanwhile, the Karakhanid Bughra Khan Harlin (Hasan) had entered the Samanid cap-
ital in 992 and temporarily occupied it (see above, Chapter 6). A few years later, the
Samanid amirate came to an end and its dominions were divided between the Ilig Khan
Nasr and Mahmud of Ghazna. In this fighting - and in the attempts, up to his death in
1005, of the last Samanid Ismacn al-Muntasir to retrieve his ancestral lands - the Seljuqs
and their partisans were to be found in the service of both sides on various occasions, fol-
lowing whoever could promise them the most plunder and pasture grounds and changing
sides without compunction. With the final triumph of the Karakhanids in Transoxania, the
Seljuqs forged links with the new regime. Arslan Isra'll gave military aid to cAlll Tegin b.
Bughra Khan Harlin, enabling him to seize control of Bukhara, married the Khan's daugh-
ter and became influential at his side. Toghril and Chagr'i, however, seem to have been
excluded from these arrangements and may have been on cool terms with Arslan Isra'il.
They and their followers were at this time in the steppes to the west of Sogdiana, the Kara
Kum desert. 6
Towards the end of the second decade of the eleventh century, in 1018 or 1019, these
Turkmens are recorded in certain Christian Syriac and Armenian sources and the late
Islamic source of Mirkhwand as having mounted, under the leadership of Chagr'i, a long-
distance raid westwards, across northern Iran to Azerbaijan and eastern Anatolia, reaching
as far as Dvin and Lake Van. The late Ibrahim Kafesoglu devoted a special study to this
episode and affirmed its historicity. 7 But Claude Cahen raised cogent objections to an expe-
dition as early as this, and placed it ten years later, the whole episode either resulting from
some chronological confusion in the sources or else from a subsequent attempt by publi-
cists or historians of the Seljuqs to glorify the exploits of Chagr'i, forebear of all the later
Great Seljuq sultans after Toghril. 8
Certainly by the mid-1020s groups of Turkmens had crossed the Kara Kum steppes and
were harrying the northern fringes of the Ghaznavid province of Khurasan. A catalyst for
these southward movements was the alliance of Mahmud of Ghazna with the Karakhanid
ruler of the east, Kadir Khan Yusuf b. Bughra Khan Harlin, against cAm Tegin in Bukhara.
This led to the temporary ejection of cAm Tegin from his city in 1025 and the consequent

5
The Arab geographer al-Maqdisi, who wrote c. 980, mentions Turkmaniyyan as harrying frontier posts
in the province of Isfijab on the middle Syr Darya, but it is not clear whether this was originally a political or
an ethnic term. See Pritsak, 1953, pp. 397-8.
6
Barthold, 1962, pp. 101-3; 1968, pp. 269-70; Bosworth, 1968, pp. 16-18; 1973, pp. 223-4.
7
Kafesoglu, 1953b.
8
Cahen, 1949, pp. 50-1.

153
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

flight from Sogdia of his Turkmen auxiliaries, under Arslan Isra'll, into the deserts to the
west. Mahmud must have felt that the Turkmens were a danger to his position in Khurasan,
for he seized Arslan Isra'll and dispatched him to imprisonment at the fortress of Kalanjar
in northern India, where he died after seven years' incarceration. 9
Deprived of Arslan's leadership, his Turkmen followers could only straggle across the
desert towards Khurasan. Various minor chieftains led individual bands, but Toghril and
Chaghr'i do not yet seem to have had sufficient prestige to impose their leadership on all
the Turkmens, and only emerged in dominant roles after the death in 1032 of Arslan Isra'll
(although the actual title of Yabghu was at that point assumed by the senior member of the
Seljuq family, Musa). Some of Arslan Isra'lis former followers, complaining of the tyranny
of their amirs and promising to act as frontier guards against further Turkmen incursions
from the deep steppes, asked Mahmud for permission to settle in northern Khurasan. The
sultan allowed 4,000 families, with their baggage and flocks, to settle near the towns of
Farawa, Sarakhs and Abiward. However, the Turkmens attacked the towns of northern
Khurasan and their herds disrupted the agricultural systems of the oases there. When in
1027 the people of Nasa and Abiward complained to the sultan about these activities,
Mahmud first sent a punitive expedition under his commander Arslan Hajib and then in
1028 came personally to Khurasan with an army, inflicting a crushing defeat on the Turk-
mens. Elements of them scattered wide, some to the Balkhan Kuh hills to the south-east
of the Caspian, some to the Dihistan steppes adjacent to Gurgan and others into northern
Persia. These last sought employment as auxiliaries with the local rulers there, threatened
by Ghaznavid expansion, such as cAla' al-Dawla Ibn Kakuya of Isfahan.
They now appear in such sources as the Ghaznavid historian Abu '1-Fadl Bayhaqi and
the later source Ibn al-Athir 10 as the so-called 'Iraqi' Turkmens, because they had pen-
etrated as far as western Iran, or ciraq cAjami. Under various leaders, among whom are
named Bugha, Kizil, Goktash, Yaghmur and Anas'i-oghlu, they speedily became a source
of chronic violence and unrest in the provinces of Rayy and Jibal. In his bid for the throne
of Ghazna in 1029, Mahmud's son Mascud enlisted Turkmens under Yaghmur at Rayy. He
later used them, as auxiliaries of the Ghaznavid regular army, for an expedition to Makran
in southern Baluchistan, but they never proved a reliable military force. Other Turkmens
remained in the Kara Kum and continued to raid northern Khurasan from there and from
Balkhan Kuh; in 1033 the Ghaznavid governor of Rayy, Tash Farrash, executed 50 of the
Turkmen chiefs, including Yaghmur. It is clear that such bands as these 'Iraqi' Turkmens
9
Nazim, 1931 , pp. 63-4; Barthold, 1962, p. 103; Bosworth, 1973, pp. 223-4; Kafesoglu, 1988, pp. 27-8.
10
Here Ibn al-Athir was very probably quoting the lost Malik-nama, an account written in Persian for
Chaghri' Beg's son, the subsequent Great Seljuq sultan Alp Arslan, as something like an official history of
the dynasty's origins and early history.

154
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

operated entirely independently of each other, in general remaining separate from those
bands acknowledging the leadership of the Seljuq family, with hostility between the two
groups rather than co-operation. The amirs of whose tyranny the Turkmens had complained
to Sultan Mahmud may have been members of the Seljuq family trying to extend their
authority over other bands. In 1035 the Seljuq leaders Musa Yabghu, Toghril, Chaghn and
Ibrahim Inal or Yinal (leader of the group of Turkmens appearing in the sources as the
Yinaliyan) approached the Ghaznavid camfd (governor) of Khurasan, Abu '1-Fadl Suri, for
permission for their 7,000 or 10,000 followers to settle at Nasa and Farawa. In return, they
promised to act as guards against fresh incursions of Turkmens from the Kara Kum and
Balkhan Kuh and to inflict punishment on the 'Iraqi' Turkmens. 11
Meanwhile, other Turkmen bands under the leadership of the Seljuq family had remained
in Sogdia, in an uneasy relationship with cAli Te gin, now as his allies, now as his foes. But
they felt their position to be increasingly untenable, and after 1029 moved away from
Sogdia to Khwarazm, settling on lands allotted to them by the Ghaznavid governor of
the province, Altuntash Hajib. Altuntash died of wounds received in 1032 at the battle of
Dabusiyya against cAli Tegin, the enemy of the new Ghaznavid sultan Mascud b. Mahmud.
Altuntash's son and successor in Khwarazm, Harlin, received less favour from the sultan
than had his father; hence by 1034 he renounced his allegiance to Ghazna and in effect
declared his independence. He allied with cAli Tegin of Bukhara and with the Seljuqs,
awarding the Turkmens further territories in Khwarazm. 12
The return movement northwards to Khwarazm brought the Seljuqs close to their ances-
tral territories at the mouths of the Oxus (Amu Darya) and Jaxartes (Syr Darya), where an
Oghuz principality had remained in existence, based on Yengi-kent and Jand, and directed
by the family of the original Yabghu of the Oghuz people. By this time the office of the
original Yabghu had devolved upon Abu '1-Fawaris Shah Malik b. cAli, who ruled from
Jand. Shah Malik raided the Seljuqs, now in Khwarazm, and is said to have killed between
7 ,OOO and 8,000 of their followers and carried back an immense booty of captives and
goods to Jand; efforts by Harlin to mediate between the two groups of Turkmens failed.
cAli Tegin, who had in general been favourably disposed towards the Seljuqs as allies
in his struggle to maintain his position against the Ghaznavids, died in 1034. His sons
Yusuf and Arslan Tegin continued the anti-Ghaznavid alliance with Harlin. A concerted
attack was planned, with the military assistance of the Seljuqs, in which Harlin and the
Seljuqs were to invade Khurasan, and the sons of cAli Tegin were to attack the Ghaznavid

11
Nazim, 1931, pp. 64-5; Cahen, 1949, pp. 56-7; Barthold, 1962, pp. 104-7; Bosworth, 1968, pp. 17-19;
1973, pp. 224-5; Kafesoglu, 1988, pp. 28-9.
12
Barthold, 1968, pp. 282-6, 295-6.

155
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

dependent principality of Chaghaniyan along the upper Oxus, then crossing the river at Tir-
midh (modern Termez) so that the two invading forces could link up at Andkhuy in what
is now northern Afghanistan. Mascud of Ghazna procured the assassination of the Seljuqs'
protector Harlin in 1035, and the double attack came to a halt when the forces of the sons
of cAli Tegin were unable to capture Tirmidh. 13
Let us now return to the Seljuqs and their followers on the northern fringes of Khurasan.
Sultan Mascud received the humbly worded request of the Seljuqs and their followers
(at this time, in a wretched condition after their defeat and dispersal by Shah Malik in
Khwarazm) to settle on these fringes as frontier guards and auxiliaries. His vizier Ahmad
b. cAbd al-Samad and his civilian advisers advised acceptance, at least until the Turk-
mens had openly shown their bad faith; but the sultan and his commanders, mindful of the
depredations of the Turkmens in Khurasan towards the end of Mahmud's reign, replied by
dispatching to Khurasan a powerful army under the Turkish commander BegtoghdL The
Seljuqs suffered initial defeats but then in August 1035, on the road to Nasa, they won a
decisive victory over the Ghaznavid army. In this instance, as was to be the case later, the
superior mobility and lighter equipment of the Turkmens enabled them to defeat the more
heavily armed but cumbersome Ghaznavid forces, hampered as they were by their long
baggage train and their inability to manoeuvre in the desert without extensive supplies.
Begtoghd'i only returned to Nishapur with difficulty. The Seljuqs acquired an immense
booty of money, arms, clothing, horses, and so on, and were amazed at their own victory
over the seemingly invincible Ghaznavids, ascribing their victory to Divine Providence and
to Begtoghd'i's poor tactics rather than to their own abilities. Mascud now had no choice
but formally to award to Toghrll, Chaghn and Musa Yabghu the regions of Nasa, Farawa
and Dihistan, with the insignia - standards, robes of honour, etc. - and style of dihqans
(here, governors on behalf of the sultan).
Mistrust nevertheless prevailed between the two sides. The sultan endeavoured to detach
Musa Yabghu from the other leaders by, among other things, awarding him the additional
title of Inanj, and marriage alliances were proposed, with wives from the Ghaznavid mil-
itary and official class for the Seljuqs. But Toghnl and Chaghn, in particular, remained
suspicious. New demands were made, this time for a grant of the regions of Merv, Sarakhs
and Abiward, on the grounds that the Turkmens' existing pasturelands were inadequate.
Mascud could certainly not agree to hand over a rich city like Merv to the nomads; more-
over, the Seljuqs were making a nuisance of themselves in other ways, raiding through
Khurasan as far as Sistan and entering into relations with enemies of the Ghaznavids

13
Pritsak, 1953, pp. 407-8; Barthold, 1968, pp. 297-8; Bosworth, 1973, p. 238; Kafesoglu, 1988,
pp. 30 -2.

156
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

like Ismacil Khandan (brother of the murdered Harlin b. Altuntash), who had taken over
Khwarazm and assumed the historic title of Khwarazm Shah, and the Karakhanid Bughra
Khan Muhammad b. Yusuf Kad'ir Khan. Ghaznavid armies were now sent to Khurasan
under the commander Sti-bash'i and also to Herat, while the sultan himself, failing to com-
prehend the seriousness of the situation in the west, campaigned in India. Sti-bash'i's army
was again soundly defeated in May 1038 and he fled with the remnants of his forces to
Herat. 14
The way was now open for the Seljuqs to enter the capital of Ghaznavid Khurasan,
Nishapur. Ibrahim Inal appeared there with 200 horsemen later in May 1038 as the advance
guard for what was to be the first Seljuq occupation of the capital. According to the sources
- which may, however, exaggerate the degree of political sophistication shown by the Turk-
mens at this time - Toghrll proclaimed himself successor to the sultan. He seated himself
on the latter's own throne in the suburb of Shadyakh, performed such sovereign acts as
presiding over the hearing of complaints (mazalim) and assumed regal titles including,
it is related in the Malik-nama, that of al-sultanal-muc azzam (Exalted Sultan). Toghrll is
said to have tried to restrain the Turkmens from their natural instinct to pillage, but to
have dissuaded his brother Chaghr'i from this only with difficulty. The cAbbasid calilph al-
Qa'im (1031-75) subsequently sent an envoy to Toghrll, implying some degree of recog-
nition for the Seljuqs in Khurasan. The Seljuqs are said already to have styled themselves
mawalf amfr al-mu'minfn (Clients of the Commander of the Faithful) when they were
established on the fringes of northern Khurasan in 1035, but such phrases were stereotypes
and need not have implied at this point any direct connection between the Seljuq chiefs
and Baghdad. 15
Mascud now endeavoured to concert operations against the Seljuqs by allying with the
original Oghuz Yabghu, Shah Malik of Jand, against the rebellious Ismacn Khandan in
Khwarazm and his Seljuq allies. (Shah Malik was ultimately victorious in Khwarazm, in
early 1041, but Mascud himself was killed shortly afterwards.) An army of 50,000 troops
was prepared for Khurasan. Some of the Seljuq leaders were fearful of their position there,
and of what they still regarded as Ghaznavid invincibility, and proposed a withdrawal west-
wards to Rayy and Jibal. It seems that the Seljuqs may still have placed Sultan Mascud's
name in the khutba (Friday worship oration) of Nishapur alongside their own leader's
name. The Seljuqs were defeated by the Ghaznavid army in summer 1039 and compelled
to withdraw into the Kara Kum desert. Nishapur was reoccupied by the sultan and his army
towards the end of 1039, and a great campaign against the Seljuqs was prepared for spring

14
Cahen, 1949, pp. 55-61; Bosworth, 1973, pp. 241-3; Kafesoglu, 1988, pp. 32-4.
15
Cahen, 1949, pp. 61-3; Bosworth, 1973, pp. 244-5, 252-68; Kafesoglu, 1988, p. 34.

157
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

1040. Khurasan had already been denuded of food and fodder by the incessant warfare and
the trampling of the nomads' herds. The Ghaznavid army, with all its impedimenta, includ-
ing war elephants, was thus ill-equipped for the campaign, and suffered severely from the
lack of water and supplies in the heat of the desert between Sarakhs and Merv.
A historic battle, which was to be decisive for the future of the Iranian lands and for
the establishment of Seljuq power in the ancient lands of Islamic civilization, took place
at the ribat (outpost) of Dandanqan in May 1040. In this clash, 16,000 lightly armed but
mobile Turkmens overcame the ponderous Ghaznavid army, by now demoralized and badly
affected by hunger and thirst. Mascud fled and ultimately withdrew towards northern India,
fearing the loss of the whole of his possessions in Khurasan and Afghanistan to the Turk-
mens - he was killed by a revolt among his troops while en route for Lahore. (Mascud's
despair proved premature, for his son and eventual successor Mawdud re-established the
Ghaznavid position at least in the eastern half of Afghanistan: see above, Chapter 4, and
below, Part Two.)
The victorious Toghril set up his throne on the battlefield of Dandanqan as amir of
Khurasan. Letters announcing the victory were dispatched to various Karakhanid rulers,
including to the sons of cAli Tegin, to Muhammad b. Nasr cAyn al-Dawla of Uzgend and
to Ibrahim b. Nasr Bori Tegin, who had been harrying the Ghaznavid lands on the upper
Oxus; to the rulers of western and central Iran; and to the cAbbasid caliph in Baghdad. The
fat'h-nama (announcement of victory) to al-Qa'im detailed the oppression to which the
Turkmens had been subjected at the hands of Mahmud and Mascud and promised faithful
allegiance by the Seljuqs to the caliph and the cause of orthodox Sunni Islam; at the top of
the letter was inscribed the ancient bow-and-arrow symbol of the Turks. 16
Toghril was now a territorial sovereign and not merely a chief of nomadic bands. In
accordance with the old Turkish practice of a patrimonial concept of power, with senior
members of the ruling family sharing out governorships and territories, the supreme ruler
Toghril now made various delegations of his authority in the form of grants. His brother
Chaghr'i Beg Dawud was allotted all of Khurasan and all those lands north of the Oxus that
he might conquer, while Merv was to become his capital. Musa Yabghu was subsequently
to make his headquarters in Herat and extend his power southwards to Sistan, where the
local Nasrid Maliks became Seljuq vassals.
The Seljuqs were complete strangers to the business of ruling a territorial state which
had long-established administrative traditions rooted in the Perso-Islamic past. They were
therefore unable to do without the services of local Khurasanian secretaries and officials

16
Barthold, 1962, pp. 107-8; 1968, pp. 302-3; Bosworth, 1968, pp. 19-23; 1973, pp. 251-2; Kafesoglu,
1988, pp. 35-6.

158
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

Uust as the Turkish Ghaznavids had recourse to former officials of the Samanids of Tran-
soxania and Khurasan) for the actual running of the extensive lands they now controlled.
Many officials moved from the service of the Ghaznavids to that of the new regime. In
his first occupation of Nishapur, Toghril had relied greatly on the aid of a local magnate,
the Salar of Buzgan, Abu '1-Qasim Buzganl. Various sources give lists of Toghril's later
viziers, including among others Abu '1-Qasim al-Juwayni; the ra'fs al-ru'asa' (head of the
town notables) of Nishapur, Abu cAbd Allah Husayn, from the distinguished local family
of the Mikalis; and the camfd al-mulk (chief secretary) Abu Nasr Muhammad al-Kunduri,
a former official of the Ghaznavids, who went on to serve Toghril's successor Alp Arslan.
Chaghr'i had as his vizier Abu cAli Shadhan, and among the officials in his administration
was a brother of the famous Ismacili author and traveller Nasir-i Khusraw. 17
The next years were spent consolidating the Seljuq position in the eastern Iranian lands
and making new conquests in central and western Iran. Responsibility for the first had
been allotted to Chaghr'i Beg. He laid siege to Balkh (defended by its governor, Altuntash)
on behalf of the Ghaznavids, and despite counter-measures by the new Ghaznavid sultan
Mawdud b. Mascud, captured it, together with other towns and fortresses of the regions
of Badhghis, Guzgan and Tukharistan. While Chaghr'i was preoccupied with affairs in
Khwarazm (see below), Mawdud was able temporarily to recapture Herat from Musa Inanj
Yabghu, but lost it again. Chaghr'i's son Alp Arslan now undertook operations in the upper
Oxus districts of Tirmidh, Qubadiyan, Wakhsh and Qunduz until a stable frontier was
eventually established in northern Afghanistan during the 1050s between the Ghaznavids
and the Seljuqs (see below, Part Two). Ertash, the brother oflbrahim Inal, imposed Seljuq
control over Sistan in the name of Musa. Chaghr'i's son Kawurd was encouraged to expand
southwards through Kuhistan and was able to set up in Kirman, in place of the local branch
of the Buyid dynasty, a semi-autonomous amirate which was to endure for nearly one and
a half centuries (see below, Part Two).
Shah Malik of Jand had by 1041 driven out of Khwarazm the Khwarazm Shah Ismacn
Khandan, who was compelled to seek refuge with the Seljuqs. Chaghr'i led an army into
Khwarazm in 1043 which expelled Shah Malik from the province; he fled southwards
through Khurasan to Makran on the Arabian Sea coast, but was captured by Ertash (who
was then operating in Sistan), handed over to Chaghr'i and killed. In this way, the ancient
rivalry between the senior branch of the Oghuz under the Yabghus of Yengi-kent and
Jand and the new force of the Seljuqs was finally ended by the total victory of the lat-
ter. Khwarazm was placed under a Seljuq governor, and little more is heard of the province
during Chaghr'i' s lifetime except for a revolt there whose suppression required Chaghr'i' s

17
Bowen, 1957; Bosworth, 1973, pp. 257, 262-3, 264-6.

159
ISBN 978-92-3-103467-1 Contents THE SELJUQS IN THE ISLAMIC LANDS

presence. On this occasion, he also received the submission of the 'amir of Kipchak', who
became a Muslim and married into Chaghr'i's family. 18
In the west, the lands for which Toghril made himself responsible, Toghril's kinsman
Ibrahim Inal overran Dihistan, Gurgan and Tabaristan. He brought under obedience to
Toghr'il the Turkmens formerly headed by Arslan Isra'll who, under Kizil Beg, had estab-
lished themselves in Rayy and northern Iran. Under new pressure, these Turkmens moved
westwards towards Azerbaijan and the Byzantine and Armenian frontier. Hence in 1042-3
Toghr'il was unable to come to Rayy personally, receive the city from Ibrahim Inal and
begin the reconstruction of the city's buildings, which had been much devastated in the
previous warfare. A new government headquarters (dar al-imara), was built, together with
a mosque and a madrasa; Toghril now moved his capital from Nishapur to Rayy, and in
this year (A.H. 434) began minting coins there. 19
Ibrahim Inal moved westwards to conduct operations against the Kurdish c Annazids and
other local powers on the Iran-Iraq-Byzantium frontiers. Toghr'il himself marched against
the Kakuyid ruler of Isfahan, Faramarz b. c Ala' al-Dawla Muhammad, and made him
his vassal, soon afterwards making Isfahan his own capital. In subsequent years, Toghr'il
endeavoured to reduce the power in western and southern Iran of the Buyids, who were,
fortunately for his purposes, undergoing a period of internal strife and rivalry between
the contending princes, al-Malik al-Rahim Khusraw Rruz b. Abu Kalijar of Iraq and his
brother Fiilad Suttin of Fars. By 1052 Turkmen raiders had penetrated into Buyid Fars. It
was almost inevitable that one of the warring brothers should call on the Seljuqs for aid;
hence in 1053 Fulad Suttin agreed to place Toghril's name in the khutba of his capital
Shiraz, and in the following years a group of Turkmens took over Khuzistan. Al-Malik
al-Rahim's seven-year rule in Baghdad (1048-55) was marked by continual violence and
rioting. Hence in 1055 Toghr'il assembled forces in Jibal and Kurdistan with the proclaimed
intention of making the pilgrimage to Mecca and of combating the Shfite Fatimids of
Egypt and Syria. He entered Baghdad with his army in December 1055, deposed the
Buyid ruler and undertook operations against the Fatimids' supporters in Iraq. Eventu-
ally, in 1058, he appeared in Baghdad again and on this occasion, met for the first time
the caliph al-Qa'im. In a series of splendid ceremonies, the c Abbasid bestowed on Toghr'il
the honorific titles rukn al-dawla (Pillar of the State), qasfm amfr al-mu'minfn (Partner
of the Commander of the Faithful) and malik al-mashriq wa 'l-magbrib (King of the East
and West), together with robes of honour in the c Abbasid colour of black and two crowns

18
Pritsak, 1953, p. 408; Bosworth, 1968, pp. 49-52; Koymen, 1972, pp. 3-9; Bosworth, 1977, pp. 9-30,
49-52; Men;il, 1980, pp. 11 et seq.; Kafesoglu, 1988, pp. 36-8.
19
Miles, 1938, pp. 196 et seq.

160
ISBN 978-92-3-103467-1 Contents Historical survey

signifying rule over the Arabs and the c Ajam or non-Arabs, and the khutba in the Great
Mosque of Baghdad was made in Toghril's name. In this way, the caliph was relieved of
his enemies in Iraq, and Toghril's position exalted as his deliverer. 20
There now begins the de facto dual arrangement, of such importance for the future
constitutional development of the Islamic world, whereby the sultan is recognized as the
secular ruler and the caliph-imam remains the moral and spiritual head of the Islamic com-
munity. This arrangement, although initially seen as novel and disturbing, was ultimately
to be recognized by Sunni Muslim constitutional and legal theory.

Part Two

THE CONSOLIDATION OF THE SELJUQ

SULTANATE IN IRAN (1055-1118)
(C. E. Bosworth)

Historical survey
After his formal assumption of the sultanate at Baghdad in 1055, Toghril was during his
latter years largely occupied with consolidating his family's power in the Iranian lands
west of Khurasan and in the Arab lands of Iraq and Mesopotamia. Khurasan and the east
were left to his brother Chaghn Beg under Toghril's supreme overlordship, even though
in practice Chaghn seems to have been left very much to himself in his territories (in the
sources, Chaghn remains a distinctly more shadowy figure than Toghnl).
In Chaghn' s time, the province of Sistan, straddling what is now the border between Iran
and Afghanistan, remained under the indigenous family of the Nas rid Maliks of Nimruz,
as vassals of the Seljuqs, required on occasion to furnish troop contingents for the Great
Seljuq army. By 1048 Kirman in south-eastern Iran had passed from the hands of its former
masters, the Buyids, into those of the Seljuq prince Kara Arslan Kawurd b. Chaghn Beg,
and a line of Kawurd's descendants now followed in Kirman, forming an autonomous
Seljuq dynasty. For well over a century, the province was to enjoy a period of peace and
prosperity under its new rulers, favoured by its position on trade routes connecting Central
20
Bosworth, 1968, pp. 45-8; Kafesoglu, 1988, pp. 39, 42-3.

161
ISBN 978-92-3-103467-1 Contents Historical survey

Asia and Khurasan with the Persian Gulf and Indian Ocean ports. This favourable status
was only ended when, after 1170, real power in the principality passed to Turkish slave
commander atabegs (tutors to young Seljuq princes); also, the province was invaded by
bands of Oghuz Turks deflected southwards from Khurasan by the fighting there between
the Khwarazm Shahs and the Ghurids (see below, Part Three, and Chapter 8), so that the
line of Seljuq princes was ended in 1186 when the Oghuz leader Malik Dinar took over in
Kirmn. 21
In the lands to the south of the upper Oxus, what later became northern Afghanistan, a
rough, north-south boundary was established between the Seljuqs and the rulers whom
they had supplanted in eastern Iran, the Ghaznavids. This was to remain substantially
the boundary between the two great powers for over fifty years, since Ibrahim of Ghazna
(1059-99) - with the exception of one or two occasions, see below - eschewed irredentist
campaigns aimed at recovering the Ghaznavids' lost Khurasanian provinces.
Since Toghril was childless, it was Chaghr'i Beg's son Alp Arslan who in 1063 took
over as supreme sultan of the Seljuq empire - like his uncle, with the formal approval of
the c Abbasid caliph in Baghdad, who granted him the honorific titles of c Adud al-Dawla
and Diya' al-Din. The decade of Alp Arslan's rule, together with the twenty-years' rule of
his son Malik Shah, represent the zenith of Great Seljuq power: their empire reached an
extent unparalleled since the heyday of the c Abbasid caliphate in the later eighth and early
ninth centuries. The two reigns of Alp Arslan (1063-73) and of Malik Shah (1073-92) may
be considered as a unity, the unifying factor being the directorship of the continued, day-to-
day running of the state by the great vizier, Nizam al-Mulk. A native of Tus in Khurasan, he
exemplified the class of professional Iranian secretaries and officials upon whom the Seljuq
sultans-as incomers into the Islamic world from the Central Asian steppelands-wisely
relied for the administration of their vast empire. Nizam al-Mulk had grown up in Ghaz-
navid Khurasan and had spent some years in Ghazna. Thus when he entered the service of
Chaghr'i Beg and Alp Arslan he brought with him into the Seljuq administration an element
of continuity, that of the old-established Perso-Islamic government tradition whose origins
went back to the Sasanian period. He was to expound these traditions in a masterly and
effective way in his treatise on statecraft, the Siyasat-nama [Book of Statecraft]. 22
At the outset, Alp Arslan secured his position in the east by placating his elder brother
Kawurd of Kirman, although an expedition thither was necessary in 1067 when Kawurd
withdrew his allegiance. Although he restored it in the face of Alp Arslan's superior mili-
tary might, Kawurd was never fully reconciled to his subordinate position. On Alp Arslan's

21
Men;il, 1980; Kafesoglu, 1988, pp. 65-6; Bosworth, 1994, pp. 376-86.
22
EI2 , 'Nizam al-Mulk'.

162
ISBN 978-92-3-103467-1 Contents Historical survey

death he was to rebel against the new ruler Malik Shah, only to be defeated and then exe-
cuted by the victor in 1073. For most of his reign, Alp Arslan was occupied with affairs
in the west, including the frontiers of the empire in the Caucasus, and it was in the west
that he achieved his famous victory over the Byzantine emperor Romanus Diogenes at
Mantzikert (Malazgird) in 1071. The eastern frontiers were strengthened by the granting
out, in 1066 when the sultan proclaimed Malik Shah as his heir, of various districts there as
appanages for Seljuq princes: for example, Khwarazm to his brother Arslan Arghun, Balkh
to his brother Sulayman, Merv to his son Arslan Shah and Tabaristan to lnanj Yabghu.
Alp Arslan personally led expeditions into the Kipchak steppes of Central Asia, as far as
J and on the lower Syr Darya, and to the Mang'ishlak peninsula on the eastern shore of the
Caspian Sea. 23
On his father's death, Jalal al-Dawla Malik Shah secured the support of the army by
distributing wealth from the treasury of Nishapur, the capital of Khurasan. He continued
Alp Arslan's policies and in some ways surpassed the latter's triumphs. As noted above,
an equilibrium had been established between the Ghaznavid and Seljuq empires, and this
was only briefly disturbed in 1073 when Ibrahim of Ghazna attempted, vainly as it turned
out, to regain former Ghaznavid territory in northern Afghanistan. Seljuq cultural influence
grew in the Ghaznavid empire during these decades, with the Ghaznavid sultans assuming
for themselves the established Seljuq title al-sultan al-mucazzam (Exalted Sultan) in addi-
tion to their normal ones of amir and Malik. Marriage links were forged and lbrahim's
son Masciid (the subsequent Sultan Masciid III, 1099-1115) married one of Malik Shah's
daughters, Jawhar Khatun. 24
Malik Shah's activities in the east of his empire mainly involved relations with the
Karakhanids, except for the two occasions (in 1080 and 1084) when he had to deal with
revolts in Khurasan by his brother Tekish b. Alp Arslan, the second one ending in the rebel's
imprisonment and blinding. One of Malik Shah's most forceful wives was the Karakhanid
princess Terken Khatun, who was, at the Seljuq court, for long the focus of opposition to the
vizier Nizam al-Mulk and his policies. Soon after his accession, the sultan led an expedition
to the capital of the Western Karakhanid Khanate, Samarkand, in order to punish Shams
al-Mulk Nasr b. Ibrahim Tamghach Khan, who had tried to profit from the succession
uncertainties on Alp Arslan' s death by invading Tukharistan to the south of the upper Oxus.
For several years, nothing is heard of Seljuq-Karakhanid relations (although we know
that Seljuq cultural influence spread within Transoxania during this period, as seen in the
Karakhanid adoption of Seljuq-type titulature) until towards the end of Malik Shah's reign.

23
Barthold, 1968, pp. 313-14; Bosworth, 1968, pp. 54-6; Koymen, 1972; Kafesoglu, 1988, pp. 45-9.
24
Bosworth, 1977, pp. 50-8.

163
ISBN 978-92-3-103467-1 Contents Historical survey

At this point, he was persuaded by the orthodox Sunni religious opponents of the Khan
Ahmad b. Khidr in 1089 to invade Transoxania and depose the Khan. He then pushed
on beyond the Syr Darya to Talas and into Semirechye in order to impose his overlord-
ship on the eastern branch of the Karakhanids, receiving at U zgend the submission of the
Khan of Kashghar, Hasan or Harlin b. Sulayman Tamghach Bughra Kara Khan. Thus at
this moment, the khutba was made in Malik Shah's name from northern Syria to East
Turkistan. Tribal and family dissensions continued within the Karakhanid lands, requiring
Malik Shah's intervention, and at some point restoring Ahmad b. Khidr to his former throne
in Samarkand (he was later, in 1095, to be arraigned and executed by his old opponents, the
religious classes in Samarkand, on the grounds that he had adopted Ismacili doctrines). 25
Malik Shah's death in 1092 inaugurated some twelve years of confusion and internecine
warfare within the western Iranian and Iraqi lands of the Seljuq empire, for the dead sul-
tan's two sons Berk-yaruk and Muhammad quarrelled over the succession. Berk-yaruk, the
candidate of the Nizamiyya, the sons and supporters of the vizier Nizam al-Mulk (who
had been assassinated shortly before the sultan's own death), eventually prevailed, but he
was never to be undisputed master of the united sultanate. Various members of the Seljuq
family seized the opportunity to intrigue or to assert their own claims, including the former
rebel Tutush, Berk-yaruk's uncle, now in touch with elements in his former appanage of
Tukharistan, and his son, and another uncle, Arslan Arghun, in Khwarazm. Military cam-
paigning by Berk-yaruk and his supporters eventually made firm his power in Khurasan,
to which in 1097 the sultan appointed his half-brother Sanjar as governor, providing him
with an atabeg and a vizier (see below, Part Three). 26
Affairs among the Karakhanids in Transoxania had been somewhat confused and trou-
bled after the death of Ahmad Khan, but in 1097 Berk-yaruk, now suzerain, confirmed
the succession in Samarkand of a succession of ephemeral rulers: Sulayman b. Dawud b.
Tamghach Khan, Kad'ir Khan Ibrahim (1097), Mascud b. Muhammad (1097-9) (both of
whom married daughters ofBerk-yaruk) and Jibra'il or Jibril b. cumar (1099-1102). Later
in 1097, however, Berk-yaruk had to leave the east for western Iran and Iraq, essentially the
seat of his power. Before departing from Khurasan, he appointed as governor in Khwarazm
the Turkish commander Qutb al-Din Muhammad b. Anushtegin Gharcha'i to replace the
dead Ekinchi b. Kochkar, in this manner inaugurating the line of shahs in Khwarazm from
the line of Anushtegin which was to endure for over a century (see below, Part Three).
It was Sanjar who was now to be responsible for the maintenance of Seljuq authority in

25
Kafesoglu, 1953a; Barthold, 1956, pp. 97-8; 1968, pp. 314-18; Bosworth, 1968, pp. 66-7, 87-102;
Kafesoglu, 1988, pp. 49-55.
26
Sanaullah, 1938, pp. 82-113; Bosworth, 1968, pp. 102-13; Kafesoglu, 1988, pp. 56-9.

164
ISBN 978-92-3-103467-1 Contents The structure of the Seljuq state in the east

the east. He continued to exercise power there, from his capital at Balkh, when his full
brother Muhammad Tapar b. Malik Shah succeeded to the Great Seljuq throne after Berk-
yaruk's death in 1104. Sanjar continued to acknowledge his constitutional dependence on
the supreme Sultan Muhammad, who enjoyed on his coins the title al-sultan al-muc azzam,
and was content to style himself on coins malik al-mashriq (King of the East)-a subordi-
nation which was, however, to be abandoned on Muhammad's death in 1118 (see below,
Part Three). 27

The structure of the Seljuq state in the east

The eastern Iranian lands of the Seljuq state were governed, like the rest of the empire, from
the central Great Dfwan, the sultans' executive organ, which was, for nearly thirty years,
as noted above, under the direction of Nizam al-Mulk, aided by his sons and partisans. All
of these made up the body which came to be known, especially after Nizam al-Mulk' s own
death, as the Nizamiyya. Hence during Malik Shah's reign, the vizier appointed several
of his sons to strategically important provincial governorships, where trusty subordinates
were required for putting his administrative measures into practice. Two of his sons, Shams
al-Mulk cuthman and Jamal al-Mulk Mansur, were governors in Merv and Balkh respec-
tively at certain times. 28
All through his official career, Nizam al-Mulk was concerned to buttress the author-
ity of his masters, the Seljuq sultans, against such external foes as the Fatimid caliphs
of Egypt and Syria and internal dissidents like the Ismacilis of Daylam and the Elburz
mountains region and of Kuhistan in southern Khurasan. Military operations were under-
taken on various occasions against the Ismacilis within the empire, but on the ideologi-
cal and intellectual level, Nizam al-Mulk's name is associated with the founding and the
spread of Sunni orthodox madrasas, named Nizamiyyas in his honour. The institution
of the madrasa had flourished in the eastern Iranian lands for many decades before the
vizier's time, so he was far from being an innovator here. 29 But Nizam al-Mulk may have
hoped that his newly founded madrasas - which for the east were located in Balkh, Herat,
Merv, Nishapur and Amul in Gurgan, according to the later biographer of Shafici scholars,
Taj al-Din al-Subki-would produce theologians and lawyers who could uphold orthodoxy
against sectarian currents like Muctazihsm and Ismacilism, and also officials and secretaries
who could implement his own policies for the Seljuq empire's future. For staffing these

27
Barthold, 1968, pp. 318-19, 323-4; Bosworth, 1968, pp. 135-6, 138-9.
28
Bosworth, 1968, pp. 56-7, 69-70; Klausner, 1973; E/2 , 'Nizam al-Mulk'.
29
On educational policies and the spread of the madrasa in general, see Volume IV, Part Two, Chapter 1,
of the present work.

165
ISBN 978-92-3-103467-1 Contents The structure of the Seljuq state in the east

colleges, Nizam al-Mulk made especial use of scholars from Khurasan, for Khurasan
and Transoxania had long been bastions of Sunni orthodox theology, law and scholarship. 30
The Seljuq military forces at this time comprised not only a professional army, with
a nucleus of ghulams (slave soldiers) but also contingents of free troops, mercenaries.
Turks were naturally dominant, and held many of the high commands, but the professional
army was multi-ethnic, and included Armenians, Greeks, Slavs, Arabs and others. In his
Siyasat-nama, Nizam al-Mulk positively commends the use of Daylamites, Khurasanians,
Georgians and Shabankara'i (Kurds) from southern Iran in the army. But at the side of
this professional, standing army - supported, at least in western Iran and Iraq, by a system
of iqtacs (land grants) - the sultans still relied to a great extent on the descendants of
their original, Oghuz tribal following, the Turkmens. The military power of these last was
particularly important for the defence of Khurasan and the eastern marches, the border
lands beyond which lay such powers as the Ghaznavids and the Karakhanids, and beyond
the latter, the peoples of the Inner Asian steppes and forests. These frontier lands were
often granted as appanages to Seljuq princes or to the proteges of Nizam al-Mulk, but their
actual defence fell largely to the Turkmens. Since they had first appeared on the fringes
of Ghaznavid Khurasan (see above, Part One), the Turkmens had been assigned grazing
grounds and rights for their herds there, and these rights continued under the Seljuqs, since
they provided the livelihood and the maintenance of these unpaid, auxiliary troops. The
tribal nomads thus had a definite place within the economic and agricultural structure of
such eastern provinces as Khurasan, Gurgan and Dihistan. It was felt that it should be the
sultans' care to conciliate and to look after the interests of their Turkmen supporters, since
these had been the original mainstay of the Seljuq family's power, states Nizam al-Mulk
again in his treatise - possibly with the implication that the Turkmens' complaints and just
claims were no longer being listened to properly or dealt with immediately, now that a
ruler like Malik Shah relied increasingly on his professional, standing army. The sultans
did, however, continue to give regular feasts (sholen) for their supporters, but Nizam al-
Mulk states that Malik Shah's failure to provide such a feast for the Chigil tribesmen of
the Karakhanid army, when he campaigned in Transoxania in 1089, caused him a loss of
prestige there. 31

30
Bosworth, 1968, pp. 70-4; Lambton, 1968, pp. 214-17.
31
Sanaullah, 1938, pp. 18-35; Bosworth, 1968, pp. 79-84; Lambton, 1968, pp. 224-39.

166
ISBN 978-92-3-103467-1 Contents Historical survey

Part Three

THE EASTERN SELJUQ SULTANATE (1118-57)

AND THE RISE AND FLORESCENCE OF THE
KHWARAZM SHAHS OF ANUSHTEGIN'S LINE UP
TO THE APPEARANCE OF THE MONGOLS
(1097-1219)
(C. E. Bosworth)

Historical survey
c Adud al-Dawla Ahmad Sanjar (Turkish sanjar. 'he who pierces, thrusts') governed the
eastern provinces of the Great Seljuq empire for some sixty years, being appointed in
490/1097, while still a boy, by his half-brother, the Great Seljuq sultan Berk-yaruk after the
unsucessful revolt in Khurasan and the death of Arslan Arghun b. Alp Arslan. He remained
there, as boy and man, until his death shortly after escaping in 1156 from the custody of
the Oghuz of Khurasan. During the civil strife in western Persia and Iraq between his elder
brothers Berk-yaruk and Muhammad Tapar (see above, Part Two), Sanjar generally took
the side of his full brother Muhammad, but from the constitutional aspect he regarded
himself as governor only of the eastern provinces and as subordinate to the supreme sultan
in the western lands, calling himself on his coins merely a Malik and acknowledging Berk-
yaruk and then Muhammad as al-sultan al-muc azzam. 32
When Muhammad died in 1118, however, Sanjar refused to consider himself subor-
dinate to his nephew in the west, Mahmud b. Muhammad. As the senior member of the
Seljuq family, both his de facto power and his position under Turkish tribal custom gave
him a claim to the supreme sultanate even though this had previously been held, for eighty
years, by the Seljuq who controlled western Iran and Iraq. The squabbling sons of Muham-
mad b. Malik Shah were too divided and militarily weak to dispute Sanjar's position, and

32
Bosworth, 1968, p. 135; Lowick, 1970, pp. 244-5.

167
ISBN 978-92-3-103467-1 Contents Historical survey

they had generally to place Sanjar's name plus his title of al-sultan al-mucazzam on their
coins before their own names and titles. At the outset, the only serious opposition to San-
jar' s claims came from Mahmud, but in 1119 Sanjar marched westwards with a powerful
army (whose commanders included, besides Sanjar himself, four vassal rulers), defeated
Mahmud near Sawa in northern Jibal and marched onwards to Baghdad. When peace was
made, Mahmud agreed to Sanjar's supremacy and was made the latter's heir (in the event,
he died long before Sanjar did), but he had to relinquish to Sanjar the Caspian provinces of
Mazandaran and Qumis and the town of Rayy, the key point for control of northern Persia,
and to agree to the reappointment of Sanjar's shihna (military governor) in Baghdad. 33
On Mahmud's death in 1131, his brothers Mascud, Toghril and Seljuq Shah success-
fully disputed the succession of Mahmud's young son Dawud, but were unable to agree
among themselves as to who should be sultan. They laid the question before Sanjar, as
senior member of the dynasty. Sanjar's favoured candidate was Toghril b. Muhammad,
but his preoccupation with events in Transoxania at this time (see below) prevented him
from providing Toghril with much military support. Toghril died soon afterwards in 1134,
allowing Mascud to succeed in the west and to reign for twenty years, the longest reign of
a Seljuq there since Malik Shah's time. Sanjar's last major intervention in the affairs of the
family in the west had been his defeat of Mascud at Dinawar in 1132, but thereafter affairs
in Khurasan and Transoxania increasingly claimed his attention. 34
Sanjar continued to exercise the overlordship over the Karakhanids of Transoxania first
imposed by his father Malik Shah (see above, Chapter 6), but had on various occasions to
intervene with his army across the Oxus against recalcitrant Khans. At Tirmidh in 1102
he had stemmed the invasion of a Karakhanid claimant, Kad'ir Khan Jibra'il or Jibril of
Balasaghun and Talas, placing on the throne in Samarkand Muhammad II Arslan Khan.
But towards the end of the latter's long reign, in 1130, Sanjar came with an army to rein-
force the Khan's faltering authority in Samarkand. Disputes occurred, with the Seljuq army
plundering part of the Karakhanid capital and with the sultan finally placing on the throne
his nominees: first, Hasan Tegin b. cAli; then briefly in 1132 Muhammad Arslan Khan's
brother, Ibrahim Tamghach Bughra Khan (who had been brought up at Sanjar's court); and
then, possibly in the same year, Muhammad Arslan Khan's third son, Mahmud. Mahmud
was Sanjar's nephew since his mother Terken Khatun, wife of Muhammad Arslan Khan,
was Sanjar's sister. The fortunes of Sanjar and Mahmud were to be closely interwoven
over the ensuing years; when Sanjar was captured by the Oghuz in 1153, Mahmud was

33
Koymen, 1954, pp. 5 et seq.; Bosworth, 1968, pp. 119-20, 135-7.
34
Koymen, 1954, pp. 174 et seq.; Bosworth, 1968, pp. 124-8.

168
ISBN 978-92-3-103467-1 Contents Historical survey

recognized by the Seljuq army in Khurasan as interim sultan of Khurasan and, after San-
jar' s death in 1157, likewise as legitimate ruler there until his own death in 1162. 35
As ruler of Khurasan, Sanjar was also concerned with the neighbouring great power
to his east, the Ghaznavids. They were ancient enemies of the Seljuqs during the period
when the Seljuqs were taking over Persia and western Afghanistan but had largely been
at peace with them since 1059 and the peace agreement - essentially one which divided
Afghanistan with a north-south line between the two imperial powers - made by the Seljuq
Chaghr'i Beg and the Ghaznavid Ibrahim b. Mascud I. Over the following decades, there
was a considerable Seljuq cultural penetration of the Ghaznavid lands, visible for instance
in numismatic patterns, titulature of the rulers and literary trends. The inaccessible and
largely independent mountain region of Ghur in central Afghanistan passed into the Seljuq
sphere of influence during the early part of Sanjar's reign after a raid into it by the sultan.
According to the Ghurid historian Minhaj-i Siraj Jlizjani, the Shansabani Malik of Ghur,
clzz al-Din Husayn, sent to Sanjar as annual tribute the specialities of the region, includ-
ing arms and armour and fierce dogs (see below, Chapter 8). The once-mighty Ghaznavid
empire was by now moving towards what it in fact became in its final years, essentially
a north Indian power rather than one of the eastern Iranian lands. A succession dispute
between Arslan Shah and Bahram Shah, the sons of the Ghaznavid sultan Mascud III b.
Ibrahim, allowed Sanjar to extend direct Seljuq suzerainty over the now somewhat trun-
cated Ghaznavid empire (see above, Chapter 5). On Arslan Shah's accession to the throne
in 1115, Bahram Shah had escaped to Khurasan and had appealed to Sanjar for help. The
Seljuq ruler marched eastwards with a formidable army, defeated Arslan Shah outside
Ghazna, despite the presence of awesome war elephants in the latter's army, sacked the
capital Ghazna and placed Bahram Shah on the throne in 1117. Bahram Shah agreed to
become a vassal of Sanjar, and to pay an annual tribute of 250,000 dinars and to place
Sanjar's name first in the khutba and on the coinage - the first time that the Seljuq khutba
had ever been heard in the Ghaznavid dominions. For some thirty years, Bahram Shah
acknowledged this subordinate status, only once becoming restive when in 529/1135 San-
jar and his other vassal, the Khwarazm Shah Ats'iz, came with their forces from Balkh
against Ghazna, expelling Bahram Shah to India before the latter returned and agreed to
reassume his vassalage. 36
Along the northern fringes of Khurasan, Sanjar found himself at odds for the first
time with another line of his vassals, the Turkish Khwarazm Shahs. The old lines of

35
Pritsak, 1953-4, pp. 49 -53; Koymen, 1954, pp. 158 -63; Barthold, 1968, pp. 319-22; Bosworth, 1968,
pp. 138-40.
36
Bosworth, 1968, pp. 157-9; 1977, pp. 89-100.

169
ISBN 978-92-3-103467-1 Contents Historical survey

Iranian Afrighid and Ma'munid shahs had been swept away by the Ghaznavids in the early
eleventh century; Mahmud and Mascud I of Ghazna had appointed Turkish slave comman-
ders from their own army, Altuntash and his sons, as governors there with the ancient title
of Khwarazm Shah. The Seljuqs continued this pattern of domination over Khwarazm:
jutting out as it did into the Central Asian steppes, the region was not only strategically
important to the sultans as a bastion against the pagan Turks of the Kipchak steppe but
it was also significant as the springboard for raids into the recruiting grounds for Turkish
auxiliary and slave troops. Sultans like Alp Arslan and Malik Shah had led punitive expe-
ditions into these steppe regions on various occasions, such as that of the first ruler in 1065
into the U styurt area and the Mang'ishlak peninsula to the east of the Caspian Sea against
the K'ipchak (see above, Part One).
In order to secure these important regions, Malik Shah had appointed the keeper of
the royal washing bowls (tast-dar), his slave commander Anush-tegin Gharcha'i, as tit-
ular governor at least in Khwarazm. During Berk-yaruk's reign, the sultan appointed in
1097 another Turkish ghulam, Ekinchi b. Kochkar, with the historic title of Khwarazm
Shah. When, in that same year, Ekinchi was killed, Berk-yaruk nominated in his stead
Anushtegin's son Qutb al-Din Muhammad as governor, and Muhammad's tenure of power
there (1097-1127) inaugurates the fourth and most brilliant line of hereditary Khwarazm
Shahs. This dynasty eventually built up, as the Seljuq empire in the east tottered to its
close, the most powerful and aggressively expansionist empire in the Persian lands, in the
end defeating their rivals for control of Khurasan, the Ghurids of Afghanistan, threatening
western Persia and Iraq and the cAbbasid caliphate itself, and only disintegrating under
the overwhelming military might of the Mongol invaders in the opening decades of the
thirteenth century.
Qutb al-Din Muhammad was a faithful vassal of Sanjar's, assiduous in attendance at
the Seljuq capital of Merv, but it was his son and successor cAla' al-Din Ats'iz (1127-56)
who was the real founder of the dynasty's might and splendour. He also attended San-
jar's court regularly, accompanying him, for example, on the campaign to Samarkand of
1130, and securing the northern and western frontiers of Khwarazm against the K'ipchak
and other marauders. But relations with Sanjar started to deteriorate as Ats'iz gradually
built up his own military strength and began identifying himself with the particular inter-
ests of his power base in Khwarazm. Sanjar later accused his vassal of indiscriminately
killing, together with pagans, Muslim ghazfs and murabitun (dwellers in ribats and fron-
tier fighters) in Mang'ishlak and at Jand on the lower Syr Darya (although this could pos-
sibly be an a posteriori justification for Sanjar's subsequent actions against Ats'iz). The
Khwarazm Shah rebelled openly in 1138, flooding lands in the Oxus valley to impede

170
ISBN 978-92-3-103467-1 Contents Historical survey

the advance of the Seljuq army, but this failed to halt Sanjar's progress. He defeated the
Khwarazmian army at Hazarasp on the Oxus, executed Ats'iz's son Atllgh, drove out Ats'iz
from Khwarazm, occupied the province and left a Seljuq prince there as governor accompa-
nied by an atabeg. However, as on earlier occasions when the attempts of an outside power
like the Ghaznavids to impose its authority on Khwarazm had provoked a national reaction
there, the Khwarazmian people now showed their resentment of alien domination. As soon
as Sanjar returned to Merv, they rose and expelled the Seljuq occupying troops. Meanwhile
Ats'iz returned from his refuge in Gurgan and took the offensive, invading Transoxania and
attacking the Seljuq garrison in Bukhara in 1139--40. 37
A new power now intervenes in the affairs of Central Asia: the Kara Khitay (the Kitan
or Liao of the Chinese sources), possibly of Mongol origin and certainly stemming from
the region of eastern Mongolia and northern China before they started to move westwards
and southwards into Semirechye and Transoxania (see below, Chapter 11 ). From their base
at Balasaghun, this pagan people began, under their Gtir Khan (Universal Ruler) Yeh-lti Ta-
shih, to attack the various Turkish tribes and amirs of western Turkistan, pagan and Muslim
alilke, inevitably coming up against the Karakhanids. Already in 1137 Mahmud Khan of
Samarkand had been defeated in a battle with the Kara Khitay at Khujand in Ferghana.
Four years later, internal disputes within the Western Karakhanid amirate led to one side in
the conflict, disaffected Kaduk tribesmen, calling in the Kara Khitay. Mahmud appealed to
his kinsman and suzerain Sanjar, who invaded Transoxania with a large army, but he was
defeated in 1141 by the Kara Khitay in a bloody battle on the Qatwan steppe of U srushana,
on the middle Syr Darya. Sanjar and Mahmud Khan fled to Khurasan, abandoning Tran-
soxania to the incomers, who went on to invade Khwarazm and to make Ats'iz their vassal.
Accordingly, while Sanjar's defeat was clearly opportune for Ats'iz, it seems improbable
that the Khwarazm Shah had, as several of the Islamic sources state, incited the Kara Khi-
tay to invade as an act of revenge on Sanjar for the sultan's killing of his son Atllgh. At this
point, Ats'iz himself raided into Khurasan, but was driven back by a Seljuq counter-invasion
of Khwarazni which penetrated to the capital Gurganj and compelled the Khwarazm Shah
to disgorge the treasuries which he had previously looted from Mervin 1143-4. Yet once
again, Khwarazm proved too hostile for the Seljuq troops to remain there. 38
The eventual downfall of Seljuq power in the east, however, did not result from the
attacks of external foes like the Kara Khitay or from the rebelliousness of ambitious
vassals like the Khwarazm Shahs, whose military strength was still inferior to that of
37
. Koymen, 1954, pp. 312-45; Kafesoglu, 1956, pp. 44-50; Barthold, 1962, pp. 126-7; 1968, pp. 323-6;
Bosworth, 1968, pp. 140-4.
38
Koymen, 1954, pp. 323 et seq.; Barthold, 1956, pp. 100-5; Kafesoglu, 1956, pp. 51-7; Barthold, 1968,
pp. 326-7; Bosworth, 1968, pp. 146-50; E/2 , 'Kara Khitay'.

171
ISBN 978-92-3-103467-1 Contents Historical survey

Sanjar - it was the result of an explosion of discontent within Khurasan itself, largely
caused by the policies of Sanjar's aides and officials there. Khurasan and the steppes
to the south-east of the Caspian Sea, in Gurgan and Dihistan, contained extensive pas-
ture grounds which supported numerous groups of nomadic, tribally organized Turkmens.
Some of these had probably been driven southwards into the Khurasanian fringes by recent
upheavals in the Central Asian steppes, including pressure from the Khwarazm Shahs and
the Kara Khitay. Others were descendants of the Oghuz tribesmen whose dynamic had
originally brought the Seljuqs to power in Khurasan a century before. The sultans had
accordingly always felt certain obligations towards these kinsfolk of theirs, often making
special administrative arrangements for them in the regions where they were particularly
numerous, appointing special shihnas and ru'asa' (sing. ra'fs) (chiefs) to act as channels of
communication between the nomads and the Seljuq state, whose dominating Perso-Islamic
ethos was largely alien to the Turkmens (see below).
These arrangements now came under severe strain because of the financial exigencies
arising from Sanjar's military adventures, which became increasingly expensive and elabo-
rate after 1135: he is said to have disbursed 3 million dinars for his Transoxanian campaign
of 1141, not counting the cost of gifts and robes of honour for various local potentates. The
burden of taxation in order to pay for these fell on sedentaries and nomads alike, but the
Oghuz in the upper Oxus regions of Khuttalan and Tukharistan finally rebelled against
the tax demands and the harsh collecting methods of the shihna over the Turkmens there,
the slave commander clmad al-Din Kumach of Balkh. Despite placatory approaches from
the Oghuz, Sanjar insisted on mounting punitive expeditions against them, but he was
twice defeated, forced to evacuate his capital Merv and finally captured by the nomads in
1153. The Oghuz bands swept through Khurasan, attacking the towns and showing partic-
ular violence and hostility towards members of the Seljuq administration and the religious
institution, closely linked with the Seljuq state. A general climate of insecurity was cre-
ated in both town and countryside, in which other anti-social elements, such as the cayyars
(bands of urban rowdies and vigilantes) took advantage of the breakdown of authority to
advance their own interests.
Sanjar was carried round by his Oghuz captors for three years, apparently in humili-
ating circumstances and enduring hunger and deprivation, until he managed to escape in
1156 to Tirmidh and Merv. But a year later, he died at the age of 71, and with him, the
authority of the Seljuqs in eastern Persia ceased; to his contemporaries it seemed like the
end of an epoch. Only in western Iran, essentially in Jibal, and in Kirman, did members
of the Seljuq family retain power for another thirty-eight and thirty years respectively.
The western Seljuqs were increasingly circumscribed, however, by the growth of Atabeg

172
ISBN 978-92-3-103467-1 Contents Historical survey

principalities in provinces like Fars, Azerbaijan, Arran and Armenia and by the later twelfth-
century upsurge of c Abbasid caliphal authority in Baghdad and Iraq, while the Kirman
Seljuqs fell victim to similar Oghuz pressures to those which had affected Khurasan and
which had caused Sanjar's downfall, involving Oghuz plunder raids on Kirman. 39
The Seljuq army in Khurasan had been left leaderless on Sanjar's capture. It now offered
the throne in Khurasan to the refugee Karakhanid Mahmud Khan, the son of Sanjar's sis-
ter Terken Khatun, hence with Seljuq blood in his veins; the Seljuq sultan in the west,
Muhammad II b. Mahmud, agreed to this and sent an investiture diploma. In fact, real
power in Khurasan was falling into the hands of the Seljuq princes, who over the next
years parcelled out the towns and districts of the province among themselves, especially
as Mahmud was unable firmly to establish his authority - he was even prepared at one
point to summon assistance against the Oghuz of Khurasan from the Khwarazm Shah Ats
'iz - and he died in 1162. Sanjar's former slave commander, Mu'ayyid al-Din Ay Aba (d.
117 4 ), had already risen to prominence in Nisha-pur. He established firm and just rule
in the town, being termed by the contemporary historian of Bayhaq, Ibn Funduq, as 'the
emperor of Khurasan, king of the East', and he eventually recognized Mahmud Khan as his
suzerain. Another of Sanjar's former slave commanders, Ikhtiyar al-Din Ay Tak, took pos-
session of Rayy and, as nominal vassal of the western Seljuq sultan Muhammad II, made
himself a power in northern Persia; while yet another former Seljuq general, Baha' al-Din
Toghril, took over Herat. These experienced and capable commanders were helped by the
Oghuz bands' own disunity and low level of political sophistication; the Oghuz could win
individual battles but could not establish a state. The rule of the Seljuq princes was only
curtailed in the 1170s by the westwards expansion of the Ghurid sultans, who then dis-
puted control of Khurasan with the Khwarazm Shahs and with the pagan Kara Khitay, who
had a foothold south of the Oxus in Balkh and Tukharistan. It was actually the Khwarazm
Shahs who triumphed there in the first decade of the thirteenth century, but they were only
to enjoy power for a short time before the whole dynasty was swept away by the incoming
Mongols.
The Khwarazm Shah Ats'iz remained loyal to Sanjar after the latter's capture, and on
his escape from the Oghuz sent Sanjar a message of congratulation; but Ats'iz died shortly
afterwards in 1156. He had by that date established the Khwarazmian state through his
skilful adaptation to the superior military strength of the Seljuqs and the Kara Khitay and
by the extension of his territories and his influence northwards into the K'ipchak steppe
as far as Mang'ishlak. This region was to be important as a reservoir of manpower for the

39
Koymen, 1947; Lambton, 1953, pp. 57-9; Barthold, 1968, pp. 329-30; Bosworth, 1968, pp. 152-5;
Lambton, 1968, pp. 246-7.

173
ISBN 978-92-3-103467-1 Contents Historical survey

Khwarazmian army when Ats'iz's successors embarked on grandiose policies of expansion

throughout the Iranian world and beyond to the Near East and to India; according to one
authority, Ats'iz even entertained the plan of conquering the town of Saksin on the lower
Volga in southern Russia. 40
Ats'iz's son Taj al-Dunya wa 'l-Din II Arslan (1156-72) had more potential freedom
of action after the death of his Seljuq suzerain Sanjar, but was at first largely preoccu-
pied with affairs in the lands north of the Oxus. Here, the Kara Khitay were his suzerains,
but these last were little disposed to interfere in Khwarazmian affairs provided that the
stipulated tribute was forwarded to the Gtir Khans at their ordo (military camp) in Semi-
rechye. The Khwarazm Shahs especially coveted the domains of the Karakhanids - who
were likewise vassals of the Gtir Khans - in Transoxania. Disputes within the Karakhanid
dominions between the Khans and contingents of their Karluk tribal followers culminated
in the murder of Mahmud Khan's brother Ibrahim III Tamghach Khan of Samarkand, by
the Karluk in 1156. When his successor c Ali b. Hasan Tegin Chaghn Khan took draconian
measures against the tribesmen, II Arslan seized the opportunity to invade Transoxania in
1158, ostensibly on behalf of the Karluk, and the Khan was compelled to take back the Kar-
luk chiefs into his service. II Arslan's troops also appeared in Khurasan against the former
Seljuq princes there, and lbn Funduq mentions the Khwarazmian army's presence at Bay-
haq and Nishapur in 1166-7, although full-scale Khwarazmian intervention in Khurasan
was only to come later. 41
It was the reign of c Ala' al-Din Tekish b. II Arslan (1172-1200) which brought the
Khwarazm Shahs to new heights of power. Khwarazmian arms vigorously combated
Ghurid attempts to annex Khurasan for themselves, succeeded in bringing about the final
demise of the Seljuq sultanate of western Persia and penetrated across Jibal and Kurdistan
to the borders of Iraq. Tekish, who had been governor of Jand, owed his throne to sup-
port from the Kara Khitay in the succession struggle with his brother Sultan Shah after II
Arslan's death. The outcome of this struggle was that Sultan Shah was driven southwards
from Khwarazm into Khurasan, where he attempted, with considerable success, to carve
out a principality for himself at the expense of the local Turkish commanders in towns
like Sarakhs, Nishapur, Tus and Merv. But once firmly on the throne in Khwarazm, Tekish
sought an opportunity to throw off the Kara Khitay yoke - comparatively light though it
was - by adducing the increasing harshness of the Kara Khitay tax-collectors and their

4
°
Koymen, 1954, pp. 424 et seq.; Kafesoglu, 1956, pp. 65-72; Barthold, 1968, pp. 330-1; Bosworth, 1968,
pp. 155-7, 185-7.
41
Pritsak, 1953-4, pp. 53-4; Kafesoglu, 1956, pp. 73-83; Barthold, 1968, pp. 332-7; Bosworth, 1968,pp.
185-8.

174
ISBN 978-92-3-103467-1 Contents Historical survey

continuing paganism. This made possible an appeal to the local Muslim religious classes
for support for the Khwarazm Shahs in the work of jihad (holy war).
The result was an alliance of Sultan Shah with the Kara Khitay, who tried to invade
Khwarazm but were blocked by the time-honoured expedient of the defenders opening
the canal barrages and dykes of the Oxus valley irrigation network, thereby flooding the
terrain. Later, however, in the 1190s, Tekish was to seek Kara Khitay aid in his struggle
with the Ghurids, for the Kara Khitay hoped to regain control of Balkh and Tukharistan,
formerly tributary to them. These hopes were dashed by a decisive victory of the Ghurids
over the Kara Khitay on the banks of the upper Oxus, allowing the Ghurids to occupy the
towns of Khurasan and to install at Merv Tekish's grandson and rival for power, Hindu
Khan b. Nasir al-Din Malik Shah (1198-1200).
Many of Tekish's diplomatic and military efforts were to be devoted to the situation
north of the Oxus and in the steppes and then, towards the end of his reign, to expansion
westwards into northern Iran, with Iraq as his ultimate goal. In general, he aimed at con-
ciliating the Turkmens of the steppes to the north and west of Khwarazm, partly through
marriage alliances between members of his own family and the local Khans' families; thus
Tekish's wife Terken Khatiin, mother of his son and successor c Ala' al-Din Muhammad,
was from either the Kangll or the Yemek Turkish tribe and was the daughter of Kipchak
Khan. Tekish drew on these Turkmens as troops for his forces, even though many of them
were still pagan, and it was these unassimilated barbarians who made the Khwarazmian
army a byword for cruelty and violence among the peoples of northern and western Iran.
But diplomacy did not always work, and at least one punitive expedition by Tekish against
the Kipchak in the vicinities of Jand and S'ignak on the lower Syr Darya is recorded (1195
-6). In Transoxania also, there is mention of an expedition to Bukhara in 1182 and possibly
another one with the same goal in 1198 (reported by the chronicler Ibn al-Athir, but whose
historicity was nevertheless doubted by Barthold), when the local population of Bukhara
supported the Kara Khitay garrison until the town was stormed by the Khwarazmians. 42
The disappearance of the Seljuqs from Khurasan and the shrinkage of the territory con-
trolled by the last Seljuq sultans in western Iran, hemmed in as they were by various Atabeg
principalities such as that of the Ildenizids of Azerbaijan and Arran and the Salghurids of
Fars, led Tekish to dream of establishing a Khwarazmian empire which would dominate
Iran proper as well as the lands north of the Oxus. The sultan who was to be the last of the
Seljuq line in Persia, Toghrll III b. Arslan, was a young, vigorous and effective comman-
der, who by 1192 had made firm his control over Jibal and had defeated the Atabeg ruler
in Rayy, Kutlugh Inanch. But the latter now appealed to Tekish for help, thus giving the

42
Kafesoglu, 1956, pp. 84-108, 113-21; Barthold, 1968, pp. 337-46; Bosworth, 1968, pp. 188-92.

175
ISBN 978-92-3-103467-1 Contents Historical survey

Khwarazm Shah a pretext to intervene in the affairs of Iran. Tekish came to Rayy with an
army and demanded that his name should be placed in the khutba of western Persia imme-
diately after the claiph's name, but had to return to Khurasan in the face of a threatened
invasion by Sultan Shah. Toghril was accordingly able in 1193 to clear the Khwarazmian
garrison out of Rayy. But when Tekish returned the following year, Toghril, having refused
to negotiate with the shah, was defeated outside Rayy and killed in 1194; henceforth, the
Seljuq dynasty survived only in Anatolia as the sultanate of Rum, centred on Konya.
Tekish now occupied the whole of Jibal and parcelled it out as iqtifs for his comman-
ders. The proximity of the powerful and aggressive shah to the cAbbasid territories in Iraq
disquieted the caliph al-Nasir ( 1180-1225). He nevertheless deemed it prudent to bow to
Tekish's evident military superiority and sent the shah an investiture patent for the sul-
tanate of Iraq (i.e. clraq cAjami, or western Iran), Iran (i.e. the remainder of the country,
essentially Azerbaijan, Fars, Khurasan and Kirman) and Turkistan. Tekish was active in
northern Iran, especially against the Ismacilis of Daylam, but died in 1200; on hearing the
news, the people of Jibal rose and massacred all the Khwarazmians they could find. 43
The new shah, Tekish's son cAla' al-Din Muhammad (1200-20), continued his father's
anti-caliphal policy. At the end of his life, Tekish had demanded that his son's name
be placed in the khutba at Baghdad, but subsequent preoccupations in the east with the
Ghurids, the Kipchak and the Kara Khitay prevented Muhammad from enforcing his-
claimed rights against al-Nasir. Not until 1217, the very eve of the Mongol invasions, was
Muhammad ready to march westwards. By now, he knew from captured diplomatic cor-
respondence that al-Nasir had in the past incited the Ghurids against him and had tried to
have him assassinated by the Ismacilis. Muhammad was aware that his anti-caliphal poli-
cies would deprive him of support from the Sunni majorities in Iran and Iraq. Hence he now
adopted a pro-Shfite policy, obtaining a fatwa (legal opinion) from compliant cuzama' that
al-Nasir was unfit to rule and that the cAbbasids had usurped the caliphate from the house
of cAll, and proclaiming a Sayyid as anti-caliph. His army proceeded to cross the Zagros
mountains barrier down to the plains of Iraq, but snow-storms of unparalleled severity held
it up in Kurdistan, and news of the appearance of the Mongols on the eastern fringes of his
empire caused Muhammad to abandon his plans to overthrow the cAbbasid caliphate. 44
During the earlier part of his reign, Muhammad's eastern policy involved conciliating
the Kara Khitay in order to leave him free to eject the Ghurids from Khurasan, where
Ghurid rule had proved unpopular. After the death of Mucizz al-Din Muhammad Ghurl
43
Kafesoglu, 1956, pp. 123-44; Barthold, 1968, pp. 346-8; Bosworth, 1968, pp. 181-3; Mason, 1972,pp.
99-100, 102-4; Hartmann, 1975, pp. 75-8.
44
Kafesoglu, 1956, pp. 199-205, 214- 21; Bosworth, 1968, pp. 183-4; Mason, 1972, p. 104; Hartmann,
1975, pp. 76-85.

176
ISBN 978-92-3-103467-1 Contents Historical survey

in 1206, the power of the Ghurids declined perceptibly. The new sultan in Ghur, Ghiyath
al-Din Mahmud, had to acknowledge the Khwarazm Shah as his suzerain and place his
name in the khutba and on the coinage. Khwarazmian control was now imposed over all
the towns of Khurasan, and the Bawandid local ruler in the Caspian provinces was made
a Khwarazmian vassal. Muhammad could therefore now dispense with Kara Khitay sup-
port, especially as the latter were distracted by the rebellion in Semirechye of the Naiman
Mongol chief Ktichltig and the revolt of the Kara Khitay's Muslim vassals in East Turk-
istan. Securing the northern frontiers of Khwarazm by a successful campaign against the
Kipchak, Muhammad turned to Transoxania and allied with the last Karakhanid ruler of
Samarkand, cuthman Khan b. Ibrahim, and other local magnates who were discontented
with Kara Khitay financial demands, against their suzerains. The fact that the Kara Khi-
tay were distracted by Ktichltig's revolt meant that the Gtir Khan was unable to main-
tain his occupation of Samarkand (probably in 1209-10) and was defeated near Talas
by the combined forces of cuthman Khan and cAla' al-Din Muhammad. By this time,
an advance force of Chinggis Khan's Mongols had appeared in northern Semirechye; the
Gtir Khan died; and Muslim Khwarazmian rule was established throughout Transoxania.
Yet this domination speedily proved unpopular. The people of Samarkand rose against
the Khwarazmians there and slaughtered them. cuthman Khan attempted to renew his
connections with the Kara Khitay, but brought down on himself the Khwarazm Shah's
wrath. This culminated in a general massacre of Karakhanid family members in 1212,
thereby extinguishing the western branch of the dynasty almost completely (a branch seems
to have persisted in Ferghana, with its capital at Uzgend, for some years more: see above,
Chapter 6).
cAla' al-Din Muhammad asserted his power against the Kipchak and incorporated S'ig-
nak into his empire, but he was less successful against Ktichltig, who had taken over most
of the former Kara Khitay territories. Even after his Talas victory over the Naiman chief,
Muhammad was unable to bring relief to the Muslim population of the town of Balasaghun
in Semirechye or to protect Ferghana; and he was equally impotent to protect the Muslims
of East Turkistan or Kashgharia against Ktichltig's fiercely anti-Muslim policies there. 45
The defeat of Ktichltig by Chinggis Khan's forces in 1218, his flight into Badakhshan
and his death there only postponed the hour of reckoning for the Khwarazm Shah. There
had been relations between him and Chinggis in 1215, when cAla al-Din Muhammad
had sent an embassy to the Mongol Khan in northern China, and an indecisive military
encounter in (?)1219 between Muhammad's forces and Chinggis' eldest son JOchi in the

45
Pritsak, 1953-4, pp. 56-9; Barthold, 1956, pp. 106-9; Kafesoglu, 1956, pp. 187-93, 223-9; Barthold,
1968, pp. 349-70; Bosworth, 1968, pp. 192-5.

177
ISBN 978-92-3-103467-1 Contents The structure of the eastern Seljuq state

steppes to the north of the Aral Sea. But the immediate cause of large-scale intervention in
Trans-oxania by the Mongols was Muhammad's ill-advised execution of Chinggis' envoys
at Utrar, which brought down on his head the full violence of Mongol military might in the
later part of 1219. The ensuing years saw the speedy disintegration of the Khwarazmian
empire, culminating in the death in Diyar Bakr in 1231 of the refugee last Khwarazm Shah,
Muhammad's son Jalal al-Din(?) Mingburnu; for all these events, see below, Chapter 12.

The structure of the eastern Seljuq state

As noted above, Sanjar was originally a Malik subordinate to the Great Seljuq sultan Berk-
yaruk and then Muhammad b. Malik Shah. After the latter's death, Sanjar acquired, by
seniority, headship of the main line of Seljuqs in Persia and Iraq, with the title of a/-
sultan al-muc azzam appearing on his coins minted at Isfahan and Rayy from 1118 onwards.
Attached to his personal court, which was usually in the capital Merv, Sanjar had a fully
developed administration headed by the dzwan-i acza (Supreme Ministry). It was presided
over by a series of viziers, of whom eight are known, from Shihab al-Islam c Abd al-Razzaq
(in office 1117-22) to Nizam al-Mulk Hasan. Naturally, high officials like viziers were nor-
mally Persians or Arabs, though from 1122 to 1124 Sanjar had a Turkish vizier, Muham-
mad b. Sulayman Kashghari Yighan or Toghan Beg. The vizier's colleagues included the
usual array of officials concerned with accounting and financial checking procedures, the
mustawfi(chief accountant) and the mushrif (inspector); those manning the chancery, under
the tughra'f (chief secretary), combining the departments of the tughra (official insignum
on documents) and of insha' (correspondence); and the carid, the head of the dfwan-i card
(department of the army).
Administrative and military control over a town or province of the empire was exercised
by the sultan's appointee, the shihna, who was concerned with public security and the
regulation of crime. Various religious and judicial officials, including the qadf Uudge) and
the khatfb (preacher at the Muslim worship), were appointed directly by Sanjar. All these
figures - as also the commanders of the army, mainly military slaves or freedmen - were
supported materially by the system of iqtacs, with the military holder of such an estate
additionally being liable for military service; often, no duration is mentioned for these
tenures, although one cannot thereby conclude that the grants were necessarily for life,
still less that they were hereditary.
Provincial government (wilayat, iyalat) under Sanjar was exercised through centrally
appointed officials with designations like waif, na'ib, shihna or ra'fs. The last of these had
an additional role as the representative of the urban notables - from the leading families

178
ISBN 978-92-3-103467-1 Contents The structure of the Khwarazmian state

of whom the ra'fs was generally chosen - and of his fellow-townsmen vis-a-vis the central
government, above all in matters relating to taxation. From Sanjar's reign, we possess
documents on the nomination of provincial governors for Gurgan and its dependencies,
Mazandaran, Rayy, Balkh and its dependencies, Merv, Tus and Dihistan. There were also
domains which belonged personally to the sultan, i.e. crown lands (amlak-i khass, amlak-i
khalisat), with their own special administration, headed by a wakfl (intendant), and their
own financial organs. 46
The relations of Sanjar and his amirs with the Oghuz nomads, who were an appreciable
element of the population in Khurasan, northern Jibal and the Caspian provinces, have
already been touched upon, and special arrangements were made for the government of
those areas where they predominated. The Oghuz felt that they had a particularly close
relationship with the sultan, with a right of approaching him directly. At the time when
Sanjar's governor in Balkh, clmad al-Din Kumach, was increasing his harsh demands on
them (see above), the Oghuz protested, according to Rawandi, the historian of the Seljuqs,
'We are the specially close subjects [raciyyat-i khass] of the sultans and we do not come
under the jurisdiction of anyone else.' Towards the end of Sanjar's reign, the numbers of
these Oghuz in Khurasan seem to have increased, and the problem of how to find a place for
them in the administrative and social structure of the Seljuq empire became acute; Sanjar's
failure to solve it was a major factor in the decline and disappearance of the eastern Seljuq
empire by the end of his reign. 47

The structure of the Khwarazmian state

Material on the internal structure and administrative system of the Khwarazmian state of
the line of Anushtegin is sparse. It stems mainly from the volumes of insha' literature,
comprising official letters, investiture diplomas, and so on, stemming from authors who
worked under the Khwarazm Shahs like Baha' al-Din Muhammad Baghdad!, who was
head of the dfwan-i ins ha' for cAla' al-Din Tekish and who left us an important collection,
the Kitab al-Tawassul ila 'l-tarassul; and, above all, the famous littreateur Rashid al-Din
Muhammad Watwat (d. ?1182-3), who served Ats'iz, 11 Arslan and finally Tekish as chief
secretary and who compiled two collections of rasa'il (epistles), the cAra'is al-khawatir
and the Abkar al-ajkar.
It is a reasonable assumption that the central and provincial administrations of the
Khwarazm Shahs, who were themselves originally provincial governors for the Great

46
Lambton, 1957; Horst, 1964; Lambton, 1968, pp. 247-69; Klausner, 1973.
47
Lambton, 1957; 1968, pp. 245-7.

179
ISBN 978-92-3-103467-1 Contents The structure of the Khwarazmian state

Seljuqs, were modelled on those of the Seljuqs in Iran and Iraq. The hub of the admin-
istration was undoubtedly the dfwan-i ac la headed by the vizier, with its ancillaries, the
chancery, the office of the mustawfiand that of the carid-i lashkar (inspector-general of the
army), as noted above. In the middle decades of the twelfth century the dfwan-i insha' was
headed by the great poet and stylist, Rashid al-Din Watwat. Only towards the end of the
dynasty's existence did cAla' al-Din Muhammad decide -in c. 1218 according to Nasawi-
to dispense, within the territories that he controlled (his estranged mother, Terken Khatun,
was the real ruler in Khwarazm itself), with a vizier. He appointed instead, as an execu-
tive body, a council of six (sitta min al-wakfldariyya), who included the wakfl of the khass
(crown domains) and the katib al-insha' (chief secretary). The accelerated course of events
after this year, with the irruption of the Mongols into Transoxania and beyond, does not
allow us to estimate how these new administrative arrangements, involving a more direct
personal management by the Khwarazm Shah, would have worked.
As mentioned above, provincial government was in the hands of waifs or shibnas
appointed by the Khwarazm Shah: those of Farah (in northern Sistan), Turkistan and
Bukhara are mentioned under 11 Arslan, and those of Jand, Barjanllgh-kent (= Barjllgh-kent
on the Syr Darya between Jand and S'ignak) and its dependencies, Gurgan, Dihistan and
Khwarazm itself under Tekish. The region of Jand, the base for Khwarazmian raids into
the K'ipchak steppe, was especially important and was often entrusted to the Khwarazm
Shah's eldest son and heir. One of the functions of the waif was the organizing of corvee
labour for state requirements (shahkar, bfgar). As in Seljuq Khurasan, the ra'fs was the
channel of communication with the local populations, and the appointment of ru 'asa' by
the Khwarazmian chancery is recorded for Dihistan, Gurgan and Khwaf in Kuhistan. 48
The historical sources say little of daily life and the condition of the ordinary people
at this time. They report widespread devastation in the Khurasanian towns by the Oghuz
towards the end of Sanjar's reign. The reports of thousands of casualties may be exagger-
ated, but the destruction of several mosques, madrasas and important libraries in Nisha-
pur is attested, such as that of the Sabuniyya madrasa, where a copy of the celebrated
100-volume compendium of Qur' anic sciences, made nearly two centuries before for the
Saffarid amir Khalaf b. Ahmad, perished, according to the biographer al-Samcani. Simi-
lar losses continued in the ensuing disorders in Khurasan even after Ay Aba had restored
some order. Khwarazm itself continued to flourish agriculturally at this time, while ethni-
cally becoming more and more Turkicized, so that the indigenous Iranian Khwarazmian
language shrank until it apparently disappeared in the fourteenth century. The geographer
and traveller Yaqut, writing just after the Mongol devastation of Khwarazm, stated that

48
Kafesoglu, 1956, pp. 205-14; Horst, 1964.

180
ISBN 978-92-3-103467-1 Contents The structure of the Khwarazmian state

when he had been in Gurganj in 1219, he had never seen a richer or fairer city in the world
than the Khwarazmian capital; and he found the Khwarazmian countryside extraordinarily
fertile, filled with settlements which had markets and an abundance of food. Soviet exca-
vations there seem to show an extension of cultivation based on irrigation canals during the
twelfth century. 49
As the Khwarazm Shahs built up their realm into what was, by the end of the twelfth
century, the most powerful empire of the eastern Islamic world, the pace of centraliza-
tion within the state increased. The urgent need for money to finance military expansion
led to much hardship and disaffection, certainly outside Khwarazm itself. Wherever the
Khwarazmian armies penetrated, they established a reputation for violence and extor-
tion which made them highly unpopular and a focus for popular hatred; in none of the
provinces they conquered did the Khwarazm Shahs ever succeed in creating a bond of
interest between themselves and their subjects. In their sporadic attempts to expand into
Transoxania at the expense of the Karakhanids and Kara Khitay, the Khwarazm Shahs were
nevertheless often at pains to conciliate the powerful orthodox Sunni religious authorities
in the Transoxanian towns. At Bukhara, for instance, an influential line of local cuzama'
the Al-i Burhan, rose to power in the first half of the twelfth century as ru'asa' of the
city, holding also the religious title of sudar (sing, sadr; 'eminences'). At times, they held
more authority within the town than the Karakhanid secular rulers, and in 1207, Burhan
al-Din Muhammad b. Ahmad, called Sadr-i Jahan, actually collected taxation there on
behalf of the Kara Khitay, as the real rulers in the town. When Tekish was at Bukhara in
1182, he acknowledged in his edicts the spiritual authority of the religious leaders there;
and although, when Tekish and c Ala' al-Din Muhammad embarked on their anti-caliphal
policies in western Iran, they could no longer count on Sunni Muslim religious support,
within Transoxania they were able to pose as the defenders of Islam against the pagan
Kara Khitay. 50

49
Tolstov, 1953, pp. 295-317; Bulliet, 1972, pp. 76 -81.
50
Elr, 'Al-e Borhan'.

181
ISBN 978-92-3-103467-1 Contents The region of Ghur and lslamization ...

THE GHURIDS*
K. A. Nizami

Contents

The region of Ghur and the beginnings of Islamization 182

The rise of the Ghurids as an independent power . 185
The apogee of the Ghurid sultanate . . . 186
The Bamiyan amirate . . . . . . . . . . 189
The Ghurid sultanate as a world power . 190
The political and social organization of the Ghurid state . 191
Cultural developments . . . . . . . . . . . . . . . . . . 194

The region of Ghur and the beginnings of lslamization

The mountainous region situated to the east and south-east of Herat and the south of
Gharchistan and Guzgan was known as Ghur. It comprised the basins of the upper Hari
Rud, the Farah Rud, the Rud-i Ghur and the Khash Rud together with the intervening
mountain chains. 1 Geographic configurations had a profound influence on historical and
cultural developments in Ghur. Geography led to the fragmentation of political power as
the entire region could not be controlled from one centre. Each fortress exercised inde-
pendent sway over the area immediately under its control and patriarchal traditions struck
deep roots. It was as late as the time of Qutb al-Din Muhammad (d. 1146-7) that a por-
tion of Ghur - the petty principality of Warshada on the Hari Rud - developed a capital
at Firuzkuh. 2 Ghur had no compact or continuous areas of habitation, but only scattered

* See Map 5.
1
Kohzad, 1951-4.
2
Vercellin, 1976, pp. 337-40.

182
ISBN 978-92-3-103467-1 Contents The region of Ghur and lslamization ...

population pockets. The landscape was studded with fortified places and towers where
people could defend themselves. 3 Cultural movements in the neighbouring areas had only
a peripheral impact on the region. The waves of Muslim conquest touched the fringes of
Ghur several times during the Umayyad period but the region did not come under Islamic
cultural influence. As late as the tenth century, the people of the region were said to be
'bad-tempered, unruly and ignorant' .4
The Iranian dialect of the people inhabiting Ghur differed from the dialect of Khurasan.
During his campaign into Ghur in 1020, Prince Mascud of Ghazna had to employ local
interpreters to communicate with the people. Since they pronounced the name of the
Prophet Muhammad as 'Hamad', they became known as Hamadis after their conversion to
Islam. The ethnic background of the Ghurid people is shrouded in myth and legend. After
the Saffarid invasions of Zamin-Dawar and Bust, the region became exposed to tribes and
peoples of different ethnic backgrounds. Contact with Ghazna led to the infiltration of
Turkish tribes from the surrounding areas. Later on, Ghuzz and Khalaj ethnic elements set-
tled on the fringes of the region, gradually breaking its cultural isolation and diversifying
its ethnic composition.
The extension of Islam and its cultural institutions, and the conversion of Ghur, took a
long time. As late as the end of the tenth century, the population of Ghur was for the most
part heathen. According to the geographer al-Istakhri, it was the biggest pagan enclave
within the borders of Islam. 5 It was probably as the result of missionary activity from
Khurasan that the movement of the Karramiyya, 6 a pietistic and ascetic form of Sunni
Islam especially strongly represented in Nishapur, was established in Ghur in the course
of the tenth and eleventh centuries. It was followed by the Shansabani Maliks, including
in the later twelfth century the brothers Ghiyath al-Din Muhammad and Mucizz al-Din
Muhammad, until they later switched to the mainstream Shafici and Hanafi law schools
of Sunni Islam. The nature of the imperfect conversion is best illustrated by the fact that
sometimes the names were Muslim but the people led the life of pagans. The anonymous
author of the Hudud az-cazam [The Limits of the World] (c. 982) says, 'In the days of old
this province of Ghur was pagan (kafir); now actually most of the people are Muslims.' 7 As
the years passed, however, three centres came into prominence in the valley of Hari Rud:
Firuzkuh (the capital of Shansabani power), Jam and Chisht.

3
Bosworth, 1961 , p. 118.
4
Anon., 1937, p. 110.
5
Bosworth, 1961 , pp. 120-1.
6
For the Karramiyya, see Barthold, 1968, pp. 289-90; Bosworth, 1960; El2 , 'Karramiyya'.
7
Anon., 1937, p. 110.

183
ISBN 978-92-3-103467-1 Contents The region of Ghur and lslamization ...

The mountains of Ghur had a large number of iron-ore workings and those of other met-
als. According to Togan, the entire mountain region from Ghur and Kabul to the land of the
Karluk was metal-working. 8 It exported armour, weapons and war equipment to neighbour-
ing areas. The chief fortress of Ghur was known as Pul-i Ahangaran (Bridge of the Black-
smiths). The Ghaznavids and the Seljuqs exacted tribute from Ghur in the form of arms,
cuirasses and the ferocious watchdogs bred locally. Sultan Masciid of Ghazna employed
Ghurid officers as specialists in siege warfare. While it was still pagan, Ghur supplied
slaves to the markets of Herat and Sistan. 9 It was widely known for horse-breeding. The
region of Ghur thus possessed two of the most important requisites of war in the Middle
Ages - horses and iron - and the Ghurids took full advantage of them.
According to Ghurid legendary tradition, Zahak was the first ancestor of the Shansabani
dynasty. Faridun and Shansab, the eponymous founder of the dynasty, were said to be
descendants of Zahak. While no systematic account of the early history of Ghur is avail-
able, it appears that the penetration of Islam was a slow process. The Arab historian al-
Tabari refers to a campaign in 667 by al-Hakam b. cAmir, the governor of Khurasan, and
Ibn al-Athir records details of an expedition undertaken against Ghur in 725. The purpose
of these occasional incursions seems to have been to obtain slaves and booty, and no per-
manent implantation of Islam resulted.
The legendary tradition, as expressed by the early thirteenth-century historian of the
Ghurid dynasty Jiizjani in his Tabaqat-i Nasirf, our prime source for the entire history of the
dynasty, holds that a Shansabani prince, Amir Banji, subsequently came to prominence. He
was the ancestor of all the Shansabani amirs who occupied the Ghurid lands, and secured
legal sanction for his authority from the caliph Harlin al-Rashid, who conferred upon him
a covenant and a standard and gave him the title of qasfm amfr al-mu'minfn (Partner of the
Commander of the Faithful). Jiizjani is silent about the successors of Banji until the advent
of Amir Suri, who in the later ninth century came into conflict with the Saffarids. Yacqub b.
Layth, the Saffarid ruler of Sistan, conquered Zamin-Dawar, Bust and Rukhkhaj, but Ghur
was saved by the inaccessibility of its mountains.
In c. 979 the Samanid overlord of northern and eastern Afghanistan, Nuh b. Mansur,
dispatched a force to conquer Ghur, but except for a few forts, no significant conquests
could be made. When Sebtiktegin was governor of Ghazna and Zabulistan on behalf of
the Samanids (977-97) (see above, Chapter 4), he made several attacks on Ghur. After
some initial set-backs, he established his authority in eastern Ghur and was recognized
as suzerain by Muhammad b. Suri of Mandish. After the death of Sebtiktegin, Muhammad

8
Togan, 1936,pp.33-4.
9
Bosworth, 1961 , p. 118.

184
ISBN 978-92-3-103467-1 Contents The rise of the Ghurids ...

withheld the payment of tribute, plundered caravans and blackmailed the subjects of Sultan
Mahmud in the neighbouring provinces.
In 1011 Sultan Mahmud sent Altuntash and Arslan Hajib, governors of Herat and Tus
respectively, on an expedition into Ghur. The Ghaznavid forces marched on Ahangaran.
Muhammad b. Suri entrenched himself in inaccessible hills and ravines, but the Ghaznavid
army routed the Ghurids, and Muhammad b. Suri and his son Shith were taken prisoner.
Another of Muhammad b. Suri's sons, Abu cAli by name, had remained on good terms
with Sultan Mahmud when his own father was at loggerheads with him, so that Mahmud
rewarded Abu cAli by placing Mandish under him. Mahmud thus brought eastern Ghur
under his control. Then in 1020, he sent his son Mascud, at that time governor of Herat, to
subdue Nab, the north-western part of Ghur; according to the Ghaznavid historian Bayhaqi,
Mascud was the first to penetrate to the interior of this part of Ghur.
Amir Abu cAli was, however, overthrown at some date in the 1030s by his nephew
cAbbas who established himself in Ghur. Some notables of Ghur approached Ibrahim b.
Mascud of Ghazna, who marched on Ghur with a large army. As soon as his army appeared,
the forces of Ghur went over to him and cAbbas was handed over to Ibrahim. He was
removed and the government of Ghur was placed in the hands of his son, Muhammad, who
now regularly paid the tribute to his Ghaznavid overlords. During the time of Muhammad's
son, Malik Qutb al-Din Hasan, tribal conflicts created chaos in Ghur; it was Hasan's son
and successor Malik clzz al-Din Husayn (1100-46) who restored peace and order in the
region. While Sanjar was Seljuq ruler of the eastern Iranian lands, he fought against the
ruler of Ghur and took clzz al-Din Husayn prisoner. Later, however, the sultan sent him
back to his native land; and thereafter, clzz al-Din Husayn regularly sent tribute - which
included war equipment, armour and the finely bred guard dogs of Ghur - to Sanjar in his
capital at Merv.

The rise of the Ghurids as an independent power

The period of Shansabani expansion began with the seven sons 10 of clzz al-Din Husayn.
They divided their patrimony among themselves and consolidated their authority in and
around Ghur, so that the history of Ghur as an imperial power begins with them in the mid-
twelfth century. Fakhr al-Din Mascud founded the dynasty of the rulers of Bamiyan and
Tukharistan; Baha' al-Din Sam became the amir of Ghur and Firuzkuh; Sayf al-Din Suri
established himself at Ghazna; Sultan cAla' al-Din Husayn became the sovereign over
Ghur, Ghazna and Bamiyan; Shihab al-Din Muhammad Kharnak established himself at

10
Habib and Nizami (eds.), 1970, pp. 153-5.

185
ISBN 978-92-3-103467-1 Contents The apogee of the Ghurid sultanate

Macdin; and Shujac al-Din cAli became the amir of Jarmas. The seventh son Qutb al-Din
Muhammad's adoption of the title of malik al-jibal (Lord of the Mountains) was the first
expression of his ambitions. He founded Firuzkuh and built a fortress there, while Sayf
al-Din Suri made Istiya his capital.
Relations between the brothers became strained, however. Qutb al-Din Muhammad
went over to Bahram Shah of Ghazna (1117-57), but he was poisoned there, 11 leading to a
war of revenge between the Ghurids and the Ghaznavids and the savage sacking of Ghazna
(see below). When Qutb al-Din had left for Ghazna, Baha' al-Din Sam came to Firuzkuh
from his territory of Sanga and gave orders for the construction of strong fortresses in Ghur,
the Garmsir, Gharchistan and the mountain tracts of Herat. He further married a daughter
of Malik Badr al-Din of Kidan, also of the Shansabani family; she was the mother of the
later sultans Ghiyath al-Din Muhammad and Mucizz al-Din Muhammad.
Baha' al-Din Sam had, when he set out for Ghazna, entrusted Ghur to his brother cAla'
al-Din Husayn. The latter, on hearing of his brother Qutb al-Din Muhammad's death in
1146-7, set out towards Ghazna to accomplish what his brother had not been able to
achieve. Bahram Shah alerted the troops of Ghazna and Hindustan. The two armies met
and the Ghurids employed their defence tactics of the karwab (a screen made of raw bul-
lock hides with both sides stuffed with cotton and used as a protective wall). The stratagem
worked: Bahram Shah's son Dawlat Shah was killed, and the army of Ghazna was routed.
cAla' al-Din took Ghazna by storm, setting fire to it for seven nights and days, and at Bust
he destroyed the palaces and buildings of the Mahmudi (i.e. Ghaznavid) dynasty. He thus
came to be known as Jahan-Siiz (Incendiary of the World). Mahmud's Ghazna, the 'bride
of cities', and many of its fine buildings and libraries, disappeared in a bloodbath.

The apogee of the Ghurid sultanate

Malik Baha' al-Din Sam's sons, Ghiyath al-Din and Mucizz al-Din, now come into promi-
nence. Ghiyath al-Din adopted a policy of fraternal co-operation and eventually gave the
title of sultan to his brother Mucizz al-Din and to his nephew Malik Shams al-Din Muham-
mad of Bamiyan, son of Malik Fakhr al-Din Masciid. He also established contact with
the cAbbasid caliph in Baghdad, whose moral support added to his prestige. cAla' al-Din
Husayn had imprisoned the two brothers, but his son Sultan Sayf al-Din had set them
free. Ghiyath al-Din rendered great service to Sayf al-Din in dealing with the Oghuz (or

11
According to one historian, Qutb al-Din had once supported the previous ruler in Ghazna, Arslan Shah,
against his brother Bahram Shah. See Khan, 1949, pp. 44-5.

186
ISBN 978-92-3-103467-1 Contents The apogee of the Ghurid sultanate

Ghuzz), and when Sayf al-Din died, the amirs and Maliks of Ghur and Gharchistan gave
their allegiance to him.
On Ghiyath al-Din's accession to the throne in Firuzkuh in 1163, Mucizz al-Din was
entrusted with the territories of Istiyan and Kajuran. The two brothers lost no time in devis-
ing a stratagem to kill their rival, Abu '1-c Abbas, who was supported by the refractory ele-
ments of Ghur. His death strengthened Ghiyath al-Din's position, but then his uncle Malik
Fakhr al-Din Mascud coveted the throne of Firuzkuh. However, he was unable to achieve
military dominance over the two brothers, who eventually allowed him to return to his
principality of Bamiyan. Ghur was now firmly in the hands of Ghiyath al-Din, who then
extended his power southwards into Zamin-Dawar. After the death of Taj al-Din Yildiz, a
military slave of the Seljuq Sultan Sanjar, the leaders and notables of Herat invited Ghiyath
al-Din, and the latter established himself there also. Later, he married a daughter of Sul-
tan cAla' al-Din Husayn in order to consolidate his family position. When Mucizz al-Din
returned from campaigns in Sistan, Ghiyath al-Din made over Teginabad (i.e. the region
of Kandahar) to him. Ghiyath al-Din then dispatched exploratory raiding parties to Kabul,
Zabul and Ghazna, which the Oghuz had wrested from the last Ghaznavid ruler Khusraw
Malik (1160-86), by driving him to Lahore; Ghazna was recovered in 1173. Ghiyath al-Din
placed Mucizz al-Din on the throne of Ghazna and himself returned to Firuzkuh.
Two years later, in 1175, the armies of Ghur and Ghazna advanced to Herat and occu-
pied it, also conquering Pushang. These successes so enhanced Ghiyath al-Din's prestige
that the Nasrid Malik of Nimruz (i.e. Sistan) became his vassal. The Oghuz Maliks of Kir-
man also submitted to him. Ghurid control was now extended over much of Khurasan and
the sultan's name was pronounced from the pulpits and inscribed on the coinage there.
Ghiyath al-Din was at the height of his prestige when he became involved with the
Khwarazmian claimant Sultan Shah, who had reached his court after having been driven
out of his lands by his brother, the Khwarazm Shah Tekish. The conflict between the
Ghurids and members of the Khwarazm Shah's family went on for several months,during
which Mucizz al-Din defeated Sultan Shah. On Tekish's death in 1200, Ghiyath al-Din
and Mucizz al-Din occupied Nishapur and assigned it to Malik Diya' al-Din, son of Abu
cAli and son-in-law of Ghiyath al-Din. In 1200 Merv was taken and Malik Nasir al-Din
Muhammad Kharnak installed there; Sarakhs was assigned to Taj al-Din Zangi, the son of
Malik Fakhr al-Din Masciid.
Ghiyath al-Din died in 1202 aged 63 and was buried in Herat, but he was survived for
four more years by his brother Mucizz al-Din. The latter had won two historic battles in
1192, one on the banks of the River Murghab, which led to the rout of the Khwarazm

187
ISBN 978-92-3-103467-1 Contents The apogee of the Ghurid sultanate

Shah, and the second at Tara' in in Panjab, which opened the gates of northern India for the
Ghurid armies. The Ghurid conquests now extended as far as the frontiers of Kashmir.
The Ghurid incursions into India had begun in 1175 when Mucizz al-Din marched
towards Multan and overthrew the renascent Carmathians there. He then occupied Uchch
and in 1178 led an army into Gujarat against the Hindu ruler of Nahrwala (the Anhilvada of
Indian geography), who had a formidable army. He defeated the Ghurid forces at Kayadra
near Mount Abu and made their retreat extremely difficult. Thereafter Mucizz al-Din
changed his plans and decided upon a thrust through Panjab, which had been in the hands
of the last Ghaznavid sultan, Khusraw Malik. In 1176 Peshawar was taken. In 1182 Mucizz
al-Din marched against Daybul in Sind and conquered the whole area up to the sea coast;
the Sumera ruler there acknowledged his suzerainty. The conquest of Lahore was com-
pleted in 1186 after three successive expeditions and Khusraw Malik was induced, under
the protection of a treaty, to surrender. He was treacherously put to death by the Ghurids,
along with his son Bahram Shah, thereby ending some two centuries of Ghaznavid power.
All the strategic areas which provided the Ghurids with a springboard into India were now
in the hands of Mucizz al-Din.
A confrontation with the Chahamanas, who ruled the territory extending from Ajmer
to Delhi, now followed. In 1191 Mucizz al-Din besieged and captured Bhatinda. The
Chahamana ruler Prithvi Raja appeared there to recover the fortress, however, and in a bat-
tle fought at Tara'in in 1191, Mucizz al-Din was utterly defeated and seriously wounded.
A Khalaj soldier rescued him from the battlefield and helped him to reach Ghazna, while
Prithvi Raja invested the fortress of Bhatinda and recaptured it after 13-months' siege. 12
Mucizz al-Din refused to take this defeat as final. After making preparations extending
over a whole year, he returned with a force of 120,000 cavalry. This time he defeated
his Chahamana adversary. Govinda Raja was killed and Prithvi Raja captured, so that the
Chahamana kingdom now lay at the Ghurid ruler's feet. Important military points, like
HansI, Kuhram and Sarsuti, were occupied and garrisoned, and the whole of the Siwalik
territory was brought under control. From Peshawar to Hansi, the entire region was now
under the control of Mucizz al-Din.
In 1196 Bhima Deva, the ruler of Nahrwala, endeavoured to retrieve Ajmer from Ghurid
control. The Ghurid commander Qutb al-Din Aybak was besieged in Ajmer for several
months, but Mucizz al-Din dispatched a relieving force. Bhima Deva retreated to Gujarat,
but Aybak pursued him towards Nahrwala and routed his forces. Bhima Deva managed to
escape, but thousands of his soldiers were put to the sword or taken prisoner. The victory

12
For the Indian campaigns of the Ghurids, see Habib and Nizami (eds.), 1970, pp. 132-90; Nizami, 1961 ,
pp. 75-88.

188
ISBN 978-92-3-103467-1 Contents The Bamiyan amirate

at Tara'in was a major triumph for the Ghurids in India. Their general Aybak occupied
Meerut, Baran and Delhi in 1192. Soon afterwards Mucizz al-Din again came to India and
conquered Thankar and Vijayamandirgarh. The ruler of Gwalior accepted his suzerainty. In
1197 Aybak conquered Badaon and in 1199-1200 Malwa. Two other Ghurid generals, the
Turkish slave commander Baha' al-Din Toghril and Muhammad Bakhtiyar KhaljI, played
a significant role in the extension of Ghurid authority in India. Toghril consolidated the
possessions of Gwalior and Bayana, while Muhammad Bakhtiyar Khalji supplanted the
Gahadavala chiefs and carried the Ghurid banners into Bihar and Bengal. Emboldened by
his victories, he pushed ahead towards the Himalayas and Tibet, but this proved a disaster
in his otherwise successful career.
Mucizz al-Din was assassinated in India in 1206 and his Turkish slave generals - Qutb
al-Din Aybak, Taj al-Din Yildiz and Nasir al-Din Qubacha-now rose to prominence. They
speedily quarrelled among themselves, however, so that Ghazna became detached from the
Ghurid possessions. According to the official Ghurid historian Fakhr-i Mudabbir, Aybak
was formally invested with viceregal powers in 1206 and was appointed walf az-cahd (heir
apparent) by his master, Mucizz al-Din. For some three years he had to content himself with
the positions of Malik and sipahsalar (commander-in-chief), 13 at first clinging loyally to
his background of service to the Shansabanis, but he then was able to establish his own
authority in India when it became apparent that the unity of the Ghurid empire had been
irretrievably shattered by the Khwarazm Shahs.

The Bamiyan amirate

The Shansabanis established their control over Tukharistan and the mountain tracts of
Bamiyan soon after their successes in Ghazna. As noted above, this region was known
for its treasures and its mines of gold, silver and precious stones. cAla' al-Din Jahan-Suz
installed his eldest brother, Malik Fakhr al-Din Mascud, there in 1145. Subsequently, the
latter's son Shams al-Din Muhammad established his authority in Balkh, Chaghaniyan,
Wakhsh, Jarum, Badakhshan and the hill tracts of Shughnan, acquiring from Ghiyath al-
Din the title of sultan for himself also. His son Baha' al-Din Sam, who succeeded him in
1192, was highly respected by scholars and literati alike; Fakhr al-Din al-Raz!, the famed
Muslim philosopher, Qur' anic commentator and theologian, was associated with his court
for many years. Baha' al-Din Sam became the focus of Shansabani hopes, and the Maliks
and amirs of Ghur, Ghazna and Bamiyan, who now looked upon him as their head, invited
him to Ghazna; but he fell ill and died, almost simultaneously with the assassination of

13
Epigraphica Indo-Moslemica, 1911-1 2, p. 2.

189
ISBN 978-92-3-103467-1 Contents The Ghurid sultanate as a world power

Sultan Mucizz al-Din in 1206. Thereafter, Shansabani fortunes in their own homelands
began to wane.
The Ghurid army commanders invited cAla' al-Din and Jalal al-Din, the sons of Baha'
al-Din Sam, to Ghazna in order to occupy the throne. Jalal al-Din placed his brother on the
throne and himself returned to Bamiyan. The fabulous treasures of Ghazna were divided
between the two brothers, but during the following years, they quarrelled over the pos-
session of Bamiyan. Finally, Sultan cAla' al-Din Muhammad Khwarazm Shah marched
against Jalal al-Din and had him put to death in 1215, thus extinguishing the Ghurid
dynasty in its homelands of Afghanistan.

The Ghurid sultanate as a world power

At the end of the twelfth century and the beginning of the thirteenth, the Ghurid state
extended from Herat in Afghanistan to Lakhnawti in Bengal and touched the borders of
Tibet and Kashmir. Barthold rightly observes that, 'the brothers Ghiyath al-Din and Shihab
al-Din ... raised their kingdom to the rank of a world power' .14 They had to deal, with
varying degrees of success, with a number of powers - the Ghaznavids, the Seljuqs, the
Oghuz, the Kara Khitay, the Khwarazm Shahs, the Hindu Chalukyas and Chahamanas,
and the Khokars, among others - taking full advantage of the decline of the Seljuqs and
the Ghaznavids. Following the death of Seljuq Sanjar in 1157, they struggled hard, but
ultimately unsuccessfully, for the control of Khurasan. 15 When Ghurid power in its own
homelands declined, their Indian acquisitions flourished and paved the way for the emer-
gence of the Delhi Sultanate. This was to be the most powerful state in northern India until
1526, when Babur replaced it by the Mughal empire (see below, Chapter 14). Thus Ghurid
Dehli became the repository of the Muslim culture of Central Asia.
The decline of the Seljuqs and the Ghaznavids, noted above, created a political vacuum
that the Khwarazm Shahs and the Ghurids struggled to fill. Ghurid ambitions in Khurasan
were blocked by the Khwarazmians, who also coveted the province and who, in the long
run, could bring greater military resources to bear in the struggle. Ghiyath al-Din sought
the Khwarazm Shah, Tekish's help against the latter's rival Sultan Shah, but the Ghurids
had to deal single-handed with Sultan Shah. In 1190-1, however, Sultan Shah and his
Turkish ally Toghril of Herat were defeated and Herat was annexed to Ghur. Sultan Shah
died the following year and his possessions in northern Khurasan were annexed by Tekish,
so the latter for a while controlled the whole of Khurasan. His advance into western Iran

14
Barthold, 1968, p. 338.
15
Ibid., p. 339.

190
ISBN 978-92-3-103467-1 Contents The political and social organization

worried the cAbbasid caliph, who sent emissaries to the Ghurid sultan urging immediate
war against the Khwarazm Shahs. The opportunity came in 1200 when Tekish died, his
enterprise in western Iran was aborted and his son cAla' al-Din Muhammad succeeded
to power in Khwarazm. In 1201 Ghurid troops entered Khurasan and captured Nishapur,
Merv, Sarakhs and Tus, reaching as far as Gurgan and Bistam. Kuhistan, a stronghold of the
Ismacilis, was plundered and all Khurasan was brought temporarily under Ghurid control.
This success proved to be short-lived: cAla' al-Din Muhammad recovered Nishapur and
other Ghurid acquisitions, and Herat came under his control in 1201. He was, however,
anxious for peace with the Ghurids so that he could combat the threat from the steppes
of the Kara Khitay (see below, Chapter 11 ). The Ghurids, for their part, were anxious to
recover Khurasan. They occupied Herat and other towns, but cAla' al-Din Muhammad
took advantage of Mucizz al-Din's brief absence from Herat due to the death of his brother
Ghiyath al-Din in 1202, defeated the Ghurid army besieging Merv and relieved the city.
In retaliation Mucizz al-Din invaded Khwarazm and besieged the capital of cAla' al-Din
Muhammad, but had to retreat. The Kara Khitay pursued him and inflicted a crushing
defeat on the banks of the Oxus near Andkhud (modern Andkhoy) in 1204. Mucizz al-Din
managed to reach his capital Firuzkuh safely, but Andkhud was a disaster for the Ghurids,
who now retained only Herat and Balkh of their conquests. Mucizz al-Din did not lose
heart, and was planning a full-scale invasion of Transoxania when developments in Panjab
attracted his attention. He ordered his kinsman, the ruler of Bamiyan, to prepare for the
campaign and to arrange the construction of a bridge over the Oxus, but he could not
undertake this campaign as he was assassinated in 1206 at Damyak, while on his way back
to Ghazna.
The death of Mucizz al-Din heralded the end, a few years later, of his Ghurid empire
which had spanned the Hindu Kush. Not long afterwards, his nephew and successor at
Ghur, Ghiyath al-Din Mahmud, had to acknowledge the suzerainty of the Khwarazm Shah.
When Mahmud died, the Shansabani lands were absorbed into the Khwarazmian empire,
and the dissolution of Ghurid power was complete when Mucizz al-Din's governor, Yildiz,
was driven out of Ghazna.

The political and social organization of the Ghurid

state
The Ghurid political authority, initially organized on a patriarchal basis in an inaccessible
mountain region, slowly acquired some of the features of a state and briefly became one of
the greatest empires of the Islamic Middle Ages, stretching from Herat in Afghanistan to

191
ISBN 978-92-3-103467-1 Contents The political and social organization

western Bengal in India. Its diversity at the height of its power is an interesting sociological
phenomenon. The Firuzkuh area was essentially patriarchal, with strong tribal traditions;
the Ghazna region had for two centuries nurtured the traditions of the Turco-Iranian monar-
chy; and Hindustan was under a decentralized feudal system of government. The rise of
the Kara Khitay in Transoxania and the Oghuz of Khurasan added a new dimension to
the situation. 16 Thus the Ghurid empire, from Ghur to Lakhnawti, comprised a multiplic-
ity of cultural traditions. The Ghurids, with their political skill, used this cultural situation
to their advantage so that when the Mongols devastated Central and Western Asia, the
Ghurids' Indian acquisitions became the repositories of their cultural heritage. Thus it was
in early thirteenth-century Lahore that the litterateurs Muhammad cAwfi and Fakhr-i Mud-
abbir worked.
Like other Sunni powers of the eastern Islamic world, the Ghurids, beginning with
Sultan Ghiyath al-Din, thought it expedient to establish contact with the cAbbasid court
and receive further confirmation of their authority. Al-Mustadi' (1170-80) and al-Nasir
(1180-1225) both granted robes of honour to Ghiyath al-Din and the imperial nawbat (mil-
itary band salute) was introduced five times a day; all these acts of recognition enhanced
the sultan's prestige.
The idea of a capital could arise only when some sort of integration had been effected
among the tribal pockets of power and when the network of castles and towers had acquired
a level of administrative cohesion. The fact that Firuzkuh only emerged as a Ghurid capital
in the time of Qutb al-Din Muhammad shows the protracted nature of efforts necessary
for this. The title of sultan was assumed by Amir Suri after his accession to the throne
of Ghazna, where the institution of monarchy had earlier taken a new shape under the
Ghaznavid Sultan Mahmud. The usual titles in Ghur had been the modest ones of amir
and Malik. The extension of authority beyond Ghur necessitated the creation of a winter
capital, and one in the warm region of Zamin-Dawar was adopted for this purpose.
Tribal traditions, ethnic considerations and the exigencies of the situation all influenced
Shansabani principles of succession. Malik Fakhr al-Din Mascud, though the eldest of the
seven brothers, was not allowed to occupy the throne of Ghur because his mother was
of Turkish origin. The two brothers Ghiyath al-Din and Mucizz al-Din simultaneously
enjoyed the title of sultan, and Mucizz al-Din accepted the seniority of his brother, who
was known as al-sultan al-aczam (Supreme Sultan). After Ghiyath al-Din's death, Mucizz
al-Din came to be called al-sultan al-aczam. The Shansabani amirs shared political power
and often worked in collaboration, under the over-lordship of a leading member of the fam-
ily. That the vassal-master relationship was frequently under strain reveals, however, the

16
Habib and Nizami (eds.), 1970, pp. 185-7.

192
ISBN 978-92-3-103467-1 Contents The political and social organization

fragile nature of the system. The vassals were expected to pay regular tribute and inscribe
the name of the suzerain on the coins; otherwise, they exercised all authority in their ter-
ritories. At a time when political loyalties were frequently opportunistic, the system of
vassalage ensured some sort of political collaboration, if not loyalty. The administrative
arrangements visualized by Mucizz al-Din for his Indian acquisitions seem to have com-
prised three or four local commanders who were independent of each other, but subject
to himself. Later, the institution of the iqtt{ (revenue assignment) developed in India and
helped the integration of feudal units into a central organization. 17
The Turkish military slave Qutb al-Din Aybak showed respect for legal forms and tradi-
tion when he waited for a letter of manumission from Ghiyath al-Din Mahmud, on whom
the legacy of Mucizz al-Din had devolved. Both Aybak and another commander after him,
Iltutmish, firmly demonstrated that they were not prepared to share political authority with
anybody and stood for a centralized power in northern India. They recast Ghurid politi-
cal traditions in the light of the Indian situation, and when the cAbbasid caliph granted a
manshar (patent of authority) to Iltutmish, the Ghurid possessions in India achieved recog-
nition as an independent political entity.
Militarily, the Ghurids had certain advantages. First, as noted above, they had iron and
horses in abundance. The Ghaznavids appreciated their production of arms and the Indians
hailed them as asvapatis (Lords of the Horse). 18 Second, the Ghurid sultans could dis-
pose of a nucleus of bellicose Ghuri and Khalji tribesmen from the core of their empire,
Afghanistan. Their numbers were limited, however; hence they had to be supplemented
by purchasing Turkish military slaves, presumably stemming ultimately from the Inner
Asian steppes. With all these forces, the sultans were able to make headway in northern
India against the strenuous resistance of Rajput and other military elements in the armies
of the Indian princes. However, the sultans chose to fight on two fronts, one in northern
India and the other in Khurasan and Central Asia. In the long run, they were not able to
sustain prolonged warfare in both spheres of action; and in the second sphere, they were
at a numerical disadvantage compared with their enemies the Khwarazm Shahs, who could
call upon vast reserves of Turkish manpower from the steppes around Khwarazm and
beyond. Hence the Ghurids failed to make permanent conquests in Khurasan and even-
tually lost even their heartland of Ghur to the Khwarazm Shahs; but their commanders,
as epigoni of the Ghurid sultans, successfully laid the foundations of the first large-scale,
dynamic implantation of Islamic political and military control in the Indus-Ganges plains

17
SeeNizami, 1961 ,pp.128-31.
18
Nizami, 1961 , p. 82.

193
ISBN 978-92-3-103467-1 Contents Cultural developments

of northern India, an achievement of lasting significance for the history of the subcontinent
(see below, Chapter 14).
The institution of the slave household assumed importance under Mucizz al-Din, who
treated his slaves as his sons, and in course of time they became the linchpin of post-Ghurid
organization in India. 19 Government machinery in the earlier period was confined to the
management of essential government functions, but when Ghazna came under Ghurid con-
trol, it was natural that the administrative institutions as developed by the Ghaznavids
should be adopted. A certain number of features of the Seljuq administrative system were
also taken over. Thus in India the Ghurid, Ghaznavid and indigenous Indian traditions
coexisted.
The vizier was the head of the civil administration. He had no judicial functions but
had a supervisory jurisdiction over the army. The qadf al-qudat (supreme judge) was the
head of the judiciary, with numerous subordinate qadfs, including a qadf for the army.
In India the office of sadr-i jahan (or sadr al-sudur) looked after religious affairs. The
most important officer of the household was the amfr-i hajib, the master of ceremonies at
court who conducted notables and officials to the royal presence. During the time of Sultan
Nasir al-Din Mahmud of Delhi, the office of wakfl-i dar is also mentioned. The amfr-i
shikar was the chief huntsman. The sar-i jandar commanded the king's bodyguards. The
sipahsalar was the supreme commander of the forces, and in India the actual organization
of the army and its commissariat was the responsibility of the carid-i mamalik. As noted
above, the Ghurid armies were multi-ethnic, so that the army of Qutb al-Din Aybak in India
comprised Turks, Tajiks of various kinds and also locally recruited Indian soldiers.

Cultural developments
Ghur lacked any urban life until a comparatively late date. It was contact with Ghazna, the
hub of the intellectual world on the eastern fringes of Islam, which initiated the Ghurids
into the cultural life of Iran and Central Asia. Amir Abu cAli ordered the construction of
many public buildings, including mosques and madrasas. Malik cAbbas built numerous
fortress-like villages in Ghur. Qutb al-Din Muhammad founded the fortress and city of
Firuzkuh. Baha' al-Din Sam erected strong fortresses in Ghur, the Garmsir, Gharchistan
and Herat, keeping strategic needs in view. A castle constructed at Wadawajzd by Sultan
Ghiyath al-Din was so impregnable that it survived the onslaught of the Mongols. The
remains of a Ghurid madrasa, constructed in 1165-76, have recently been unearthed by
archaeologists in Gharchistan. It 'exhibits numerous parallels to many structures of the

19
Habib and Nizami (eds.), 1970, pp. 194-7.

194
ISBN 978-92-3-103467-1 Contents Cultural developments

twelfth century in western Khurasan, Central Asia, Ghazna and Sistan'. 20 The discovery
by Andre Maricq in 1957 at Jam in Ghur of what are possibly the minaret and citadel of
Firuzkuh has also thrown valuable light on Ghurid architectural traditions.
The earliest Shansabani ruler to take any interest in academic pursuits was Amir cAbbas,
who was interested in astrology and raised a lofty castle with twelve towers for his astro-
logical studies. His son, Amir Muhammad, extended his patronage to men of culture and
learning. Sultan Ghiyath al-Din founded many institutions for the ShafiCis. Baha' al-Din
Sam was respected for his patronage of scholars: according to the Ghurid historian Jiizjani,
'there was no Muslim sovereign who was a greater cherisher of learned men'. As men-
tioned previously, Fakhr al-Din al-Razi was associated with Baha' al-Din Sam's court for
a considerable time and wrote his Risala-yi Baha'iyya in his name, while during the time
of Ghiyath al-Din he wrote another treatise entitled Lata'if-i Ghiyathiyya.
The religious life of the people of Ghur passed through interesting phases: as noted
above, the pietistic sect of the Karramiyya was influential for many years. Initially, the fol-
lowers of the Karramiyya had received encouragement in Khurasan from Sebtiktegin and
Mahmud of Ghazna. But contact with Ghazna, Herat and other centres of Muslim culture
slowly changed the religious complexion of Ghur and its adjoining territories, and during
the course of the twelfth century, the Shansabanis started to abandon their patronage of the
Karramiyya. There were a number of encounters with the Karramiyya leaders, who were
strongly opposed to Fakhr al-Din al-Razi but who had a considerable popular following in
Ghur. In the end, Ghiyath al-Din adopted the ShafiCi school of law, while Mucizz al-Din
became a Hanafi.
The Ghurids destroyed Ghazna, but in India their role was more constructive. 21 The
intellectual heritage of Central Asia, both in the form of scholars and of books, reached
India during this period and flourished under Ghurid patronage. Thus the Ghurid occupa-
tion of northern India had a social and cultural significance in the broader framework of
Central Asian history.

20
See Casimer and Glatzer, 1971 , pp. 53-68.
21
SeeNizami, 1961 ,p. 85.

195
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

THE UIGHURS, THE KYRGYZ AND THE

TANGUT (EIGHTH TO THE THIRTEENTH
CENTURY)*
D. Sinor, Geng Shimin and Y. I. Kychanov

Contents

THE UIGHURS IN MONGOLIA AND THE KYRGYZ 196

THE UIGHUR KINGDOM OF KOCHO . . . . . . . . . . . . . . . . . . . . . . . . 206
THE TANGUT HSI HSIA KINGDOM (982- 1227) . . . . . . . . . . . . . . . . . . 212

Part One

THE UIGHURS IN MONGOLIA AND THE KYRGYZ

(D. Sinor)

The first mention of the Uighurs (under the name Hui-ho and various graphic variants)
appears in Chinese sources and refers to the early seventh century A.D. when this people
lived on the banks of the Selenga river and was subjected to the Turks. The Chinese viewed
the Uighurs as the descendants of the Hsiung-nu, the dominant power on the steppe from

* See Map 2.

196
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

about 200 B.c. to A.D. 48. This may indeed have been the historical truth, but there is no way
to substantiate it, since the Chinese sources tend to ascribe Hsiung-nu origin to any of the
numerous steppe peoples with whom the Chinese had contacts over succeeding centuries.
It is thus not surprising that the description given of the Uighurs by the Chin T' ang-shu
[Old T'ang Annals] follows the traditional pattern of characterization of the steppe peoples:
They have no chiefs, no permanent dwellings; they wander in search of water and pasture.
These men are of an evil disposition and cruel. They are excellent riders and archers and most
rapacious. Brigandage is their livelihood.

More specific, and certainly trustworthy, is the information that the Uighurs were subjects
of the Turks, who relied on them 'to govern the wild northern regions' .1
Around 630, at a time when Turk power was on the wane, under the leadership of
P'u-sa the Uighurs became more assertive. Although the Chinese characters used to write
his name are identical with those transcribing the Buddhist term boddhisatva, there is no
other indication that he or his family were Buddhists. On P'u-sa's death his son T'u-mi-tu
assumed the title of Kaghan; he was murdered in 648. His son P' o-juan (for him, as for
his father, we have names only in Chinese transcriptions) died some time between 661 and
663, leaving the reins of government to his sister who, defeated by the troops of the Chinese
emperor T'ai-tsung, disappears from the stage. Nothing is said about the activities of the
four Uighur chiefs whose names are known for the period between 680 and 741; their
people, together with other tribes of the T'ieh-le confederation to which they belonged,
lived within the borders and under the sway of the Eastern Turk empire.
In 7 44 the Kaduk, Basmil and Uighur tribes formed an alliance to overthrow Turk
rule. The victorious coalition was first headed by the Basmil Alp Bilge Kaghan, but he was
soon eliminated by an Uighur-Karlukjoint action. Shortly afterwards, it was the turn of the
Kaduk to be ousted, but they were still a force to be reckoned with and remained hostile
to the Uighurs. The action undertaken by the three peoples should be labelled 'revolt'
rather than 'invasion'. The first Uighur rulers considered themselves continuers of the Turk
tradition, and claimed legitimacy by linking themselves with Bumin Kaghan, the founder
of the First Turk empire. The difference separating Turks from Uighurs must have been
purely political. As is clearly shown by the inscriptions commemorating the deeds of their
great men, Turks and Uighurs spoke the same language, used the same runic-type script
and lived within the same geographic boundaries. Were it not for their name, the Uighurs
would be indistinguishable from the Turks. The Uighur state in Mongolia was, in fact,
the Third Turk empire. Its first ruler, Kutlugh Bilge Ktil Kaghan (744-7), and his son and
successor El-Etmish Bilge Kaghan (747-59), also known as Bayan Chur, made a point
1
Chavannes, 1903, p. 87.

197
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

of cultivating friendly relations with China. Uighur troops were instrumental in the T'ang
reconquest in 757 of Lo-yang, which had fallen into the hand of the rebellious An Lu-shan.
A year later, in recognition of his services, the Kaghan received for his wife the Princess
Ning-kuo, daughter of Emperor Su-tsung.
We might pause here for a moment to direct our attention to a process of urbanization
gaining strength among the Uighurs. The Shine-usu inscription - written in Uighur in 759
or 760 and celebrating the deeds of El-Etmish Bilge Kaghan - mentions two Uighur cities.
One of these, located at the confluence of the Orkhon and Ballkllg rivers, was the Kaghan' s
residence. It is usually referred to by its recent Mongol name, Karabalghasun. Its founda-
tion probably goes back to the times of the Turk empire. The Shine-usu inscription also
reports that El-Etmish Bilge Kaghan entrusted some Chinese and Sogdians with the build-
ing at the Selenga river of a city called Bay Ballq. The city of the Uighur Kaghan, most
likely Karabalghasun, was described in some detail by the Arab traveller Tamim b. Bahr,
who visited it, probably in 821. 2 Located within a conurbation which included cultivated
tracts, the town itself had twelve huge iron gates and must have been a bustling place. 'The
town is populous and thickly crowded and has markets and various trades,' writes Tamim b.
Bahr about the city, though he does not give its name. The majority of its inhabitants were
Manichaeans. One cannot establish with any certainty what prompted the Uighur rulers to
engage in the building of cities. Of course, both of the two dominant foreign influences -
Chinese and Sogdian - must have favoured this development, but, for reasons of prestige,
the Kaghan might have felt the need for an official, fixed place of residence. He himself
- so we are told - might have preferred to stay in a tent, though admittedly a luxurious
one. Some sixty years after the decision to build Bay Ballq had been taken, Tamim b. Bahr
would find the Kaghan's 'golden tent' at a distance of some 30 km from Karabalghasun.
From the marriage of El-Etmish Bilge Kaghan and Princess Ning-kuo issued Bogu
Kaghan (759-79), possibly the central figure in Uighur history. Following in his father's
footsteps, he gave substantial aid to the Chinese, once more recovering on their behalf the
city of Lo-yang in 762, though at a price: the devastation wrought upon the city by the
unruly Uighur troops. The story of their behaviour in China, which they were supposed
to help, is a miserable record of brutality and destruction. It is to be observed that the
Uighurs now held a position of vantage in regard to the Chinese empire that none of the
other nomadic powers of Mongolia ever occupied. If the emperor was able to cope with his
internal foes, it was because he could rely on the Uighurs.
Without any doubt, the most important event of Bogu Kaghan's stay in Lo-yang was his
conversion by Sogdian religious to Manichaeism. In his subsequent actions, Bogus Kaghan

2
On Tamim b. Bahr's journey, see Minorsky, 1948.

198
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

displayed the zeal usually shown by recent converts. An Uighur Manichaean text gives a
highly idealized picture of the enthusiasm with which the Uighurs are said to have accepted
their ruler's announcement of his conversion:

At that time when the divine Bogti Kaghan had thus spoken, we the Elect and all the people
living within the land rejoiced. It is impossible to describe this our joy. The people told the
story to one another and rejoiced. 3

The most troublesome aspect of the Kaghan's conversion was that it resulted in the Sog-
dians gaining overwhelming influence in matters of policy. The above-mentioned text
attributes to the Kaghan the promise: 'If you, the Elect, give orders, I will follow your
words and requests.' Such an attitude might have justified the title zahan-i Mani (Emana-
tion of Mani) given to Bogti Kaghan in a Pahlavi fragment, 4 but was unlikely to impress
either the shamans who felt that their influence was being threatened, or indeed the rank
and file of a Turkic people of warriors who resented the arrogant meddling of a bunch of
foreigners in the affairs of their state. Bogti Kaghan had fallen under the influence of Sog-
dians who were more interested in their own prosperity than in that of the state. According
to the Hsin T'ang-shu [New T'ang History], the Sogdians' 'property flourished and they
accumulated a very large amount of capital'. Ultimately, they overplayed their hand by
attempting to induce Bogti Kaghan to invade China. When he remained impervious to
the arguments put forward by his uncle Tun Bagha Tarkhan against such an adventurous
undertaking:

Tun Bagha became annoyed and attacked and killed him and, at the same time, massacred
nearly two thousand people from among the kaghan's family, his clique and the Sogdians. 5

Uighurs and Sogdians had developed a commensalism that, most of the time, benefited
both as long as the latter did not overstep the rules of prudence dictating that a parasite
should not exploit its host to the point of death, for the destruction of the host entails its
own death. Commensalism presupposes a moderation in greed by the parties involved that
was not always displayed by the Sogdians. Following Tun Bagha' s coup d'etat in 779, some
of them (one wonders whether these were the real culprits) had to pay dearly for their past
actions, but the severe measures taken against the Sogdians produced only a temporary
eclipse of their role. They were soon back, wheeling and dealing, acting as intermediaries
between Uighurs and Chinese, self-appointed diplomats representing their own interests
first of all. It should be remembered that, although Sogdians were instrumental in the spread

3
See Bang and Gabain, 1929, p. 8, lines 52-6.
4
Muller, 1912.
5
Mackerras, 1972, p. 89.

199
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

of Manichaeism among the Uighurs, international trade was typically their principal field
of activity.
As with so many other features of Uighur political life, symbiosis with the Sogdians
was inherited from the Turks. It will be remembered that under the First Turk empire, the
Sogdians attempted to secure for themselves the monopoly of the silk trade, and with this
aim in view they were ready to send to war none other than the Turks. There is no evidence
that under the Uighurs attempts were made to sell silk directly to Byzantium, but it seems
unlikely that the vast quantities obtained from the Chinese were absorbed by the internal
market. To obtain silk, two ways were open for the Uighurs, and they made use of both.
The most marketable commodity produced by the Uighur economy was, of course, the
horse, for which there was a permanent demand among the Chinese military. The Uighurs
were ready to provide the mounts in exchange for silk - a conventional transaction. But
they were not satisfied with supplying a genuine want; they had recourse to what must
be called forced trade, foisting on to the Chinese more horses than were needed by them
and of lesser quality. In most instances, the Chinese sources give only the number of silk
ligatures paid to the Uighurs but rarely indicate the horse/silk ratio. The Hsin T' ang-shu
gives the following picture of this pseudo-trade as practised under Emperor T' ai-tsung:

The Uighurs took even greater advantage of their services to China by taking as a price forty
pieces of silk for every horse they brought in as tribute. Every year, they sought to sell several
tens of thousands of horses . . . The horses were inferior, weak and unusable. The Emperor
gave them [the Uighurs] generous presents, wanting by this means to shame them, but they
did not recognize this. They came again to the capital with 10,000 horses, but the emperor
could not bear to place this burden on his people once again, so he paid for only 6,000 of
them. 6

There is reason to think that, though heavy, the burden was not as unbearable as the Chinese
wanted the Uighurs to believe. According to their own records, between 780 and 829 the
Chinese paid 2,012,000 pieces of silk to the Uighurs. 7 If we reckon 40 pieces of silk per
horse, the silk delivered by the Chinese would have sufficed for the purchase of only 50,300
horses over a period of almost half a century. Yet we know that trade was much more brisk,
since under T' ai-tsung the number of horses to be imported officially was set at 10,000 per
annum. 8 The discrepancy may be explained by the Chinese habit of being in arrears with
the payments, a practice greatly resented by the Uighurs. According to Uighur reckoning in

6
Mackerras, 1972, p. 87.
7
Mackerras, 1969, pp. 238-9.
8
Jagchid, 1989, p. 177.

200
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

781, the Chinese owed them 1,800,000 pieces of silk, corresponding to the price of 53,000
horses. 9
An unpleasant and by no means negligible concomitant of the horse-silk trade was the
overbearing, arrogant, unruly behaviour of the Uighur and Sogdian merchants in China. It
led to numerous conflicts and, among the officials as well as the people, it left a lasting
dislike of these greedy barbarians. It was to manifest itself to the detriment of the Uighurs
and Sogdians once they ceased to be shielded by the might of the Uighur empire.
Traditionally, the Kaghan's leadership was dependent on what he could deliver to his
followers, but none of the coveted goods were home-produced; they had to be obtained
from or in China, either through loot or trade. Another way to satisfy the 'insatiable greed'
of his entourage was to allow them to participate in the lavish embassies that were sent to
the T'ang court with or without proper justification. In less than 100 years - between 745
and 840-116 such embassies came to China on the pretext of 'rendering tribute' ;10 some of
them comprised more than 1,000 members, among them the wives of the more important
dignitaries. The jockeying for a place in these diplomatic missions must have been intense.
To some extent, the visits of privileged Uighurs and Sogdians to the splendour of the T'ang
court were counter-productive; they whetted rather than satisfied the participants' appetite
for luxury goods.
Let us now return to the reign of Tun Bagha Tarkhan, who ruled under the name or title
Alp Kutlugh Bilge Kaghan (779-89) and did his best to remain on good terms with the
Chinese. His was not an easy task, since Emperor Te-tsung (779-805) had no sympathy for
the Uighurs, having been humiliated by them in his youth. Patient diplomacy and the advice
of his counsellors slowly overcame the emperor's hostility and, as a token of renewed
confidence, the princess of Hsien-an was given in marriage to the Kaghan in 787. Alp
Kutlugh Bilge Kaghan died in 789, but the princess remained among the Uighurs until
her death in 808, the Khatun (consort) of four successive Kaghans. Te-tsung had good
reasons to make conciliatory gestures towards the Uighurs; their help was sorely needed
in the quasi-permanent conflict between the T'ang and the Tibetans. The two immediate
successors of Alp Kutlugh Bilge Kaghan were murdered. A measure of stability returned
only under the rule of A-eh' o, whose regnal title was Kutlugh Bilge Kaghan (790-5), and
who was probably still a child on his accession to the throne.
At about this time, there appeared on the scene of Uighur history a new dignitary called
the ta-hsiang (Grand Minister) or ta-chiang-chun (Grand General) by the Chinese. He is
probably identical with the il ogasi (Glory of the Land/Nation) appearing in a Uighur text.

9
Mackerras, 1972, p. 93.
10
Ibid., pp. 221-38.

201
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

There is clear evidence to show that, at least under Kutlugh Bilge Kaghan, effective power
rested with the il ogasi. It should not then come as a surprise that, at the death of the youth-
ful Kutlugh Bilge Kaghan, the il ogasi ascended the throne under the grandiloquent regnal
title of Ai tangrida tiltig bolmish alp qutlugh ulugh bilga huai-hsin Kaghan (795-805 or
806). The death of Kutlugh Bilge Kaghan spelled the end of the Yaghlakar dynasty, which
had ruled ever since the establishment of the Uighur empire; the Huai-hsin Kaghan (to use
the short form of his name) belonged to the Adiz clan. There is some doubt concerning the
very existence of his successor who, if the Huai-hsin Kaghan indeed ruled until 808, might
be a 'ghost' created by the confusion of our sources.
The period of rule of Ai tangrida qut bulm'ish ktiltig bilga pao-yi Kaghan (809-21),
on the other hand, is quite well documented. It was during his reign that the trilingual
(Chinese, Uighur and Sogdian) inscription of Karabalghasun was erected, bearing witness,
inter alia, to a renaissance of Sogdian and Mani-chaean influence among the Uighurs.
It might be that this was one of the reasons why the Chinese - deeply distrustful of the
Sogdians - refused the Pao-yi Kaghan's request for an imperial bride, on the flimsy pretext
that they could not afford the expenses of such a wedding. 11 The true cost of the marriage
of a princess was estimated at 200,000 pieces of silk, the annual tax revenue of a large sub-
prefecture of the south-east. It was the view of Li Chiang, president of the Ministry of Rites,
that the benefits of a marriage alliance with the Uighur Kaghan would amply justify such
expenditure. 12 By the time the emperor ceded to persistent Uighur demands, the Kaghan
had died and, finally, Princess T' ai-ho became the bride of the new Uighur ruler, Kiln
tangrida ulugh bulm'ish ktichltig bilga ch'ung-te Kaghan (821-4). The third princess to
marry an Uighur ruler, T' ai-ho was to stay among the Uighurs for some 22 years, a pawn
in the complicated end-game of Sino-Uighur relations. She became the wife successively
of Ai tangrida qut bulm'ish alp bilga chao-li Kaghan (824-32), murdered by his entourage,
and of his successor Ai tangrida qut bulm'ish alp ktiltig bilga chang-hsin Kaghan (832-9),
who - menaced by an attack of the Sha-t' o tribes and by a conspiracy engineered by one
of his ministers - committed suicide.
Uighur fortunes had sunk very low. Exceptionally heavy snowfalls, causing widespread
famine, contributed to the disintegration of the Uighur body politic. The rule of Ho-sa
(839-40) was too short to allow him to receive an appointed name. He lost his life when
the Kyrgyz, who had been engaged for twenty years in a struggle with the Uighurs and
were this time led by the renegade Uighur general Ktiltig Bagha, attacked and destroyed
Karabalghasun, the Uighur capital. This sequence of events marks the end of the Uighur

11
Chavannes and Pelliot, 1911- 12, p. 282.
12
Mackerras, 1972, p. 113.

202
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

empire of Mongolia, though mopping-up operations were to continue for some time. Two
Uighur factions sought Chinese help. One of these was led by Oga Kaghan (841-7), who
had the support of thirteen tribes and in whom we must see the last legitimate Uighur
ruler. There is a whiff of Shakespearian tragedy in the destinies of these men, their swords
broken, seeking help from the Chinese who had never seen in the Uighurs anything but
rapacious barbarians. Chinese efforts were limited to repeated attempts to rescue Princess
T'ai-ho. Once she was safely back among her people, Li Te-yu, a distinguished civil ser-
vant of the T'ang, summarized well the policy to follow: 'Now that we have obtained the
princess, we should do battle with the Uighurs again, exterminating them completely, so
that none remain to cause later calamities.' 13 The fall of the Uighurs, so often praised by
the Chinese as their staunchest ally, caused no regret in China.
The Kyrgyz who put an end to the Uighur empire on the Orkhon represented a different
type of civilization. Although their language was Turkic, fairly close to Turk and Uighur,
the Kyrgyz chapter of the T' ang-sku (217B) describes them as strong and tall people, red-
haired, white-faced and green-( or blue-)eyed, with a dislike for dark hair and dark eyes. The
Persian historian Gardizi, writing in the mid-eleventh century, confirms this description.
Individuals with dark features were thought by the Kyrgyz to be the progeny of the rene-
gade Chinese general Li-ling, who in the first century B.c. defected to the nomads. Unlike
the Uighurs or the Turks, the Kyrgyz were not considered descendants of the Hsiung-nu
by the Chinese. In fact, one of the earliest detailed mentions of the Kyrgyz (called Ch'ien-
kun at that time) refers to their defeat in c. 49 B.c. at the hands of the Hsiung-nu ruler, the
shan-yu Chich-chih.
The T'ang-shu remarks that the Ch'ien-kun - ancestors of the Kyrgyz called in T'ang
times Hsia-ch'ia-ssu - were 'mixed Ting-ling'. This indication is of little help and has
caused much confusion among superficial historians. It was assumed that the Ting-ling
constituted the 'blond' element among the Kyrgyz. Probably the opposite is true since,
in Maenchen-Helfen's words, 'there is not the slightest evidence for attributing to the
Ting-ling the characteristics of the white race which the T' ang-shu gives to the Ch'ien-
kun' .14 There seems to be general agreement that Ting-ling is the Chinese rendering of the
same name which later appears in the form of T'ieh-Ie, the designation of a tribal confed-
eration to which, as we have seen, the Uighur (among other Turkic peoples) belonged. One
could then speculate that the Turkic element was brought into the process of Kyrgyz ethno-
genesis by the Ting-ling. (Archaeological publications - based on the acceptance of the
unjustified theory proposed by historians - usually attribute to the Ting-ling the Europoid

13
Drompp, 1986, p. 300.
14
Maenchen-Helfen, 1939, p. 83.

203
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

component.) The population of the Tasht'ik culture (first century B.c.-fifth century A.D.) in
the Yenisei valley was an amalgam of Europoids and Mongoloids, as is clearly shown by
the clay death masks discovered. It is also beyond doubt that the Mongoloid component
of the Tashtik population was superimposed upon the Europoid population which it had
probably conquered. While it seems unlikely that the Kyrgyz adopted a Turkic language as
late as the eighth century, it is certain that their language contained non-Altaic (Samoyed
or Palaeoasiatic) elements, such as the name of a special type of iron.
Some aspects of their civilization clearly set the Kyrgyz apart from the rest of the Turco-
Mongol inhabitants of the steppe. Thus, for instance (and the T' ang-shu makes a special
point of this), when mourning, the Kyrgyz did not lacerate their faces. Also, the original
title of their ruler was a-je, not in use by any known Inner Asian people. The title Kaghan
came to be used only following the victory over the Uighurs. The description of their cus-
toms in T'ang times clearly shows that the Kyrgyz were a forest-dwelling people, very
different in their lifestyle from the 'Hsiung-nu type' steppe-dwelling pastoral nomads. We
are informed that, in their forests, the pine trees grew so tall that an arrow could not reach
their peaks. Their land was marshy in the summer and covered with deep snow in the
winter, so that the Kyrgyz could use skis (mu ma, or 'wooden horse') for the hunt. In the
winter, when the cold was so severe that the rivers froze to half their depths, the Kyrgyz
lived in huts covered with bark. Their tribute to the Chinese consisted of pelts. They prac-
tised tattooing. They were shamanists, and the shamans were called by their Turkic name,
gam or kam. The early twelfth-century geographic treatise of Marvazi describes in some
detail the doings of a shaman, and he notes that the Kyrgyz bum their dead. 15 Gold, tin
and iron could be found in their land, and the Kyrgyz were skilled in the manufacture of
arms which they purveyed to the Turks in lieu of tribute. Chinese information concerning
Kyrgyz prowess in the manufacture of arms is confirmed by archaeological finds and also
by petroglyphs on which the representation of plate armour is clearly visible. The Kyrgyz
used caparisons to protect their horses, and their pointed helmets are reminiscent of the
'lion helmets' used in T'ang China.
Without doubt, in the course of their early history, the Kyrgyz moved from west to east.
At the time when the Hsiung-nu shan-yu defeated them, the Ch'ien-kun (as the Kyrgyz
were then called) were located to the north-west of the Western Hsiung-nu and to the north
of K'ang-chti. According to the Han shu [History of the Former Han] (Ch. 94B), in the
middle of the first century B.c. the Ch'ien-kun lived at a distance of 7,000 li to the west
of the Hsiung-nu headquarters. The location of K' ang-chti poses some problems - at some
periods the name was applied to Sogdiana, at others to Samarkand itself - but it is beyond

15
Minorsky, 1942, p. 30.

204
ISBN 978-92-3-103467-1 Contents THE UIGHURS AND THE KYRGYZ

doubt that it lay far to the west and, most probably, somewhat to the north of the T'ang-
period country of the Kyrgyz. Muslim historians tend to place the Kyrgyz in the north-
eastern extremity of the known world. According to the anonymous Persian geography,
the Hudad az_c alam [The Limits of the World] (A.D. 982), only China and the Eastern
Ocean were east of the Kyrgyz, while to the north of them lay the 'Uninhabited Land of
the North where people cannot live on account of the intensity of the cold' . 16 The distance
separating the Kyrgyz from the Chinese made any contact tenuous, and communications
were made difficult and often cut off by the Turk or Uighur empires wedged between the
two. According to the T' ang-shu, caravans needed 40 days to cover the distance separating
Karabalghasun from the Kyrgyz country.
It is difficult to assess with any degree of certainty to what extent the Kyrgyz actually
ruled over the ancient land of the Uighurs. For demographic as well as economic reasons
- they were not numerous enough and they were not a typical steppe people - they did
not fill the vacuum left by the disappearance of the Uighur empire. Although, perhaps for
lack of any contender, nominal Kyrgyz rule over Mongolia lasted for about a century, their
power base remained in the Yenisei region, in present-day Tuva, where was located their
capital Kemjkath (uncertain reading). Savinov distinguishes five cultural groups within the
Kyrgyz empire, namely those of Tuva; east Kazakhstan; the Gorno Altai; the Minusinsk
basin; and the Krasnoyarsk region. 17
From about 924 onwards, the Kitan had effective control of the land that used to be the
centre of the Uighur empire. This is evidenced by the offer made to the Uighurs of Kan-
chou by A-pao-chi (alias T' ai-tsu), ruler of the Kitan, to return to their former homeland.
By then happily settled in their new country, the Uighurs declined the offer. The arrival
of the Kitan and the withdrawal of the Kyrgyz from that region marks the end of Turkic
preponderance in what was to become Mongolia.

16
Anon., 1937, p. 96.
17
Savinov, 1984.

205
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

Part Two

THE UIGHUR KINGDOM OF KOCHO

(Geng Shimin)

After the destruction of the Uighur Khanate in 840, some tribes migrated west to Gansu,
where they established the Uighur kingdom of Ganzhou (later, in 1026 to be conquered by
the Tangut. The Yughur nationality in the vicinity of Jiuquan, Gansu province, are their
present-day descendants). Another important branch (fifteen tribes) migrated westward to
the area of Beshballk (in Chinese called Beiting; its ruins are at Hubaozi in Jimsar county,
Xinjiang Uighur Autonomous Region) north of the T'ien Shan mountain range, crossed the
T'ien Shan southward and occupied the Turfan basin where the Uighur kingdom of Kocho
was established (c. 860-1284). Shortly afterwards the Uighurs expanded their power to
include the Yanqi (Argi, Karashahr) and Qiuci (Kuci, Kuchar) areas. Historical records
concerning the Uighur kingdom of Kocho are very sparse. The Chinese records are frag-
mentary because the chaotic situation in China during the period of the Five Dynasties or
Ten Kingdoms (907-60), and the weak government of the Sung dynasty (960-1279), did
not allow much attention to be paid to the Western Regions. Arab and Persian sources had
little interest in a land that was not yet Islamic and the records written in Uighur were
mostly destroyed after the introduction of Islam into the Tarim basin.
The Uighur name 'Kocho', originating from the Chinese 'Gaochang', refers to the city
as well as to the area. The ruins of the ancient city of Gaochang, with its imposing walls,
stand 40 km east of the Turfan county administrative centre. The Turfan region, located in
the north-eastern part of the Tarim basin, was an important line of communication between
East and West and had since ancient times been a key post on the Silk Route.
In the north of the area lies the T'ien Shan, with its snow-capped peaks such as Bogda
and Qara-uchin. The streams and rivers, formed by melting snow, provided abundant water
that was conducted to the fertile farmlands by underground aqueducts called by the Persian
term karfz. As early as the second century B.c., Turfan became a prosperous oasis with a
developed agriculture, producing wheat, barley, rice, maize and beans as well as cotton,

206
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

grapes, melon, Hami melon (a kind of musk melon), sesame, and so on. To the north of
the T'ien Shan, there are vast tracts of excellent natural (alpine) pasturelands, where large
herds of horses, sheep, cattle and camels can graze. There are also many wild animals such
as Asiatic wild ass, yak, large-headed sheep, Mongolian gazelle and antelope.
The earliest identifiable inhabitants of the Turfan area were probably the Ytieh-chih
(Juzhi), speaking a particular Inda-European language formerly called in the scholarly lit-
erature 'Tokharian fa:. During the period of the Northern and Southern Dynasties (420-581 ),
a large number of Chinese entered the Turfan area from inland China. One of them named
Jujia of Jincheng, Gansu, founded the Ju dynasty (493-640), which lasted for nearly one
and a half centuries. Chinese civilization had a great influence upon the local people.
According to Chinese sources, during the fifth-sixth centuries, in the Turfan area the 'bar-
barian' script (i.e. the local script) as well as Chinese characters were used. In schools, the
Chinese classics like The Book of Songs, The Confucian Analects and The Book of Filial
Piety were taught, but they were all recited in the native ethnic language. Judging from
Turkic manuscripts dating from the eighth century, it seems that even before the Uighur
immigration, there was a Turkic presence in the Turfan region. The arrival of large num-
bers of Uighurs accelerated the Turkicization process. There was also a substantial Sogdian
population living in the area.
The original inhabitants of Karashahr and Kucha also spoke Tokharian (those living
in Karashahr speaking dialect A, those of Kucha using dialect B). From ancient times,
Karashahr and Kucha, two city-states on the northern edge of Tarim basin, have had a
developed economy and culture. The Chinese pilgrim Hstian-tsang (600-40) gives a vivid
description of Kucha in his famous Da Tang xi-yu-ji [Records of Travels to the Western
Regions in Great T'ang] (Ch. 1):

The Kuci state measures over 1,000 li [l li = 500 m] from east to west and more than 600 li
from south to north. The capital has a circumference of 17-19 li and produces millet, wheat,
rice, grapes, pomegranates, pears, peaches and apricots; gold, copper, iron, lead and tin are
mined. The climate is mild and the manner of the people is polished. Their script originated
in India, but a few changes have been made. Their music and dance are famous. They wear
brocade or plain cloth, cut their hair short and wear turbans. They use gold or silver coins and
small copper coins as currency ... There are more than 100 Buddhist temples, and more than
5,000 monks and novices, and all belong to the Sarvastivadin sect of Hinayana.

The indigenous inhabitants of all these three areas were Europoids. In the words of the
Pei-shih [Annals of the Wei Dynasty] (Ch. 97), 'The people of the states west of Turfan all
have deep-set eyes and high noses.'
After the Uighurs moved into these three areas, a process of fusion of different ethnic
groups began through intermarriage. Because the Uighur and other Turkic groups enjoyed

207
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

both political and numerical superiority, the indigenous population was gradually Turki-
cized and the Uighur language triumphed over the native Tokharian language and became
a kind of lingua franca in these areas. The languages of the indigenous population gradu-
ally fell into disuse and died out, though the indigenous substratum had a great influence
on the Uighurs in ethnological, cultural and linguistic respects. Under the influence of the
relatively advanced economic system and culture of the native people, the Uighurs gradu-
ally gave up their nomadic life and turned to a settled, urban or agricultural existence and
created in Kocho a brilliant civilization.
The sovereign of the Uighur kingdom of Kocho took the title idiqut ( < iduq qut, His
Holy Majesty). It seems that this title was taken from another Turkic tribe, that of the Bas-
mil, who lived in the Beshballk area before the Uighur immigration. To foreign countries,
the Uighurs of Kocho called their land 'Great, Good-Fortunate State'. Under the idiqut
there were high-ranking officials such as nine ministers, ulchi, tarqan, sangun, tutuq as
well as many officials of middle and lower rank such as the alchi, bagi, and so on. Further,
there were freemen, craftsmen and merchants. In the countryside, the landlords possessed
extensive farmlands and water resources, with a class of poor peasants and tenant farmers
beneath them. Uighur civil documents show that the remains of a slave system existed. 18
The Manichaean and Buddhist monasteries also owned extensive farmlands, with many
dependent households.
As far as relations with neighbouring countries were concerned, in the east, after the
Liao dynasty (907-1125) and the Sung dynasty (960-1279) were established, there were
frequent diplomatic and commercial contacts between them and the Uighurs of Kocho. In
the west, relations between the Karakhanids (c. 960-1213, founded by the Turkic Karluk)
and the Uighurs of Kocho were strained. Although both sides were Turkic, speaking a
common language (with only some dialectal differences) and at the outset using the same
Uighur script, the former were Muslims while the latter were Buddhists. Because of the dif-
ference in religion, the two sides were extremely antagonistic. The idiqut of Kocho put up a
determined resistance to the expansion of the Karakhanids; until the thirteenth-fourteenth
centuries the influence of Islam could not be felt beyond Kucha.
At the beginning of the twelfth century, Kocho came under the rule of the Western Liao
(1124-1211), who installed ajianguo (supervisor). In the early thirteenth century (1209),
the Uighur king Barchuk Art Tegin voluntarily submitted to Chinggis Khan, founder of the
Mongol empire. Because of this, his realm was particularly well treated by Chinggis, and
preserved its original boundaries.

18
Geng Shimin, 1984, pp. 7-18.

208
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

In the second half of the thirteenth century, the Haydu-Duwa rebellion, lasting for nearly
forty years, gave the Uighur kingdom of Kocho its death blow. In 1270 the summer capital,
Beshballk, was taken by the rebels, and the Uighur idiqut retreated to Kocho city. Some
years later, in 1284, he was forced to flee to Yongchang in Gansu. By this time, the kingdom
existed in name only; its domain was incorporated into the Chaghatay Khanate.
In A.D. 982 the Sung envoy Wang Yande had visited Kocho. His account in the Shi
gaockang ji [Records of an Embassy to Kocho] constitutes the most important historical
source for the study of the Uighur kingdom during this period. He gives a vivid description
of Kocho in the Sung shi [History of the Sung dynasty] (Ch. 490):

The area has no rain or snow and is extremely hot, and when the hottest season arrives the
inhabitants all move into caves dug in the earth . . . Their houses are whitewashed, and water
from Jinling [Golden Mountain] flows through them and is circulated through the capital
city to water the gardens and turn mills. The area produces the five cereal grains, but it lacks
buckwheat. The nobles eat horse-meat and the common people eat goat or fowl. In music they
make much use of the pipa [a kind of four-stringed lute] and the konghou [an ancient plucked
instrument with seven strings like the harp]. They produce sable, cotton and brocaded cloth.
They are fond of archery and riding. The women wear oiled caps which are called sumuzhe.
They use the calendar promulgated in the seventh year [719] of the reign of Kaiyuan of the
T'ang dynasty ... They are fond of excursions and always take along musical instruments.
There are more than fifty Buddhist temples with inscribed boards given by the T'ang imperial
court. In the temples are kept The Buddhist Tripitaka, Rhyming Book of T'ang, The Dictionary
of Chinese Characters, The Chinese Buddhist Dictionary ... and other works ... There is an
imperial library which holds imperial letters and orders from Tang Taizong (627-49) and
Tang Xuanzong (712-56), which is very carefully locked. In addition, there are Manichaean
monasteries and Persian monks ...

When Wang Yande arrived in Kocho, the Uighur king Arslan Khan was away in his summer
quarters north of the T'ien Shan, leaving his maternal uncle to take care of state affairs.
Thus Wang and his party crossed over the T'ien Shan to go to Beshbalik On the road,
he saw many herds of horses belonging to the royal family grazing on the steppes which
stretched as far as 100 li. Horses of different colours were divided into separate herds for
grazing, an indication that the Uighur upper classes still maintained the old, traditional
nomadic habits.
According to the accounts of Uighur sources, on ceremonial occasions the Uighur king
sat on a golden chair (orgin) placed on a platform (Uighur taucang <Chinese daochang)
and decorated with pearls and jewels. He wore a red robe and a crown (Uighur didim <
Greek diadema).
Among the Uighurs who moved into the Turfan region, the Uighur script gradually
replaced the Old Turkic so-called 'runic' alphabet. As urban life and trade developed, the

209
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

use of the Uighur language and script became general. Not only were a large number
of religious (Manichaean, Christian and Buddhist) works translated into Uighur, but the
language was also extensively used in everyday life, which was strongly influenced by
Chinese culture. Regarding the calendar, for example, the Uighurs used the Chinese system
of the tiangan (Ten Heavenly Stems) as well as the dizhi (Twelve Earthly Branches) to
designate years. In this period, Uighur borrowed many Chinese words, such as maka or
'ink' (<Medieval Chinese mak), kuin or 'scroll, volume' ( < M. Chin, kuian ), etc.
As early as the time of the Uighur Khanate in Mongolia, in 762, Mani-chaeism was
accepted by the Uighur nobility. After their migration to the Turfan region, Manichaeism
was still maintained for a period among the Uighurs, as shown by the many Manichaean
mural paintings as well as Uighur Manichaean manuscripts (often decorated with beautiful
miniatures) found in the Turfan area. Nestorian Christianity also had followers among
the Uighurs. In the Nestorian sites of Turfan, a fresco depicting the rites of Palm Sunday
has been discovered. In addition, several Uighur fragments, some Nestorian writings and
a story about three Persian Magi visiting Bethlehem have been unearthed. The 'Iranian
monks' mentioned by Wang Yande must refer to Nestorian priests. The translation into
Uighur of Aesop's Fables might also be related to the spread of Nestorianism among the
Uighurs.
The religion that spread most widely among the Uighurs, however, was Buddhism.
Under the influence of the original inhabitants, the Uighurs gradually converted to that reli-
gion and large numbers of Buddhist classics were translated into Uighur. During Buddhist
festivals, at mass gathering places around temples, a variety of activities such as story-
telling, as well as dramas with Buddhist content, were performed. The 'banquet with dra-
matic performance' attended by Wang Yande, and the discovery of the primitive Buddhist
drama the Maitrisimit written in Uighur testify to this point. 19 Many Buddhist sites, grot-
toes (such as Bezekllk, Tuyok, Kizil and Kumtura), as well as the large quantity of Uighur
manuscripts, further indicate that Buddhism flourished extensively in the Uighur kingdom.
Since the 'more than fifty Buddhist temples' in Turfan were given boards inscribed in
Chinese, indicating that Buddhist works were kept there, we may conclude that Chinese
Buddhism had a profound influence on Uighur Buddhism. Having originally spread from
Xinjiang to inland China, Buddhism now returned to the area from China.
Because of their conversion to Buddhism, the Uighur nobility and even the common
people took to building temples, making statues, painting frescoes and copying sutras as
a kind of charitable and pious deed (in Uighur called buyan < Sanskrit bunya). These are
precious works of art representing the high level attained in art and culture by the Uighurs at

19
Gen Shimin and Klimkeit, 1988.

210
ISBN 978-92-3-103467-1 Contents THE UIGHUR KINGDOM OF KOCHO

that time. The frescoes give a vivid picture of the daily life of the Uighurs. Their residences
are enclosed by walls; flowers and trees are planted in the courtyards. Men and women hold
long-stemmed flowers in their hands. Men have garments and headdresses in T'ang style,
they wear half-length boots and have various daily utensils hanging on their waistbands.
Women wear long gowns with T'ang floral patterns in Transoxanian style.
If the mural paintings depicted the lifestyle of the rich families, the Uighur commoner's
life was reflected in a variety of civil documents written in Uighur. They are prime sources
for the study of the socio-economic condition of the Uighur kingdom of Kocho. Through
these documents (up to now several hundred of them have been discovered), we can clearly
see how peasants and farm labourers were exploited by the landlords.
In the Uighur kingdom of Kocho, large numbers of Buddhist works were translated into
Uighur, mostly from the ancient Karashahr-Kuchaean language as well as from Chinese.
During the Yuan or Mongol dynasty (1279-1368) some were also translated from Tibetan.
The surviving corpus of Uighur Buddhist texts shows that nearly all the main Buddhist
works (including sutras, vinayas and abhidharmas) were translated into Uighur. Important
among them were Mahayana works such as The Golden Light Sutra, The Lotus Sutra, The
Garland Sutra and Sthiramati's Commentary on the Abhidharmakosa, all translated from
Chinese. But there are also Hinayana texts, such as the Maitrisimit Nam Bitig, translated
from 'Tokharian fa: and many fragments of Agamas. Also during the Yuan dynasty, many
tantric works were translated from Tibetan and Chinese. Among the Uighur Buddhist trans-
lators, Master Singu Sali is worthy of particular mention. Born in Beshballk, he lived in
about the tenth-eleventh centuries. From the Chinese he translated into Uighur The Golden
Light Sutra, The Biography of Xuan-zang and many other Buddhist works. His transla-
tions read smoothly, his vocabulary is rich; he not only had a good command of classical
Chinese, but was a master of his own mother tongue; it seems that he also knew Sanskrit.
The coexistence of various religions resulted in a fusion of different faiths. For example,
in Uighur Buddhist texts the Indian gods Brahma and Indra are called by the Manichaean
names Azrua and Hormuzd respectively, indicating a degree of fusion between Buddhism
and Manichaeism. Furthermore, a type of calendar written in Sogdian and used by the
native Manichaean believers indicates an assimilation of Sogdian, Chinese and Uighur
cultural elements. The name of every weekday was written in Sogdian and then each name
was supplemented by the transcribed Chinese names of the 'Ten Heavenly Stems' such as
jia, yi, bing, and so on. Then again there were written in Sogdian the twelve animal names,
such as mouse, ox, tiger, rabbit, and so on, which were used by the Uighurs. Finally, the
names of the Chinese wuxing (five elements), metal, wood, water, fire and earth, were
translated into Sogdian and matched with two-day periods.

211
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

In sum, it may be said that during almost 400 years of the Uighur kingdom of Kocho's
existence, great achievements in the social, economic and cultural areas can be credited to
the Uighurs. At the same time, the period was characterized by further fusion in ethnic,
linguistic, religious and cultural elements within Uighur society. 20

Part Three

THE TANGUT HSI HSIA KINGDOM (982-1227)

( Y. I. Kychanov)

The active rivalry between Tibet and China from the seventh to the ninth century altered
the destinies of the Qiang and Xianbei populations occupying the area of the modem
provinces of Qinghai and Sichuan. The Tuytihun state of the Xianbei, which had lasted
for more than 300 years, was destroyed by the Tibetans; and some of the Dangxiang,
a people of Tibetan-Burman ethnic origin living in the Sunpan region in north-western
Sichuan, moved north under the pressure of the Tibetans, eventually reaching the Ordos. In
the ninth century, the ruling clan of the Dangxiang - better known to European scholars by
their Mongol name, the Tangut - became firmly established in the centre of the Ordos with
their capital in the town of Siachou (in the western part of the modem Heng Shan district
of Shanxi province). The centre of the Ordos, known as the Land of the Five Regions, had
acquired almost full independence during the tenth century. The Tangut population formed
an absolute majority there and the consolidation of the independence of the region with
its Tangut population contributed, first, to China's breakup into a number of states during
the period of the Five Dynasties and, second, to the emergence in the north of the pow-
erful Kitan Liao state which supported the ruling Tangut house of Toba in its striving for
independence.
The unification of China under the Sung dynasty in 960 naturally raised the question
of what would happen to the Tangut possessions of the Toba. When, in the year 982, the
Sung court attempted to bring those regions under its control, one of the members of the
ruling Tangut Toba clan, Jiqian, openly opposed China and, following the example of the

20
For the genealogy of the kings of the Uighur kingdom of Kocho, see Geng Shimin and Hamilton, 1981 ,
pp. 10-54.

212
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

independent Kitan state, began the struggle to establish a Tangut state. He banked on con-
flict between the Liao and the Sung, and this proved to be justified since in 989 the Kitan
court gave Jiqian the hand of a Kitan princess and recognized him as Wang of the Hsia state.
In 997 Jiqian succeeded in establishing his authority over the area of the Five Regions. In
the year 1002 the Tangut captured the town of Linchow (the modem town of Lingwu in
the Ningxia-Huizu Autonomous Region). Some time at the end of the tenth century or the
beginning of the eleventh (1001 according to the Sung shi) the term Hsi Hsia began to be
employed to designate the Tangut state. The new state could only expand to the west and
south-west, although there was quite a large ethnically similar Dangxiang (Qiang) popula-
tion in the areas of China to the east of Hsi Hsia bordering on Jiqian's domain and in the
southern Liao regions bordering on Hsi Hsia. These regions could not be wrested by force
from the Sung and the Liao. Throughout the first third of the eleventh century, the Tangut
therefore waged war to the west.
Jiqian was killed during a war with the Tibetans in the year 1004. His successor Li
Deming, realizing that without peace with China it would be difficult to achieve any success
in the west, concluded a peace treaty with the Sung in 1006. He agreed to accept from
the Sung the post of tzedushi (military governor), thus acknowledging himself to be in
Chinese service, and received the title of Wang Xiping (Pacifier of the West). The treaty of
1006 between Hsi Hsia and the Sung was, however, a sworn treaty which signified Sung
recognition of the Tangut state or, as modem Chinese historians have written, 'the Sung
emperor Shen Zong recognized the special position of Li Deming' .21 Li Deming began
to prepare for official acceptance of the title of emperor. In 1016 he declared his father
Jiqian emperor posthumously and in 1028 declared his son as heir to the imperial throne
and his son's mother as empress. Xingqing, the modem town of Yinchuan (Ningxia-Huizu
Autonomous Region), was declared to be the state's new capital.
Peace with China and the support of the Liao enabled the Tangut to conduct success-
ful wars against the Tibetans and the Uighurs. In 1028 territories including the towns of
Lanzhou (Wuwei) and Hangchou (Chane) and the modem province of Gansu were added
to Hsi Hsia. In 1036 the Hsi Hsia dominions were extended further westwards to the area
of modem Dunhuang and the edges of the Hami oasis. The territory of the Tangut state
included the entire province of Gansu in its western part, the western aymaks of the mod-
em Inner Mongolian Autonomous Region located to the south of the Gobi desert, the Ala
Shan and the Helan Shan as well as the entire area of the Ningxia-Huizu Autonomous
Region and the western regions of the modem province of Shanxi in the Ordos. The border
with the Kitan ran along the southern edge of the Gobi desert and the northern branch of

21
Zhong Kan et al., 1979, p. 25.

213
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

the Yellow River, where it forms the Ordos bend. The western branch of the Yellow River
was entirely within Hsi Hsia territory but the border with China did not extend as far as its
eastern branch, as Sung forces not only denied the Tangut access to the river but even, in
1067, took from them the town of Suizhou (the modern town of Suide in Shanxi province).
In the year 1136, following the defeat of the Jurchen by the Sung, the Tangut incorpo-
rated into their state the area of the modern town of Xining in the province of Qinghai.
Altogether this was a vast territory, containing fertile land that was suitable for agriculture
(the valley of the Yellow River, the valleys of the rivers Heitui and Xining, the north-
western edge of the loess plateau) and good pastureland on the plains of the Ordos and Ala
Shan and in the mountain regions. The present-day climate of these regions is continental
with a cold winter, a very hot summer and insufficient precipitation. There is evidence,
however, that the climate was milder and more humid in the period from the eleventh to
the thirteenth century. Rice, cotton and apricots were grown in a number of regions within
the territory of the Hsi Hsia. The mainstay of the economy consisted of agriculture and the
raising of livestock. The country was self-sufficient in grain, with the possible exception
of rice, and exported livestock and livestock products. Camels and horses from Hsi Hsia
were particularly highly valued. The extensive mountain regions provided a reliable source
of ores and the country experienced no shortage of iron although, according to some evi-
dence, the Tangut lacked sufficient quantities of copper. Salt extraction was a flourishing
activity and salt from the Ordos lakes was exported to China.
In the year 1031 Li Deming died and was succeeded by his son, Yuan Hao. The first
Tangut emperor and a reformer, he was to play an outstanding role in the ultimate consol-
idation of the Tangut state deep in the interior of Asia. In 1032 Yuan Hao abandoned the
use of the Sung reign titles, used for the purpose of chronology, and adopted as the title
of his reign Xiandao (Clear Path). This was an important step towards the formation of
the imperial system of administration and the first of Yuan Hao's reforms, of which there
were ten according to historians. The second reform was the decree of 1033 concerning
the introduction of a single hairstyle (tufa) for all men. This was the Kitan male hairstyle
in which most of the hair on the head was shaved off, leaving only a forelock and locks
on the temples. This was an important political act as a single hairstyle (similar to the later
Manchu requirement for all Chinese adult males to wear pigtails) was a symbol of the
submission and unity of all the sovereign's subjects.
Under the third reform, the area of the country's capital was elevated to the status of ju
(conventionally, a 'department'), the same status as the capitals of Liao and Sung. The city
was renamed Xingqing (Celebration of the Ascendancy). At the same time, central gov-
ernment institutions on the Chinese model were introduced: a state secretariat (chunzhu)

214
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

for the administration of civil affairs; a privy council (shumi) for the command of the army
and the conduct of military matters; and a censor's office (yushitai), an organ of control.
The Tangut did not introduce the system of 'six ministries' (liu bu) adopted in Liao, for
example, but the ministries or departments established by Yuan Hao met all the require-
ments for the household administration and record-keeping of the imperial court; for the
economy, especially agriculture and livestock-raising; and for the maintenance of law and
order throughout the country. The fifth reform introduced uniforms for officials. These
reforms were enacted in 1033 and were followed three years later by a precise definition of
the system of compulsory military service (the sixth reform), the ordering of the country's
administrative subdivisions into twelve military-political districts (the seventh reform) and
the introduction, also in 1036, of an original Tangut script.
The Tangut language was tonal and rich in homonyms. The Tangut therefore devised
a logographic-syllabic hieroglyphic script. This script looked very unlike Chinese, but its
resources were based on the Chinese writing system and the characters were constructed
using the six methods for the formation of characters laid down by Chinese philology of
the period. Chief among these were the ideogram (huii) and the phonogram ( hsing shen)
methods. This eighth reform was not only the most important event in the cultural history
of the Tangut people but also a milestone in the cultural history of East and Central Asia.
During the previous stage in the cultural history of East Asia, the Koreans, Japanese and
Vietnamese merely adopted the Chinese script, perhaps introducing some of their own
indigenous characters and methods of transcribing their mother tongue, but the tenth to
the twelfth century saw the Kitan, the Tangut and the Jurchen adopting the resources of
Chinese writing to create their own original scripts. The invention and introduction of the
script led directly in 1038 to another event of enormous cultural significance extending
beyond the limits of Tangut culture, when a special group of thirty-two Buddhist monks
was entrusted with the translation of the Buddhist canon into the Tangut language. The
ninth and tenth reforms were also cultural in nature. In 1037 Tangut (the Tangut's own
self-appellation was Mi-nyag) and Chinese schools were established; the year was also
marked by the introduction of Tangut music to the court as official music for ceremonial
purposes in the belief that 'the wise ruler should conform to popular customs'.
On 10 November 1038 Yuan Hao officially declared himself emperor and bestowed
posthumous imperial titles on his grandfather, Jiqian, and his father, Li Deming. Since the
Tangut state organized itself as an empire on the Chinese model during the reign of Yuan
Hao, who was also the first openly to adopt the title of emperor, a number of contemporary
historians have put the date at which the Tangut Hsi Hsia state was first established at 1032
(the year in which Yuan Hao was enthroned) or 1038 (the year in which he took the title

215
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

of emperor). His assumption of that title created a new situation in East Asia and in east
Central Asia. A third Son of Heaven appeared alongside the Sung and Liao emperors. The
Sung emperor recognized the imperial title of the Liao emperor and, by the Xianyuan treaty
of 1005, Sung China paid the Kitan a sizeable tribute under the guise of 'assistance'. The
recognition of one more Son of Heaven would have further undermined Chinese views
of universal monarchy and the exclusive position of the Son of Heaven on Earth. The
Sung court therefore refused to recognize the imperial title of Yuan Hao. This led to the
Tangut-Chinese war of 1040-4.
The war demonstrated that neither side could count on a definitive military victory. The
title became the subject of negotiations which turned on whether Yuan Hao would adopt a
traditional title (kaghan, shan-yu) rather than the Chinese imperial title of huang-ti. Matters
were further complicated by the interference of the Kitan, who threatened the Sung in 1042
and forced China to increase the payments to Liao. Having obtained what they wanted from
the Sung, the Kitan also began to threaten Hsi Hsia. In the circumstances, Hsi Hsia and the
Sung concluded a peace agreement whereby Yuan Hao renounced the use of the title of
huang-ti (emperor) in the international arena and in relations with China, retaining the
imperial title (utszu in the Tangut language) within the boundaries of Hsi Hsia. The Sung
court recognized him as chou (sovereign) and paid an annual tribute of 255,000 units of
silk, silver and tea. Yuan Hao acknowledged his status as 'junior' or, as the translation has
it, 'vassal' (chen), and received a seal from the Sung emperor. Although the Hsi Hsia state
preserved its actual independence, the outcome of the war ultimately favoured the Sung.
A final peace with the Sung had still not been concluded when Yuan Hao was drawn into
a war with Liao. China, for which the Kitan state represented the main enemy, occupied a
position in the Tangut-Kitan war which, on the whole, was more favourable to Hsi Hsia.
The Tangut inflicted a military defeat on Liao and the two sides made peace in 1045. In
January 1048 Yuan Hao was killed. There followed a period in the history of the Tangut
state, during the minority of the emperors Liangzu (1048-67), Pinchang (1067-86) and
Qian-shun (1086-1139), when power was held by empress-regents and the Lian clan to
which the mothers of Pinchang and Qianshun belonged. China attempted to exploit the
internal disturbances in Hsi Hsia in order to destroy the Tangut state. Of the three major
wars between the Sung and Hsi Hsia (1069-72, 1081-6 and 1096-9), the war of 1081-6
proved the most serious for the Tangut state. An army of over 300,000 invaded the territory
of Hsi Hsia. Nevertheless, the overall outcome of the war did not favour the Sung and,
according to one source, 'the Sung forces perished ingloriously (Sung shi, Vol. 486, p.
3792). In 1094 the young Emperor Qianshun mounted something resembling a coup d'etat

216
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

and destroyed the Lian clan, restoring to power the imperial clan of Toba (the Tangut
Weimin or Ngwemi).
The economy and culture of the Tangut state began to flourish from the beginning of
the twelfth century. To a certain extent, this was due to the emergence of a new power to
the north-east, that of the Jurchen, which contained the activity of Liao and attracted the
attention of the Sung as China hoped to use the Jurchen to destroy the Kitan. The destruc-
tion of Liao in 1125, the subsequent fierce struggle between the Sung and the Jurchen,
and the emergence of the Jurchen Chin state with which Hsi Hsia managed to establish
friendly contacts, led to the flowering of the Tangut state in the middle and second half of
the twelfth century during the reign of Qianshun's son, Emperor Renxiao (1139-93), who
was half-Chinese by birth.
The problem of Hsi Hsia's 'own path' was finally resolved during the reign of Renxiao.
Under Yuan Hao the concept of their 'own path' had been expressed in the establishment
of the Tangut state's ideology on the twin pillars of Buddhism and the copying of the prin-
ciples of the Chinese state system. This was accompanied by propaganda advocating their
own path as opposed to the Tibetan and Chinese paths. Chinese official and bureaucratic
customs (ritual, clothes) were openly rejected in favour of Tangut customs. Under Yuan
Hao's successors, the empress-regents and their supporters inclined towards Buddhism
and the reduction of Chinese influence. In contrast, the ruling house, after recovering real
power, demonstratively returned to Chinese ways. Qianshun assigned equal status to Bud-
dhist and Chinese influences in his policy but Renxiao attached greater importance to the
Chinese Confucian model in state administration. The cult of Confucius was officially
adopted in Hsi Hsia in the year 1146. Orders were given to construct Confucian temples
in all regions of the country and Confucius was venerated, receiving the homage due to an
emperor; he was accorded the title of Emperor Wenxuan. Confucian canonical texts were
industriously translated into Tangut. Buddhism also continued to develop actively during
the reign of Renxiao but could no longer claim to be the principal state religion. It was
assigned a role as preserver of the well-being of the dynasty. In the history of the Tibeto-
Burman peoples there were three major states: Tibet, Burma and the Tangut state. Two of
these states chose Buddhism as their sole ideology and developed as theocratic states. This
did not happen in the case of the Tangut state, which was able to choose its 'own path'
in this area. The main reason was the proximity of Hsi Hsia to China, whose institutions
exerted an overpowering influence on the Tangut.
The Tangut state was multinational in the modern sense. The peoples forming the
bulk of its population were the Tangut (self-designation, Minyag), Chinese, Tibetans and
Uighurs. There was no difference in the rights enjoyed by these peoples in the area of

217
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

public life since in terms of hierarchical precedence, the official with the higher rank was
always considered senior. Only among officials of the same rank was the Tangut always
considered senior. The codex of the Tangut state, a 20-volume collection of the country's
laws amounting to 1,460 articles, came into force during the reign of Renxiao. The text,
which has been preserved to the present day, testifies to the highly developed legal sys-
tem of the Tangut state. While Tangut law developed along the lines of Chinese law (as
did Korean, Japanese and Vietnamese law), like those legal systems it exhibited certain
original features.
Tangut society was made up of people with and without personal freedom (pkhinga,
or men without personal freedom, and nini, or women without personal freedom). Free
people were divided into those who served ('those of rank' in Tangut terminology) and
those who did not serve ('ordinary people' in Tangut terminology). Members of the ruling
dynasty of the N gwemi clan and the emperor's kin occupied a particular position in society.
According to Chinese law, the aristocracy had no rights deriving from their origin but
it is thought that there were a number of Tangut clans/families, membership of which,
while it did not constitute an entitlement to additional legal rights, nevertheless ensured
respect and preference in state service. The nucleus of the ruling class consisted of the
emperor's clan, his relations and the greater part of the Tangut bureaucracy. Attached to
them were the rich 'proprietors', landowners and livestock owners who were not in state
service. They exploited the slaves (pkhinga and nini) and ordinary peasants to whom they
evidently leased land. The trend towards the establishment of tenancy relations, which was
also very evident in the neighbouring state of Sung China, was possibly dominant. The
landowners paid a land tax in grain, hay and brushwood and performed compulsory labour
for the state. Land, livestock and other belongings were either the property of individuals
or of the sovereign (the state). In accordance with the traditions of Chinese law, certain
restrictions were sometimes placed on the sale of land by landowners, especially with
regard to the choice of purchasers. Irrigation played a major role in agriculture. The service
responsible for the maintenance of the canals was state-run, and these canals were repaired
and built by people working under the system of compulsory labour. The length of the
period of service depended on the amount of land held by the worker. Livestock-raisers
paid a tax in livestock, wool and milk products. Depending on the number of their livestock,
they supplied both horses and military equipment, or military equipment alone for their
army service. The army was divided into regular and auxiliary forces who provided support
for the activities and guaranteed the fighting efficiency of the regulars. The auxiliary troops
may have included engineers.

218
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

The twelfth century saw the flourishing of Tangut culture. The entire Tripitaka, pre-
served to the present day in material from the dead town of Kara-Koto (rediscovered in
1909 by P. K. Kozlov) and in a small number of texts from the period of the Mongol
Yuan dynasty, was translated into Tangut. Printing by means of wood engraving developed
within the country and a proportion of what was printed were state texts produced by a
special Printing Office. Although Buddhist painting developed under a strong Chinese and
Tibetan influence, it none the less retained an independent identity. Various types of Tangut
dictionaries have been preserved to the present day: the Tangut were proud of their mother
tongue, literature and sayings. In addition to the Buddhist canon and the Confucian classics,
Chinese military treatises and a number of moralizing texts were translated into Tangut. It
may be assumed that there was a fairly high level of literacy among the population of Hsi
Hsia, particularly among officials and, more generally, among people connected with state
service (for which literacy was essential) or with Buddhist communities. Literate people
were apparently bilingual (with a knowledge of Tangut and Chinese) and information has
survived indicating that a knowledge of three languages (Tangut, Tibetan and Chinese) was
required of Buddhists. A clear strain of patriotic pride in their culture and state is evident
in the works of Tangut authors.
During the reign of Renxiao's successor, Chunyu (1193-1206), a new threat appeared
to the north of the Tangut state, the power of Chinggis Khan. At the end of the twelfth
century, the Tangut intervened in the internal struggles of the Mongols and it was possibly
for this reason, as well as their obvious weakness in comparison with the Jurchen state of
Chin, that Hsi Hsia was the first state against which the Mongols conducted campaigns
outside Mongolia. They devastated the western regions of the Tangut state. Defeat in war
against the Mongols cost Chunyu his throne. He was succeeded by Anquan ( 1206-11). The
Mongols attacked Hsi Hsia again in the winter of 1207-8 and in 1209 laid siege to the cap-
ital of the Tangut state. The Tangut endeavoured to establish a military alliance against the
Mongols, requesting aid from Chin, but they received no assistance. Anquan secured peace
by acknowledging himself to be a vassal of Chinggis Khan and by giving him his daugh-
ter in marriage. He saved the country but lost his throne. His successor Eunxu (1211-23)
was drawn into a war with Chin by the Mongols in the year 1214. That war lasted until
1224 and, although it was not conducted intensively, it none the less appreciably weakened
the forces of both states in the face of the Mongol onslaught. In 1217 the Mongols again
besieged the capital of the White High Great State of Hsia (such was the splendid official
title of the Tangut state) but the siege dragged on and they left after demanding that the
Tangut should take part in Chinggis Khan's westward campaign. The Tangut refused and
Chinggis swore to settle his score with them on his return from the campaign. In the spring

219
ISBN 978-92-3-103467-1 Contents THE TANGUT HSI HSIA KINGDOM (982-1227)

of 1226 the main force of the Mongol army, led by Chinggis Khan in person, fell upon Hsi
Hsia. In the autumn of 1227 the Mongols accepted the capitulation of the last ruler of Hsia
and swiftly executed him. The downfall of Hsia coincided with the death of Chinggis Khan
himself and the population of Hsi Hsia was therefore massacred with particular brutality.
Tangut culture perished with the Tangut state and the process leading to the disappear-
ance of the Tangut as a people was initiated. The Mongols replaced the Tangut state with
a Tangut region and the capital of Hsia was renamed Ningxia ('Peaceful Xia' in Chinese).
The Mongol prince who ruled the region of Tangut at the end of the thirteenth century was
a follower of Islam and made every effort to convert the population under his control to
that religion. Various peoples who settled in the Tangut region adopted Chinese as their
lingua franca, accepted Islam and formed a unified group of Chinese-speaking Muslims.
During the Ming dynasty they developed into the ethnic and religious group of Muslims
in north-west China which to the present day constitutes the core of the population of the
Ningxia-Huizu Autonomous Region of the People's Republic of China.

220
ISBN 978-92-3-103467-1 Contents THE WESTERN HIMALAYAN STATES

10

THE WESTERN HIMALAYAN STATES*

A. H. Dani

Contents

The Trakhan dynasty of Gilgit 222

The Maglot ruling family of Nager 225
The Ayash ruling family of Hunza 225
The Kator royal family of Chitral . 225
Baltistan . . . . . . . . . . . . . . 225
Relations with Tibet, Kashgharia and the trans-Pamir regions 227
Relations with Kashmir . . . . . . . . . . 227
Long-term socio-economic developments 228
Socio-religious developments . . . . . . . 231

The western Himalayan states, better described as the trans-Himalayan states, 1 lie prac-
tically south of the ranges of Chinese Turkistan. Sandwiched between high ranges, the
entire region is unaffected by the monsoon climate and protected from northern blizzards.
Although the Indus river runs across it, this does not make for the unity of the region.
The river is too deep and rocky for navigation and its banks are too high and sloping for
normal habitation. The hill ranges subdivide the region into smaller river valleys, with the
human population organized into smaller communities and forming separate cultural units.
Each of them also developed petty states in the course of history; hence we talk in terms
of Himalayan states and not of one state. They all lie in the western Himalayan zone to the
south-west of Tibet. 2

* See Map 7.
1
See Dani, 1989a, Ch. 1, for details.
2
Maqbul Ahmad and Raja Bano, 1984, pp. 194-7.

221
ISBN 978-92-3-103467-1 Contents The Trakhan dynasty of Gilgit

There is no one historical name for the entire region. Leitner coined the term 'Dardistan'
in the nineteenth century, since the languages spoken in the different valleys have all been
grouped under the Dardic family - with the exception of Balti, spoken in Baltistan, which
is affiliated to Tibetan. Jettmar3 has argued for 'Bolor', a term known to Arabic scholars
and also to the Chinese of the medieval period. But its exact connotation is not clear, nor
is it comprehensive enough to embrace the whole region. During British rule, two terms,
'Gilgit' and 'Baltistan', were applied to two political divisions, and 'Chitral' was placed
under the Malakand Political Agency. The Government of Pakistan refers officially to the
region as the Northem Areas, but that excludes Chitral, which is now a district of the
North-West Frontier Province.
The medieval history of the region begins with the emergence of the Turkish ruling
families, whose dates are not certain. 4 However, they appear to have replaced the ancient
dynasty of the Patola Shahis in Gilgit some time in the middle of the eighth century. At
this time, several historical events may be noted in the region. The Tibetans are known
to have advanced into Gilgit via Baltistan. To counter them, a Chinese general of Korean
origin, Kao-Hsien, marched with his army into the region in 743. He restored the rule of
the Patola Shahis, but how long thereafter the Shahis ruled is not certain. According to the
anonymous author of the Hudad az-c alam [The Limits of the World], the king Balurin Shah
regarded himself as the son of the sun god. Al-Biriini calls them Bhatta Shah and refers to
them as Turkic tribes. 5 An Arabic inscription attributed to the time of the c Abbasid caliph
al-Ma'miin (813-33) speaks of a victory over the rulers ofWakhan and Bolor. 6 These items
of information do not clarify the issue of the origin of the Turkic tribes, but they certainly
speak of the change of dynasty that must have taken place long before the time of al-Biruni.

The Trakhan dynasty of Gilgit

According to traditional history, 7 the main ruling dynasty in Gilgit is known as the Trakhans.
It is from this dynasty that the rulers of Nager and Hunza derive their origin. Similarly, the
rulers of Yasin, Punial and Chitral had close connections with them. It is only the history
of Baltistan which had its separate role to play, although that also had links with Gilgit
on several occasions. Tradition traces the origin of the Trakhans to an imaginary Kayani
prince of Persia, by name Azur J amshid, who is said to have fled here after the Arab

3
Jettmar, 1979, pp. 39-70.
4
Dani, 1987, Introduction.
5 Sachau, 1910, Vol. l,pp.207-8.
6
Ghafoor, 1965, pp. 4-12.
7
See Dani, 1987.

222
ISBN 978-92-3-103467-1 Contents The Trakhan dynasty of Gilgit

conquest of Persia and secretly married Nur Bakht Khatun, the daughter of the Buddhist
king Sri Badat. A son was born, who was named Kark or Garg. Although Azur Jamshid is
credited with having overthrown Sri Badat and succeeded to his throne, he chose to abdi-
cate sixteen years later, after handing over the throne to his queen, who ruled until her son
grew up. The son, Kark, held power for fifty-five years and was succeeded by Rajah Sau
Malik.
There was a quick succession of rulers after this event. Rajah Sau Malik was succeeded
by his son Rajah Shah Malik, also known as Glit Kalika (or Malik), i.e. Malik of Gilgit,
followed by Glit Kalika's son Deng Malik, and finally the latter's son Khusraw Khan. The
rule of this last king is said to have ended in 997, a date that may be accepted. Khusraw
Khan married a princess from Badakhshan and the presence of a Badakhshani princess
must have led to some social changes in the Gilgit ruling family. It is important to note that
this Turkic family came from the north; thus there was a strong northern influence in the
royal house of Gilgit by the end of the tenth century.
Rajah Khusraw Khan's son, Rajah Haydar Khan, succeeded to the throne in 997. His
cousin, Shah Hatam (or Shah Tham), governed the Nager and Hunza valleys. Shah Tham
tried to assert his independence, but being pressed by Haydar' s forces he fled to Baltistan
via the Hispar glacier. Rajah Haydar Khan was succeeded in 1057 by his son Nur Khan,
who spent his time in religious devotion and entrusted the work of administration to his
ministers. In 1127 he abdicated in favour of his son Shah Mirza, who died in 1205, leaving
the throne to his son Tartora Khan, who had two queens. The first, called Shah Begam, was
of the royal family and gave birth to Torra Khan, but the second queen, a commoner from
the Darel valley, proved to be jealous and cunning. She managed to poison the king and
snatch the throne for herself in 1236, but ruled only for five years. All her attempts to kill
her stepson Torra Khan were foiled by a minister hailing from the Hodur valley. She met
her death and was succeeded in 1241 by Rajah Torra Khan, who is supposed to have started
the dynastic name Trakhan. Before this ruler the dynasty was known as the Kayanis.
Rajah Torra Khan's reign (1241-75) was marked by great events. His own stepbrother,
Shah Ra'is Khan, born of the Dareli queen, fled from Gilgit and took shelter with the king
of Badakhshan, whose name is given as Taj Mughal, perhaps to be corrected as Tajdar-i
Mughal, who is said to have been a follower of the Ismacili Shfite sect. The Badakhshan
ruler invaded Gilgit, snatched Chitral from Torra Khan and placed Shah Ra'is Khan on Chi-
tral' s throne. This was the beginning of the Ra'isiyya dynasty there. According to tradition,
Ismacilism was introduced into the Gilgit region by Taj Mughal, but this lacks confirma-
tion. In any event, the ruler of Gilgit is said to have abandoned Ismacilism. As a result,
there was a second invasion, but Gilgit forces defended their territory steadfastly. In the

223
ISBN 978-92-3-103467-1 Contents The Trakhan dynasty of Gilgit

course of this campaign, Rajah Torra Khan died; he was succeeded by his son, Sau Malik
II (1275-1345).
The Mongol invasion of Gilgit appears to be related to two invasions of Kashmir, the
first by Dulcha (or Zulchu) in c. 1319 and the second by Urdul (or Achal) in 1326. 8 As a
result of these invasions, there was a dynastic change also in Kashmir and we note there
the foundation of an independent Muslim sultanate by Shah Mir. Seen in this perspective,
this interrelation of events brings the histories of Gilgit and Kashmir together and con-
firms that Sau Malik II and Shah Mir were contemporaries. From this time on, there was
a close relationship between the rulers of Gilgit and those of Kashmir. Sau Malik II was
succeeded by his son Chilis Khan (1345-59). He also enjoyed good relations with Kashmir
and encouraged commerce with the neighbouring countries of Badakhshan, Transoxania,
Kashmir and Afghanistan. He married Malika Hashim Begam, a daughter of Shah Ra'is
Khan of Chitral. His own daughter was later married to Shahzada Shah Khan, a grand-
son of Shah Ra'is Khan. Chilis Khan was succeeded by Rajah Firdaws Khan, who ruled
until 1397. He brought large numbers of artisans and craftsmen from Kashmir and built
the Qilca-yi Firdawsiyya in Gilgit. Firdaws was succeeded by Khusraw Khan II, who ruled
until 1422. He added a tower to his father's fort at Gilgit, which was known as Khusraw
Khan-i Shikar.
Khusraw Khan II had two sons, Rajah Malik Shah and Dula Shah. While the first ruled
from 1422 to 1449, the second became his commander-in-chief. The latter had a handsome
son, called Lili Gashpur, who married the king's daughter. According to tradition, they had
twin sons whose backs were joined together at birth. When they grew up, they became
bitter rivals. One of them called Jamshid, also known as Maglot, obtained Nager as his
possession, while the second (called Sahib Khan alias Girkis) received Hunza. 9 This is the
mythical story of the creation of the Nager and Hunza states.
The next ruler of Gilgit, Torra Khan II (1449-79), son of Rajah Malik Shah, was a
man of great consequence. He continued good relations with Kashmir and was so fond of
Kashmiri arts and crafts that he invited a group of Kashmiri craftsmen to settle in Gilgit in
Mohalla Kishrot. He was followed by a quick succession of rulers, the most important of
whom, Shah Ra'is Aczam, ruled until 1561.

8
Hasan, 1959, pp. 34-7; Sufi, 1949, Vol. 1, pp. 117-18, 128-9; Parmu, 1967, pp. 82-4.
9
For details, see Dani, 1989a, pp. 170-4.

224
ISBN 978-92-3-103467-1 Contents Baltistan

The Maglot ruling family of Nager

The ruling family derived its name from Maglot, a nickname of Prince Jamshid, ruler of
the Nager valley and founder of a royal house from about 1440. At first the Maglot rulers
built a fortified village, called Muko-Kot or Nager Khan, where they lived until 1894, when
it was destroyed by the Nager river. There was a perpetual struggle between the rulers of
Nager and Hunza until Shah Kamal came to the throne in 1559 and began a line which in
general had peaceable relations with the rulers of Gilgit.

The Ayash ruling family of Hunza

The origin of the Hunza royal house has been traced to the earlier Girkis. This state, being
close to Wakhan and Xinjiang, had many historical links with them. As Hunza was always
at war with Nager, their mutual struggle constantly affected the royal house. The succession
to Girkis is variously given in different traditions. Finally, Ayasho (alias Shah Khan) was
brought from Wakhan and placed on the throne. He was succeeded by Silum (or Salim)
Khan. Another account gives Mayhuritham as the next ruler. Mayhuritham fled to Wakhan
and gave his daughter in marriage to the ruler of Wakhan. A son born to them bore the name
of Ayasho II and was crowned in Hunza; he married a Skardu princess, a daughter of c Abd
al-Khan. Ayasho II also sent artisans from Baltistan, who built the palatial forts of Baltit
and Altit in Hunza; the tower of the Altit fort gives the date 955/1548. The local historian
Qudrat Allah Beg accordingly states that, after this marriage of Ayasho II, contacts between
Hunza and Baltistan increased.

The Kator royal family of Chitral

We have earlier seen how Shah Ra'is Khan became the ruler of Chitral and founded the
Ra'isiyya dynasty. This dynasty continued to rule in Chitral for nearly 300 years. Accord-
ing to the unpublished manuscript, the Shahnama-i Chitral, this dynasty came to an end in
the sixteenth century and was succeeded by that of Sangin c Ali. His grandson, Shah Kator
I, founded the Kator dynasty in Chitral in 1585 by usurping power from the Ra'isiyya
family.

Baltistan
Baltistan spreads upwards from the Indus river and is separated from Ladakh by the Siachen
glacier. The river makes a great lake around the city of Skardu, the headquarters of Baltistan

225
ISBN 978-92-3-103467-1 Contents Baltistan

district. Three ruling families were important: the Makpons of Skardu, the Amachas of Shi-
gar and the Yabghus of Khaplu.

THE MAKPONS OF SKARDU

Francke has related the history of the Tibeto-Dard kingdom (500-1000). 10 The involve-
ment of the Tibetans in this region must have followed the consolidation of the dynastic
rule in Tibet by its founder, Songtsen-gampo (d. 649 or 650). We have distinct evidence
of Tibetan inscriptions and the existence of the Tibetan form of Tantric Buddhism all over
Baltistan. Only towards the end of the ninth century or at the beginning of the tenth, after
the dissolution of this dynastic power, could a new power arise in Baltistan. It was linked
with the implantation there of a new ethnic element. Two factors are important here: first,
the expansion of the Uighur Turks in Xinjiang and their interest in controlling the trade
across the Karakorum pass towards India; and second, the establishment of Turkish author-
ity in Gilgit, Hunza and Nager. The local traditions speak of an immigrant from Kashmir,
who, after marrying a local princess, started the line of Makpon rulers of Skardu. 11 Seven
generations of rulers followed the founder of the dynasty in Skardu. This ruler was fol-
lowed in turn by Khokhar Singa, then Ghotachon Singe, next Bahram Shah and finally
Makpon Bokha, whose date of accession to power is given as c. 1500. He is credited with
having been the real founder of the state power of Skardu and he in fact founded the city
of Skardu, so-named because it means 'low land between high places', those of Shikri and
Satpara. His residential seat was at Kharpocha fort, towering high above the surrounding
plain. He is also known to have provided a new socio-economic base for the district by
importing people and craftsmen from Chilas and Kashmir.

THE AMACHAS OF SHIGAR

Shah Tham came to Shigar from Hunza and accepted service with the last local ruler,
Mashido. It is quite possible that he was his minister, i.e. amatya (a word of Sanskrit
origin), from which the family title Amacha is possibly derived, although others take the
title to be of Chinese origin. 12 Ten generations separate Shah Tham and Gori Tham. It
was in the time of this last ruler that Sayyid c Ali Hamadani came to Shigar, converted the
local ruler, spread Islam and built the Ambariq mosque. Gori Tham was followed by Ghazi
Tham, then c Ali Mir and then Ghazi, who ruled from 1490 to 1520. He was a contemporary
of Makpon Bokha of Skardu and it was during his time that the famous Sufi, Shams al-Din
1
° Francke, 1907, Chs 3, 5.
11
Vigne, 1842, Vol. 1, p. 251; Biddulph, 1977, p. 144.
12
See Thomas, 1935, Vol. 2, p. 191.

226
ISBN 978-92-3-103467-1 Contents Relations with Kashmir

clraqi, came to Shigar; he died there in 1525. Subsequent rulers became subject, in the
seventeenth century, to the Mughal emperors in Delhi.

THE YABGHU RULERS OF KHAPLU

The very title Yabghu suggests a Turkish origin for the rulers of Khaplu. Tradition gives
the history of these rulers as starting with Bag Nanthal in c. 850 and continuing with a list
of nineteen rulers up to Aczam Khan Shah. In the fifteenth century an invasion of Baltistan
by the Timurid Abu Sacid Sultan is mentioned.

Relations with Tibet, Kashgharia and the trans-Pamir

regions
It was after the expulsion of the Tibetan forces from Gilgit and Baltistan that the new states
of the medieval period came into existence. Thereafter, the Gilgit region was almost cut
off from Tibet, but the Baltistan states were in close contact with Ladakh, which was then
a close ally of Tibet. The rulers of Skardu and Khaplu waged many wars against those
of Ladakh. The state of Hunza and the Baltistan states were likewise in direct contact
with Kashgharia. In the earlier history of Hunza, the Hunza princes went to Yarkand and
obtained territorial rights in the Raskam area. The most important event, however, was
the invasion of the Timurid Abu Sacid Sultan into several parts of Baltistan, particularly
Khaplu and Shigar, and the imposing of his temporary suzerainty. It was probably threats
from the north that later persuaded the Mughal emperors of Delhi to advance into Kashmir
and still later into Baltistan and Ladakh. But the most important relationship of the local
rulers was that established with the royal family of Badakhshan.

Relations with Kashmir

In the medieval period, there was no warfare between Kashmir and the states of this region,
although some of the local rulers on occasion fled to Kashmir, and with military help from
there, were able to reconquer their former kingdoms. On the other hand, right from the
beginning there were numerous cultural and commercial relations with Kashmir. We have
seen earlier how the rulers of Gilgit and Baltistan imported artisans and craftsmen from
Kashmir and settled them there, and these introduced the Kashmiri style of architecture
and also brought in decorative elements to ornament the wooden mosques and tombs. Of
great cultural and religious importance was the advent of the Sufi Shaykh Sayyid cAll

227
ISBN 978-92-3-103467-1 Contents Long-term socio-economic developments

Hamadani and his disciples through Kashmir into Baltistan, thereby injecting a new ele-
ment of mysticism into the religious life of the local people.

Long-term socio-economic developments

All the ruling families from Chitral to Gilgit, Hunza, Khaplu and Skardu were of Turkic
ethnic origin, although they came to speak the languages of the regions where they settled.
These Turkic tribes migrated here soon after the Arab conquest of Central Asia in the
eighth to the ninth century, retaining at this time their shamanist and Buddhist beliefs; only
later were they converted to Islam in this very region. However, the local population is not
ethnically Turkic, nor does it have any ethnic relationship with the Chinese. Except for
the Balti-pa, i.e. those who speak the Balti language in Baltistan and Burushaski, they all
belong to the lndo-Aryan group as far as their languages are concerned. In this wide group,
the particular sub-branch has been termed Dardic. This penetration of lndo-Aryan tribes
into the far-flung region of western trans-Himalaya has kept Chinese political and cultural
influence out of this region. Moreover, the Aryan character of the land has been further
shielded by the Turkic peoples of the neighbouring Xinjiang province of western China.
It is likely that the relationship with these Turkic tribes has been a factor in the cultural
connections of the population of this part of these areas and Xinjiang; this may explain the
acceptance of Turkish royal families as rulers.
Apart from the ethnic penetration from the north, however, there is a deeper linkage with
the Buddhist and Hindu south as far as cultural and religious aspects are concerned. Yet
this cultural penetration did not establish close, friendly relations with Kashmir, probably
because of political rivalries. Linguistic, palaeographic and even socio-cultural differences
can be traced to this one main factor. On the other hand, a connection with Taxila and Gand-
hara and even with the Indo-Gangetic plains is more firmly based. In spite of these discrep-
ancies, the influence of Sayyid c Ali Hamadani and his disciples from Kashmir became an
enduring feature in the Islamic penetration of the region. Yet we find northern influences
imperceptibly breaking the cultural unity, and it is this peculiar feature that explains the
distribution pattern of Muslim sects in the northern areas of modern Pakistan.
At present, three distinct socio-political patterns are easily recognizable. Starting from
the west, we first meet with the so-called 'republican' tribes localized in the valleys of
Tanigr, Darel and Chilas. These may be extended still further south-west on either side of
the Indus river, which, in British times, was referred to as Yaghistan (Land of Rebellion,
i.e. no-man's-land), but is today included in Kohistan district. This is the hard-core area
of the orthodox Sunni Muslims. Some early epigraphic evidence reveals the existence of

228
ISBN 978-92-3-103467-1 Contents Long-term socio-economic developments

small kingdoms here, but at some point in the medieval period these kingdoms disappeared
and the different tribes established their own tribal system of political management, which
the British Indian records characterized as 'republican'. These political units had no kings,
but tribal elders negotiated political relations with superior military powers. Local defence
was organized by able-bodied young men hailing from various families. When these tribes
moved into Gilgit, however, they accepted the authority of the local raja. It appears that
the rajas of Gilgit, Yasin and even Chitral raided into this territory, and, depending on their
might, exercised control over the area. It was probably under these circumstances that these
tribes moved into Gilgit. At a later period the British encouraged the migration of Pashtun
tribes from the direction of Swat into the no-man's-land in order to create a manageable
political force in that area. It was probably as a result of this Pashtun penetration that the
jirga system of tribal assemblies became widespread in this sub-region. Even the ruler of
Swat advanced right up to Kandia valley with his soldiers, and finally the western part of
the Indus river fell under his control. However, the social system of the indigenous local
tribes is fundamentally different from that of the Pashtuns.
The second important sub-area contains the Burushaski-speaking population in Hunza.
(There is also the Wakhi-speaking population, who migrated from across the Pamirs at a
much later date.) The Burushaski-speaking population had a much wider distribution in the
past but, according to the latest research, they have been pushed into their present home by
the Shina-speaking population. The Burushaski population is fundamentally different in
behaviour patterns, attitudes, traditional customs, and even in social categorization, from
the population in Gilgit, Chilas and other western areas. Burushaski society is freer. The
women have greater mobility and better social standing and status; they move freely in
society and share equal responsibility with the men. The ruling families and their associated
elites have established a political authority, but the gulf between this upper structure and
the mass of the Burushaski population is easily recognizable to anyone who lives in this
society and notes the great influence exercised by the local population on the ruling elite.
The difference between the Hunzakuts on the one hand, and the Gilgit and Chilas on the
other, was so great that early British Indian writers applied the term 'Yashkum' to the
Hunzakuts and 'Shin' to the western tribes, although these two terms have nothing to do
with ethnic groups; they have, rather, socio-economic and feudal connotations. It is this
Burushaski-speaking population who are now Ismacilis.
The third important sub-area is Baltistan, which is a melting-pot of influences from
Tibet, Ladakh, Kashmir and even beyond from the further Indian side. The Tibetan and
Ladakhi influence is largely seen in the local language, geographic names, physical types
and probably also in the social behaviour patterns. As it is open to Kashghar through the

229
ISBN 978-92-3-103467-1 Contents Long-term socio-economic developments

Soltoro and Karakorum passes, the northern influence is no less important. Its relationship
with Kashmir, however, is known to date back to the Mughal period. Nevertheless, the rul-
ing families have had a blood relationship with those of Hunza because the Hispar glacier
on the north permits a direct route from Shigir to Nager and Hunza. With all these geo-
graphic features and political connections, it is mainly Kashmir and Kashghar that have
left an indelible influence as far as Islam is concerned. That is probably because Baltistan
lies midway on the direct route between Kashmir and Kashghar. Although as a result of
Kashmir's influence Islam arrived here quite early, it is probably the Kashghar channel that
introduced both Shicism and the Niirbakhshi Sufi order, the two dominant religious currents
that now exist. It would be interesting to discover how much of the Tibetan social system
is still current among the original population of this sub-area. But the ruling elite, which
extended its political authority here, made many compromises with the local population.
The three divisions given above show the socio-cultural cleavage between the Shina-
speaking people and the Burushaski-speaking people; at the same time, these two differ
from the Baltis in physical type as well as in various aspects of cultural life. In social
classification, apart from the professional groups, the two dominant classes are those of
Shins and Yashkuns, who exhibit a deep-seated rivalry and hatred for each other. The origin
of this hostility is not known, but it appears to be rooted in some kind of landowning over-
lordship, a system that has survived from the pre-Muslim period. These two categories are
not much known in Baltistan. As against them, there is the class known as dams, who are
low-class immigrants from Kashmir.
As the traditional economy depended primarily on agriculture and pastoralism, the two
systems are unevenly distributed in the different valleys. River irrigation is not much devel-
oped since it is difficult in a hilly country. Instead, melted glacier water is distributed and
channelled from great heights to the lowest-level fields, which range in terraces from top to
bottom. The amount of such cultivable land is very limited, hence there have been constant
wars among tribes for the possession of land. Ownership is common rather than individual,
but within the common lands, individuals have their own rights and privileges. It is these
rights that are disputed by Shins and Yashkuns. However, common ownership is limited
within a particular valley. It is this system that has led to perpetual rivalry and wars; the
enmity between Nager and Hunza is proverbial. Similarly the Chilasis, Gilgitis and people
of Yasin have traditionally been bitter rivals.

230
ISBN 978-92-3-103467-1 Contents Socio-religious developments

Socio-religious developments
At present, the entire population of the Northern Areas is Muslim, but Islam did not spread
here all at one time and from one direction. Islamization was a gradual process, and as a
result of the influences from the neighbouring areas of north, south and west, the nature
of Islamic beliefs and practices differs from one sub-region to another. These practices are
themselves greatly influenced by surviving cultural rituals and behaviour patterns inherited
from the past. Thus we may note the legacy of pre-history in the form of rock art that has
continued up to the present day; in the beliefs in fairies and their hold in the minds of the
local people; in the reverence paid to the standing rocks, monoliths and high mountains,
including the devastating effect of the glaciers; in the animal sacrifices, ceremonies and
dances that are quite distinct from the Islamic cld-i Qurban; in the manner in which graves
and tombs are built, and the dead interred there; and in the local dress, ornaments, forms
of pots and pans, and many household objects that are firmly interwoven into the pattern
of a socio-economic life that is rooted in the horticultural and pastoral productive system
of the region. Each one of these aspects deserves a special study, as they have continued to
characterize socio-cultural practices differently in the different sub-regions.
Today, the Muslims of the region are either Sunni, Ismacili, Twelver Shicite or adherents
of the Nurbakhshi Sufi order. There is no definite information on the advent of Ismacilism
before c. 1830. However, the disciples of Sayyid cAli Hamadani reached the region in
late medieval times. Credit must go to Shams al-Din clraqi and his family members, now
lying buried in Skardu, Shigir and Kiris in Baltistan, for their continued missionary activity
in this area. Although it is disputed whether Shams al-Din was a Nurbakhshi adherent
or a Shfite, his connection with Kashmir is well established. There is another family of
missionaries - Mawlana Sayyid Mahmud Shah Tusi and his brother, Mir Sayyid cAli Tusi,
now buried at Skardu and Kuwardu respectively - who are known to have come from
Persia through Yarkand in the sixteenth century and are mainly responsible for the spread
of Shfism. It is from Baltistan that five preachers - Sayyid Sultan cAli cArif, Sayyid Shah
Wall, Sayyid Shah Afdal, Sayyid Akbar Shah and Sayyid Shah Ibrahim - are known to
have travelled towards Gilgit and spread Shicism there; their tombs are found at Gilgit,
Danyor and Gulmit.

231
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

11

THE KITAN AND THE KARA KHITAY* 1

D. Sinor

Contents

The rise of the Kitan 232

The rise of the Kara Khitay 239

The rise of the Kitan

Kitan history does not fit easily into the usual pattern of national histories. The subject of
this chapter is not the history of one nation-state defined by more or less stable geographic
frontiers; rather, we shall follow the destiny of one people which, though maintaining its
feeling of identity over a period of some eight centuries, has not only twice changed its
country but has also undergone a remarkable cultural and linguistic metamorphosis.
Before the fifth century, under a different name or names, the Kitan belonged to the
heterogeneous group called by the Chinese T'ung Hu (Eastern Hu), where Hu may be
considered a generic name for barbarians. Chinese sources first mention the Kitan, under
this name, in connection with a defeat inflicted on them in 406 by the Chinese Wei dynasty.
At that time the Kitan lived on the upper course of the Liao river, but the military reverse
prompted them to move into the Sung-mo area of what was to become Jehol, the north-
eastern province of China. The Wei-Kitan conflict was, one might say, almost a family
feud. The Northern Wei dynasty (386-538) was of alien (Tabgach) origin, a name better
known in its Chinese transcription: t' o-pa. The Tabgach, Kitan, and Shih-wei were all

* See Maps 2 and 5, pp. 427 and 432-3.

1
Or Karakitay. Editor's note: the form 'Kara Khitay' is used throughout this Volume for consistency with
the Islamic sources.

232
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

detached clans of the Hsien-pi. Their language was Mongol with some archaic features
and they may have spoken different dialects.
The Kitan tribes occupied a region which incorporated the eastern slopes of the Khingan
mountains and also the plains crossed by the Shira Muren. Their ancestral territory was
marginal to the rich pastures of the steppe but still suitable for pastoralism. The Kitan
took full advantage of this, concentrating on horse-breeding, with horned cattle and sheep
being of secondary importance. There was also land suitable for agriculture, and the forests
and rivers of Manchuria provided for extensive hunting and fishing, both activities being
pursued on a grand scale. More importantly, the Kitan's country was an ore-rich land;
mining and metallurgy in general were to play a major role in Kitan history. All in all,
the land on which they lived allowed the Kitan to develop a diversified economy, with a
modicum of industrialization unparalleled anywhere in medieval Inner Asia.
From the outset, the Kitan appear as a conglomerate held together by economic or
political interests rather than by sharing the same culture. The composite character of their
civilization is reflected in their ancestral legend, according to which the Kitan descended
from the union of a man riding a white horse along the Lao-ha river and a maiden travelling
in a cart drawn by a grey ox along the Shira Muren. At Mount Mu-yeh, at the confluence
of the two rivers, the man and the woman united; the descendants of their eight sons were
to form the original eight tribes of the Kitan. It should be noted that the text of the Liao-
shih [History of the Liao] which provides this information allows the interpretation that
procreation resulted from the union of the two rivers. A white horse and a grey ox remained
sacrificial animals among the Kitan, probably symbolizing the two main components of
their culture. The motif of the birth of a mythical ancestor at the confluence of two rivers
appears also in connection with Buqu Khan, legendary ruler of the Uighurs.
Fragmentary legends recorded by understandably sceptical Chinese historians would
suggest that at least some of the early Kitan rulers led a segregated life and remained inac-
cessible to the people. One was said to be a mere skull, covered with felt, emerging from
his tent and taking human shape only for the ceremonial occasions when the white horse
and the grey ox were sacrificed. Shamans played an important role in the ceremonies con-
nected with these sacrifices. The habit of covering the ruler with a felt rug survived in the
investiture ceremony of the later Kitan (Liao) emperors. Another ruler, equally sequestered
and appearing only when the need arose, was reported to have had a boar's head and was
dressed in pigskin; clearly he was perceived as a pig, an animal which places at least some
of the Kitan tribes firmly in a forest-dwelling, eastern Tunguz cultural group. Both rulers
vanished when their secret identity was improperly interfered with. A third mythical ruler
was said to have owned twenty sheep. Each day he ate nineteen of them, but the next day

233
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

there were twenty again. The story of 'King Skull' and 'King Boar' permits the assump-
tion that early in their history, the Kitan were ruled simultaneously by two men: one, a
purely ceremonial, sacral king living in seclusion; the other, perhaps an elected individual,
in whose hands real power lay.
In the middle of the eighth century, the Kitan are often mentioned in the exploits of the
celebrated rebel Ngan Lu-shan, sometimes as enemies and at other times as allies. By the
ninth century, the eight Kitan tribes were governed by a chieftain elected from among the
members of the Yao-lien family for a period of three years. This term could be cut short
if, while he was at the helm, extraordinary calamities struck the community. The Yao-lien
family gradually lost its grip, and in 872 A-pao-chi of the Yeh-lti (also called I-la) clan
was elected. He became the founder of the Kitan empire. Twice re-elected, at the end of
his third term A-pao-chi was reluctant to abandon the reins of power, and, with the consent
of the tribes, organized an independent tribe centred on the 'Chinese City', situated on the
Luan river. Inhabited by Chinese who had fled the disorders which marked the fall of the
T'ang empire, the region was not only suitable for agriculture, but also had considerable
mineral resources such as iron, copper, gold and salt. A-pao-chi's father had already a
keen perception of the advantages to be derived from industrial progress. He had paid
particular attention to the development of metallurgy and, combining the skills of the Shih-
wei tribe (reputed specialists in working metal) with those of the recently settled Chinese,
he established iron smelteries. Perhaps even more important was A-pao-chi's acquisition
of the lake which provided the salt used by the Kitan tribes. The economic power of A-
pao-chi enabled him to gain the support of an ever-increasing group of Kitan; the killing
in an ambush of all the rival chieftains secured his power. In 907 A-pao-chi assumed the
Chinese name T' ai-tsu and declared himself emperor of the Kitan. It should be noted that
there is some doubt whether A-pao-chi ever used the title Kaghan, certainly not current
among the early Kitan rulers.
A-pao-chi (T'ai-tsu) thus became the first ruler (907-26) of a dynasty which in 947
took the name Liao - adopted in memory of the Kitan homeland located at that river -
and ruled over northern China until 1115. The investiture of the ruler followed practices
well attested among other Inner Asian peoples. In the course of the ceremony the future
emperor gallops off, 'falls' from his horse, to be covered - as mentioned above - with a
felt blanket. In the Turk ceremony of investiture, as recorded by the Chou shu, the order
of events was reversed, the future Kaghan being first bundled in a felt rug and then put on
horseback. In Kitan practice, having faced what must be viewed as an ordeal, the ceremony
culminated in the emperor being raised on a felt carpet.

234
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

The Liao-shih also records the precursory signs marking the birth of a proper founder of
a dynasty. Born of a sun-ray, so our sources tell us, A-pao-chi had the body of a3-year-old
and could, from the moment of his birth, crawl on all fours. It took him but three months
to walk and one more year to speak and to foresee the future. He was undoubtedly an
exceptional man: the ancestral hero of an ascending dynasty. Similar signs of precocity
characterize the legendary Oghuz Kaghan, eponymous ancestor of the Oghuz tribes.
The diplomatic skills of A-pao-chi were considerable. In the words of the Liao-shih, he:
treated all tribes kindly, made the rewards and punishments equitable and abstained from
wanton military campaigns. He benefited his people by pursuing their interests. The herds
flourished and both the government and the people were sufficiently provided for. 2

Kitan history did not unfold in a vacuum and, from the very beginning, the relationship
with China was the dominant factor in external relations. The first recorded Kitan embassy
reached the Wei court in 468. In 479 some Kitan tribes, fleeing a Juan-juan invasion, sub-
mitted to the Wei. Contacts became more frequent in T'ang times when, as we have seen,
the Kitan were increasingly inclined to take advantage of the weakening of central power.
A-pao-chi's rise coincided with the fall of the T'ang dynasty, followed by a period of politi-
cal upheaval known in Chinese history as that of the Five Dynasties (907-60). The growing
importance of the Kitan is shown by the fact that in 906 the founder of the short-lived Liang
dynasty (907-12) thought it opportune to send envoys to A-pao-chi.
It was with Kitan help that in 936 the Later Chin (one of the Five Dynasties) could
establish their rule. In 938 they had to repay this service in kind by ceding to the emperor
T'ai-tsung (T'ai-tsu's successor; 927-47) sixteen prefectures located in northern China,
thereby extending Kitan territory well into China proper. In 938 Peking became the Kitan
Southern Capital (Nan-ching) for which in 1012 they revived its old name, Yen-ching.
With the founding of the (Northern) Sung dynasty in 960, a barrier to further south-ward
Kitan expansion was erected, and the treaty of Shan-ytian (1005) established a long-lasting
peace on the basis of equality between Liao and Sung. The history of Liao rule over China
is beyond the scope of the present chapter.
There is evidence of Kitan contacts with the great nomad empires centred on Mongolia.
In the first part of the sixth century, under attack by the Kao-chti-li (called somewhat later
Kao-Ii, i.e. the Koreans) and the Juan-juan, at that time the dominant power in Mongolia,
some Kitan sought refuge with the T' o-pa dynasty of the Northern Wei. Gaining strength,
the Kitan then invaded the short-lived Northern Ch'i dynasty (550-70) but in the process
they suffered the loss of some 100,000 men. By that time the Turks held sway in Mon-
golia. The KOl Tegin inscription in Mongolia lists the Kitan among the peoples hostile to
2
Wittfogel and Feng, 1949, p. 127.

235
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

the Turks though they were represented at the funeral of that Turk prince. The Tonyuquq
inscription, of a slightly later date, speaks of a Tokuz Oghuz-Chinese-Kitan triple alliance
against the Turks. It is impossible to establish the exact nature of Turk domination over the
Kitan. Interestingly, Kitan-Turk hostility survived the fall of the Turk empire of Mongolia;
we know of Kitan attacks against the Western Turks in 916 and 983. It is not clear what was
meant by the 'Turks' since by that time their states, both Eastern and Western, had long
since disappeared. However, it is known that some Turk groups survived the dissolution
of their empire. In the middle of the tenth century, Hu Ch'iao, a Chinese in Kitan service,
mentions two groups of Turks: one, called the 'Turks of the shan-yu', living to the west
and north of the Kyrgyz; the other far to the north in a very cold country. It seems likely
that it was with the former group that the Kitan had their conflicts.
Relations with the Uighurs, whose empire succeeded that of the Turks in Mongolia in
745, were more friendly and more complex. Theoretically at least, the Kitan were sub-
jects of the Uighur empire of Mongolia. But Uighur rule may have rested light on the
Kitan, as may be concluded from the following anecdote. In 924, on the occasion of a
visit to Karabalghasun, the old capital of the former Uighur empire, A-pao-chi ordered
that an inscription in honour of the Uighur Bilge Kaghan be erased and replaced by a
trilingual (Kitan-Turkic-Chinese) text extolling the deeds of the defunct Kaghan. If the
historian may be allowed to suggest a seemingly far-fetched analogy, many former sub-
jects of empires that no longer exist - such as those of the British or of the Habsburgs -
look with a measure of nostalgia on the days of their 'subjugation'. It was in this frame of
mind that A-pao-chi offered the Uighurs of Kocho the possibility of returning to their for-
mer homeland from which they had had to flee when, in 840, the Kyrgyz put an end to their
empire in Mongolia. By 924 the Uighurs had been settled for almost a century in a land
where they were content and so they did not take up the offer. Unfortunately the inscription
erected by A-pao-chi has not survived nor do we know which of the several Bilge Kaghans
it was intended to honour; the text might have thrown light on the conjectures just adum-
brated. The date of 924 is often referred to in the scholarly literature as that at which the
Kitan, having defeated the Kyrgyz, took possession of Mongolia. It may be more accurate
to suggest that A-pao-chi took advantage of a political vacuum created by the gradual with-
drawal of the Kyrgyz into their homeland in the Yenisei region. What is certain is that by
924 the Kyrgyz evacuation of the Orkhon region must have been completed. Kitan-Uighur
political co-operation continued over the years and, almost a century later, a joint diplo-
matic action was undertaken aimed at establishing contact with the Ghaznavids. About this
more will be said later in this chapter.

236
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

Beyond strictly political links, the Kitan and the Uighurs were connected in more than
one way, particularly following the Uighurs' expulsion from Mongolia. The Kitan tribal
confederation comprised several tribes of Uighur origin, such as the 1-shih, second in rank
only to the imperial Yeh-lti clan. Also of Uighur origin was the Hsiao clan which provided
consorts to the ruling Yeh-lti. The power of the consort was almost as great as that of
the emperor. The empress accompanied her husband at ceremonial hunts and participated
in public ceremonies such as the annual sacrifice at Mount Mu-yeh where, it should be
recalled, not only the male but also the female ancestor of the Kitan was duly honoured.
T'ai-tsu's widow refused to follow her deceased husband into the grave- as tradition would
have demanded - and got away with it. Her act of defiance marked the end of the custom.
Among the powerful women in the Liao-shih mention should be made of Hu-lien, wife of
the second son of T'ai-tsung. In 994 she led a campaign against the Tatar and ten years
later founded K'o-tun ch'eng, i.e. the 'city of the Khatiin [the empress]' near the Orkhon
river. Kitan women of her stature presage the long-ruling female regents of the Kara Khitay
and the forceful Chinggisid consorts of the thirteenth century.
The Liao-shih records for 924 the arrival of Uighur messengers to the Kitan court. Since
the short notice raises several questions, it is given here in the translation of Wittfogel and
Feng:

Uighur messengers came [to the court], but there was no one who could understand their
language. The empress said to T'ai-tsu, 'Tieh-la [a younger brother of the emperor] is clever.
He may be sent to welcome them.' By being in their company for twenty days he was able to
learn their spoken language and script. Then he created [a script of] smaller Ch'i-tan charac-
ters which, though fewer in number, covered everything. 3

The first, somewhat surprising, fact that emerges from this notice is the seeming ignorance
of the Uighur language in the Kitan court. Yet one should not attach too much importance
to the statement, which may simply indicate that, at the precise time of the messengers'
arrival, and in the emperor's immediate entourage, there was no one who could speak
Uighur. For all his cleverness, Tieh-la could not have learned Uighur in twenty days; it
must be assumed that he knew it already.
The second part of the statement has been widely interpreted to mean that the 'small'
characters of the Kitan script were, in fact, those of the script used by the Uighurs. Recent
research 4 indicates otherwise. The Kitan created two scripts, at present neither of them
completely deciphered, which in the sources are called respectively 'large' and 'small'.
Texts written in each of them have survived, and to the layman's eye both are similar to

3
Wittfogel and Feng, 1949, p. 243.
4
See Ch. 9 by Gyorgy Kara in Vol. IV, Part Two, of this History

237
ISBN 978-92-3-103467-1 Contents The rise of the Kitan

Chinese characters. The superficial similarities notwithstanding, the systems underlying

the two scripts are different and the 'small' script has nothing to do with any of the scripts
used by the Uighurs. Tieh-la is probably justly credited with the invention of the 'small
script'; the link with the arrival of the Uighur messengers is coincidental.
Uighur cultural influence on the Kitan manifested itself in various ways and included
the teaching of the cultivation of melons, a rather sophisticated process, since it involved
the fertilizing of the plants with cow dung and their protection with mats. Important and
cordial as Kitan-Uighur relations may have been, the Kitan took care to keep their distance
from the politically insignificant Uighurs of Kocho. On two occasions Uighur requests for
imperial brides were rejected.
The rise of Kitan power did not go unnoticed in the Islamic world. In 923 a Persian
delegation, and in 924 an Arab embassy, arrived at the court of T'ai-tsu. The Kitan seem
to have been slow to respond to these overtures; the first recorded Kitan embassy to the
west visited Ghazna in 1026 or 1027. It was clearly prompted by the news reaching the
Kitan that Mahmud of Ghazna (998-1030) had embarked on a series of conquests which
brought the eastern limits of the Ghaznavid state close to the Kitan sphere of interest. The
Liao emperor Sheng-tsung (982-1031) decided to send an exploratory mission which was
joined by an Uighur mission. Arabic sources mention their arrival which they interpret
either as proof of the Kitan's fear of Ghaznavid might or even as their desire, in Gardizl's
words, 'to place themselves at his service'. In his work written c. 1120, the geographer and
ethnographer Marvazi gives the text of Sheng-tsung's letter to Mahmud:
The Lord of Heavens has granted to us (many?) kingdoms upon the face of [this] wide earth
and placed us in possession of regions occupied by numerous tribes. In our capital we enjoy
security and act according to our will. Anyone in the world who can see and hear cannot
help seeking friendship and close relations with us. Our nephews from among the amirs
of the nearer regions constantly and without exception send their envoys, and their letters
and presents follow upon one another. [Only] he [Mahmud] until now has sent no envoy
or messenger, while we hear of his excellence in strength and courage, of his outstanding
position in might and elevation, of his supremacy over the amirs by awe, of his control of the
provinces by might and authority and of his peace in his homeland according to his own will.
As he enjoys such a glorious position it is a duty for him to write his news to the Supreme
Khan than whom there is none higher beneath the heavens, and to treat him with consideration
according to his state. So we have taken the initiative, limiting ourselves to the dispatch of
this lightly equipped envoy rather than someone who would exceed him in rank and equipage,
in view of the greatness of the distance and the length of time [necessary] for covering it. 5

Marvazi also gives the text of the message of the Uighur Khan, much more modest in tone
but equally friendly. 'We ardently desire', writes the Khan, 'that love and respect should be
5
Minorsky, 1942, pp. 19-20.

238
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

established between [us].' Judging by Marvazl's text, Mahmud was certainly not overawed
by the Kitan initiative. In Marvazi's words:

When the two letters were presented to Mahmud and he saw what stupidity they contained...
he did not find it possible to grant what was requested with regard to the establishment of
sincere relations and correspondence, and he dismissed the envoys, saying to them: 'Peace
and truce are possible only so far as to prevent war and fighting. There is no faith uniting
us that we should be in close relations. Great distance creates security for both of us against
perfidy. I have no need of close relations with you until you accept Islam. And that is all.' 6

Harsh though Mahmud's reply was, it certainly contained a core of realism. There was
little likelihood of a Ghaznavid-Liao conflict.
The fate of the Liao was not sealed by any distant menace, nor by a sudden change in the
modus vivendi established with the Sung. The danger arose in the Kitan's own Manchurian
hinterland where the Tunguz Jurchen, for many years subjects of the former, revolted
and under the leadership of A-ku-ta (1068-1123) conquered most of northern China. The
short-lived Jurchen dynasty of the Chin (1115-1234) was established byA-ku-ta's younger
brother, known as the emperor T' ai-tsung.

The rise of the Kara Khitay

Thus, following a rule of just over two centuries, Kitan dominion over northern China
ended. Needless to say, the demise of the Liao did not imply the disappearance of the Kitan
people in its homeland or even in the conquered territories. The integration of the Kitan
into the new Jurchen Chin state does not belong to the history of Central Asia and, hence,
cannot be dealt with in this chapter. We must, however, follow the destiny of the Kitan in
Central Asia where, showing a remarkable resilience, they embarked on the foundation of
a new state, the third in the course of their known history.
The success of the enterprise was due to the extraordinary determination, as well as
the political and military skills, of Yeh-lti Ta-shih, a descendant in the eighth generation
of T' ai-tsu. A highly educated man, Yeh-lti Ta-shih was put on his mettle in a number of
military engagements with the Sung in which he revealed an uncommon military talent. He
fully realized the plight of the Liao, taken in the fearful pincer of the Jurchen Chin advanc-
ing from the north and the Sung from the south. With more than half his territory occupied
by the Jurchen, and one major city after the other taken, the emperor T'ien-tsu had to flee
and in 1125 was eventually captured by the Chin. In the general maelstrom of war, amid
intrigues and counter-intrigues and contending, ephemeral, self-styled emperors, Yeh-lti

6
Minorsky, 1942, p. 21.

239
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

Ta-shih operated with great skill and a good measure of ruthlessness, putting expediency
above every other consideration. Captured by the Chin, and perhaps toying with the idea of
collaboration with the victors, he escaped and made his way to T'ien-tsu. Appalled by the
emperor's incompetence and unwilling to undertake military operations that were doomed
to failure, in 1124 - while T'ien-tsu was still alive - Yeh-lti Ta-shih set out with a small
band of followers to create a state of his own. According to Chinese sources, he started
out with 10,000 horses. Since one must count at least 2 horses per warrior, his following,
though fast growing, was very small indeed.
There is some doubt whether, at this moment, he took the title of king or emperor; but
what is certain is that, simultaneously, he had himself called Gtir Khan, a dignitary title
he was the first to use. It is a composite of the conventional term Khan and the adjective
gur (universal). It could well be that by the adoption of an unusual title, at the outset
of his second career, the Kara Khitay ruler wished to distinguish his regime from those
customary in Inner Asia. There is a twentieth-century parallel in Hitler's and Franco's
choice of the unconventional terms Fuhrer and Caudillo respectively. In later years, the
title Gtir Khan was assumed by Jamuka, an erstwhile companion of Temtijin (later to be
known as Chinggis Khan), who became his principal antagonist. Gtir Khan was also the
title of the ruler of the probably Turkic-speaking Kerait, who in or about 1008 adopted
Nestorian Christianity.
Among the many qualities of Yeh-lti Ta-shih, his powers of persuasion must have ranked
second only to his steely determination. As a fugitive with no territorial base, in order
to gain any credibility he had to raise an army willing to follow him on a road not yet
charted leading to a state not yet founded. That he brilliantly succeeded in this undertaking
gives the measure of his genius. First, Yeh-lti Ta-shih turned north where he obtained the
modest support of the White Tatars. He then advanced towards Mongolia where, inK' o-
tun city by the Orkhon, to an assembly of tribal chiefs and heads of Kitan prefectures, he
delivered a rousing speech requesting their help in the extermination of the Chin and the
re-establishment of the Liao empire. It is a matter for speculation whether, at that time,
Ta-shih's prime objective was the restoration of the Liao or the creation of a new empire
resting on the military resources of the steppe peoples. His immediate further actions would
suggest that he opted for the second alternative.
As an outcome of the successful meeting in K' o-tun, Yeh-lti Ta-shih now had an army
of some 10,000 horsemen - a respectable force when well led, but insufficient for large-
scale operations against the Chin or any other major power. There was a need for fur-
ther allies and, to avoid risky confrontations, Yeh-lti Ta-shih decided to proceed westward.
The sources report that before embarking on the long expedition, Yeh-lti Ta-shih dutifully

240
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

performed the traditional sacrifice of the grey ox and the white horse, a telling sign of Kitan
cultural continuity carried from the Manchurian forests, through two centuries of rule over
China, towards the steppes of Central Asia. In 1125 - the date is uncertain - he requested
permission from the Uighur ruler of Kocho to pass through his territory towards the west.
Referring to the erstwhile Kitan offer that his ancestor T' ai-tsu had made to the Uighurs to
re-establish them in Mongolia, Ta-shih emphasized the time-honoured friendship between
Uighurs and Kitan. The Uighur king of Kocho agreed to the passage of the Kitan troops
through his territory. According to the Liao-shih:
He presented him with six hundred horses, a hundred camels, and three thousand sheep; and
to prove his sincerity gave some of his sons and grandsons as hostages; and declaring himself
a vassal of Ta-shih, accompanied him as far as the boundary of his realm.7

Whether the co-operation between the kingdom of Kocho and the Kitan was really as
harmonious as the sources wish us to believe is of secondary importance. It can be taken
for granted that, at a minimum, the advance of Yeh-lti Ta-shih towards the west was not
impeded by the Uighurs. How long this crossing took cannot be established; it might have
taken a few years during which Yeh-lti Ta-shih prepared his forces for the conquest of West
Turkistan.
Before we follow this Kitan advance, we should pause for a moment to examine a
change in their name that occurred almost concurrently with this migration. First, it should
be mentioned that, although Chinese historians shed few tears over the fall of alien, con-
quest dynasties, that founded by Yeh-lti Ta-shih has continued to be listed also in the Chi-
nese annals under the name of Hsi (Western) Liao.
For reasons unknown, in Muslim sources the Kitan who migrated to the west appear
under the name of Kara, i.e. 'Black' Kitai or Kitay. The first question that emerges is: why
has the final -n in the name been replaced by a -y? (The value of the final -i as written in the
conventional spelling of Kitai is, in fact, that of a -y as in the English 'yes'.) The 'Kitan'
pronunciation of the name is vouchsafed by the Turk, Chinese and Tibetan transcriptions,
though the first of these suggests a palatalized nsound, similar to gn in French 'agneau'. It
so happens that the final-nor -n became-yin the language of the Uighurs of Kocho. Most
probably the populations of West Turkistan became acquainted with this name through
the intermediary of Uighur and adopted its Uighur form. The fact that the Mongols of
the thirteenth century continued to use the original 'Kitan' form supports this explanation.
The exact significance of the epithet 'black' (kara) in tribal or personal names is not fully
established, but the more likely interpretation is that it is an honorific of sorts widely used
among Central Asian peoples in conjunction with other ethnonyms.
7
Bretschneider, 1911, pp. 214-15.

241
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

We have seen that the first Kitan attempt to establish contacts with the Islamic world
foundered on Ghaznavid arrogance. Now, almost exactly one century later, Ghaznavid
power was a shambles and the Kitan ruler a virtual fugitive. Yeh-lti Ta-shih was now faced
by the Turkic tribal confederation ruled by the Muslim Karakhanid dynasty. Since about
the middle of the eleventh century there were in fact two Karakhanid states (see above,
Chapter 6) and, through geographic necessity, the advancing Kitan first had to face the
eastern branch ruled since 1103 by Ahmad b. Hasan Arslan Khan. The two foci of this
state were the cities of Kashghar and Balasaghun, the latter located in the Chu valley and,
in subsequent centuries, completely destroyed.
It should be emphasized that the reconstruction of Kara Khitay history is, at best, tenta-
tive. The relevant Chinese and Muslim sources are often in contradiction and it is possible
that occasionally both are wrong. It would appear that, advancing in the Tarim basin, at
least one wing of Yeh-lti Ta-shih's army met with little luck and, probably in 1128 near
Kashghar, suffered a defeat at the hand of the Karakhanid ruler Ahmad b. Hasan (Arslan
Kara Kaghan). Ta-shih was more successful in another push westward, through Kyrgyz
territory. The character of this move, whether peaceful or warlike, cannot be determined; a
Khitay tribe still lives in southern Kyrgyzstan. Nor is it known with any degree of certainty
which of the two Kara Khitay wings - if indeed it is justified to speak of such a division
- built the city by the River Emil of which, in the mid-thirteenth century when the Persian
historian Juwayni wrote his work, 'some traces still remained' .8
Yeh-lti Ta-shih left his country and moved westward with the avowed intent of gathering
sufficient forces to re-establish Liao power over China. Perhaps, as time passed, he found
this aim unrealistic. The dispatch in 1134 of an expeditionary force which he did not join
himself was more an empty gesture than a real effort to achieve this aim. It could also
be that he thought that the creation of a new empire further west offered him a better
opportunity to leave his mark on history. Whatever the true motive, he made no serious
effort at a reconquest. Unmistakably, Yeh-lti Ta-shih's ambition was to carve himself an
empire in the west.
Internal dissent within the Eastern Karakhanid state facilitated Yeh-lti Ta-shih's search
for a territorial base. The Karakhanid ruler, Ibrahim II b. Ahmad, unable to cope with
his unruly Kangll and Kaduk subjects, sought and obtained Yeh-lti Ta-shih's help. This
assistance, if this is what it was, took the form of the occupation of the Karakhanid capital
Balasaghun in the Chu valley and the de facto elimination of Karakhanid rule in the region.
In Juwayni's words, 'the Gtir Khan proceeded to Balasaghun and ascended the throne that

8
JuwaynI, 1958, Vol. 1, p. 355.

242
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

had cost him nothing' .9 Yeh-lti Ta-shih had thus acquired a capital city; the marauding
refugee was now in possession of a fixed territorial base. Unfortunately, the date when this
happened cannot be established with any degree of certitude. Ibrahim II, while recogniz-
ing Kara Khitay suzerainty, was allowed to remain in place, occupying the more modest
position of Ilig-i Ttirkmen (Ilig-i Turk?), perhaps a concession to the predominantly Turkic
population of the Kara Khitay empire.
The neutralizing of the Eastern Karakhanids did not satisfy the ambitious Gtir Khan,
who now embarked on a western expansion of his domain. In 1137, near Khujand, he
defeated the Karakhanid Mahmud II, a nephew of Sanjar, the powerful Seljuq sultan.
Clearly, Yeh-lti Ta-shih was bent on further conquest. Taking advantage of a conflict between
Mahmud and his Kaduk subjects, in which the latter called on him for help, the Gtir Khan
inflicted a crushing defeat on the joint forces of Mahmud II and Sanjar at a battle fought
north of Samarkand in the Qatwan steppe in September 1141. News of this debacle reached
participants in the Second Crusade and, it is usually assumed, became the basis of the cre-
ation of the legend of Prester John, a fictional Christian king of the east bringing sorely
needed relief to the Crusaders battling with the Seljuqs. Despite his victory at Qatwan,
which brought the whole of Turkistan under Kara Khitay control, Yeh-ltih Ta-shih did
not assume direct governance; Karakhanid amirs, recognizing his suzerainty, continued to
rule. Effective power was in the hands of the Burhan family of Bukhara, the so-called sudur
('eminences'), Sunni religious leaders who dealt directly with the Gtir Khans.
Yeh-ltih Ta-shih died in 1143 and was succeeded by his widow, T'a-pu-yen (the empress
Kan-t'ien by her honorific title), who acted as regent for 8 years until Ta-shih's son Yeh-lti
1-lieh (1151-63) could ascend the throne. During his reign, Kara Khitay involvement in the
perennial conflicts in Transoxania continued. In 1158, answering a request for help from
the Khan of Samarkand, a Kara Khitay contingent of 10,000 men was sent to his rescue.
It was headed by the Ilig-i Ttirkmen, the deposed Karakhanid Ibrahim II. He showed no
more courage this time than when he had yielded to Yeh-lti Ta-shih; faced by the forces of
the Khwarazm Shah 11 Arslan, he sued for peace.
A census taken under 1-lieh counts 84,500 households with men aged 18 or older. This
figure probably does not include the population of the conquered territories. According
to Kitan traditional reckoning - which the Kara Khitay probably followed - each house-
hold had 2 adult, arms-bearing men. In the decisive battle fought on the Qatwan steppe,
the Kara Khitay forces numbered probably fewer than 20,000 men. Whatever the value of
these figures, it is certain that the Kara Khitay element constituted but a small fraction of
the population under its control. It could not influence in any significant way either the
9
Ibid.

243
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

economy or the cultural structure prevailing in the land. While ruling over China, the Kitan
(Liao) had operated within a fairly homogeneous Chinese population whose vast majority
did not object to the alien rule for the simple reason that it did not seriously interfere with
their way of life. The tasks faced by the Kara Khitay state were very different. Here too,
the Yeh-lti clan and its followers represented a minority within a much larger population
which, however, was far from homogeneous. Here, in Wittfogel's words, 'a limited imper-
ial domain [was] surrounded by a vast agglomeration of vassal peoples, sedentary as well
as nomadic' . 10 The tradition and the values of this Kara Khitay island constituted a con-
glomerate of Chinese and Kitan customs, the latter kept alive by the Yeh-lti clan, inheritors
of a centuries-old tradition and know-how of government.
Very little is known about the cultural fabric of the Kara Khitay. For lack of evidence,
even their language cannot be established with certainty. It seems unlikely that the use of
Mongol, if at all present, was widespread and possibly the language in Yeh-lti Ta-shih's
entourage was Chinese. The chancery used Chinese script, and Kara Khitay copper coins
were minted on a Chinese pattern and carried Chinese characters. What is known for cer-
tain is that the Kara Khitay were not Muslims and that, by replacing the Karakhanids in
Balasaghun, they superimposed an alien culture on the local population. The vast majority
of the Kara Khitay subjects were Muslims, speaking a Turkic or Iranian language.
One might argue that with the appearance of the Kara Khitay state, an alternative politi-
cal structure to the traditional Inner Asian Kaghanate was emerging, a process symbolized
by the claim to universality as expressed by the title of the ruler. It was to respond to the
needs of a state in which ethnic, religious and cultural diversity excluded the possibility of
a strong sense of solidarity. In this, as also in chronological order, the Kara Khitay were the
precursors of the Mongols. Both demanded little other than recognition of their rule and,
beyond this, the payment of taxes. By the side of the native ruler stood the vigilant Kara
Khitay official whose task it was to ensure regular payment. The political aims of the Kara
Khitay appear to have been modest, based, as they probably were, on a correct assessment
of their military strength. There is every indication to show that in organization and com-
bativeness the Kara Khitay armies were on a par with their Central Asian counterparts:
they won some battles, they lost others.
In vivid contrast to some of their Muslim vassals, the Kara Khitay rulers followed a pol-
icy of religious tolerance, a feature which was also to characterize the rule of the Mongols.
The Yeh-lti dynasty was probably Buddhist but did not try to impose this persuasion on its
subjects. Of particular interest is the florescence of the Nestorian Christian Church, whose
missionary activity continued unhampered by the Gtir Khans. In the twelfth century the

10
Wittfogel and Feng, 1949, p. 665.

244
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

city of Kashghar was a Nestorian metropolitan see and followers of this creed were still
encountered in the region by Marco Polo in the thirteenth century. Among the Christian
gravestones with Syriac inscriptions discovered in the Chu valley, the earliest were erected
in the Kara Khitay period.
At the death of 1-lieh, power was assumed by his younger sister P'u-su-wan, known
as the Empress Ch'eng-tien (1164-77). The Kara Khitay continued the tradition of strong
female participation in the running of state affairs mentioned above in connection with
T'ai-tsu's widow. Ch'eng-tien did not see her role as merely that of caretaker until such
time as 1-lieh's son could ascend the throne. Under her rule, hostilities against 11 Arslan
continued until his death in 1172 and she even intervened in the struggle for the throne
of his sons. Ch'eng-tien's demise was not caused by any political or military actions on
her part, but by her sentimental involvement with her husband's younger brother. She had
her husband murdered and, in turn, fell victim to the vengeance of her father-in-law. He
staged a coup in which both the empress and her lover perished. 1-lieh's second son, Yeh-lti
Chi-lu-ku (1179-1211), then became the ruler.
Under the new Gtir Khan, Kara Khitay involvement in Khwarazmian affairs continued
to follow the earlier pattern: interference in struggles for the succession and more or less
successful attempts at collecting taxes. There is no need to enumerate here the battles
fought with Kara Khitay participation on behalf of or against successive Central Asian
rulers. Let us just mention the Kara Khitay help given to the Khwarazm Shah cAla' al-Din
Muhammad in 1204 in his struggle against the Ghurid Mucizz al-Din. Since Yeh-lti Ta-
shih's arrival in Balasaghun, the Gtir Khans had been deeply influenced by and, in their
turn, had been able to influence events in Semirechye, Transoxania and even south of the
Oxus. But the winds of history were blowing from the east and were soon to disturb the
usual pattern of Kara Khitay politics.
Simultaneously with the events just described, far away in the foothills of the Altai, in a
decisive battle fought in 1204, Chinggis Khan (still called Temtijin) crushed the Mongol-
speaking but Christian Naiman. Their leader Tayang Khan fell in the battle, but his son
Ktichltig escaped. In 1208 Chinggis Khan, implacable, led another raid against the Naiman,
then living on the upper Irtysh. Once again, Ktichltig escaped the general massacre and
sought refuge with the Kara Khitay. He was well received, married the Gtir Khan's daughter
and abandoned Christianity in favour of Buddhism.
However hurried Ktichltig's flight might have been, he was accompanied by a substan-
tial number of Naiman whose arrival upset the delicate internal balance prevailing within
the core of the Kara Khitay empire, and thus posed a serious threat to its very existence.
Historians, relying mainly on Juwayni' s description of the events, usually depict a scenario

245
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

in which the Kara Khitay fell victim to a joint, concerted attack led from the east by the
Naiman under Ktichltig, and from the west by c Ala' al-Din Muhammad. This was not the
case. Perhaps the best summary of the events is given by Li Chih-chang in his narrative of
the Taoist sage Ch'ang-chun's journey to the west, where he was to join Chinggis Khan:
When the Naiman tribe was defeated, they took refuge with the Ta-shih [i.e. the Kara Khitay];
but having recuperated their strength they presently seized the land that had sheltered them,
while the western part of the empire was shorn off and allotted to the Khwarazm Shah. 11

Indeed, Ktichltig's grab for power is more like a classical palace coup led by a vigorous
son-in-law against his aging father-in-law than a military attack from the outside. Armed
clashes between their respective followers did occur, but when in 1211 Ktichltig emerged
the victor in the conflict, he did not assume the title Gtir Khan, and Yeh-lti Chi-lu-ku,
who remained the nominal ruler, was allowed to die a peaceful death in 1213, two years
after Ktichltig's revolt. For the Franciscan John of Plano Carpini, who travelled through
the region in 1246 and was generally well informed, the Naiman and the Kara Khitay were
allies who had been defeated by the Mongols.
It could be that it was the threat of a new conflict with the Mongols that prompted
Ktichltig to strengthen his position within the Kara Khitay polity by eliminating the Gtir
Khan. One of Chinggis Khan's leading generals, Khubilay Noyan (not to be confused with
the Great Khan Khubilay), who operated in the Semirechye region around 1210, must
certainly have been aware of Ktichltig's ascent to power and it can be taken for granted
that he warned Chinggis, unless - a distinct possibility - Khubilay had already been sent to
the region to report on Ktichltig's actions. A revival of Naiman strength was not something
Chinggis Khan would tolerate and steps were taken to eliminate Ktichltig, who had escaped
him twice. This time the restless Naiman was to meet his fate at the hand of the Mongols.
Ktichltig was unable to hold together the Kara Khitay inheritance. Internally, his anti-
Muslim religious policy alienated his Muslim subjects; in his external affairs he found
himself in a desperate situation, trapped as he was between the Khwarazm Shah and the
Mongols. The Uighur ruler Barchuk, whose predecessors had been so friendly to Yeh-lti
Ta-shih, exasperated by Ktichltig and sensing the approaching Mongol tide, changed his
allegiance, killed the Kara Khitay governor in 1209 and submitted to Chinggis. A Mongol
army under the command of Jebe, another of Chinggis Khan's chief lieutenants, advanced
on Kashghar, where the population, incensed by Ktichltig's religious persecution, rose in
revolt. To meet the challenge, Ktichltig went to Kashghar but found that the forces against
him were too strong to be mastered. To save his life, Ktichltig fled for a third time, but on
this occasion his luck deserted him: captured in Badakhshan, he was decapitated.
11
Waley, 1931 , pp. 88-9.

246
ISBN 978-92-3-103467-1 Contents The rise of the Kara Khitay

Ktichltig was neither a great statesman nor a great warrior. Moreover, the legitimacy
of his rule was doubtful and he certainly lacked the charisma so characteristic of the Yeh-
lti clan. One could speculate as to what would have happened had the Mongol advance
reached the Kara Khitay state still ruled by a Yeh-lti. The advancing Mongol armies con-
tained important Kitan contingents. Let us also recall that Yeh-lti Ch'u-tsai (1189-1243),
trusted adviser of Chinggis Khan and Ogedey, belonged to the same clan. He accompa-
nied the conqueror on the campaign to the west that would destroy Khwarazm. Ch'u-tsai's
record of his journey, though far too short, informs us that in November 1221 he was in
Balasaghun, of which he briefly notes that it used to be the capital of the Hsi Liao, i.e. the
Kara Khitay. One wonders what his thoughts might have been there, just ten years after his
relative, the last Gtir Khan, had been deposed. Ch'u-tsai had a keen interest in and feel-
ing for the various cultures - Kitan, Mongol, Chinese - all of which he served with great
distinction. On his above-mentioned journey, the Taoist sage Ch' ang-chun met another
member of the Yeh-lti clan, namely A-hai, the civil governor of Samarkand installed by the
Mongols. His son and successor Yeh-lti Mien-ssu-ko was governor of Bukhara.
Clearly, the Yeh-lti were great survivors, a quality that was apparently characteristic
of the Kitan people whom they had served with such distinction. Following the collapse
of their second great state, that of the Kara Khitay, for centuries to come the Kitan were
able to preserve a measure of their national identity in a diaspora that extended well into
eastern Europe. References to the Kitan can be found in the heroic poetry of such north-
west Siberian peoples as the Voguls and the Ostiaks, and in hydronyms of the same region.
As a tribal name, Kitan, Katay, Kitay and their variants can be found among the Kalmucks
of the seventeenth century, located to the west of the Ural river; among the Bashkirs of the
Volga region; and even among the Tatars of the Crimea. Toponyms found as far west as
Moldavia testify to the former presence of Kitan groups and Hungarian chronicles of the
thirteenth and fourteenth centuries locate the Kitan on the banks of the Don. 12
The most enduring trace of Kitan power, however, is the very name Cathay, China's
medieval Latin appellation, still current in many modern usages and in the Russian name
of that country. The search for fabulous Cathay was a principal incentive for the great
discoveries of the fifteenth and sixteenth centuries. The history of the Kitan constitutes a
truly extraordinary chapter in the history of the world.

12
Sinor, 1995, pp. 264-7.

247
ISBN 978-92-3-103467-1 Contents THE MONGOLS AND THEIR STATE ...

12

THE MONGOLS AND THEIR STATE IN

THE TWELFTH TO THE THIRTEENTH
CENTURY*
Sh. Bira

Contents

The socio-economic and political situation . . . . . . 249

Chinggis Khan and the founding of the Mongol state 251
Chinggis Khan's campaigns of conquest: The foundation of the Mongol empire 257
The Mongol empire during the reign of Chinggis Khanr's successors 259
The Yuan empire of the Mongols and its fall . . . . . . . . . . . . . 264

The geographic setting of Mongolia was of vital importance in shaping its unique history
and civilization. The vast mountainous-steppe zone of Mongolia forms part of two impor-
tant regions of world civilization, the oases of Central Asia and the so-called Eurasian
steppe belt stretching from the Danube to the Great Wall of China. From early times, Mon-
golia was at the crossroads of world communications. Two great highways, the Great Silk
Route and the Eurasian steppe corridor, also known as the Silk Route of the Steppes, linked
Mongolia with the centres of civilization of East and West. The birth of nomadic civiliza-
tion was an appropriate response to the physical challenges in that specific part of Central
Asia.
It was in the late twelfth and early thirteenth centuries that the Mongols proper appeared
on the stage of history. Occupying the main regions of Mongolia, they represented the great
majority of its inhabitants. The problem of the genesis of the Mongols is not yet settled, but
what is most probable is that the Mongol ethnic group of people had been formed as a result

* See Map 6, pp. 434-5.

248
ISBN 978-92-3-103467-1 Contents The socio-economic and political situation

of the long historical and ethno-cultural processes in which various peoples who inhabited
the Mongolian steppes took part from very early on, and which was likely to have been of
Altaic provenance (Proto-Mongol, Proto-Turkic and Proto-Tungus). The name 'Mongol' is
apparently mentioned in Chinese sources from the fourth century A.D. as Shi Wei Meng-gu
or Shi-Wei Mongol. In the period of the rise of the Mongols, the most important fact was
that the greater part of the tribes living in Mongolia were sufficiently alike, both ethnically
and in their way of life, for them to be moulded together into a highly organized nomadic
society.

The socio-economic and political situation

By the thirteenth century, the traditional social system of the nomads in Mongolia had
undergone considerable change, with a decay of the primitive clan system held together
by blood relationships; consciousness of membership and a sense of solidarity with the
nomadic society now tended to be determined by shared socio-economic interests. 1 The
clan lineage or the common descent group, usually referred to as obuq in Mongolian,
had already fallen into decay as a nucleus of social structure in Mongolia. The more the
clans propagated themselves, the more they developed into other groups and subdivisions
(yasuns). The Secret History of the Mongols and Rashid al-Din's history give many con-
crete examples of the splitting of clans into subgroups and their reassembling into larger
units and tribes (aymaks). On the other hand, the great changes in the traditional nomadic
system could not but cause the centripetal tendencies that encouraged the development of
another process, i.e the unification of all separate social groups or divisions into larger units
or tribal confederations, or even state formations of varying size, under the supremacy of a
leading clan lineage, such as that of the Borjigids. With the development of Mongol soci-
ety, the coalescence of clans and their subdivisions into tribes and tribal confederations had
become ever more vital.
The decay of the kinship structure in Mongolia went hand in hand with a process of
social stratification. Private ownership of the country's main wealth - cattle - became the
basic criterion for social position within Mongol society; this was concentrated in the hands
of a small group of people, while the greater part of the population remained as depen-
dants. Hence there developed two main classes: the nobility (noyad) and the commoners
(karachu). In its initial stage, the nobility might have largely consisted of those who held
various traditional titles - Khan (chief, king), mergen (meaning 'an excellent archer', later
'a wise man'), baghatur (brave, hero), sechen (wise), ejen (lord), beki (honoured shaman),

1
Vladimirtsov, 1938; Jagchid and Hyer, 1979, Ch. 6.

249
ISBN 978-92-3-103467-1 Contents The socio-economic and political situation

singgum, and so on. The authority of the Khan, the highest rank in the hierarchy, grew with
the increase in the common interests of the nobility. Whenever a confederation of tribes
emerged, a supreme leader with the title of Khan was declared.
The majority of the population comprised the commoners, karachu or irgen and haran
(aran); the last two terms are mentioned in The Secret History and other sources. They
included different categories of people: boghol (so-called slaves), members of defeated
aymaks (jadaran) and poor cattle-breeders. The Mongol term boghol included various dif-
ferent categories. Corresponding to the real meaning of slaves were the so-called Otog
boghol (prisoners of war), who were used as servants in the households of the privileged.
Another category, the unaghan boghol, may be interpreted as 'slaves by origin'. This cate-
gory of boghol had almost nothing to do with real slaves. In nomadic society, the practice
existed of families or individuals presenting young men to a prominent leader, usually a
qan, as a token of friendship or submission. Here, the institution of boghol reflected a
vassalage-type relationship rather than the subordination of slavery. The so-called omchi
boghol (personal slaves) in effect represented the followers or vassals of their lord, and
were different from the institution of nuker. Nuker (pl. nukud) means 'friend', and the term
referred to a member of a group of warriors who freely declared themselves to be the
'men' of a chosen leader, irrespective of their origin or tribal affiliation. In the beginning,
a nuker was a loyal companion in battle, but later he assumed the special character of vas-
sal; the institution played an essential role in the formation of vassalage-type relations in
Mongolia.
The transformation of the clan system in Mongolia caused great changes in the country's
economic structure. Under the conditions of a primitive kinship society, in which cattle, its
main wealth, were the common ownership of clans, the nomads moved collectively or
by kuriyen, i.e. the community of clans tied together by blood. The kuriyen (or 'circle')
was the traditional form of economic organization of the nomads, reflecting the nomadic
custom of erecting tents in the form of a circle, with the tent of the clan leader in the
centre. The kuriyen was a specific collective institution for the joint ownership of cattle
and pasturelands, but when private cattle ownership developed and the clan system fell into
decay, this form of economic structure did not meet the requirements of daily life; private
cattle-owners preferred to nomadize by smaller groups, i.e. by ayil. Hence the kuriyen form
of management of the cattle-breeding economy changed into the ayil form, suitable for the
private ownership of cattle and grazing land. Thus by the advent of Chinggis Khan in the
latter half of the twelfth century, most of the Mongols had adopted the aymak tribal system
of social order. Numerous nomadic tribes are known to have inhabited the main parts of

250
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

Mongolia. Some of them formed separate and often rival tribal confederations, for example
the Kerait, Naiman, Merkit and Tatar, each with its own Khan. 2
Special mention should be made of the Mongols whose pastures were in the southern
valleys of the mountain of Burqan Qaldun in the Khentei mountains, the headwaters of
three rivers, the Onon, Kerulen and Toul, a region considered to be the cradle of the Mon-
gols. The Daychud and Jalayr, who may be called the 'Three River Mongols', constituted
the virtual nucleus of the Mongol people; they had started to play a prominent role in
the country's history well before the time of Chinggis Khan. Tribal tradition attributes the
major events pertaining to the initial period of Mongolian history solely to the Mongols
of these regions. According to The Secret History, from the very beginning the ancestors
of the Mongol Khans had chosen the lands of Burqan Qaldun for their principal camping
ground. The legendary forefather of Chinggis Khan's clan, Btirte-chinua, who is said to
have been born 'having his destiny from Heaven above', together with his spouseGo'a-
maral, came from over the sea and encamped at the head of the Onon river, at Burqan
Qaldun. 3
The centre of gravity of Mongol history in the twelfth and thirteenth centuries had
already shifted to the eastern regions of Mongolia. It was during the reign of Qabul Khan,
who probably lived in the first half of the twelfth century and was the great-grandfather of
Chinggis Khan, that the Mongols of the Three River regions became powerful enough to
be united into a state-like confederation (ulus). It is not certain when Qabul Khan died, but
there is no doubt that this ulus continued to exist after his death. He is known to have been
succeeded by at least two Khans, Ambaqay and Qutula. But soon after the death of the latter
in the late twelfth century, the confederation of the Three River regions disintegrated due to
the constant feuds between the Mongols and the Tatar. The Chin dynasty of China's policy
of playing off one group of the 'barbarians' against another contributed to this decline.

Chinggis Khan and the founding of the Mongol state

Qutula was probably the last Khan of the Mongols of the Three River regions. In any event,
the historical records do not mention any other Khan up to the rise of Chinggis. Yestigey,
Chinggis Khan's father, although he was the grandson of Qabul Khan and one of the most
influential noblemen, was not of sufficient stature to be called Khan. He served Qutula
Khan loyally and displayed courage in numerous battles, and held the title of Baghatur,
which he must have inherited from his father, Bartan Baghatur.

2
Shirendev et al., 1966.
3
Cleaves, 1982, p. 11.

251
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

It is not certain when the future Chinggis Khan was born. According to most Chi-
nese and Mongolian records, he was born in 1162 (a Year of the Horse) on the Deltigtin-
Boldog watershed on the upper reaches of the River Onon. 4 Rashid al-Din and other
Persian sources place his birth in 1155 or 1167 (a Year of the Pig). The date of his
death in 1227 is, however, certain. He is said to have been born at the moment when his
father returned from a successful campaign against the Tatar. Following an ancient custom,
Yestigey gave his son the name of the captured chief of the Tatar, i.e. Temtijin, which may
have meant 'blacksmith'. Although the extensive information on the early life of Ching-
gis Khan in The Secret History is almost certainly romanticized, it is likely that he had to
overcome numerous difficulties and hardships in the steppe, left to the mercy of fate after
losing his father in early childhood. As Lattimore has pointed out:

Chinggis Khan was a genius but not a savage, illiterate but not ignorant. He was born into a
tradition that embraced war as a profession and also included a sophisticated knowledge of
the political and economic uses of power. All his natural talent would not have got him very
far, however, if he had not been born into this tradition at a propitious moment and in just the
right geographical region. 5

Indeed, Chinggis Khan was a real son of his time, for his advent coincided with crucial
changes in the nomadic society of the Mongols, as mentioned above, and the appearance
of such an able figure possibly accelerated the development of this historical process. The
secret of his unprecedented career was perhaps that he could make the best use of the situa-
tion that prevailed in the nomadic society of his time. This last engendered both centrifugal
and centripetal tendencies. On the one hand, different tribes tried to be separate and to pro-
vide for their independence existence, which led to mutual feuding. On the other hand, the
more powerful a tribe became, the more it endeavoured to incorporate others, resulting in
fewer but larger tribal units.
To achieve his purpose, Chinggis Khan resorted to a great variety of means and tactics.
First, he was clever in manipulating traditional tribal politics, deriving great benefit from
his skilful use of some old tribal institutions, such as those of the nuker, the anda and oth-
ers. Thanks to such tested devices, the young Temtijin was able to save his family from the
humiliating position in which it found itself after it was deserted by its tribesmen follow-
ing the death of Yestigey. He was further able to restore the leading role of the Borjigids.
The anda was the oldest form of alliance in the kinship society, by which warriors were
sworn to blood loyalty with one another. In Chinggis Khan's time, the oath of anda lost its
primary feature and could be found among warriors belonging to different tribes or clans.

4
At present, this is known as Gurban-nuur in Dadal-sum territory, Khentei-aymak, in eastern Mongolia.
5
Lattimore, 1963a, p. 57.

252
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

What was needed was that those who wanted to become andas shared common interests;
they had to confirm their oath by tasting each other's blood in order to symbolize their
close brotherhood. At the beginning of his career, Temtijin was able to win over to his side
many talented and loyal nukers and andas.
The fact that the young Temtijin enjoyed the patronage of the mighty Toghril Khan
greatly enhanced his position. The Kerait Khan placed a considerable force at his disposal;
and several kinsmen of Temtijin and some tribal chieftains, like Jamuka of the Jajirad,
joined him and were ready to help. In 1185 Temtijin, with his united forces, easily defeated
the Merkit and secured abundant loot, freeing his wife Barte (who had previously been
abducted by them). This was the first serious victory which Temtijin achieved over his
enemy with the aid of his andas, Toghril Khan and Jamuka. After his victory over the
Merkit, Temtijin became a notable figure among rival chieftains in the steppe. He renewed
his anda brotherhood with Jamuka, who was at this time stronger than Temtijin and no
less ambitious and energetic: in the future, these two young men were to become the main
rivals for power.
The anda brotherhood between Temtijin and Jamuka did not last long. According to
The Secret History, after the victory over the Merkit they lived together for one and a
half years on very friendly terms. But it seems more probable that during this period they
kept an eye on each other and did their best to increase their forces by attracting adherents.
When the anda alliance between them broke up, Temtijin found himself in a much stronger
position. Altogether, twenty-three groups of kinsmen and noblemen from some twenty-
three clans and tribes had come over to Temtijin when he encamped at Ayil Qaraghana on
the Kimuragha stream. Among those who rallied to his rising standard were such hereditary
representatives of the Mongol nobility as Da' aritay, the grandson of Qabul Khan, the fourth
son of Bartan Baghatur, his uncle Altan Odchigin, the third son of Qutala Khan, Quchar,
Yestigey's nephew, the son of Neken-taysi, Sacha-beki, Qabul Khan's grandson and others.
Temtijin had become virtual overlord of the Borjigids, the noblest clan, from which
the Mongol royalty derived its origin. He was again in possession of the original home of
his tribe. It was the Mongol nobility which provided both moral and physical support for
Temtijin to become Khan. Thus it says in The Secret History, 'When Altan, Quchar, and
Sacha Beki took counsel with one another together and spoke unto Temtijin, saying "We
shall make thee to become khan".' 6 Temtijin's accession to the throne took place in 1189
at Lake Koko of Qara Jtirtigen, on the Senggtir stream, within Mount Gtirelgti in the upper
reaches of the rivers Onon and Kerulen, near Burkhan Qaldun. The official title 'Chinggis'

6
Cleaves, 1982, 123.

253
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

was conferred on Temtijin by the shaman Kokochti Teb-Tenggeri. 7 The etymology of his
name has been explained differently by scholars, but the most convincing interpretation
is that 'Chinggis' means 'fierce, hard, tough'; thus ' Chinggis Khan' means 'the Fierce
Ruler', not 'the Universal Ruler' .8 One has also to assume that the Mongol shaman called
his militant leader by the name Saqiyusun (Defender).
According to The Secret History, Chinggis Khan then instituted ten court offices, but,
with the exception of that of Cherbi, the names of the officials are not given and only their
duties are indicated, even though without precision. As the chief advisers of the Khan, the
duty of maintaining order in the meetings very probably devolved upon them. The two
offices of 'guardians of the assembly' were occupied by Bo'orchu and Jelme, to whom
Chinggis reputedly said, 'Being my shadow, this settles my mind, so let this be in my
thoughts.' 9 Chinggis's guard was organized more formally in 1203, after the victory over
the Kerait, when Chinggis became the chief personage in eastern Mongolia. Seventy men
were selected for the day guard and eighty for the night guard; altogether, they constituted
the protective guards or kesigten (sing. kesik, meaning 'turn', 'relief').
The enthronement of Chinggis meant the restoration of the state confederation of the
Mongols of the Three River regions which had fallen into decay after Qutula Khan. From
now on, Chinggis Khan could act as the lawful ruler of all the Mongols. But to become
the genuine lord of the nation in Mongolia, he had to achieve the real unification of all
the people on a country-wide scale, and this took him from 1189 to 1206. The majority of
the tribes continued to be separate and were not initially disposed to recognize Chinggis
Khan's rule over the whole country. Among them were such powerful tribes as the Tatar,
the Kerait, the Naiman and others, but by manoeuvring between the conventions and oppo-
sitions of the tribal system, Chinggis was finally able to unite all the peoples living under
felt tents.
The second stage in Chinggis Khan's rise to power began with his victory over the
Tatar as a result of the successful campaign that he had undertaken in 1196 in alliance with
Toghril Khan of the Kerait and the Chin dynasty of China. By subjugating the Tatar, Ching-
gis not only took vengeance on his family's enemy but eliminated the threat to the south-
ern side of his domain. However, these successes alarmed his rivals, first of all Jamuka. In
1201 Jamuka was proclaimed Gtir Khan at the kurultay (general council) of the leaders of
the Daychud, Qonqirad, Ikira, Qorlo, Qatagin, Oyirad and Naiman, against the Wang Khan
(Toghril) of the Kerait and his son, the allies of Jamuka. Chinggis Khan was victorious over

7
JuwaynI, 1958, Vol. 1, p. 39; Rashid al-Din, 1952, p. 167.
8
Rachewiltz, 1989, p. 288.
9
Cleaves, 1982, 123-4; Barthold, 1968, pp. 38-23.

254
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

the Wang Khan and the latter was killed, but Chinggis still had to overcome an alliance of
opponents led by Jamuka until the latter was also killed. Thus Chinggis finally triumphed
in the tribal wars which had continued for more than twenty years.
Chinggis Khan now held all Mongolia, having subjugated all the tribes of the Mongolian
steppe. To guarantee his right to rule over the entire country, in the Year of the Tiger (1206)
he called a kurultay at the head of the Onon river, where he 'set up a white standard with
nine tails' and was once again proclaimed Chinggis Khan. 10 This event signified the birth
of a Mongol power that stretched some 1,600 km from east to west, from the Khingan
mountains to the Altai range, and more than 960 km from north to south, from Lake Baikal
to the southern margins of the Gobi desert along the Great Wall of China.
Between 1206 and 1211 Chinggis Khan was engaged in establishing and reorganizing
the civil and military administration. The head of state was the Khan, who was declared
to have a mandate from Mongke Tengri (Everlasting Heaven). The second most important
office was that of Guiong (Chinese Kuo-Wang), or 'Prince of the Realm'. This title was
conferred on Muqali, the most devoted companion of Chinggis. The office of Supreme Jus-
tice or jarghuchi was introduced: it was given to Sigi Qutuqu, a Tatar by extraction, who
had been adopted as a boy by Chinggis Khan's mother and was one of the best-educated
men of his time. The office of beki, which designated the chief shaman, the highest reli-
gious authority, was also set up; it was occupied by Ustin-ebtigen. He was instructed 'to
ride on a white horse, wear white raiment' and 'choose a good year and moon' .11 The
Uighur Tatatungha, the keeper of the seal of the Naiman Khan, held the same office at the
court of Chinggis Khan and was also commissioned to teach the Khan's sons reading and
writing.
The Khan's guard, or kesigten, was also reorganized. It consisted of kebte 'ul (night
guards), qorchin (day guards) and turgha'ud (bodyguards). The number of each of these
corps of guards was raised to 1,000 men, making a total of up to 10,000 men. It was
decreed that each son of a 'leader of a 1,000' had to bring with him 1 kinsman and 10
companions; the son of a 'leader of 10' and freemen in general had to bring 1 kinsman and
3 companions. The guard was subject to severe discipline, but its members enjoyed great
privileges: a combatant private in the guard stood higher in rank than a 'chief of 1,000' in
the army, non-combatants in the guard stood higher than a 'chief of 100'. Members of the
guard who committed a crime could not be punished by anyone except Chinggis himself.
The bodyguard was not only the personal guard of the Khan and the core of the army; it was
also a sort of military school which allowed the Khan personally to test the future leaders of

1
° Cleaves, 1982, 202; Ishjamts, 1974, pp. 48-61.
11
Cleaves, 1982, 216

255
ISBN 978-92-3-103467-1 Contents Chinggis Khan and the Mongol state

his military forces. The army was reorganized according to the traditional decimal system.
Chinggis Khan appointed 95 'noyans (chiefs) of 1,000'; the names of these noyans are
listed in The Secret History. 12
Administratively, Mongolia was divided into three large tumens ('myriads'): the Left,
the Right and the Centre, each of them in turn consisting of tens, hundreds and thousands.
The main tumen was the Left tumen commanded by Gui Ong Muqali. The other two tumens
were headed by Bo' orchu and N aya. The Central tumens occupied the main area of Mon-
golia; the Right tumens, the lands near the Altai mountains; and the Left tumens, the lands
up to the Khingan mountains. 13 Chinggis appointed leaders for all the tribes and clans from
among his personal followers and his family, thus laying the framework of the new Mongol
empire and destroying the old tribal system; Chinggis Khan's own clan, the Borjigids, with
its vassals and followers, now became the supreme clan of the Mongols.
One of the most important measures undertaken by Chinggis Khan in the field of the
civil administration was the codification of laws under the title of Yeke Jasa (The Great
Law). Although this work has not yet come to light, data from various sources prove
beyond doubt the existence of a written version. According to an authoritative source, the
history of c Ata' Malik Juwayni:

In accordance and agreement with his [Chinggis's] own mind, he established a rule for every
occasion and a regulation for every circumstance, while for every crime he fixed a penalty.
And since the Tatar peoples had no script of their own, he gave orders that Mongol children
should learn writing from the Uighurs, and that these Yasas and ordinances should be written
down on rolls. These rolls are called the Great Book of Yasas and are kept in the treasury of the
chief princes. Whenever a khan ascends the throne, or a great army is mobilized, or the princes
assemble and begin to consult together concerning affairs of state and administration thereof,
they produce these rolls and model their actions thereon, and proceed with the disposition of
armies or the destruction of provinces and cities in the manner therein prescribed. 14

In general, the Great Jasa represented a code of laws which is said to have been prescribed
by Chinggis Khan for the various spheres of social life and in military, organizational and
administrative affairs. It also dealt with religious beliefs, court ceremonial, civil rules, gen-
eral conduct and justice, and so on. Thus it laid down the juridical basis for the newly born
Mongol state. Moreover, with the creation of the Mongol empire, it eventually became the
most authoritative handbook of Mongol jurisprudence, to be strictly followed throughout
the expanse of the empire. Its authority was so great that, even after the fall of the empire,
it had some appeal for statesmen in the countries of Central Asia (as for Timur, known

12
Ibid., 202.
13
Ishjamts, 1974, pp. 59-61.
14
JuwaynI, 1958, Vol. 1, p. 25.

256
ISBN 978-92-3-103467-1 Contents Chinggis Khan's campaigns of conquest

to the West as Tamerlane, and others), serving as a political and moral instrument for the
justification of their expansionist ambitions.
According to the Sino-Mongolian inscription of 1346, in the fifteenth year of Chinggis
Khan's reign, i.e. in 1220, the capital city of Mongolia, Karakorum, was founded in the
valley of the Orkhon river, 15 demonstrating that Chinggis wished to rule his empire from
Mongolia.

Chinggis Khan's campaigns of conquest: The

foundation of the Mongol empire
Shortly after the creation of the political and military machinery in his own country, Ching-
gis Khan embarked on an expansion of his power. It is difficult to say whether from the
outset he had any serious intentions of conquering the great settled civilizations outside
Mongolia. The order of his conquests shows that, having settled internal affairs, he first
incorporated within his state all other nomadic peoples who lived outside Mongolia and
whose way of life was similar to that of the Mongols, rather than any of the settled peo-
ples. In 1207 the nomadic tribes in the valleys of the Selenga and the Yenisei were added
to his dominions. In 1209 the Uighur Turks, who had some four centuries earlier created
their own empire in Mongolia, with a capital in the valley of the Orkhon river, and who
had migrated south-westwards to the oases of the Tarim basin after the fall of their empire,
were peacefully incorporated. It is true that Chinggis Khan twice (in 1210 and in 1214-15)
campaigned against the Chin empire of China. But by doing so, he probably wished to
demonstrate his might and fame rather than to subjugate China, for he soon desisted; but
he continued to bring under his control the nomadic tribes of the Eurasian steppes. He con-
quered practically all the nomadic peoples of Turkic origin up to the north-eastern fringes
of Persia. In 1218 the Kara Khitay of Transoxania submitted to his power almost without
opposition. Thus the lands of East Turkistan and some other areas of Central Asia came
under the rule of the Mongols.
Chinggis Khan's domains now bordered directly on the great empire of the Khwarazm
Shah c Ala' al-Din Muhammad, who at this time controlled most of northern Persia and
Transoxania (the latter having been largely taken over from the Kara Khitay in 1210).
Central Asia in this period was divided into an eastern part ruled by the Mongol Khan,
while its western part was under the power of the Turkish Khwarazm Shahs. It was perhaps
natural, though not inevitable, that these two political forces should be rivals. Chinggis
Khan is said to have recognized the Khwarazm Shah as ruler of the west, as he himself

15
Cleaves, 1952.

257
ISBN 978-92-3-103467-1 Contents Chinggis Khan's campaigns of conquest

was the ruler of the east, and to have expressed the hope that peace would be maintained
and trade promoted between the two empires. But the Mongol Khan pointedly addressed
the Khwarazm Shah as his 'son', hardly the treatment of an equal by an equal. There is no
material available to ascertain the real intentions of Chinggis Khan towards his neighbour.
Whatever his true motives, between 1219 and 1224 Chinggis Khan embarked on his
campaigns of conquest against the Khwarazmian empire, which was at the time affected
by internal discord and feuds. As a result, such Transoxanian and Khurasanian cities as
Samarkand, Bukhara, Urgench, Utrar, Nishapur, Balkh and Merv were devastated by the
armies of Chinggis Khan.
Thus the people of the north-eastern Islamic world, mostly of Turkic stock, were brought
under the rule of the Mongol Khan. In their pursuit of the defeated Khwarazm Shah, two
Mongol generals Jebe and Stibetey reached the Caucasus in 1221, defeated the Georgian
king George IV Lashen and emerged on to the southern Russian steppe. In the spring of
1223, at the battle of the River Kalka, the Mongols crushed the combined Russian and
Kipchak forces, but did not really exploit their success. At the end of that year, the armies
of Jebe and Stibetey rejoined the forces of Chinggis Khan. The Khan himself returned to
Mongolia in 1225. Chinggis Khan's last campaign ended with the subjection of the Hsi-
Hsia (Tangut) in 1227. Soon afterwards, he fell ill and died in the same year. His body was
taken home to be buried in the Khentei mountains.
It is likely that at the outset Chinggis Khan did not have a clearly formulated policy
of conquest; but if one judges his wide-ranging conquests by their practical outcome, it is
possible that what we might now call a grand strategy lay behind his military actions. In
Lattimore's view, Chinggis was anxious to avoid the classic mistake of previous barbarian
rulers of the steppe who, as soon as they had formed an effective nomadic confederation,
succumbed to the temptation to invade northern China and to establish themselves there.
This sequence of events generally created, Lattimore suggests, a power vacuum within the
steppe. This was duly filled by the next nomadic general to form a confederation, and he in
his turn would then invade China and expel his predecessor. By contrast, Chinggis Khan's
strategy was first to form his confederation, secondly to neutralize temporarily the danger
from China, and then to return to the steppe to mop up and incorporate all the remaining
Turko-Mongol peoples, thus ensuring that no power vacuum would be created and that,
when China was conquered, the steppe would be retained as well. 16
Indeed, Chinggis, a true son of the nomadic culture, remained loyal to the ideals of his
background until the very end of his life and resisted the allurements of settled civilization.
He bequeathed to his successors a great empire consisting mostly of the nomadic peoples

16
Lattimore, 1963b, pp. 6-7; Morgan, 1986, p. 73.

258
ISBN 978-92-3-103467-1 Contents The Mongol empire ...

who inhabited the vast area extending throughout Central Asia with its centre in Mongolia.
He knew how to take advantage of those achievements of the sedentary civilizations which
could be beneficial to his empire; thus the Sogdian-Uighur script was adopted for the writ-
ing of Mongolian and use made of it in his chancery. With the conquest of the Uighurs and
the Kara Khitay, Chinggis could use the administrative skills of these peoples, who had
much in common with the Mongols as regards their nomadic mode of life - and these were
the intermediaries who transmitted the acquisitions of Islamic and Chinese civilization to
the Mongols. Muslims played a particularly prominent part in the service of the Mongol
Khan; their activities were greatly stimulated by the atmosphere of religious toleration and
the policy of unhampered trade and communications.
There were, of course, many factors behind the successes of Chinggis's conquests, not
least the role of the Khan himself, a shrewd politician and a military genius. Light cavalry,
comprising tough, swift-footed Mongol horses and archers, was always the main force,
but Chinggis also took over military techniques and improved them with the help of Chi-
nese and Muslim experts. He even used gunpowder in siege warfare, sapping and mining
operations, during his western campaigns.

The Mongol empire during the reign of Chinggis

Khan's successors
Contrary to Alexander the Great, whose Graeco-Asian empire did not even survive his
death, Chinggis Khan left a great empire which was capable of continuing to function in
both time and space, and had many successors from his own family to continue his imperial
policy. During their reign, the Mongol empire became the largest continuous land empire
that had so far existed in history. At its greatest extent, it stretched from the Far East to
eastern Europe. Military expeditions were mounted into mainland South-East Asia, and
even against Java and Japan, but without success. As a whole, the empire lasted for well
over a century, and some parts of it survived for much longer.
The successors of Chinggis Khan, although they proclaimed on every occasion their
adherence to the commandments of their great predecessor, in fact departed from his fun-
damental principle of staying outside 'civilization' and not sacrificing the ideals of the
nomads for their sake. Ogedey (1229-41), Gtiytik (1226-48) and Mongke (1251-9) went
on expanding their empire into the sedentary lands (for further details, see Chapter 13
below).
The unprecedented territorial expansionism of the Mongol nomads undoubtedly caused
great upheavals and distress for large numbers of people, although the bloodshed and

259
ISBN 978-92-3-103467-1 Contents The Mongol empire ...

destruction caused to the settled civilizations may not have been so widespread as some ter-
rified contemporaries depicted it. Like all conquerors, Chinggis Khan could calmly exter-
minate people by the thousands if he considered it necessary for the consolidation of his
rule, but none of his actions shows any sign of useless or stupid cruelty, 17 and he was far
from being a savage warrior, blindly ferocious and conquering for the sake of plunder. 18
At the same time, we must understand the feelings of hatred and horror that the conquered
peoples naturally had - and still have today - towards their enemy.
The problem of consolidating and administering such a great empire was the most diffi-
cult task the Mongols had ever faced. Nevertheless, judging by their actions, it is clear that
they did their best to secure their rule for as long as possible. Characteristically, the first
successors of Chinggis Khan tried to keep the centre of their empire in Mongolia itself. But
to do this proved much more difficult for them than for Chinggis Khan. The conquest of
Persia and China involved them in the governance of two great sedentary societies, and it
was then that the Mongol Khans encountered the problem of reconciling two incompatible
ways of life - a nomadic existence and a sedentary civilization. This was a problem that
had never previously been solved and it proved to be a major cause of the decline of the
Mongol empire.
Nevertheless, Chinggis Khan's successors managed to set up an imperial organization
whose unity endured for forty years after the death of the founder, with the supremacy of
members of Chinggis Khan's family extending over several generations in the successor
states. How was this achieved? Yeh-lti ch'u-ts' ai, the great Kitan adviser of the Mongol
Khans, is said to have repeated to Ogedey Khan the old Chinese admonition: although the
empire had been conquered on horseback, it could not be ruled from horseback. No doubt
the Mongol Khans realized this when faced with the problem of how to maintain their rule
over the conquered lands. First of all, they depended on what had already been achieved
by their great predecessor in the field of empire-building, while modifying and developing
some of its institutions.
In ideology, the first successors of Chinggis Khan followed tradition and maintained the
belief that the Khan ruled by the mandate of Heaven (Tengri); the forefather of the Altan
Urugh (Golden Kin) Bodonchar was considered to have been born from Light. 19 They also
paid attention to spiritual factors in their policy for the subjugation of different peoples,
adhering to the following instruction: 'Having seized the body, hold the soul. If the soul

17
Barthold, 1968, p. 461.
18
Lattimore, 1963a, p. 62.
19
Cleaves, 1982, 20.

260
ISBN 978-92-3-103467-1 Contents The Mongol empire ...

is held, the body will not go anywhere.' 20 In this connection reference should be made to
the religious policy of the Mongol Khans. Religious fanaticism was alien to the Mongols;
they pursued a policy of religious tolerance in their multinational empire. Some scholars
have held that this was determined simply by the Mongols' indifference or ignorance; but
it may rather have been a premeditated approach necessitated by 'holding the soul' of the
subjugated peoples belonging to different ethnic groups and beliefs. It may well be that the
Mongols' religious tolerance was influenced by the attitude of their nomadic predecessors,
like the Uighurs and the Kitan, towards the great variety of religions coexisting in Central
Asia.
The first Chinggisid rulers endeavoured to strengthen their control all over the con-
quered countries by consolidating the rule of their sons and relatives in their own domains,
granted as appanages. But the more the empire expanded, with its various parts ruled by
different agents, the more necessary it became for the supreme imperial power to avert
the danger of discord and disunity. In this connection, some traditional institutions of the
nomadic society acquired particular significance and were modified and strengthened in
conformity with the new requirements. Thus the kurultay, the ancient political institution
of the nomads, now assumed greater importance than it had ever had previously. It became
a true assembly of the Mongol elite, princes and nobles acting on the basis of old traditions
and customs in order to handle the most important matters of state, such as the election of
a Khan, the question of war, the establishment of law and issues of policy. All the great
Khans, including Chinggis himself, had to be proclaimed at a kurultay especially convened
for this purpose, and quite a number of kurultays are known to have been convened in order
to discuss other important military and governmental affairs of the Mongol empire.
The Mongol empire was created through military conquest and the Mongol Khans
regarded the army as the basic institution of the empire. The organization based on the
decimal system was not only maintained for several generations of Mongol Khans, but
also served as the model for the armies of their followers and pretenders to the heritage of
the members of Chinggis Khan's family. An important new element of the Mongol army
structure during the Chinggisid period was the institution of the tamma. Tamma forces
were originally established by order of the central imperial government for the purpose of
maintaining control of the conquered territories. Some tamma armies ultimately became

20
According to Lubsan-Danzan, 1937, p. 46, this instruction emanated from Chinggis; but according to
another source written in the so-called 'Square script', it came from Qubilay or Khubilay Khan (album Najm
az-caja'ib, in Istanbul University Museum, in Ligeti, 1972, p. 123).

261
ISBN 978-92-3-103467-1 Contents The Mongol empire ...

the nuclei of the permanent military forces of the empire's subsidiary Khanates, such as
Htilegti's 11 Khanate in Persia. 21
Among the numerous institutions faciliting control within the empire, the communi-
cations system should be mentioned. The Mongols were among the first to introduce a
transcontinental network of communications, thus encouraging the movement of peoples
and ideas. The real initiator here was Ogedey Khan: 'I made one to establish post stations
for that Our messengers, hasting on the way, make speed, and again for that We make them
to convey Our needs and necessities.' 22 Having consulted with his brothers, Ogedey insti-
tuted the jam (yam) system, setting up post-stations within his dominions. It was further
extended by his brothers Chaghatay and Toluy and by his nephew Batu to include the lands
under their direct rule. For the first time in history, a network of post-stations was estab-
lished covering the whole of Central Asia; its efficient functioning impressed European
travellers such as Marco Polo. 23 The structure of the system was based on post-stations
(jams), established at stages equivalent to a day's journey. The stations held horses and
stocks of fodder for those who travelled. Official envoys or messengers had to carry a spe-
cial authorization tablet called in Mongolian gerege, made of wood, silver or gold, in order
to make use of the system. Normal traffic might travel some 40 km a day, but express
messengers could go very much faster, covering up to 300-500 km a day. 24
As regards Mongol rule over the great sedentary societies of Persia and China, it should
be noted that the Mongols invented several institutions and offices which not only func-
tioned efficiently, but left a noticeable imprint on the civil administration and the govern-
ment of the conquered countries. One of the key institutions in local administration was
the office of the darughachi. This system was set up in all the Mongol-ruled regions of
Inner Eurasia, i.e. Persia, China and Russia, with the purpose of controlling the conquered
territories. The term darughachi (in Russian, darugha or its Turkic equivalent baskak; in
Persian, darugha) was widely known all over the empire. The Mongol preference for the
hereditary transfer of the office of darughachi was valid in all the parts of the empire. But
in most cases, for instance in China, where problems arose due to the insufficient num-
bers of Mongols capable of holding the office, the Mongol Khans enlisted the services
of Western Alans and Central Asians in order to guarantee Mongolian and non-Chinese
predominance in the local civil bureaucracy. 25 This made it easier for Uighur, Persian and
other Asian Muslims to gain high positions in the Mongol bureaucracy in various parts of

21
Morgan, 1986, pp. 94-5.
22
Cleaves, 1982, 281.
23
Ricci, 193 1, pp. 152-7.
24
Ibid., p. 157.
25
Endicott-West, 1989.

262
ISBN 978-92-3-103467-1 Contents The Mongol empire ...

the empire. Although the office of darughachi was first entrusted by Chinggis Khan with
mostly military tasks, its main function gradually developed into that of the chief civil offi-
cial stationed in the conquered territories; thus one of the primary duties of darughachis in
Persia, Russia and Central Asia was the collection of tribute.
The Khans introduced various forms of taxation in the regions of their empire. From his
examination of the Persian sources of the Mongol period, Petrushevsky has concluded that
some forty-five different terms for taxation were used, though this does not imply that there
were forty-five distinct taxes, since the terminology varied at different times and places. 26
The original Mongol taxation may be divided into tribute (alba) and levy (qubchighur or
qubckur). In 1229 Ogedey Khan issued the first decree with the aim of regulating the alba:

[As we are] sitting on the [throne made] ready by Our father the Kaghan, not making the
people to suffer, for [lack of] broth, from these peoples, in year after year, let one give one
sheep of two years old of [every] flock. Let them, bringing forth one sheep from one hundred
sheep ... 27

Alba and qubchur were paid in kind. With the conquest of the sedentary populations of Per-
sia and Central Asia, however, qubchur gained a rather different meaning and it became the
term used for a poll tax, either fiat-rate or graduated, fixed and imposed on the conquered
sedentary peoples. 28
The first three successors of Chinggis Khan were committed to making Mongolia the
centre of their empire. Such a policy could not but favour the political, economic and
cultural revival of the country, although this did not continue for long. Mongolia had
become a vortex of great events and innovations, with extended relations with other coun-
tries, particularly those of Central Asia. For a while, it became a meeting-place of differ-
ent peoples, cultures and religions. Its capital Karakorum was a cosmopolitan city where
Christians, Muslims, Buddhists and such nationals as Hungarians, Alans, Russians, Geor-
gians, Armenians and, of course, Chinese and Central Asians, mingled freely. The city
itself was mainly built by captured artisans. Thanks to their knowledge and abilities, Uighur,
Persian and other Central Asian Muslims occupied high posts in the Mongol bureaucracy;
their role in the ruling of the empire was no less essential than that of the military lead-
ers and their strategies for conquering other countries. Mongolia became directly involved
in the caravan trade between East and West, with a liberal policy by the Mongol Khans
towards trade. The jam system of communication greatly favoured commercial travel, and
the caravan trade was generally protected and encouraged. Muslim merchants of western

26
Petrushevsky, 1968,p.529;Morgan, 1986,p. 100.
27
Cleaves, 1982, 279.
28
Morgan, 1986, p. 100.

263
ISBN 978-92-3-103467-1 Contents The Yuan empire of the Mongols and its fall

Central Asia co-operated with Mongolian Khans and the nobility and were particularly
active in money-lending and tax-collection. All this tended to come to an end with the
creation of the Yuan empire in China, however, and the centre of trade, as of most other
activities shifted from Karakorum to Khanbalik

The Yuan empire of the Mongols and its fall

After the death of Mongke Khan in 1259, internecine disputes broke out among the Toluyids
themselves. In 1260 the two sons of Mongke, Arig-boke and Qubilay, simultaneously had
a kurultay convened and each declared himself Khan. In the Mongol traditionalist view,
Qubilay's rise to power was illegal and a four-year fratricidal war between the two broth-
ers ensued, with Arig-boke wishing to retain the centre of the empire in Mongolia. After
his defeat, the struggle was continued by Qaydu until his death in 1303.
During the reign of Qubilay Khan (1260-94), the Mongol empire underwent a great
transformation. The transference of the empire's capital from Karakorum to Khanballk (or
Dai-du) in northern China, and the adoption of the Chinese title Yuan (The Origin) for the
Mongol reign in China, meant a change in the traditional policy of Qubilay's predecessors.
Theoretically, Qubilay Khan was the ruler of the whole empire and tried to avail himself of
his rights in all the subjugated countries. But in reality, in his time and particularly in the
reign of his successors, the component parts of the Mongol empire progressively achieved
such a degree of independence that it was difficult for the Khan to claim the status of
universal ruler. It is probable that Qubilay Khan realized this and regarded himself more
as a Mongol-Chinese emperor than as a universal sovereign. Despite having settled in
China, Qubilay did not forget that he was a Mongol emperor and did his best to make
the Mongols an elite group in the conquered country. Having thus secured Mongol rule
in China, it became possible for a small, non-Chinese ethnic group to govern the huge
Chinese sedentary society for almost a century.

264
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

13

CENTRAL ASIA UNDER THE RULE OF

SUCCESSORS*
B. Akhmedov revised by and D. Sinor

As a result of both major and minor wars waged with great effectiveness between 1188
and 1206, Chinggis Khan (1155-1227) laid the foundations of a new state destined to play
a major role in the history of the peoples of Central and East Asia, and of eastern Europe.
Chinggis Khan liquidated the Tangut state of Hsi Hsia (982-1227), located on the frontiers
of the contemporary Chinese province of Gansu and the western part of Shanxi, and the
Jurchen Chin empire which covered the territory of north-east and northern China. Having
defeated the Chin, the Mongols directed their conquest against the powerful state of the
Khwarazm Shahs. On their way, they crushed the Kara Khitay empire, which was by then
under the leadership of Ktichltig. Between 1219 and 1224 the Mongols conquered, one
after the other, Utrar, Binakat, Khujand, Bukhara, Samarkand, Gurganj and other Transox-
anian towns, garrisons of the Khwarazm Shah c Ala' al-Din Muhammad (1200-20), and
established their rule in Transoxania and Khwarazm.
Already during his lifetime, the vast empire of Chinggis Khan was divided into domains
(ulus) which he assigned to his sons JOchi, Chaghatay, Ogedey and Toluy. His eldest son
JOchi (who predeceased his father) received an appanage ranging from the Irtysh river
'as far as Mongol hoofs had beaten the ground', and the lower Syr Darya (the towns of
Si'gnak, Barchkent and Yengi-kent) and north-western Khwarazm. JOchi's summer seat
was on the Irtysh, while his winter quarters were on the lower Syr Darya. Chaghatay
received Kashghar, Yeti Su and Transoxania. Later, under Baraq Khan (1266-c. 1271), the
Chaghatayids were to spread their power over northern Afghanistan as well. Chaghatay's
seat was on the Ili river. Western Mongolia and Tarbaghatai were assigned to Ogedey, who

* See Map 6, pp. 434-5.

265
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

resided in Chughuchak. The youngest son Toluy inherited his father's former ulus, i.e.
Mongolia proper. His seat was on the banks of the Kerulen river.
The founder of the Golden Horde 1 was JOchi's son Batu (1236-56), the conqueror of
eastern Europe who also played an important role in the political life of the entire Mon-
gol empire under the Great Khans Ogedey (1229-41), Gtiytik (1246-8) and Toluy's son
Mongke (1251-9). Batu and his successors ruled over vast territories not only in Transoxa-
nia but also in Iran; it was a huge empire, whose exact frontiers cannot be exactly defined.
In the northeast, the Golden Horde included Volga Bulgharia; in the north the frontier fol-
lowed that of the Russian principalities. In the south, the territory of the Golden Horde
included the Crimea, the Caucasus up to Darband, occasionally Baku, and also northern
Khwarazm with the town of Urgench. The frontier followed the steppes from the Dniestr
to western Siberia and the lower Syr Darya. The capital of this state under Batu was Saray-
Batu (Old Saray), located not far from Astrakhan, while under Berke Khan (1257-66) it
was Saray-Berke (New Saray), located on the Aktuba, a branch of the Volga.
As far as the ulus of Chaghatay (1227-42) was concerned, initially it comprised only
the lands from the country of the Uighurs in the east to Samarkand and Bukhara in the
west. But Chaghatay held it from the Great Khan, or Kaghan, only as inju (crown land),
civil power being exercised, on his behalf, by the Khwarazmian Mahmud Yalavach and,
after his transfer to China (after 1239), by his son Mascud Beg (d. 1289). Military power,
including duties such as taking a census of the population and collecting taxes and exac-
tions, was in the hands of Mongol officials called darughachi or tammachi (see above, p.
292). Judging by the fact that Mahmud Yalavach was able to prevent the Mongol comman-
ders Ildiz-noyon and Jighan-khorchi, who had crushed the rebellion in 1238 of Mahmud
Tarahi (see below), from plundering Bukhara and slaughtering its inhabitants, he must have
enjoyed great power in the ulus of Chaghatay, and the local Mongol officials were clearly
obliged to abide by his orders. Mahmud Yalavach and his son Mascud Beg were account-
able only to the Great Khan Ogedey. The following story by the historian Rashid al-Din is
characteristic:

It is said, that during the reign of Ogedey kaghan, Chaghatay ... gave some of the provinces
of Transoxania, which, by the command of the kaghan were under the control of Yalavach,
to someone else. Yalavach reported the matter to Ogedey who sent an order to Chaghatay
rebuking him and ordering him to write an answer. Chaghatay wrote in his reply: 'I acted from
ignorance and without guidance. I have no answer that I can write, but since the kaghan has
ordered me to write I dared to write this.' The kaghan was pleased and accepted this excuse;
and he gave that province to Chaghatay as injU. Thereafter Yalavach came to visit Chaghatay,

1
A name apparently given to them by the Russians, although Russian and Polish-Lithuanian sources
usually refer to it simply as 'the Great Horde', see Bosworth, 1996, p. 253.

266
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

who rebuked and abused him. Yalavach said to Vazir [one ofChaghatay's viziers]: 'I should
like a word with you in private.' And when they were closeted together he said to Vazir: 'I
am the kaghan's minister and Chaghatay cannot put me to death without consulting him. If
I complain of you to the kaghan he will put you to death. If you will set matters to right
with me, well and good; otherwise I shall denounce you to the kaghan. And if you repeat
these words to Chaghatay I will deny them however much I am questioned, and you have no
witness.' On this account, Vazir was forced to put the matter to rights. 2

Soon afterwards, around 1239, probably because of his strained relations with Chaghatay,
Mahmud Yalavach was transferred to China and his son Mascud Beg was appointed governor-
general of Turkistan.
The political position of the Chaghatay ulus until the reign of Kebek Khan (1309;
1318-26) was unstable. Chaghatay's grandson, Qara Htilegti (the son of MotOghan, killed
at the siege of Bamiyan in 1221), did not rule for long (1241-7) and he was deposed by the
Great Khan Gtiytik (1246-8), who enthroned Yesti Mongke (1247-52), Chaghatay's son.
Yesti Mongke was Gtiytik's intimate friend, but he spent his time carousing and, like many
Khans of the house of Chaghatay, did not effectively participate in governing the coun-
try. All power was in the hands of his wife Toqashi and of the Tajik vizier Baha' al-Din
Marghinani, son of the Shaykh al-Islam of Ferghana.
After Gtiytik's death his widow Oghul Qaymish acted as regent until 1251. In the forth-
coming election of the Great Khan, Batu decided to lend his support not to one of her
sons but to Toluy's son Mongke, and had him elected at a kurultay (general council) held
in Karakorum in 1251. As might have been expected, the princes of Ogedey's lineage
opposed this decision and were supported by Yesti Mongke, Chaghatay's fifth son and the
ruler of the Chaghatay ulus. Following Mongke's accession to the throne, many of those
who had opposed his election were executed, including Oghul Qaymish. The deposition
of Yesti Mongke was also proclaimed and Qara Htilegti was appointed in his place. How-
ever, on his way to the ulus Qara Htilegti died and it fell to his widow Orqina Khatun to
have Yesti Mongke put to death. Toqashi Khatun was also executed. Qara Htilegti's widow
and her under-age son Mubarak Shah were appointed to the Chaghatay ulus, but they were
considered only nominal rulers. In fact, power was in the hands of Mascud Beg, who ruled
in the name of Batu and Mongke.
Hence, after the Karakorum kurultay of 1251, the Chaghatay ulus came to be split into
two: East Turkistan, the Kulja region and Yeti Su (Semirechye ), apparently, together with
the north-eastern part of Ferghana, came within the Kaghan's sphere of influence. In Tran-
soxania and the western part of Ferghana the Golden Horde's influence was preponderant,

2
Boyle, 1971 , p. 156.

267
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

judging by the indirect evidence, in particular the returning by Batu of all Temtir-Malik's
properties in Khujand to his son. In the words of the Franciscan William of Rubruck, the
frontier between Mongke's and Batu's domains followed the steppe between Talas and the
Chu river, east of the Alexander mountain range.
To save the Chaghatay ulus from annexation by the Golden Horde, Alughu (1260-4),
son of Baydar, son of Chaghatay, continued the struggle against Berke Khan. Finally, a
serious blow was delivered to the Golden Horde: a 5,000-strong garrison of the Golden
Horde encamped at Bukhara was, in the words of Wassaf, 'withdrawn from the town to the
steppe and exterminated; their property, wives and children were confiscated' .3 An account
by Rashid al-Din also deserves attention. He relates that under Alughu, the Chaghatayid
army defeated Berke Khan's troops encamped near Utrar: 'He [Alughu] assembled the dis-
persed troops, and then fought with Berke, crushed him and pillaged Utrar.' 4 Thus Alughu
brought Transoxania under Chaghatayid rule. Alughu' s successor, Mubarak Shah, a Mus-
lim, was, as already mentioned, a weak ruler, who soon after his enthronement was deposed
by his cousin Baraq.
Under Baraq (1266-71), the Chaghatay ulus became somewhat stronger. Baraq tried
to carry on an armed struggle against Qaydu ( Ogedey's grandson), and Mongke-Temtir
(1267-80) of the Golden Horde, who were allies. Once, on the banks of the Syr Darya, he
succeeded in defeating them, but subsequently suffered a defeat himself. The war with the
11 Khan Abaqa (1265-82) did not bring Baraq success either. But in the last years of his
life, the Chaghatay ulus was once again divided into distinct spheres of influence, those of
Qaydu, in the 1260s, of Mongke-Temtir of the Golden Horde, and of Baraq who received
only about two-thirds of Transoxania.
Following the short reigns of Negtibey (Nikpay in the Persian sources; c. 1271) and
Toqa-Temtir (1272-91), Qaydu enthroned in the Chaghatay ulus Baraq's son Duwa Khan
(c. 1282-1307), with whose name the restoration of Andijan and its becoming the capital of
Ferghana are associated. Duwa Khan was a true ally of Qaydu and actively participated in
his military campaigns in Mongolia proper; he also interfered in the internecine struggles
of the JOchids in the White Horde and, after Qaydu's death (in the autumn of 1301), he
enjoyed great authority within the territories of his successor Chapar. Accordingly, Duwa
Khan can be seen as the true founder of the Chaghatayid state. After his death, disturbances
again began in the Chaghatay ulus. The reigns of Konchek Khan (son of Duwa Khan), who
was enthroned near Almali'k, in a small town of Sabqu-bala, and Talighu (son of Qadami,

3
Barthold, 1968, p. 490.
4
Ibid.

268
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

son of Bori, son of Mittigen, son of Chaghatay) were not long. They ruled for hardly more
than two years, marked by a revolt of princes led by Kursabe, a descendant of Ogedey.
To some degree, Duwa Khan's son Kebek Khan succeeded in curbing the separatist
tendencies of his relatives. At a kurultay held in 1309, he had his elder brother Esen Buqa-
Boge (1309-18) elected as Khan. Kebek Khan and his brother Yesti-Boge were able to
annex the largest part of Qaydu's domains and to some extent stabilize, at least temporar-
ily, the socio-political situation in the country. Kebek Khan (who succeeded his brother
and ruled from 1318 to 1326) holds a special place in the history of the Chaghatay ulus.
For example, his name is linked to the currency and administrative reforms which played
an important role in the development of feudal statehood in Central Asia. His name is also
linked to the building and restoring of the towns of Transoxania. Among his new construc-
tions there was, for example, a palace (in Mongolian, qarshi) located near Nasaf around
which a whole town later grew up. Among the towns restored by Kebek Khan was ancient
Balkh, 'which, from the time of the Great Sahib-Qiran [i.e. Chinggis Khan] was deserted
and turned into a tangle ofreeds'.
The administrative and monetary reforms of Kebek Khan were aimed at putting an end
to confusion and checking the abuse by the various officials and speculators. The admin-
istrative reform divided the country around Bukhara and Samarkand into tumens, and in
Ferghana and East Turkistan into orchins (literally 'near', 'around', 'surrounding'), i.e. a
region located around the capital. As for the monetary reforms, the systems of 11 Khanid
Iran and the Golden Horde were utilized as models. The weight of 1 kebek dinar was 2
mithqals and 1 kebek dirham was equal to 1/3 of a mitbqal. The administrative and currency
reforms of Kebek Khan were only superficial, however, and internal problems remained.
The new monetary unit became known as kebek, a term that survives in the Russian word
kopek.
The paucity of sources makes it difficult to give a detailed picture of the social, eco-
nomic and cultural aspects of life within the Chaghatay ulus. Fragmentary evidence pro-
vided by historical sources such as the works of Juwayni, Rashid al-Din, Wassaf and Jamal
Qarshi allows us to say only the following. The Chaghatay ulus was a decentralized state,
with governors appointed by the Kaghan (for the settled regions, until 1289) and rulers of
provincial districts, i.e. princes assisted by special officials, the darughachi or tammachi,
the representatives of Mongol power. Leaders of local origin such as Mahmud Yalavach
and Mascud Beg from Khwarazm, Habash cAmid from Utrar, Baha' al-Din Marghinani,
the vizier Yestinte Mongke from Ferghana and others actively participated in government.
Transoxanians were also involved in the political life of China under the Yuan dynasty
(1279-1368). Among them there were, for example, Mahmud Yalavach, his son cAli Beg

269
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

and Yacqub son of cAli Beg, Khwarazmians by origin, Shams al-Din Sayyid Aja and his
son cAla' al-Din who came from Almali'k, Baha' al-Din from Qunduz and others.
Because of favourable climatic conditions, Transoxania had long been a region of devel-
oped agriculture based on artificial irrigation. The principal crops were cotton, grain,
gourds, alfalfa and grapes. Furthermore, the region played a major role in the transit trade
linking China, the Near and Middle East and Europe. Crafts and trade had developed in the
towns there: besides Bukhara and Samarkand, there were also Khujand, a residence of the
Kaghan's deputies in Transoxania such as Mahmud Yalavach and Mascud Beg; Uzgend
where, as under the Karakhanids and the Kara Khitay, the Kaghan's treasury was kept;
Andijan, developed by Duwa Khan and turned by him into the capital ofFerghana; Marghi-
nan, which was a centre for many scholars and poets; Isfara, the home of a talented poet
of the thirteenth century, Sayf al-Din Isfaraghi (d. between 1261 and 1267); and Kuba (the
present Kuwa), the home town of Rukn al-Din Qubani (thirteenth century) and others.
As mentioned above, the Chaghatayids had only inju rights in the ulus, that is, they
had only the right to make use of the revenues. As far as the taxes and duties levied on
the subject population are concerned, we have only very general indications. For example,
Rashid al-Din gives information on the main taxes levied on landowners and nomads. Thus
the amount of mal (land tax) was 10 per cent of all the harvest yield, and a tax levied on
nomads (qubchighur) was 1 per cent of 100 head of cattle. The targhu, a kind of trade
duty, was also collected. At first, taxes and duties were paid in kind, but from the 1250s,
after the introduction of various types of Mongol monetary units and the coinage issued
by Mongke (1251-9), and later, particularly from 1270 onwards, duties came to be paid
in cash. Coinage was minted in many large towns, such as Almali'k, Bukhara, Samarkand,
Utrar, Taraz, Kashghar, Tashkent, Ush, Marghinan, Ak-tepe, Uzgend and Khujand.
With regard to the socio-economic life of the people of Transoxania under the
Chaghatayids, the revolt of the population of the Bukhara district, which took place in
1238 and was led by an artisan called Mahmud Tarabi, is worthy of attention. The cause
of the revolt was the suffering of the masses caused by the oppression of local landowners,
further aggravated by the outrages of the Mongol officials and tax-collectors. As related
by Juwayni and, following him, Ulugh Beg, the revolt began in the village of Tarab near
Bukhara and very quickly spread throughout the region. Several thousand rebels, armed
with sticks, spades and axes, moved towards Bukhara. Some Mongol high officials fled to
Karmina, others pretended to side with the rebels. These officials planned to kill Mahmud
Tarabi on his way to Bukhara and thus to stifle the revolt, but the plot was unsuccess-
ful. Eventually, the rebels occupied Bukhara and encamped on the height of Abu Hafs,
situated on the northern side of the city. Mahmud Tarabi was brought to Malik Sanjar's

270
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

palace and proclaimed caliph. The rich Mongol high officials who had no time to flee were
arrested and executed and their properties were distributed among the poor. The Mongol
commanders who had fled to Karmina assembled several scattered Mongol detachments
and came out against the rebels, but were defeated. Nevertheless, because the rebel leaders
- Mahmud Tarahi, his brothers Muhammad and c Ali, the scholar-theologian Shams al-Din
Mahbubi and others - were inexperienced, the revolt did not spread beyond Bukhara. Tak-
ing advantage of this, Mahmud Yalavach sent out forces from Khujand and defeated the
insurgents; altogether some 21,000 men were killed. Thus the revolt was suppressed, but it
showed the Mongols that the people hated the regime and could muster enough strength to
challenge the entire establishment.
An essential feature in the life of the Chaghatay ulus in the middle of the thirteenth
century was a growing conflict between Chaghatay's descendants who governed the vari-
ous regions. As far back as the times of Mubarak Shah (1266) and Baraq (1266-71), some
princes had aspired towards the establishment of stable links with the settled population
of Transoxania. Thus Mubarak Shah moved from Yeti Su to the valley of Ahangaran, and
Baraq at first moved to Chaghaniyan, where in 1266 his election took place. Conversely,
a strong group of the military-nomadic aristocracy headed by Qaydu, Yasa'ur and Buzan
favoured a nomadic way of life; they repeatedly attacked the settled regions of the country,
pillaged the population and burned towns and villages.
After Kebek Khan, the Chaghatay ulus was again involved in internecine warfare. Thus
in one single year (1326), Elchigidey and Dua-Temtir, both sons of Duwa Khan, followed
each other on the throne. Tarmashirin (1326-34), who, because of his adherence to Islam,
was called c Ala' al-Din (Grandeur of the Faith), finally settled in the western part of the
Chaghatay ulus and no longer came to Almalik The eastern part of the ulus, which also
included a part of Ferghana, fell under the power of nomadic feudal lords and Khans: Buzan
(Dua-Temtir's son) and Changshi (Abughan's son), both of them grandsons of Duwa Khan,
and Yestin-Temtir (Changshi's brother; 1334-8) ruled in name only.
In the 1340s the Chaghatay ulus finally disintegrated into two parts: Moghulistan
(which included Yeti Su, the eastern part of Ferghana and East Turkistan) and Transoxania
proper. The western part of the ulus also included eastern Khwarazm. The years 1340-70
witnessed an aggravation of disturbances and internecine wars and the feudal disintegra-
tion of the ulus into smaller, independent domains. In the main regions of the western part
of the ulus, power was seized by tribal leaders. Thus Kish and its regions fell into the
hands of the amir Hajji Barlas; Bayazid J alayir took possession of the Khujand region; the
Balkh region passed into the hands of Husayn, grandson of the Turkish amir Kazaghan,
killed in 1358 while hunting; and in Shiburghan, the standard of independence was raised

271
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA UNDER THE RULE ...

by Muhammad Khwaja Apendi, the leader of the Naiman tribe. There were also regions
where power was in the hands of local feudal nobles: the sadrs (leaders of the Muslim
religious class) took possession of Bukhara and its regions, the local Kaykhusraw Sayyids
of Termez took possession of Khuttalan, etc. A similar situation prevailed in the eastern
part of the ulus in Moghulistan. Here also, in spite of the firm hold of the family-tribe
tradition, disturbances and internecine struggles began. The Khans no longer wielded real
power: it was wholly in the hands of nomadic feudal lords. It was of this state of affairs
that the clever and enterprising Timur (known to the West as Tamerlane ), the son of Barlas
Beg Taraghay, took advantage.

272
ISBN 978-92-3-103467-1 Contents THE DELHI SULTANATE

14

THE DELHI SULTANATE*

Riazul Islam and C. E. Bosworth

Contents

THE ESTABLISHMENT OF THE SULTANATE IN THE THIRTEENTH CENTURY

AND THE NATURE OF THE NEW STRUCTURES IN INDIA 274
Background . . . . . . . . . . 274
Qutb al-Din Aybak (1206- 10) 275
Shams al-D!n Iltutmish (1210-36) 275
Nasir al-D!n Mahmud (1246- 66) and Ghiyath al-D!n Balban (Ulugh Khan) (1266- 87) 276
The end of Turkish supremacy . . . . . . . . . . . . . . . . . . . . . . . 277
Jalal al-D!n Firiiz Khalj1 (1290-6) and c Ala' al-Din Khalj1 (1296-1316) . 277
Income levels among the ruling and scholarly elites 281
Agrarian conditions in the fourteenth century 282
The political structure of the state . . . . 283
Social and economic developments . . . 283
THE DELHI SULTANATE , 1316-1526 284
The Tughluqids (1320- 1412) 284
The Sayyids (1414-51) 291
The L6d1s (1451- 1526) 293

* See Map 7, p. 437.

273
ISBN 978-92-3-103467-1 Contents Background

Part One

THE ESTABLISHMENT OF THE SULTANATE IN

THE THIRTEENTH CENTURY AND THE NATURE
OF THE NEW STRUCTURES IN INDIA
(Riazul Islam)

Background
Under the Ghaznavid Sultan Mahmud and later the Ghurid Mucizz al-Din, during the
period lasting from the death of Harsha (646-7) to the Turk invasions of northern India, the
socio-political configuration was dominated by a number of factors which help to explain
the rapidity of the Muslim conquest. First, the feudal-like system clearly favoured the rulers
and the ruling classes at the expense of the peasantry. Second, the Rajputs - mostly of for-
eign origin, but gradually absorbed into the fighting caste of the Hindus - who emerged
as a political force after the fall of the Pratiharas, had a passion for war and often went to
war to enhance their prestige. The Rajput political structure, feudal and hierarchic in char-
acter and lacking a strong central force, encouraged fissiparous tendencies. The Rajputs'
narrow vision, even narrower loyalties, and endless and purposeless internecine fighting,
contributed to the military and political particularism which prevented a collective response
against foreign invasions during the eleventh and twelfth centuries. Third, no strong cen-
tral authority for the entire subcontinent existed. The Rashtrakutas of the Deccan extended
their authority to the north; the Pratiharas, with Kanawj as their seat, from Panjab eastward;
and the Palas from Bengal westward. This led to the formation of three large and separate
kingdoms in the Deccan: one in the north, one in the east and one in the west. Much of
their strength was wasted in mutual warfare. The predecessors of the Pratiharas, the Gur-
jaras who had ruled over Panjab and Marwar, are given credit for stalling the Arab east-
ward expansion from Sind. The Pratihara dominance of northern India, which had acted
as a shield against external aggression during the major part of the ninth and tenth cen-
turies, now disintegrated, leaving India exposed to foreign invasions. After the Ghaznavid

274
ISBN 978-92-3-103467-1 Contents Shams al-Din lltutmish (1210-36)

and, especially, the Ghurid invasions (see above, Chapters 5 and 8), Islam spread from its
foothold in the extreme north-west of the subcontinent into much wider regions.

Qutb al-Din Aybak (1206-10)

Mucizz al-Din Ghiiri's leading slave generals succeeded him: Yildiz at Ghazna, Qutb al-
Din Aybak at Lahore and Qabacha at Uchch. Aybak was undoubtedly the late sultan's most
trusted lieutenant and thus his main successor in India. But his four years of stewardship of
the Ghurid Indian dominions were marked by his struggles against Yildiz, the Turkish ruler
of Ghazna; against Qabacha, who controlled Sind and Multan; and against the rebellious
Hindu Rajahs, who wanted to throw off the Muslim yoke. Aybak's accidental death during
a game of polo in 1210 ended a promising career, but his role as lieutenant during Mucizz
al-Din's life, and later as his successor, entitles him to an important place in the formative
history of the Delhi Sultanate.

Shams al-Din lltutmish (1210-36)

Iltutmish ascended the throne of Delhi in difficult and markedly uncertain circumstances.
The defiant attitudes of the senior slave generals like Qabacha and Yildiz, the revival of
resistance among the Hindu ruling classes, and above all, the threat from the growing power
of the Chinggisid Mongols across the North-West Frontier, posed great challenges. The
Khaljis in Bengal and Bihar withdrew their allegiance. Iltutmish displayed great intrepidity
in the face of all these difficulties and showed a shrewd sense of strategy and timing in
tackling the various problems. He humbled the hostile Turkish generals; overcame Hindu
resistance; re-established his authority in the eastern provinces; and, through a combination
of strategy and luck, succeeded in saving his kingdom from the Mongol onslaught.
Iltutmish, the first sovereign ruler of Delhi, is rightly considered the founder of the
Sultanate of Delhi. He is given credit for creating durable foundations, organizing the
administration and evolving statesmanlike basic political policies. In 1229 al-Mustansir,
the cAbbasid caliph of Baghdad, conferred a mandate of authority on Iltutmish; this was an
event of considerable significance, for it made the Sultanate of Delhi a legally and morally
recognized state in orthodox Muslim eyes. Himself a man of piety and learning, Iltutmish
maintained cordial relations with the cuzama' and the mashayikh (Sufi leaders and saints),
thereby achieving acceptability and legitimacy for his new sultanate.
None of Iltutmish's five successors - two sons, one daughter and two grandsons who
followed each other in quick succession - proved to be capable leaders. The Mongols kept
pressing on the frontier and both Lahore and Multan were subjected to raids and spoliation.

275
ISBN 978-92-3-103467-1 Contents Nasir al-Din Mahmud (1246-66) ...

Provincial governors found an opportunity to extend their autonomy and the Hindu rulers,
in particular the Rajputs, showed signs of disaffection.

Nasir al-Din Mahmud (1246-66) and Ghiyath al-Din

Balban (Ulugh Khan) (1266-87)
One of the notable developments in the post-Iltutmish period is the emergence of a group of
nobles - all slaves of Iltutmish - called the Ghulaman-i Chihilgani (possibly meaning 'the
slave commanders who each commanded forty slaves'), who attained a dominant position
in the court. For thirty years the 'Forty' held the royal power in commission and reduced
the sultan to a figurehead. Among the powerful 'Forty', the dominant figure of Ghiyath
al-Din Balban emerged. He had gained considerable power even before the accession of
Sultan Nasir al-Din Mahmud, the last ruler of the line of Iltutmish. Shortly after Nasir al-
Din's accession, Balban, now called na'ib-i mamlakat (viceroy), in effect assumed power
as regent, reducing the sultan to a titular ruler. During the two decades that he was at the
helm as na'ib-i mamlakat, Balban tried to stem the rot that had set in during the decade of
anarchy (1236--46).
Having served the sultanate at all levels, Balban had an intimate knowledge of the man-
ner in which it functioned and its sources of strength and weakness. He was thus able to
identify its core problem. He believed that the weakness of the crown lay at the root of
all the maladies of the state. His ideas on monarchy, government and religion, expressed
in his speeches to his sons and nobles, are sometimes labelled his 'political theory'. The
various elements of his thinking, though not elaborate or comprehensive enough to be con-
sidered a theory, are nevertheless coherent. Balban displayed great vigour and ruthlessness
in crushing political rivals and rebels and punishing refractory governors and local chiefs.
The ordinary people, in general, were not affected. With his blind belief in the supreme
value of nasab (good birth), however, Balban would not employ men of ordinary birth in
the army and the administration.

REORGANIZATION OF THE ARMY

The government depended essentially on force, or the threat of force, in order to preserve
its authority. Thus it was natural for the army to receive the utmost attention. With a view
to improving its efficacy as a striking force, Balban gave high priority to the reorganization
and expansion of the army. There is also an indication that Balban endeavoured to change
the payment of the soldiers' salaries from iqtac s (assignments of land; what in later, Mughal
times were to be calledjagfrs) to cash payments.

276
ISBN 978-92-3-103467-1 Contents Jalal al-Din FTruz KhaljT ( 1290-6) ...

The decades following the end of Iltutmish's reign saw a marked increase in Mongol
pressure on the western frontiers of the Delhi Sultanate. The governors of these regions,
ill-supported by the central government, were helpless in the face of Mongol inroads. By
the time that Balban came to the throne, large parts of Sind and Panjab were under Mongol
occupation. With his reorganized army, Balban made the defence of the frontiers a prior-
ity, his contribution here being twofold. First, he cleared Sind of the Mongol adventurers,
recovered Lahore and Multan and built a special force to protect the frontiers. Thus he held
a firm line against the Mongols. Second, following a realistic defence policy, he compro-
mised by holding a line between the Beas and the Ravi rivers, leaving large parts of western
Panjab in Mongol hands. During Balban's reign, the Mongols never attempted to proceed
beyond the Ravi and the security of Delhi and the central provinces was never under threat.

The end of Turkish supremacy: The Khalji revolution

Balban was succeeded in 1287 by his grandson Kay Qubad, who took the title of Mucizz al-
Din. This young, handsome, pleasure-loving and inexperienced sultan paid little attention
to the administration and soon lost all control of the affairs of state. The rising Khalji
clan soon replaced the house of Balban, and Jalal al-Din Rruz Khalji, an old officer of
Sultan Balban, ascended the throne of Delhi. The Khaljis, too, it is now fairly certain, were
originally of Turkish origin, but were Iranized because of their long stay in the steppelands
of Afghanistan. (The Turks did not consider the Khaljis their peers.) The fact that the
Khaljis did not demonstrate any racial elitism of their own enabled them to build a wider
political and social base for their 'new monarchy'. The change in the social base of power
was so pronounced as to justify the term 'Khalji revolution'.

Jalal al-Din Flruz Khaljl (1290-6) and cAla' al-Din

Khaljl (1296-1316)
Jalal al-Din Rriiz Khalji 's six-year reign was marked by incompetence and pusillanim-
ity. His lenience towards robbers and rebels, his half-hearted fight against the invading
Mongols, and his failure to seize the prestigious Ranthambor fortress from the Rajputs,
marked him as a ruler unsuited to the times. His ambitious nephew and son-in-law cAli
Garshasp showed little compunction in disposing of his uncle and ascended the throne as
cAla' al-Din Khalji.
To understand fully the reign of cAla' al-Din, one should look back at the thirteenth cen-
tury and take note of the salient socio-economic trends. The striking motif is the continuity
of the institutions. Not only were the conventional Indian methods of revenue-collection

277
ISBN 978-92-3-103467-1 Contents Jalal al-Din FTruz KhaljT ( 1290-6) ...

(mainly a simple produce-sharing system) largely left unchanged by the new rulers, but
even the collection agents, the ra'fs (chief), the chaudkrfs (heads of parganas, groups of
villages) and the patwarfs (village accountants), were mostly retained. The new rulers who
had taken over the immense lands were short of manpower and in need of funds; thus they
adopted the methods most likely to ensure rapid success. The state's demand for revenue
was deliberately kept at a low rate - one-fifth of the produce - and the countryside was
largely left undisturbed.
Forewarned by a number of rebellions early in his reign, c Ala' al-Din took prompt mea-
sures to forestall further trouble. First, in order to keep himself posted of all important
occurrences in the capital and the provinces, he strengthened the dfwan-i barfd (intel-
ligence department). Next, acting on the idea that 'wealth and rebellion are twins', he
adopted measures to extract as much wealth as possible from his subjects. From the rural
chiefs he demanded full taxes, while for the peasants the state demand for revenue was
increased. Finally, in order to keep the nobles from uniting against him, he issued strict
orders forbidding them to assemble or intermarry without royal permission.
In the military sphere, c Ala' al-Din's achievements fall into two categories: the war
against the invading Mongols and the conquest of the unsubdued Indian territories. During
the thirteenth century, the Mongols were so powerful that even a strong ruler like Bal-
ban had to adopt a defensive policy and accept a frontier line that was not particularly
favourable. c Ala' al-Din faced two Mongol attacks on Delhi, including a siege of the city;
but on both occasions the Mongols retreated. Other Mongol invasions directed at Panjab
and the Ganges valley were also defeated. Hence by the end of the first decade of his rule,
he had ensured protection from external aggression for his dominion. The death in 1306 of
Duwa Khan, the Chaghatayid ruler of Transoxania and the main inspiration behind these
invasions, may have also contributed to the decrease of Mongol pressure on India.
c Ala' al-Din's twenty-year reign entailed ceaseless military activity in India. The result-
ing acquisitions can be classed under three headings: areas recovered, territories freshly
conquered and annexed, and states subdued but not annexed. The most noteworthy recov-
ered areas were Jaysalmir, Ranthambor and Malwa. The most substantial and significant
newly conquered territory was Gujarat, for its annexation brought the sultanate a province
rich in natural resources as well as the benefits of extensive maritime trade. Chitor, too, was
conquered and annexed, but after a short period was placed under a loyal Rajput dynast.
The states subdued but not annexed include the three kingdoms of the Deccan and southern
India: Deogir ruled by the Yadavas; Telingana ruled by the Ganapatis; and Dwarsamudra
ruled by the Hoysalas. c Ala' al-Din's main goals regarding these rich kingdoms were to
obtain as much tribute as possible and to secure their submission to Delhi's suzerainty.

278
ISBN 978-92-3-103467-1 Contents Jalal al-Din FTruz KhaljT ( 1290-6) ...

Otherwise, the Rajahs were left free to manage their internal affairs. The general Malik
Kafiir, who was thrice sent to subdue the three kingdoms, met with unqualified success
and c Ala' al-Din's policy of establishing imperial hegemony, rather than direct rule, over
the distant Deccan proved eminently successful. There is no adequate explanation as to
why c Ala' al-Din made no attempt to conquer and annex Bengal, which was still ruled
by Balban's descendants; but for a reign of twenty years, his military achievements were
substantial.
A factor of paramount importance in c Ala' al-Din's far-reaching conquests and his suc-
cess in dealing with the problem of the Mongols was the quality and size of his army.
First of all, he did not suffer from the constraints which Balban had imposed in order to
limit the strength of the cadre of commanders. Talent and loyalty were the only criteria
by which c Ala' al-Din judged the men of the armed forces. He increased the strength of
the main wing of the army, the cavalry, to 475,000 well-equipped troops who were paid
directly from the treasury. Furthermore, he made the rules of annual muster more strin-
gent, thereby ensuring the preparedness of the troops. However, the expense of the salaries
for an army with such a large cavalry element would soon have exhausted the treasury.
To overcome this problem, c Ala' al-Din introduced price controls to ensure that a soldier
could live reasonably well on a lower scale of pay.
c Ala' al-Din was the first sultan to give serious thought to the reorganization of the
revenue system. While devising a plan, he kept in mind the following well-considered
objectives: to maximize the government's revenue, to equalize the burden of taxation on
the various sectors of the rural population, and to minimize the dangers of a rebellion by
the nobles and of rural discontent. He also introduced the rule of measurement of land
(which of course was familiar in India); this largely replaced the rule of sharing thepro-
duce, known for many centuries as bata'f. Being a realist, he did not impose the rule of
measurement on the entire realm, but only on a well-defined and carefully chosen core
of the sultanate. Clearly, a fixed and stable rule of measurement was in the government's
interests as it helped to ensure a stable level of revenue. Bata'f, on the other hand, favoured
the peasants, for under it they paid in proportion to what they produced. By shifting to
the land-measurement method, the sultan increased the pressure on the peasants to pro-
duce more. Furthermore, he increased the rate of the state's demand to 50 per cent of the
calculated produce, thereby more than doubling the rate (compared to 20 per cent under
Iltutmish and later Rruz Shah Tughluq). However, he also made provisions for revenue
exemptions in cases of crop failure resulting from natural calamities. Firmly insisting on
the principle that 'the burden of the strong shall not fall on the weak', he lightened the tax
burden on the peasants. In effect, he forced the superior rural classes (variously called kut,

279
ISBN 978-92-3-103467-1 Contents Jalal al-Din FTruz KhaljT ( 1290-6) ...

muqaddam and chaudhr() to pay their taxes themselves rather than pass the burden on to
the peasantry. He also abolished all the tax exemptions that they had previously enjoyed.
As a result, they were no longer in a position to oppress the peasantry at will. Thus although
the peasants lost under c Ala' al-Din, they also gained some advantages. The evidence is
the fact that, during the two decades of the sultan's reign, no rebellion occurred in the rural
areas.
c Ala' al-Din's strong rule alone cannot, however, account for the absence of articulated
discontent, but recorded cases of harsh treatment and punishment under his administration
mostly refer to urban political rebels and corrupt administrators, and not to rural mal-
contents. In any case, the peasants were left enough of their produce to enable them to
survive from one year to the next. The muqtacs (executive heads of provinces responsible
for the collection of revenues) and the staff of the dfwan-i wizarat (revenue department)
were, if found guilty of laxity or dishonesty in the fulfilment of their duties, treated with
marked severity; even governors were not spared physical beatings. As a result, first, the
collections became effective and regular; second, the lag between collections and deposits
was reduced; and third, the village people were saved from the extortions of the revenue
staff.

IQTJi.cs (ASSIGNMENTS OF LAND)

In lieu of salary, an assignment of land, or iqtac (sometimes simply a portion of the land
revenue), was granted to state employees. It saved the administration from having to keep
ready cash for the monthly salary payments, and substantially reduced the amount of paper-
work. The sultans of Delhi prior to c Ala' al-Din had followed this convenient and simple
method. Its main disadvantage, however, was that it enabled the recipients of large iqtac s
to gain extensive personal influence and thus become an impediment to the operation of
state power. It also provided loopholes for recipients to enjoy the benefits of the iqtacs
without fulfilling their obligations. c Ala' al-Din strongly disliked the system. Within five
years of his accession, he issued orders for the withdrawal of iqtac s, as well as other grants,
and their inclusion into khalisa (state-administered lands). This was a far-reaching change.
Financial benefits aside, it increased the authority wielded by the state over the bureaucracy
and the nobility. It seems, however, that the practice of giving iqtacs was not completely
abandoned, but was now restricted to special cases in which the sultan wanted to empha-
size the executive authority of a minister who had been entrusted with an important and
difficult task.

280
ISBN 978-92-3-103467-1 Contents Income levels and scholarly elites

MARKET-CONTROL REGULATIONS

One of c Ala' al-Din's most important measures - and one which has attracted a great deal
of attention - was the control of prices. It was introduced for the purpose of employing a
larger army on a lower scale of pay. Of all the requisites of the troops, the most important
single item was food. It therefore constituted the first of the four sectors of price control.
The prices of wheat and other commodities were fixed and elaborate arrangements were
made to ensure adequate supplies in the markets and to maintain huge reserves. The other
three sectors of price control were: (a) horses, ponies, cattle and slaves; (b) cloth and fruit;
and (c) articles for domestic consumption and personal use.
The prices of the various items in the four sectors were not changed during the rest of
the sultan's reign. Aside from the firmness of c Ala' al-Din's administration, other features
supported the system. First, prices were only fixed after very careful consideration and
were generally reasonable. The price of the most important item, wheat, was fixed at 7-1/2
jftals per man (1 man of 14 seers= approx. 13 kg). From the days of Balban to the reign of
Firuz Shah Tughluq towards the end of the fourteenth century, the price of wheat remained
stable, ranging from 7 to 8 jftals per man (except during periods of famine). Second, in
another controlled sector, we learn that prices were so fixed as to ensure a fair margin of
profit for the producer/seller. Both these features, namely the approximation of the fixed
price to the normal price and the allowance of profit to the producer, greatly contributed
to the stability of c Ala' al-Din's market-control arrangements. In addition, the sultan took
care to ensure that the market was never short of supplies. During periods of scarcity,
rationing was enforced. Through these devices, prices were kept steady at the fixed rate,
even under famine conditions. c Ala' al-Din also made sure to appoint men of honesty and
impartiality to the hisba (market control) staff. Apart from the troops stationed in Delhi,
the main beneficiaries of the system were men of modest income and of the lower salary
group in the capital. This explains the concentration within Delhi of a large number of
scholars, craftsmen and men of the various professions.

Income levels among the ruling and scholarly elites

During the initial phase of conquest, large areas and entire provinces were assigned to the
nobles, in order to collect revenue and consolidate the sultan's hold on the territory. The
muqtacs (assignee-governors of these territories) tended to wield a wide range of pow-
ers. Iltutmish rectified the situation by bringing the provincial governors under the central
authority and subjecting them to a certain financial control. But in general, the nobles con-
tinued to enjoy the benefits reaped from the iqttf s. Out of the revenues collected from the

281
ISBN 978-92-3-103467-1 Contents Agrarian conditions ...

territories assigned to him, the muqtac kept a portion for himself and his household, used
another portion towards maintaining his contingent of troops and sent the balance (jawadil)
to the central exchequer. The muqta0 s obligations included maintaining military contin-
gents and placing them at the sultan's service when needed. By the time of c Ala' al-Din
Khalji, the muqtacs had been made fully accountable to the central revenue department,
and in general, the nobles were no longer given iqtacs but cash salaries. Subsequently,
under the Tughluqids (see Part Two below), the system of iqtacs was revived, but some
restrictions were introduced. The nobles lived in great luxury and style. They comprised
three main grades: the Khans, who were paid 1 lac of tankas, the Maliks, who were paid
50-60,000 tankas, and the amirs, who received 30-40,000 tankas. At the beginning of
the thirteenth century, even soldiers (called iqtac-dars) were given iqtacs; but when it was
found that they converted the land into milk (private property), the practice was gradu-
ally discontinued. Payment to cuzama' and mashayikh was made in various forms: regular
stipends, assignments of 'dead land brought to life', assignments of cultivable land and
assignments on the jizya (poll tax on non-Muslims) of a particular locality. Land not given
in assignment and reserved for the state treasury was known as khalisa. Specific amounts
of land were assigned for the sultan's personal and household needs, but were not treated
as royal property.

Agrarian conditions in the fourteenth century

The evolution of agrarian conditions during the thirteenth century and the agrarian reforms
of c Ala' al-Din Khalji have already been noted. Ghiyath al-Din Tughluq (1320-5) reverted
to the 'produce-sharing' method which, as noted earlier, favoured the peasantry. He low-
ered the rate of state revenue demand and abolished several agrarian excesses. He miti-
gated the harshness of c Ala' al-Din's measures concerning the kuts, the muqaddams and
the muqtacs. His short reign probably brought considerable relief to the rural population.
The impact of Muhammad b. Tughluq's (1325-51) experiments with the agrarian econ-
omy, such as the sudden increase of the rate demanded in the Do' ab, the rotation of crops
and the granting of loans to the peasants, was lost in the plethora of revolts; in the agrarian
sector, as in other sectors, Muhammad b. Tughluq left only confusion and anarchy. It took
Firuz Shah Tughluq (1351-88) six years to survey the entire land and prepare new esti-
mates of revenue. He too decided to adhere to the 'produce-sharing' method. By making
a substantial addition to the water supply through canals and innumerable wells, he made
an enormous contribution to gardens and cultivable land and thus ensured a substantial
increase in the supply of cereals and fruit.

282
ISBN 978-92-3-103467-1 Contents Social and economic developments

Firuz Shah made extensive assignments to the nobility, officers, and men and institutions
of learning and piety; these inevitably diminished the khalisa lands, thereby weakening the
state financially. The impression of overall prosperity - in which the village peasants were
also beneficiaries - is due to the notable and sustained increase in production and to the
long period of general peace. At the same time, it was also a period of lax administration,
during which nobles and officers would misappropriate public funds, fail to pay dues and
thus become rich and powerful at the expense of the state.

The political structure of the state

Dynastic monarchy was a structure with which Indians had long been familiar. From 1210,
when the Delhi Sultanate was formally founded, the sultans steadily gathered more and
more powers; it can fairly be stated that a sultan was generally a more powerful ruler
than a Hindu ruler of earlier centuries. The process reached its climax under c Ala' al-Din
Khalji, who effectively controlled the empire and ran it as if it were a village. Muhammad
b. Tughluq, however, went too far and suffered a set-back.
The sultan was assisted by a body of ministers who managed their respective depart-
ments under royal orders. The most important departments were those of religion and jus-
tice, of the army, of the intelligence service and the imperial post, and of finance and
revenue, which was looked after by the most important minister, the vizier. For adminis-
trative purposes, the sultanate was divided into provinces, with the executive head of each
province serving as governor. His powers were considerable, yet limited by the central
government, especially in financial matters. When the government at Delhi was weak, the
governors, especially those of the distant provinces, tended to assume more powers and
run their provinces autonomously; some were tempted to declare independence. Depend-
ing upon the circumstances, a rebel governor might face the gallows or become the founder
of a new provincial dynasty.

Social and economic developments in the fourteenth

century: urbanization, crafts, etc.
The trend towards urbanization, which had begun in the thirteenth century, continued apace
during the following century. Both the state chronicles and the accounts of foreign visitors
such as the Moroccan c alim (scholar) and traveller lbn Battiita confirm this. Two interesting
pieces of evidence are the constantly increasing size of the congregational mosques and the
organization of regular transportation into the city of Delhi, with fixed charges from and

283
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

to various points; the latter indicate the growth in the size of the city. Ibn Battuta declared
that Delhi was the largest city not only in India, but in the entire Islamic East.
The increase in population and the growth of a large number of cities led to the devel-
opment and diversification of industries and crafts. Of particular importance were cotton
fabrics, silken stuffs, carpets, woollens, ironware, leatherware and sugar-making. Indian
hardware achieved great fame, producing damascened steel which had a worldwide repu-
tation. Many other industries and crafts are mentioned in the context of the royal workshops
or of the taxes imposed on the industries. The scale of diversification of food production
can be grasped from Ibn Fadl Allah's Masalik al-absar, with its mention of 21 varieties
of rice and 65 varieties of sweets. In trade and commerce, the most notable groups were
the karwaniyan (banjaras), who distributed large quantities of grain all over the land and
are continually mentioned in chronicles and in Amir Khusraw's historical mathnawfs. The
merchants, especially the famous Multani merchants, who were concerned with internal
as well as foreign trade, also played an important role. They organized the import of fine
cloth for c Ala' al-Din's Saray-i c Adl market. The sahas (bankers), the Multani money-
lenders and the sarrafs (money-changers) provided banking services which greatly facil-
itated commercial transactions in the country. The increased pace of production led to
certain technological advances. The introduction of the cotton-carder's bow and the spin-
ning wheel, for example, contributed to the expansion of the textile industry. The introduc-
tion of the true arch, dome and vault facilitated the construction of large buildings; Diya'
al-Din Baran! mentions that c Ala' al-Din Khalja employed as many as 70,000 craftsmen
for the construction of his buildings.

Part Two

THE DELHI SULTANATE, 1316-1526

(C. E. Bosworth)

The Tughluqids (1320-1412)

With the murder of Qutb al-Din Mubarak Shah in 1320, the line of the Khalji sultans of
Delhi came to an end, and his assassin, his Hindu convert slave Khusraw Khan Barwari,

284
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

ascended the throne as Sultan Nasir al-Din. But his reign was cut short by the rebellion
of Ghazi Malik Tughluq, governor of Dipalpur in Panjab, who had risen to prominence
under the Khaljis, utilizing resentment against the ascendancy of the Hindus in the state
under Khusraw Khan: in 1320 Nasir al-Din was defeated and killed by Ghazi Malik, who
ascended the throne as Ghiyath al-Din (1320-5). The line of sultans which he inaugurated
is conveniently referred to as the Tughluqids, although Tughluq was almost certainly a
personal name of Ghazi Malik rather than a Turkish ethnic or tribal name. 1
Ghiyath al-Din thus came to power posing as the saviour of the faith from Hindu threats
to subvert Islam, although Nasir al-Din's failure had stemmed from his personal incapacity
to rule rather than from outraged Islamic sentiment. Hence Diya' al-Din Baran! presents
Ghiyath al-Din as the paragon oflslamic rulers, 2 although the Sufi hagiographic tradition
is less enthusiastic because of the new ruler's differences with the Chishfi mystic Nizam
al-Din Awliya' .3 Ghiyath al-Din's main tasks were to restore internal order and to pull
together the empire after the financial chaos and the centrifugal administrative forces at
work during the previous reign. These he achieved by recovering land grants (iqtt{ s, or
jagfrs) which had been lavishly distributed by his predecessor, by campaigning against
the Hindu rulers of Orissa and Macbar (Madura) (this last province conquered in 1323)
and by securing the vassalage of the Muslim sultanate of Bengal in 1324. Thus on his
death in 1325, the sultanate had been once more consolidated and its frontiers extended
considerably beyond those of Khalji times. 4
Ghiyath al-Din's son, Muhammad b. Tughluq (1325-51), consummated this work of
consolidation and expansion during his long reign, and under him the Delhi Sultanate
reached its greatest extent; his reign marks a watershed in the history of the sultanate.
He is certainly one of the great figures of medieval Indo-Muslim history, yet Professor K.
A. Nizami has written of him:

His reign of twenty-six years is a fascinating but tragic story of schemes and projects correctly
conceived, badly executed and disastrously abandoned. His ingenious mind was as quick in
formulating new plans as it was slow in understanding the psychology of the people. He
could never establish that rapport and mutual understanding with his subjects, which was so
necessary for the implementation of his schemes. 5

Historians such as cisami and Baran! adopted hostile attitudes to him and stigmatized him
as an impractical visionary. Yet Muhammad was in fact a vigorous commander and man of

1
Habib and Nizami, 1970, p. 460.
2
Hardy, 1960, pp. 35-6.
3
Habib and Nizami, 1970, p. 482.
4
Ibid., pp. 460-83.
5
Ibid., p. 484.

285
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

action. 6 In 1327 he embarked on one of his most controversial and innovatory actions, the
founding of a second capital of the sultanate in Deogir, renamed Dawlatabad, in the north-
ern Deccan (near modern Awrangabad in Maharashtra province), in which many members
of the Muslim administrative and religious elites of Delhi were willy-nilly resettled. In
this way Muhammad parted company with the Khalji policy of exercising suzerainty over
the Deccan from outside, and, from this new military base, he apparently planned a more
activist policy within the Deccan. Whether this was his express intention or not, the policy
speedily proved a failure and the division of central authority within the sultanate has been
criticized by later historians as having had, in the longer term, an adverse effect on the
sultanate's unity and effectiveness. 7
Soon after Muhammad's accession, the Tughluqid army raided Peshawar and the moun-
tains beyond, but had to retire because of the lack of food and fodder there. It seems to
have been this raid which in c. 1329-30 provoked the last major invasion of India by the
Chaghatayids, whose territories in the North-West Frontier region and eastern Afghanistan
had just been threatened. Under their Tarmashirin Khan, the Mongol forces entered Panjab
and reached the Jumna. Peace was made, but Muhammad seems to have entertained the
grand design of attacking the Chaghatayids in 'Khurasan', a vague term in Indo-Muslim
usage of the times. Barani speaks of a campaign against the 'Qarachil mountains', which
has often been taken to refer to the Himalayan regions of Garwhal and Kumaon but which
might well refer to Kashmir, at that time considered to be within the Chaghayatid sphere of
influence; the venture was, at all events, unsuccessful. One side-effect of Muhammad's pol-
icy vis-a-vis the Chaghatayids was that his realm became a haven for many Turco-Mongol
chiefs and soldiers fleeing from Tarmashirin's strongly anti-Muslim measures within the
Khanate, and contingents of Turco-Mongols appeared in the Tughluqid army later in his
reign. 8
After a certain number of successes, however, a reaction set in and in the latter part
of his reign, Muhammad had to deal with no fewer than twenty-two rebellions in different
parts of the empire. These involved the permanent loss to the sultanate of several provinces.
Bengal and Macbar (Madura) regained their independence; Multan, Sind and Gujarat were
disaffected; above all, the new policy towards the Deccan clearly failed when cAla' al-
Din Hasan Bahman Shah constituted the Bahmanid sultanate there after 1347. Hence at

6
Hardy, 1960, pp. 36-9.
7
Habib and Nizami, 1970, pp. 506-15.
8
A. Ahmad, 1964, pp. 17-19; Habib and Nizami, 1970, pp. 498-500; Jackson, 1975.

286
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

Muhammad's death, the sultan of Delhi possessed no authority in central and southern
India beyond the Vindhya range. 9
The causes of this decline are various. Muhammad had clearly aroused discontent in the
state by his policy of opening the doors of the army and the administration to new sectors
of talent. In pursuit of this broadening of his power base, he did, as mentioned above,
encourage dissident Mongol amirs to come to his court. He further admitted converts from
Hinduism - as of course had his predecessors - and this was resented by the old Muslim
Turkish families and by the c ulama', both classes ever jealous of their own positions and
interests. The sultan's policy of attracting strangers to India and of honouring them for their
capabilities was approved by Ibn Battiita (mentioned above), who reached India in 1333
and the Delhi court in the following year (as an outsider himself, he benefited greatly from
it). 10 Muhammad was keen to establish links with the c Abbasid faineant caliphs now living
in Cairo under the tutelage of the Mamluks, receiving their emissaries and placing their
names on his coins, presumably in the hope of strengthening the aura of Islamic legitimacy
for his rule; but the caliphate was by this time such a pale and ineffectual shadow of its
former self that c Abbasid approval does not seem to have brought Muhammad any tangible
benefits in the eyes of his contemporaries. 11
Of more immediate damage to Muhammad's image as a divinely mandated ruler were,
first, his strained relations with the religious classes of India (although the accounts by
contemporary chroniclers of a decline in religious life at Delhi as a consequence of the
move to Dawlatabad are clearly much exaggerated) and, second, his general reputation
as a stem, even bloodthirsty ruler, whose anger and violence did not spare recalcitrant
religious scholars and Sufis - as Ibn Battuta notes in a fair-sized list of those executed by
the sultan. 12 But Muhammad's attempts to encourage agriculture, especially in the wake
of a disastrous famine in the Delhi- Do' ab region in 1335-6, to reform the coinage by
introducing a low-denomination copper and brass coinage (perhaps in response to heavy
drains of precious metal resulting from military campaigning and/or to some economic
crisis not made explicit in the sources) 13 and to establish a secure base for the Islamization
of the Deccan at Dawlatabad show him as a man of some vision who was trying to follow a

9
Habib and Nizami, 1970, pp. 523-37.
10
Habib and Nizami, 1970, pp. 562-5; Ibn Battuta, 1971 , p. 671; Nizami, 1985, pp. 115-20; Siddiqui,
1992, pp. 27-8.
11
Qureshi, 1958, pp. 33-6; A. Ahmad, 1964, pp. 8-9; Habib and Nizami, 1970, pp. 537-8; Ibn Battuta,
1971, pp. 679-82.
12
Ibn Battuta, 1971 , pp. 695-706; Nizami, 1985, pp. 200-1.
13
Brown, 1922; Wright, 1936; Habib and Nizami, 1970, pp. 515-19.

287
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

coherent policy but was held back by inadequate resources, refractory human material and
personal failings. 14
Muhammad's nephew Firuz Shah (1351-88) had a more pacific and conciliatory tem-
perament and his thirty-seven-year reign gave India a period of general relaxation and
peace after the storm and stress of Muhammad b. Tughluq's reign. This newfound tran-
quillity was signalled by the prohibition, on Firiiz Shah's accession, of what Barani calls
siyasat, i.e. the infliction of harsh punishments and torture which the severe and blood-
thirsty Muhammad had used with such abandon as instruments of state policy. 15
The new sultan was nevertheless by no means averse to military glory and success, and
aimed at restoring the control lost by Delhi over the provinces. Unfortunately, he lacked
military skill and the ruthlessness required of a great commander. His two invasions of
Bengal (in 1353-4 and 1359-61) gained virtually nothing. He attacked Hindu rulers in
Orissa and at Nagarkot-Kangra and led a long and costly campaign against the Samma
chiefs of Thatta on the Indus and lower Sind and against Gujarat (in 1365-7), asserting the
suzerainty of Delhi there; but the whole enterprise was later regretted by the sultan for the
losses in manpower and treasure involved. An invitation from discontented elements in the
Bahmanid sultanate to intervene in the Deccan was, on the advice of the sultan's veteran
vizier, the Khan-i Jahan Maqbiil, wisely refused and Firiiz Shah henceforth abstained from
military adventures. 16
In general, Firuz Shah showed himself more concerned with the arts of peace, and this
inevitably led to a decline in the organization and fighting qualities of the army during
the last twenty years or so of his reign. Much of the army's preparedness and military
effectiveness had rested on the periodic reviews (card) of the cavalry, their weapons and
their mounts by the official entitled the rawat-i card. The standards attained were recorded
in the registers of the dzwan-i card (military department) of the administration; it was on
the basis of performance on these occasions that salaries and allowances were issued. 17
The system had been rigorously upheld by such sultans as c Ala' al-Din Khalji, Qutb al-
Din Mubarak Shah and Muhammad b. Tughluq, when military efficiency had been the
criterion for financial rewards. Firiiz Shah, however, granted extensive hereditary iqtacs to
the army commanders rather than paying them in cash, a reversal of previous practice; and
since the troops now collected their salaries directly from the cultivators, the door was open
to extortion, oppression and corruption throughout the countryside, as the state could no
longer threaten to withhold salaries in the case of military unpreparedness or inadequate

14
See on his reign in general, Husain, 1938.
15
Hardy, 1960, pp. 37-8; Habib and Nizami, 1970, pp. 576-7.
16
Habib and Nizami, 1970, pp. 562-600.
17
Qureshi, 1958, pp. 136 et seq.

288
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

training. 18 The sultan, meanwhile, buttressed his personal authority by the acquisition of
a large body of personal slaves, the bandagan-i Ffruz-Shahf: their numbers stationed in
the capital and in the provinces were implausibly put by cAfifi at 180,000. 19 It is true that
the more deleterious effects of the new trends in military organization and payment were
delayed by the abilities of Firiiz Shah's ministers, who included men of high calibre such
as the two Khan-i Jahans, father and son, and cAyn-i Mahru.
The adverse effects of the new system took time fully to emerge. It was only after Firiiz
Shah's death in 1388 that it became apparent that the decay of a highly trained, centrally
paid, salaried army meant that assignees of lands often had inadequate military force with
which to collect the revenues from their iqtt{s in the face of rebellious provincial gover-
nors, recalcitrant Hindu chiefs, and others. For the same reason, the central administration
in Delhi could not collect its own share of the iqtac s, that proportion which was kept back
from the assignees for the expenses of running the state. Hence when the Turco-Mongol
conqueror Timur (Tamerlane) appeared in India from Central Asia a decade after Firiiz
Shah's death, the Delhi Sultanate's military and financial resources were totally inadequate
for opposing him.
Firuz Shah's relaxation of central control in several spheres of state activity, and his
amelioration of the harsh and oppressive policies of the preceding reign, were meritorious
measures, although one consequence of them seems to have been a spread of corruption
in the administration once fears of draconian punishment had disappeared. In religion, the
sultan held strictly orthodox Sunni views. He deferred to the cuzama'; he was the last Delhi
sultan to receive formal investiture from the puppet cAbbasid caliph in Cairo; he destroyed
newly erected Hindu temples; he persecuted the extremist Shfites and the Ismacilis; he
exacted the jizya, albeit at a low rate, from Brahmans, hitherto exempt; and he abolished
mukus (pl. of maks; non- Qur' anic taxes), although it is reasonable to assume that, as had
always happened on previous occasions when these were abolished, the state soon found
itself unable to do without the revenue and the old taxes and abuses crept back in. 20
The sultan's pacific policies may have brought some beneficial results for the masses
of the population, if only because of the decreased need to finance military campaigns.
Whether the price of provisions remained stable and affordable during his reign has been
disputed by modern historians; prices were certainly much higher than they had been in,
for example, cAla' al-Din Khaljl's time half a century or so before (see Part One above). It

18
Habib and Nizami, 1970, pp. 579-81.
1
9 Ibid., pp. 600-1, 619.
20
Qureshi, 1958, pp. 36-7, 129-30, 244-7; A. Ahmad, 1964, pp. 9-10; Habib and Nizami, 1970,
pp.578-9,609-12.

289
ISBN 978-92-3-103467-1 Contents The Tughluqids (1320-1412)

was as a builder of public buildings and endower of charitable institutions that the sultan
achieved particular fame. Around Delhi, he laid out many gardens and orchards, and within
the city he completed a Friday mosque and the Madrasa-yi Firuz-Shahi in 1352, as well as
a Sufi khanaqah-cum-madrasa (convent-cum-college) for the noted Sayyid Najm al-Din
Samarqandi. In 1359 Firuz Shah founded the city of Jawnpur, possibly named after his
kinsman Muhammad b. Tughluq's pre-accession title of Jawna Khan, and he further built
a new city in the Delhi district, named Firuzabad after himself; it did not, however, survive
the Timurid onslaught soon afterwards. 21
After Firuz Shah's death in 1388, the remaining twenty-five years of Tughluqid rule
were filled with a series of ephemeral sultans; none save one of the last, Nasir al-Din
Mahmud Shah II (1394-5; 1399-1412) ruled for more than two or three years. Thesul-
tanate was, in fact, in a state of disintegration, racked by disputes over the succession
and the allocation of power; thus in 1395 Mahmud Shah II was ruling in Delhi while
his rival, Nasir al-Din Nusrat Shah (like Mahmud Shah a grandson of Firuz Shah), held
power at Firuzabad. By this time, many of the muqtac s had achieved virtual independence.
The Bahmanid sultanate had flourished under its able second ruler, Muhammad Shah I
(1358-75), and his successors. The Hindu ruler of Vijayanagar in the south-eastern tip
of the Deccan had already succeeded in extinguishing the petty Muslim principality of
Macbar (Madura) soon after 1378. The governor of Malwa in central India, Hasan Dilawar
Khan, ceased to forward any tribute to Delhi after 1392. He sheltered the fugitive Tugh-
luqid sultan Mahmud Shah II when Timur invaded India in 1398, but in 1401 proclaimed
his independence, thus inaugurating the powerful sultanate of Malwa, based on its capital
Mandu, which was to endure for over a century until conquered by the sultans of Gujarat.
In Jawnpur, the eunuch commander of the sultanate, Malik Sarwar, who already held the
title of sultan al-skarq (Ruler of the East), was in 1394 sent to Jawnpur to quell disaf-
fected Hindus there; he extended his power over most of the Ganges valley east of Delhi,
including Bihar, as an independent ruler. The progeny of his adopted son and successor,
Malik Mubarak, founded the principality of the Sharqi sultans which was to last for some
eighty years until Sultan Bahliil Lodi reincorporated Jawnpur within the Delhi Sultanate.
In Gujarat, the sultanate commander Zafar Khan had been sent to restore order there, but
first his son Tatar Khan assumed power in 1403 and then Zafar Khan himself in 1407 -
at a time when the Tughluqid dynasty was largely impotent, the sultans not having minted
coins for six years - assumed independent authority as Sultan Muzaffar I. He reigned in

21
Habib and Nizami, 1970, pp. 585-9, 601-16.

290
ISBN 978-92-3-103467-1 Contents The Sayyids (1414-51)

Gujarat until his death in 1411, after which his descendants enjoyed power for almost two
centuries until the Mughal conquest of Akbar the Great in 1583. 22
The catalyst for all these losses and secessions from the empire, which reduced Tugh-
luqid control virtually to the Delhi region alone, so that it became a capital city without an
empire, was Timur's invasion of 1398-9, which culminated in the sack of Delhi and the
flight of the sultan. 23 When the Turco-Mongol armies at last withdrew, real power in Delhi
lay not so much in the hands of the restored Tughluqid Mahmud Shah II as in those of
his Afghan minister, Mallu Iqbal Khan. The former north-western provinces of the empire,
including Panjab and Multan, gave their allegiance to Timur and then to his successor
Shah Rukh. The governor of the western frontier region, Sayyid Khidr Khan, was in 1414
to seize power at Delhi and inaugurate the shortlived Sayyid line of rulers there.

The Sayyids ( 1414-51)

Mahmud Shah II died in 1412, and there was a two-year interlude during which power in
Delhi was held by a former Tughluqid commander, Dawlat Khan. Delhi was then captured
by Sayyid Khidr Khan, who had since the early 1390s governed the province of Multan
and had maintained himself there against Mallu Iqbal Khan when the latter was governor
of Lahore and Panjab; now, in 1405, Khidr Khan defeated and killed Mallu at Ajodhan.
As long as a legitimate Tughluqid, Mahmud Shah II, reigned in Delhi, Khidr Khan could
make no headway, but the sultan's death provided him with an opportunity.
Whether Khidr Khan really was a sayyid (descendant of the Prophet) is doubtful. The
reference to this status in the near-contemporary Tarfkh-i Mubarak Shahfof Yahya Sirhindi
is at best vague; in its favour, however, may be Timur's earlier appointment of Khidr Khan
as his governor in Delhi, suggesting that he was regarded as a sayyid, given Timur's special
regard for the descendants of the Prophet. The growth of the claim may have been encour-
aged by Khidr Khan's undoubtedly benevolent rule (1414-21) and his skill as a military
commander. 24
It is not easy to characterize the Sayyid dynasty, with its four members only, but early
fifteenth-century Muslim India clearly shows the transition from a strong centralized rule
by the early Delhi Sultanate dynasties to a more diffused system of government, with
strong tendencies to particularism and regionalism and a much reduced role of Delhi in
Indo-Muslim political and military affairs. The period of the Sayyids was full of mili-
tary campaigns, often against petty rebels and chieftains who had witheld taxation, but the
22
Husain, 1963 ; Habib and Nizami, 1970, pp. 620-9; E/2 , Qjawnpur; GuQjarat; Macbar; Malwa; Sharkis.
23
Roemer, 1986, pp. 69-70.
24
Habib and Nizami, 1970, pp. 635-6; E/2 , Sayyids.

291
ISBN 978-92-3-103467-1 Contents The Sayyids (1414-51)

Sayyid rulers showed little vision in extending their power beyond the vicinity of Delhi,
the upper and middle Do'ab, with a more tenuous authority over Panjab and Multan. Thus
the amount of revenue available to the state depended largely on the success or failure of
these punitive expeditions and holding operations, and also on the rulers' ability to control
a powerful and ambitious Turkish military nobility which had benefited from the power
vacuum at the centre under the last Tughluqids to increase its own influence. The Sayyids
were themselves conscious, it seems, of enjoying a lesser status and prestige than their
predecessors. Khidr Khan ruled as a Timurid vassal and Shah Rukh was recognized in the
khutba (Friday worship oration) and on the coinage of Delhi. Only after 1417, and with the
Timurid monarch's permission, did Khidr Khan add his own name to the coinage, previ-
ously having been content to restamp the coins ofFiruz Shah Tughluq and his successors. 25
Nor did he ever claim the exalted title of sultan but only that of rayat-i acza (Most Exalted
Standard-[bearer]).
Khidr Khan's seven-year reign was full of campaigns localized, however, in the northern
Indian plain and on its fringes: against Hindu Rajahs in Katahr (in the later Rohilkhand),
Gwalior, Itawa (in the Kanawj region) and Mewat (in Rajasthan); in repelling an attack on
Nagawr by the sultan of Gujarat, Ahmad I (1411-42), son of Zafar Khan Muzaffar I; and
against rebellious Turkish troops of the sultanate, the Turk-bachchas. 26
Khidr Khan's son and heir, Mubarak Shah (1421-34), was the ablest ruler of his line.
It is clear that he felt himself in a stronger position than his father from the fact that he
adopted the title of sultan, placed his own name in the khutba and issued coins. Even so,
he faced much the same problems, with additional challenges from the Khokars of Panjab
(in 1421-2 and 1428), and with further campaigns required against Katahr, Mewat and
Gwalior - on more than one occasion, against all of these places. The provincial Muslim
kings of India were now strong and ambitious enough to challenge Delhi, and Mubarak
Shah clashed with Alp Khan Hushang of Malwa (1405-35), who was menacing Gwalior,
in 1423; and with the Sharqi ruler of Jawnpur, Shams al-Din Ibrahim (1402-40), who
was threatening Bada'un and Itawa, in 1428. He was successful in repelling incursions of
Turco-Mongols from Kabul, instigated by Shah Rukh's governor there, Mascud Mirza, in
concert with their Khokar allies (in 1431 and 1433), and in relieving pressure on Lahore.
Like certain other previous rulers in Delhi, Mubarak Shah had the idea of founding a new
city on the banks of the Jumna, Mubarakabad, in 1433, but his plans were cut short when he
was assassinated in 1434 by partisans of his discontented vizier, the Hindu convert Sarwar
al-Mulk. 27
25
Wright, 1936.
26
Habib and Nizami, 1970, pp. 630-40.
27
Ibid., pp. 641-58.

292
ISBN 978-92-3-103467-1 Contents The Lodls ( 1451-1526)

The dead sultan's adopted son was raised to the throne in Delhi as Muhammad Shah
(1434-45), but was not given full allegiance by the great men of the state until Sarwar
al-Mulk, regarded as the instigator of Mubarak Shah's murder, had himself been killed.
Muhammad Shah ruled over a reduced, disordered realm with powerful rival princes on
its fringes. In 1440 the ruler of Malwa, Mahmud Shah I Khalji (1436-69), marched almost
to the gates of Delhi and was only defeated and repulsed with the aid of the governor of
Sirhind in Panjab, the Afghan Bahliil Lodi. Bahliil's power was increased by the grant
to him of Lahore and Dipalpur, and in the last two years of Muhammad Shah's reign he
rebelled and at one point even besieged Delhi.
When Muhammad Shah died, his son c Ala' al-Din c Alam Shah (1443-51) was recog-
nized in Delhi, but with even less power than his father; his inability to control any ter-
ritories beyond those within a 30-km radius of Delhi led to the witticism, (az Dihlf ta
Palamlpadshahf Shah cAzam) [c Alam Shah's rule extends only from Delhi to Palam] (Palam
being the site of the modern international airport of Delhi). In 1448 he decided to withdraw
to Bada'un, where he had previously been governor, abandoning Delhi. The military lead-
ers then took over power and in 1451 offered the throne to the most vigorous figure in
the now truncated sultanate, Bahliil Lodi. Since c Alam Shah was content with the par-
gana of Bada' un, Bahliil left him there in peace until the former Sayyid ruler died in 147 6,
when Bada'un was briefly annexed to Jawnpur by the Sharqi Husayn Shah (1459-79). The
Sayyid dynasty thus came to an end after a somewhat unremarkable thirty-seven years,
notable only as a further stage in the disintegration of the Delhi Sultanate. 28

The Lodls (1451-1526)

The Lodis sultans represent the first Afghan dynasty ruling in Delhi - the heart of lndo-
Muslim authority in northern India - since the time of the Ghurid sultans (originally
from Ghur in central Afghanistan) some three centuries previously. Even before the rise
to fame in the eastern Iranian lands of these Shansabani Maliks of Ghur (see above, Chap-
ter 8), Afghans had taken part in the Muslim raids and forays down to the Indian plains,
attracted by prospects of rich plunder, and the 'Afaghina' are mentioned among the troops
of Mahmud of Ghazna (see above, Chapter 5). The constituting of the Ghurid empire, tran-
sient though it was, brought in its wake large numbers of Afghan soldiers of fortune to
India, with especial areas of concentration in the middle Indus valley, Panjab and parts
of the Do' ab. They played a role in political and military affairs under the Khaljis and
Tughluqids second only to the predominant Turks in the forces of these ethnically Turkish

28
Habib and Nizami, 1970, pp. 659-63.

293
ISBN 978-92-3-103467-1 Contents The Lodls ( 1451-1526)

sultans. During the Timurid invasions of India, Afghans fought on both sides. Sultan Shah
Lodi aided the founder of the Sayyid line of Delhi sultans, Khidr Khan, against Mallu Iqbal
Khan and was rewarded with the governorship of Sirhind and its dependencies in Panjab,
plus the title of Islam Khan.
During the reign of Sayyid Mubarak Shah, the Lodi power base was extended, and after
Sultan Shah Lodi was killed, his younger son Bahliil inherited this. Bahliil managed to fight
off an attack by Sayyid Muhammad Shah's army and was diplomatic enough to conciliate
the ruler in Delhi and thereby to retain Sirhind and its adjuncts; and when in 1440 Sultan
Mahmud Shah I Khalji of Malwa attacked Delhi (see above), Bahliil provided a force of
8,000 Afghans and Turco-Mongols to ward him off, receiving in return the title Khan-i
Khanan. Nevertheless, Bahliil shortly afterwards revealed his own designs on Delhi and
the heart of the sultanate, fruitlessly besieging the city and assuming for himself the title of
sultan. The death of Sayyid Muhammad Shah and the accession of the even weaker c Ala'
al-Din c Alam Shah facilitated the fulfilment of his ambitions, as recorded above, so that
in 1451 Bahliil was able to ascend the throne in Delhi as Abu 'l Muzaffar Bahliil Shah
(1451-89), the first in a line of three Lodi sultans, the first and second of whom enjoyed
what were by contemporary standards long reigns. 29
Initially, Bahliil's position as sultan was by no means firm. The tribal and social tradi-
tions of his Afghan supporters favoured a more diffused allocation of powers in the state
rather than a centralized monarchy on the Khalji or early Tughluqid pattern, and Bahliil
had to take this into account; there are, in any case, no indications that Bahliil wished to
be an autocrat, withdrawn from his own folk, and he handled with care the body ofperma-
nently ambitious nobles and military commanders around the Delhi court. Moreover, there
was still a representative of the Sayyid family at Bada'un as a possible focus of discon-
tent, especially as the Sharqis in Jawnpur, whose territories marched with those of Delhi,
viewed themselves, because of matrimonial links with the Sayyids, as in large measure
heirs of the Sayyids in northern India. It thus behove Bahliil to proceed with caution. He
gained an access of prestige from defeating a Sharqi invasion at Narela outside Delhi in
1452 and hostilities with Mahmud Shah (1440-57) and Husayn Shah (1458-79) were to
fill the greater part of his reign. His success attracted considerable numbers of Afghan
troops from Roh (i.e. the North-West Frontier region and the adjacent mountain regions
of eastern Afghanistan), and with these, Bahliil inflicted a series of defeats on Husayn
Shah in 1479, culminating in the expulsion of the Sharqi ruler from Jawnpur to Bihar and
Bengal. Jawnpur was now reunited with the Delhi Sultanate after an independent existence
of nearly ninety years. This success allowed Bahliil to mount an invasion of Malwa and to

29
Habib and Nizami, 1970, pp. 664-72.

294
ISBN 978-92-3-103467-1 Contents The Lodls ( 1451-1526)

humble various Hindu princes at Gwalior and in the middle and lower Do' ab, such as the
ruler of Itawa.
When Bahllil died in 1489 he had reigned for thirty-eight years and had, by his crafty
diplomacy and military skill, placed Lodi authority on a firm footing. Occupied as he was
with frequent wars, he seems to have been content to let the adminstrative and revenue-
collecting system run on the same lines as those of later Tughluqid and Sayyid times; but
he did introduce, at a time when gold and silver for minting had already become scarce
under the Sayyids, a billon tanka, the bahliilf, which remained current until the time of
Akbar. 30
Before his death, Bahllil had allocated various parts of his realm as appanages for his
sons and other Afghan relatives and connections. Thus his son Barbak received Jawnpur;
Aczam Humayun received Lucknow and Kalpi; Khan-i Jahan received Bada'un; Nizam
Khan was given Panjab, Delhi and the upper and middle Do' ab, and so on. It was Nizam
Khan who finally emerged on his father's death as head of the Lodi family, taking the regnal
name of Sikandar and ruling for nearly thirty years (1489-1517). His most formidable task
was to make his rule acceptable to his numerous relatives, many of whom had their own
ambitions for the throne, and to the Afghan military classes at large. This he achieved by
campaigns which reduced his relatives to submission, by defeating the dispossessed Sharqi
of Jawnpur, Husayn, near Benares in 1494 and by humbling the latter's ally, Husayn Shah
(1494-1519), sultan of Bengal, these successes enabling him to take over the province of
Bihar. In the direction of central India, he twice successfully attacked Rajah Man Singh of
Gwalior (in 1501and1506). When Malwa was racked by succession disputes on the death
there in 1511 of Sultan Nasir al-Din Shah, Sikandar was tempted in 1513 to intervene on
behalf of a rival to Nasir al-Din's successor Mahmud Shah II (1511-31) and his Rajput
adviser Medini Ra 'i; but he achieved little beyond the capture of Chanderi (on the borders
of Malwa and Bundelkhand).
When Sikandar died in 1517, he left behind a prosperous kingdom with a considerable
degree of internal security. Being himself a poet in Persian, with the takhallus (nom de
plume) of Gulrukhi, he was also a patron of scholars and literary men. Among the most
tangible legacies of his reign was his re-foundation in 1504 or 1505 of the ancient town of
Agra and his decision to tum it into his capital city and military headquarters. 31
Sikandar's eldest son Ibrahim (1517-26) succeeded him, but he could only make firm
his power after a succession struggle with his brother J alal Khan of Kalpi. The latter
had originally been assigned the governorship of the former kingdom of Jawnpur in a

30
Wright, 1936; Habib and Nizami, 1970, pp. 673-88.
31
Habib and Nizami, 1970, pp. 689-701.

295
ISBN 978-92-3-103467-1 Contents The Lodls ( 1451-1526)

power-sharing agreement which Ibrahim speedily abrogated, driving Jalal Khan into
Gwalior and Malwa, eventually to be captured and killed. lbrahim's overbearing behaviour
soon aroused the fears and resentment of the military nobility, apprehensions strengthened
by such arbitrary acts as the sultan's arrest and imprisonment of the respected religious
leader Miyan Bhu' a. Ibrahim had already lost much prestige and military materiel in a dis-
astrous conflict with the Rajput potentate of Mewar, Rana Sanga. Various rebellions of the
Afghan commanders now erupted. That of Islam Khan, son of Aczam Humayun Sarwani,
was subdued, but a focus of opposition arose around Bahadur Khan Nuhani in Bihar, where
Bahadur Khan himself assumed the title of Sultan Muhammad and minted his own coins.
Further, the commanders of Panjab wrote to the Mughal Babur at Kabul in 1525, inviting
him to invade India. Babur occupied Lahore and came to face Ibrahim on the battlefield at
Panipat (the first of three important battles in Indian history there) in April 1526. Despite
an inferiority in numbers, Babur's effective use of his cavalry and of a protective laager
of linked carts carried the day and Ibrahim was killed on the field, the only Delhi sul-
tan thus to die. The Delhi Sultanate accordingly expired, with the ending of the Afghan
line of the Lodis (although Afghan domination in northern India was to be briefly revived
by the Suris). It was eventually to be replaced by the Mughal empire created, after some
vicissitudes, by Babur's son Humayun and his successors. 32
The Lodi sultanate had provided prosperity and stability until lbrahim's failure to work
with the Afghan nobility, who provided the military basis for the regime, brought about
military defeat and the dissolution of the whole sultanate. There had been a considerable
renaissance of learning during Sikandar's reign, including the translation into Persian of
Sanskrit works. He had encouraged the Persianization of the administration, which entailed
a wider learning of Persian by its Hindu officials, some of whom attained a high degree of
proficiency in that language. Sikandar had a particular interest in music. On the other hand,
he had a reputation for fierce Sunni orthodoxy and intolerance towards the Hindus, despite
the fact that his own mother was a Hindu. He had temples torn down, and erected in their
place mosques and other buildings or else he turned them into caravanserais. At Nagarkot,
he is said to have had idols broken up and the pieces used as butchers' weights. The chance
of securing more general support for the sultanate in northern India, outside the Muslim
ruling class, was thereby lost. It was to be the Mughal Akbar who, half a century or so later,
was to endeavour to establish a greater community of interest between rulers and ruled. 33

32
Habib and Nizami, 1970, pp. 702-9.
33 cAbdu '1-Halim, 1961; E/2 , LOdis.

296
ISBN 978-92-3-103467-1 Contents THE REGIONS OF SINO ...

15

THE REGIONS OF SINO, BALUCHISTAN,

MULTAN AND KASHMIR:
THE HISTORICAL, SOCIAL AND
ECONOMIC SETTING*
N. A. Baloch and A. Q. Rafiqi

Contents

THE RULERS OF SIND, BALUCHISTAN AND MULTAN (750-1500) . . . . . . . 298

The c Abbasid period and the Fatimid interlude (mid-eighth to the end of the tenth
century) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
The Period of the Ghaznavid and Ghurid Sultanates (eleventh and twelfth centuries) . 301
The era of the local independent states . . . . . . . . . . . . . . . . . . . . . . . . . 304
KASHMIR UNDER THE SULTANS OF THE SHAH MIR DYNASTY . . . . . . . 310

* See Map 4, 5 and 7, pp. 430-1, 432-3, 437.

297
ISBN 978-92-3-103467-1 Contents The c Abbas id period

Part One

THE RULERS OF SINO, BALUCHISTAN AND

MULTAN (750-1500)
(N. A. Baloch)

From 750 to 1500, three phases are discernible in the political history of these regions.
During the first phase, from the mid-eighth until the end of the tenth century, Sind, Baluchis-
tan and Multan - with the exception of the interlude of pro-Fatimid ascendency in Mul-
tan during the last quarter of the tenth century - all remained politically linked with the
c Abbasid caliphate of Baghdad. (Kashmir was ruled, from the eighth century onwards, by
the local, independent, originally non-Muslim dynasties, which had increasing political
contacts with the Muslim rulers of Sind and Khurasan.) During the second phase - the
eleventh and twelfth centuries - all these regions came within the sphere of influence of
the powers based in Ghazna and Ghur. During the third phase -from the thirteenth to the
early sixteenth century - they partly became dominions of the Sultanate of Delhi, which
was in itself an extension into the subcontinent of the Central Asian power base. Simul-
taneously, local sultanates independent of Delhi also emerged. Besides, the explosion of
Mongol power in Inner Asia had repercussions in these regions.

The cAbbasid period and the Fatimid interlude

(mid-eighth to the end of the tenth century)
When the c Abbasids supplanted the Umayyads in 750, Sind, the easternmost province
of the caliphate, included Makran and Turan and Qusdar (western and central Baluchis-
tan), Sind proper (including Kachh) and Multan (southwestern Panjab). Further expan-
sion and consolidation followed, beginning with the conquests of the caliph al-Mansur's
(7 54-7 5) energetic governor, Hisham b. c Amr al-Taghlibl. During his six-year tenure of
power (768-74), Hisham achieved several victories. Throughout the early c Abbasid period,
Sind continued to receive regular governors and the province enjoyed internal peace. Later,

298
ISBN 978-92-3-103467-1 Contents The c Abbas id period

the authority of the caliphate grew weaker, leading to the establishment of some five inde-
pendent Arab principalities in Mansura, Multan, Turan and Qusdar, Makran and Mashkey.

THE HABBART AMIRATE OF MANSURA

In the strife that erupted in Sind in 841-2, the local chief cumar b. cAbd al-c Aziz al-
Habbari emerged victorious. In 854 al-Mutawakkil appointed him as governor and he held
this position during the caliph's reign, but in the wake of the disorder following the death
of al-Mutawakkil in 861, cumar, though continuing to read the khutba (Friday worship
oration) in the name of the cAbbasid caliph, established himself as an independent ruler in
Mansura. Thus cumar became the founder of the Habbari dynasty. He, his son cAbd Allah
(who was ruling in 883) and his grandson cumar (who was in power at the time that al-
Mascudi visited Sind in 914-15) were effective rulers. Caravan routes from eastern Persia
led to Mansura and further on into the subcontinent. During the tenth century, the capital of
the Habbaris continued to flourish, as confirmed by the reports of al-Istakhri, Ibn Hawqal
and al-Maqdisi, who all visited it. In Mansura, the Habbari dynasty lasted until 1025, when
Sultan Mahmud of Ghazna overthrew their last ruler, Khafif.

THE BAN0 MUNABBIH AMIRATE OF MULTAN

The Banu Munabbih, who claimed to be of Qurayshite stock, had established themselves
in Multan at about the same time that the Habbaris had done in Mansura. At the opening
of the tenth century, Ibn Rusta was the first to report on the well-established rule of the
Banu Munabbih in Multan. According to a report recorded by al-Biruni, one Muhammad
b. al-Qasim b. Munabbih established himself in Multan after his victory there. He probably
belonged to the house of Jahm b. Sama al-Shami, who had allegedly settled in 'Kashmir'
(sic) as far back as 712-14 and whose descendants had reportedly continued to flourish
there. Muhammad b. al-Qasim attained prominence in the later ninth century and wrested
power, probably from a rebel deputy of the cAbbasids, around 861--4. He and his successors
gave allegiance to the cAbbasids and recited the khutba in the caliph's name. During their
long dynastic rule, which remained unchallenged for over a century, the Banu Munabbih
brought power, prestige and prosperity to Multan, as confirmed by the geographers who
visited it. After the middle of the tenth century, the power of the dynasty began to be
eroded due to 'Carmathian' (i.e. pro-Fatimid) propaganda, which was gaining momentum.
Thus it seems that the rule of the Banu Munabbih came to an end during the years 982-5.
In Turan- Qusdar and Makran, the two westernmost divisions of Sind, central cAbbasid
authority broke down earlier than in Mansura and Multan. cimran al-Barmaki was the last
cAbbasid governor who, under the caliph al-Muctasim (833-42), had led his forces from

299
ISBN 978-92-3-103467-1 Contents The c Abbas id period

Mansura to Qusdar and asserted his authority in that turbulent region. But cimran was killed
in Mansura in 842; the provincial administration collapsed and the two amirates, Qusdar
and Makran, and the third smaller one of Mashkey (see below), emerged as independent
entities.

THE RULERS OF TURAN

Turan (central Baluchistan), with its capital at Qusdar (Khuzdar), was governed by the
Habbaris of Mansura until the end of the ninth century. Then, early in the tenth century,
the chief Mughira b. Ahmad established himself independently in Turan, changing its cap-
ital from Qusdar to Kijkanan (Kalat), a fertile district producing grapes, pomegranates and
other winter fruit, but no dates. Mughira did not recognize the supremacy of the Habbaris
since he read the khutba 'only in the caliph's name'. Mughira was succeeded by his brother
Mucin b. Ahmad, who ruled during the time in which Ibn Hawqal wrote (mid-tenth cen-
tury). During his reign, the administration was far from satisfactory; his deputy, Abu '1-
Qasim al-Basri, had appropriated all powers - administrative, judicial and military - and
under these circumstances the radical, egalitarian sect of the Kharijites occupied the region
soon afterwards. These hard-pressed sectarians had sought refuge in the far-away fringes of
the caliphate ever since al-Muhallab b. Abi Sufra had expelled them from Iraq and south-
ern Iran. From the ninth century onwards, they succeeded in establishing themselves in the
regions between south-eastern Iran and Sind.
The Kharijites occupied Qusdar in about 971, set up their own principality and ruled
independently, without recognizing the cAbbasid caliph. Writing in 982, the anonymous
author of the Hudud az-cazam [The Limits of the World] observed that the residence of
the 'king of Turan' was in Kijkanan (Kalat). The fact that soon afterwards, these 'kings'
changed their capital from Kalat to Qusdar is confirmed by al-Maqdisi. That these just
sovereigns were Kharijites who were then being called 'caliphs' is borne out by a con-
temporary report recorded by the qadf Abu cAli al-Tanukhi (d. 994), according to which a
Kharijite 'caliph' was ruling the country from Qusdar, which was at that time (before 994)
a stronghold of the Kharijites.

THE BAN0 MAcDAN DYNASTY OF MAKRAN

The Kharijites apparently became significant in Makran towards the end of the ninth cen-
tury so that al-Mascudi observed, 'Makran is the land of the lawless Kharijites.' Soon
afterwards, in about 951, Makran fell into their hands and the ruler clsa b. Macdan was
called by the Indian title Maharaj by the people. Coastal Makran had trade links with the

300
ISBN 978-92-3-103467-1 Contents Ghaznavid and Ghurid Sultanates Period

Indian towns of the littoral and the fact that Maharaj meant 'Supreme Sovereign' led the
Kharijite ruler to assert his independence by taking this exalted Indian title. Obviously, his
being recognized as 'Great King' contravened the cAbbasid caliph's position as sovereign.
clsa thus became a notable member of the independent Banu Macdan dynasty that ruled
Makran from Kiz/Kej (Kech) for at least a century and a half, from Saffarid to Ghaznavid
times.

MASH KEY

Situated between Kirman and Makran, the petty principality of Mashkey existed in the
middle of the tenth century. Its independent ruler, Mudar b. Raja', read the khutba only in
the name of the cAbbasid caliph. Later, by about 985-6, when al-Maqdisi was writing, the
principality seems to have been annexed by the ruler of Makran and to have become a part
of the administrative district of Panjgur.

The Period of the Ghaznavid and Ghurid Sultanates

(eleventh and twelfth centuries)
Under pressure from the strongly orthodox Sunni new powers of the Ghaznavids and then
the Ghurids in what is now Afghanistan (see above, Chapters 5 and 8), neither the Khari-
jite rulers of Makran and Qusdar (who recognized no sovereign but God) nor the Multan
rulers (who came to recognize the Fatimids of Egypt) were able to continue their sectarian
independence much longer; they compromised by submitting when vanquished, but then
reasserted their independence when left to themselves. In 971, when the Buyid cAdud al-
Dawla's military power prevailed in Tiz and Makran, the Macdanid ruler accepted Buyid
suzerainty. After 977-8, however, with the decline in power of the Buyid dynasty and
when Sebtiktegin had attacked and annexed Qusdar, Macdan transferred his allegiance to
the Ghaznavids, first to Sebtiktegin and then to Sultan Mahmud. However, Macdan soon
involved himself in the politics of Central Asia. The Karakhanid conqueror of Bukhara, the
Ilig Nasr, entered into a secret pact with the ruler of Qusdar, stipulating that Macdan would
rise in rebellion against Ghazna when the Karakhanid invaded Khurasan. This accordingly
happened in 1011, but Sultan Mahmud marched against him, laid siege to the town of
Qusdar and seized the ruler, who now paid tribute and delivered fifteen elephants and a
substantial indemnity in cash. In return, the sultan allowed him to retain his principality as
a vassal of Ghazna.
On Macdun's death in 1025, his younger son clsa usurped power, forcing the elder
son Abu '1-c Askar Husayn to flee to Sistan. In 1031 Sultan Mascud sent a powerful army

301
ISBN 978-92-3-103467-1 Contents Ghaznavid and Ghurid Sultanates Period

against clsa, who was killed. Abu '1-c Askar Husayn, who now succeeded him, eschewed
Kharijism, extended his power and read the khutba in Masciid's name. This prince was
a man of learning, well-versed in medicine, and wrote a treatise on left-side hemiplegia
(i.e. paralysis). He ruled Makran successfully and for a long time, until a date beyond
1058(?). After his death, one might assume that the Kharijite faction would have reasserted
its power, but subsequently the Ghurid sultan Ghiyath al-Din Muhammad (1173-1203)
attacked and annexed Makran, putting an end to the local dynasty.

MULTAN: THE FATIMID INTERLUDE

Ismacili propaganda was introduced into Sind and Hind from Yemen in 883. In Sind it
did not become effective while the Habbarls were ruling there. In Multan, the Fatimid
dtFfs (propagandists) sent by al-Mucizz (952-75) succeeded by the middle of the tenth
century. In 965 al-Mucizz wrote an encouraging letter and sent seven mission flags to the
dt{f J alam, who subsequently gradually subverted the power of the Banii Munabbih. At the
time when Ibn Hawqal was writing, power was still in the hands of the ' Qurayshite ruler of
the Banii Sama' (i.e. the Banii Munabbih). According to the report in the Hudud az_cazam,
by about 982 the Qurayshite ruler of the Banii Sama dynasty was still ruling in Multan, but
was reciting the khutba in the name of the Fatimid caliph. By then the pro-Fatimid forces
under Jalam had presumably come to dominate, compelling the ruler to change allegiance
from the cAbbasids to the Fatimids. Subsequently, during 982-5, Jalam attacked, defeated
and killed the Banu Sama ruler. This is to be inferred from the following statement of al-
MaqdisI: 'They read the khutba in the name of the Fatimid and do not do anything except
by his order.' He makes no mention of the Qurayshite/Banii Sama/Banii Munabbih ruler in
Multan, and also confirms that the famous Multan idol in the temple of the sun god was
still there.
Jalam's next target was this temple of Aditya. In order to gain sufficient power to fight
back in case the destruction of the idol brought avenging forces from Kanawj and other
Hindu states against him, he seems to have proceeded slowly. It was some time after 985
that Jalam destroyed the idol. According to the report preserved by al-Biriini:

Jalam broke the idol into pieces, killed its priests, converted the temple mansion, which stood
on an elevated platform, into a new Jamie Mosque, and ordered the old Jamie Mosque to be
shut down, from hatred against anything that had been built under the Umayyads. 1

It is not known how long Jalam governed Multan, but there being no further mention of him
in any record, one can assume that he soon died or was eliminated by rival dt{zs. Shaykh

1
Al-BirunI, 1910, Vol. 1, p. 116.

302
ISBN 978-92-3-103467-1 Contents Ghaznavid and Ghurid Sultanates Period

Hamid, who was possibly heading a less intransigent faction, rose to power and won the
confidence of Sebtiktegin, who left him to rule Multan. But his grandson(?) Dawud b. Nasr
b. Hamid later aligned himself with Anandpal, the Hindushahi ruler of Wayhind, against
Sultan Mahmud of Ghazna, who then stormed Multan in 1006. Dawud escaped, with his
treasure, to an island in the Indus. The city elders in Multan sued for peace on payment
of the tribute, which was granted. However, the Carmathians (pro-Fatimid elements) were
not spared and their mosque was razed to the ground. The sultan appointed Sukhpal, 'the
nephew of the Shah' (of Kabul), as his governor and he himself hurried back to Khurasan to
repel the Ilig Khan's invasion. Thereupon, Dawud returned to Multan and wrested the fort
from Sukhpal. Early in 1011-12, as disturbances flared up again, Sultan Mahmud attacked
Multan, captured Dawud, imprisoned him and annexed Multan to the Ghaznavid sultanate.
Sultan Mahmud's two expeditions had broken the power of the pro-Fatimid elements
in Multan, though some remained underground and continued to foment the occasional
rebellion. Back in Egypt, with the disappearance of al-Hakim in 1017, the power of the
Fatimids was shaken and consequently the strength of their agents in distant Multan also
declined. Al-Hakim's partisans of the Druze faction, founded by Hamza b. cAll in 1017,
made some attempts to reinvigorate pro-Fatimid elements in Multan through Baha' al-Din
al-Muqtana. On Sultan Mahmud's death in 1030, al-Muqtana found it opportune to revive
contacts, and in 1034, addressed a letter to the influential local chief lbn Sumar (i.e. 'Raj pal
son of Sumar', of the house of Abu '1-Futuh or Abu '1-Fat'h Dawud (see above, Chapter 5).
But lbn Sumar's faction had already dissociated itself from pro-Fatimid elements and had
gained the confidence of the new Ghaznavid sultan, Mascud, so that the Younger Dawud
(al-Asghar) was granted a pardon and freed from prison on lbn Sumar's recommendation.
In his letter, al-Muqtana praised lbn Sumar and reminded him of the fidelity of his elders,
Dawud al-Akbar and others, and, warning him not to be misled, urged him to rise and
play an active role. lbn Sumar's response is not recorded, but the elements led by Dawud
the Younger remained active underground and rose in open rebellion when they found the
local administration weak. Thus on the death of Sultan Mascud in 1041, they succeeded
for a while in capturing the fort of Multan, but fled before the forces dispatched by the
new Sultan Mawdud. The people of Multan surrendered the fort and agreed to perform the
khutba in the names of the cAbbasid al-Qadir and of the Ghaznavid Mawdud.
Subsequently, during the period of the Ghurid sultanate, Multan remained peaceful
except for one rebellion attributed to the Carmathians, against whom Sultan Mucizz
al-Din Muhammad took action in 1175 and delivered Multan from their hands.

303
ISBN 978-92-3-103467-1 Contents The era of the local independent states

The era of the local independent states (thirteenth to

early sixteenth century)
THE SULTANATE OF MAKRAN

The sultanate of Makran, with Kej as its capital, emerged out of the Ghaznavid/Ghurid
dependencies of Makran and Qusdar by the turn of the twelfth century and included both
the former amirates. In Jiizjanl's Tabaqat-i Nasirf, Taj al-Din Abu Makarim of Makran
is referred to as a Malik of the two Ghurid sultans Ghiyath al-Din and Mucizz al-Din.
The foundation for the emergence of the Makran sultanate was laid in the twelfth cen-
tury by Malik Hasan, who was succeeded by his son Abu '1-Makarim Khusraw Shah. The
latter's own two sons Taj al-Din and Nusrat al-Din, after the death of Mucizz al-Din in
1206 assumed the title of sultan and ruled jointly for a long period. With the integration
of Makran and Qusdar, a strong power emerged with maritime and commercial trade links
and the pastoral resources from the vast hinterland. The religious base of the society was
apparently now orthodox Sunni. No one drank wine, but both the clds and the Nawruz
festivals were celebrated with enthusiasm. Both the Persian and the 'Makrani' (?Baluchi)
languages were used. With the Baluch as the backbone of their military power, the sul-
tans of Makran were able to inflict a crushing defeat on the invading Oghuz. This victory
added to the power and prestige of the Makran rulers, and Makran was soon compared to
Khurasan. The rulers actively promoted maritime commerce, and merchants and mariners
from Makran reached the east Asian shores. According to one tradition, the main port
became popularly known as 'Jawadar' (Gwadar), i.e. gateway to Java. Among others, the
learned Abu Is'haq of Makran settled at Pasai in Sumatra, where he became known as Abu
Is'haq al-Makrani al-Fasi (i.e. of Makran and then of Pasai). Thus Makran, like Sind, con-
tributed to the early commercial and cultural contacts between the region and South-East
Asia.
The references in the court poet Siraji 's panegyrics to the four sons of Sultan Taj al-
Din and the two sons of Sultan Nusrat al-Din indicate that the line of successors probably
continued; there is nothing to show that Makran was under any other rulers up to the
fourteenth century.

MULTAN UNDER NASIR AL-DiN QABACHA (1206-28)

After its annexation to the Ghurid sultanate, Multan became an administrative province
of the succeeding Sultanate of Delhi. On Sultan Qutb al-Din's death in 1210, Nasir al-Din
Qabacha, the governor ofMultan, became independent. He ruled successfully, extended his
power and consolidated the kingdom; and he succeeded in blocking the Mongol inroads

304
ISBN 978-92-3-103467-1 Contents The era of the local independent states

into Multan which commenced with their pursuit of Sultan Jalal al-Din Khwarazm Shah
in 1222. The Khwarazm Shah wanted a foothold in Multan, but Qabacha stood firm until
the sultan left for Sind on his way to Iraq. In 1222 Qabacha faced the Mongols, who
besieged the Multan citadel for forty days but failed to occupy it, and then withdrew. Dur-
ing Qabacha's reign (1206-28), education developed and colleges were founded; his court
became a rendezvous for the learned, and some of the earliest Persian works in the subcon-
tinent, such as c Awfi's literary anthology Lubab al-albab and c Ali: Kufi's Fat'h-nama on
Sind history (translated from Arabic) were produced. In 1228 Sultan Iltutmish of Delhi
attacked Qabacha, and on the latter's death in that same year, Multan was once again
annexed to Delhi.

THE LANGAH SULTANATE OF MULTAN (1437-1525)

With Ilutmish's victory, Multan became a province of the Delhi Sultanate and remained so
for the next two centuries. Timur's sack of Delhi in 1398 shook the sultanate and led to the
collapse of the central administration and the province of Multan passed into the hands of
the Langahs and the whole of Sind was possessed by the sultans of Sind.
There is much confusion in the sources about the identity of the Langahs and the begin-
ning of their rule in Multan. According to the historian c Abd al-Haqq, the author of the
Tarfkh-i Haqqf (written in 1592-3), with the decline of the power of the sultans of Delhi,
Budhan Khan of Sind, the chief of the Baluch tribe of the Langah, assembled his force at
Uchch and invaded Multan. He expelled the khan-i khanan, occupied the fort, took the title
of Mahmud Shah in 1437 and became the first ruler of the independent state of Multan.
He ruled for sixteen years and laid the foundations of the Langah sultanate. His son Sultan
Qutb al-Din succeeded him and further consolidated its power. On his death in 1469, his
son Shah Husayn ascended the throne and became the most illustrious ruler of the line,
bringing peace and prosperity to the country during the thirty years of his reign (1469-98).
Shah Husayn increased his military power by inviting and settling in his territory a con-
siderable body of Baluch. He wrested the principality of Shorkot from Ghazi Khan and
extended his control up into Chiniot. His prestige rose when he repelled the invasion of
Multan by the Delhi forces under Barbak Shah and Tatar Khan. However, his grandson
Mahmud was killed when the ruler of Sind, Shah Hasan Arghun, invaded Multan in 1525:
having lasted for almost ninety years, the rule of the Langahs came to an end.

S0MARA RULE IN SINO (c. 1050-1360)

The Sumaras, who had long been settled in south-western Sind had, through their frater-
nization and alliance with the Banu Tamim Arabs in Sind, gained political influence under

305
ISBN 978-92-3-103467-1 Contents The era of the local independent states

the Habbari rulers of Mansura. With the fall of the last Habbari ruler Khafif in 1025, Sind
came under Ghaznavid rule, though effective authority was not maintained. During the
reign of Sultan c Abd al-Rashid (1049-52), the Sumaras met together in their stronghold of
Thari (in the present Badin district) and declared their independence. They chose a leader
called Sumara as their first ruler.
The exact chronology of the Sumara rulers is obscure. According to later histories and
traditions, there were between nine and twenty-one rulers, but no unanimity exists concern-
ing their regnal periods. As verified by external sources, the earliest Sumara ruler whose
name figures in history was Sinan al-Din Chanesar, who was ruling Daybul (in southern
Sind) in 1224 when Jalal al-Din Khwarazm Shah passed through Sind. The last Sumara
ruler, Hamir, son of Doda, was killed by the Sammas some time before 1365 (see below).
The Sumara amirate of Sind lasted for more than three centuries (c. 1050-1360). For
most of this period the Sumaras held southern Sind and the territories east of the Indus,
which extended northwards and halfway to Multan, and included the western part of
Bikaner and Cutch in the south. At the peak of their power, during the reign of Hamir
I, their rule is said to have extended up to Marwar in the south-east and to the boundaries
of Gujarat in the south. The later Sumara rulers remained under constant pressure from
the sultans of Delhi, who controlled Sind's northern provinces of Bakhar and Siwistan
and, on occasion, intervened in the Sumaras' internecine disputes. In general, Sind under
the Sumaras remained independent and the people united in resisting outside interference
even when their rulers were at odds with each other. This is confirmed by the tradition
which is epitomized in the epic of Dodo Chanesar. Composed in different versions and
narrated by professional minstrels over the centuries, it may be counted among the world's
most famous epics. The Sumara period was one to which Sind tradition traces the origin of
some of the great romances and stories, which became the pillars on which the edifice of
classical Sindhi poetry and literature rests.
The fall of the Sumaras was mainly due to their dwindling economic base. Their pros-
perity had depended on the waters of the Puran channel, which was then the main course of
the Indus. As the Indus began to flow along a more westerly course, the volume of water in
the Puran diminished, adversely affecting the agricultural prosperity of the Sumara lands.
On the other hand, the changed course brought prosperity and power to the Sammas who
supplanted them.

THE SAMMA SULTANATE OF SINO (1350-1520)

The sultanate was founded by the Samma chief Unnar, an erstwhile functionary of the
Delhi Sultanate. He first rose in revolt in Shewan in 1333-4, a few days before lbn Battuta's

306
ISBN 978-92-3-103467-1 Contents The era of the local independent states

arrival. Later, he assumed the title of Sultan Firuz al-Din and became the undisputed ruler
of Thatta after Sultan Muhammad b. Tughluq of Delhi died while attacking Thatta in 1351,
so that the imperial army returned to Delhi. To preserve their independence, the Samma
rulers now had to contend with pressure from both Delhi and Central Asia.
Unnar was jointly succeeded by his son Sadr al-Din Shah Banbhnia and his brotherc Ala
' al-Din Shah Jiina. In order to weaken the authority of Delhi, which was exercised through
the governor in Multan, Banbhnia aligned himself with the Mongols who were attacking
Multan from the north. He also remained on the offensive internally against the Sumaras
who ruled eastern Sind (i.e. the territory to the east of the Indus). He finally attacked and
killed Hamir, son of Doda and last Siimara ruler, who was supported by the governor of
Multan on orders from Delhi. In 1365 Sultan Rruz Shah of Delhi marched against Thatta,
but a political settlement was reached when Banbhnia surrendered in 1366. The sultan
agreed that the Sammas could rule Sind as his vassals, but he held Banbhnia and later
his son Tamachi as hostages in Delhi. On Rriiz Shah's death in 1388, however, Delhi lost
control over the Sind province and an independent sultanate of Sind became a reality.
Beginning with Unnar, fifteen Samma 'community chiefs' (jams) ruled as sultans of
Sind from Thatta. 2 Tamachi, who returned from Delhi in 1388 (Banbhnia having died on
the way), ruled with the title of Sultan Rukn al-Din Shah. The next illustrious ruler was
Jam Tughluq Jiina (1428-53), who, in order to counterbalance Delhi, entered into matri-
monial relations with the neighbouring sultans of Gujarat. The alliance between the two
littoral sultanates gave an impetus to maritime trade. Numerous allusions in the classical
Sindhi poetry of the seventeenth and eighteenth centuries show that Sindhi merchants and
mariners frequented Java and penetrated beyond Perlak, the capital of the Samundara state
in eastern Sumatra. The last illustrious ruler of the line was Sultan Jam Nizam al-Din Shah.
During his long reign (1462-1508), education spread and commerce and agriculture pro-
gressed; and under the commander-in-chief Darya Khan, a scion of the Lashari (Baluch)
community, who enabled the bulk of the Baluch soldiery to become the backbone of the
army, the sultanate became militarily strong.
The political events in Central Asia during the latter half of Sultan Jam Nizam al-Din's
reign had an impact on the affairs of Sind. In the wake of the Thirty Years' Rind- Lasharl
War, the Sind army had occupied Sibi while the Arghiins from Kandahar had reached
Shal (Quetta) and had made incursions through the Bolan pass. The battle of Jaliigir (in
the Bolan pass), during which Muhammad Beg, brother of Shah Beg Arghun, was killed,
proved decisive and the Arghuns never again attacked Sind during the lifetime of Sultan
Jam Nizam al-Din. After his death in 1508, however, the situation changed. Hard-pressed

2
See Baloch, 1954.

307
ISBN 978-92-3-103467-1 Contents The era of the local independent states

by Babur in Kandahar, Shah Beg Arghun decided to attack Sind. After successful initial
raids and the occupation of Sibi, Bakhar and Sehwan, he conquered Thatta in 1521. With
Arghiin's victory, the sovereignty of the Sammas in Sind comes to an end. The ruler Jam
Firuz surrendered, but Arghun allowed him to rule southern Sind from Thatta. In 1528,
however, Jam Firiiz fled to Gujarat and sought refuge with Sultan Bahadur Shah.

THE BALUCH PEOPLE, THEIR MIGRATIONS

AND THEIR PRINCIPALITIES (750-1500)

There are no written records concerning the origin of the Baluch people or the chronology
of their migrations. Indirect evidence comes partly from the later histories but mainly from
Baluch ethnography and their oral tradition, particularly as embodied in their classical
poetry, although this imaginative material must obviously be used with caution.
The Baluch are not mentioned in Islamic geographic and historical sources until the
tenth century, and then shortly afterwards, in Firdawsl's Shah-nama; they are usually linked
with the Kufichis or Qufs as predatory peoples, apparently still pagan, living in the moun-
tains of south-eastern Persia, from which they preyed on Muslim caravans. The Buyid and
Ilyasid rulers of Fars and Kirman led punitive expeditions against them, and from then
onwards they probably gradually became Islamized. In the eleventh century, they must
have moved eastwards into what is now Baluchistan, doubtless after the stronger power of
the Seljuqs took over Kirman and reduced the opportunities for raiding and banditry, i.e.
after 1040. Since the central highlands of Baluchistan were by the eleventh century occu-
pied by the non-Indo-Aryan, Dravidian Brahuis, the Baluch tended to bypass this region
and make for the Indus valley, towards Sind, Multan and Panjab. Linguistic evidence shows
that Baluch was originally a northern dialect of Iranian, placing the homeland of the Baluch
people somewhere south of the Caspian Sea; their migrations into south-eastern Persia may
have been due to pressure from the warfare of the later Sasanians with the Hephthalites
which racked eastern Persia, although this is wholly undocumented.
The last phase of substantial Baluch migrations took place at the turn of the twelfth
century and continued into the wake of the upheavals caused by the Turkish-Mongol inva-
sions from Central Asia. According to one tradition, forty-four biilaks (clans) moved. In
effect, this was a mass exodus, from Sistan and Kirman to their main concentration in
Makran. On the one hand, the Baluch strength there became the backbone of the sultanate
of Makran, and on the other, the increase of the Baluch population in Makran led to an
extensive migration south-eastwards. According to tradition, five main sub-stocks from the
progeny of Jalal Khan, namely Rind, Lasharl, Kora'i, Hoat and Jatoi, became identifiable at

308
ISBN 978-92-3-103467-1 Contents The era of the local independent states

this stage, and from their power base of the sultanate of Mak.ran they sought new pastures
and new horizons.

THE PRINCIPALITY OF KALMAT

The Hoat, who were the first to leave, followed a twofold movement; those who went south-
wards, along the coastal belt, established their principality in the central littoral region, with
Kalmat as their capital. They actively participated in sea trade, and Kalmat became a pros-
perous state and served as a supporting base for the onward advance of the later Kalmati
group into the Habb and Indus valleys.

PRINCIPALITIES OF THE DERAJAT

The other Hoat sub-stocks of the Doda'is and the Chandias, as also the Kora'is and the
Jatois, migrated north-eastwards. Avoiding the colder highlands of Kalat, they descended
from the Mullah pass into the plains of Kachchi-Gandava and from Harbab and other passes
into the Indus valley. The Doda'is, migrating further north, eventually reached and occu-
pied the eastern slopes of the Sulayman mountains. By the end of the fifteenth century,
under their chief Suhrab Khan Doda'i, they were powerful enough to form the core of the
military power of the Langah sultanate of Multan (see above). They also vigorously devel-
oped the settled areas and founded the flourishing market towns of Dera Ghazi Khan ( 1494)
and Dera Ismacil Khan, the capitals of their two principalities, which the Hoat-Doda'i clan
ruled for some two centuries.

THE PRINCIPALITY OF KALAT

Leaving Mak.ran a century after the Hoat, the clans of the Rind-Lashari confederacy reached
central Baluchistan by the middle of the fifteenth century. Kalat was already a Baluch prin-
cipality, having been conquered by the early migrating Baluch tribes. It was then ruled by
Mir cumar, son of Miro of the Mirwarri dynasty of the Brahuis. When Mir cumar blocked
the Rind-Lashari advance, he was killed in the ensuing battle and Kalat was thus occu-
pied and ruled by the Rind-Lasharl confederacy. Because of its cold climate and meagre
resources, inadequate to sustain the bulk of its people, the confederacy does not seem to
have stayed in Kalat for long. The Rind and allied clans descended into the plains of Sibi
through the Bolan pass, while the Lasharls and their allied clans, passing through the Mul-
lah pass, spread into the plains of the Kachhi-Gandava country. There the Rinds, led by Mir
Chakar, and the Lasharls, led by Mir Gwahram, quarrelled in a dispute concerning horse
racing. This resulted in the long-drawn-out battles of the Thirty Years' War, which became

309
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

the main theme of classical Baluch poetry. The Rinds sought help from the Timurids of
Herat, and Mir Chakar or his emissary is said to have visited the court of Sultan Husayn
Bayqara (1469-1506). The Lasharis received support from the Samma ruler of Sind, Jam
Nizam al-Din (1462-1508), whose commander-in-chiefDarya Khan, alias Mubarak Khan,
was a scion of the Lashari family. The Thirty Years' War sapped the energies of the once
powerful Rind-Lashari confederacy. The Arghiins, who ruled Kandahar on behalf of Herat
and were supporting the Rinds, found it opportune to extend their power into the plains of
Sibi and eventually into Sind. Under pressure, the Rinds and the Lasharis disengaged, the
Lasharis going to Thatta and thence to Gujarat and Mir Chakar leading his people to the
Multan region.

Part Two

KASHMIR UNDER THE SUL TANS OF THE SHAH

MIR DYNASTY (1339-1561)
(A. Q. Rafiqt)

It is probable that before the first Muslim sultanate - known as the Shah Mir dynasty
- was established in Kashmir, Muslims had already settled the area, but the process only
accelerated after the establishment of the dynasty in 1339. The Muslim invaders, first Arabs
and then Turks, had invaded Kashmir on many occasions, but failed to conquer it. In 713,
when the Arab general, Muhammad b. al-Qasim, occupied Multan, he was said to have
marched against 'the frontiers of Kashmir, called Panj Nahiyat', but any putative threat to
Kashmir was removed when Muhammad was recalled by the caliph al-Walid I (705-15)
to his court. Later, some time after 7 57, Hisham b. c Amr al-Taghlibi, the Arab governor of
Sind, in vain attempted to conquer the valley of Kashmir.
Although the mountains proved barriers to would-be conquerors, they did not prevent
adventurers and refugees from entering Kashmir. c Ali b. Hamid al-Kufi, for example, states
in the much later source of the Chach-nama that Muhammad Alafi, an Arab mercenary who
had served Dahir (d. 712), the ruler of Sind, sought refuge in Kashmir. The ruler of Kash-
mir, Chandrapida, received him well and bestowed on him the territory of Shakalbar. After
Alafi's death, his estate was inherited by one Jahm, who, according to al-Kufi, built many

310
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

mosques there. This account, if true, would imply that there were a number of Muslims
already in Kashmir by that time.
We do not, however, find concrete contemporary information regarding Muslim influ-
ence until the early eleventh century. Sultan Mahmud of Ghazna invaded Kashmir on two
occasions, in 1014 and 1016, but his efforts to penetrate the valley were defeated by the
strong fortresses of Loharkot and a timely snowfall. According to the Ghaznavid historian,
Abu '1-Fadl Bayhaqi, however, Mahmud - while in pursuit of Narojaipal (Trilochanpal),
who had received military assistance from Samgramraja, the ruler of Kashmir (1003-28)
- plundered one of the valleys to the south of Kashmir and converted a large number ofits
people to Islam. Kalhana's twelfth-century metrical chronicle of Kashmir, the
Rajatarangfnf, also describes this invasion, but does not speak of a conversion to Islam.
The statement of Bayhaqi is doubtless an exaggeration.
It is, however, possible that some of Mahmud's soldiers, finding it difficult to cross the
mountains towards the plains of India, stayed behind and settled in Kashmir. It is after
these Turkish invasions that Kalhana refers, for the first time, to the presence of Turkish
Muslim soldiers in Kashmir when describing the reign of Harsa (1089-1111); later rulers
also employed Turkish mercenaries. From the account of the Venetian traveller, Marco
Polo, it appears that by the end of the thirteenth century, there was a colony of Muslims in
Kashmir, for he says, 'The people of the province [Kashmir] do not kill animals nor spill
blood; so if they want to eat meat they get the Saracens who dwell among them to play the
butcher.' 3
The Hindu rulers of Kashmir seem to have been munificent and hospitable to the Mus-
lim soldiers of fortune, who continued to enter the valley until the establishment of Muslim
rule; it was one of these Muslim adventurers, Shah Mir, who established the first Muslim
sultanate in Kashmir. The Kashmiri and Mughal historians recount different legends about
the ancestry of Shah Mir. According to Jonaraja, Shah Mir was the descendant of Partha
(Arjuna) of Mahabharata fame. Abu '1-Fadl cAllam!, Nizam al-Din and Firishta also state
that Shah Mir traced his descent to Arjuna, the basis of their account being Jonaraja's
Rajatarangfnf, which Mulla cAbd al-Qadir Bada'uni translated into Persian at Akbar's
orders. It is likely that either Jonaraja, in order to glorify the family of his patron (Zayn
al-c Abidin, a direct descendant of Shah Mir: see below), or Shah Mir, after coming to the
throne, worked out an apocryphal genealogy connecting himself with the legendary heroes
of the past; this was a common practice with rulers and dignitaries of those days. Accord-
ing to some Persian chronicles of Kashmir, Shah Mir was a descendant of the rulers of
Swat, but it is more probable that his ancestors were of Turkish or Persian origin and had

3
Yule and Cordier, 1903 , p. 167.

311
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

migrated to Swat. Shah Mir arrived in Kashmir in 1313, along with his family, during the
reign of Suhadeva (1301-20), whose service he entered. In subsequent years, through his
tact and ability, Shah Mir rose to prominence and became one of the important personalities
of the time. Later, after the death in 1338 of Udayanadeva, the brother of Suhadeva, he was
able to assume the kingship himself and thus laid the foundation of permanent Muslim rule
in Kashmir. Dissensions among the ruling classes and foreign invasions were the two main
factors which contributed towards the establishment of Muslim rule in Kashmir. Because
of the long period of weak reigns and internal troubles, the Lavanyas and Damaras (the
local chiefs) had become the most powerful element in the valley; they continually rose in
rebellion and prevented the growth of a strong centralized government. Suhadeva seems to
have played off these chiefs against each other and thereby established his authority over
the whole of Kashmir. But at the same time, he alienated the Brahmans, the traditional
class of officials, by imposing taxes on them.
Meanwhile, in 1320, Zulju or Dhu '1-Qadr Khan invaded Kashmir at the head of a large
army. The sources regarding the origin of Zulju are not unanimous. According to Jonaraja,
he was a 'commander of the army of the great King Karmmasena' (who is unidentified).
Elsewhere, however, Jonaraja calls Zulju 'the king of the mlechchhas', meaning that he
was a Muslim. The Mughal historian Abu '1-Fadl c Allam! holds that Zulju was the 'chief
commander' of the ruler of Kandahar, and Nizam al-Din and Firishta call him the mfr-
bakhsh (paymaster-general) of Kandahar. The Persian chronicles of Kashmir assert that
Zulju was a Mongol from Turkistan, which could be correct since the Mongols had not only
repeatedly invaded Kashmir prior to this time, but, if we believe Rashid al-Din, had even
succeeded in temporarily subjugating the country. The chiefs did not come to the aid of
Suhadeva and he was left alone to face the invader. He tried to save his kingdom by paying
the Mongols a large sum of money to withdraw from the country; but Zulju's appetite
for plunder merely increased. Suhadeva himself fled to Kishtwar, leaving the people at
the mercy of the invader. The Mongols plundered and enslaved the people, burnt down
buildings and destroyed crops. After a stay of eight months, Zulju left the valley through
the Banihal pass, where he perished along with his prisoners in a heavy snowfall. Famine
was the natural consequence of the wholesale destruction of the stores of grain and of
standing crops by the invading army.
Zulju's invasion proved to be a turning-point in the history of Kashmir and contributed
towards the establishment of Muslim rule there, for Rinchana rose to power in its aftermath.
He was originally from Ladakh, where his father had been chief. Fearing an attack on
his life, Rinchana had sought refuge in Kashmir, where he was employed by Suhadeva' s
commander-in-chief, who had shut himself up in the fort of Lar during Zulju's invasion.

312
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

After Zulju's departure, this commander, Ramchandra, tried to establish his own authority,
but Rinchana treacherously had him murdered and his family imprisoned, and seized power
himself.
The fact that Rinchana was able to rise from the position of a refugee to that of a
sovereign clearly demonstrates the state of anarchy and discord which prevailed in Kashmir
at the time. Rinchana, however, proved an able ruler and restored peace and prosperity to
the country. The most important event of his reign was his conversion to Islam, which has
been variously recorded. According to Jonaraja, Rinchana wanted to become a Hindu, but,
on the grounds that he was a 'Bhotta' (Tibetan Buddhist), the Brahman Devasvami refused
to initiate him into Hinduism. This story seems to have been invented by Jonaraja, however,
resentful that Rinchana had accepted Islam, for, if Rinchana had wished to become a Hindu,
there should have been no difficulty for him, especially since he was a king. According to
a popular version of the story, supported by most of the medieval Muslim scholars of
Kashmir, Rinchana accepted Islam because of 'divine grace'. It is said that after Rinchana
came to the throne, he held discussions with both Hindu and Buddhist priests, in order to
ascertain the 'Truth', but none could satisfy him. Finally, he decided to accept the religion
of the first person whom he should see the next morning. That person was Sayyid Sharaf
al-Din, a Suhrawardi Sufi saint, who at the time was offering prayers near the royal palace.
Rinchana immediately went to him, and, after inquiring about his religion, accepted Islam.
In reality, it is more probable that Rinchana's conversion to Islam was prompted by
political reasons. In the absence of co-operation from the Hindus, only the Muslims in
Kashmir would support Rinchana's newly acquired kingdom. It is not, therefore, unlikely
that Shah Mir, who, according to Jonaraja, was 'a lion among men', persuaded Rinchana
to accept Islam. Abu'l-Fadl c Allami, who made a careful study of the history of Kashmir,
confirms the fact that Rinchana accepted Islam because of his intimacy and association
with Shah Mir, whom he appointed his minister. His decision to embrace Islam might
also have been influenced by the penetration of Islam into the countries outside Kashmir,
particularly with the conversion to Islam of the Mongol 11 Khanid Ghazan Khan in Persia in
1295. Whatever the truth, Rinchana's conversion to Islam must be seen in a wider context,
and not just as the result of either a Hindu refusal to take him into their fold or of a chance
meeting with Sayyid Sharaf al-Din. Rinchana (or Sadr al-Din, the Muslim name which
he adopted) died in 1323. Soon after his death, Hindu rule was once again established
in Kashmir under Udayanadeva, the brother of Suhadeva, who nevertheless bestowed the
territory of Kramarajya and other districts on Shah Mir's two sons, c Ali Shir and Jamshid.
Meanwhile, Kashmir was once again threatened by a foreign invasion. According to
Jonaraja, the invader was 'Achala', supported by 'the lord of Mugdhapura', whom it is not

313
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

possible to identify. The Persian chronicles assert, however, that it was an army of the Turks
(lashkar-i turk). Modem scholars have not identified the Turks of these chronicles, but it
seems likely that they were Turco-Mongols who had previously, and on several occasions,
invaded the valley. The incompetent Udayanadeva fled to Ladakh, leaving his wife Kota
Ran! to face the invader. With the help of Shah Mir and Bhikshana, a Hindu noble, she
repulsed the enemy. After the enemy had withdrawn, Udayanadeva returned and regained
the throne, but his cowardly flight had greatly impaired his prestige. His relations with
Shah Mir did not remain cordial and he began to suspect his loyalty. Because of his heroic
stand against the invader, Shah Mir had become exceedingly popular among the people.
As a result, he became politically ambitious and, according to Jonaraja, he 'did not deem
the king even as grass'. He had already taken steps to win over the leading chiefs to his
side. He bestowed on them his daughter and granddaughters in marriage and made large
gifts to them, waiting for an opportunity to assume the kingship himself. It came soon,
in 1338, as Kota Ran! took the reins of government into her own hands after the death
of Udayanadeva. Realizing the extent of Shah Mir's ambition, she raised Bhikshana to
prominence as a counterpoise to him and transferred the capital from Srinagar, where Shah
Mir had a considerable following, to Andarkot.
The rise to power of Bhikshana was an open challenge to Shah Mir. He did not, however,
make his feelings public, but feigned illness and soon removed his political rival by having
him assassinated. Later, Shah Mir sent a proposal of marriage to Kota Rani, which she
rejected, perhaps thinking it beneath her dignity to marry a man who had been in her
service. After Shah Mir, with the help of the chiefs, successfully besieged her, however,
she surrendered and accepted the proposal of marriage. Even so, as she had married him
under pressure, Shah Mir suspected her loyalty and imprisoned her. He ascended the throne
himself in 1339, under the title of Sultan Shams al-Din.
Shah Mir's coup firmly established Muslim rule in Kashmir. The details of the admin-
istrative machinery that he created are not known, but drastic changes cannot have been
made at that time. The Muslim community of Kashmir was a minority, with no outside
contacts or support. Power remained, as before, in the hands of the Hindu chiefs, with
whose help Shah Mir had established himself on the throne. In order to increase the num-
ber of his supporters and to check the ambitious chiefs, who had been the main cause of
confusion and disorder in the preceding reigns, Shah Mir patronized the families of the
Chaks and Magres, who were of indigenous origin. According to Jonaraja, Shah Mir made
gifts to certain chiefs; it seems that, following the pattern of the Turkish sultans of Delhi,
Shah Mir assigned iqtif s (land grants) to his supporters.

314
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

Shah Mir was succeeded in 1342 by his eldest son, Jamshid, who had gained consider-
able experience in the art of administration during the reigns of his father and Udayanadeva.
However, in the field of statesmanship he was no match for his younger brother, c AH Shir,
who won over a number of important nobles and deposed him within a year (1343); he
died two years later. c AH Shir styled himself Sultan c Ala' al-Din. He ruled for about twelve
years, but very little is known about his reign. From Jonaraja's account, it appears that he
was a just and able ruler. He founded the town of c Ala' al-Dinpora, now a part of Srinagar,
and made it his capital. He died in 1355 and was succeeded by his son Shivasvamika, who
assumed the title of Shihab al-Din. Shihab al-Din was one of the ablest rulers of the Shah
Mir dynasty. From a military point of view, his reign has been described as the most glori-
ous epoch in the history of Kashmir. He not only curbed the growing power of the feudal
chiefs and consolidated his position, but also undertook military expeditions.
After Lalitaditya (724-61), Shihab al-Din was the first ruler of Kashmir whose army
campaigned outside the kingdom. Jonaraja and the Persian chronicles of Kashmir have
given a highly exaggerated account of his conquests; the Kashmiri chroniclers implausibly
attribute to Shihab al-Din the conquest of territories such as Pakhli, Swat, Sind, Multan,
Kabul, Ghazna, Kandahar, Badakhshan and some parts of Transoxania. It is much more
feasible that he conquered and annexed to his kingdom (as is also claimed) Baltistan and
Ladakh. His most memorable campaign, however, is said to have been launched against
Rriiz Shah Tughluq of Delhi, with an encounter on the banks of the Sutlej in which
neither side secured a decisive victory: the peace agreement allotted the territories from
Sirhind to Kashmir to Shihab al-Din, and the rest, lying to the east, to Rriiz Shah; and mar-
riage alliances were also contracted. The Kashmiri chronicles do not corroborate, whether
directly or indirectly, Shihab al-Din's supposed external conquests. Moreover, bearing in
mind the geographic location of Kashmir and the limited resources and numerical strength
of its army, such vast conquests were impossible. In fact, the chronicles magnified his mil-
itary exploits, which must have been limited to Gilgit and Baltistan in the north, Ladakh in
the east and Kishtwar, Jammu and other hill states in the south.
Shihab al-Din was undoubtedly a great ruler who governed his kingdom efficiently. In
1360 the valley suffered badly from a devastating flood. The sultan provided prompt relief
and, in order to prevent similar future calamities, built a new town on higher ground near
Kohi-Maran, which he named Lakshminagar, after his queen Lakshmi. However, some
of his measures were less conducive to the welfare of his people; thus he ordered hanjfs
(boatmen) to serve him gratis for seven days every month. Towards the end of his reign he
came under the influence of Lasa, the daughter of Queen Lakshmi's sister, who succeeded

315
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

not only in alienating the sultan from his queen, but also managed to have the sultan's two
sons exiled. The last days of Shihab al-Din were not happy and he died in 1373.
Shihab al-Din was succeeded by his younger brother Qutb al-Din (Rindal), who was
an efficient and highly cultured ruler. It was during his reign that the great Persian saint
and scholar Sayyid cAli Hamadani (1314-85) arrived in Kashmir in 1381 and was, with a
large number of his followers, warmly received by the sultan. Sayyid cAli travelled widely
in the valley and asked a number of his prominent disciples to settle in places that were
great Hindu centres of the time. These followers established khanaqahs (dervish convents),
which led to the emergence of a whole network of centres for the preaching and teaching of
Islam. Thus Sayyid cAli's arrival gave a great impetus to the spread of Islam in Kashmir.
Because of their different attitudes towards non-Muslims, however, relations between Qutb
al-Din and Sayyid cAli did not remain cordial. At the time, the majority of the Kashmiri
people were non-Muslims and high government officials were also Hindus. Thus political
exigency demanded that the sultan should follow a policy of conciliation towards his non-
Muslim subjects. The intransigent Sayyid cAli did not approve of this, and finding that the
sultan was not responsive to his teachings, left the valley in 1385.
Qutb al-Din died in 1389 and was succeeded by his son Sikandar (who was a minor),
with the latter's mother, Queen Sura, acting as regent. She appears to have been a woman
of courage and ability, but by the time Sikandar took over the reins of government, his
chief minister Rai Magre had grown ambitious and powerful and soon openly challenged
the authority of the sultan. However, the latter marched against him and Rai Magre was
captured and imprisoned.
It was during Sikandar's reign that another wave of Sufi saints and cuzama' arrived,
headed by Mir Muhammad Hamadani (1372-1450), the son of Sayyid cAli, who arrived in
Kashmir in 1393. Mir Muhammad's arrival in Kashmir marked a turning-point in its his-
tory. In the early years of his reign, Sikandar had followed the policy of tolerance towards
non-Muslims as practised by his predecessors, but after the arrival of Mir Muhammad,
his attitude changed and a strictly orthodox policy was introduced. The selling of wine,
(public) dancing of women, music and gambling were prohibited and the jizya (poll tax)
on non-Muslims was imposed for the first time. Hindus were prevented from applying the
tilak (religious mark) on their foreheads and the custom of sati (immolation of a Hindu
widow on her husband's funeral pyre) was banned. Thus Jonaraja writes, 'The good for-
tune of the subjects left them and the king forgot his kingly duties and took delight, day
and night, in breaking images.'
When Timur invaded India in 1398, however, Sikandar showed great maturity as a
statesman. In order to avert the invasion of his kingdom, he sent an envoy to Timur

316
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

professing submission. Timur was satisfied with this and directed the envoy to ask Sikan-
dar to join him at Dipalpur. Accordingly, Sikandar left Srinagar, but upon reaching Jabhan,
Timur's ministers demanded a contribution of 30,000 horses and 10,000 durusts of gold,
and he returned to Kashmir to collect the items demanded. When Timur came to hear of
this, he reprimanded his ministers for having demanded a tribute far beyond the capacity
of the Kashmiri ruler's resources. He informed Sikandar that he should merely present
himself to the conqueror on the banks of the Indus. Upon receiving this message, Sikan-
dar again started out from Srinagar, but on reaching Baramulla, he learnt that Timur had
already crossed the Indus; hence he returned to Srinagar, providence having saved Kashmir
from a great scourge.
After the death of Sikandar in 1413, his eldest son Mir Khan ascended the throne with
the title of cAli Shah. cAli Shah was ignorant of the art of government and was dom-
inated by his chief minister, Siiha Bhatta (Sayf al-Din), who had accepted Islam at the
hands of Mir Muhammad during the reign of Sikandar. But Siiha Bhatta soon died in 1417
and shortly afterwards Shah! Khan, cAli Shah's younger brother, became chief minister.
Like his father, cAli Shah was very religious and decided to give up the throne to per-
form the pilgrimage to Mecca, entrusting the reins of government to Shah! Khan. But upon
reaching Jammu, its ruler, who was his father-in-law, persuaded him to return to his king-
dom. Inc Ali Shah's absence, Shah! Khan had revealed his own ambition: he defeated cAli
Shah at Thana in 1420, and at Srinagar declared himself sultan, assuming the title of Zayn
al-c Abidin.
Zayn al-c Abidin was undoubtedly the greatest of all the Muslim sultans of Kashmir.
Although he followed his religious duties strictly and showed great respect for Muslim
saints and scholars, he did not allow this to interfere with the administration of the country.
Aware that the government needed broad-based support, he allowed complete freedom of
worship to all his non-Muslim subjects. He offered high positions to able and meritorious
non-Muslims in the administration. He also celebrated the Hindu festivals and banned the
slaughter of cows. The jizya, which had been imposed by Sikandar, was reduced from
2 pales to 1 nominal masha and then dropped altogether. Other taxes which had been
imposed only on non-Muslims in previous reigns were also abolished. All Hindus who had
left the valley during the reign of Sikandar were encouraged to return to Kashmir and the
Brahmans were given rent-free lands. Zayn al-c AbiDin thus proved to be the most tolerant
and benevolent Muslim ruler in the history of Kashmir. His concern for his people's welfare
and development led him to lay out a large number of canals which helped not only to
reclaim marshy lands for cultivation, but also to provide irrigation facilities for parched
areas.

317
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

The sultan also reformed the administrative system. Corrupt judges were severely dealt
with and venality among the officials was rooted out. Similarly, crime was ruthlessly put
down. In order to prevent fraudulent property transactions, Zayn al-c Abidin introduced a
system of registration of important documents. He provided his subjects with a code of
laws, which he had engraved on copper plates placed in villages and towns for the infor-
mation of the general public. The prices of commodities were also inscribed on copper
plates and placed in public markets. The sultan was a great patron of learning. He extended
royal patronage to both Persian and Sanskrit scholars, establishing a translation bureau in
which Sanskrit works were translated into Persian and vice versa. He took a keen interest in
the spread of education. One of the famous centres of learning was the seminary of Shaykh
Ism~nl Kubrawi, to which students came not only from India but also from Kabul and Tran-
soxania. The sultan was a great builder: among his works were bridges and rest-houses, for
the convenience of traders and travellers. His most magnificent edifice, however, was his
palace at Rajdan (Nawshahr, in Srinagar), which, according to Mirza Haydar Dughlat, had
'twelve storeys, some of which contain fifty rooms, halls and corridors. The whole of this
lofty structure is built of wood.'
It is, however, for Zayn al-c Abidin's encouragement of arts and crafts that his name
has become immortal. He not only revived traditional arts and crafts, but also introduced a
number of new ones, inviting teachers and craftsmen from Persia and Central Asia to train
local artists and artisans. As a result of his liberal patronage, Kashmir became a 'smiling
garden of industry'. The country made significant progress in wood-carving, paper-making,
papier mache, and silk, shawl and carpet weaving. 'In Kashmir one meets with all those
arts and crafts', writes Mirza Haydar, 'which are in most cities uncommon ... In the whole
of Transoxania, except in Samarkand and Bukhara, these are nowhere to be met, while in
Kashmir they are even abundant. This is all due to Zayn al-c Abidin.'
The sultan had four sons. One of them, Adam, had fallen out of favour when he revolted
against his father in 1459; hence on Zayn al-c Abidin's death in 1470, another son, Hajji
Khan, ascended the throne with the title of Haydar Shah. Haydar Shah appointed his
younger brother, Bahram, as chief minister, and his own son, Prince Hasan Khan, as heir
apparent. In order to strengthen his position, he married Prince Hasan to Hayat Khatun,
the daughter of Sayyid Hasan Bayhaqi, a leading and powerful Sayyid. Yet Haydar Shah
soon succumbed to pleasure and drinking and neglected the administration of the kingdom.
This resulted in internal unrest and the declaration of independence of the tributary states,
such as Poonch, Rajauri and Jammu; Prince Hasan was asked to proceed against the rulers
of these states and successfully subjugated them. In 1472, when Haydar Shah died after a
brief reign, the minister Malik Ahmad Yattii declared Prince Hasan as ruler, with himself

318
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

as chief minister. The new sultan (known as Hasan Shah) began his reign well, but power
struggles among the various groups of the nobility started as soon as he had come to the
throne.
After the accession of Sultan Qutb al-Din in 1373, a large number of saints and scholars
had started to pour into Kashmir from Persia and Central Asia. Warmly received by the
rulers, these immigrants included the Bayhaqi Sayyids, who had arrived in Kashmir dur-
ing the reign of Sultan Sik:andar. The Hamadani Sayyids - Sayyid cAli and his son, Mir
Muhammad - and their disciples were mainly teachers and preachers and sought the help
of the sultans and their nobles to spread the faith of Islam. But the Bayhaqi Sayyids gen-
erally focused their energy on establishing family ties with the ruling house and the high
government officials; soon after their arrival in Kashmir, they made matrimonial alliances
with the royal family, and with the help of these, they obtained important positions in the
administration and also took an active part in the intrigues and rebellions which followed
the death of Zayn al-c Abidin.
Malik Ahmad Yattu, chief minister of Hasan Shah, although very loyal to the sultan,
was over-ambitious. He in time turned against the Bayhaqi Sayyids, who, because of their
relationship with Hasan Shah, had secured a commanding position in the state administra-
tion. But before taking on the Sayyids, Malik Ahmad Yattii strengthened his position by
marrying his adopted son, Tazi Bhatt, to the sister of Jahangir Magre, the commander-in-
chief of the army; Tazi Bhatt now became the leading figure in the anti-Sayyid campaign.
The sultan, fearing an open revolt, exiled all the leading members of the Bayhaqi family,
but eventually the Bayhaqi Sayyids were recalled from Delhi and Sayyid Hasan Bayhaqi
became chief minister.
Like his father, Sultan Hasan Shah was a heavy drinker. When the sultan died, on 19
April 1484, Sayyid Hasan Bayhaqi placed his own 7-year-old grandson, Muhammad Shah
(the son of Hasan Shah and Hayat Khatun), on the throne. With the accession of Muham-
mad Shah, Kashmir witnessed, on the one hand, a bitter struggle for power between the
Kashmiri nobility and the Bayhaqi Sayyids, and, on the other, a civil war between Muham-
mad Shah and a rival claimant, Fat'h Khan. For over a quarter of a century, complete
confusion and anarchy, involving virtually the entire nobility, existed in the valley.
The Bayhaqi Sayyids, who now enjoyed absolute power, began to harass the prominent
Kashmiri nobles. In retaliation, in 1484, the nobles made a surprise attack, killing fifteen
Bayhaqi Sayyids, including Sayyid Hasan, the chief minister, who was now succeeded
in this office by his son Sayyid Muhammad. The Kashmiri nobles soon reorganized and
made another attack on the Sayyids, defeating them and forcing them to leave the country
once again, but soon succumbed to internal discord with a series of intrigues and coups.

319
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

Eventually, Fat'h Khan (who ruled as Fat'h Shah) became sultan of Kashmir, actually
for the second time in 1505, and appointed Shams Chak as his chief minister. But rivals
procured Shams Chak's murder, making Musa Raina chief minister in his place. Musa
Raina was an able administrator, but since he had been converted to Shfism by Mir Shams
al-Din clraqi, many Sunni nobles turned against him and he was killed while trying to flee
from the valley. Thus religious differences were added to the melee of personal conflicts. In
1514 Fat'h Shah was deposed and Muhammad Shah came to the throne for the third time.
He had only ruled for one year when the Chak nobles, headed by Kaji Chak, dethroned him
and caused him to flee to Panjab. Fat'h Shah now became sultan for the third time in 1515,
but died in exile shortly afterwards. Muhammad Shah, already on his way to Kashmir, was
warmly received and declared sultan for the fourth time in 1517.
Although the civil war between Muhammad Shah and Fat'h Shah had thus come to an
end, peace still eluded the country. Ambitious nobles continued to quarrel among them-
selves. Muhammad Shah appointed Kaji Chak as chief minister. Disgruntled nobles used
Fat'h Shah's three sons against Kaji Chak, but were unable to overthrow him. In 1528-9,
Nazuk Khan, the third and only surviving son of Fat'h Shah, with the support of a Mughal
army supplied by the emperor Babur, successfully invaded Kashmir and was briefly hailed
as sultan. He only lasted until 1530, for Abdal Magre, the chief minister, released Muham-
mad Shah from prison and enthroned him as nominal ruler again, with the leading nobles
dividing the kingdom among themselves. The Mughals were given gifts and requested to
leave.
Soon afterwards, Kashmir was attacked from the north-east by Mirza Haydar Dughlat,
who, in 1532, was deputed by Sultan Sacid of Kashghar to conquer Ladakh. Mirza Hay-
dar occupied Ladakh and then proceeded to Kashmir, entering it without much resistance
in January 1533. For some three months the ravages of Mirza Haydar's army continued
relentlessly until the cuzama' encouraged the Kashmiris to defend themselves by issuing
a decree which proclaimed that fighting against the invaders was not only permissible but
obligatory. Their attacks soon wore down Mirza Haydar and his troops: he made peace
with the Kashmiri nobles and in May 1533 he left Kashmir by the same route as he had
come.
In 1537 Sultan Muhammad Shah died and his second son, Shams al-Din II, wasen-
throned. But once again power struggles between the different groups of the Kashmiri
nobility began. Shams al-Din soon died and Ismacn, another son of Sultan Muhammad
Shah, became sultan. One group called for help from Mirza Haydar, who had entered the
service of Humayun after the death of Sultan Sacid. Since he had, in 1533, conquered
Kashmir without much resistance, Mirza Haydar prevailed on Humayun to let him proceed

320
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

to Kashmir once again. He entered Srinagar on 22 November 1540, without encountering

any resistance, and Sultan Ism~fll Shah fled to seek help from Shir Shah Sur in Delhi.
In the beginning, Mirza Haydar showed great respect for the Kashmiri nobles, especially
the Chaks, who were mainly Shfites. Out of regard for the Chak nobles, he visited the tomb
of Mir Shams al-Din clraqi at Zadibal in Srinagar, but relations soon became strained. To
uproot the influence of the Chaks, Mirza Haydar embarked upon an anti-Shfite policy; the
tomb of Mir Shams al-Din clraqi was razed to the ground and his son, Shaykh Daniyal, was
beheaded on the grounds that he had reviled the first three caliphs. Mirza Haydar claimed
that no one now openly dared to profess Shfism as a result of his policy of persecution.
Meanwhile, he began to neglect the local nobility and became increasingly dependent on
the support of his own followers, whom he appointed to responsible positions in the admin-
istration. This led to a revolt in which he was killed in October 1551. There is no doubt
that Mirza Haydar's intolerant policies brought untold misery upon the people of Kashmir.
However, he was not only a brave soldier but also a great patron of culture; he was largely
responsible for the revival of arts and crafts, which had been languishing after the death of
Zayn al-c Abidin; it was in Kashmir that he composed the most famous of his works, the
Tarfkh-i Rashfdf, a history of the Mughals of Central Asia.
Mirza Haydar's death did not bring an end to the scramble for power among the nobles.
Nazuk Shah was allowed to continue as sultan, but the real power was in the hands of his
chief minister. Dawlat Chak secured the ascendancy and in late 1552 deposed Naziik Shah
and enthroned Ibrahim Shah, the son of Muhammad Shah, for the second time. Although
Dawlat Chak tried to win over other factions of the nobility, he managed to arouse jealousy
within his own family, leading to his fall and blinding. Ghazi Chak became chief minister,
eliminated various members of the ruling dynasty and in 1561 himself assumed the title of
Sultan Nasir al-Din Muhammad Ghazi Shah, thereby laying the foundations of the Chak
dynasty. After some 222 years, the rule of the Shah Mir dynasty had thus ended. After Zayn
al-c Abidin, a succession of weak and worthless rulers had exposed the country to internal
revolts and external invasions. The Shah Mir dynasty only paved the way for a powerful
faction of the nobility, the Chaks, to unseat it and assume the reins of government.
During the period of the Shah Mir dynasty, changes of far-reaching significance took
place in the life and conditions of the people of Kashmir. Many elements of Persian and
Central Asian culture were introduced into the life of Kashmiris and the continuous waves
of Muslim missionaries, artisans and fortune-seekers immigrating from Persia and Central
Asia inevitably influenced Kashmiri society. Hindu influence, hitherto dominant in the
court, now began to decline, with the place of the Brahmans taken over by newly arrived
Muslims. Sanskrit, which had received royal patronage for many years, was replaced by

321
ISBN 978-92-3-103467-1 Contents KASHMIR UNDER THE SUL TANS ...

Persian. The Bayhaqi Sayyids, who wielded great influence in the court, 'neglected men
[who were] learned in the vernacular and in Sanskrit', in the words of Srivara. However,
Sanskrit continued to be the literary language of the Hindu elite, although non-Muslims
found that their prospects of employment and promotion were enhanced by a knowledge
of Persian and so started to learn it. Bhattavatara, a scholar of Zayn al-c Abidin's time,
who was enamoured of Firdawsi's Shah-nama, composed the Jainavilasa, which contains
the sayings of the sultan. Srivara translated Jami 's Yusuf u Zulaykha and entitled it Katha-
Kautuka. Hence, in the course of time, Hindu society was split into two groups: the Persian-
speaking Hindus, who were called Karkun (the class of officials), and the Sanskrit-speaking
Hindus, who included the Pandits (religious scholars). The families of Sanskrit-studying
and Persian-studying Hindus did not intermarry, but formed endogamous groups.
The Persianization of the administration had a cultural counterpart. Although Sultan
Qutb al-Din had refused to promote the missionary activities of Sayyid cAH Hamadani,
he nevertheless followed his advice and gave up dressing in the Hindu fashion. From
Jonaraja's account, it appears that, by the time of Sikandar, Hindus too had adopted Muslim
dress. Besides dress, the diet of the Kashmiris also underwent a change. Lamenting these
changes, Jonaraja remarks, 'As the wind destroys the trees, and the locusts the shali crop,
so did the Yavanas [Sayyids] destroy the usages of Kashmira.' Similarly, Srivara ascribes
the misfortunes of the people of Kashmir to their acceptance of changes in their way of
life. But these protests were in vain; the influence of Persian and Central Asian culture
continued to increase day by day. The immigrants from Persia and Central Asia were also
responsible for establishing madrasas, several of them in Srinagar, which encouraged the
dissemination of Islamic spiritual and intellectual values.
On the whole, the attitude of the Shah Mir dynasty towards its subjects was one of con-
sideration.Non-Muslims embraced Islam for various reasons and under various pressures,
but the administration did not, in general, create a situation in which the people felt forced
to abandon their former way of life. The intolerant attitudes adopted by Sultan Sikandar
and Mirza Haydar Dughlat were exceptional rather than usual for rulers in Kashmir. The
period was, on the whole, one of peaceful change, from the ancient Hindu system to a Per-
sianized form of Islamic society; as a result, culturally, Kashmir became part of the Persian
and the Central Asian world.

322
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA TIMUR 1370

16

CENTRAL ASIA UNDER TIMUR FROM

1370 TO THE EARLY FIFTEENTH
CENTURY*
K. Z. Ashrafyan

Contents

Formation of Timur's empire 325

The Sarbadar movement . . 326
Balkh and Samarkand under Timur . 327
Reorganization of Timur's army 330
Timur's military campaigns . . . 332
Socio-economic conditions under Timur 339
Urban development, crafts and trade 342
Relations with west European rulers 345
The succession struggle . . . . . . . 346

One of the most extensive military empires in the medieval Islamic East was that of Timur,
the fruit of his long years of campaigning and the resultant destruction of many towns and
regions. Into this empire were incorporated, in addition to Transoxania and Khwarazm, the
regions around the Caspian Sea, Iran, Iraq, part of the southern Caucasus, and the territory
of present-day Afghanistan and northern India. The heart of the empire was Transoxania,
incorporated after the death of Chinggis Khan, and under the terms of the arrangements
made by him, into the appanage of Chaghatay.
At the beginning of the fourteenth century, bitter disputes arose between the princes
of the other territories of the former Mongol empire and the Chaghatay Khans. In their

* See Map 8.

323
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA TIMUR 1370

struggle to increase their power, the Khans sought support not only among the leaders of
the Mongol tribes and clans (leaders known as noyans in Mongolian, begs in Turkish),
but also among the local feudal landowners and to some extent the urban notables. Kebek
Khan (1318-26), settling, contrary to nomadic custom, in Transoxania, in the valley of the
Kashka Darya river, built himself, at a distance of 2farsakhs (about 12 km) from Nakhshab
(Nasaf), a palace (qarshi in Mongolian) around which the town of Karshi later grew up.
The monetary system he introduced, in imitation of that of the II Khanid ruler of Iran,
Ghazan Khan (1295-1304), of the house of Htilegti, was designed to extend trade rela-
tions and curb the abuses of officials and swindlers. In accordance with Mongol tradition,
Kebek Khan divided Transoxania into military-administrative districts, or tumens (in Per-
sian orthography, taman), that is, '10,000' (the original meaning being a group of 10,000
fighting men or a territory providing that number of warriors). The holdings of many local
landowners became tumens, and the landowners themselves hereditary governors.
An important step in the acceptance of local cultural tradition by some of the Mon-
gols was the adoption of Islam by Tarmashirin (1328-34), brother and successor of Kebek
Khan. This gave rise to a new wave of dissatisfaction on the part of the nomadic Mon-
gols. Tarmashirin was killed and the headquarters of the Khanate was transferred to Semi-
rechye. The upholder of the 'settler tradition', and the last Chinggisid of the Chaghatayulus
(domains in Mongolia), Kazan Khan, brought the seat of the Khanate back to Transoxania
only to be killed in a battle in 1346 against one of the Mongol leaders, the amir Kazagan.
Kazagan, not being a Chinggisid, did not assume the title of Khan, and ruled on behalf of
the titular Khan of Chinggis's line. Kazagan's power did not extend beyond Transoxania.
The remainder of the disintegrating ulus of Chaghatay (Semirechye and East Turkistan)
came to be called Moghulistan; here the military-nomadic aristocracy of the Mongol tribes
held undisputed sway, under the leadership of Khans of the Dughlat clan. In 1358 Kazagan
was killed by one of the noyans of the Khan of Moghulistan, Kutlugh Timur. Transoxania
was now divided into a few mutually hostile fiefs, belonging to the leaders of the Mongol
and local nobility. Nizam al-Din Shami, author of the earliest of the better-known accounts
of Timur's life, the Zafar-nama [Book of Victories] written on Timur's instructions and
during his lifetime, names some of them. Kazagan's grandson Amir Husayn ruled part of
the wilayat (region) of Balkh, together with the town of Balkh. The remainder of this region
belonged to the head of the Sulduz tribe. Kish and its region were under the sway of Hajji
Barlas. In Shiburghan, Badakhshan and Khuttalan, similar independent leaders established
themselves. Their dissension and strife, according to the historian, sowed confusion in the
affairs of Transoxania. 1

1
Nizam al-Din ShamI, 1937- 56, p. 15.

324
ISBN 978-92-3-103467-1 Contents Formation of Timur's empire

Formation of Timur's empire

In 1360 and 1361 the Khan of Moghulistan, Tughluq Timur, campaigned in Transoxania.
One of his followers was the future 'conqueror of the world', the 'second Alexander', as the
authors of eulogistic histories dubbed their hero, Timur of the Barlas tribe. 2 This Mongol
tribe had settled as early as the beginning of the fourteenth century in the valley of the
Kashka Darya, intermingling with the Turkish population, adopting their religion (Islam)
and gradually giving up its own nomadic ways, like a number of other Mongol tribes in
Transoxania.
The official histories of Timur - written at his command by Nizam al-Din Shami, by
Ghiyath al-Din cAli, the author of the Journal of Timur's Campaign in India, and by the
historian Sharaf al-Din cAli Yazdi, who lived at the court of Timur's son and successor
Shah Rukh - say nothing about his early years, although it is known that he was born
in 1336. According to Ruy Gonzalez de Clavijo, ambassador of the king of Castile, who
visited Samarkand, Timur's capital, in 1404, the father of the 'Emperor of Samarkand'
(el Emperador de Samarcante) was a notable personage (home fidalgo), but not wealthy,
and had no more than 2 or 3 horsemen in his service. Timur had approximately the same
number ('4 or 5') of hired horsemen. With their help he seized from his neighbours 'one
day a sheep, the next day a cow, and when he was able, he feasted with his followers'.
Gradually, thanks to Timur's bravery and 'magnanimity', a force of as many as 300 grew
up around him, and with them 'he began to scour the countryside, robbing and stealing all
he could, for himself and his followers; he also travelled the roads robbing merchants' .3
The historian lbn cArabshah, who is extremely hostile to Timur, writes that his father was
a shepherd.
After the attack on Transoxania, Tughluq Timur heaped 'all manner of attentions and
kindnesses' on Timur. The Kashka Darya tumen of the fugitive Hajji Barlas was bestowed
on him in 1361. But Timur soon broke off relations with the Khan of Moghulistan and his
son Ilyas Khoja, who had been left as ruler in Transoxania.
Now that he was a ruler and amir of a rich tumen, Timur made contact in 1361 with
one of the pretenders to power in Transoxania, the chieftain of Balkh, Amir Husayn, seal-
ing the relationship by marrying his sister Ulday Turkan-aga. At the head of their troops,
the amirs carried out predatory raids on the territories of their neighbours. During one of
these they were taken prisoner by the Javuni-Qurbani Turkmens and escaped by a lucky

2 Ghiyath al-Din cAlI, 1915, pp. 133, 139, 183.

3
Clavijo, 1881 , pp. 238-9.

325
ISBN 978-92-3-103467-1 Contents The Sarbadar movement

chance. 4 In a later battle in Sistan - where the local prince had invited the amirs to help
fight his enemies - Timur received arrow wounds in the arm and leg. Lamed for life, he was
given the nickname Timur-Leng (literally 'Iron Cripple'), rendered by European writers as
Tamerlane.
The increase in power of amirs Husayn and Timur threatened Ilyas Khoja. Expelled
from Transoxania after the death of Tughluq Timur, he was proclaimed his successor in
Moghulistan and in 1365 reappeared in Transoxania with a large army. On the banks of
the River Chirchik near Tashkent, a battle was fought that has gone down in history as 'the
battle in the mud'. Heavy rains had turned the ground into a bog, in which the horses of
the allied forces stuck fast. But the Mongols spread large pieces of felt under their horses'
hooves, and their cavalry, manoeuvring freely, carried the day. The amirs fell back towards
Samarkand and then retreated across the Oxus (Amu Darya).

The Sarbadar movement

Meanwhile, a host of jete (bandits), as the Mongol nomads were called in Transoxania,
appeared in the vicinity of Samarkand. In those days, the city was unfortified and its cap-
ture was, it seems, inevitable. An active role in the organization of the defence of the
city was played by the townspeople themselves, among whom the ideas of the Sarbadars
had spread (see further, below). The Sarbadar movement, with its varied composition -
it embraced artisans, the urban poor, peasants and small landowners and contained mes-
sianic Shicite elements - had originated in Khurasan. Having ousted Mongol power from a
number of towns and districts, the Sarbadars in 1337 set up their own state with its centre
at Sabzavar which survived until 1381. A Sarbadar state also sprang up in Mazandaran
(1350-92). The ideas of the Sarbadars were evidently propagated secretly in Transoxania
too; it is known that the ideologue of the Sarbadar movement, Shaykh Khalifa, won many
adherents in Sabzavar and preached for some time not only in Khurasan but also in Central
Asia. According to the fifteenth-century historian Hafiz-i Abrii, author of the anonymous,
no longer extant, History of the Sarbadars, Sarbadar (in Persian, literally 'head in the
noose' or 'gallows-bait') was the name adopted by the members of anti-Mongol popular
movements who were ready to die in the cause of deliverance from tyranny. However, it is
not impossible that participants in uprisings were called 'gallows-bait' by their opponents,
who considered that the 'rebels' deserved the gallows.
The historians also gave the name 'Sarbadars' to the citizens of Samarkand who had
taken the control of the city into their own hands. Their leaders were the madrasa

4
Nizam al-Din ShamI, 1937- 56, pp. 20-1.

326
ISBN 978-92-3-103467-1 Contents Balkh and Samarkand under Timur

student Mawlana-zada; the headman of the cotton-scutchers' district Abu Bakr; and
Mawlana Khurdak-i Bukhari, nicknamed 'the archer' and known as a brave man of respec
ted lineage. Night and day, the citizens of Samarkand worked to fortify the alleyways of
the town, erecting barricades and leaving only the main artery open to free passage. Enter-
ing by this route, the Mongol horsemen found themselves in a trap: the townsmen attacked
them from all sides, showering them with arrows, stones and sticks. The Mongols, unused
to street fighting, retired. On the following days they renewed their onslaught. Unable to
claim victory, they dug in around the city for a long siege, but an epidemic broke out and
many of their horses died. The Mongols were therefore compelled to leave the environs of
Samarkand, where the conflict among the various social groups seems to have intensified.
It is not impossible that the Sarbadar faction (representing the democratic elements of the
population) attempted to put into effect their doctrines, which were similar to those popular
among the Sarbadars of Khurasan, i.e. equal rights to property and repeal of taxes that were
contrary to the sharf a. Hence, no doubt, the indignation of the historians at the actions of
the Samarkand Sarbadars. In the words of Khwandamir, the Sarbadars 'followed the path
of wickedness and sedition and laid avaricious hands on the property of the citizens'. 'O
God,' implores Sharaf al-Din c All Yazdi, 'let not a beggar become a respected man.'
News of the retreat of the Khan of Moghulistan's troops from Transoxania reached
Timur, who hastened to inform Amir Husayn. But the allies did not march immediately
on Samarkand, which remained in the hands of the Sarbadars, whose 'extreme audac-
ity', in the words of Khwandamir, alarmed them. Meeting at the village of Baghlan, the
amirs worked out a plan of action. Through their agents they established contacts with the
'Samarkand gallows-birds', welcomed them with gifts and noble apparel and issued a doc-
ument in the name of Amir Husayn recognizing the Sarbadar leaders as lawful chieftains.
Only in the spring of 1366, several months after the Mongols' departure from Transoxania,
did Timur and Husayn march on Samarkand. Anticipating no easy victory, they resorted to
treachery and invited the Sarbadar chiefs to meet them at the village of Kani Gil. On their
arrival, bearing gifts, the Sarbadars were seized and put to death. If the historian c Abd al-
Razzaq Samarqandi is to be believed, Timur interceded for Mawlana-zada and saved him
from the gallows, but the others were executed. 5

Balkh and Samarkand under Timur

Amir Husayn established himself in Samarkand, assuming the rulership of Balkh also.
Timur took control of Kish (Shahr-i Sabz) and Karshi. Their alliance quickly soured and

5
Stroeva, 1949; Gafurov, 1972, pp. 477-81.

327
ISBN 978-92-3-103467-1 Contents Balkh and Samarkand under Timur

turned to enmity. Each saw the other as the one resolved to strike the first blow. Timur
besieged Balkh: the fortifications were breached and on 10 April 1370, the city was taken.
Husayn tried to hide in one of the minarets beyond the fortress wall, but was captured and
killed by one of Timur's allies. 6 It would seem that a contributing factor in Timur's success
was Husayn's unpopularity with the inhabitants of his domains. In the words of Clavijo,
'this king of Samarkand was not loved by his subjects, especially the ordinary people, the
city-dwellers and some notables' .7 The fortress was demolished and the city sacked by
the army, which also received a share of the rich treasury seized from Husayn. After the
capture of Balkh, Timur summoned, in the city itself, a kurultay (meeting of the military
chiefs and nobles of the tribes). To the title of amir was added the honorific Gtiregen,
since Timur married one of Husayn's widows, Mulk-khanum (Bibi-khanum), who was the
daughter of the Kazan Khan, a descendant of Chinggis Khan.
In deference to Mongol tradition, Timur, in the words of Nizam al-Din Shami, handed
over control of the state to the house of Chaghatay and restored their 'rights'. He pro-
claimed Soyurghatmish, a descendant of Chaghatay, as Khan. 8 After Soyurghatmish's
death in 1388, another Chinggisid, Sultan Mahmud, was appointed titular Khan by Timur.
When this Khan also died in 1402, Timur named no successor and continued to mint coins
bearing the name of Sultan Mahmud; his name was also commemorated in the khutba (Fri-
day worship oration), and the signature of the Khan appeared on the yarlighs for raising
armies and announcing campaigns. 9
In 1370 Timur made Samarkand his capital, 'the residence of the sultans, home of
the Khans, dwelling of the saints, homeland of the dervishes or Sufis and capital of the
learned'. He 'brought under his sway the whole of Transoxania' and 'set in order the
affairs' of the region so that 'there was no room for rebellion there'. 'The people became
prosperous, and, it seems, both the nobility and the mob were pacified under the generous
and kindly protection of the Khan.' 10 In these words, the historian-panegyrist tells how
Timur became the sole ruler of Transoxania.
The main prop of Timur's administration was the warlike nomadic and semi-nomadic
Mongol nobility of Transoxania (particularly the Barlas). These tribes were called
Chaghatays, since they inhabited the ulus of Chaghatay, although they did not belong to

6
Nizam al-Din ShamI, 1937- 56, pp. 59-60.
7
Clavijo, 1881, p. 240.
8
Nizam al-Din ShamI, 1937- 56, pp. 57-8, 65.
9 Ghiyath al-Din cAlI, 1915, pp. 54, 55, 125, 126.
10
Ibid., pp. 17-18.

328
ISBN 978-92-3-103467-1 Contents Balkh and Samarkand under Timur

his house, as Clavijo correctly notes. 11 The Mongols' contemptuous name for them was
qaraunas (half-breeds). According to Clavijo:

These Chaghatays are especially favoured by the king [Timur]; they can go everywhere they
want with their herds, graze them, sow crops and settle anywhere they wish, winter or sum-
mer; they are free and pay no tribute to the king, for they serve him in war when he calls on
them. And let it not be thought that they leave their wives, children and herds anywhere; they
take with them everything they have, whether they go to war or move from place to place. 12

Conscious of the need to broaden his social base, Timur sought the support of the local
landowning nobility, the urban notables and the Muslim clergy. This trend was reflected
in his predilection for Islam and the sharf a even to the detriment of the Yasa of Chinggis
Khan. It is not by chance that the historians portray Timur as an enthusiastic defender of
Islam (which quite possibly he was not), a ghazf (fighter for the faith). 13 It is well known
that Timur was generous to the representatives of the Muslim clergy, even in conquered
countries. When he sacked a town, he forbade the pillage of the property of Muslim reli-
gious institutions and severely punished anyone guilty of it. 14 As his spiritual adviser he
chose the descendant of the Prophet Muhammad, Sayyid Baraka, who, legend has it, fore-
told his victories.
Nevertheless, in Timur's kingdom, where Islam held undisputed sway, neither paganism
nor Mongol custom was completely superseded. To orthodox Muslims, Timur's Chaghatay
soldiers, wearing their pigtails, seemed kafirs (unbelievers, hence 'savages'). At Timur's
court, contrary to Muslim law, feasts were held with copious entertainment, wine and mer-
riment; princesses and queens were present at the banquets, and sometimes gave banquets
themselves. On the walls of Timur's court at Samarkand, Koksaray, Clavijo saw cornices
with representations (contrary to the precepts of the sharf a) of figures from Timur's vic-
tories, his sons, grandsons and amirs. 15 At the weddings of princesses only a few elements
of Mongol ceremony were observed (for example, cups of koumiss, or fermented mare's
milk, were served). Official correspondence was carried on in Persian, but among the clerks
whom Timur kept by him there were 'some clerks, who read and can write in Mongol char-
acters for all his purposes' . 16

11
Clavijo, 1881 , p. 214.
12
Ibid., pp. 195-6.
13
Ghiyath al-Din cAn, 1915, pp. 153 et seq.
14
Ghiyath al-Din cAlI, 1915, p. 132; Sharaf al-Din YazdI, 1972, pp. 241-2.
15
Clavijo, 1881, pp. 289-90.
16
Ibid., p. 201.

329
ISBN 978-92-3-103467-1 Contents Reorganization of Timur's army

Reorganization of Timur's army

Timur saw in the waging of uninterrupted wars of conquest the principal means of increas-
ing his power. 'Not the whole inhabited part of the world is valuable enough to have two
masters' are the words attributed to him by the historian Sharaf al-Din c Ali Yazdi. In these
wars - a source of wealth and booty - the military nobility of the tribes, and the tribes
themselves, were deeply involved in support of Timur. But the purpose of the war was not
only to win booty but also to gain control of the major trade routes linking Europe and
western Asia to India and China. Both the tribal military nobility and the urban patricians
- the local landowning nobility who were engaged in trade and the town's leading traders
and money-lenders - took part.
The instrument of conquest was the army. Its basic striking force was the cavalry,
recruited from the tribal population. The infantry was recruited from the settled peoples
(tajfks), 17 who were assigned to work the catapults, battering-rams and other siege-engines
used in the storming of cities. To defend themselves against the enemy, the foot-soldiers
would form up in front of their troop, dig ditches, build defensive shields and screens, and
under cover of these would shoot arrows and throw spears at the foe. 18
The basic weapons of pitched battle were bows, both large and small (the arrows being
kept in quivers of plate metal). For close combat the soldiers used swords, curved sabres
(shamshfrs), throwing spears, and clubs. By the second half of the fourteenth century,
firearms began to make their appearance in the Near East, but nothing is known of their
use in Timur's armies. The soldier's equipment included shield, body armour and helmet.
On the occasion of Timur's triumphal entry into his Samarkand palace (el castillo ), Clav-
ijo saw carried in front of the amir, at his command, all the weapons and armour used
during the time he had been away from the town. Among other items there were '3,000
suits of body armour, decorated with fine cloth, of very fine workmanship; though they
do not make them strong enough and do not know how to temper iron [properly]'. There
were also many helmets 'round and tall'. Attached to the helmet was a plate, two fingers
in breadth, that descended to the wearer's brows and protected the face against sideways
strokes. At that time Timur gave out armour and helmets to 'knights and other individuals'
(a los caballeros y otras personas). 19 The coat of mail - an expensive article - was obvi-
ously worn only by the amirs and Timur himself. More popular, but still only affordable
by the captains, was the kayak, a protective cloak of velvet, lined with close rows of metal

17 Ghiyath al-Din cAlI, 1915, pp. 168, 212.

18
Nizam al-Din ShamI, 1937- 56, pp. 121, 159.
19
Clavijo, 1881 , p. 293.

330
ISBN 978-92-3-103467-1 Contents Reorganization of Timur's army

plates, fastened in such a way that the cloak retained its elasticity and did not hamper the
wearer's movements.
When fortifications were to be stormed, siege-engines were used, and by means of cat-
apults or ballistas breaches were made in the walls, tunnels were dug under them and fires
started in the tunnels. To storm fortress walls, the warriors used siege-ladders and ropes,
which they threw over the projections of the walls. They were skilled at building pontoon
bridges, which after the river had been crossed were usually demolished. Before battle was
joined, the king's great drum and the war drums were beaten, the kettledrums sounded and
the trumpets blown. The air reverberated with the soldiers' war cries. The kalima, 'There
is no God but Allah', was heard and the thunderous cry of 'Allah is great' 'made the hearts
of the huge lion and the mighty elephant shrink'. 20
The decimal system traditional among the Mongols was adopted by Timur also. The
army was divided into tumens ('ten thousands'); mingliks (in Mongolian) or hazars (a
Persian word that came into use early among the Mongols and Turks), that is to say, thou-
sands; yuzlUks (hundreds); and onluks (tens). Small formations (up to 500 men) were called
khoshuns in Transoxania. The yuzlUks and the mingliks were commanded by yuzbashis and
minbashis. The larger formations were commanded by amirs and princes (mfrzas). Those
dispatched to raise armies in the provinces were known as the tavachis; they had other
important functions as well, for example, the division of spoils among the warriors, the
assignment of sectors for digging of defensive ditches around the camp, and so on.
The army had two wings, the centre (here, as a rule, was to be found the headquarters,
or borgah) and a vanguard, surrounded by outposts (kanbuls), which were often the first
to engage the enemy. The various formations had different coloured clothing, red, white,
blue, etc. Scouts (khabargfrs) reported on enemy movements. The army was accompanied
by porters (kachars). The army camp, if it was pitched close to the enemy's position, was
surrounded by entrenchments, and large defensive works (chapars), as well as mobile tow-
ers, were erected. In extreme cases, linked carts (arabas) were also used as well as screens.
The sources give inflated numbers for Timur's army; Sharaf al-Din c Ali Yazdi says that
there had not been so large an army since the days of Chinggis Khan. In the campaign
against China in 1404, the army allegedly numbered 200,000 fighting men. It was encum-
bered by a large baggage-train, as the fighters were accompanied by their families. 'These
folk,' says Clavijo, 'wherever they are called to war by their king, set off immediately with
all their belongings, their flocks, their wives and children.' 21

20
Ghiyath al-Din cAlI, 1915, p. 154.
21
Clavijo, 1881 , p. 191.

331
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

In the baggage-train were the yurtchis, whose task it was to set up the yurts and lay out
the encampment 'in accordance with accepted rules' ,22 and a whole 'army' of traders and
artisans. According to Clavijo:
In this horde [of Timur's] there are always butchers and cooks, selling boiled and roasted
meat, and others who sell barley and fruit, and bakers, who set up their ovens, knead and
sell bread. Whatever artisans or craftsmen are needed can be found in the horde, all located
in their separate streets. And that is not all; wherever the army goes they take with them
baths and bath attendants, who set up stalls, build houses for the iron, that is, hot baths, with
cauldrons inside for storing and heating water, and everything needful. 23

Because of this cumbersome baggage-train, the army moved slowly, especially when it had
to transport booty and drive along captives. The soldiers therefore carried out raids lightly
burdened, leaving the baggage-train behind.
Timur, though mainly continuing Mongol military traditions, introduced certain inno-
vations in the dispositions and tactics of his army. According to Sharaf al-Din Yazdi, when
Timur was preparing for his decisive encounter with Toktam'ish in 1391, he disposed his
forces (seven traditional kuls, or corps) 'in such a way that no one had ever seen or heard
the like'. Timur's regular inspections of the army helped to maintain the men's fighting
spirit and tighten up discipline, qualities that were responsible for the success of many of
their military undertakings.

Timur's military campaigns

Believing two rulers for the inhabited world to be one too many, Timur devoted his own
time as ruler to constant and pitiless wars of conquest. The Turco-Mongol military chiefs
who were Timur's mainstay were the chief beneficiaries of the seizures of ever more lands
and riches, while the local magnates of Transoxania were also drawn into supporting him
by the prospect of the advantages to be gained. No sooner had Timur consolidated his hold
on Samarkand than he advanced against the White Horde, which occupied the territory
from the region of Yangi-Talas to the borders of Kashghar. The White Horde formed part
of the ulus of JOchi (the eldest son of Chinggis Khan), the other part of which consisted of
the Golden Horde. This division stemmed from the way in which the Mongol forces had
originally been organized: the Golden Horde manned the right flank, the White Horde the
left. As time went by, the White Horde broke away and came under the rule of its own
Khans, who were constantly raiding Turkistan and Transoxania. Urus Khan (1364-83)
became so powerful that he attempted to unite both parts of the ulus of JOchi under his
22
Ghiyath al-Din cAlI, 1915, p. 159.
23
Clavijo, 1881, pp. 233-4.

332
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

own authority. This was of great concern to Timur, who fought repeatedly against the
White Horde. 'He [Timur] frequently did battle with them [the Mongols] of the White
Horde, until finally they chose the right path and accepted the role of cringing servility and
expressions of submission.' 24
In 1372 Timur launched a campaign against Khwarazm. The Sufi dynasty of the Kun-
grat tribe, which had joined forces with the White Horde, had made Khwarazm its base
in the 1350s and early 1360s. The pretext for the campaign was the seizure by Husayn
Sufi, the founder of the dynasty, of Khiva and Kath in southern Khwarazm, which formed
part of the ulus of Chaghatay. Timur, who laid claim to the whole ulus of Chaghatay,
demanded the return of the captured territories. When this was refused, he advanced into
Khwarazm. Kath was captured and Husayn Sufi shut himself up in the fortress of Urgench,
where he soon died. His successor, Yusuf Sufi, entered into peace talks with Timur but
after the latter had left Khwarazm, he retook Kath. Timur mounted a second campaign
against Khwarazm (1373-4), but no actual fighting took place because Yusuf Sufi offered
his apologies. Southern Khwarazm passed into the hands of Timur. In 1388, however, the
ruler of Khwarazm led a revolt stirred up by Toktam'ish. The son of an amir of the White
Horde who had ruled over Mangi'shlak and was slain by Urus Khan for insubordination,
Toktam'ish, with Timur's support, had succeeded after several reverses in routing Urus
Khan and gaining control of the White Horde in 1379.
In 1380-1 Toktam'ish was victorious over Mamai, the Khan of the Golden Horde (who
had just been defeated by Prince Dimitri Donskoy of Moscow at the battle of Kulikovo ),
thus uniting the White and Golden Hordes under his own rule. Timur had been hoping to
establish a vassal relationship between himself and Toktam'ish, but found in him a strong
and wily adversary. In 1387-8 Toktam'ish invaded Transoxania, acting in collusion with
the ruler of Khwarazm, who withdrew his allegiance to Timur. Timur seized Urgench, the
capital of Khwarazm, in 1388 and overthrew the Sufi dynasty. He ordered the inhabitants to
move to Samarkand and had the plundered and devastated city razed to the ground and its
site sown with barley. Only a handful of fine buildings remained of what had been a great
city. In 1391 Timur ordered the restoration of Urgench, but only one part of the city was
rebuilt. 25 As a result of the subjugation of Khwarazm, all the lands of Central Asia with
the exception of Semirechye and the lower reaches of the Syr Darya fell into the hands of
Timur.
In 1381 Timur unleashed his forces on the Kart principality in northern Afghanistan.
The founder of this local dynasty was Rukn al-Din (d. 1245), who had been appointed ruler

24 Ghiyath al-Din cAlI, 1915, p. 30.

25
Sharaf al-Din YazdI, 1972, p. 32.

333
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

of Ghur by Chinggis Khan. In 1248 Herat had become the capital of the much strengthened
principality of the Karts. When the II Khanids established themselves in Iran, the Kart
rulers, who had previously been subordinate to the Great Khans in Karakorum, became
vassals of the II Khanids. The collapse of the II Khanid state shortly after the death of Abu
Sacid (1318-35) enabled the Karts to gain their independence. After the havoc wrought
by the Mongols, irrigation systems were gradually restored and towns and villages rose
again from the ruins. Herat at that time was a major trading and crafts centre. In early
1381 Timur's forces appeared beneath its walls. Having cut it off from Ghur and other
Kart possessions, they proceeded to besiege it. The ruler of Herat, Ghiyath al-Din Pirc All,
withdrew into the citadel, but the townsfolk did not support him and took no active part
in the defence of the city - no doubt because of Timur's promise to spare the lives and
property of those who did not resist. 26
Although, realizing his helplessness, the ruler came to Timur's tent and threw himself on
his mercy, the same fate befell Herat as other captured towns. A heavy tribute was imposed
on it and many leading citizens were deported to Transoxania. Timur confirmed Ghiyath
al-Din's position as ruler of Herat, but it was a nominal appointment only, since he had
to remain in constant attendance upon Timur, at the latter's insistence. Shortly afterwards
he was executed on Timur's orders even though he had made no attempt to forswear his
allegiance. 27
Such was the oppression imposed by the conquerors that in 1382-3 the citizens of Herat
rose in rebellion. They were led by a native of Ghur, nicknamed GhiirThachcha (Son of
Ghur), and their rebellion was supported by strong detachments of fighting men from Ghur.
The officials appointed by Timur to administer Herat were driven out and the troops gar-
risoned there were slaughtered. Timur entrusted the task of putting down the revolt to his
son M'iranshah, who routed the contingent from Ghur at the battle of Herat and overran the
city. 28 Taking the bulk of his army with him, Timur left Herat for Kalat in Khurasan. The
rulers of Kalat, Merv and several other cities, terrified by Timur's successes, submitted to
him.
Timur now turned his attention to subduing the Sarbadarid amirate, whose spiritual
leaders were Shaykh Khallfa and subsequently his pupil Hasan Jiirl. Concealed beneath
the mystical doctrine - with Shicite overtones - that they preached was a call to throw
off the yoke of the II Khans and the powerful Iranian nobles who supported them. The
followers of this teaching called themselves the Sarbadars (see above). An uprising broke

26
Masson and Romodin, 1964, pp. 24-5.
27
Nizam al-Din Shami, 1937- 56, Vol. 1, pp. 82-4.
28
Bartol'd, 19 18, p. 29.

334
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

out in 1337 in Bashtin (Khurasan) under the leadership of a follower of Hasan Jiiri, a local
landowner called cAbd al-Razzaq. The punitive expedition mounted by the II Khanid wazfr
of Khurasan was defeated by the rebels, and the wazfr was captured and put to death. The
uprising spread to other towns and villages of Khurasan. Sabzavar, which was taken by
the Sarbadars, became their headquarters. In 1338 Wajih al-Din Mascud, the successor of
cAbd al-Razzaq, assumed the title of sultan.
Towards the middle of the fourteenth century, under their ruler Yahya Qarabi, the Sar-
badars wrested Tus and Mashhad from the Mongols and on 13 December 1353, invited by
Tughay Timur Khan to his encampment at Gurgan, they took control of the Mongol camp
and executed the II Khan. As time went by, there were increasing divergences between
the radical wing of the Sarbadar movement, represented by craftsmen, town-dwellers and
the peasantry of the surrounding countryside, and such moderate elements as small local
landowners. As the struggle for power intensified, one cAll Mu' ayyad laid claim to the
supreme authority. Seizing power in 1364, he inflicted a crushing defeat on the Sarbadars
representing the rank and file of the population. The loss of their support weakened his own
position, however. In the war with the Kart ruler he lost the eastern lands of the Sarbadars,
including Nishapur. From the west he was threatened by one of the Mongol amirs, Wali,
who had consolidated his position in Astarabad.
In the conflict between cAll Mu' ayyad and Amir Wall, the fortunes of war fluctuated.
There were even periods when the adversaries became allies, as a result of the revolt of the
radical wing of the Sarbadars in Sabzavar in 1378, under the leadership of the Darvish Rukn
al-Din. In alliance with Amir Wall, cAll Mu'ayyad put down the revolt, but enmity soon
sprang up again between them. Amir Wall besieged Sabzavar, whereupon cAll Mu' ayyad
took a step that was to prove fatal. In 1381 he appealed for help to Timur, who was quick
to take advantage of this convenient pretext for interfering in the affairs of the Sarbadarid
state. cAll Mu' ayyad met Timur in Sabzavar as his humble vassal. He was allowed to retain
his title, but was invited to the conqueror's headquarters, while one of Timur's lieutenants
was appointed to Sabzavar. Shortly afterwards, cAll Mu' ayyad was stealthily slain on the
orders of his new overlord.
Two years later, in 1388, an uprising against Timur took place in Sabzavar and the
surrounding area under the leadership of Shaykh Dawiid Sabzavari. It was mercilessly
put down; the town's fortress was demolished and, on the orders of Timur, 2,000 peo-
ple were walled up alive in the towers. That same year, having pacified the 'rebels' of
Sabzavar, Timur annexed Kandahar and a number of other towns and districts in Khurasan
and Afghanistan. The following year, he conquered Gurgan, thereby putting an end to the
reign of Amir Wall.

335
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

Between 1386 and 1404 Timur's hordes repeatedly raided the Trans-caucasian coun-
tries from northern Iran. The kingdom of Georgia, which had gained its political indepen-
dence after the death of II Khan Abu Sacid in 1335, was witnessing a period of economic
and cultural expansion. Georgia put up a stiff resistance to Timur, who crossed its bor-
ders with his troops seven times, but Tifiis fell to him and in 1404 King George VII was
compelled to acknowledge Timur's suzerainty. Armenia, which had lost its own statehood
and become part of the J alayirid principality, had gone into decline under the rule of Turk
and Kurd nomadic tribal overlords and Timur's invasion proved to be a fresh calamity for
the country. An eyewitness describes events as follows: 'By hunger, the sword, captivity,
indescribable tortures and inhuman treatment they [Timur's hordes] turned the populous
Armenian province into a desert.' 29
In 1386 Timur captured Tabriz in southern Azerbaijan, a city that had been laid waste by
Toktam'ish the previous year. Sultan Ahmad Jalayir, the ruler of the Jalayirid lands, which
comprised the bulk of Persian Iraq with the cities of Ramadan, Qazvin and Sultaniyya, as
well as Kurdistan, southern Azerbaijan, Karabagh, Armenia and Arabian Iraq, fled, leav-
ing his domains to their fate. Timur's next victims were the Muzaffarids (1313-93). In
1387 Timur captured their capital Isfahan. The exactions of the tax-collectors appointed
by him caused the city to rebel. The main insurgents were the craftsmen and the poor, led
by a blacksmith. They were mercilessly suppressed: historians recount that Timur's sol-
diers received orders to deliver a prescribed number of severed heads, and minarets were
built with the heads of 70,000 slaughtered citizens. 30 The struggle with the Muzaffarids in
southern Iran continued until 1393. By 1392 Timur's hordes had conquered the Sayyid state
in Mazandaran, which had come into being in the 1340s during the mass uprising against
the Mongols. Its ruler was Sayyid Qawam al-Din, and in social structure and ideology it
differed very little from the Sarbadars.
The conquest of Iran was completed by 1393 and the country was divided into two
vicegerencies. Timur's son Shah Rukh was made vicegerent of the region, comprising
Khurasan with Gurgan, Mazandaran and Sistan, with its centre in Herat, while his brother
Miranshah became the vicegerent of western Iran, including Azerbaijan and Armenia, with
its centre in Tabriz. Turkic-speaking nomadic tribes were brought in from Central Asia to
settle in northern Iran and Azerbaijan.
The purpose of Timur's conquests was not merely to acquire loot but to gain con-
trol of the lucrative major international trade routes. The Golden Horde, subordinate to
Toktam'ish, was stationed astride the trade routes leading from Europe and Asia Minor to

29
Thomas ofMetsopc, 1957, p. 58.
30
Nizam al-Din Shami, 1937- 56, Vol. 1, p. 105; Sharaf al-Din YazdI, 1972, p. 320.

336
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

Central Asia, Iran, Mongolia and China. After seizing control of the Golden Horde inA
1381, Toktam'ish put an end to the rivalries that had tom it apart. He put a great deal of
effort into maintaining the Horde's dominion over the Russians. His chief adversary was
Prince Dimitri Donskoy, whose policy consisted in unifying the Russians around Moscow.
Toktam'ish succeeded in winning over the princes of Nizhniy Novgorod and Ryazan to his
side and in 1382 attacked Prince Dimitri, laying waste north-eastern Russia. As the chron-
icler Nikon records, Moscow, at that time 'a great and wondrous city at the height of its
wealth and glory', was captured by Toktam'ish and put to the torch. Dimitri Donskoy was
forced to pay tribute to the Khan. 31
The struggle between Timur and Toktam'ish was a long and stubborn one. It was only
after three major campaigns, in 1389, 1391and1394-5, that the Golden Horde was finally
crushed. During the last of these campaigns Timur destroyed Astrakhan and other towns
along the Volga, including Berke Saray, the capital of the Golden Horde, and ravaged
the Crimea with fire and the sword. Toktam'ish fled to Bulghar on the Volga and, after
Timur's departure, fought with other contenders for the throne of the Golden Horde until
his death in 1406 or 1407. The defeat inflicted on the Golden Horde by Timur was a blow
from which it never recovered. For the next three decades or thereabouts, trade between
the Mediterranean and Asia was confined to routes passing through Iran, Bukhara and
Samarkand, which were controlled by Timur and the Timurids.
Towards the end of the fourteenth century, the Delhi Sultanate, once a powerful state,
entered a period of decline. Following the death of Sultan Firuz Shah Tughluq (1351-88),
the heirs of the house of Tughluq battled successively for the throne. In 1394 Nasir al-Din
Mahmud was put on the throne by one of the noble factions, but his real power extended
no further than the district round the capital and some adjacent regions (see above, Chapter
14, Part Two). Timur's Indian campaign was heralded by the appearance under the walls of
Multan of the forces commanded by his grandson, Pir Muhammad, who overran and looted
this wealthy city. In September 1398 Timur himself crossed the Indus. Reducing towns and
fortresses to 'heaps of ashes and debris' as they went, his forces headed for the capital,
Delhi. Before the decisive battle on the banks of the Jumna (17 December 1398), Timur
ordered the execution of all prisoners held by his armies - the sources speak of 100,000
captives - fearing that they would side with the Sultan of Delhi during the fighting.
The battle for Delhi was bloody: 'The battlefield was piled high with mountains of
dead and wounded ... blood flowed in streams.' 32 Sultan Nasir al-Din Mahmud fled to
Gujarat. On 18 December the khutba was read out in the mosques of Delhi, mentioning Pir

31
Nasonov, 1940, p. 136.
32 Ghiyath al-Din c AlI, 1915, p. 119.

337
ISBN 978-92-3-103467-1 Contents Timur's military campaigns

Muhammad by name. The inhabitants of the city resisted the intruders, who were looting
and pillaging, seizing prisoners and killing: 'Hindu heads were piled as high as they could
go and their bodies became food for wild animals and birds.' It took several days to escort
the captives out of the city; among them were several thousand master craftsmen, including
stonemasons whom Timur intended to use for the construction of mosques in Samarkand. 33
On 1 January 1399 the warriors began to leave the city. They overwhelmed and pillaged
several further provinces and towns in north-western India, including Mirath (Meerut) and
Kangra. Timur recrossed the Indus in March 1399 and had soon left India behind. As
his vicegerent over Multan, Lahore and Dipalpur he appointed Khidr Khan Sayyid, who
mounted the throne in ruined Delhi in 1414 and founded the short-lived Sayyid dynasty.
While his Indian campaign was in progress, Timur lost many of the cities he had previ-
ously seized, including Baghdad and Mosul. In September 1399 he rode out of Samarkand
at the head of his armies on a new western campaign. Leaving Transoxania by northern
Iran, he led his horde across the Trans-caucasus, ravaged Azerbaijan, Armenia and Geor-
gia and made for Syria, which together with Egypt was ruled by the Mamluk sultans.
Factional strife among the latter prevented the military forces of the two provinces from
uniting, with the result that only the amirs of Syria opposed Timur. They were defeated in
battle near Aleppo, which was captured and laid waste. After taking several more Syrian
cities, Timur laid siege to Damascus in January 1401. Failing to receive assistance from the
Mamluk sultan Faraj (1399-1412), Damascus threw itself on the conqueror's mercy. It was
sacked and set on fire. Timur did not stay long in Syria, but left the province in a ruinous
and devastated condition so that it was several years before the land recovered, while the
Mamluk state became weakened by factional strife between the sultan and the great amirs.
Timur, meanwhile, in 1401 turned eastwards to deal with the Jalayirid capital Baghdad,
besieging the city for six weeks and sacking it savagely. 34
Heading northwards, Timur entered Asia Minor. As a result of the wars of conquest
of the Turkish sultans Murad I (1361-89) and his son Bayazid I (1389-1402), nicknamed
Yild'inm ('the Thunderbolt') because of his lightning military successes, the Ottoman state
had become the most powerful in Anatolia and the Balkans by the beginning of the fif-
teenth century. Bayazid I had completed the subjugation of Macedonia, Bulgaria and Asia
Minor (except Cilicia and the Greek empire of Trebizond), but Timur's hordes, having
been triumphant in Syria, turned northwards, invaded Asia Minor and reached Ankara.
Timur demanded the sultan's submission. In response, Bayazid marched against him. In
the decisive battle of Ankara on 20 July 1402, the betrayal of the sultan by the levies from

33
Ibid., pp. 127-9.
34
Sharaf al-Din YazdI, 1972, pp. 262-3.

338
ISBN 978-92-3-103467-1 Contents Socio-economic conditions under Timur

the former Asia Minor amirates (which had been defeated by the Ottomans and incorpo-
rated into their sultanate) tipped the balance. The Ottoman army was routed and Bayazid
sought safety in flight, but he was taken prisoner together with his two sons. In the spring
of 1403, learning that Timur intended to carry him off to Samarkand, he took his own life.
The invasion by Timur had serious consequences for the Ottoman state. In order to weaken
it, he dismembered it, restoring the independence of seven out of the ten amirates of Asia
Minor, while the territory that remained in the possession of the Ottomans was divided up
among Bayazid's four sons, among whom internecine warfare soon broke out; the fall of
Constantinople to the Ottomans was thus postponed for half a century. 35
Timur's last great military undertaking was to have been his projected campaign against
China, but this was interrupted by the death of the 'conqueror of the world' in 1405,
aged 70.

Socio-economic conditions under Timur

Timur's empire was a conglomeration of states and tribal territories. The peoples inhab-
iting them belonged to a variety of civilizations and represented various stages of socio-
economic evolution. Well-developed relationships of dependence existed side by side with
disintegrating tribal ones; the traditions of centuries-old Islamic Persian government struc-
tures with loose Turco-Mongol systems of state organization in the process of growth;
and regions of ancient urban and farming culture with the nomadic life of the steppe. The
nomadic way of life showed remarkable stability. Clavijo writes:
These folk, who live in tents and other shelters, possess only their tents and they wander
summer and winter over the plains. In summer they go where water is to be found, and they
sow their grain and cotton ... and the king with his whole army wanders the plains summer
and winter alike. 36
Timur, a tribal chieftain, became the ruler of a great empire of which the component parts
possessed no uniformity of economic and cultural life. This in turn prevented the growth of
a centralized system of government and the establishment of effective control by the state
over the territories included within its boundaries. Instead it gave rise to the extremely
widespread practice of bestowal as soyurghals of regions and districts upon members of
the ruling house, the sons and grandsons of Timur, his retainers and military commanders,
as well as those common soldiers who had distinguished themselves in war.
The soyurghal (in Mongol, literally 'gift' in the widest sense of the word, viz. gifts of
clothing, weapons, money, rank, privilege) in the form of a grant of land, in exchange for
35
Petrosyan, 1990, pp. 25-8.
36
Clavijo, 1881 , p. 190.

339
ISBN 978-92-3-103467-1 Contents Socio-economic conditions under Timur

service (often military) to the state, was analogous to the later iqtif of the Mongol period.
From being, in the pre-Mongol period, a non-hereditary 'grant of support' or tenure of land
with the right to appropriate a part of the taxes due to the state, the iqtif no later than the
beginning of the fourteenth century became de facto a military fief, linked to the hereditary
ownership of lands and irrigation installations; the owner enjoyed immunity from taxation.
These rights were enjoyed by the holder of the soyurghal not only de facto but de jure,
that is, with the sanction of the state in fifteenth-century documents, particularly the 1417
charter of the Ak Koyunlu Kara Yusuf given in the Sharaf-nama chronicle. 37
Unfortunately, not a single document attesting to a grant in soyurghal by Timur has been
preserved. Nevertheless, to judge from narrative sources, such grants were quite frequent
in his time and the recipients enjoyed the rights enumerated above. Individuals enjoying
tax exemption were known as tarkhans. Nizam al-Din Shami reports that in 1390-1, at
the time of the campaign in the Dasht-i Kipchak against the Golden Horde, Timur singled
out, favoured and rewarded all those who had displayed bravery and ordered them to be
exempted from taxes. The order was given:

that they be not hindered from approaching his greatness, that they and their children be
immune from prosecution for up to ten misdemeanours, that their horses be not taken for ful-
filment of transport obligations (ulagh) and that they be considered exempted and discharged
of all obligations (takalluj). 38

Yet in spite of the substantial rights enjoyed by the holders of soyurghals, they remained
a form of servile landholding under the aegis of the state, in its capacity as the supreme
proprietor of the lands conveyed. In addition to these lands, the state and the sovereign had
at their disposal state or crown lands (khalisa-yi mamlaka).
In the second half of the fourteenth century and the early part of the fifteenth, appre-
ciable landed properties remained at the disposal of private individuals and represented
their private holdings (Arabic-Persian mulk, pl. amlak). In Transoxania, as in other neigh-
bouring lands, amlak were both peasant (i.e. based on private labour) and landlord hold-
ings. Unlike the soyurghals, the amlak were generally subject to state taxation and were
divided, according to their area, into kharaj (land tax) and cushr. It was only at the end of
the Timurid period that the 'unencumbered' mulk (i.e. the mulk exempt from taxes, known
as mulk-i hurr, mulk-i khalis or mulk-i hurr wa khalis) became widespread. 39
Under Timur, practices of land transfer inherited from the past were preserved, as was
the other form of ownership of the waqf type by the Muslim religious foundations -

37
Petrushevsky, 1949, pp. 233-4.
38
Nizam al-Din ShamI, 1937- 56, p. 123; see also Sharaf al-Din YazdI, 1366/1987, pp. 379-80.
39
Makhmudov, 1966, p. 53.

340
ISBN 978-92-3-103467-1 Contents Socio-economic conditions under Timur

mosques, madrasas, mazars (burial shrines), and so on. At times of political instability,
one of the most widespread types of waqf was the waqf-i awladf (family or hereditary
waqj), when those in whom the waqf was vested collected their share from the property
conveyed as waqf. The only document of Timur's reign that has come down to us, how-
ever, is a waqf conveyance drawn up, in the opinion of its editor, 0. D. Chekhovich, no
later than 1383. Under Timur, the waqfs evidently enjoyed tax exemption, but under his
successors a different practice obtained.
Of peasant raciyyat (land tenure) in the time of Timur, little is known from authentic
sources. From indirect data it may be assumed that there already existed in Transoxania a
rural community, known in the Muslim East by the Arabo-Persian name of jamacat, and
communal institutions. In the waqf deed mentioned above, the donor names, among the
lands bordering on the landed properties being transferred in the waqf, 'the land of a spec-
ified community' (zamfn-i jamac at-i muc ayyan) and 'land at the disposal of a village'. 40
Slave labour for cultivation, as well as in the trades, though practically extinct in the pre-
Mongol period, obtained a new lease of life during the Mongol conquest and the conquest
by Timur. Tens and hundreds of thousands of prisoners (bardas) were enslaved. In India,
in particular, after Timur's capture of Delhi, prisoners continued to be led out of the city
gates for several days: 'Each warrior led out of the city 150 men, women and children,
considering them [his] prisoners, so that the least of the soldiers found himself with 20
captives.' 41
We do not know to what extent the Yasa of Chinggis Khan forbidding the nomads to
leave the il (large tribal grouping) of their leader spread among the settled population of
Transoxania. It is known from Clavijo's journal that, at the passage across the Oxus (Amu
Darya), the frontier guards demanded a certificate showing whence and for what purpose
the traveller was leaving the confines of 'the kingdom of Samarkand'. Entry into the capital
area was free, since Timur made every effort to populate that region. Nevertheless, serfdom,
in the sense of a legally sanctioned peasant status, was unknown in Transoxania, as in
many other countries of the East, possibly because of the absence of private estates of
great landowners. However, the peasants were bound to the land by the burden of taxation.
The basic tax was the land tax, mal or kharaj, but nothing is known of any govern-
ment rescripts fixing its level. A kharaj at the rate of one-third (two dangs, or two parts
in six) was levied, according to Sharaf al-Din Yazdi, in Iraq and Azerbaijan, which in
Timur's time were governed by his son Miranshah. This appears to have been the offi-
cially established rate in Timur's empire. In addition to the kharaj, the peasants paid other

4
41
° Chekhovich, 1951 , p. 59.
Ghiyath al-Din cAlI, 1915, p. 129.

341
ISBN 978-92-3-103467-1 Contents Urban development, crafts and trade

taxes, considered to be formally 'illegal' since they were not prescribed by the sharf a.
Fifteenth-century writers refer to levies by a whole army of officials - the dabitan, the
muhassilan, the darughan, the mushrifan and others. Other taxes that seem to have been
levied were a capitation tax (dud{) and a poll tax (sar-shumar); a particularly heavy burden
was the ulagh (literally 'beast of burden'), the obligation to provide government couriers
with relay horses and also to furnish heavy transport.
Clavijo was witness to the cruel treatment of the population by the knights (caballeros)
accompanying the embassy of the king of Castille:

Those who saw them on the road and realized that they were royal servants, and guessing
that they were the bearers of some royal command, would take to flight, as if pursued by the
devil; and those who were in their shops selling their wares, hid them and also fled, locking
themselves in their houses, and passers-by would say to one another, elchi, which is to say,
envoys, for they knew already that with the coming of envoys black days were in store for
them; and they would flee as if the devil were at their heels .... They act thus not only towards
(foreign) ambassadors, but also whenever anyone is on royal business; for whatever someone
on royal business does, everyone must keep silent, and raise no objection ... ; for that reason
the emperor (Sefior) is so feared throughout the whole country, that it is a marvel. 42

The jurisdiction of the central dfwan (council) extended only over Transoxania. The rulers
appointed by Timur in the conquered regions had their own staff of officials.
Timur could not but be aware that the volume of revenue collected by the state was
directly dependent on the condition of agriculture. This was the reason for the measures
taken during his reign for the repair and construction of irrigation works. Extensive irriga-
tion works were carried out on the Mughan steppe and in the Kabul valley. 43 The fifteenth-
century historian and geographer Hafiz-i Abril gives a list of twenty canals, nine of which
bore the names of high officials of Timur's government, perhaps the builders of the canals
in question.

Urban development, crafts and trade

The huge receipts flowing into Timur's treasury from taxes in Transoxania, the more or
less regular inflow of tribute from the conquered territories governed by his deputies, the
heavy taxes laid upon the subject populations and, finally, the undisguised robbery that
followed his conquests enabled him to spend large sums on the beautification of his cap-
ital, Samarkand, to which were relocated, in addition to material wealth, craftsmen and
scholars, poets and painters. Samarkand became a great centre of craftsmanship and trade,

42
Clavijo, 1881, pp. 189, 190.
43
Bartol'd, 19 14.

342
ISBN 978-92-3-103467-1 Contents Urban development, crafts and trade

medieval learning and culture, and surrounding settlements were drawn upon to enhance
the greatness of the capital. The fortress of Samarkand was completed, palaces surrounded
by immense gardens were erected in the town and its environs and work was started on
the construction of the Shah-i Zinda necropolis, a complex of mausoleums of the feudal
aristocracy and the family of Timur.
Across the whole city was laid 'a very wide street', with shops on either side; at fixed
intervals reservoirs were situated. According to Clavijo, the builders worked in relays, so
that the work went on uninterruptedly day and night:
Some broke down houses [i.e. those situated on the roadway], others levelled the ground, still
others did the building, and all made such a noise day and night that they seemed very devils.
In less than 20 days so much had been done that it was a marvel.

Construction also began on several other cities of Transoxania. In particular, Urgench,

destroyed by Timur in his battle with the Khan of the White Horde and sown with barley,
was restored. In Kish, the mosque and palace of Ak-Saray were built. Clavijo considered
the interior apartments of the palace, finished in gold and azure and faced with glazed tiles,
to be 'astonishing work'. 'Even in Paris,' he adds, 'where there are past masters, this work
would be considered very beautiful.' 44
Essential to the growth and productive activity of the cities was a rich agricultural hin-
terland:
Illis land [writes Clavijo of the region around Samarkand] is rich in all things: grain, fruit
and various meats ... bread is as cheap as it could possibly be, and there is no end to the rice
available. So rich and abundant are the town and its surroundings that one can only marvel. 45

To the city from its hinterland came not only foodstuffs but also industrial crops, especially
cotton, which served as the raw material for weaving, the most important branch of urban
handicrafts. The towns produced a variety of metal articles: weapons, all kinds of imple-
ments, copper vessels, and so on. There was a widely varied production by wood-workers,
leather-workers, potters, jewellers; in the towns lived and worked cotton-carders, stone and
alabaster carvers, carpet-makers, bakers, cooks, confectioners and other tradesmen.
The craftsmen were a heterogeneous group, since many of them hailed from distant
towns and countries and had been brought to Transoxania as captives. The employment
of artisans from a variety of countries for building work contributed in no small degree
to the blending on Transoxanian soil of the artistic and architectural styles of the various
peoples of the East. The bulk of urban craftsmen were free tradesmen. They were orga-
nized in guilds and worked in their own shops, with the help of apprentices and members
44
Clavijo, 1881, pp. 209, 279
45
Ibid., p. 285.

343
ISBN 978-92-3-103467-1 Contents Urban development, crafts and trade

of their families. For heavy or dirty work, the craftsmen might engage ancillary work-
ers. There were also craftsmen of royal workshops (karkhanahs), especially armourers. In
Koksaray, one of the villages on the outskirts of Samarkand, surrounded by a deep ditch
and to all intents and purposes impregnable, Timur kept his treasury. Here 'the emperor
kept about 1,000 captive craftsmen, making armour, helmets, bows and arrows, working
the year round' .46
Samarkand and a series of other Transoxanian towns became major centres of interna-
tional trade. Through them passed the most important trade arteries, linking China and
India with Europe and the Near East. The development of the external caravan trade
involved not only merchants but the Turco-Mongol nomadic and semi-nomadic tribal nobil-
ity and many representatives of the local landowning aristocracy, among them the Muslim
clergy. To Samarkand came 'from Rush [Rus] and Tartary hides and linen, from China
silken stuffs, which in that country are made better than anywhere else; and especially
satins, considered to be the best in the world'; from India were brought spices, nutmeg,
cloves, cinnamon and ginger. If Ghiyath al-Din c Ali is to be believed, the trade routes,
which had previously been 'impassable' because of 'the depredations of thieves and ban-
dits', during Timur's reign were made safe and merchants traversed them freely. 47
Despite the high level of development of their craftsmanship and trade, the cities of
Transoxania remained completely in the power of the ruling lords. These owned houses,
lands, caravanserais and other lucrative property. Shihnas (municipal officials with specific
police powers as well as other functions) were appointed by Timur himself or his local
representatives. The craft and trade guilds, though they took decisions in matters relating to
production and some everyday social questions concerning their members, had no political
rights whatever. Their headmen, who were responsible for the collection of taxes and the
discharge of obligations, were often subjected to physical coercion and other forms of
oppression on the part of the ruling powers. Representative institutions of citizens and a
civil law, which had grown up in Europe, were non-existent in Transoxania. Cities and
country localities alike were handed over by the monarch as gifts to the princes and amirs.
The power of the ruling interests was especially manifested in a variety of levies upon
the trading and artisan population. A special tax levied on craft and trade was the tamgha,
which in the Mongol period came to replace the zakat, the sharfca-sanctioned tax tradi-
tionally levied in Muslim countries on such activities. Tolls were exacted from travelling
merchants at river crossings and in mountains passes; 'a great revenue' was derived from

46
Clavijo, 1881 , pp. 289-90.
47 Ghiyath al-Din cAlI, 1915, p. 28.

344
ISBN 978-92-3-103467-1 Contents Relations with west European rulers

the tolls at the famous 'Iron Gates' on the caravan route from India to Samarkand. 48 The
arbitrariness of the authorities was a source of great distress to the townsfolk. An example
was the laying, by Timur's order, of a commercial highway from one end of Samarkand to
the other. The princes (mfrzas) in charge of the project:
started the works and began to demolish the houses, to whomever they might belong, situated
in the places where the sovereign (Sefior) had ordered the road to be laid; so that the owners,
seeing their houses being demolished, collected up their goods and everything they had and
fled.

The expense of building the trade road was, apparently, laid on the townsfolk themselves.
At all events, Clavijo states that 'the people working here received payment from the
town' .49 As the fancy took him, Timur would force the townsfolk to take part in the tri-
umphal celebrations he held.

Relations with west European rulers

Timur's wars of conquest drew the attention of west European rulers, for they saw of the
'Tartar conqueror' a possible counterpoise to the growing military strength to the Ottoman
Turks, who as early as 1389, after winning a victory on the plain of Kosovo in Serbia,
seized that country along with Bulgaria and were threatening Hungary and Constantino-
ple. The emperor of Byzantium, the Genoese ruler of Galata (the Frankish suburb of Con-
stantinople) and the French king Charles VI Valois appealed to Timur for help against the
Turks. For his part, Timur sent an embassy (its details are unknown) with gifts to Genoa
and Venice. After his victory over Sultan Bayazid In 1402, he informed Charles VI in a
letter of 1 August 1402 of the 'deposition' of the sultan and proposed that merchants from
the realms of both rulers should visit each other's countries, 'for peace is strengthened by
trade' .50 A letter in like vein was addressed by Timur to the English king Henry IV. In
reply, the king wrote that the news of the victory over Bayazid, 'our enemy and yours',
'aroused in us a feeling of great solace and great joy'. However, this correspondence did
not lead to the establishment of systematic diplomatic relations. One reason for this was the
death of Timur and the incipient feudal quarrels among his successors, together, of course,
with the great distances involved.
Timur's warlike activity was attentively observed by Henry Ill, king of Castile, who as
early as 1402 sent envoys to the East with orders to collect information 'about the mores,
customs, religion, laws and strength of these far-off nations'. This embassy was received
48
Clavijo, 1881 , pp. 201, 204-5.
49
Ibid., pp. 278-9.
50
Umnyakov, 1969, pp. 179 et seq.

345
ISBN 978-92-3-103467-1 Contents The succession struggle

by Timur during the festivities held by him to celebrate his victory over Bayazid In 1402.
The amir lavished gifts on the envoys. 'Mahomet al-Kazi', as Clavijo calls Timur's envoy,
was sent back to Spain with them to deliver the letter from 'Lord Tamurbeg'. A return
embassy led by Clavijo left Spain by sea on 25 May 1403 for Constantinople and Trebi-
zond, whence they travelled overland through Iran to Samarkand. On 8 September 1404
they were received by Timur. The envoys were shown marks of respect, but there quickly
came an order for them to quit Timur's court and return to their country. This unexpected
tum of events was brought about by the start of preparations for a new invasion, this time
of China, where Timur's army arrived on 27 November, just a few days after the departure
of the Spaniards from Samarkand. 51
The great army raised by Timur set out, but the severe winter of 1404-5, proved fatal
for the aging conqueror and, on arrival at Utrar in February 1405, he suddenly died.

The succession struggle52

After Timur's death, conflict broke out among the members of his house, the rulers of
the state domains and the provincial governors. They refused to accept Pir Muhammad,
Timur's grandson, who had named himself Timur's successor. The empire, lacking a single
economic base and resting on the sole authority of Timur as military leader and on his
despotic power supported by methods of terror, quickly began to disintegrate. Shah Rukh
(1405-47) declared himself an independent ruler, governing a large domain that included
Khurasan and the Herat region.
A serious rival to Shah Rukh appeared in the person of Khalil Sultan (Timur's grandson
and the son of Miranshah), the former ruler of Tashkent. He had the support of several
powerful amirs and grandees of Samarkand, who handed over to him the keys of the city,
the fortress and Timur's rich treasury. Miranshah, with the intention of supporting Khalil
Sultan, marched on Samarkand from Iran at the head of his army. Khalil Sultan's position
was made more complicated by raids by the Khans of the Golden Horde, who had taken
Khwarazm and marched as far as Balkh. Nevertheless, in February 1406 he managed to
defeat, in a battle fought near Karshi, the army of Pir Muhammad and Ulugh Beg (son of
Shah Rukh), who had concluded an alliance against him. Several months later Pir Muham-
mad died at the hand of an assassin.
Meanwhile, the position of Khalil Sultan in Samarkand had appreciably worsened and
he did not have enough troops to undertake further military conquests. The advancement

51
Umnyakov, 1969, pp. 190-5.
52
For the chronology of the Timurids, see Bosworth, 1996, pp. 270-2.

346
ISBN 978-92-3-103467-1 Contents The succession struggle

of new men under his rule annoyed Timur's amirs. One of them, Khudaydad, took Khalil
Sultan prisoner and occupied Samarkand. Shah Rukh marched on Samarkand. Learning
of his approach, Khudaydad left the city, taking Khalil Sultan with him. On 13 May 1409
Shah Rukh entered Samarkand and dealt severely with the amirs of the warring groups
there. Leaving the city, he appointed Ulugh Beg ruler of Samarkand and another of his sons,
Ibrahim Sultan, as ruler of Balkh. Other Timurid princes also received fiefs in Transoxania
from Shah Rukh. In the spring of 1410, unwilling to recognize the authority of Ulugh Beg,
they raised a rebellion and, to the west of Samarkand near Kizil Arvat, they defeated Ulugh
Beg and his regent, Amir Malik Shah. Returning to Transoxania, Shah Rukh put down the
rebellion. Ulugh Beg (1409-49) became in effect the sole ruler of Transoxania, although at
the demand of Shah Rukh he contributed troops to his campaigns, and on the coinage and
in the khutba Shah Rukh's name appeared along with that of Ulugh Beg.
In western Iran and Azerbaijan, governed in Timur's reign by his son Miranshah (killed
in battle in 1408), the Jalayirids established themselves in power with the help of the Kara
Koyunlu (the 'Black Sheep' Turkmens). But it was only in 1410 that the Jalayirid Sultan
Ahmad died in battle against his former allies. Power over Azerbaijan, Armenia and Arab
Iraq then passed into the hands of the Kara Koyunlu dynasty. Shah Rukh, whose effective
power extended over Khurasan and Gurgan, Mazandaran, Sistan and the region of Herat,
Kandahar and Kabul, waged several wars against the Kara Koyunlu. His next campaign in
1435 brought him victory and Jahanshah Kara Koyunlu (1436-67) acknowledged himself
to be Shah Rukh's vassal.
A series of military operations by Shah Rukh were directed against two nomadic states
that had sprung up to the north and north-west of Timur's disintegrating empire. A con-
siderable threat was posed by the Uzbek nomads, whose state arose out of the fragments
of the Golden and White Hordes; the Uzbeks carried out predatory raids against Transox-
ania, Gurgan and Astarabad. The Chaghatayid Khans of Moghulistan, although occupied
with internal disturbances, strove to take control of Ferghana and Turkistan. From time to
time, the Afghan tribes rebelled; their subjection to Shah Rukh was in many cases purely
nominal.
Shah Rukh's efforts were also directed against the rebellious scions of the house of
Timur, his own grandsons, to whom he had granted fiefs in soyurghal. They were allowed
to keep their own courts, but had to hand over part of their revenues to Shah Rukh and
fulfil certain other feudal obligations on pain of punishment. In 1414-15 Shah Rukh's
grandson Iskandar was deprived of his soyurghal that included Isfahan, Ramadan, Luristan
and Fars; in 1414 Husayn Bayqara was evicted from Qum, Kashan and Rayy. In 1446 Shah

347
ISBN 978-92-3-103467-1 Contents The succession struggle

Rukh mounted what was essentially a punitive expedition against his grandson, Sultan
Muhammad (son of Baysunqur), to whom he had once awarded a large fief.
In his efforts to increase his power, Shah Rukh relied for support mainly on the settled
Tajik Persian landowning nobility, particularly the civil bureaucracy. He had close ties with
the Muslim clergy, particularly the shaykhs of the Sufi order of the Naqshbandiyya. To this,
he owed the reputation of being the ideal Muslim ruler. At the same time he was severe
with his amirs. Some of them, including several of the most prominent, he put to death
for a variety of offences against his authority. The customs and standards embodied in
the Mongol Yasa and observed in Timur's time were, in Shah Rukh's reign, relegated to
oblivion and the law of the sharf a achieved unconditional supremacy. 53
During Shah Rukh's reign, his capital Herat became a great centre of commerce, crafts
and culture. Many outstanding poets and painters, scholars and historians lived and worked
there. Their activities were encouraged by Shah Rukh's son and effective co-ruler, Baysun-
qur (d. 1433), who officially occupied the post of chief wazfr. Through his efforts, the
library was founded there. In it worked scholars and philologists, calligraphers, minia-
turists and bookbinders. Herat was embellished by architectural monuments bearing the
stamp of the influence of Samarkand at the end of Timur's reign. To Herat came not only
merchants from many countries, but also foreign embassies, most notably from China. In
return, in 1420 Shah Rukh sent to China envoys of his own, among whom was the artist
Ghiyath al-Din. His diary of the journey to China was used by the historian Hafiz-i Abril.
In 1441-2 c Abd al-Razzaq Samarqandi went to India on Shah Rukh's orders. In his histor-
ical work he describes the route from Hurmuz to India and gives a vivid account of many
cities in the south Indian state of Vijayanagar. Its ruler sent a return embassy to Herat, with
a view to increasing trade between Transoxania, Khurasan and India.
Under the outward brilliance of the state, however, one can discern clear signs of weak-
ness. Although Shah Rukh's historians describe his reign as a time when the raciyyats
(peasants) were 'freed from cares' and found 'tranquillity', this conflicts with the known
facts, above all the growth of popular rebellion. As early as 1405 a Sarbadar uprising took
place in Sabzavar and was suppressed by Shah Rukh. In the following year, a similar upris-
ing engulfed Mazandaran. The peasants and craftsmen failed to make common cause with
the small local landowners, as had happened in the fourteenth century. The ideological
underpinning of their action was provided by the doctrines of the 'extremist' Shfite sects,
who preached, in particular, the common ownership of land and other property and a Utopia
of social equality. On 21 February 1427, in the Friday mosque of Herat, an attempt was
made on the life of Shah Rukh, who was gravely wounded in the stomach. The attacker was

53
Bartol'd, 1914, p. 32.

348
ISBN 978-92-3-103467-1 Contents The succession struggle

a pupil of Fadl Allah Hurufi, the founder of one of the extremist Shfite sects in Khurasan,
the Huriifiyya; Fadl Allah Hurufi had earlier been banished by Timur to Azerbaijan, where
Timur's son Miranshah had put him to death with his own hands.
Towards the end of Shah Rukh' s reign, his state presented a picture of political fragmen-
tation. In Khurasan alone, says c Abd al-Razzaq Samarqandi, were there 'severalpadishahs';
the padishah (king) of Azerbaijan and both Iraqs, Jahanshah Kara Koyunlu, seized the
region of Astarabad and Sabzavar.

349
ISBN 978-92-3-103467-1 Contents The aftermath of Timur's death

17

THE TIMURID STATES IN THE

FIFTEENTH AND SIXTEENTH
CENTURIES*
R. G. Mukminova

Contents

The aftermath of Timur's death . 350

The rule of Ulugh Beg . . . . . 351
The power struggle following Shah Rukh's death 353
The realm divides . . . . . . . . . 354
Agriculture, livestock and hunting 356
Landownership and taxation 357
Towns, handicrafts and trade 361
Revolts . . . . . . . . . . . 364
The Dasht-i Kipchak at the end of the fifteenth century and the beginning of the sixteenth 364

The aftermath of Timur's death

The great empire of Timur had been maintained by force of arms and by Timur's skill
as a ruler and military leader. With his death in February 1405, the unity of the realm
collapsed. Timur's sons and grandsons who had governed the provinces strove for inde-
pendence; attempts to lead the realm by Timur's grandson Pir Muhammad (designated
by his grandfather as his successor) failed. In 1406 he was killed. The representatives of
former dynasties, overthrown by Timur, were not slow to take advantage of the situation.

* See Map 8.

350
ISBN 978-92-3-103467-1 Contents The rule of Ulugh Beg

As a result of a long struggle in Azerbaijan, Armenia and western Persia, the state of
the Kara Koyunlu (1410-68) was established; its rulers pursued a policy hostile towards
the Timurids. Only later, in 1435, did Shah Rukh make their ruler his vassal (see above,
Chapter 16).
A bitter struggle raged throughout Transoxania. Samarkand was seized by Khalil Sultan
(Timur's grandson and the son of Miranshah; 1405-9), but in fact he did not control even
the whole territory of Transoxania: the steppes and Sawran were in the hands of Amir
Berdi Beg; Tashkent, Khujand and Ferghana were in the hands of Amir Khudaydad. From
late 1405 to early 1406, Khwarazm was ruled by Edigti of the Golden Horde, later by the
Uzbek sultans; Khurasan with its adjacent regions was ruled by Shah Rukh (1405--47), who
had constantly to put down the revolts of governors. In May 1409 he was able to occupy
Samarkand; his nephew, Khalil Sultan, who had been captured by Khudaydad some time
earlier, was made to surrender.

The rule of Ulugh Beg

Shah Rukh handed over the government of Turkistan to Ulugh Beg (1409-49), his eldest
son, then aged 15, and appointed Shah Malik (d. 1426) as his tutor. However, at Ulugh
Beg's insistence, the power-loving Shah Malik was recalled to Herat in 1411. Some time
later, Khwarazm was handed over to him as a soyurghal (land grant). Ulugh Beg's real
name was Muhammad Taraghay, but from his youth he had been called Ulugh Beg (Great
Prince) and it is under this name that he has gone down in history. Like his grandfather,
Ulugh Beg had the title of Gtiregen because he was the Khan's son-in-law and under him
there remained some Chinggisid faineant Khans to whose names labels were issued. The
name of the sovereign Shah Rukh was mentioned in the khutba, and money was also coined
in his name. The most important questions were solved by Ulugh Beg only with the consent
of his father. When, in 1427, he mounted a campaign against the Dasht-i Kipchak and suf-
fered a defeat, Shah Rukh temporarily debarred him from the government of Transoxania.
Nevertheless, Ulugh Beg mainly pursued a policy that was independent of Herat. Accord-
ing to Dawlat Shah, 'Ulugh Beg ruled independently over Samarkand and Transoxania for
forty years.'
Ulugh Beg made a number of attempts to unite the separate areas of the Timurid realm.
He was able to banish Ahmad b. cumar Shaykh b. Timur (who had ruled over Ferghana)
and, at some time, with the assistance of his father, to recall this same Ahmad from
Kashghar (where he had tried to consolidate his power) to Herat. Thus Ulugh Beg assumed
power over Ferghana in 1414 and Kashghar in 1416. He also tried, by diplomatic means,

351
ISBN 978-92-3-103467-1 Contents The rule of Ulugh Beg

but unsuccessfully, to defend his territories from raids by the nomadic tribes. Shortly after
having come to power with Ulugh Beg's support, these tribal rulers became his politi-
cal opponents, not infrequently uniting with forces hostile to both Ulugh Beg and Shah
Rukh. For example, the rulers of Moghulistan, 1 struggling against Ulugh Beg, more than
once formed a unified front with the rulers of Badakhshan, who refused to recognize the
supreme power of Shah Rukh.
In February 1425 Ulugh Beg, at the head of a force assembled at Tashkent, departed
on a campaign against Moghulistan. The campaign was successful and a record of it was
inscribed at the Jilanut gorge. On their way home, the participants in the campaign loaded
special carts with pieces of nephrite which Timur had previously tried to bring back and,
having delivered them to Samarkand, used them for Timur's tomb. But as noted above,
Ulugh Beg was unable to establish good relations with the nomads of the steppe bordering
the northern regions of his domains. Baraq Khan, one of Urus Khan's grandsons, who with
his support had been enthroned in the ulus (domain) of the Dasht-i Kipchak, laid claim to
the towns along the Syr Darya, alleging that before Timur's time these areas had belonged
to JOchi's descendants. Hence in 1427 Ulugh Beg, having consulted with his father and
received his support in the person of his brother Jiiqi (d. 1444-5), embarked on a campaign
to the north. However, at S'ignak his army was crushed. Pursuing the Timurid army, the
conquerors penetrated into Transoxania and ravaged many towns and villages. The defeat
at S'ignak, together with an attempt to close the Samarkand gates to Ulugh Beg and his
temporary removal from power, played a decisive part in his subsequent political activity.
From that time onwards he never again mounted a campaign in person, although troops
belonging to the chiefs of nomadic Uzbeks and amirs of Moghulistan more than once
attacked the areas subjected to him. During Ulugh Beg's last years, the Moghuls were able
to attack Andijan and Kanibadam with impunity.
According to the historian c Abd al-Razzaq Samarqandi, when Shah Rukh handed over
the government of Transoxania to Ulugh Beg, he advised him to protect landholders from
violence and unfairness and to keep the army ready for action, paying the soldiers' wages
on time. Meanwhile, in order to protect villagers and town-dwellers alike, to maintain the
irrigation system, to prevent officials from robbing the subjects and to guard the boundaries
of the realm, it was necessary to establish a strong central government. Despite the attempts
of various amirs to increase their own power, Ulugh Beg's forty-year reign was marked by
a relative stability that contributed to the economic growth of his state.
Ulugh Beg took measures to develop both domestic and foreign trade, with India and
China in particular. His monetary reform of 1428 played a role in encouraging internal

1
The north-eastern part of the Chagatayid ulus, separated from it in the mid-fourteenth century.

352
ISBN 978-92-3-103467-1 Contents The power struggle following Shah Rukh's death

trade. His reform of the coinage and organization of the circulation of copper money
resulted in a ban on the old coins in 1420, and the putting into circulation of new ones;
the weight of the new coins was also altered in order to regulate the circulation of money.
Bukhara was later granted the monopoly of striking copper coins, though at the end of
Ulugh Beg's life, there was a clear decentralization of the minting of coins. 2
It was in the time of Ulugh Beg that the Registan Square, lying at the crossing of
the six main roads through Samarkand, was laid out. Hundreds of local master crafts-
men and workers, as well as those who had been collected by force from all regions
at the time of Timur's reign, erected madrasas, mausoleums (including the mausoleum
of the astronomer Qadi-zade Rumi, d. 1437), trading premises, bath-houses and bridges
at Samarkand, Bukhara and Ghijduwan. On Ulugh Beg's orders, a garden was laid out.
In it there was a pavilion: 'the whole lower part of its walls was made of china. It was
called chfnfkhana. The china was delivered from China by one man.' 3 After 1427 Ulugh
Beg mainly gave himself up to scientific studies. Samarkand became a centre of scien-
tific thought and here, at the observatory built under his guidance, worked Qadi-zade
Rumi, Ghiyath al-Din Jamshid, Mucin al-Din Kashani and cAla' al-Din Abu 'I-Hasan cAla
Qushchi. Literary figures such as Lutfi, Saqqaqi, Khayali and others wrote their works
there - Ulugh Beg was fond of poetry and music.

The power struggle following Shah Rukh's death

After Shah Rukh's death in 1447, the heir apparent Ulugh Beg won the dynastic struggle
but, unable to consolidate his power in Herat, was forced to return to Samarkand. Shah
Rukh's body, brought back by him, was buried in Timur's mausoleum. The power struggle
in Herat contributed to the breaking-off of relations between Ulugh Beg and his son cAbd
al-Latif. In autumn 1449 at Dimashq, near Samarkand, there was a battle between the two
men; Ulugh Beg's army suffered a defeat and, on the orders of cAbd al-Latif, he left for
the pilgrimage to Mecca. On his way, near Samarkand, he was killed on 25 (or 27) October
1449 at the instigation of cAbd al-Latif, who went down in history as a patricide; he himself
was to reign for only six and a half months.
Subsequently, the struggle for succession raged mainly between cAbd Allah b. Ibrlahim
b. Shah Rukh and Abu Sacid (son of Miranshah; 1451-69). In 1451 Abu Sacid, supported
by the forces of Abu '1-Khayr Khan, leader of the nomadic Uzbeks, crushed cAbd Allah's
army and seized Samarkand. Mainly with support from the tribal chiefs, Abu Sacid then

2
Davidovich, 1965, p. 298.
3
Babur, 1905, p. lOa.

353
ISBN 978-92-3-103467-1 Contents The realm divides

struggled against various pretenders, among whom was Sultan Husayn Bayqara, a future
ruler of Khurasan (see below). Abu Sacid tried to consolidate his power through gaining
the support of the religious classes: under his rule, great influence was exerted on state
affairs by the Shaykh al-Islam of Samarkand, who subsequently had to yield his posi-
tion to cubayd Allah Khoja Ahrar (1404-90), leader of the Naqshbandiyya Sufi order and
later a powerful political figure. In 1454 and 1470 Khoja Ahrar organized the defence of
Samarkand; he was the initiator of a peace deal at Shahrukhiyya between Sultan Ahmad,
cumar Shaykh b. Timur and Mahmud Khan b. Yunus Khan (d. 1487), who had all been at
war with each other.
Abu Sacid was able to seize Khurasan twice. After the second conquest in 1458-9,
he made Herat the capital of his realm and assigned Transoxania to his eldest son, Sul-
tan Ahmad (d. 1494). According to the agreement with the ruler of the Kara Koyunlu,
Jahanshah (1436-67), the great central Persian desert (Dasht-i Kavir) was recognized as a
border between their realms. Thus Transoxania and Khurasan with the adjacent provinces
formed the Timurid realm, without Abu Sacid, however, being able to achieve his aim of the
formation of a centralized state. Afraid of increasing the power of the Ak Koyunlu, who had
replaced that of the Kara Koyunlu, Abu Sacid mounted a campaign into Azerbaijan early
in 1469, but this ended in the defeat of his forces, his capture and execution. Subsequently,
some disgraced Timurids sought the protection of the rulers of the Ak Koyunlu.

The realm divides

With Abu Sacid's death, the Timurid realm finally split into two sections, Khurasan and
Transoxania, with their adjacent provinces. Power in Khurasan, with Herat as its capital,
went to Sultan Husayn Bayqara (1469-1506), Mansur's son, and the great-grandson of
cumar Shaykh b. Timur; he ruled the country, with a few intervals, for nearly forty years.
Sultan Ahmad Mirza was considered the nominal governor of Transoxania. For some time,
he was able to subjugate Tashkent, Sayram, Khujand and Ura-ttibe. But though he held
power for twenty-five years, 'his will was in the hands of the Beg' .4 Under him, Samarkand
was not the recognized capital of Transoxania; almost every town with its province had
its own independent governor. Sultan cAli Mirza (killed in 1501) was in Bukhara; Fer-
ghana, with the towns of Andijan, Ush, Marghelan, Isfara, Khujand, Askhi and Kasan, was
assigned to cumar Shaykh, the fourth son of Abu Sacid, and Khujand (attributed by some
historians of that time to Ferghana), was also part of his dominions; Tashkent, with its

4
Babur, 1905, pp. 19a, 24a.

354
ISBN 978-92-3-103467-1 Contents The realm divides

surrounding area, was ruled by the Khan of Moghulistan. The rulers of these provinces
were often at war with each other.
The lands along the upper Oxus (Amu Darya) - Hisar, Termez, Chaghaniyan, Khuttalan,
Qunduz and Badakhshan - were assigned to Mahmud Sultan (d. 1495), another of Abu
Sacid's sons, even during the reign of his father. However, the local amirs of these provinces
also pursued an independent policy. After Sultan Ahmad's death, Mahmud Sultan moved to
Samarkand and reigned there for some five or six months, reportedly attempting to regulate
the collection of taxes and strengthen his army. But with the deaths of Ahmad, Mahmud
and cumar Shaykh, all occurring during the space of a year, civil strife intensified. The
richest amirs tried to make use of the child Timurids, preferring to enthrone the weakest
of them. The young Timurid Baysunqur Mirza's coming to power in Samarkand roused
the governors of other provinces. Sultan All Mirza left Bukhara on a campaign against
Samarkand, but the inhabitants of the city put up a fierce resistance. In the spring of 1497
Baysunqur Mirza engaged in a counter-attack against Bukhara. When Sultan cAll learned
of this action, he went towards Baysunqur Mirza at the head of an army. Meanwhile, the
Andijan Begs decided to take advantage of the situation. Putting the young Babur at the
head of their forces, they set off to conquer Samarkand; having occupied the surrounding
mountains and valleys, they bided their time. Khoja Ahrar's sons, Khojagi Khoja (Khoja
cAbd Allah) and Khoja Yahya, took an active part in the struggle for Samarkand, leading
local groupings competing among themselves for power.
At the end of the fifteenth century, the Tarkhan Begs came to exert a great influence
upon state affairs. Struggling to gain influence over the Timurids, they opposed the old
aristocracy, including Ahmad Khoja Beg (who protected cAllshir Nawa'I when the lat-
ter was in Samarkand), Abu '1-Makarim and Khojagi Khoja b. Khoja Ahrar. In 1496 the
Tarkhan Begs suffered a defeat, but continued to play a decisive part in state affairs until
Shaybani Khan's seizure of power in 1500. The resultant wrecking of the economy and the
destruction of the foundations of the state were noted by contemporaries; starvation and
poverty, and the deaths of many of the poor, contrasted sharply with the nobles' splendid
garments, gold and silver plates and dishes, and seemingly infinite numbers of sheep and
thoroughbred horses.
Internecine wars had been occurring in the south as well, with opposition against Sultan
Mascud, the eldest son of Sultan Mahmud Mirza: his most powerful opponent was Khusraw
Shah, the ruler of Qunduz. Although power was more centralized in Khurasan, Sultan
Husayn Bayqara also had to restrain his sons' aspirations for independence, most notably
his eldest son Bad~ al-Zaman. Objecting to a redistribution of lands in 1497, Badie al-
Zaman came out against his father. Sultan Husayn was compelled to cede Balkh and the

355
ISBN 978-92-3-103467-1 Contents Agriculture, livestock and hunting

territory from the Oxus to the Murghab to him; in those towns (Qunduz, Baghlan, Termez,
Hisar, Qubadiyan, Khuttalan and Badakhshan) controlled by Khusraw Shah, the name of
Badie al-Zaman was included in the khutba and inscribed on coins. 5 In 1498 the other sons
of Sultan Husayn rose against him. A significant role in the restoration of peace between
Sultan Husayn and his sons, Badie al-Zaman in particular, was played by the great author
eAlishir Nawa'i.
On the whole, in the fifteenth century and early in the sixteenth, the weakening of central
power and the breaking down of the realm into separate domains were clearly discernible
processes. The attempts by some Timurids to consolidate the central power had no lasting
success and merely stimulated further internecine warfare which eventually resulted in the
conquest of Transoxania by the Shaybanids.

Agriculture, livestock and hunting

Farming in the Timurid realm, as elsewhere in the East, was based on artificial irrigation.
The main provider of irrigation was the state, so that the proper functioning of the irrigation
system and the good state of farming were dependent on the vitality of the central authority;
with the break-down of the state and the increased frequency of nomadic raids, agriculture
also declined.
The Zarafshan valley was the main farming region of Transoxania. Com, fruit and veg-
etables were also grown in the fields of Ferghana, the Tashkent oasis, the Surkhan Darya
and Kashka Darya regions. Khuttalan (Kulab) and Hisar were famed for their rich har-
vests of corn and melons. Melons from Khwarazm and Bukhara enjoyed special fame.
Wheat, barley, millet and maize were grown and vast areas were sown with luceme. The
manufacture of various cotton fabrics in all areas led to the spread of the cotton plant; a
sixteenth-century document mentions ghuza-puli, monetary taxes on cotton. Suburban gar-
dens sprang up all around Samarkand and Tashkent. Special varieties of grapes, apples and
plums, exported to other regions, were prized. In former times, the 'golden peaches' of
Samarkand had been sent to the Chinese emperor. Plants from which dyes (madder, henna
etc.) were obtained, as well as herbs, also deserve mention. The leaves of the mulberry
tree served as food for silkworms. Among the various branches of stock-raising, sheep-
breeding was especially developed. There were many varieties of sheep; the Hisar meat
and suet sheep enjoyed special renown. Karakul pelts were exported from the Bukhara and
Karshi regions. The great number of horses, used in war, shows the level of horse-breeding;
camel-breeding was also widespread, especially in the Karshi region.

5
Khwandamir, 1857, p. 286.

356
ISBN 978-92-3-103467-1 Contents Landownership and taxation

The Timurids and their retinues took a great interest in hunting. According to Babur, in
the region adjacent to Kabul there lived a detachment of slaves who were hunters, together
with 2-300 of their families; one of Timur's descendants had brought these slaves from the
environs of Multan. Their main task was to care for the hunting birds, of which the most
powerful amirs kept several hundreds. A famous Baqi Tarkhan who 'grew up among the
nobility and rich' kept 700 such birds. In honour of the nobles who set out on the hunt,
ghazals (poetry) were written and inscribed on the kettledrums; Ulugh Beg's setting out on
the hunt is mentioned in the writings of his contemporaries.

Landownership and taxation

Property in land was divided into three main categories: state, waqf and mulk. The revenues
from state lands went to the maintenance of the administrative machinery and the army
and for rewarding amirs for their service, for the most part military; they received lands for
guarding the frontiers, participation in wars and the subjugation of rebellious lords.
Under the Timurids, one of the forms of transferring state lands into conditional grant
lands, the soyurghal, became widespread (see above, Chapter 16). They were granted, as
a rule, to members of the ruling dynasty, statesmen of high rank and military leaders.
Thus Shah Rukh transferred Hisar-i Shadman to Muhammad Jahangir; Kabul, Kandahar
and other provinces of Afghanistan to Muhammad's son, Sind-Qaydu-Bahadur; and the
province of Shiraz to Ibrahim Sultan. He also granted soyurghals to many military lead-
ers. Ulugh Beg gave Balkh as a soyurghal to c Abd al-Latif. Sultan Husayn Bayqara also
widely practised the granting of soyurghals. In almost all cases, such grants included the
instruction to use profits received from a province for the maintenance of the armed forces
and for preserving order. Soyurghals were often transferred from one province to another.
Ulugh Beg, for example, before he received Transoxania, was twice granted a soyurghal
in different provinces of the realm. Some of the holders of soyurghals simultaneously pos-
sessed tarkhan charters, which gave them a right to privileges additional to tax immunity.
At the same time, the older term iqtac continued to be used in the sixteenth and seventeenth
centuries in the realm of the Shaybanids and Astrakhanids with the meaning of soyurghal.
Another conditional land grant was the tuyul. A tuyuldar gained the right to acquire tax
receipts, collected by officials from the population of a locality and having been transferred
into tuyul, but he had no right to govern them. A tuyul was usually allotted to military
leaders in lieu of pay for their services. According to a waqf charter, at the beginning of
the sixteenth century, the right to the revenues from land made into a tuyul belonged to a
group of tax-collectors, the yasakiyan. 6
6
Mukminova, 1966, p. 287.

357
ISBN 978-92-3-103467-1 Contents Landownership and taxation

The personal estates of a sovereign and the members of a ruling dynasty were defined
by the term inju; their vassals and servants, who lived there, were also included in this
category. The profits derived from the inju lands went to the maintenance of the court.
People, with their lands, frequently passed under the patronage of a ruler or high-ranking
amir. Information given by cAlishir Nawa'I shows that in Khurasan in the late fifteenth
century, juridically free people tried to save themselves from extortion and plunder by
entering into inju relationships. It should be noted, however, that the sources seldom define
the Timurids' own domains as being in the category of inju, possibly because a sovereign
and the members of his family came to regard what belonged to the treasury as their own
property. As a result, by the end of the fifteenth century, the difference between state lands
and inju lands had disappeared.
The mulk lands belonged, in the main, to landed proprietors. A certain part of the land,
in the form of small plots, was in the hands of peasants. In the first case, these were the
lands with leaseholders; in the second case, lands directly cultivated by their owners, that is,
mulkdars - in essence, these were two different categories of land. Under the last Timurids,
there was a concentration of mulk plots and villages in the hands of both secular and cler-
ical landowners. Among them was Kokeltash, who had lands even outside the Timurid
realm. cubayd Allah Khoja Ahrar was a large landlord; as a contemporary notes, 'his
lands, fields and estates, as well as herds of livestock and other movable and unmovable
personal property, were extraordinarily great and incalculable'. In the Samarkand region
alone, he possessed 30,000 tanaps of cushr lands. 7
In Transoxania in particular, tarkhan holdings - that is, freedom from taxation for the
holder of mulk lands if he held a tarkhan charter, and the granting to him a number of other
privileges - became widespread in the last quarter of the fifteenth century. A tarkhan was
released from punishment for up to nine offences, and only after the ninth was he pun-
ished. He also enjoyed free access to the Khan. To enjoy the privileges of a tarkhan was
recognized as conferring high status. Many tarkhans were connected with the Timurids
by family ties. Under Sultan Ahmad Mirza, the greatest amir in the state was considered
to be Darwish Muhammad Tarkhan. In 1496 Abd al-c Ali Tarkhan was appointed a hakim
(governor) of Bukhara, with Bukhara, Karakul and Karmina designated as his soyurghal;
the fate of the last Timurids often depended on this 'authoritative, cruel, lewd and haughty'
person. 8 Ultimately, the transfer of state lands into soyurghals and the spread of the immu-
nity of tarkhan holdings led to a loss of revenues to the treasury and to the decentraliza-
tion of the realm. The Timurids' attempt to halt this process of disintegration gave only

7
Ivanov, 1945, pp. 11-13.
8
Babur, 1905, p. 22a.

358
ISBN 978-92-3-103467-1 Contents Landownership and taxation

temporary results; on the whole, the trend of the transformation of conditional land grants
into hereditary holdings continued.
Waqf lands were formed as a result of donations in favour of religious, charitable and
other institutions. In theory, they were considered inviolable and hence, in comparison with
the other forms of landownership, had a more firm juridical basis. They were cultivated by
peasant leaseholders, who paid rent. According to a contract concluded with the mutawallf
of the waqf, tax-farmers received the right to collect the tax over a fixed period; needless
to say, there were many abuses. 9 Property of different types and extents was made into
waqfs. In 1420 Ulugh Beg allocated in favour of the madrasa in Samarkand, built by his
own order, lands, crops and irrigation canals; the income they brought 'exceeds many times
the expenses of the madrasa' .10 According to a document of 1464, half of the village of
Saray Malik - with its adjacent lands, as well as 20 pairs of oxen used for the cultivation
of the land, 32 slaves, 12 camels and other types of property - was transferred to a waqf
for the tomb of Abu Sacid's daughter, Khawand Sultan Bika. Khaja Ahrar transferred to a
waqf about 400 plots, as well as 64 villages and 30 country gardens; and property endowed
by him also included 11 houses with gardens, some tens of trading and handicraft estab-
lishments, bath-houses, and mills in Samarkand, Bukhara, Tashkent and the Kashka Darya
provinces and in Afghanistan.
Peasants constituted a major part of the population; they were usually defined by the
term raciyyat, although they were not a homogeneous socio-economic grouping. The most
widespread form of exploitation of the peasants was metayage, and the typical peasant-
leaseholder was a metayager who cultivated state, waqf and mulk lands belonging to well-
to-do landowners. The cultivation of land on the base of metayage rent (in most cases,
through an intermediary) served a number of peasants as the main, and in some cases the
only, source of income; only a small part of the rural population constituted peasant-owners
of mulk lands. Some land was occupied by nomadic cattle-breeders. The specific character
of their economy demanded annual migrations to seasonal pastures. Close contacts with
the settled population brought about interaction between the two ways of life.
Taxes from the peasants constituted the foundations of the state's income. The chiefland
tax was the kharaj, levied from a part of the harvest mainly in kind, with rates that usually
exceeded the fixed norms. Constant military action absorbed the treasury's receipts. Under
the Timurids, the repeated levy of the kharaj was practised, especially at times when there
were frequent changes of ruler. Thus after Shah Rukh's death, the kharaj was collected
from the population of He rat in favour of Shah Malik' s court, and then again when he was

9
Mukminova, 1966, pp. 46-51.
10 cAbd al-Razzaq SamarqandI, 1941-9, p. 421.

359
ISBN 978-92-3-103467-1 Contents Landownership and taxation

removed in favour of Mirza Ibrahim; kharaj was collected for the third time when Abu
Sacid captured Herat. In 1458, when the power of Mirza Ibrahim was restored for a short
time, he ordered the collection of the kharaj for that year; and in the same year, it was
collected for the second time on the orders of Jahanshah Kara Koyunlu, who had captured
Khurasan. 11
Besides the kharaj, there were many other taxes - 'permanent', 'legal' and 'extraordi-
nary' - which sprang up in periods of decentralization of power. Among them there were:
c awarizat, levied for the defrayment of extraordinary expenses; darughan, a tax in favour
of the head of the local aristocracy; dudf, a hearth tax; sar-shumar, a poll tax; and pfshkash
(formally voluntary, but in fact obligatory), gifts to high-ranking officials. Presentation of
the pfshkash was considered an acknowledgement of supreme power; Sultan Mahmud (the
ruler of Khuttalan) and Saray Khoja Khalil (the ruler of Sali) demonstrated their obedi-
ence to Shah Rukh by sending him pfshkash. There were many taxes designated for the
maintenance of tax-collectors and officials of the taxation apparatus (mushrifan, muhas-
silan, dabitan, etc.), the irrigation administration, messengers and so on, as well as for the
provisioning and regular maintenance of the army, which were levied in some cases even
in times of peace. The enforced conscription of ordinary inhabitants to local levies was
practised. When Babur found he had too few soldiers, he ordered people from the suburbs
of Andijan to be taken, using 'kindness or violence'.
Handicrafts and market trading taxes constituted a significant part of the treasury's
income. According to the author of the sixteenth-century Matlab al-talibfn, the sum of
the state revenues received from Samarkand constituted 1,600,000 tamghas. The tax from
urban handicrafts and urban business, as well as domestic and foreign trade, was defined by
this term tamgha. In the fifteenth century, the tamgha collected from the large trading and
handicraft centres of Transoxania, in particular from Samarkand, brought in good returns
to the treasury. The word tamgha was also used in the collective meaning of various taxes.
At the same time, road taxes and taxes for crossing the rivers had their concrete denomi-
nations: rahdarf, kishtibanf, etc. Mention of the terms baj and rahdarf in one and the same
nishan allows us to see these two taxes as different kinds of road duties.
The tamgha related to taxes was not specified in Muslim law, the sharf a, and the Mus-
lim clergy declared the abolition of the tamgha to be a deed that was pleasing in the sight
of God. In 1460-1, at Khaja Ahrar's request, the tamgha was abolished by Abu Sacid
in Samarkand and Bukhara. In order to encourage the development of handicrafts, and
domestic and foreign trade, the Timurids abolished the tamgha several times; however, the
high returns from the tamgha did not allow them to renounce it wholly and its collection

11
cAbd al-Razzaq SamarqandI, 1941-9, pp. 107-8, 148, 286.

360
ISBN 978-92-3-103467-1 Contents Towns, handicrafts and trade

was always resumed. Collection of the tamgha, as of some other duties, was farmed out
as muqataca (tax-farming), in which the tamgha was usually levied at a rate exceeding
the fixed norm. Some Timurids let influential local lords have the tamgha. For example,
Baqi Beg Chaghaniyani, brother of the independent prince Khusraw Shah (killed in 1506),
received all the tamgha of Kabul, though the sovereign of that province was nominally
considered to be Babur. 12 Part of the above-mentioned taxes was levied in cash. The docu-
ments mention taxes thus levied, judging by their denominations: ghuza pull (cotton tax),
pillya pull (cocoon tax), kuknar pulf (poppy tax), sabzf pulf (carrot tax), kharbuza pulf
(water-melon tax), and so on.
Even in the years of relative centralization of the realm, it was common for the normal
rate of tax-collection to be exceeded. A convincing example is given in c Abd al-Razzaq
Samarqandl's Matlac al-sacdayn: the taxable population was to bring to the treasury one
hen instead of one egg, one sheep instead of one mana of meat, ten manas of barley instead
of one mana, one kharwar instead of one sack of straw. The author notes that, in that year
(1407-8), these products were priced excessively high.

Towns, handicrafts and trade

Progress in the manufacture of handicrafts could be observed in Herat and Samarkand,
the capitals of the two realms (although they were in fact independent from each other).
According to Babur, fifteenth-century Samarkand was a well-built town, with such hand-
icrafts as weaving, garment-making, dyeing, pottery, the processing of metals and paper-
making. Some towns specialized in certain wares. Samarkand, for example, was famous for
its high-quality paper and kermezi velvet, which was exported to other towns and countries.
Babur notes, 'This town has one peculiarity seldom met in other towns: every handicraft
has its own market and they do not mix with each other. This is a good custom.'1 3 Bukhara
was similar. Antony Jenkmson, an English envoy and the representative of a trading com-
pany, wrote in the middle of the sixteenth century: 'every handicraft has its special place
here and a special market'. He reported that in Bukhara there were 'many stone houses,
temples and structures luxuriously built and gilded' . 14 By the end of the fifteenth century
the role of Tashkent had grown significantly and it produced metalware, earthenware, fab-
rics and articles of leather, the development of many of these handicrafts being largely con-
ditioned by trade with the neighbouring steppe nomads. In Shahr-i Sabz, Termez, Andijan,
Khujand and Hisar, a lively trade was carried on, as well as between the towns and local
12
Babur, 1905, p. 159a.
13
Babur, 1905, p. 47b.
14
Jenkinson, 1886.

361
ISBN 978-92-3-103467-1 Contents Towns, handicrafts and trade

villages. Herat, as a capital, played a major role in the political and cultural life of the
area, and handicrafts and trade flourished. According to Babur, in every branch of the arts,
all the residents of Herat sought to perfect their work. Jami, c Alishir Nawa'i and Bihzad
all composed their literary and artistic works in Herat at this time. Meanwhile, the ruler
Husayn Bayqara 'occupied himself only with having a good time, day and night' .15
The urban craftsmen used raw and semi-finished materials delivered from landowners
and the stock-breeding regions. Some kinds of semi-finished products were brought from
more distant countries. The handicrafts of the period were characterized by the division of
labour and a small commodity production which can roughly be divided as follows: articles
intended to satisfy the needs of the numerous local urban inhabitants; articles for sale in
the adjacent villages and nomad encampments; goods for the narrow aristocratic circle
(such as costly fabrics, jewellery, some kinds of weapons and metal goods, rich clothes,
splendid manuscripts, etc.); and products exported to foreign countries as well (fabrics,
writing paper, weapons, ready-made clothes). Under the Timurids, measures were taken
to improve the condition of the roads, and trading houses, bridges and caravanserais were
also built. Shah Rukh considered guaranteeing the safety of the caravan routes as one of
the duties of the owners of soyurghals.
In the fifteenth century, Samarkand was the largest Central Asian trading town and
played a major role in the circulation of goods between East and West. By the ancient
highways of the Great Silk Route came the most varied assortment of goods, such as
articles made by Transoxanian skilled craftsmen from raw and semi-finished materials.
Via Samarkand came goods from other towns: the Bukharan alacha bows (the 'Bukha-
ran yellow bow' is even mentioned in the Buryat epic); wines, considered the strongest in
Transoxania; dried plums, used as a purgative; the Shahr-i Sabz (Kish) unbleached calico;
koshma, a kind of salt known for its specific quality; light-blue and grey astrakhan and
the black astrakhan from Karshi and Karakul; and sal ammoniac and santonica (used as a
medicament) from Tashkent. As in earlier times, sandanachi fabric (senden) was exported
as far as Novgorod. From this point, traders of the Teutonic Knights' Order brought it to the
towns of Europe, where it was considered one of the most serviceable fabrics for common
citizens. Thoroughbred horses, including 'Turkic horses', were sent to Khiva from Tran-
soxania; Ulugh Beg presented a Chinese emperor with 'a black horse having white legs'.
From one country to another, tree seedlings were delivered; both fresh and dried fruit was
exported, with Samarkand apples enjoying particular fame.
Contemporary documents testify to a high level of trade. For example, the largest
towns had sarraf-khanas (houses for the exchange of foreign currency) and the sarrafs

15
Babur, 1905, p. 166a.

362
ISBN 978-92-3-103467-1 Contents Towns, handicrafts and trade

(money-changers) also issued cheques. Relationships were established between the vari-
ous trading companies, and by means of a cheque issued in one town a person could draw
money in another. Subsequently, the Eastern (originally Arabic) word 'cheque' became
widespread in the commercial world of western Europe. There are also references in the
sources to the street of the sarrafs, the Chaharsu-i Sarrafan, and to a mosque and a bath-
house of sarrafs. Although this information relates to Bukhara in the sixteenth century,
such establishments had certainly existed in Bukhara and Samarkand at an earlier period.
According to Badr al-Din Kashmirl, the Bukharan sarraf-khana was 'well-known and
famous'. There existed, moreover, the practice of giving capital for investment (mudaraba)
that was legalized by an act made before the qadf. The Timurids, including Mirza Ulugh
Beg, themselves took part in such trading. Ulugh Beg gave money to merchants who, in
their tum, gave it back to him with a share of their profit. On Ulugh Beg's orders, a car-
avanserai named the Mirza'! was built in Samarkand; the income received from it was
applied to Ulugh Beg's madrasa.
During the fifteenth century, particularly in the first quarter, there were extensive com-
mercial and diplomatic relations with China. Khalil Sultan, Shah Rukh, Ulugh Beg and
even their provincial governors fitted out caravans whose journeys lasted, on average,
some nine months. Silk fabrics including kim-khab (kamka), atlas and taffeta were brought
from China; porcelain, silver, mirrors and paper were also brought from China, though in
Herat and Samarkand in particular, paper of high quality was manufactured locally. 16 Wares
brought to China from adjacent countries were also called 'Chinese'; they included san-
dalwood, Tibetan musk and other wares carried along the Silk Route. Timurid merchants
in their tum dispatched locally made fabrics, horses and camels to China. Babur reports on
the mutual advantages derived from the trade between the two countries. Their respective
rulers were both concerned about political conditions in the neighbouring countries; a letter
of 1412 to Shah Rukh, in which the Chinese emperor advises him to patronize his nephew
Khalil Sultan, shows that the emperor was aware of the succession struggle in the Timurid
realm.
Trade connections with India were also close, according to Babur:

On the way between Hindustan and Khurasan there are two trading towns: one is Kabul,
another is Kandahar. The caravans from Ferghana, Turkistan, Samarkand, Bukhara, Balkh,
Hisar and Badakhshan arrive in Kabul, and caravans from Khurasan arrive in Kandahar. Kabul
province lies halfway between Hindustan and Khurasan. It is a very good trade market. Seven,
eight or ten thousand horses are brought to Kabul every year. Ten, fifteen or twenty thousand
merchants lead caravans from lower Hindustan. From Hindustan there are delivered slaves,

16
Mukminova, 1966, pp. 95-103.

363
ISBN 978-92-3-103467-1 Contents The Dasht-i KYpchak

white fabrics, sugar, dyestuffs and medicaments. Many merchants are not satisfied with the
profit of thirty for ten or forty for ten.

Babur also describes the foreign trade of Kabul: 'In Kabul you can find wares from
Khurasan, Iraq, Rum and China; it is as if it were the entrepot of Hindustan.' 17 From
Kashmir came locally woven cloth (shaki), which subsequently became famed in Europe,
and medicinal herbs; from Azerbaijan, mineral oil; from the Arab countries, prayer rugs,
special towels and toothpicks; from Egypt, fabrics; from Turkey, weapons; and from the
Kazan Khanate, leather. In the fifteenth and sixteenth centuries, merchants from Bukhara,
Samarkand and Herat visited Moscow. Several goods were imported from European coun-
tries: white and coloured farangs, woollen stuffs from England, and spectacles intended
for c Abd al-Rahman Jami, are mentioned in the sources.

Revolts
In the fifteenth century, a number of local revolts took place in Khurasan and Transoxania.
The impact of the Sarbadar movement of the previous century was still felt (see above,
Chapter 16). On the whole, these revolts were distinguished by their religious and mystical
overtones, but they were also connected with the propagation of ideas concerning com-
munal landownership and utopian ideals of social equality. In 1427 a clandestine Huriifii
group organized an attempt on Shah Rukh' s life. Social oppression, the transfer of the right
to land rents from the ruler's treasury to local landed magnates, and the more frequent
raids by nomads, all led to an increase in internal population migrations. Mass riots in
Samarkand, Bukhara, Karshi and Shahr-i Sabz are noted in the sources for the end of the
fifteenth century and the beginning of the sixteenth; but such actions were mainly localized
and dispersed, and were of a spontaneous, non-organized character.

The Dasht-i K"lpchak at the end of the fifteenth century

and the beginning of the sixteenth
The region north of Transoxania, the Dasht-i Kipchak or Kipchak steppe, containing the
realm of Abu '1-Khayr Khan (1428-69), also included towns such as S'ignak, Yasa, Uzgend,
Arkuk and Ak-Kurgan. In this steppe, a leading position among the military-nomadic rul-
ing class was occupied by the chosen Khans, Chinggis Khan's descendants, who controlled
the winter and summer pasture grounds:

17
Babur, 1905, p. 129a.

364
ISBN 978-92-3-103467-1 Contents The Dasht-i KYpchak

In all of these families there are many respectable Khans; every family of the great and emi-
nent descendants of Chinggis Khan had sultans and the most notable of them was called
Khan, that is, the supreme sovereign. 18

A significant role was also played by the tribal nobility of the Begs of non-Chinggisid
origin, who during the fifteenth century sought to control handicraft and trading centres.
At the call of the leader of the ulus, the nomads would depart on campaigns of plunder, a
large part of which came to the sultans and tribal chiefs. The historian Muhammad Salih
describes expensive Altai squirrel and ermine fur coats which the Dasht-i Kipchak amirs
wore at the time of one of the campaigns into Moghulistan. Cattle-breeding was the main
occupation of the Dasht-i Kipchak tribesmen, and wealth was reckoned by the numbers of
sheep and cattle, as well as horses.
From early times, the trade routes connecting Central Asia with eastern Europe - the
Volga Bulghars and their successors and Russia in particular - and the roads to China
passed through the Dasht-i Kipchak. There was consequently a symbiosis between the
nomadic tribes and the neighbouring sedentary peoples who supplied them with numerous
manufactured products. The same demand for handicrafts, and cloth in particular, led the
steppe inhabitants to raid the settled regions when these last were undergoing periods of
internecine war, with the resulting interruption of trade connections.
The population living along the river valleys was occupied in farming, but campaigning
raiders usually went along these valleys and damaged the local economy. There is, for
example, a report of the forced collection of foodstuffs by Shaybani Khan's soldiers among
the inhabitants of the Arkuk fortress in 1509:

In spite of the fact that it was the middle of the winter, the time of the highest price of bread
and scarcity of food products, innumerable forces . . . settled around the area of the fortress
and with haste made their way into the houses of the town for the collecting of foodstuffs and
everything that was necessary for the forces ... 19

Attempts by nomadic forces from the steppes to conquer the settled regions increased under
Muhammad Shaybani, Abu '1-Khayr Khan's grandson (b. in 1451; killed 1510). In many
years of wandering, he and his soldiers had served different rulers, taking advantage of
the conflicts between the rulers of Transoxania, Moghulistan and Khurasan. Sometimes
serving the Samarkand ruler as commander of a hired detachment, sometimes acting as a
vassal of the Khan of Moghulistan, Muhammad Shaybani gradually rose to the position
of a Khan of the 'Turkistan realm'. In 1486, during one of his campaigns, he captured
(and held for a period) some fortresses of Khwarazm. According to an anonymous author,

18
Fadl Allah b. Ruzbihan KhunjI, 1976, p. 22.
19
Fadl Allah b. Ruzbihan KhunjI, 1976, p. 70a.

365
ISBN 978-92-3-103467-1 Contents The Dasht-i KYpchak

Sultan Husayn Bayqara, whose territories included Khwarazm, sent women's veils to the
amirs of Urgench after the defeat.
As the sources attest, numerous tribes took part in the conquest of Transoxania by
Shaybani Khan: they included the Qushchi, Naiman, Uighurs, lchki, Durman, Kayat, Tuman,
Mangh'it, Kungrat, Khitay, Tangut, Tatar, Jalayir, Karluk and Sulduz. According to Muham-
mad Salih, the core of the tribes surrounding Shaybani Khan was formed by the Shaybans,
stemming from the former ulus of Shayban. The number of Uzbek forces who participated
in the campaign against Transoxania reached several thousand. According to Muhammad
Haydar, before the conquest of Bukhara and Samarkand, Shaybani Khan had only 2-3,000
men, but after the occupation of these towns their number rose to 56,000.
After the Shaybanid conquest of the towns of Transoxania, the Timurid amirs are men-
tioned among the forces, though their role was now diminished. The Timurids had been
unable to oppose the forces of the Dasht-i Kipchak rulers, and with internecine warfare
and mutual distrust, no power was able to present a united front against the invaders. Hence
by 1503 Shaybani Khan's forces had seized Samarkand, Bukhara and Tashkent. In 1504
Ferghana and Khusraw Shah's domains were conquered, followed by Khwarazm in 1505.
Husayn Bayqara's death favoured the subsequent success of the Shaybanids. In 1506 Balkh
was occupied and Herat in 1507. Thus the realm of the Shaybanids now included Transox-
ania, the towns of Khwarazm through which passed the trade routes to the lower Volga
and further on to Russia, and Khurasan, with its transit trade routes to western Persia and
north-western India.

366
ISBN 978-92-3-103467-1 Contents Religious groups of the Middle East ...

18

POPULAR MOVEMENTS, RELIGIOUS

TRENDS AND SUFI INFLUENCE ON THE
MASSES IN THE POST-cABBASID
PERIOD
K. A. Nizami

Contents

Religious groups of the Middle East in early Islamic times 367

The rise of Sufism and the Sufi orders in Central Asia 370
Socio-religious and politico-religious movements 376
Sufi orders in India . . . 379
The Mahdaw1 movement 382

Some of the popular movements which made their appearance in the early Islamic period
had political objectives that were camouflaged by religious masks; others had religious
objectives but were expressed through forces of political discontent (for a consideration of
these movements in early Islamic Persia and Transoxania, see above, Chapters 1 and 2).

Religious groups of the Middle East in early Islamic

times
From the point of view of the origin of sects in Islam, the crystallization of religious atti-
tudes, the codification of law and the development of Islamic religious sciences, the Cen-
tral Asian region was of great significance. At least two of the most important collections
of the traditions of the Prophet Muhammad - the Sahfh of al-Bukhari and the Sahfh of
Muslim - were compiled there. One of the most eminent commentators on the Qur' an,

367
ISBN 978-92-3-103467-1 Contents Religious groups of the Middle East ...

al-Zamakhsharl (d. 1143), also produced his work there. As a result of the transmission
of Greek science by the Nestorian Christians, the development of philosophical trends in
Islam and the reaction to Greek thought found an echo in Central Asia; thus the theologian
Fakhr al-Din al-Raz"i worked in Transoxania, Khwarazm and Afghanistan.
With the arrival of the Mongols in Central Asia and Persia, the II Khans in general
became supporters of the sunna once they had converted to Islam. Ghazan Khan
(1295-1304) took a keen interest in the family of the Prophet and made substantial waqf
endowments to help the sayyids (descendants of the Prophet). It is said that clashes between
the two main Sunni schools in the East - the Shaficis and the Hanafis - drove some people
to Shicism, while others became disillusioned with Islam altogether. Disgusted with the
quarrel between some Hanafi scholars and the Shafici Qadi Nizam al-Din cAbd al-Malik
of Maragha at the court of Oljeytti, the Mongol amir Kutlugh Shah is said to have turned
to another Mongol amir and said, 'Why have we abandoned the Yasa of Chinggis Khan
and the religion of our forefathers and accepted this religion of the Arabs which is divided
into so many sects?' 1 It was also during the II Khanid period that the theology of the Ithna
cAsharl (Twelver) Shicites was further elaborated; Nasir al-Din Tusi (d. 1274) and his dis-
ciple al-c Allama al-Hilli (d. 1326) played an important role in the consolidation of Shicite
canonical law.

THE NESTORIAN CHRISTIANS

The Nestorian Church was the main bearer of Christianity in Persia, Afghanistan and Cen-
tral Asia. Its head was known as 'Patriarch of the East' and his followers engaged in
extensive missionary activity in the region. The Nestorians consequently became the most
widespread of all the Christian churches in the East. Under the cAbbasids, the Nestorians
enjoyed protection and peace. Under the Catholicos Timotheus I (780-823), a contem-
porary of Hamn al-Rashid (786-809), their church extended its sphere of influence and
activity. The Nestorian patriarch was given the right of residence in Baghdad, a privilege
denied to the Jacobites. In 762 the Nestorians had moved their see to Baghdad, where
a Christian quarter, known as Dar al-Rum ('abode of the Greeks'), grew up. Under the
Catholicos' jurisdiction there were seven metropolitans, each with two or three bishops
under him, in order to look after the religious needs of the Christians. The patriarch-elect
received his investiture from the cAbbasid caliph, who recognized him as the official head
of all Christians in the caliphate.

1
Hafiz-i Abru, cited in Bausani, 1968, p. 544.

368
ISBN 978-92-3-103467-1 Contents Religious groups of the Middle East ...

The Nestorians were the first to promote Greek science and philosophy by translating
Greek texts into Syriac and then into Arabic. They were also the first to introduce Greek
medicine into Baghdad and most of the personal physicians of the cAbbasid caliphs were
Nestorians. It appears that religious debates took place between Nestorians and Muslims
at the cAbbasid court. Al-Kindi's famous Risala [Treatise] contains an account of a debate
held in about 819, before al-Ma'mun (813-33), on the comparative merits of Islam and
Christianity. The text of an apologia for Christianity delivered by a patriarch of the Nesto-
rians in 781 before al-Mahdi (775-85) has also come down to us. There is a story (totally
unproven and highly improbable) that a certain Ahmad b. cAbd Allah b. Salam translated
the Bible into Arabic during the time of Harlin al-Rashid. Al-Muqtafi (1136-60) granted a
charter of protection to the Nestorians in 1138. The Nestorian Church made notable con-
versions among the Turkish tribes and among the Mongols in the latter's first advances
westwards, and in the late thirteenth century there was even a Mongol Catholicos. The
church generally prospered until the fall of Baghdad to the Mongols in 1258; subsequently,
the Catholicos left Baghdad and settled in northern Iraq. But after 1295, when the Mon-
gol Khan became a Muslim, the Nestorian Church rapidly declined; its missionary activity
among the Mongols, the Turks and the Chinese also came to an end.

MANICHAEISM

Manichaeism, the doctrine professed by Mani, originated in the first half of the third cen-
tury in Mesopotamia, a region where a number of different religions and philosophical
schools, notably Christianity, Judaism and Zoroastrianism, were present. Among the prac-
tical writings of Mani was the Khwastwanift, a handbook of formulas for the confession
of sins, which has come down to us in both Sogdian and Uighur versions from Central
Asia. Al-Yacqubi, al-Tabari, al-Mascudi, al-Biruni and others refer to this faith in their
writings. In the beginning, Islam was tolerant of Manichaeism. Later, cAbbasid persecu-
tion pushed the Manichaeans eastwards, in the direction of Transoxania; hence it was in
Khurasan, Khwarazm and Sogdiana that the Manichaean faith gathered strength, and Sog-
diana became an outpost for the dissemination of Mani's gospel as far as northern China
and Central Asia. The Uighur Turks accordingly adopted Manichaeism as their official
religion in 763, but in 840, with the fall of the Uighur empire, the role of Manichaeism
received a set-back in Central Asia. However, it was to survive alongside Taoism and Bud-
dhism until the fourteenth century.

369
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

ZOROASTRIANISM

Under the c Abbasids, Zoroastrianism (or Mazdaism) continued to survive and it was in
the reign of al-Ma'mun that the most important Pahlavi work on Zoroastrian theology, the
Denkard, was composed. Under the strongly orthodox al-Mutawakkil (847-61), the situa-
tion changed and Zoroastrians were persecuted, so that many of them emigrated to India.
Indeed, the Arab conquest of the seventh century and the subsequent spread of Islam in
Iran adversely affected the fortunes of Zoroastrianism in general. From being in Sasanian
times the proud masters of society, the Zoroastrian priests became inferiors on the defen-
sive. Also, certain aspects of Zoroastrianism were attacked in some of the socio-religious
uprisings in which Islamic and Manichaean elements joined hands.

JUDAISM

Judaism developed freely in the territories under the caliphs and there was a large Jewish
colony in Baghdad. In 1164 Benjamin of Tudela visited this colony, where he noticed
ten rabbinical schools and twenty-three synagogues; the exilarch (leader of the Jewish
community in exile) exercised control over the Jews of the region. When Benjamin was
on his way to an audience with the caliph, a herald marched ahead of him making the
announcement, 'Make way before our lord the son of David.' 2
In the eighth and ninth centuries a new power, that of the Turkic Khazars, rose to promi-
nence north of the Caspian and the Black Sea. Judaism became the official religion of the
Khans and the nobility of this state, but large numbers of Christians (both Greek and Syrian,
including Jacobites) and Muslims also lived there. There were said to be thirty mosques in
the town of Itil. Situated between the worlds of Islam and Christianity, Khazaria developed
an international trade network. Rome, Antioch and Byzantium were the great silk reposi-
tories of the Mediterranean world, but a certain amount of raw silk came from Central Asia
via Khazaria. At these times, all along the Silk Route, Nestorian, Jacobite and Muslim
communities flourished.

The rise of Sufism and the Sufi orders in Central Asia

In the tenth and eleventh centuries, the majority of the Turkic Kaduk and Oghuz peo-
ples were converted to Islam and they founded the Karakhanid and Seljuq empires respec-
tively. During the Karakhanid period, the Hanafi school of law and the Maturldi school
of theology were established in Transoxania, and a new Turkish literature inspired by

2
Benjamin of Tudela, 1841.

370
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

Persian Islamic literature gradually came into being (see above, Chapter 6). The
Karakhanids encouraged the diffusion of Islam from Transoxania into the Tarim basin and
towards the northern steppes. Sufi preachers such as Shaykh Ahmad Ata Yasawi (d. 1166)
played an important role in spreading Islam among the nomadic peoples.
The widespread dislocation - social, moral and political - which followed the irruption
of the Oghuz and Mongol tribes was profoundly disturbing for mystically inclined people
and a trend to transform the individualistic approach of the Sufis into a wider movement
was discernible. In the eyes of the great thirteenth-century poets Sacdi, Rumi and Amir
Khusraw, humanity was suffering: 'Men were multiplying but humanity was languish-
ing,' they lamented. The mystics were determined to resuscitate society, check its moral
degeneration and awaken people's moral sensibilities. They pressed into service the mystic
concept of walayat (spiritual jurisdiction assigned to a mystic deputy) and tried to con-
centrate spiritual and moral forces in order to deliver vast regions, area by area and unit
by unit, from intellectual anarchy and moral chaos. Significantly, the 'extremist' Shicite
Ismacilis had also worked out a programme of sending dtfzs (propagandists) to different
areas specifically assigned to a worker to propagate their dogmas. But while the diffs
worked secretly, the Sufi teachers established open contact with the people. The organi-
zation of silsilas 3 (chains of spiritual authority) in the Central Asian regions transformed
mystic activity into a powerful movement there. The eventual conversion of the Mongol
rulers, as well as the change in the religious demography of the region, was largely due
to the efforts of Sufis belonging to different silsilas - the Khwajagan, the Qadiriyya, the
Kubrawiyya and, at a later date, the Naqshbandiyya. The Chishtiyya - so-named after the
place of origin of its founder in Chisht on the Hari Rud in Afghanistan - developed in India
and played an important role in the cultural life of the country. Some disciples of Shaykh
Shihab al-Din Suhrawardi also came to India from Iraq and organized the order there. It
was probably under Sufi influence that the Hindu Bhakti movement (see below) appeared
in the subcontinent and had an effect on the religious life of the lower classes of society.
Of the saints who played an important role in the religious life of Central Asia, the
names of Shaykhs Abu Sacid b. Abi '1-Khayr, Yusuf Hamadani, Ahmad Ata Yasawi, cAbd
al-Qadir Gilani, Shihab al-Din Bakharzi, Baha' al-Din Naqshband and cubayd Allah Khoja
Ahrar deserve special mention. Naqshbandi pilgrimages - particularly those of Khwaja
Baha' al-Din and Jami - awakened religious feelings and fervour all along the route they
travelled. The sermon meetings of cAbd al-Qadir Gilani attracted large numbers of people
to his mystic fold, while Ahmad Ata Yasawi's Turkish verses became known throughout

3
For the silsilas, see Trimingham, 1971.

371
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

Central Asia, so that the activities of the Sufi saints gave a new impetus to spiritual life in
the region.

SUFI TEACHERS

Abu Sacid b. Abi '1-Khayr (d. 1049) was born at Mayhana, a small town between Abi-
ward and Sarakhs. 4 A contemporary of Abu '1-Qasim al-Qushayri, lbn Sina (Avicenna)
and Sultan Mahmud of Ghazna, he had to elaborate his mystic principles at a time of
socio-religious turmoil. Besides leaders of other faiths - Zoroastrians, Christians and Jews
- he came into conflict with the cuzama' of Nishapur, the followers of the Hanafi and
Shafici schools, the Shicites and the Karrami preachers. The c ulama' of Nishapur allegedly
approached Sultan Mahmud for a religious decree ordering the shaykh and all his followers
to be hanged in the market-place of Nishapur, but they did not succeed. Instances of his
having allegedly converted his Zoroastrian physician and a rich Jew are recorded in his later
hagiographic biography, the Asrar al-tawhfd. Shaykh Abu Sacid kept an open house and an
ever-ready table and entertained people lavishly and on a large scale. He was sometimes in
debt on account of his extravagant expenditure. His visitors included people from all walks
of life, rich and poor, and two Seljuq leaders, Toghril and Chaghr'i Beg, allegedly attributed
their rise to his blessings. The shaykh believed that nothing ensured greater rewards on the
Day of Judgement than bringing happiness to the hearts of men; hence he returned evil
with good and treated friend and foe alike. Fond of samac(mystic songs), the authenticity
of his quatrains - which were very popular in medieval khanaqahs (dervish convents) -
has been questioned, though in Delhi these verses were recited by eminent saints as his
genuine compositions. He was endowed with great intuitive intelligence (firasa), which
helped him to extend his influence over people of all sorts; in India, his views were cited
in mystic circles with great approbation. 5
Yusuf Hamadani (d. 1140), another dynamic Sufi of the period, inspired many founders
of mystic silsilas. The spiritual mentor of Ahmad Ata Yasawi and Khwaja cAbd al-Khaliq
Ghujuwani, it was he who encouraged cAbd al-Qadir Gilani to preach in public. He worked
at Merv and Herat for more than sixty years to propagate Sufi principles, distinguishing
himself in bothfiqh (Islamic law) and philosophy.
Ahmad Yasawi (d. 1167) was the founder-saint of the Yasawiyya. 6 Such was his popu-
larity among the Turks of Central and South-East Asia that people reverentially called him
'Ata' (father). His claim to descent from the cAlid Muhammad b. al-Hanafiyya, although

4
See Nicholson, 1921 , pp. 1-76; Bosworth, 1984, pp. 79-96.
5
See, e.g., Amir Khwurd, 1885, p. 411.
6 EI2 .

372
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

certainly spurious, further added to his prestige. 7 His dfwan (collection of poems) and his
hikam (aphorisms), which breathed inspiring mystic sentiments, were extremely popular
in what is now Uzbekistan; the twelfth-century Sufi poet Farid al-Din cAttar refers to him
as 'Pir-i Turkistan'. 8 His tomb at Yasi was built by Timur, who had great faith in his con-
tinuing spiritual powers. It is said that it was through his long and sincere effort that many
Turks were converted to Islam; according to al-Waciz, he was the chief of their saints and
many of their mashayikh (spiritual leaders) traced their spiritual pedigree to him. 9
cAbd al-Qadir Gilani (d. 1166) was a very prominent figure in the annals of Islamic
mysticism. 10 Endowed with rare powers of persuasion and eloquence, he attracted thou-
sands of people to his sermon meetings, which were held outside the city due to the lack
of space for the huge crowds that flocked to listen to him. According to cAbd al-Haqq,
as many as 400 people would sit in his meetings recording his discourses with pen and
paper. 11 Gilani spoke in both Arabic and Persian. He relied on a madrasa to impart knowl-
edge and a ribat (hospice) to provide spiritual training. In Ghur, Gharchistan, Bamiyan
and Khurasan, where the Karramiyya dominated the religious scene, his teachings paved
the way for the rejection of their allegedly anthropomorphic ideas. He is reported to have
converted large number of Jews and Christians to Islam and to have reformed thousands of
people who held heterodox or heretical views. 12 To strive for the welfare of society was,
in the eyes of the shaykh, a religious and spiritual obligation. 'Whoever fills his stomach
while his neighbour starves is weak in his faith,' he declared. He considered service to
humankind an act of spiritual value and all people to be 'the children of God on earth'. His
philanthropic spirit attains sublimity when he says that he 'would like to close the doors
of hell and open those of paradise to all mankind'. His Ghunya is imbued with a spirit
of deep humanism and concern for society. He gave a wide berth to the rulers of his day;
the Seljuq Sultan Sanjar allegedly offered the province of Sistan to cover the expenses of
Gilani' s khanaqah, but the shaykh refused to accept it. 13
Abu Najib cAbd al-Qahir Suhrawardi (d. 1168), an erudite scholar who for some
time taught at the famous Nizamiyya madrasa of Baghdad, laid the foundations of the
Suhrawardi order and explained in his Adah al-murfdfn the principles which should deter-
mine the teacher-disciple relationship. 14 His nephew Shihab al-Din Suhrawardi (d. 1234),

7
Sharaf al-Din YazdI, 1885- 8, pp. 9-10.
8
cAttar, 1297/1879, pp. 182-3.
9
Fakhr al-Din b. Kashifi al-Waciz, 1911 , p. 9.
10
Nasr (ed.), 1991 , pp. 6-25.
11 cAbd al-Haqq DihlawI, 1892, p. 12.
12
Ibid., p. 13.
13
Ibid., p. 198.
14
See Hasan Miyan, 1911.

373
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

spiritual mentor of the great Persian poet Sacdi, extended the order's sphere of influence.
He was responsible for laying down the principles for the organization of khanaqahs in his
famous practical manual, the cAwarif al-mac arif, which was imparted by Sufi teachers to
those disciples who desired to establish independent ribats and to concentrate on dissemi-
nating Sufi principles; thus the Indian mystic Farid Ganj-i Shakar of Pakpatan taught it to
his disciples and also prepared a summary of it. Suhrawardi's other work, the al-Nasa'im
al-fmaniyya wa-kashf al-fada'ih al-yunaniyya, contains a critique of Greek philosophy.
cAbd al-Qadir Gilani considered him the last of the great thinkers to come from Iraq. 15
Baha' al-Din Naqshband (d. 1390), founder of the Naqshbandi order, was born near
Bukhara. He visited many contemporary centres of culture, including Bukhara, Samarkand
and Nasaf, as an itinerant Sufi. It was due to his efforts that the order eventually became
popular in Central Asia and thousands of people were attracted to his mystic fold; he
advised his followers to act upon the sharf a and to hold fast to the sunna of the Prophet.
His life and teachings are recorded in the Anfs al-talibfn, and according to the Hungarian
traveller Vambery, as late as the nineteenth century people still came from long distances
to visit his tomb.
cubayd Allah Khoja Ahrar (d. 1490), one of the most distinguished saints of the Naqsh-
bandi order, was born at Chach (Tashkent). 16 His teachings had a profound impact on the
people of Central Asia, and the Timurid princes and the Shaybanid Uzbeks, as well as the
common people, held him in high esteem. Babur himself, his maternal uncle, Yunus Khan
Moghul, and others belonged to the order, and Babur turned the saint's Risala-yi Walidiyya
into Turkish verse. Khoja Ahrar believed in combining political prestige with spiritual
excellence. Jami, Mir cAlishir Nawa'i and other eminent figures of the age belonged to the
Naqshbandiyya.
Najm al-Din Kubra (d. 1221) was another seminal figure of the period; his influence
extended from Khwarazm up to the Aral Sea, so that the Kubrawi order became a factor
of great significance in Central Asian life. Born in 1145 in Khiva, Kubra received instruc-
tion in the Suhrawardi order from Shaykh Ruzbihan al-Wazzan al-Misri and cAmmar b.
Yasir al-Bidlisi. At the instance of Riizbihan, he worked in Khwarazm, and on account
of his qualities as a mystic teacher became known as walf tarash ('designer of saints').
He left a number of treatises on mystic principles and practices, the most important being
the Fawa'ih al-jamal wa-fawatih al-jalal, which deals with the ecstatic experiences of
a mystic. 17 Among his successors, the names of Majd al-Din Baghdad!, Baha' al-Din

15
Jami, 1915, p. 420.
16
Jami, 1915, pp. 364-70; Fakhr al-Din b. Kashifi al-Waciz, 1911 , pp. 220 et seq.
17
SeeNajm al-Din Kubra, 1957.

374
ISBN 978-92-3-103467-1 Contents The rise of Sufism and the Sufi orders in Central Asia

Walad, Najm al-Din Daya Razi, Sayf al-Din Bakharzi, cAla' al-Dawla Simnani, Nur al-
Din Isfara'ini, Sacd al-Din Hamuya and Sayyid cAli Hamadani deserve special mention.
Majd al-Din Baghdadi (d. 1219) came from a village in Khurasan, while Baha' al-
Din Walad, the father of Jalal al-Din Rumi, was born at Balkh in c. 1145. The Mongol
threat drove Walad westwards and he travelled extensively until he reached Anatolia. A
compilation of his discourses, the Mac arif, has survived, which reveals many aspects of his
influence on the thought and style of Rumi.
Najm al-Din Daya Razi (d. 1256) came to Khwarazm from his native city of Rayy and
joined the discipline of Najm al-Din Kubra. In 1221 he moved to Asia Minor, from where
he is reported to have undertaken a diplomatic mission for the cAbbasid caliph with the aim
of encouraging various Muslim rulers to put up a concerted resistance against the Mongols.
He is the author of the Bahr al-haqa'iq [Ocean of Divine Realities], a commentary on the
Qur'an, and the Mzrsad al_cibad min al-mabd'a ila 'I mac ad [The Path of God's Servants
from the Beginning until the Return to Him], which deals with mystic ideals and gnosis.
Two other Kubrawi saints who exercised a great influence on contemporary Muslim
society were Sayf al-Din Bakharzi and his grandson Abu '1-Mafakhir Yahya. Bakharzi had
a khanaqah at Bukhara; Berke, the Khan of the Golden Horde, embraced Islam at his hands
and Ibn Battuta heard Persian and Turkish Sufi poems being recited at his khanaqah. Yahya
expounds his mystic principles in his Fusus al-adab [Bezels of Refinement].
cAla' al-Dawla Simnani (d. 1336), a disciple of Nur al-Din Isfara'ini, was born at Sim-
nan in northern Persia in 1261. His khanaqah, which was known as Sufi-abad-i Khudabad,
became a lively centre of mystic activity in the region. Simnani criticized the idea attributed
to Ibn cArabi of wahdat al-wujud ('the unity of being'); however, in his Chihil majlis he
approvingly quotes Sacd al-Din Hamuya' s remark that Ibn cArabi was a 'boundless ocean'.
Besides his own contribution to exegetic studies, Simnani completed the unfinished com-
mentary of Najm al-Din Kubra.
Sacd al-Din Hamuya (d. 1253) and his disciple cAziz al-Din Nasafi were Imami Shfites,
with Hamuya having his khanaqah at Bahrabad in northeastern Persia. 18 Nasafi was the
author of a treatise entitled Insan-i kamil [The Perfect Man].
A branch of the Kubrawiyya even reached India, where it gained fame as the Fir-
dawsiyya. One of Sayf al-Din Bakharzi's successors was Badr al-Din Samarqandi, whose
disciple Najib al-Din Muhammad (d. 1300) migrated to Delhi and set up a Firdawsi centre
there. Shaykh Sharaf al-Din Yahya Maneri (d. 1371) popularized the silsila in Bihar; his
letters contain a lucid exposition of mystic ideas and his Malfuzat [Table-talk] has also
been preserved. Also connected with cAla' al-Dawla Simnani was Sayyid cAli Hamadani

18
Najm al-Din Kubra, 1957, pp. 383-5.

375
ISBN 978-92-3-103467-1 Contents Socio-religious and politico-religious movements

(d. 1385) (see above, Chapters 10 and 15), who arrived in Kashmir in 1380 with several
hundred disciples. His commentary on the Fusus al-hikam of Ibn cArabi, his views about
the obligations of rulers as contained in Dhakhfrat al-mulak and his Risala-i Futuwwatiyya
provide valuable information for students of religious history. Sayyid Muhammad b. cAbd
Allah (d. 1464), known as Nur-bakhsh, a disciple of Hamadani's disciple Khwaja Is'haq
of Khuttalan, founded the Nurbakhshi order, although Timur's son Shah Rukh thrice threw
him into prison for his heterodox views; he claimed to have received the esoteric teach-
ings of cAli through Imam Jacfar al-Sadiq. Among various works, he wrote al-Risa/a al-
JCtiqadiyya in support of his religious concepts. The Nurbakhshiyya was introduced into
Kashmir by Shams al-Din clraqi, who adopted secret methods of propaganda like those of
the Ismacilis. Amir Mirza Haydar Dughlat first tolerated, but later persecuted, the sect for
its heterodox views.

Socio-religious and politico-religious movements

CAYYAR ACTIVITY

Paramilitary vigilante groups known as c ayyars arose in many parts of the Eastern Islamic
lands during this period. Sometimes they appeared as ghazfs (fighters for the faith), at
others they formed a disruptive element in towns and usurped power when the government
was weak. Often they degenerated into bands of robbers who indulged in terrorist activities,
making the richer sections of the population targets of their brigandage. The Qabus-nama
ranks soldiers, cayyars and the people of the bazaar as the last groups of people to possess
futuwwa (chivalry).

The perfect soldier [remarks Kay Kawus] is like the perfect cayyar, but generosity, hospitality,
probity should be greater in a soldier ... obedience and humility are a virtue in a soldier, [but]
a fault in an c ayyar. 19

As noted in Chapter 2 above, Yacqub b. Layth, the founder of the Saffarid dynasty, was
originally an c ayyar. The c ayyars were active in various cities of Central Asia, although
Baghdad and Persian towns like Nishapur were the main centres of their activity; in 1154
the Oghuz rebellion in Khurasan provided the cayyars with an opportunity to ransack and
plunder Nishapur.

19
Kay Kawus b. Iskandar, 195 1, p. 143.

376
ISBN 978-92-3-103467-1 Contents Socio-religious and politico-religious movements

THE SARBADAR UPRISING

During the 11 Khanid period, the Sarbadar uprising created great unrest in Khurasan and
the adjacent regions. Although it was principally a revolt against fiscal burdens, Shfite,
Sufi, Mahdist, Messianic and chiliastic concepts all went into its making. The obligation to
billet officials (nuzul) - which meant the duty to take into one's house military personnel,
officers and their staff, and to feed and entertain them - had been abolished by Ghazan
Khan (1295-1304), but during the time of his descendant Abu Sacid (1316-35) the oblig-
ation reappeared. The uprising of the Sarbadars in Khurasan in 1337 was provoked by the
unbridled licence of a Mongol messenger who stopped for lodgings at a village of Bashtin
and demanded wine and a woman. According to Rashid al-Din Fadl Allah, the obligation
had assumed such dimensions that people purposely kept their houses in a dilapidated state
in order to escape it. The rioters murdered the official and rebellions subsequently broke
out at a number of places. The Sarbadars continued in western Khurasan until 1381.
Two extreme views have been expressed about the role of the Sarbadars. Petrushevsky
considers them a social revolutionary movement, while others see them as 'a robber state'.
They doubtless included elements of both. It is said that some of them made contributions
to the architecture of Sabzavar, including a Friday mosque and a warehouse affording work
for artisans. According to Scarci Amoretti:

The Sarbadar opposition comprehended many kinds of heterogeneous elements; there were
not only peasants and plebeians from the towns, but also members of the local landed gen-
try, who provided the first two military leaders of the movement ... cAbd al-Razzaq and
Mascud ... These non-plebeian and essentially Iranian elements, faithful to the Firdawsian
traditions, did what they could to provide an opposition to the administrative aristocracy. 20

Certain Sufi orders seem to have been linked with the Sarbardar movement, including
the Shaykhiyya and the Jiiriyya in Khurasan. The Shaykhiyya were followers of Shaykh
Khallfa (killed in 1335), who was a Mazandarani by origin and a disciple of cAla' al-
Dawla Simnani. At Sabzavar in Khurasan, Khallfa founded a silsila which many people of
the town joined. The Sunni jurists considered them heretics; they implored the 11 Khan Abu
Sacid to get rid of the shaykh and he was secretly murdered. Hasan Jiirl, who succeeded
Khallfa, was of peasant origin and he gave the movement a more markedly Shfite and
militaristic character. He had large numbers of supporters in Nishapur, Tus, Khabushan,
Abiward and other places and joined forces with the Sarbadars in helping to create the
curious Shfite 'republic of Sabzavar'; but he was arrested in 1338 and died shortly after-
wards (see further, above, Chapters 16 and 17).

20
Amoretti, 1986, pp. 612-13.

377
ISBN 978-92-3-103467-1 Contents Socio-religious and politico-religious movements

Mir Qiwam al-Din Marcashi of Mazandaran and his followers were inspired by the
Shaykhiyya-Jiiriyya tarfqa (Sufi order) to found another order. Hasan Jiiri had granted the
title of 'shaykh' to Marcashi's father, cizz al-Din Sujandi, but the latter died while return-
ing from Sabzavar to Mazandaran and his son succeeded him as head of the Mazandarani
branch of the tarfqa. Marcashi became the head of a mass movement at Amul and founded
a miniature Shfite state; his order, like the Shaykhiyya-Jiiriyya, had definite Shfite over-
tones.

SOME OTHER MOVEMENTS

At some time during the years 1283-9, a maker of sieves, Mahmud by name, appeared in
the village of Tarab (near Bukhara), claiming that he possessed magical powers to cure the
sick and that he received messages from the spirits. Mahmud Tarabi (as he was known)
occupied Bukhara with the help of the peasants and artisans and, according to Juwayni,
his adherents included nobles and learned Muslims; the Chaghatayid Mongol Khans had
to send a force to quell this rebellion (see further, above, Chapter 13). Another movement
which had subversive social ideas was that of the pfr (mystical guide) Yacqub Baghbani
and Shaykh Habib. 'Under a veil of mysticism, coupled with reported stories of miracles
and the apparitions of angels, prophets and saints, there lay concealed', alleges Rashid al-
Din, 'the ancient way of the thinking of Mazdak.' The conspiracy was discovered and its
leaders were executed.
The activities of some heterodox sects and groups in northern India during the four-
teenth century are referred to by Firuz Shah Tughluq (1351-88) in his Futuhat. About the
Shfite heretic groups, he writes:
the Shicites, otherwise called Rajidis, invited people to [their] cult and wrote treatises and
books ... They openly reviled ... the Rightly Guided Caliphs and cA'isha Siddiqa, [and] the
great Sufis ... and committed pederasty and called the Holy Qur' an the Mulhaqat-ic Uthmanr
... When it was established that they were leading others astray, ... I inflicted capital pun-
ishment on them and ... burnt their books in public.

Firuz Shah goes on to refer to antinomians and libertines (ibahatfs), and their alleged inces-
tuous practices; he ordered that their leaders be beheaded. One Ahmad Bihari became the
focus of another heterodox movement and his followers deified him; Firiiz Shah punished
both him and his followers. A man called Rukn declared in Delhi that he was the promised
Mahdi, but was executed along with his followers.
In Persia, the Hurufi and the Nuqtawi sects appeared and attracted many intellectuals. 21
Fadl Allah Astarabadi (d. 1394) founded the Hurufiyya, claiming to be the sahib al-zaman
21
Nizami, 1989, pp. 54-6.

378
ISBN 978-92-3-103467-1 Contents Sufi orders in India

(Lord of the Age), and he held that the word 'Allah' is written on the human face. A number
of Hurufi tracts - the Adam-nama, the Kursf-nama, etc. - reveal the aberrant nature of
their beliefs. Hurufi ideas later reached Turkey and influenced the mystic thought of the
Bektashis. The Nuqtawi sect was founded by Mahmud Pasikhani Gilani (d. 1428). He
considered an atom of dust (nuqta-yi khak) to be the origin and the first element of human
life; all other elements rose out of it. The Nuqtawis looked upon their leader as the Mahdi
and called him insan-i kamil (the Perfect Man). In India a century and more later, Akbar
and Abu '1-Fadl cAllam!, among others, were much impressed by Nuqtawi ideas. 22

Sufi orders in India

Many of the Sufi orders which subsequently flourished in India had connections with
the Central Asian lands. As noted above, the founder of the Chishfi order came from
Afghanistan; the Shattari silsila was brought to India from Persia by Shaykh cAbd Allah
Shattari (d. 1465); and Khwaja Baqi Billah came from Kabul and introduced the Naqsh-
bandi order into India during the closing years of Akbar' s reign.
Mucin al-Din Hasan Sijzi introduced the Chishfi order into India before the Ghurids
occupied the country. 23 He had travelled widely in Central Asian lands and at Kharvan, in
the district of Nishapur, he met Shaykh cuthman and decided to join his discipline. For
twenty years he accompanied him on his arduous mystic journeys, coming into contact
with saints like cAbd al-Qadir Gilani, Najm al-Din Kubra and Najib al-Din cAbd al-Qahir
Suhrawardl. Having roamed all over the Eastern Islamic lands, parts of which had not
yet recovered from the aftermath of the Kara Khitay and Oghuz invasions and were soon
to be ravaged by the Mongols, he turned to India and settled at Ajmer, then a citadel of
Rajput power. His concept of religious devotion and his pantheistic approach established
ideological bridges between Islam and Hinduism; the highest form of devotion in his eyes
was 'to redress the misery of those in distress, to fulfil the needs of the helpless and to feed
the hungry'. He therefore advised his followers to 'develop river-like generosity, sun-like
affection and earth-like hospitality' .24 His profound humanism thus provided the motive
power behind the Chishfi organization.
Five saints of the first cycle of the Chishfi silsila- Qutb al-Din Bakhtiyar Kaki, Farid al-
Din Ganj-i Shakar, Nizam al-Din Awliya', Mucin al-Din Chishfi and Nasir al-Din Chiragh
- carried the message of the order far and near. Kaki was immersed in pantheistic philoso-
phy. His presence in Delhi helped the cultural development of the newly established capital
22
Nizami, 1989, pp. 58-61.
23
SeeNizami, 1961 ,pp.180-219.
24
Amir Khwurd, 1885, pp. 45-6.

379
ISBN 978-92-3-103467-1 Contents Sufi orders in India

of Irano-Turkish power there. Shakar's ribat in Ajodhan- a meeting-place of several trade

routes - attracted high and low alike to his fold and a number of backward tribes of the
region embraced Islam as a result of his teachings. Nizam al-Din Awliya', who lived and
worked in Delhi for more than half a century, spread the Chishti order throughout India
and sent several hundred of his khalifas (deputies) to different parts of the country; his
successors established ribats in Bengal, Gujarat, Malwa and other places. 25 He taught the
sublimation of desire through nourishing cosmic emotion and was a believer in pacificism
and non-violence, preaching patience, tolerance and forgiveness. 26
The Suhrawardi order was established in India by Shaykh Baha' al-Din Zakariyya' of
Multan, a disciple of Shaykh Shihab al-Din Suhrawardi, many of whose disciples had
migrated to India. 27 Shaykh Jalal al-Din Tabrizi, Qadi Hamid al-Din Nagawrl, Sayyid
Nur al-Din Mubarak Ghaznawi, Baha' al-Din Zakariyya', Mawlana Majd al-Din Hajji
and Shaykh Diya' al-Din Rumi were among his celebrated representatives there. Tabrlzi
established his khanaqah in Bengal and attracted non-Muslims to his fold; the devotion
of a section of Hindus to him is clearly evinced in the Sanskrit work the Shakasubho-
daya. Nagawri's erudition and learning were widely recognized in academic circles, but
the ecstatic element so dominated his life that he preferred to pass his time engaging in
mystical songs and dances rather than in attempting the arduous task of organizing a spiri-
tual order. Ghaznawi delivered sermons on the duties of a Muslim ruler at the court of the
Delhi sultan Iltutmish. 28
As mentioned above, Baha' al-Din Zakariyya' (d. 1262) set up the Suhrawardi order
in India and founded a large khanaqah at Multan. His mystic ideology differed from that
of the Chishtis in certain respects. He believed in keeping sufficient material resources
in the khanaqah and also maintained contact with the state; succession in his khanaqah
was determined on the basis of heredity. 29 His descendants Sadr al-Din c Arif and Rukn
al-Din Abu '1-Fat'h increased the area of their influence, but the silsila remained confined
mainly to Panjab and Sind. Sayyid Jalal al-Din Surkh Bukhari (d. 1291) established a
strong Suhrawardi centre at Uchch. A native of Bukhara, he was attracted to Multan by the
reputation of Baha' al-Din Zakariyya'; many tribes of Uchch claim that he was responsible
for their conversion to Islam. His grandson, Sayyid Jalal al-Din Makhdum-i Jahaniyan
(d. 1383), made the Suhrawardi order a factor of great importance in the life of Sind; his

25
Nizami, 1992.
26
BaranI, 1860- 2, p. 344. The shaykh's conversations, Fawacid al-fu'ad, throw light on his religious and
ethical ideals; see Nizam al-Din Awliya', 1992.
27
Nizami, 1961 , pp. 220-9.
28
Ibid., pp. 160-2.
29
Nizami, 1957, pp. 109-49.

380
ISBN 978-92-3-103467-1 Contents Sufi orders in India

Malfazat [Table-talk], contained in the Siraj al-hidaya and Jamtaz-cuzam, throws light on
his mystic ideals and instructional methods.
The Shattari order was introduced into India by Shah cAbd Allah Shattari (d. 1485), a
saint of strange ways. He used to put on royal dress, while a huge retinue of his disciples
followed him in military attire, carrying banners and drums and announcing in every village
and town that they happened to visit, 'Is there anyone who wishes to be shown the way to
God?' He settled and later died at Mandu in Malwa, but his work was continued by his
successors, particularly Muhammad Ghawth of Gwalior, who was held in high esteem by
Babur and Humayun; the latter was introduced to several astrological and occult sciences,
particularly those concerning the heavenly bodies, by Ghawth. He wrote the Jawahir-i
khamsa, Kalfd-i makhzan, Damayir, Basa'ir, Kanz al-tawhfd and Bahr al-hayat, and had
close contact with Hindus; Tansen, the most famous Hindu musician of the age, was among
his disciples. In his Bahr al-hayat, the Persian translation of a Sanskrit work the Amrit
Kund, he presented Hindu mystic concepts in Muslim terminology. 30
The activities of these orders paved the way for the rise of the Hindu Bhakti move-
ment in India during the fifteenth and the sixteenth centuries. 31 Saints of the Bhakti school
visited Sufi centres and exchanged views with them, leading to the emergence of new atti-
tudes. The movement spoke out against the prevailing vices of society, religion and the
state; its devotees emphasized the unity of the godhead, rejected ritual and ecclesiastic for-
malities, threw open the doors of religious education to all irrespective of caste, colour or
creed, and treated all human beings as made of the 'self-same clay'. The opening of the
door of divine communion for all people, particularly those to whom this had long been
denied by the caste-ridden Hindu society, was a revolutionary step imbued with great possi-
bilities for spiritual regeneration. In fact, many of the saints of the Bhakti school came from
the class of weavers, cobblers, tanners, carders, and so on; the poet Kabir was a weaver by
trade, Sa'in a barber, Raidas a leather-worker and Dadii a cotton-cleaner.
Apart from the fact that the Bhakti saints - like Kabir, Nanak, Chaitanya, Ramdas,
Ramananda and Dadu Dayal - had come into contact with the Sufis and were influenced
by their teachings, study of the themes of the literature left by the Bhakti saints reveals the
impact of Persian mystic literature, language and ideas; thus the Bhakti saints adopted Sufi
terminology and drew inspiration from the works of cAttar, Sac di and Rum!. The Guru
Granth of the Sikhs contains scores of Persian and Arabic words, and it seems that Guru
Nanak, the founder of Sikhism, had come to acquire a knowledge of many subtle concepts
of Islamic mysticism, including the teachings of Shaykh Farid al-Din Ganj-i Shakar.

30
Nizami, 1950, pp. 56-70.
31
Nizami, 1985, pp. 295-304.

381
ISBN 978-92-3-103467-1 Contents The Mahdawl movement

The Mahdawl movement

The Mahdawi movement in India was a powerful expression of Messianic expectations.
Sayyid Muhammad of Jawnpur (d. 1504) declared himself to be the promised Mahdi.
Unable to settle permanently at any one place due to the fierce opposition to his views, he
travelled to many areas throughout India and Central Asia, including Kabul, Badakhshan,
Andkhud, Herat, Kandahar, Baluchistan and Makran. He finally settled at Farah in
Afghanistan, where he died, and there sprang up a colony of Mahdawis around his tomb.
Sayyid Muhammad was deeply disturbed at the moral and spiritual decline of Muslim soci-
ety. He preached the rejection of material lust (tark-i dunya); resignation to the will of God
(tawakkul); frequent repetition of the names of Allah (dhikr); and migration (hijrat) from
place to place as an antidote to materialist attractions. The Mahdawis set up their commu-
nity centres, known as da'iras, and carried on their missionary activities with great zeal;
Sayyid Muhammad's successors included Shaykh c Ala'i and Miyan Mustafa. 32

32
Nizami, 1989, pp. 42-51; Qamaruddin, 1985.

382
ISBN 978-92-3-103467-1 Contents SOCIO-ECONOMIC DEVELOPMENT ...

19

SOCIO-ECONOMIC DEVELOPMENT:
FOOD AND CLOTHING IN EASTERN IRAN
AND CENTRAL ASIA
N. Kasai and S. Natsagdorj

Contents

THE EASTERN ISLAMIC LANDS, FROM IRAN TO THE FRONTIERS WITH

CHINA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
Food and diet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
Dress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
MONGOLIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Shelter, crafts and dress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Food and diet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393

383
ISBN 978-92-3-103467-1 Contents Food and diet

Part One

THE EASTERN ISLAMIC LANDS, FROM IRAN TO

THE FRONTIERS WITH CHINA
(N. Kasai)

Information regarding food and clothing can only be gleaned from sporadic mentions in
the Arabic and Persian sources and, for the later period, from details gathered by European
envoys and travellers like Clavijo and Marco Polo.

Food and diet

The accounts of Islamic geographers mention the existence of markets and the varied prod-
ucts of the cities. Khurasan, for example, is described as a region with good weather, fertile
land, green plains, numerous sheep, delicious fruit, and a wide variety of foodstuffs. With
its markets and bazaars of shoemakers, drapers, craftsmen and so on, Nishapur was well
known as a commercial centre and had every sort of fruit, vegetable, cereal, meat and
bread. Tabaran, the district's main town, was famous for its abundance of fruit and cheap
food. Abiward gained recognition for its fertile soil and other blessings and Khawaran was
known for its salted meat. Balkh was considered incomparable among the Persian cities: all
kinds of fruit (including citrus fruit and grapes), cereals such as wheat, rice and barley, and
walnuts, almonds and vegetables were grown there. Thus it was known as the granary of
Khurasan and Khwarazm. Of the adjacent regions, Qumis was noted for its pomegranates,
Damghan for its red apples, Sistan for its meat, fruit and saffron and Tabaristan for its
various kinds of fruit, grains and cereals, its wild birds and seafood, its pickles and jams.
Khwarazm, celebrated for its hospitable people, was also noted for its cuisine. The
region had an abundance of fish oil, nuts, honey, jujube, large raisins, sesame and milk
products (mainly a kind of cheese called rahbfn), meat, frozen fish and a famous variety of
watermelon, which was placed in ice-filled lead containers and exported as far as Baghdad
and the caliphal court. The Khwarazmian baranj melon was famous for its sweetness and

384
ISBN 978-92-3-103467-1 Contents Food and diet

taste. As long as the caliph al-Ma'mun (813-33) lived in Khurasan, these melons were
brought to him by courier. The dried melons of Merv were also well known and were
exported, as were the dried plums of Khwarazm. Transoxania was famed as the only land
that never faced famine. Its cotton, wool and silk cloth, and cloaks made out of fox, sable
and grey squirrel fur, were exported to other countries. The district of Bukhara was so
fertile that 1 acre (0.405 ha) was able to provide the livelihood of an entire family. Much of
its agriculture depended on irrigation, by means of which rice, corn and much cotton was
grown. Its sister city Samarkand was rebuilt and embellished after the Mongol devastations.
The Spanish envoy Clavijo, who visited it at the beginning of the fifteenth century, during
the time of Timur, praised it for its green fields, wide squares and extensive markets, full
of a great variety of foods, including raw and cooked meat.
Staple items in the diet of the ordinary people included bread made with kneaded flour
of wheat, barley, millet, maize or rice. Some parts of Khurasan, especially Merv, were
noted for their bread. In addition to bread made from wheat flour, there was a type using
flour, raisins and a mixture of fruit, that was exported to other regions. The most popular
bread was made from wheat, as recommended by ancient physicians; barley bread was
mainly the food of the deprived, and a sign of poverty.
The most commonly eaten meat in these regions was mutton. Alone or mixed with
other foods, whether fresh or salted, it could be served cooked, grilled or broiled. As eating
meat was not against the law, a great variety of domestic and wild animals were raised. In
Turkistan, fish was a popular dish. Khwarazm was noted for its frozen fish, called sbargh;
Bukhara for its fresh and salted fish; Bayhaq for its plump poultry; Sarakhs for its camel
meat; and Tabaristan for its numerous domestic and predatory animals and its great vari-
ety of poultry, wild birds and fish. The salted fish, rubaytha, was exported from Iraq to
Khurasan. Many people appreciated poultry such as chicken and pigeon; the rich served
broiled pheasants and partridges at their parties. Chicken was also among the medicinal
foods prescribed by doctors. It was quite common to breed hens, pigeons, partridges and
many other birds for their eggs and flesh.
The Turks of the Central Asian steppes ate the meat from their sheep and from the wild
animals they hunted, while the Mongols ate, both cooked and raw, the meat of various
domestic and wild animals, including horses, donkeys, dogs, cats, pigs, wolves, foxes,
snakes and rats. Both the meat and the milk of mares were considered excellent dishes. The
Mongols sometimes cut the flesh of horses and sucked their blood in an emergency. Non-
Muslims killed animals by spearing their chest and shoulders. The Great Khan Ogedey's
enforcement of this law was so severe that no Muslim could slaughter an animal for four
successive years.

385
ISBN 978-92-3-103467-1 Contents Food and diet

A favourite dish consumed by all classes was harfsa; it was prepared with fatty meat,
rice, millet or husked wheat and sugar. Merv was known to have an especially good kind.
Another favourite, surkha-ba, made with meat, crushed grain and vinegar, spread from
Persia to Iraq, where it was popular and known as sikbaj. Bazmaward, another Persian dish,
made with cooked meat, eggs and leeks rolled in a thin sandwich of bread, was supposedly
popularized in Baghdad by the Persian Barmakids. Cheese and other milk products such
as yoghurt were consumed everywhere.
Cooking flavourings and spices included pomegranate juice, sour grapes, dried lemons,
lemon juice, vinegar and sour as well as sweet herbs; these were used in soups, various
kinds of cooked rice and in cooked, grilled and broiled meat. Jams, pastries, candies, dates,
rhubarb, pistachios, shelled almonds, walnuts, seeds and dried fruits were other items that
either accompanied meals or were eaten together with dried nuts and seeds between meals.
Edible earth found in Zuzan, Kuhistan and Nishapur was among the rare and valued prod-
ucts which were exported to distant places and offered at the courts of kings.
Drinking has a long history in Persia and many Persian, Turkish and Mongol rulers
were wine-bibbers. Devout Muslims avoided drinking wine, but medical texts praised its
medicinal use. The Mongol Khans were particularly addicted to drinking wine at festivals
and parties. In 1241 excessive wine caused the death of Ogedey. The 11 Khan Abaqa, who
was mentally disturbed, also drank himself to death. During ceremonies at the court of
Qubilay Khan, a large golden barrel full of wine was placed on a big chair, encircled by
smaller pitchers full of mare's or camel's milk and ordinary wine. Both male and female
guests drank the wine. In the homeland of the Khitay, at the time of the Mongols, a type of
wine was made by fermenting rice and adding seasoning; when heated, it was particularly
intoxicating. At the time of the 11 Khanids, even when they became Muslims, wine was so
widespread among all classes that Ghazan Khan was forced to forbid drinking in public
places. He ordered, 'Whoever is found drunk in cities and bazaars must be arrested and
punished.'
Fuqqac (beer) was made from malt or dried grapes; a non-alcoholic version was pre-
pared from sugar, honey, syrup of sugar and ice. Nabzdh (date wine) was made of dates or
raisins and drunk by all classes in society. According to the historian Ibn Isfandiyar:

In Tabaristan there are colourful wines of yellow, white and red colours as if they were fenu-
greek, ruby or rose-water. They are nutritious, useful, good-smelling and without causing any
headache or troubles after intoxication.

It seems that these drinks were non-alcoholic and made of various plants and fruits of the
reg10n.

386
ISBN 978-92-3-103467-1 Contents Food and diet

Milk, mainly from sheep, cows and camels, was the main drink not only of the nomadic
peoples, but also of villagers and city-dwellers. Mongols and Turkic nomads liked mare's
milk best. Dugh (yoghurt diluted with water) was an everyday drink served with meals.
There were various colourful dishes, fruits and drinks used to welcome guests and messen-
gers on happy occasions and at parties. Accustomed to their own local foods or from fear
of being poisoned, most of the great men took their personal cooks with them on trips and
missions. The Saffarid ruler of Sistan, Yacqub b. Layth, ate plain, ordinary food, mostly
barley bread, leeks, onions, fish and a little rice with a sort of cream or starch jelly. In his
kitchens, 20 sheep were slaughtered every day and put in 5 big copper pots to cook. Yacqiib
himself ate from this food and divided the rest among his servants, army commanders and
friends. It was he who told the envoy of the cAbbasid caliph that while he (Yacqub) was
satisfied with a piece of dried bread and a slice of onion, he would never make compro-
mises where the caliph was concerned. cAmr b. Layth, Yacqub's brother and successor, had
so many kitchen utensils that 300 camels and horses could hardly carry them. But at the
time of his captivity in the camp of Amir Isma.C-i:l Samani, his ration was just a daily piece
of meat which a dog reputedly once snatched out of his hands.
During festivities at the court of Masciid of Ghazna, a vast array of dishes was set out
on tablecloths - meat from birds and wild animals, fish, pickles, jams, thin bread and wine.
The great quantity of food consumed at the court of the Seljuq Sultan Sanjar in Merv
required the Oghuz Turkmens to bring 24,000 sheep to the kitchens every year. When Ibn
Battiita visited the court of the Khans of the Golden Horde in Khwarazm in the fourteenth
century, the dishes were as follows: broiled chickens, crane, young pigeons, a kind of but-
tered bread called kelfcha or kak, and sweet paste; there were also many kinds of fruit, such
as grapes, excellent melons and pomegranates, served in silver as well as gold containers.
Many of the Mongol rulers favoured horse meat, especially chopped tripe, and the whole
head of a sheep. Flesh of predatory animals and birds was another dish of Mongol com-
manders. In the autumn of 1222, when Chinggis Khan spent the winter in Samarkand, his
sons Ogedey and Chaghatay went bird-hunting near the mouth of the Zarafshan and sent
50 camel-loads of various kinds of hunted birds to him every week.
Sufi mystics avoided the usual types of food and drink, and were content with a bare
minimum; they lived mainly on broad beans and bread with salt or olive oil, and, as an
ascetic exercise, went for long periods without eating meat. The diet of the masses was
restricted and often inadequate. Grain and cereals, such as wheat, rice barley, vetch, lentil
and beans, were cooked with the cheap meat of camels or cows and made into soup; broad
beans were also common among the lower classes of society. Barley bread and millet were

387
ISBN 978-92-3-103467-1 Contents Dress

the food of the poor. Dates, aubergines, onions, carrots, potatoes, green vegetables, maize
and millet bread were common foods among villagers.
As to eating habits, most people would simply put out all the kinds of food together
so that family and guests could eat whatever they liked; but in general, there was only
one cooked dish with bread. At court or in hostelries, however, the dishes were not all
put out at once. Instead, a list of available dishes was offered and each person ordered
what he wanted. Upper-class people had their own containers or bowls for eating, while
the common people usually shared a bowl or tray. The 'tablecloth' was a piece of cloth or
leather, or a big copper or wooden tray. The Mongol way of eating was crude. At parties
and receptions everyone was given a slice of meat which was eaten without being cut -
cutting meat was not allowed, even when it was offered to a guest by the host. The left-
over meat was kept, inside a skin, for use at the next meal. Meal times were usually as
follows: breakfast in the morning, lunch at noon and supper in the early evening, to allow
the food to be digested before sleeping. People were advised to eat warm food only twice
a day in order to keep fit and healthy.
Contemporary works on ethics describe the correct manners when eating. They include
washing one's hands and saying bismillah al-hamdu lillah before and after every meal;
stopping eating before one is stuffed; beginning and ending a meal with some salt; taking
small pieces of bread from a dish, taking care to chew them well; not opening one's mouth
wide; not licking one's fingers; and sipping water rather than drinking it all at once. People
were expected to eat in a happy mood, to speak of topics of common interest and to talk
about righteous people, rather than remaining silent. Attending parties without an invita-
tion, going towards the table, looking at friends and at the dishes while eating, leaving the
table before the dishes have been collected, and a host ending his meal before his guests,
were all considered impolite. Regarding the order of courses at the end of a meal, fruit was
eaten first, as physicians considered it better for the digestion, then pastries; and finally
cold water was drunk.

Dress
In this region of vast lands, varied climate and diverse products and ways of living, class
differences had a great effect on people's clothing. Governors, officials, soldiers, aristo-
crats, scholars, judges, craftsmen, farmers, and so on, each had their own dress; religious
affiliation and national and ethnic differences were also a major influence on the type or
style of clothing. However, because of the exiguous sources, especially about the regions
with which we are concerned, a survey of dress is difficult. The works of writers such

388
ISBN 978-92-3-103467-1 Contents Dress

as Dozy regarding Islamic dress are mainly concerned with clothing in the c Abbasid and
Fatimid lands, essentially Iraq, Syria, Egypt, Africa and Spain. Nevertheless, geographers
and travellers of the Eastern Islamic lands give some useful information about styles of
dress in such regions as Khurasan, a flourishing centre of silk, wool and cotton textile
manufacture of the time.
Nishapur was noted for its various kinds of cloth and clothing, exported to far-off
regions, where kings and great men chose them for their attire. The city's products included
white cloth, various kinds of turbans, scarfs, silk undershirts and other types of hair and
cotton cloth which were exported as far as Iraq and Egypt. Most towns had their own spe-
cialities, such as mulham, the half-silk cloth of Merv. Bukhara produced various kinds of
silk cloth and had famous weaving factories. Most of Bukhara's taxes were paid to Bagh-
dad in this cloth. The types of cloth known as Bukharf, which were heavy and strong, were
bought by Arabs in great numbers. Carpets, rugs, silk cloth, bedding, mats and prayer rugs
were exported from Bukhara to other regions, and especially to Iraq. In Khwarazm, sable
and grey squirrel furs, goat skins, carpets, bed coverings, silk, silk caps and cotton robes
were produced in large numbers and the surplus was exported. Samarkand was noted for its
tents, silk, wool and furs; its robes were exported to Turkistan. In Tabaristan, many kinds
of wool, silk, linen and cotton robes were found. Sistanis wore three or four turbans on
their heads in red, yellow, green or white, mostly of silk. The people of Sogdiana wore
skilfully ornamented silk waistcoats and jackets, boots, and hats with sharp brims. In the
tenth century, the people on the northern frontier regions of Transoxania dressed like the
neighbouring Turks.
The oldest piece of silk from this region dating back to the Islamic era, now to be found
in the Louvre in Paris, belongs to the Samanid period and was woven c. 985 for a ruler
in Khurasan. White robes and other silk articles of clothing, together with precious head
coverings, were among the tribute sent from Khurasan to the court of Harlin al-Rashid
(786-809) in Baghdad. The successors of Chinggis Khan wore gold-woven robes; the
Mongols' dress had previously consisted mainly of animal skins. Soon afterwards they
took to a sack-like garment that was loose on the left side; the right side was tied at the
shoulder (see further, Part Two below).
11 Khanid dignitaries in Persia wore furs and leather hats. Ghazan Khan gave orders for
turbans to be worn, on religious grounds, but was unsuccessful in enforcing this measure.
Mongol women wore long trousers under their sack-like garments and tall, basket-like hats
covered with a piece of cloth. In an attempt to curb the nobles' luxurious lifestyle, the
11 Khan Gaykhatu forbade the wearing of gold-woven garments. Nevertheless, dignitaries
wore new robes at parties and festivities. The furs of sables, grey squirrels, ermines and

389
ISBN 978-92-3-103467-1 Contents Shelter, crafts and dress

other animals were essential materials for the garments of the Mongols and the Turks of the
steppes. At his birthday festivities, the Great Khan Qubilay donned gold-woven garments;
20,000 of his courtiers attended the ceremony, wearing golden and brightly coloured gar-
ments made of costly silk ornamented with pearls and gold. Timur, however, wore a plain
silk robe and a long white hat with a Badakhshan ruby on its top, surrounded by precious
pearls and jewels.
During the ninth century, great merchants wore the taylasan (a head-shawl whose end
did not fall below the chin). The lower classes, however, did not wear this garment. Cooks
wore garments resembling boiler-suits, servants carried napkins and towels, water-carriers
wore short trousers, meat-roasters wore a loincloth or napkin, while traders and artisans
wore loose-fitting garments and farmers wore thick cotton dresses and colourful turbans.
Muhtasibs (municipal inspectors) watched over the type and the state of cleanliness of
craftsmen's dress.
The choice of colours for flags, government dress and badges usually had political sig-
nificance. Thus the caliph al-Ma'mun changed the traditional black of the cAbbasids to
the green of the cAlids when he chose cAli: al-Rida as his heir, but later returned to the
cAbbasid black. In these as in other parts of the Islamic world, black garments were tra-
ditionally used for mourning ceremonies, but in some quarters white was the symbol of
mourning. When mourning the death of the cAbbasid caliph al-Qadir (991-1031), Mascud
of Ghazna wore a white robe and turban and all the court retainers and chamberlains also
attended in white dress.

Part Two

MONGOLIA
(S. Natsagdo!J)

Shelter, crafts and dress

The Mongols' nomadic existence necessitated dwellings that could easily be dismantled.
The yurt, a felt-covered framed structure, fulfilled these needs. John of Plano Carpini, who

390
ISBN 978-92-3-103467-1 Contents Shelter, crafts and dress

travelled through the region in the mid-thirteenth century, describes the Mongol yurts as
follows:

Their shelters are round, shaped like tents and made of twigs and thin sticks; at the top in the
centre is a circular window that admits light and lets out the smoke, for there is always a fire
in the middle. The walls and roofs are covered with felt, and the doors are also made of felt.
Some of the shelters are large; others are small, according to the people's wealth or poverty.

Certain yurts, permanently mounted on wagons and unable to be dismantled like the normal
kind, were called ger-tereg, or 'wagon-yurt'. The yurts varied in size: whereas just 1 ox was
needed to transport a small yurt, 3 or more were required for a large one. Khans and noyans
(chiefs) had special wagon-yurts (ord-ger tereg = 'palace wagon-yurt'); measuring up to 9
m in width, they were drawn by 22 oxen.
In addition to the yurt, conventional dwellings and public buildings (temples, monas-
teries, etc.) had been known to the population of Mongolia for some time. These structures
were built of a variety of materials. In the wooded areas of northern Mongolia, for exam-
ple, timber was used, evidence of which is provided by the remains of beams found in
the Kitan settlement in Khentei. In steppe areas, there were frame-and-post structures with
various types of filling, structures of sun-dried brick with foundations and occasionally
walls of stonework. In settlements dating from the thirteenth and fourteenth centuries or
earlier, we find the remains of buildings where a large, fiat, fired grey brick was used in the
construction of extensions. Grey, fiat, broad brick has been found in the Kitan settlements
of Mongolia (tenth and eleventh centuries).
Crafts played a significant part in the economy and culture of the Mongolian nation.
There were several stages in the development of Mongolian crafts; and in the course of
these, Mongolian craftsmen assimilated new methods and a certain amount of specializa-
tion thus took place. Workmen who fashioned a variety of objects from wood and iron
were called darhan. According to the references in The Secret History of the Mongols,
the darhans were divided into tergech, craftsmen specializing in the construction of wag-
ons, and modoch, who specialized in woodwork. Huyagiin darhuul, who specialized in the
manufacture of weapons for the Khans, resided permanently at their courts.
The abundant metal deposits in the Mongol lands encouraged the development of a
knowledge of the physical properties of ores and metals and the skill of processing them
for specific purposes, attested by literary and archaeological finds. The results of the exca-
vations at Karakorum and other Mongol settlements confirm the existence of specifically
Mongol crafts and trades, including the casting of iron and bronze; other workshops were
principally concerned with the production of various types of weaponry. The stamped
bronzeware, silverware, ornaments from horses' harnesses and bronze mirrors found at

391
ISBN 978-92-3-103467-1 Contents Shelter, crafts and dress

Karakorum testify to the skill of the Mongol smiths, metalworkers and jewellers. The high
level of metal production there is confirmed by the technical analysis of some of the articles
found; analysis of samples of white cast iron and steel shows that the white iron melted at
a temperature of 1350 °. Experts believe that it would have been impossible to achieve this
temperature by hand-operated bellows; a motor force would have been necessary, and was
probably provided by water power reaching Karakorum from the Orkhon river.
In terms of their development, Mongol crafts and trades may largely be described as a
dependent sector of the Mongols' economic activity, on the level of cottage industry. Given
their economic system, the herders generally only made products and household items to
meet the vital requirements of their nomadic life. The various products obtained from the
livestock long remained the principal raw material for the craft industries. Sheep's wool
was used to manufacture felt; belts, harnesses, various types of vessels, clothing, headgear
and footwear were made from the skins of domesticated livestock.
The making of clothing required particular skills, such as the ability to process the hides
and manufacture thread and a knowledge of stitching techniques. The Mongols' main item
of clothing was the deli, a robe with seamless shoulders. Mongolian delis of the tenth to
the thirteenth century were very different from the modem versions. Collarless and open
from top to bottom, they wrapped over at breast level and fastened with three clasps on the
right and a single clasp on the left, where they were slit as far up as the sleeve. Married
women wore a kind of kaftan (nemreg) that was extremely wide and slit in front down
to the ground; they also wore a headdress known as a bogtog. Mongols in the thirteenth
century, as now, wore a soft material belt wrapped tightly around the waist. A belt of this
sort served as a kind of unstiffened corset, to make long journeys on the hard Mongol
saddles more bearable and to help riders maintain their posture.
Differences in the finish, style and quality of materials were apparent in the clothing
of the rich and the poor. Rich people wore clothes made of silk and wool and expensive
furs brought from various foreign countries. They lined their robes with silk floss, which
is extremely soft, light and warm. The poor made their heavy outer coats from dog or goat
skins, lining their clothing with linen or cotton. They used felt to make cloaks, saddle-
cloths and rain hats.
The footwear of the Mongol peoples had a number of characteristic features; the cut
and assembly were common to all groups. Mongol boots (tenth to the eleventh century)
had tops which enclosed the entire shin and were the same width at top and bottom. The
sole was thick and inflexible with felt padding. The rigid toe was turned upwards. This
boot was designed specifically for standing in stirrups and riding in a hard saddle at a
quick gallop.

392
ISBN 978-92-3-103467-1 Contents Food and diet

Thus the Mongols' traditional clothing was influenced by the nomadic population's
adaptation to the natural environment. It fulfilled its principal utilitarian function; it was
simple, and it provided excellent protection from the sharp variations in temperature, from
the wind and from the large number of insects which were always present around the
livestock as they grazed.

Food and diet

The basis of the Mongols' diet consisted of milk and meat, but the milk products were
varied in composition and mode of preparation. The milk of all sorts of livestock - cows,
sheep, goats, camels, mares - was used. All types, apart from mare's milk, were boiled in
a cauldron before use. The unboiled mare's milk was turned into koumiss (airak), a fer-
mented, foaming drink, made by whisking the milk for a long period. The various products
obtained from the boiled milk were either consumed fresh or conserved and stored for the
winter. Fresh milk products included taraq (soured milk), byashlaq, a type of cheese made
from unsalted curds, and orume, a thick cream cheese formed by boiling milk for a long
time in a cauldron over a slow fire. Conserved milk products included aaruul (dried curds);
when dried in the form of little balls, it was called grut, which was firmer than aaruul and
could be kept for a long time. Aaruul and grut had the same importance in the life of the
Mongol population as bread has in the lives of farming peoples.
Meat was mainly consumed in winter. The livestock raised by the Mongol peoples may
be divided into two categories: 'livestock with hot breath', i.e horses and sheep; and 'live-
stock with cold breath', i.e. cattle, goats and camels. Meat from all types of domestic
livestock formed part of the diet. In summer and autumn, game such as antelope, Mon-
golian gazelle and Siberian marmot was added. The Mongols preserved meat by drying
it; the dried meat was known as borts. Summer borts differed from winter borts in that it
dried more quickly and turned dark in the sun. Winter sun-dried borts became lighter in
the sun and was more highly valued. Barts was made from all types of meat. According to
The Secret History of the Mongols, an ancestor of the Mongols named Bodonchar prepared
borts from the flesh of wild duck, which he hung on trees. In winter, meat was not only
dried but also frozen. The chilled carcasses were jointed, wrapped in a skin and placed
in a cart, where they were kept throughout the winter. The fresh meat of cattle and other
livestock was boiled for consumption, whereas game was usually roasted. The practice of
cooking the meat of hunted animals goes back, of course, to antiquity. Among the Mon-
gols, the earliest known techniques of food preparation, i.e. roasting over a fire or using red
hot stones, were still in use in the tenth and eleventh centuries. Various seasonings were

393
ISBN 978-92-3-103467-1 Contents Food and diet

used in their food: wild onions, wild roots and grasses, caraway seeds, rhubarb and sarana.
Although the Mongols' diet also included farinaceous foods, these were not as varied as
those found in the diet of settled agricultural peoples.
The nomads' livestock-breeding activities shaped their spiritual life and their ethics.
The various genres of popular oral works formed a major element in the culture of the
Mongolian peoples, reflecting the characteristics of the surrounding landscape, economic
activity and mores of these nomadic herders. Thus Mongolian folklore was deeply rooted
in the life of these pastoral people of the steppes.

394
ISBN 978-92-3-103467-1 Contents CENTRAL ASIA

20
COINAGE AND THE MONETARY SYSTEM
E. A. Davidovich and A. H. Dani

Contents

CENTRAL ASIA 395

Coinage and the circulation of money from the eighth to the tenth century 396
Coinage and the circulation of money from the eleventh century to the beginning of
the thirteenth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
Coinage and the circulation of money under the Mongols (thirteenth and fourteenth
centuries) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
Coinage and the circulation of money in Transoxania under Timur and the Timurids
(late fourteenth and fifteenth centuries) . . . . . . . . 412
AFGHANISTAN, PAKISTAN AND NORTHERN INDIA . . . . . . . . . . . . . . 417

Part One

CENTRAL ASIA
(E. A. Davidovich)

The minting of coins and the circulation of money in the major sub-regions of Central
Asia in all the periods between the eighth and the fifteenth century show both similarities
and differences. Local variations in the composition and supply of currency were due to a
number of factors (economics, politics, traditions, the psychology of the people), among

395
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

which state borders were no longer the most important. For example, even in the relatively
centralized Samanid state (late ninth and tenth centuries) there existed a number of variants
in monetary circulation. The local variants cannot all be compared on equal terms, both
because of constraints of space and because of the differing degrees to which these variants
have been studied. We shall thus concentrate on one sub-region, that of Transoxania. For
other regions, the main differences vis-a-vis Transoxania will be noted.

Coinage and the circulation of money from the eighth

to the tenth century
After the conquest of Central Asia by the Arabs, the local mints started issuing gold, sil-
ver and bronze coins inscribed in Arabic on both sides. These coins came to be known
technically as 'Kufic', from the type of Arabic script used for their legends. Kufic gold
coins (dinars) were not minted regularly or on a large scale in Transoxania. Under the
Tahirids (821-73), dinars were minted periodically in Samarkand and Chach (Tashkent).
Even under the Samanids (tenth century), whose capital was Bukhara, the issue of dinars
in Bukhara, Samarkand and Chach remained only occasional. Samanid dinars were minted
on a markedly larger scale beyond the borders of Transoxania, particularly in Nishapur
and Muhammadiyya (Rayy). Hoards of gold coins found in Transoxania consist mainly of
externally minted Samanid dinars (mostly from Nishapur).
There was clearly more than one legal standard of fineness for dinars. 1 The bulk of the
Samanid dinars from Nishapur are of the highest standard (93-98%, and usually 96% fine
gold). Muhammadiyya issued dinars both of that high standard and of lower standards.
The official weight-standard of the dinar is known (4.26 g) and the mean weight of 4.2 g
corresponds to this. Noteworthy, however, are a significant number of coins that exceed the
weight requirement.
Tenth-century Arab geographers such as al-Istakhri and Ibn Hawqal point out that gold
coins fulfilled different functions in the various regions of the caliphate. In some (e.g. Jibal
and Tabaristan), dinars were the medium of exchange; in others (e.g. Kirman, Fars and
Transoxania), they were not. When describing the money of Bukhara, al-Istakhri notes,
'Dirhams are their coinage; they do not deal among themselves in dinars, which they treat
as goods.' 2 The geographer and traveller Yaqiit explained this matter in more detail for
the enlightenment of his thirteenth-century contemporaries: 'In the time of the Samanids,
the inhabitants of Bukhara used dirhams for trading purposes and did not deal among

1
Ehrenkreutz, 1963, pp. 253-4.
2
Al-IstakhrI, 1927, p. 314.

396
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

themselves in dinars. Gold was just another commodity.' 3 Gold coins were, until the tenth
century, clearly used in Transoxania for rewards or gifts alone. They served as treasure
and universal currency but not as a medium of exchange in domestic trade. This is attested
by the composition of the hoards, the well-preserved state of the coins, the peculiarities
of their real weights and the fact that the local mints of Transoxania issued dinars only
periodically.
Monetary circulation in the territory of Transoxania was marked by the long coex-
istence of two groups of quite dissimilar dirhams: the Kufic and what are termed the
Bukhar Khudat dirhams. The Bukhar Khudat dirhams were modelled on the drachm of
the Sasanian king Bahram V (420-38). The representation in the eighth century of the
monarch's crowned bust (obverse) and fire-altar and two guards (reverse) was quite dif-
ferent from the fifth-century design, the former being schematized and executed in dots
and dashes. The Pahlavi inscriptions on both sides of the coin have disappeared (with only
the rudiments left), but what remains is something that was featured earlier on the obverse
in front of the face of the monarch, namely a Sogdian inscription (three words), and in
one issue it is replaced by a shorter (one-word) inscription. Behind the monarch's head,
the rudiments of a Pahlavi inscription have been changed into four barbed spikes (Fig. 1: 1
- see p. 444 ). In the second half of the eighth century, these rudiments were replaced in
some issues by Arabic inscriptions (Fig. 1:2- 4 ). In the final quarter of that century the
Bukhar Khudat coins were called 'Mahdiyya' after the caliph al-Mahdi (775-85). Man-
uscript sources do not record this name but it occurs in coin inscriptions. The Mahdiyya
dirhams were of a high standard (over 70% silver). All or most were nummi subaerati (fine
silver coating on a copper core). The mean weight (3.2 g) of the eighth-century Bukhar
Khudat dirhams shows that they were minted in accordance with the local weight-standard.
They were issued by three mints: those of Bukhara, Samarkand and Chach. 4
In the eighth century there was an abundance of Kufic dirhams of the Umayyads (Fig.
1:6) and early c Abbasids in the territory of Transoxania, minted in many cities beyond
its borders. The hoards pertaining to the late eighth century include both Bukhar Khudat
and Kufic dirhams. The latter were minted from very high-standard silver and their offi-
cial weight-standard was 2.97 g. An unpublished hoard found near Samarkand contains,
together with Kufic and Bukhar Khudat dirhams, a number of drachms of the Sasanians.
Late Sasanian drachms were probably still to be found in the markets of Transoxania in the
eighth century.

3
Yaqut, 1866- 73, p. 519.
4
Davidovich, 1979, pp. 92-117.

397
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

Fig. 1. 1-5: Bukhar Khudat dirhams, eighth century (obverse, behind the king's head: 1: rudiments
of a Pahlavi inscription; 2: Arabic inscription, 'Khalid'; 3: Arabic inscription: 'Muhammad'; 4:
Arabic inscription, 'Mahdl'; 5: reverse). 6: Kufic dirham from Darabgird, 96/714-15. 7:fals of the
Samanid sovereign Ismacn b. Ahmad, 288/900-1 (enlarged).

398
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

It might have been expected that, in the ninth century, when the local minting of Kufic
dirhams in Samarkand, Chach and Bukhara became fairly regular and abundant, these Mus-
lim coins inscribed with quotations from the Qur' an would have supplanted the Bukhar
Khudat dirhams with their pre-Islamic legends. In fact, the opposite happened since the
position of the Bukhar Khudat dirhams became consolidated in the ninth and tenth cen-
turies. Written sources from those centuries distinguish three types of Bukhar Khudat
coins: Musayyabi, Muhammad! and Ghitrifi dirhams. The tribute received by the caliphate
from various cities and regions of Transoxania was initially reckoned in silver, but in the
first quarter of the ninth century it was recalculated in terms of these dirhams. It is particu-
larly interesting that the recalculation was carried out in a particular form for each region.
The table below summarizes the data provided by Ibn Khurradadhbih for the year 826:

MusayyabI (total annual Muhammad! (total annual Ghitrm (total annual

production: just under production: 1,417 ,OOO) production: 1,189,200)
755,500)
Chach with mines, Sogdiana (Samarkand), Bukhara
Khujand, Usrushana Buttam and Kash with
(partial), Turk cities mines, Nasaf, Ferghana,
on the Syr Darya (partial) Usrushana (partial)

What, then, was the difference between Musayyabi, Muhammad! and Ghitrifi dirhams?
This question has occupied many numismatists and historians, all reckoning the distinc-
tion to lie in the Arabic inscriptions on the obverse, but this has proved a blind alley
for research5 since no account was taken of the pattern of monetary circulation, the clear
information provided by some written sources and the mistakes of others, compared with
numismatic facts. The distinction has proved to lie not in the inscriptions but primarily in
the metal, together with weight, dimensions and appearance. 6
In the last quarter of the eighth century, the Bukhar Khudat dirhams underwent two
local reforms. The first concerned their issue in Samarkand and Bukhara, with a reduction
in their silver content, weight-standard and size. The reformed dirhams - also nummi sub-
aerati but with two coatings -were no more than 40% silver. These were to be the Muham-
mad! dirhams, some issues of which later bore the name 'Muhammadiyya'. The second
reform was conducted in Bukhara alone and legend ascribes it to Ghitrlf b. c Ata', governor
of Khurasan in 792-3. Tradition has it that the inhabitants of Bukhara presented him with
the clear task of giving the market a coinage that would not circulate beyond the city limits.
5 The findings are, not unnaturally, contradictory. Dirhams with the legend 'al-MahdI', for example, have
been identified by some as MusayyabI, by others as Muhammad! and by still others as Ghitrifi.
6
Davidovich, 1966, pp. 49-125, 131.

399
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

This was the origin of the Bukhar Khudat dirhams of copper or non-precious alloys with
no silver at all. These copper coins came to be called Ghitrifi (although this name did not
appear on the actual coins), and in the ninth and tenth centuries they were only minted in
Bukhara. The Musayyabi dirhams, the highest standard of the Bukhar Khudat coins, were
in fact the direct successors of those pre-reform dirhams of the eighth century containing
over 70% silver. The most detailed description of the Muhammad! dirhams is given by
al-Istakhri, according to whom they were minted 'from various metals: from iron, copper,
silver and others'. 7 In other words, the Muhammad! were low-standard silver coins, as con-
firmed by assaying (over 40% silver). Concerning the Ghitrlfi dirhams, all sources agree
that they were made of an alloy of non-precious metals and resembled copper coins minted
from various sorts of copper or from copper with an admixture. 8
Alongside the Bukhar Khudat dirhams, the late eighth century saw the start in Transox-
ania of the systematic issue of locally minted Kufic coins. The Samanids struck particularly
regular and abundant quantities of Kufic dirhams there. These were named Ismacm after
the Samanid amir Ismacn b. Ahmad (892-907). The hoards of Kufic dirhams found in
Transoxania do not just consist of local coins, however; and similarly, the hoards found
beyond its borders in other regions of the Muslim East contain a fair number of dirhams
from Samarkand and Chach. The most important feature of the period, however, was the
fact that the bulk of Kufic dirhams (including Samanid coins) circulated well beyond the
limits of the Muslim East, so that hundreds of hoards of Kufic coins have been discovered
in Europe.
Ismacm dirhams are of a high standard. The real standard has been studied by means
of two methods: atomic absorption spectrometry (6 specimens) 9 and quantitative chem-
ical analysis (104 specimens). 10 Samanid policy regarding the official standard of fine-
ness clearly underwent changes. The silver content of the bulk of dirhams of the early
group (Samarkand and Chach coins) fluctuates within the 89-96% range; lower-standard
(owing to the presence of lead) and higher-standard specimens are exceptions. The official
standard of fineness was no less than 92.5%. Characteristic of the later group is a gen-
eral lowering of the standard, coupled with a wider range of fluctuation and more random

7
Al-IstakhrI, 1927, p. 323.
8
Apart from NarshakhI, who states that the Ghitrlfi dirhams fluctuated between high- and low-standard
silver. But his tenth-century Tafikh-i Bukhara was abridged and revised in 1128 by QubawI, which gave rise
to errors and contradictions; in particular, two reforms of Bukhar Khudat coins in Bukhara were 'telescoped'
by QubawI into one.
9
Lowick, 1975, p. 124.
10
Davidovich, 1966, pp. 132-4.

400
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

deviations from even those limits. The real standard of the late Samanid dirhams has not
been studied, but outwardly they look to be of even lower standard.
The average weight of the early Ismacili dirhams corresponds to the official standard
of 2.97 g, but there was a gradual increase in the range of variations. In addition to whole
coins, the sources make mention of fragments. The composition of the hoards found in
Transoxania is specified as follows: the dirhams were split not only in Samarkand and
not only 'into basic fractions' (i.e. halves and quarters). The hoards contain a great many
fragments of the most varied sizes, shapes and weights, including very small ones. Clearly,
the splitting of Ismacm dirhams into fractions answered two purposes: the pieces stood for
lower denominations than the face value and were added to full dirhams as makeweights,
sometimes making good a substantial difference between their actual and their official
weight. Clearly, too, the question of the circulation of Ismacm dirhams as monetary units
or by weight cannot yet be resolved unequivocally. It is more plausible to speak of the
coexistence of these two forms and the preponderance of one form within different time
spans. In the early decades, dirhams clearly circulated as whole coins (which did not rule
out weight checks, especially for large sums). The criterion of weight became prevalent
later - hence the accumulation of small fragments, which in some hoards far outnumber
whole coins.
The coexistence and differing functions of the four groups of coins designated for sil-
ver circulation (Kufic Ismacm; and Bukhar Khudat Musayyabi, Muhammad! and Ghitrifi)
were determined to a great extent by the financial and fiscal policy of the state, which
rested both on objective market requirements and on a knowledge of the people's psy-
chology. An important component of this policy was the recalculation of the kharaj (land
tax) into Bukhar Khudat coinage and the firm establishment of the extent of the kharaj
in each region of Transoxania, either in Ghitrlfi dirhams (Bukhara), Muhammad! dirhams
(Samarkand, Ferghana, etc.) or Musayyabi dirhams (Chach, Khujand, etc.). Another impor-
tant ingredient was the free exchange of Bukhar Khudat for Kufic dirhams. The exchange
value of the Bukhar Khudat coins fluctuated, but by the ninth century it was usually higher
than that of the Kufic dirhams. In 835, for instance, 100 fine-silver Kufic dirhams weighed
out were worth at most 85 Ghitrlfi dirhams despite the fact that the latter contained no silver
at all. The rate for Muhammad! and Musayyabi coins was higher still. The public consid-
ered the Bukhar Khudat coins to be more stable, less susceptible to market fluctuations,
and therefore preferred them for domestic trading purposes to the Kufic dirhams.
A certain demarcation of functions occurred that provided the best response to both the
domestic and the external trading requirements of Transoxania. The Kufic dirhams fulfilled
all monetary functions, but the universal currency function prevailed and, as a result, the

401
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

bulk of them ended up in Europe, beyond the bounds of Central Asia. The Bukhar Khudat
coins were not involved in this flow, which would have hurt the local population (since their
exchange value exceeded the silver rate), and they would in any case have been unusable for
trading with Europe on account of their metal content. In other words, the Bukhar Khudat
dirhams did not possess inherently the function of a universal currency but served the needs
of domestic commerce, the overriding function being that of a medium of exchange.
Under the Samanids, outsize Kufic dirhams were minted beyond the bounds ofTransox-
ania. They differed from Ismacn1 dirhams in weight, thickness, size, standard of fineness,
and aspect. Their weight was very variable (usually falling within the 10-13 g range) 11 and
they were struck from an alloy of silver (about 70%) and copper. 12 With the large size of
the coins, the inscriptions occupy only part of their field, leaving a free strip round the rims.
These outsize coins occur in hoards found in the territory of Transoxania, which means that
they entered into the province's monetary circulation.
A considerable place in monetary trading was occupied by copper coins (julas, sing.
fals), which were issued in markedly increased numbers under the Samanids (Fig. 1:7). In
retail trading,fuliis predominated, as Ibn Hawqal (mid-tenth century) particularly empha-
sizes. Under the Samanids, fulas were as a rule issued in two denominations, the basic
unit being called an c adlf, which divided into pashfz. Copper coins were struck in many
cities of Transoxania, being intended for trade within the cities and the province, but the
composition of hoards indicates thatfulas crossed the borders of their regions. At the same
time, a particular feature of the circulation of copper coins was that local fulas and those
from other cities (whether always or at specific times is not known) were at different pari-
ties. In Bukhara in 921, for example, a fine-silver dirham fetched 24 Bukharafuliis but 36
Samarkandfuliis. In other words, in Bukhara afals from another city was worth two-thirds
of the localfals. 13
The circulation of money in other sub-regions of Central Asia differed from the variants
that had arisen in Transoxania. The main difference lay in the Bukhar Khudat dirhams of
the province. Furthermore, in some sub-regions the general level of trading and monetary
relations was different. In northern Tukharistan, for instance, Kufic dirhams and even cop-
per coins were struck only occasionally, as borne out by the number of coin finds. In overall
terms of trading and monetary relations, northern Tukharistan lagged behind Transoxania.

11
Mitchiner, 1973 ; Lowick, 1977, pp. 205-7.
12
Sellwood, 1980, pp. 176-7.
13
Davidovich, 1966, pp. 125-30.

402
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

Coinage and the circulation of money from the

eleventh century to the beginning of the thirteenth
The 'silver coin crisis' (the causes of which are not examined here) relates to a situation
that studies have already confirmed regarding the minting and circulation of coins in the
Muslim East from the eleventh to the thirteenth century. The situation displays general
features over the whole of that immense territory, and local features in various regions
and sub-regions. The general features of the silver coin crisis are: the debasing of silver
dirhams with alloys (even involving various copper or non-precious metal alloys); the acti-
vation of gold coins even in regions where they had previously played no significant role
in monetary circulation; and the persistence of this situation for some 200 years. Local
variants are characterized, first, by the variable duration of the silver coin crisis and by the
time taken and the means employed to overcome it; and, second, by the differing compo-
sition of the money supply as regards both silver and gold circulation, not only in different
states but in individual sub-regions of one and the same state. The composition of the
money supply depended primarily on the kinds of debasement of silver dirhams and the
technical methods used, and the time of introduction and organization of the circulation of
silver substitutes. It also hinged on disparities in the minting of dinars (gold coins of dif-
ferent standards of fineness and gilded silver coins). Local variants of the silver coin crisis
stemmed from interaction between the local level of trading and monetary relations and the
socio-political situation, the status of silver reserves and the possibilities of replenishing
them, the monetary and fiscal policy of the state (and sometimes of particular rulers) and
social psychology (in particular, the solidity of the traditions of monetary circulation in the
preceding period).
Transoxania, as part of the state of the Karakhanids, continued to display an extremely
original variant of minting and monetary circulation. 14 Typical here are three features,
whose combination seems on the face of it paradoxical. The first is the lack, for a century
and a half, of any regular minting of gold coins; the second is the maintenance of the
positions of Bukhar Khudat coins; and the third is the fast pace and variety of methods
used for debasing the Karakhanid Kufic dirhams.
The Karakhanid silver coins were, from the technical point of view, debased in two
ways. In some cases, a hardener (usually copper) was added to the alloy, and in others,
nummi subaerati (consisting of a copper core and coatings of silver) were issued. A quan-
titative chemical analysis of a few dozen coins of different periods from hoards found in

14
Davidovich, 1960, pp. 92-117.

403
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

Transoxania makes it possible to trace the course of changes in the standard of fineness of
Karakhanid dirhams.
In the first and early second decade of the eleventh century, dirhams were still of a high
standard. For example, in the dirhams of Ferghana (minted in the cities of Akhsikat and
Uzgend) there was a copper admixture of about 12%. In addition to locally minted dirhams,
Karakhanid coins minted in Kashghar and Yarkand were in wide circulation in Ferghana.
The early Kashghar dirhams were also of high quality, but the copper content increased
later. It is noteworthy that the dirhams of Kashghar in 1019-20 and of Yarkand in 1026-7
belong to the nummi subaerati category. The silver coatings are worn (in some instances to
the point of transparency) and in places they have come off altogether. Hence an analysis
of the silver has indicated a considerable spread in the 40-60% range. No less than 30%
of copper had thus been added. In the Chaghaniyan dirhams of 1034-5, the amount of
copper reached 73-75%, and later (1041-2) more than 80%. In the chronologically close
Samarkand coins (minted in 1046-7 under the Karakhanid Ibrahim Tamghach Khan) the
respective proportions of copper and silver were 80% and 15-16%, but the coins were
nummi subaerati, so that the silver content was originally somewhat higher.
In the mid-eleventh century, dirhams were minted from a copper and lead alloy in many
cities of Ferghana. After the conquest of Ferghana by Ibrahim Tamghach Khan, the cir-
culation of copper and lead dirhams was banned in the region and they were replaced by
fresh dirhams bearing the name of the conqueror. The new coins were minted from a silver
and copper alloy (specimens analysed from 1061 and 1067-8 contain 18-23% silver and
70-72% copper). Low-standard dirhams continued to be issued in the twelfth century. For
example, two dirhams of Mascud b. Hasan minted in Samarkand around 1166-70 proved
to be 22-25% silver.
The exchange rate of Karakhanid Kufic dirhams depended on their standard of fine-
ness. It was therefore customary for documents to contain detailed descriptions of them,
making it quite clear which particular coins were concerned. In one of the waqf docu-
ments of Ibrahim Tamghach Khan, for example, his Samarkand dirhams are described
thus: 'mu' ayyadz c adlz dirhams of the established model, liquid assets [i.e. legal tender -
Ed.] in the district of Samarkand' and '47 dirhams [are equivalent] to 1 mithqal of pure,
high-carat gold' at the time of establishment of that waqj. 15 Had silver circulation consisted
solely of low-standard Karakhanid Kufic dirhams, this would certainly have prompted the
regular minting of gold coins in Transoxania (a province with highly developed trading
and monetary relations) and their use in monetary commerce. However, the feature of cen-
tral Transoxania up to the mid-twelfth century was the already familiar Bukhar Khudat

15
Khadr, 1967, p. 329.

404
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

dirham. Of these, the Muhammad! dirhams (nummi subaerati, over 40% silver) began to
drop out of circulation over several decades of the eleventh century, often together with
Karakhanid dirhams of differing standards. Ghitrifi dirhams (of copper or non-precious
alloys) remained legal tender and the recommended medium of exchange until the mid-
twelfth century. As previously, the parity of the Ghitrlfi dirhams fluctuated but was usually
either equivalent to or above the rate for fine silver. This particular point was noted with
some surprise in the mid-eleventh century by the historian and polymath al-Biruni.
Silver was expensive at that time. In 1128, for instance, the ratio of gold to silver was
1:7 .5. Yet in that same year, 100 dirhams by weight of fine silver were worth no more
than 70 or 72 Ghitrifi copper dirhams. 16 Significantly, the area of circulation of the Ghitrifi
dirhams had increased in comparison with the previous period. Under the Karakhanids,
these dirhams went beyond the bounds of the Bukhara region and became the official
medium of exchange in many regions of the state. Such an established situation, coupled
with the high exchange value of the Bukhar Khudat copper Ghitrifi, was secured by deep-
rooted tradition and the monetary policy of the Karakhanids that was built upon it.
Major changes took place in the mid-twelfth century, with the start of regular issues
of gold coinage. The low-standard Karakhanid Kufic dirhams were replaced by dirhams
minted without any silver at all, and with just the amalgamation of a thin silver film on
the surface. These silver-coated copper dirhams were large, handsome coins (Fig. 2: 1- 4)
issued in many cities, including Samarkand, Bukhara, Uzgend, Kasan, Marghinan, Binakat
and Utrar. Some cities minted silver-coated copper coins of two values (distinguished by
size, weight, outward aspect and sometimes inscriptions). Like any tokens of value with an
imposed exchange rate, the silver-coated copper dirhams often suffered inflation, against
which the financiers of the time had two weapons: one was economic (reducing the money
supply through prohibitions and exchanges) and the other was psychological (increasing
the weight of new coin issues). A particularly vivid picture of frequent inflation and such
fiscal measures is provided by the coins of Ferghana in the second half of the twelfth
century. 17
The policy with regard to the minting of gold coins followed by the Ghaznavids, the
Great Seljuqs, the Khwarazm Shahs of Anushtegin's line, and the Ghurids, who ruled over
various parts of eastern Iran and Central Asia, was quite different. The Seljuqs, for exam-
ple, began issuing gold dinars even before the decisive victory in 1040 over the Ghaznavids
at Dandanqan. After 1040, the issuing of both high- and lower-standard gold coins, and
subsequently gilded silver dinars, was put on a regular footing.

16
Davidovich, 1960, pp. 93-8.
17
Davidovich, 1961, pp. 186-200.

405
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

Fig. 2. Drawings of silver-coated copper dirhams from the end of the twelfth century to the begin-
ning of the thirteenth. 1-3: Karakhanid dirhams, from Uzgend, A.H. 594; from Marghinan, A.H.
602; from Binakat, A.H. 602. 4: dirham from Uzgend, A.H. 609, with the names of the Karakhanid
Mahmud b. Ahmad and the Khwarazm Shah Muhammad b. Tekish. 5-6: dirhams of Muhammad
b. Tekish, from Termez, A.H. 617. 7: dirham from Samarkand, 622/1225.

Let us take a closer look at the initial stage of the emergence of the 'Ghaznavid vari-
ant' of the silver coin crisis. In every way (size, thickness, style of inscriptions, decorative
motifs), the Ghaznavid coins differed from the Karakhanid Kufic coins. The debasing of

406
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

silver dirhams also proceeded more slowly than in the state of the Karakhanids. In the
Ghaznavid dirhams of Sultan Mascud I (1030-41), the silver content was thus over 70%
and the admixture of copper did not exceed 25-27%. At that time, dirhams were still being
issued at an official standard of fineness equivalent to 'nine-and-a-half tenths', i.e. 95% sil-
ver. In modern southern Tajikistan, some hoards of Ghaznavid coins have been found from
the first half of the eleventh century (including local dirhams from the mint of Khuttalan).
A tangible sign of their circulation is the fact that the hoards contain not only whole coins
but fragments of differing sizes, shapes and weights. Together with small Ghaznavid silver
dirhams, outsize heavy coins remained in circulation (with a broad free field around the
centrally placed inscription) and the series was still in use under the Samanids.
The Ghaznavids at once began regularly issuing gold coins as well, at various standards
of fineness. For instance, the dinars of Nishapur maintained their high standard (most of
those assayed being 93-96% gold), while the dinars of Herat were no more than three-
quarters gold (the assay results being in the 67-75% range). Dinars of various standards
were variously named. For instance, Bayhaqi and al-Biruni refer to both the Nishapur and
the Herat varieties, which bears out the above factual data on the difference in their stan-
dards of fineness. However, the mere fact of minting gold coins gives no idea of their
functions and place in monetary trading (cf. dinars under the Samanids). Crucial material
for understanding the objective changes in the position of Ghaznavid gold that began in
the eleventh century is provided by the historian and official Bayhaql. An examination of
actual instances of purchases and sales, gifts and recompenses, supplies sufficient evidence
that, even in the second quarter of the eleventh century, the value-measurement function
was still fulfilled by silver dirhams, but gold dinars were already in use as a medium of
circulation and payment. 18
The Khwarazm Shah cAla' al-Din Muhammad b. Tekish (1200-20) attached great impor-
tance to the political aspect of minting money. Coins were issued in his name from many
mints of conquered territories, including Khwarazm itself, Bukhara, Samarkand, Uzgend
(?) and Farab (Utrar); and, in the south, Tirmidh (Termez), Chaghaniyan and Wakhsh.
However, Muhammad b. Tekish did not abolish the monetary systems established before
him in various states and provinces, though the geographic limits of the variants sometimes
shifted. In central Transoxania, the basis of the monetary system was, as before, the coexis-
tence of gold dinars and silver-coated copper dirhams (Fig. 2:5- 6). Tukharistan happened
to be in the area of minting and circulation of those coins. We know of the many issues of
silver-coated copper dirhams in Termez and Chaghaniyan (modern southern Uzbekistan)
and hoards of such coins have been recorded in southern Tajikistan. In the territory of the

18
Davidovich, 1980, pp. 61-2.

407
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

former Ghurid empire, gold dinars and silver dirhams of various standards were issued in
the name of Muhammad b. Tekish, but they retained their traditional appearance. There
was no break in the tradition of also issuing fulas - copper coins for retail trading - in a
number of cities.
A considerable number of hoards have been found in Central Asia, consisting basically
of gold dinars of Muhammad b. Tekish. They usually come with a small admixture of
gold coins of other rulers and dynasties. These mixed hoards include, as a rule, not only
whole and fragmented dinars but pieces of coin varying in size and weight. Clearly, gold
coins (whose weight from the outset, at the time of minting, fluctuated markedly) could not
circulate simply as coins. Weight checking at the time of payment required makeweights,
the number of which increased in the process of gold circulation. It is no coincidence that
some hoards contain more assorted fragments of dinars than whole coins.

Coinage and the circulation of money under the

Mongols (thirteenth and fourteenth centuries)
The Mongol conquest, accompanied as it was by the destruction of cities and the devasta-
tion of entire regions, with the consequent reduction and ruin of the population, also did
great harm to monetary trading. The changes in the minting of coins, the composition of
the money supply and monetary circulation need to be divided into four stages.
The first stage (second quarter of the thirteenth century) is marked by a sharp drop
in the number of mints, irregular issues of coins and a monetary circulation crisis. 19 The
coins of Samarkand provide a particularly vivid picture of the development and specific
nature of the crisis. In Samarkand, gold dinars (both anonymous and bearing the name of
Chinggis Khan) and silver-coated copper dirhams were at first minted - a system inherited
from pre-Mongol times. But a change came about in the language and substance of the
inscriptions. The outsize silver-coated copper dirhams (weighing about 6 g) started bear-
ing non-standard inscriptions in Persian. First, they assure the inhabitants of Samarkand
that the coins have been specially minted for them and so deserve their trust. In 622/1225
we find, 'This money is for circulation in Samarkand and its district' (Fig. 2:7). A repeated
statement comes in 624/1226-7 that the money is for local circulation, but there is addi-
tional psychological pressure in the legend: 'Chinggis currency', 'Khan' and once more
the name of the fearsome conqueror 'Chinggis Khan'. Clearly, the 'prompting' was to no
avail since in 1232-3 and 1233-4 the inscription became admonitory: 'Non-acceptance of
this currency in Samarkand and its district shall be a criminal offence.' As a result, there

19
Davidovich, 1972, pp. 37-47, 127-35.

408
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

was no inflation, but a deeper crisis when the population would have nothing to do with
the money. However, the threats made no impression and, as a result, the minting of silver-
coated copper dirhams had to be halted in Samarkand. The minting of small silver coins
in 1236-7 did not improve the situation since they promptly disappeared, and Samarkand
entered its moneyless decades. In Bukhara, the minting of small silver-coated dirhams was
also discontinued. In other cities of Transoxania, none were issued at all. This means that,
to begin with, commerce within cities and within the region (the volume of which certainly
fell) was mainly based on barter.
The situation in the eastern part of the possessions of Chaghatay (Chinggis Khan's son)
was somewhat different, but for special reasons. The Mongols had brought in a number
of craftsmen from Transoxania, and Chaghatay had his summer residence near the city of
Almali'k, where commerce thrived. High-standard silver dirhams and copper fulas were
regularly minted in Almalik.
The second stage (up to 1270) 20 was marked by the regular minting of gold coins
throughout the empire and of silver-coated copper dirhams in the north-east. At the kurul-
tay (general council) of 1251, Mongke (Mengti) was elected chief of the Mongols. The
new leader proclaimed a comprehensive reform in order to stabilize the situation. One of
its points was the cash payment of a standard per capita tax, the qubchur, which Mongke
ordered to be collected annually in gold dinars. In Khurasan, ruknf dinars were in circu-
lation at the time and their standard of fineness was equivalent to 4 dangs (four-sixths,
i.e. two-thirds fine gold). The qubchur was recalculated in ruknf dinars. However, since
there were no gold coins in Transoxania to comply with Mongke Khan's command, it
became urgent to start minting them. Mascud Beg, a merchant governing Transoxania and
other regions of Chagatay's realm, succeeded in this task. Many mints (in Bukhara and
Samarkand, in Khujand and Utrar, in Urdu al-Aczam, Almali'k, Shafurqan and elsewhere)
started minting small, slim dinars of low-standard gold (about 60% gold and 40% alloy).
The weight-standard was not strictly laid down. Not only whole coins but halves and small
fragments of various weights were in circulation. Because of the low standard and the exis-
tence of fragments, the gold dinars practically supplanted the silver coinage by meeting
the requirements of the various levels of monetary trading. The gold coins thus made their
appearance among the population and the qubchur was recalculated in them (by weight).
A further sign of the revival of monetary trading in the second stage was the renewed
minting of silver-coated outsize copper dirhams. In Utrar they were struck annually start-
ing in 1251-2, and they were used for commerce not just in Utrar but in the whole of Fer-
ghana and the Tashkent region. Ferghana (i.e. the city of Khujand) subsequently, in 1264-5

20
Davidovich, 1972, pp. 29-36, 48-50, 136-41.

409
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

and 1266-7, issued its own coins of this type. The minting was resumed of silver-coated
copper dirhams even in Bukhara and from 1261-2 this was done on a regular basis. Only
Samarkand had not recovered from the crisis affecting the silver-coated copper dirhams.
When such coins were finally issued there in 1264-5, they once more bore admonitory
inscriptions (but this time in Turkish). No further attempts were made, for in Samarkand
nobody was any more willing than before to deal in silver-coated copper dirhams.
The third stage (starting in 670/1271-2) 21 begins with the monetary reform of the gov-
ernor Mascud Beg, who banned the further circulation of silver-coated copper dirhams, put
an end to the minting of gold dinars and organized the issue of high-standard silver coins
in many cities. This was one of the most difficult reforms in the whole history of Central
Asia. For over two and a half centuries, there had been no regular minting there of high-
standard silver coins and isolated attempts to overcome the silver coin crisis had come to
grief. It took Masciid Beg two decades of systematic effort to succeed with the reform. In
Almali'k, as already noted, silver coins were issued in the first and second stages, but this
was an exceptional occurrence. Mascud Beg began the reform in 1271-2, which was the
very year when silver dirhams were first issued in Utrar. By the following year, 1272-3,
the silver coins were already being produced by five mints, those of Utrar, Kanjida, Taraz,
B urkhan and Almali'k.
There was an increase in the number of mints and the reform achieved its greatest
success in the course of the second decade (the 1280s), when at least fifteen mints were
in operation. In the north-east and east of Central Asia, silver coins were minted in Utrar,
Taraz, Kanjida, Burkhan, Almali'k and Bulat. In Ferghana, dirhams were issued by the
cities of Khujand, Marghinan, Andijan and Ush. The single mint in the Chach region had
three names: Chach, Binkath and Tashkent. The Mongols repeatedly sacked Bukhara in
1263, 1273 and 1276, massacring its population. The city stood empty. By 1283 Mascud
Beg had retaken Bukhara and forthwith organized the minting of silver dirhams there. At
the same time, the Samarkand mint started operating (Fig. 3: 1- 2). An important ingredient
of Masciid Beg's reform was the unification of the weight-standard (2.1 g) and the standard
of fineness (approximately 80% fine silver) for all the mints. These small, high-standard
silver coins circulated at the same parity throughout the province, regardless of the place
of issue. The silver coin crisis in Central Asia lasted longer than in the other regions of the
Near and Middle East. The consistent and firm policy of Masciid Beg helped to stabilize
and somewhat stimulate domestic trade and the life of the cities in general. The monetary
reform, restoring silver coinage to the market, was a key component of his policy.

21
Davidovich, 1972, pp. 51-114, 141-51.

410
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

.,

Fig. 3. 1-2: Chaghatayid silver dirhams after the reform of the third stage: Andijan, Bukhara
(enlarged). 3-7: Timurid silver tanga, fifteenth century (3-6: Shah Rukh from Abarkuh,
Samarkand, Herat and Sabzavar; 7: cAbd Allah from Samarkand, 854/1450--1). 8-14: anonymous
copper coins from the reform of Ulugh Beg (8: one coin from a series dated A.H. 832; 9-14: coins
from Bukhara, Samarkand, Karshi, Andijan, Shahrukhiyya and Termez).

411
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

A fresh reorganization of the minting of silver coins was carried out by Kebek Khan
(1318-26). 22 This was the fourth stage of the circulation of money in the Chaghatay ulus
(domain). Kebek's reform repeated the system introduced much earlier in Persia by Ghazan
Khan (1295-1304). The basis of the system was the outsize silver coin (dinar), equivalent
to six small silver coins (dirhams). Under Kebek, the dinar weighed just over 8 g and
the dirham therefore about 1.4 g. The dinars were dubbed kebeks and the name stuck.
Kebek's reform obviously prohibited further circulation of the anonymous dirhams of the
third stage, whose weight-standard (2.1 g) did not fit into the new scale of values.

Coinage and the circulation of money in Transoxania

under Timur and the Timurids (late fourteenth and
fifteenth centuries)
Surprisingly, the minting and circulation of gold and silver coins during this period have not
attracted the attention of numismatists and historians. No study has even been made of such
basic ingredients of monetary policy as official weight-standards and standards of fineness
of gold and silver coins. The scientific literature contains only isolated observations, and,
in essence, all writers on the subject repeat Vasmer's notion regarding the weight of silver
coins: in approximately 1390, in Transoxania and eastern Iran, Timur began minting tangas
of about 6.0 g and l/4-tangas (dirhams) weighing some 1.5 g. Under his son and succesor
Shah Rukh (1405-47), the weight of the tanga dropped to 4.72 g. 23 Meanwhile, there is
information in the written sources indicating the existence of at least five weight-standards.
Some are expressed in eastern units of weight and others are compared with the weights of
non-Muslim coins. A comparative analysis of all these data enables five weight-standards
to be singled out, both in eastern units of weight and in grams: the Timur tanga, equivalent
to 1 mithqal and 2 dangas of a mithqal (6.4 g); the Shah Rukh tanga, equivalent to 1 mithqal
and 1 1/2 dangas (6.0 g), to 1 mithqal and 1 danga (5.6 g), to 1 mithqal and 1/2 danga (5.2
g); and the post-Timurid tanga, equivalent to 1 mithqal (4.8 g) (Fig. 3:3- 7). Other weight-
standards were quite possibly used as well. The early coins of Timur, for instance, were
probably minted in accordance with a higher weight-standard. The underlying trend of the
changes was clearly in the direction of a lowering of the weight-standard. It is not clear,
however, what the general policy was regarding the weight of silver coins and whether the
weight-standards applied to the entire province or were regional. To resolve this question,

22
Masson, 1957, pp. 41-108.
23
Vasmer, 1930, pp. 680-1.

412
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

it would be necessary to study the real and mean weight of a large number of coins from
each major mint separately.
Under Timur and the Timurids, besides silver coins of the basic face value, fractions
of that value were issued. In Samarkand under Timur, for instance, small coins worth 1/4
tanga, called mfrf (their weight-standard being 1/3 mithqal, i.e. 1.6 g), were minted in vast
quantities. In Khurasan under the last Timurids, the small coins worth 1/3 mithqal (1.6
g) and 1/6 mithqal (0.8 g) were called kebekf dinars and c adlf kebekf. 24 Halves and other
fractions were minted. The silver coins of Timur and the Timurids were of a high standard.
In a number of fifteenth-century wathfqas (deeds of purchase), the standard of fineness of
the coins is defined as dah dahi ('ten-tenths', i.e. 100% silver), as reckoned at the time.
Many coins are overstruck with small cartouches bearing an inscription (rulers' names,
cities, words). Coins so modified were worth more than the plain version.
Particularly noteworthy are copper coins, since their importance in monetary trading
grew steadily throughout the fifteenth century. Four names were in use for the copper
coins: fulus, c adlf, dznar and dangf. 25 In coin inscriptions and legal documents (wathfqas,
waqf documents, etc.) they occur sometimes in isolation and more often in a variety of
combinations. The term fulus lost its plural sense and in the fifteenth century was applied
to any copper coin regardless of its value. If the terms c adlf or dangf (julas-i c adlf, fulas-i
dang()were added to the term fulus, this meant that the basic face value was implied in
the copper issue. In the second half of the fifteenth century, the term dinar became the
principal name of the basic face value and in order to distinguish the copper dinar from
the silver and gold dinars, epithets were added to it. The most common epithet became the
term fulus. Dfnar-i fulus was thus a 'copper dinar'. Local epithets were also used, like the
harawf dinar, the tabrfzf dinar and so on. The early copper coins bear the names of Timur
and his first heirs, Jahangir and Khalil Sultan. Subsequent issues are anonymous and most
of the inscription space is taken up by the name of the mint, the date of minting and words.
Examination of the many hoards of fifteenth-century copper coins has made it possible
to pinpoint periods of inflation, crises, reforms, general development trends regarding the
minting and circulation of copper coins, the period of the maximum production of con-
sumer goods and of retail trading, the features of tax policy, and much else. Because of the
way the changes took place, the minting and circulation of copper coins in the dominion
of Timur and the Timurids must be split up into five stages. 26

24
Davidovich, 1983, pp. 40-50.
25
Ibid., pp. 32-57.
26
Ibid., pp. 129-322.

413
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

The first stage (until 832/1428-9) was marked by a growing number of mints and an
increase in the money supply. Basic denomination copper coins were issued in Samarkand,
Bukhara, Karshi, Shahr-i Sabz (central Transoxania), Termez and Khuttalan (in the south),
Utrar and Shahrukhiyya (in the north) and Andijan (Ferghana valley). The biggest issue
was that of 823/1420, the average weight being approximately 4.4 g. No importance was
attached to accuracy regarding the weight of individual coins as the variance on either side
of the mean was considerable.
The second stage (starting in 832/1428-9) was the monetary reform of Ulugh Beg
(1409-49), on whom his father Shah Rukh had bestowed a huge appanage, essentially an
entire province with Samarkand as its capital. The main aim of the reform was to give the
market a constant and stable copper coin for retail trading and, at the same time, to increase
revenue from the minting of such coinage. The reform worked. In 1428-9, many mints pro-
duced coins that were identical as regards design, content and the layout of the inscription
(Fig. 3:9- 14). Their mean weight was one and a half to twice that of the pre-reform coins
of 823/1420. The 'old' coins were prohibited and new ones offered in exchange. The terms
of the exchange are not known but they were evidently not ruinous for the population since
the operation went ahead smoothly. After the massive exchange, the minting of copper
coins in many cities was centralized in Bukhara and serialized. This means that for several
decades, identical coins were minted in Bukhara, without even any change of date in the
legend, since they all read 832/1428-9 (Fig. 3:8). Coins of this series found their way into
every comer of Central Asia and were used in all markets. In addition, their centralized
minting certainly increased Ulugh Beg's revenue many times over.
Some features of the reformed minting also paved the way for inflation. In the first place,
the mean weight of the reform coins was repeatedly lowered. Over a few decades, this led
to the accumulation of coinage that was too divergent in weight, with coins of the main
denomination ranging from 1.8 to 8.6 g. Secondly, the total volume of such coin issues was
far in excess of market demand. The third stage was the elimination of inflation. The usual
method was to withdraw current copper coins and substitute specially minted new ones, but
the minting of new coins for a complete exchange is a costly and lengthy operation. During
the third stage the problem was overcome in another way, by means of overstriking. Cir-
culating coins of the series were for several years overstruck in many cities. The overstrike
was a square cartouche with the inscription of the name of the mint (Samarkand, Bukhara,
Karshi, Termez, Hisar, Khuttalan, Andijan, Shahrukhiyya, etc.) and the word dangf, the
latter indicating that only coins with this overstrike had acquired the basic face value. The
market also received halves of the mam denomination by means of another overstrike: in a
lenticular cartouche the name of the mint (Samarkand, Bukhara, Karshi, Termez, Andijan,

414
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

Shah Rukh (?)) and the word nfmdangf (1/2 dangf) were inscribed. Coins without over-
strikes were not withdrawn but served as the smallest fractions of the main denomination.
The market received copper coins of three denominations, which testifies to the expansion
of retail trading. The transformation of part of the coins of the series into lesser frac-
tions of the main denomination reduced the volume of surplus money. Inflation was thus
eliminated or checked. There was one miscalculation, however, in the overstriking opera-
tion: no selection by weight was made, and the coins of the main denomination (with the
dangf overstrike) continued to be extremely variable in weight. This was one of the causes
of the destabilization of monetary circulation.
The most conspicuous signs in Transoxania of the fourth stage (final decade of the fif-
teenth century) lay in the change of policy regarding the weight of copper coins and the
range of values. Copper coins were for the very first time subjected to a definite weight-
standard (first, 1 mithqal and 2 nukhuds, i.e. 5.2 g; later, 1 mithqal and 1 nukhud, i.e. 5.0 g)
with a very small tolerance. The regular minting in Samarkand, Bukhara and other cities,
starting in 897/1491-2, of new coins of a fixed high weight, coupled with the large-scale
withdrawal of the coins of dissimilar weights struck in the second and third stages (i.e.
coins of the 832/1428-9 series with and without overstriking), reduced and renewed the
money supply and created the important sensation of solidity. A no less perceptible sign
was the increase in the range of values. Instead of copper coins of two or three denom-
inations (the usual pattern of the previous period), the market received coins of five or
six denominations of distinct weights and sizes. In Samarkand, for instance, copper dinars
were issued in 1 1/2-, 1- and 1/2-dinar pieces, together with coins worth two-thirds and
one-third of a copper dinar.
The situation was quite different in another district, that of northern Tukharistan. The
local minting of copper coins (in Hisar) was of secondary significance since it accounted
for only a small part of the district's money supply. The main feature was the long-term
and repeated overstriking of the copper coins of Transoxania, which indicated a change of
coinage. For the sake of convenient and full exchange, temporary mints were even opened
in small cities and population centres. During one of the overstriking periods, lasting a
number of years, exchange mints were opened in nine centres of the relatively small terri-
tory of northern Tukharistan. Each new overstrike invalidated the previous one, and only
coins with the new, most recent modification were accorded full-dinar status. On each occa-
sion, the holders of the old coins forfeited part of their assets, and there were at least seven
such operations in the course of the decade. At the beginning of the sixteenth century, the
circulation of copper coins and retail trading in northern Tukharistan was consequently

415
ISBN 978-92-3-103467-1 Contents Coinage and the circulation of money ...

in a crisis situation, with the prices of staples and commodities constantly rising and the
effectiveness of such a fiscal policy decreasing as far as the revenue was concerned.
The year 907/1501-2 marked a political watershed and the start of a new, fifth stage
in the minting and circulation of copper coins. In that year, Muhammad Shaybani Khan
(1500-10), the founder of the new dynasty of the Shaybanids, definitively conquered
Samarkand. Northern Tukharistan, however, still belonged to the Timurids and was the
scene, in 907/1501-2, of a far-reaching monetary reform. It bore the stamp of a talented,
albeit anonymous, financier who grasped the extent and danger of the current monetary
circulation crisis, together with the need to take account of the psychology of the populace
in overcoming it, but did not overlook considerations of public revenue. The success of
the reform required a clear and universally comprehensible divide between old and new,
past and future, plus a range of denominations that would not ignore the steady increase
in prices. To begin with, all the old discredited coins were withdrawn. In other words, the
past was 'closed down'. In their place, the market was supplied with coins of three denom-
inations which differed in every respect (legend, language, image, weight, size, thickness)
from those withdrawn. On the 1- and 2-dinar pieces, the inscriptions were in the Persian
familiar to the local population, with the value of the coin also stated in Persian: '1 dinar',
'2 dinars'. The 2-dinar pieces were larger and thicker, and their mean weight was twice
that of the single dinars. A particularly noteworthy point is that the reform determined that
the basic coin would be not the single dinar but the 2-dinar piece, since it tallied better
with the steady increase in prices. The 2-dinar pieces depicting a gazelle and bearing a
Persian inscription constituted a series issued without any change for a number of years by
three mints, those of Hisar, Qunduz and Termez. The population at last had a steady and
enduring currency for retail trading, and the public coffers were swelled considerably by
its production.

416
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

Part Two

AFGHANISTAN, PAKISTAN AND NORTHERN INDIA

(A. H. Dant)

The first series of new coins found in this region belongs to the Arab caliphate - the
Arab-Sasanian coins, found in southern and western Afghanistan. 27 Together with Arab-
Byzantine coins, they have also been recovered in excavations at Banbhore, a port town
on the Arabian Sea coast in Sind (Pakistan). The town is generally identified with Daybul,
the city captured by the Arab general Muhammad b. al-Qasim in c. 711. One gold coin
of the caliph al-Wathiq (842-7) has been recovered. 28 The second site, al-Mansura, was
established as a mint town by the Arab governors. Here the coins were minted by the
Habbari rulers, such as al-Mansur b. Jumhur al-Kalbi and his successors. Copper coins
minted at Banbhore have also been found in large numbers. However, the most remarkable
is the local coin type of the caliph al-Muqtadir (908-32), a silver dirham, copying the bull-
and-horseman type of the Shahi rulers of Kabul. The obverse shows the figure of the caliph
on horseback and the reverse has a recumbent bull with his name above it. The Kabul ruler
is taken to be Spalapati Deva, now assigned to the Turk-Shahis and possibly defeated by
the Muslims in the time of al-Ma'mun (813-33). 29 This type is later found also in the
coins of Muhammad and Mascud I, the two sons of Sultan Mahmud of Ghazna. The coins
continued the older design, fabric and weight-standard of the Shah! originals. The type
reaffirms the continuity of monetary tradition.
The Ghaznavid rulers, initially as governors of the Samanid amirs of Bukhara, intro-
duced the coin types of the Samanids. Alptegin's coins are known only from the Andarab
mint, 30 but Sebtiktegin issued coins from Ghazna in 969, recognizing Mansur I b. Nuh
as his overlord. The coins of Sebuktegin, who acknowledged Nuh II b. Mansur as his
overlord, were struck in Balkh, Bamiyan and Farwan. But those minted at Bamiyan vary

27
Walker, 194 1.
28
Khan, 1969, p. 43.
29
See Abdur Rahman, 1979, Ch. 6.
30
Thomas, 1848b, p. 295.

417
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

considerably from the currency of his Samanid masters, and approximate in weight and
size to the local coinage of the Hindushahis of Kabul.
More varieties are seen in the coins of Sultan Mahmud (998-1030), who struck coins
in Nishapur, Herat, Farwan and Ghazna. His coins come in the same category as those
of Sebtiktegin, though his Nishapur mint coins differ materially from others in form and
value. Mahmud's coins began with displaying the simple title of sayf al-dawla, conferred
on him by Nuh II b. Mansur in 994, but in 998 he broke away from the Samanid amirs when
he assumed for himself on his coins the fully independent titles of al-amfr and al-sayyid.
To these were added the titles of yamfn al-dawla and amfn al-milla, conferred on him by
the cAbbasid caliph al-Qadir (991-1031), duly recognized on his coins. Later, he assumed
the titles of nizam al-dfn and also malik al-mamalik and malik al-mulak.
A special bilingual variety of Mahmud's coins is worth noting. The obverse shows the
usual kalima and ruler's name with his titles in Arabic, while the reverse has the Sanskrit
legend in Nagari showing an attempt to translate the kalima. The margin names the coin
as a tanka and the mint as Mahmudpur. 31 The translation of the kalima is not exact, but is
understandable in the local philosophical concept. The words avyaktam-eka (Invisible and
the One) stand for la ilaha ilaAllah and avatara (Incarnate) stands for rasal (Messenger).
The two contending successors of Mahmud continued his types and, as mentioned ear-
lier, they also copied the bull-and-horseman type. Muhammad first gave the name and titles
of his father on the obverse before he issued his own coins. Mascud first called himself walf
al-cahd (heir) before issuing his own independent coins. 32 He later started a new type of
brass coin, weighing 60 g and with the titles of al-sultan al-muc azzam malik al_c alam, these
titles apparently in imitation of those of the Seljuq sultan, who called himself al-sultan
al-mucazzam shahanshah Toghril Beg Abu Talib. 33 It is from the time of Mascud's son,
Mawdud ( 1041-8), that we find another type in silver and copper with the name and titles in
Arabic on the obverse, thus replacing the horseman, but the reverse has the bull and Nagarl
name Srf Samanta Deva. 34 The mint name Lahore given in his coin35 should be identified
with Salature in the Swati tahsfl (subdivision) of the modern North-West Frontier Province.
Ibrahim b. Mascud (1059-99) assumed still greater titles of al-sultan al-aczam, qahir al-
mulak and sayyid al-salatfn. The title al-sultan henceforth became common and was used
in the coins of Mascud III and Malik Arslan. But the two subsequent rulers, Bahram Shah

31
Thomas, 1860, pp. 160-2.
32
Thomas, 1848b, p. 335, No. 58.
33
Ibid., p. 337.
34
Ibid., p. 349, No. 91.
35
Ibid., No. 92.

418
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

and Khusraw Shah, though calling themselves al-sultan al-aczam, also acknowledged as
their suzerain the Seljuq Sultan Sanjar on their coins.
The economic influence of the dynastic change on currency is not difficult to infer. The
gold currency of the time was based on that of the Arab caliphate - on the dinar in standard
and value - but the silver and copper currency adopted the local weight-standard, thus
continuing the older exchange relationship.
The succeeding dynasty of the Ghurids followed this practice. However, the type was
now different. Ghiyath al-Din Muhammad b. Sam (1163-1203) has the horseman on the
obverse and his own name and titles in Arabic on the reverse. His brother Mucizz al-
Din Muhammad ( 1173-1206), the conquerer of northern India, shows new varieties and
the adoption of the coin types of the Rajput rulers. The fundamental change in his gold
currency is noticeable in the unusual appearance of the Hindu goddess Lakshmi on the
obverse, with his name and title Srf mad Hamfr in Nagari on the reverse. This type, as
Nelson Wright pointed out, was based on a local standard of 72 grains (or 40 rattis) and
differed from the Ghazna gold issues. 36 The sultan also inaugurated a new 'Indian gold
tankah' of double the weight but following the same standard, having a Chawhan horseman
on the reverse and his name and title in Arabic on the obverse. His billon coins show the
bull-and-horseman type with the name and title written in Nagari. Sometimes the bull is
replaced by the name in Arabic. They are based on a 32-ratti standard, different from the
standard known in Ghazna and the sultan's northern territories.
In Ghazna and other parts of Afghanistan, the local currency was continued, even after
the death of Mucizz al-Din Muhammad, by his son and successor Mahmud b. Ghiyath
al-Din Muhammad and even by the Ghurid slave commander Taj al-Din Yildiz, with the
latter's new mint possibly at Karman or Kurraman (near Parachinar in the modern North-
West Frontier Province), showing a standing bull in silver and copper coins. In Ghazna,
Yildiz gave his own name on the margin and that of his late master's name on the obverse
as al-sultan al-shahzd Muhammad b. Sam on the gold coins. Later, when the Khwarazmian
sultans advanced south of the Hindu Kush, they copied the local gold and silver coin types
from the mints of Farwan, Bamiyan and Ghazna. The silver and copper coins of cAla' al-
Din Muhammad show the horseman type with the name and titles in Arabic, while another
variety adds the title of Sikandar al-thanf, and on the reverse either a bull with Nagari leg-
end or a horseman with the Nagari title Sri Hamfra. 37 His son Jalal al-Din Mingburni also
copied the horseman type. Later, when the Mongol conqueror Chinggis Khan advanced
to Ghazna and the Indus, he continued the local silver and copper currency with Arabic

36
Wright, 1907, p. 7, Nos. 1-3.
37
Thomas, 1848b, p. 383, Nos. 15 and 16.

419
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

obverse titles: al-nasir li-dfn allah amfr al-mu'minfn and on the reverse al-cadil al-aczam
Chink(g)fz Khan or cadl kaqan al-aczam.
In northern India, the reign of the Delhi Sultan Shams al-Din Iltutmish (1211-36) marks
a period of political consolidation and, as Wright observed, it 'stands out as a landmark in
the coinage of Delhi'. After refuting the earlier views of Edward Thomas, Wright claimed
to have correctly established the value of the silver tanga of Iltutmish, which was equal to
1 tola, i.e. 12 mashas, each masha having 8 rattis. Hence the tola was equal to 96 rattis,
a weight-standard that has continued up to our own time in the case of the Indian rupee.
The second innovation was the introduction of the jital in the billon currency of the time,
replacing the old dehliwal. Wright argued that 48 jitals made 1 tanga, i.e. 1 jital was equal
to 2 rattis, and thus he succeeded in relating both tang a and jital to the prevalent masha
and ratti weight-standard of gold that had been in use in northern India until the decimal
system was adopted in our own time. 38
One may, 39 however, differ from Wright in attributing Iltutmish's horseman type of
gold currency to Bengal, in which Iltutmish assumed the title of al-sultan al-muc azzam
and called himself al-Qutbi. 40 This type is hardly known in Bengal and even the reading of
the mint name Ba-Gawr is not above doubt. It could also be Nagawr in northern Rajasthan,
to which region the coin type really belongs. If this is acceptable, Iltutmish's horseman
type of silver coins needs not be attributed to Bengal either, because it bears Nagari letters,
a script that was hardly in use in Bengal. His use of the names of three cAbbasid caliphs,
al-Nasir, al-Zahir and al-Mustansir, may be connected with his investiture by the caliph,
as this was relevant to the political conditions of the time. However, it is difficult to make
a distinction of grades of exaltedness in the use of the titles al-mucazzam and al-aczam.
Iltutmish's billon coins show many varieties in Nagari legends; one remarkable variety
gives the name of the defeated Hindu ruler Sri Chahada Deva.
The coins of the four sons and one daughter of Iltutmish all continue his silver, bil-
lon and copper currencies on the same standard, although types are restricted. The first
three sons, Rukn al-Din Firuz Shah, Jalalat al-Din Radiyya and Mucizz al-Din Bahram
Shah, are not known to have issued gold coins. The first two keep their father's name
on the coins. The new mint of Lakhnawti appears on the coins of Radiyya. cAla' al-Din
Mascud and Nasir al-Din Mahmud restarted the gold currency. In the time of Mascud,
the name of the new caliph al-Mustacsim (1242-58) occupies the place of al-Mustansir
(1226-42). Mahmud's gold coins have greater artistic merit and their counterpart in silver

38
Wright, 1936, p. 72.
39
See Dani, 1955.
40
Wright, 1936, pp. 15-16.

420
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

is also known. A new mint of Bada'un appears in his time. The billon coins of these rulers
continued the jital varieties of bull-and-horseman, but sometimes the bull is replaced by
a name in Arabic, while the reverse has the Nagari name. Mahmud also issued tiny silver
mas has and half-pieces of billon currency.
Ghiyath al-Din Balban (1266-87) continued the gold and silver issues and struck at
mints other than the capital, viz. Alwar, Lakhnawti, Sultanpur and Fakhrabad. He was the
last to use the horseman type in billon, but he inaugurated a new bilingual type in mixed
metal. He also revived the 40-ratti piece of copper so popular in the time of Iltutmish.
While his successor Kay Qubad continued the gold coinage, he also issued 2- and 4-masha
pieces of silver coins. The copper issues of the next sultan, Kayiimarth, continued the 40-
ratti standard which was passed on to the Khaljis. Another change from the time of Balban
was in the referrence to the caliph in the legend. Earlier, it used to be ft c ahd al-imam
al-mustacsim, but after the latter's murder by the Mongols in 1258, the legend came to be
simply al-imam al-mustac sim.
Jalal al-Din Firuz Khalji (1290-96) continued the gold and silver tanga and also billon
and copper of Balbanid times. His son Rukn al-Din Ibrahim used his own name on the
obverse and that of his father on the reverse of silver coins. It is from the time of 0 Ala'
al-Din Muhammad Khalji (1296-1316) that there was an abundance of gold and silver
tang as because of his conquest of the Deccan. He started the new mint of Deogir. Old Delhi
was called Dar al-Islam. All reference to the caliph's name was given up, but he himself
assumed the titles of Sikandar al-thanf, yamfn al-khilafa and nasir amfr al-mu'minfn. He
was the first to issue square forms of coins. In copper, 40-ratti fulas were the principal
coins. The coinage of Qutb al-Din Mubarak Khalji (1316-20) is remarkable for the variety
of inscriptions. He usurped for himself the position of khalifat allah and designated himself
as al-imam al-ac zam khalifat rabb al-c alamfn. On the reverse of his coins he called himself
Sikandar al-zaman, yamfn al-khilafa and nasir amfr al-mu'minfn. From his time, the square
form of coins in gold, silver and billon become more common. Among the mints, we find
the addition of Qutbabad fort and the practice of recording Hazrat Dihli, or Hazrat Dar
al-Khilafa or Hazrat Dar al-Mulk. Qutbabad has been identified with Deogir. Nasir al-Din
Khusraw continued all the three currencies and adopted the title of walf amfr al-mu'minfn.
The coinage of the Tughluqids makes an interesting study. The gold and silver tankas
continued the older tradition by Ghiyath al-Din Tughluq (1320-25), but the obverse of his
coins has for the first time the title of al-sultan al-ghazz. 41 After the conquest of Telingana,
a commemorative medal gives the mint as Mulk-i Telingana. The second commemorative
medal was issued by him as suzerain of Bengal, in which the name of the sultan on the

41
Wright, 1936, pp. 58-63, 80-1.

421
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

obverse has the title of al-sultan al-aczam, while the reverse gives to the local ruler of
Bengal, Nasir al-Din Ibrahim Shah, the title of al-sultan al-mucazzam. Ghiyath al-Din's
billon coins have bilingual legends in Arabic and Nagarl.
Muhammad b. Tughluq (1325-51) has been designated as the 'prince of moneyers'
because of the variety of his coins and the great interest that he took in their minting. No
less than fifty-one different types of his coins have been described. These are subdivided
into four classes: (a) those struck in memory of his father and recording only the latter's
name, i.e. commemorative issues; (b) those issued in his own name, i.e. normal issues;
(c) the forced currency, i.e. copper or brass token issues; and (d) those recording only the
names of the caliph, i.e. caliphal issues.
In class (a), the title of the late sultan is given as al-sultan al-sayyid al-shahfd al-ghazf.
In one coin, the mint is read as Mulk Macbar. In class (b) coins, the obverse has the kalima
and the reverse shows the title of the ruler as al-mujahid ft sabfl allah, and its margin has
the names of the first four caliphs, i.e. Abu Bakr, cumar, cuthman and cAli. The signif-
icant feature of this period was the assimilation of South India into the monetary system
of the Tughluqids. For this purpose, special types of coins were issued, e.g. the coin of
the Macbar mint with an extraordinary weight of 245 grains, which is said to represent a
4-pagoda piece, well-known in the fourteenth century in the kingdom of the Hoysalas of
South India. Muhammad b. Tughluq's novel issues include one 16-rattis (i.e. one-sixth)
higher in weight in gold, called dznar, and the other 16-rattis or one-sixth lower than the
tanga in silver, called c adlf. He also issued a dumpy type of coin having traces of South
Indian influence at a time when the capital was transferred to Deogir (or Dawlatabad).
The silver coins of Muhammad b. Tughluq are scarce. He commenced his reign with a
single billon type of 32-ratti weight and the title of al-mujahid ft sabfl allah. He is known
to have issued 4-jital pieces (known as chawghanf) and also 8-jital pieces (hashtganf). But
from a numismatic point of view, the most important is the introduction of 80-ratti billon
in 1326, a mixed-metal currency probably issued as a regular currency for the Deccan, and
an issue to replace the original silver tanga. According to Wright, this was the first step in
the degradation of the silver tanga of Iltutmish.
There are several other varieties of billon issues of Muhammad b. Tughluq, but his
copper currency is scarce. This experiment in debasement probably led to the issue of
copper or brass forced currency - class (c) - in 1329, with the name of the coin given as
tankah-i-panjahganf, dirham-duvazdihganf, hashtganf and duganf, thus fixing the value by
law (sharcf). It appears that this novel experiment aimed at bringing into line the different
monetary systems that then prevailed in northern India and the Deccan, but the attempt
collapsed because of market forgery, so that the token currency failed. It was withdrawn,

422
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

and the value was paid by the government, proving that the experiment was not due to
lack of gold reserves but for some other reason; the chief aim might have been to balance
the monetary standard throughout the empire. The last class of Muhammad b. Tughluq's -
coins class (d) - are the caliphal issues, in which he acknowledged the cAbbasid shadow
caliphs in Mamluk Cairo, first al-Mustakfi and later al-Hakim, after he received the patent
and insignia, thus assuring his legal sovereignty.
Rruz Shah Tughluq (1351-88) continued the name of the caliph on his gold coins right
up to 1383, recording the name of Abu '1-c Abbas Ahmad, then of Imam Abu '1-Fat'h and
finally Abu cAbd Allah. We also have his gold and silver coins without the name of the
caliph, in which he called himself na'ib amfr al-mu'minfn. Rruz Shah also associated his
sons with himself and allowed them to issue coins, first Fat'h Khan, then Rruz Shah Zafar
and finally Muhammad Shah. Gold and silver tangas of the earlier time continued in this
period, though silver was rare. The most popular coin was the 32-ratti billon. However, it
seems that Rruz Shah further debased the tanga and his new standard was maintained by
his successors right up to the time of Sikandar Lodi (1489-1517). 42
Rruz Shah's subordinate rulers and successors included his three sons and five grand-
sons: Fat'h Khan (d. 1371) and the latter's son Ghiyath al-Din Tughluq II; Rruz Shah
Zafar, who probably issued coins after the death of Fat'h Khan, and his son Abu Bakr, who
succeeded his father in 1389; and finally Muhammad the third son. Muhammad had two
sons: Sikandar, who ruled for a few days; and Mahmud, who had a long reign (1394-1412),
with the interruption of Timur's invasion, and a rival in Nusrat Shah, a second son of Fat'h
Khan. All of them issued gold and billon coins except Sikandar Shah, whose gold coinage
is not known, while Muhammad and Mahmud also issued silver coins. Mahmud's stormy
period of rule, because of the conflict with his cousin Nusrat Shah, led to a weakening of
the empire. After Timur's invasion of 1398-9, in particular, there was a depletion of the
state treasury and of precious metals, hence a hiatus in the issue of gold and silver currency.
For ten years, the government was left in the hands of leading nobles, who refrained from
issuing coins in their own name. Before the time of Timur, the normal gold to silver ratio
was 1: 10, i.e. 10 silver tangas exchanged for 1 gold tanga, and the copper to silver ratio
was 80: 1, but later the value of silver to copper fell.
When Mahmud died in 1413, his successors continued to issue the old currency of Rriiz
and Mahmud posthumously until, in 1429, Mubarak Shah Sayyid initiated a coinage of his
own. There was no change in the gold, silver and copper currency of this period, but gold
and silver became scarce right through the time of the succeeding Lodi rulers. Eighty-ratti
billon became common from the time of Bahliil Lodi, a revival of the currency from the

42
Wright, 1936, p. 219.

423
ISBN 978-92-3-103467-1 Contents AFGHANISTAN, PAKISTAN ...

time of Muhammad b. Tughluq; it is this currency which came to be known as Bahliili.

In the time of Sikandar Lodi, it was the billon that remained common, but by this time its
silver metal content was reduced. Twenty of these were equivalent to a 180-grain silver
coin of the time of the Sayyids. It is the Bahliili and Sikandari tangas that were available in
large quantities in the time of the last Lodi ruler, Ibrahim b. Sikandar (1517-26), and this
monetary system was passed on to his successors.

424
ISBN 978-92-3-103467-1 Contents CONCLUSION

CONCLUSION
C. E. Bosworth

The end of our period, i.e. the fifteenth century, can be seen in retrospect to mark a profound
change in the balance of power between the sedentary peoples and cultures of Central
Asia and Inner Eurasia and the nomad elements there. Previously, the advantage had lain
with the latter, for the steppe peoples had a comparative lack of interest in settled life and
agricultural activity in the settled lands, and therefore a superior mobility, enabling them
to raid into the lands of ancient culture - those of China, Afghanistan and India, and of
the Middle East - with little hindrance. Mass incursions by elements from Inner Eurasia
virtually ceased with the Mongols, for Timur's military following in the later fourteenth
century was basically not a tribal one and his focus of interest was in the settled lands of
the Middle East rather than the steppes and forest lands of Inner Asia.
With the advent of the sixteenth century, powerful states arose along the western and
southern fringes of Islamic Central Asia: the Christian Muscovite principality, the Mus-
lim Uzbek Shaybanids in Transoxania, the Safavids in Iran and the Mughals in northern
India. The acquisition of firearms by these new powers meant that the ancient superior
mobility of the steppe peoples was effectively counteracted. Hence great movements of
peoples from the heart of Asia outwards into the settled lands of Western Asia, South Asia
and China largely ended. Instead, a reverse process now began, whereby militarily supe-
rior outside forces started to affect the general political, cultural and ethnic situation of
the Asian heartland, seen, for example, in Imperial China's extension of its authority west-
wards towards Xinjiang in the eighteenth century (a process already begun under the early
Ming emperors three centuries before). Imperial Russia, by now free from the remnants
of control by the Turco-Mongol Golden Horde and its epigoni, began in the seventeenth
century to expand across Siberia to the Pacific coast and subsequently into Turkistan as far
as the Pamirs. Only such inward-looking and less accessible regions as Mongolia and Tibet
were to remain immune from such pressures. All these events will, however, be treated in
Volumes V and VI of this History.

425
ISBN 978-92-3-103467-1 Contents MAPS

MAPS
N.B. Modern cities such as Tehran, Ashgabat, Dushanbe, Islamabad, Bishkek, Almaty and
Ulan Bator are inserted in order to give some guidance as to the administrative geography.

UM

DAS,_. .r.

MAP la. Central Asia in the early Islamic period (c. 800).

426
ISBN 978-92-3-103467-1 Contents MAPS

MAP 1b. (Continued.)

427
ISBN 978-92-3-103467-1 Contents MAPS

I
w..
0
~

"' ~

MAP 2. The Uighur Khanate (745-840) and its neighbours.

428
ISBN 978-92-3-103467-1 Contents MAPS

0
I

MAP 3. The Samanids of Transoxania and Khurasan and their dependants (c. 970).

429
ISBN 978-92-3-103467-1 Contents MAPS

MAP 4a. The empire of the Ghaznavids (c. 1030).

430
 ISBN 978-92-3-103467-1 Contents MAPS

\
•

ZAB ULl'S fA 1 •
,--)
l#tHPCHA.J/
... I
,• <iii .......
I

MAP 4b. (Continued.)

431
ISBN 978-92-3-103467-1 Contents MAPS

MAP 5a. The Ghurids and their conquests and the Khwarazm Shahs (c. 1200).

432
 ISBN 978-92-3-103467-1 Contents MAPS

GOB I DESEFIT

.
T
..
,

_/'
t
HBE"T

MAP 5b. (Continued.)

433
ISBN 978-92-3-103467-1 Contents MAPS

,..•
I

'

MAP 6a. The Mongol empire in the thirteenth century.

434
ISBN 978-92-3-103467-1 Contents MAPS

~
~

z ,,,. •
; ""

41,( l r. ;
'111 ,,,,.
...

MAP 6b. (Continued.)

435
ISBN 978-92-3-103467-1 Contents MAPS

MAP 7. Northern and central India under the Delhi Sultans and the western Himalayan states.

436
ISBN 978-92-3-103467-1 Contents MAPS

MAP 8a. The Middle East, Central Asia and northern India at the time of Timur's rise to power (c.
1370).

437
 ISBN 978-92-3-103467-1 Contents MAPS

MAP 8b. ( Continuted.)

438
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

BIBLIOGRAPHY AND REFERENCES

ABBREVIATIONS

AEMAe =Archivum Eurasiae Medii Aevi, Budapest

AOHung =Acta Orientalia Academiae Scientiarum Hungaricae, Budapest
APAW =Abhandlungen der Preussischen Akademie der Wissenschaften, Berlin
A UDT<;FD =Ankara Universitesi Dil-Tarih-<;o grafya FakUltesi Dergisi, Ankara
BSOAS = Bulletin of the School of Oriental and African Studies, London
CAJ =Central Asiatic Journal, The Hague
EI2 . 1960- =The Encyclopaedia of Islam, 2nd ed., Leiden
Eir. 1980- = The Encyclopaedia Iranica, London/Costa Mesa, Calif.
EW = East and West, Rome
GJ = The Geographical Journal, London
HJAS =Harvard Journal of Asiatic Studies, Cambridge, Mass.
IC= Islamic Culture, Hyderabad, Deccan
IQ= Islamic Quarterly, London
Isl. =Der Islam, Berlin
JA= Journal asiatique, Paris
JAOS =Journal of the American Oriental Society, New Haven, Conn.
JASB =Journal of the Asiatic Society of Bengal, Calcutta
INES= Journal of Near Eastern Studies, Chicago
JRAS =Journal of the Royal Asiatic Society, London
JSFOu =Journal de la. Societe Finno-ougrienne, Helsinki
KSIIMK =Kratkie soobshcheniya Instituta Istorii Material'noy kul'tury, Moscow
MDAFA =Memoires de la Delegation archeologique fran<;aise en Afghanistan, Paris
MSFOu =Memoires de la Societe Finno-ougrienne, Helsinki
MW= Muslim World, Hartford, Conn.
NC =Numismatic Chronicle, London
SBAW = Sitzungsberichte der K. K. Akademie der Wissenschaften, Vienna
SI= Studia Islamica, Paris
SPAW= Sitzungsberichte der Preussischen Adademie der Wissenschaften, Berlin
TP = T'oung-Pao, Leiden

439
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

UAJ= Ural-altaische Jahrbiicher

ZDMG = Zeitschrift der Deutschen Morgenlandischen Gesellschaft, Leipzig/Berlin/Wiesbaden

CHAPTER 1
AZIZI, M. 1938. La domination arabe et l'epanouissement national en Iran. Paris.
BARTHOLD, W. 1935. ZwolfVorlesungen Uber die Geschichte Mittelasiens. German tr. T. Menzel.
Berlin. French tr. M. Donskis. Histoire des Tures d'Asie Centrale. Paris, 1945.
- - . 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr. T. Minorsky,
ed. with further addenda and corrigenda by C. E. Bosworth. London.
BELENITSKY, A. M.; BENTOVICH, I. B.; BOLSHAKOV, 0. G. 1973. Srednevekoviy gorod Sred-
ney Azii. Leningrad.
BOLSHAKOV, 0. G. 1976. Khronologiya vosstaniya Mukanny. In: Istoriya i kul'tura narodov
SredneyAzii, pp. 90-8. Moscow.
CRONE, P. 1980. Slaves on Horses. The Evolution of the Islamic Polity. Cambridge.
DANIEL, E. L. 1979. The Political and Social History of Khurasan under Abbasid Rule 747-820.
Minneapolis/Chicago.
EI2 , arts. Khwarazm; Khurasan; Kutayba b. Muslim; Ma wara' al-nahr.
Elr, art. Central Asia. iv. In the Islamic Period up to the Mongols.
FRYE, R. N. 1975. The Golden Age of Persia. The Arabs in the East. London.
GIBB, H. A. R. 1923. The Arab Conquests in Central Asia. London.
KADIROVA, T. 1965. Iz istorii krestyanskikh vosstaniy v Maverannahre i Khorasane v VIII-
nachale IX veka. Tashkent.
LASSNER, J. 1980. The Shaping orAbbasid Rule. Princeton, N. J.
MOTTAHEDEH, R. 1975. The cAbbasid Caliphate in Iran. In: The Cambridge History of Iran,
Vol. 4, pp. 57-89.
NAGEL, T. 1972. Untersuchungen zur Entstehung der Abbasidischen Kalifats. Bonn.
NARSHAKHI. 1954.Tarrkh-i Bukhara. English tr. and commentary R. N. Frye, The History of
Bukhara. Cambridge, Mass.
PIPES, D. 1981. Slave Soldiers and Islam. The Genesis of a Military System. New Haven/ London.
SADIGHI, G. H. 1938. Les mouvements religieux iraniens au Ir et au IIr siecle de I'hegire. Paris.
SHABAN, M.A. 1970. The cAbbasid Revolution. Cambridge.
SKRINE, F. H.; Ross, E. D. 1899. The Heart of Asia. A History of Russian Turkestan and the
Central Asian Khanates from the Earliest Times. London.
SPULER, B. 1952. Iran infriihislamischer Zeit. Politik, Kultur, Verwaltung und offentliches Leben
zwischen der arabischen und seldschukischen Eroberung 633 bis I 055. Wiesbaden.
WELLHAUSEN, J. 1902. Das arabische Reich und sein Sturz. Berlin. English tr. M. G. Weir, The
Arab Kingdom and its Fall. Calcutta, 1927.
ZARRINKUB, cABD AL-HUSAIN. 1975. The Arab Conquest of Iran and its Aftermath. In: The
Cambridge History of Iran, Vol. 4, pp. 1-56.

CHAPTER2
AMORETTI, B. S. 1975. Sects and Heresies. In: The Cambridge History of Iran, Vol. 4, pp.
481-519.
ANON. 1976. Tarrkh-i Srsten. English tr. Milton R. Gold, The Tarikh-e Srstan. Rome.
AL-ASHcARI. 1929-30. Maqalet al-Islamiyyrn. Ed. H. Ritter. Istanbul, 2 vols.

440
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

AL-BAGHDAD!. 1328/1910. Al-Farq bayn al-jiraq. Ed. M. Badr. Cairo. Partial English tr. A. S.
Halkin, Moslem Shisms and Sects, Part 2. Tel Aviv, 1935.
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BOSWORTH, C. E. 1968. Slstan under the Arabs. From the Islamic Conquest to the Rise of the
Saffarids (30-250165I-864). Rome.
- - . 1975. The Tahirids and Saffarids. In: The Cambridge History of Iran, Vol. 4, pp. 90-135.
- - . 1977. The Medieval History of Iran, Afghanistan and Central Asia. London.
- - . 1978-9. The Interaction of Arabic and Persian Literature and Culture in the lOth and Early
1lth Centuries. In: al-Abhath, Vol. 27, pp. 59-75, reprinted in C. E. Bosworth, Medieval Arabic
Culture and Administration (London, 1982), no. VIII.
- - . 1994. The History of the Saffarids of Sistan and the Maliks of Nimruz (247186I to
9491I 542-3. Costa Mesa, Calif./New York.
DAFfARY, F 1990. The Ismac1z1s. Their History and Doctrines. Cambridge.
- - . 1993. A Major Schism in the Early Ismacm Movement. SI, Vol. 77, pp. 123-39.
DANIEL, E. L. 1979. The Political and Social History of Khurasan under Abbasid Rule, 747-820.
Minneapolis/Chicago.
2
EI , arts. Ismaciliyya; Khurramiyya; Ma wara' al-nahr.
Elr, arts. cAbbasid Caliphate in Iran; cAlids of Tabarestan, Daylaman, and Gnan; Babak KorramI;
Carmathians.
FRYE, R. N. 1975. The Golden Age of Persia. The Arabs in the East. London.
GARDIZI, cABD AL-HAYY B. AL-DAHHAK. 1347/1968. Zayn al-akhbar. Ed. cAbd al-Hayy
HabibI. Tehran.
HALM, H. 1982. Die islamische Gnosis. Zurich/Munich.
IBN AL-NADIM, ABU 'L-FARAJ MUHAMMAD B. IsHAQ AL-WARRAQ. 1973. Fihrist. Ed. M. R.
Tajaddud, 2nd ed. Tehran.
MADELUNG, W. 1965. Der Imam al-Qasim ibn Ibrahim and die Glaubenslehre der Zaiditen.
Berlin.
- - . 1988. Religious Trends in Early Islamic Iran. Albany, New York.
MINORSKY, V. 1932. La domination des Dailamites. Paris.
NARSHAKHI. 1954. Tarlkh-i Bukhara. English tr. and commentary R. N. Frye, The History of
Bukhara. Cambridge, Mass.
AL-NASHI' AL-AKBAR. 1971. Masa'il al-imama. Ed. J. van Ess. Beirut.
AL-NAWBAKHTI. 1931. Firaq al-Shfa. Ed. H. Ritter. Istanbul.
NIZAM AL-MULK. 1978. Siyar al-mulak (Siyasat-nama). English tr. H. Darke, The Book of Gov-
ernment or Rules for Kings, 2nd ed. London.
AL-QUMMI. 1963. Al-Maqalat wa 'l-jiraq. Ed. M. J. Mashkilr. Tehran.
AL-RAZI. 1988. Kitab al-Zina. Ed. cA. Sallum al-Samarra'I in his al-Ghuluww wa 'l-jiraq al-
ghaliya, 3rd ed. Baghdad.
RYPKA, J., et al. 1968. History of Iranian Literature. Dordrecht.
SADIGHI, G. H. 1938. Les mouvements religieux iraniens au Ir et au IIr siecle de I'hegire. Paris.
AL-SHAHRASTANI. Al-Mila! wa 'l-nihal. Partial English tr. A. K. Kazi and J. G. Flynn, Muslim
Sects and Divisions. London, 1984. Partial French tr. D. Gimaret and G. Monnot, Livre des
religions et des sectes. Paris, 1986.
SHARON, M. 1983. Black Banners from the East. The Establishment of the c Abbasid State. Incu-
bation of a Revolt. Jerusalem/Leiden.

441
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

SPULER, B. 1952. Iran infrUh-islamischer Zeit. Politik, Kultur, Verwaltung und offentliches leben
zwischen der arabischen und seldschukischen Erorberung 633 bis 1055. Wiesbaden.
S1ERN, S. M. 1960. The Early Ismac·m Missionaries in North-West Persia and in Khurasan and
Transoxania. BSOAS, Vol. 23, pp. 56-90.
- - . 1971. Yacqilb the Coppersmith and Persian National Sentiment. In: Iran and Islam, in Mem-
ory of the late Vladimir Minorsky, pp. 535-55. Ed. C. E. Bosworth. Edinburgh.
AL-TABAR!, AB-0 JAcFAR MUHAMMAD B. JARIR. 1879-1901. Tar!kh al-Rusu! wa'l-mulak. Ed. M.
J. de Goeje et al., Leiden. 15 vols. in 3 series. English tr. by various scholars, The History of
al-Tabar!. Albany, New York, 1985-.
WELLHAUSEN, J. 1975. The Religio-Political Factions in Early Islam. English tr. R. C. Ostle and
S. M. Walzer. Amsterdam.
YARSHA1ER, E. 1983. Mazdakism. In: The Cambridge History of Iran, Vol. 3 (2), pp. 991-1024.
ZARRINKUB, H. 1355/1976. Tar!kh-i Iran bacd az Islam, 2nd ed. Tehran.

CHAPTER 3
ABU 'L-FADL BAYHAQI. 1969. Tar!kh-i Mascadr (1030-1041). Russian tr. A. K. Arends.
Moscow.
AGAJANOV, S. G. 1969. Ocherki istorii oguzov i turkmen Sredney Azii IX-XIII vv. Ashgabat.
ANON. 1930. Hudad az_cazam. Tumanskiy manuscript. Leningrad.
AL-BALADHURI. 1932. Futah al-buldan. Cairo.
BARTOL'D, V. V. 1898. Turkestan v epokhu mongol'skogo zavoevaniya, Part I, Texts. St Peters-
burg.
--.1973. Sochineniya, Vol. 8. Moscow.
AL-BIRDNI, ABU 'L-RAYHAN. 1966. Geodesy. Tashkent.
DUNLOP, M. D. 1954. The History of the Jewish Khazars. Princeton, N. J.
FAKHR AL-DIN MUBARAKSHAH. 1927. Tarrkh. Ed. E. D. Ross. London.
IBN AL-ATHIR. 1870. Al-Kami!, Vol. 8. Ed. C. J. Tornberg. Uppsala/Leiden.
IBN AL-FAQIH, Akhbar al-buldan. Photocopy of manuscript. LOIVAN (Leningrad Branch of the
Institute of Oriental Studies of the Academy of Sciences). IFV-202.
AL-IDRISI. Nuzhat al-mushtaq fi 'khtiraq al-afaq, pp. 108-9. Manuscript. Saltykov-Shchedrin
State Public Library, ArNS-176.
Istoriya Kazakhskoy SSR. S drevneyshykh vremen do nashykh dney. 1979. Alma-Ata.
JUWAYNI, cATA-MALIK. 1958. Tar!kh-i Jahan-gushay. English tr. J. A. Boyle, The History of
the World-Conqueror. Manchester. 2 vols.
JUZJANI. 1881-97. Tabaqat-i Nasir!. English tr. H. G. Raverty. London. 2 vols.
LEVINA, L. M. 1972. Keramika nijney i sredney Syrdar'i v pervom tysyacheletii nasheyery.
Moscow.
AL-KASHGHARI, MAHMUD. 1917-19. Diwan lughat al-turk. Ed. Kilisli RiFat Bey. Istanbul.
KONONOV, A. N. 1958. Rodoslovnaya turkmen. Work of Abu 'I-Ghazi, Khan of Khiva.
Moscow/Leningrad.
KUMEKOV, B. E. 1972. Gosudarstvo kimakov IX-XI vv. po arabskim istochnikam. Alma-Ata.
KUSEEV, R. G.; SHITOVA, S. N. 1963. Bashkiry. Istoriko-etnograficheskiy ocherk. Ufa.
MARWAZI. 1942. Shara/ al-Zaman Tahir Marvazf on China, the Turks and India. Ed. and English
tr. V. Minorsky. London.
AL-MAscuDI. 1894. Kitab al-Tanbih wa 'l-ishraf Ed. M. J. de Goeje, Leiden. Tr. Carra de Vaux.
Paris, 1897.

442
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

MIRKHWAND, Rawdat al-safa. 1841. Section on Chinggis Khan ed. E. Quatremere, Vie de
Djengbiz-khan. Paris.
MISKAWAYH. 1921. Tajarib al-umam. In: The Eclipse of the cAbbasid Caliphate, Vol. 2. Oxford.
AL-NASAWI. 1895. Sr:rat al-Sultan fatal al-Din Mangubirtf. French tr. 0. Houdas. Paris.
NE~RI, M. 1949. Kitab-i Cihttn-numa. Ankara.
PELLIOT, P. 1920. Apropos des Comans.JA, Series 11, Vol. 15. Paris.
Polnoe sobranie russkikh letopisey. 1962. Vol. 1. Moscow/Leningrad.
Povest' vremennykh let. 1950. Vol. 1. Moscow/Leningrad.
PRITSAK 0.1953. Der Untergang des Reiches des Oguzischen Yabgu. In: Fuad KoprUZU
armagam. Melanges Fuad KoprUZU. Istanbul.
RASHID AL-DIN. Jamie al-tawar!kh. Photocopy of British Museum manuscript. LOIVAN
(Leningrad Branch of the Institute of Oriental Studies of the Academy of Sciences) SSSR.
IFV-242.
Tarrkh-i Al-i Seljaq. Manuscript. LOIVAN (Leningrad Branch of the Institute of Oriental Studies
of the Academy of Sciences). D-116.
TOLSTOY, S. P. 1948. Po sledam drevnekhorezmiyskoy tsivilizatsii. Moscow.
AL-YAcQUBI. 1892. Kitab al-Buldan. Leiden.

CHAPTER4
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BELENITSKIY, A. M. 1960. Drevneye izobrazitel'noe iskusstvo i 'Shakhname'. (Mejdunarodniy
Kongress Vostokovedov. Doklady delegatsii SSR). Moscow.
BOSWORTH, C. E. 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual.
Edinburgh/New York.
BROWNE, E. G. 1908. A Literary History of Persia, Vol. 1: From the Earliest Times until Firdaws{.
London.
D'YAKONOV, M. M. 1951. Obraz Siyavusha v sredneaziatskoy mifologii. KSIIMK, Vol. 10.
EI2 , arts. Khwarazm; Khwarazm-Shahs; Ma wara' al-nahr; Samanids.
Elr, arts. Bukhara, ii. From the Arab Invasion to the Mongols; Central Asia. iv. In the Islamic
Period up to the Mongols.
FRYE, R. N. 1965. Bukhara, the Medieval Achievement, Norman, Okla.
- - . 1975a. The Golden Age of Persia. The Arabs in the East. London.
- - . 1975b. The Samanids. In: The Cambridge History of Iran, Vol. 4, pp. 136-61.
- - . 1996. The Heritage of Central Asia. From Antiquity to the Turkish Expansion. Princeton, N.
J.
IBN FADLAN. 1939. Risala. German tr. and commentary A. Z. V. Togan, Ibn Fadlan's Reise-
bericht. Abh. ftir die Kunde des Morgenlandes, XXIV, 3. Leipzig.
NARSHAKHI. 1954. Tar!kh-i Bukhara. English tr. and commentary R. N. Frye, The History of
Bukhara. Cambridge, Mass.
NEGMATOV, N. N. 1977. Gosudarstvo Samanidov (Maverannahr i Khorasan v IX-X vv.
Dushanbe.
RYPKA, J., et al. 1968. History of Iranian Literature. Dordrecht.
SPULER, B. 1952. Iran infrUhislamischer Zeit. Politik, Kultur, Verwaltung und offentliches Leben
zwischen der arabischen und seldschukischen Eroberung 633 bis I 055. Wiesbaden.

443
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

CHAPTER 5
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BOSWORTH, C. E. l962a. The Imperial Policy of the Early Ghaznawids. Islamic Studies. Journal
of the Central Institute of Islamic Research. Karachi, Vol. 1, 3, pp. 49-82.
- - . l962b. The Titulature of the Early Ghaznavids. Oriens, Vol. 15, pp. 210-33.
- - . 1963. The Ghaznavids. Their Empire in Afghanistan and Eastern Iran 994-I 040. Edinburgh;
2nd ed., Beirut, 1973.
- - . 1965. Notes on the Pre-Ghaznavid History of Eastern Afghanistan. IQ, Vol. 9, pp. 12-24.
- - . 1968a. The Political and Dynastic History of the Iranian World (A.D. 1000--1217). In: The
Cambridge History of Iran, Vol. 4, pp. 1-202.
- - . 1968b. The Book of Curious and Entertaining Information. The Lata'if al-Macari/ of
Thacalibr, English tr. with introduction and notes. Edinburgh.
- - . 1970. The Turks in the Islamic Lands up to the Mid-llth Century. In: Cl. Cahen (ed.),
Philologiae Turcicae Fundamenta, Vol. 3. Wiesbaden. 3 vols.
- - . 1975. The Early Ghaznavids. In: The Cambridge History of Iran., Vol. 4, pp. 162-97.
- - . 1977. The Later Ghaznavids, Splendour and Decay. The Dynasty in Afghanistan and North-
ern India I 040-I I 86. Edinburgh.
- - . 1994. The History of the Saffarids of Sistan and the Maliks of Nimruz (247186I to
9491I 542-3). Costa Mesa, Calif./New York.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual. Edin-
burgh/New York.
BOSWORTH, C. E.; CLAUSON, Sir Gerard. 1965. Al-XwarazmI on the Peoples of Central
Asia.JRAS, pp. 2-12.
BROWNE, E. G. 1906. A Literary History of Persia, Vol. 2: From Firdaws{ to Sacd{. London.
BULLIET, R. W. 1972. The Patricians of Nishapur. A Study in Medieval Islamic Social History.
Cambridge, Mass.
GANDJEI, T. 1986. Turkish in Pre-Mongol Persian Poetry. BSOAS, Vol. 49, pp. 67-75.
HABIB, M. 1951. Sultan Mahmud ofGhaznin. 2nd ed. Delhi.
KOPRULUZADE, M. F. 1934. Gazneliler devrinde ttirk §i'ri. In: Turk dili ve edebiyatl hakkinda
ara~txxxirmalar, pp. 6-32. Istanbul.
MEIER, F. 1976. Abu Sacrd-i Abul-ljayr (357-440:967-I049). Wirklichkeit and Legende.
Tehran/Liege.
MINORSKY, V. 1942. Shara/ al-Zaman Tahir Marvazr on China, the Turks and India. London.
NAZIM, M. 1927. The Hindu Shahiya Kingdom of Ohind. JRAS, pp. 485-95.
- - . 1931. The Life and Times of Sultan Mahmud ofGhazna. Cambridge.
- - . 1933. The Pand-Namah of Subuktigin. JRAS, pp. 605-28.
PRITSAK, 0. 1950. Karachanidische Streitfragen 1-4. Oriens, Vol. 2, pp. 209-28.
- - . 1953. Der Untergang des Reiches des Oguzischen Yabgu. In: Fuad KoprUZU armagam.
Melanges Fuad KoprUZU. Istanbul.
- - . 1953-4. Die Karachaniden. Isl., Vol. 31, pp. 17-68.
RAY, H. C. 1931-6. The Dynastic History of Northern India (Early Mediaeval Period). Calcutta.2
vols.
RYPKA, J., et al. 1968. History of Iranian Literature. Dordrecht.
SACHAU, E. 1873. Zur Geschichte und Chronologie von Khwarazm. SBAW, phil.-hist. Classe,
Vol. 73, pp. 471-506, Vol. 74, pp. 285-330.

444
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

SCHLUMBERGER, D., et al. 1978. Lashkari Bazar. Une residence royale ghaznevide et ghoride.
Parts IA, lB. MDAFA, Vol. 18.
S1ERN, S. M. 1949. Ismacnr Propaganda and Fatimid Rule in Sind. IC, Vol. 23, pp. 298-307.
TOLSTOY, S. P. 1953. Auf den Spuren der altchoresmischen Kultur. German tr. 0. Mehlitz. East
Berlin.

CHAPTER 6

BARTHOLD, V. 1928. Turkestan down to the Mongol Invasion. 2nd ed. tr. from the original
Russian and revised by the author with the assistance of H. A. R. Gibb. London.
BOSWORTH, C. E. 1965. The Islamic Dynasties. A Chronological and Genealogical Handbook.
Edinburgh.
- - . 1973. The Ghaznavids. Their Empire in Afghanistan and Eastern Iran. 994:I040. 2nd ed.
Beirut.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Man-
ual.Edinburgh/New York.
DAVIDOVICH, E. A. 1954. Numizmaticheskie materialy dlya istorii razvitiya feodal'nykh
otnosheniy v Sredney Azii pri Samanidakh. Trudy Akademii nauk Tadzhikskoy SSR, Vol. 27,
pp. 69-117.
- - . 1957. Numizmaticheskie materialy dlya khronologii i genealogii sredneaziatskikh
Karakhanidov. In: Numizmaticheskiy sbomik, Part 2 (Trudy gosudarstvennogo istoricheskogo
muzeya, 36), pp. 91-119. Moscow.
- - . 1968. 0 dvukh karakhanidskikh kaganatakh. Narody Azii i Afriki, Vol. 1, pp. 67-7 6. Moscow.
- - . 1970. Klad saganianskikh monet vtoroy chetverti XI. Pis'mennye pamyatniki Vostoka. Ezhe-
godnik, pp. 91-119. Moscow, 1968.
- - . 1977. Voprosy khronologii i genealogii Karakhanidov vtoroy poloviny XII v. In: Srednyaya
Aziya v drevnosti i srednevekov'e (istoriya i kul'tura), pp. 177-87. Moscow.
- - . 1978. 0 monetakh dikhkanov Ilaka kontsa X-nachala XI v. In: Drevnost' i srednevekov'e
narodov Sredney Azii (istoriya i kul 'tura ), pp. 80-100. Moscow.
- - . 1994. Poslednie Karakhanidy i Mukhammad b. Tekesh (numizmaticheskie korrektivy). Vos-
tochnoe istoricheskoe istochnikovedenie i spetsial'nye distsipliny, Vol. 2. Moscow.
GOLDEN, P. B. 1990. The Karakhanids and Early Islam. In: The Cambridge History of Early Inner
Asia. Ed. D. Sinor, pp. 348-70. Cambridge.
ISHANKHANOV, S.; KOCHNEV, B. D. 1979. Drevneyshie karakhanidskie monety. Istoriya mate-
rial'noy kul'tury Uzbekistana, Vol. 15, pp. 142-52. Tashkent.
KHADR, M. 1967. Deux actes de waqf d'un Qaraganide d' Asie Centrale. With an introduction by
Cl. Cahen. JA, Vol. 255, pp. 305-34.
KOCHNEV, B. D. 1979. Zametki po srednevekovoy numizmatike Sredney Azii, Pt. 2
(Karakhanidy). Istoriya material'noy kul'tury Uzbekistana, Vol. 15, pp. 120-38. Tashkent.
- - . 1982. Zametki po srednevekovoy numizmatike Sredney Azii. Part 4 (Karakhanidy). Istoriya
material'noy kul'tury Uzbekistana, Vol. 17, pp. 152-69. Tashkent.
- - . 1988. Zametki po srednevekovoy numizmatike Sredney Azii, Part 9 (Samanidy,
Karakhanidy, Anushteginidy). Istoriya material'noy kul'tury Uzbekistana, Vol. 22, pp.
191-205. Tashkent.
LANE-POOLE, S. 1895. The Mohammadan Dynasties. London. Russian tr. V. V. Bartol'd with
comments, Musul'manskie dinastii. Khronologicheskie i genealogicheskie tablitsy s istorich-
eskimi vvedeniyami. St Petersburg, 1899.

445
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

NASTICH, V. N.; KOCHNEV, B. D. 1988. K atributsii mavzoleya Shakh-fazil (epigraficheskie i

numizmaticheskie dannye). Epigrafika Vostoka, Vol. 24, pp. 68-77. Leningrad.
NIZAM AL-MULK. 1978. Siyar al-mulak (Siyasat-nama). English tr. H. Darke, The Book of Gov-
ernment or Rules for Kings, 2nd ed. London.
PRITSAK, 0. 1950. Karachanidische Streitfragen. 1-4. Oriens, Vol. 3. pp. 209-28.
- - . 1953. Die Karachaniden. Isl., Vol. 30, pp. 17-68.
- - . 1955. Qara. Studie zur ttirkischen Rechtssymbolik. In: Zeki Velidi Togan'a armagan, pp.
239-63. Istanbul.
ZAMBAUR, E. 1927. Manuel de genealogie et de chronologie pour l'histoire de !'Islam. Hanover.

CHAPTER 7
1. Original Sources
AL-c AZIMI. 1988. Ta'r!kh. Ed. Ali Sevim, Sel(:uklular ilgili bolUmler (H.
430-538=1038139-1143144). Ankara.
BARHEBRAEUS. 1932. Chronography. English tr. E. A. Wallis Budge. London.
BAYHAQI, ABU 'L-FADL. 1324/1945. Tar!kh-i Mascadr. Ed. Q. Ghani and cA. A. Fayyad.
Tehran.
AL-BUNDARI. 1889. Zubdat al-nusra. Ed. M. T. Houtsma. Leiden. Turkish tr. K. Burslan. Istan-
bul, 1943.
GARDIZI, cABD AL-HAYY B. AL-DAHHAK. 1928. Zayn al-akhbar. Ed. M. Nazim. Berlin.
AL-HUSAYNI, SADR AL-DIN cALI. 1933. Akhbar al-dawla al-saljaqiyya. Ed. M. Iqbal. Lahore.

IBN AL-cADIM. 1976. Bughyat al-talab fita'r!kh Halab. Partial Turkish tr. Ali Sevim. Ankara.
IBN AL-ATHIR. 1965-7. Al-Kamilfi ta'r!kh. Beirut.
IBN FADLAN. 1939. Risala. Ed. and German tr. A. Z. V. Togan, lbn Fadlans Reisebericht. Leipzig.

IBN AL-JAWZI. 1938-41. Al-Muntazamfita'r!kh al-mulak wa 'l-umam. Hyderabad.

JUWAYNI, cATA-MALIK. 1958. Tar!kh-i Jahan-gushay. English tr. J. A. Boyle, The History of
the World-Conqueror. Manchester.
JUZJANI, MINHAJ-I SIRAJ. 1963-4. Tabaqat-i Nasir!. Ed. cAbd al-Hayy HabibL 2nd ed. Kabul.
English tr. H. G. Raverty. London 1881-99.
AL-KASHGHARI, MAHMUD. 1939-41. Diwan lughat al-turk. Facsimile text and Turkish tr.
Besim Atalay. Ankara. English tr. R. Dankoff and J. Kelly, Compendium of the Turkic dialects.
Cambridge, Mass., 1982-4.
MATTHEW OF EDESSA. 1962. Chronicle. Turkish tr. Hrant R. Andreasyan, Vekayiname
(952-1136) ve Papaz Grigor'un zeyli (1136-1162). Ankara.
MIRKHWAND. 1853-6. Rawdat al-safa'. Ed. Rida Qulf Khan. Tehran.
NARSHAKHI. 1954. Tarrkh-i Bukhara. English tr. and commentary R. N. Frye, The History of
Bukhara. Cambridge, Mass.
NISHAPDRI, ZAHIR AL-DIN. 1953. Saljuq-nama. Ed. Ismacn Afshar. Tehran.
RAWANDA, SADR AL-DIN. 1921. Rahat al-sudar. Ed. M. Iqbal. London.
SIBT IBN AL-JAWZI. 1968. Mir' at al-zaman. Partial Turkish tr. A. Sevim. Ankara.
AL-cUTBI. 1869. Al-Tar!kh al-Yamin!. With commentary by Shaykh Ahmad al-ManinL Cairo.
2. Studies
BARTHOLD, W. 1935. ZwolfVorlesungen Uber die Geschichte Mittelasiens. German tr. T. Menzel.
Berlin. French tr. M. Donskis, Histoire des Tures d'Asie Centrale. Paris, 1945.

446
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

- - . 1956. History of the Semirechye. In: Four Studies on the History of Central Asia. English tr.
V. and T. Minorsky, Vol. 1, pp. 73-165. Leiden.
- - . 1962. A History of the Turkman People. In: Four Studies on the History of Central Asia.
English tr. V. and T. Minorsky, Vol. 3, pp. 73-170. Leiden.
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BOSWORTH, C. E. 1968. The Political and Dynastic History of the Iranian World (A.D. 100-1217).
In: The Cambridge History of Iran, Vol. 5, pp. 1-202.
- - . 1973. The Ghaznavids, their Empire in Afghanistan and Eastern Iran, 994:I040, 2nd ed.
Beirut.
- - . 1977. The Later Ghaznavids: Splendour and Decay. The Dynasty in Afghanistan andNorth-
em India I 040-I I 86. Edinburgh.
- - . 1994. The History of the Saffarids of Sistan and the Maliks of Nimruz (247186I to
9491I 542-3). Costa Mesa, Calif./New York.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual. Edin-
burgh/New York.
BOWEN, H. 1957. Notes on Some Early Seljuqid Viziers. BSOAS, Vol. 20, pp. 105-10.
BULLIET, R. W. 1972. The Patricians of Nishapur. A Study in Medieval Islamic Social History.
Cambridge, Mass.
CAREN, Cl. 1949. Le Malik-Nameh et l'histoire des origines Seljukides. Oriens, Vol. 2, pp. 31-65.

EI2 , arts. Alp Arslan; Berk-Yaruk; Ghaznawids; Ghi1rids; Ilek-Khans; Kara Khitay; Khw arazm-
Shahs; Malik Shah; Muhammad b. Malik Shah; Nizam al-Mulk; Ra'is. 2. In the Sense of
'Mayor' in the Eastern Islamic Lands; Saldjuks; Sandjar.
Elr, arts. Al-e Borhan; CagrI Beg. - -
HARTMANN, A. 1975. Al-Nasir li-Dfn Allah (I I80-I225). Politik, Religion, Kultur in der spaten
cAbbasiden Zeit. Berlin/New York.
HORST, H. 1964. Die Staatsverwaltung der Groselaqen und orazmsahs (1038-1231). Wiesbaden.

KAFESOGLU, I. 1953a. Sultan Melik~ah devrinde BUyUk Se[(;uklu imparatorluu. Istanbul.

- - . 1953b. Dogu Anadolu'ya ilk Sel<;uklu aklm (1015-1021) ve tarihl ehemmiyeti. In: Fuad
KoprUZU armagam. Melanges Fuad KoprUZU, pp. 259-74. Istanbul.
- - . 1955. Se[(;uklu ailesinin men~e'i hakkxxxinda. Istanbul.
- - . 1956. Harezm~ahlardevleti tarihi (485-61711092-1220). Ankara.
- - . 1988. A History of the Seljuks. Ibrahim Kafesoglu's Interpretation and the Resulting Contro-
versy. English tr. and Introduction G. Leiser. Carbondale and Edwardsville, III.
KLAUSNER, C. L. 1973. The Seljuk Vezirate. A Study of Civil Administration
1055-1194.Cambridge, Mass.
KOYMEN, M. A. 1947. Btiytik Sel<;uklular imparatorlugunda Oguz isyani; and Btiytik Sel<;uklu
imparatorlugu tarihinde Oguz istilasi. AUDT<;FD, Vol. 5, pp. 159-73, 563-620, with German
tr.pp. 175-86,621-60.
- - . 1954. BUyUk Se[(;uklu imparatorlugu tarihi. Vol. 2, Ikinci imparatorluk devri. Ankara.
- - . 1957-8. Sel<;uklu imparatorlugunun kurulu§u. AUDT<;FD, Vol. 15-16.
- - . 1972. Alp Arslan ve zamam. Istanbul.
- - . 1976. Tugrul Beg ve zamam. Istanbul.
LAMBTON, A. K. S. 1953. Landlord and Peasant in Persia. A Study of Land Tenure and Land
Revenue Administration. London.

447
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

LAMBTON, A. K. S. 1957. The Administration of Sanjar's Empire as Illustrated in the cAtabat

al-kataba. BSOAS, Vol. 20, pp. 367-88.
- - . 1968. The Internal Structure of the Saljuq Empire. In: The Cambridge History of Iran, Vol.
5, pp. 203-82.
LOWICK, N. M. 1970. Seljuq Coins. NC, 7th ser., Vol. 10, pp. 241-51. Also in idem, Coinage and
History of the Islamic World. Variorum Reprints. London, 1990, no. VI.
MASON, H. 1972. Two Statesmen of Mediaeval Baghdad. Vizir Ibn Hubayra (499-560
AH/1105-1165 AD) and Caliph an-Nasir li Din Allah (553-622 AH/1158-1225 AD), The
Hague/Paris.
MER\:IL, E. 1980. Kirman Se[(;uklulan. Istanbul.
MILES, G. C. 1938. The Numismatic History of Rayy. New York.
NAZIM, M. 1931. The Life and Times of Sultan Mahmud ofGhazna. Cambridge.
PRITSAK, 0. 1953. Die Untergang des Reiches des Oguzischen Yabgu. In: Fuad KopriilU
armagam. Melanges Fuad Kopriilii, pp. 397-410. Istanbul. Also English tr. in idem, Studies
in Medieval Eurasian History. Variorum Reprints. London, 1981, no. XIX.
- - . 1953-4. Die Karachaniden. Isl., Vol. 31, pp. 17-68, also in idem, Studies in Medieval
Eurasian History. Variorum Reprints. London 1981, no. XVI.
RASONYI, L. 1939. Sel<;uk admm men§e'ine dair. Belleten, Vol. 3, pp. 377-84.
SACHAU, E. 1873. Zur Geschichte und Chronologie von Khwarazm. SBAW, phil.-hist. Classe,
Vol. 73, pp. 471-506; Vol. 74, pp. 285-330.
SANAULLAH, M. F. 1938. The Decline of the Saljuqid Empire. Calcutta.
SOMER, F. 1967. Oguzlar (Tiirkmenler). Ankara.
TOGAN, A. Z. V. 1946. Umumf Tiirk tarihine giri§.
TOLSTOY, S. P. 1953. Auf den Spuren der altchoresmischen Kultur. German tr. 0. Mehlitz. East
Berlin.
TURAN, 0. 1965. Se[(;uklular tarihi ve Tiirk-Islam medeniyeti. Ankara.
- - . 1971. Se[(;uklular ve Islamiyet. Istanbul.
UZUN\:AR~ILI, I. H. 1941. Osmanlr devleti teskilatrna medhal. Istanbul.

CHAPTER 8
ANON. 1937. Hudad al-calam. The 'Regions of the World', a Persian Geography 372 A.H. 982
A.D. English tr. and commentary V. Minorsky. London.
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BOSWORTH, C. E. 1960. The Rise of the Karamiyyah in Khurasan. MW, Vol. 50, pp. 5-14.
- - . 1961. The Early Islamic History of Ghur. CAJ, 6, pp. 116-33.
- - . 1963. The Ghaznavids. Their Empire in Afghanistan and Eastern Iran 994: 1040. Edinburgh;
2nd ed., Beirut, 1973.
- - . 1968. The Political and Dynastic History of the Iranian World (AD. 1000-1217). In: The
Cambridge History of Iran, Vol. 5, pp. 1-202.
BOSWORTH, C. E. 1977. The Later Ghaznavids, Splendour and Decay. The Dynasty in
Afghanistan and Northern India 1040-1186. Edinburgh.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual. Edin-
burgh/New York.
CASIMER, M. J.; GLATZER, B. 1971. Sah-i Mashad. A Recently-Discovered Madrasa of the
Ghilrid Period in Garistan (Afghanistan). EW, N.S. Vol. 21, pp. 53-68.

448
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

EI2 , arts. Ghurids; Karramiyya.

Epigraphica Indo-Moslemica. 1911-12. Government of India Archaeological Survey. Calcutta.
HABIB, M. 1981. Collected Works. Politics and Society during the Early Mediaeval Period,Vol. 2.
Ed. K. A. Nizami.
HABIB, M.; NIZAMI, K. A. (eds.). 1970. A Comprehensive History of India, Vol. 5: The Delhi
Sultanate (A.D. /206-I526). New Delhi.
HABIBULLAH, A. B. M., 1961. The Foundation of Muslim Rule in India, 2nd. ed. Allahabad.
AL-ISTAKHRI. 1927. Kitab al-Masalik wa 'l-mamalik. Ed. M. J. de Goeje, 2nd ed. Leiden.
JUZJANI. 1881-99. The Tabakdt-i-Nasiri. A General History of the Muhammadan Dynasties of
Asia, including Hindustan, from A.H. I94 (8IO A.D.) to A. M. 658 (/260 A.D.). English tr. H. G.
Raverty. London. 2 vols.
KHAN, G. M. 1949. A History of Bahram Shah of Ghaznin. IC, Vol. 13, pp. 62-91, 199-235.
KOHZAD, A. A. 1951-4. Along the Koh-i Baba and Hari-Rud. Afghanistan, 6 parts. Kabul.
MARICQ, A; WIET, G. 1958. Le minaret de Djam. La decouverte de la capitale des sultans
ghorides (XIr-XIIr siecles). MDAFA, Vol. 16. Paris.
NAZIM, M. 1931. The Life and Times of Sultan Mahmud ofGhazna. Cambridge.
NIZAMI, K. A. 1961. Religion and Politics in India During the Thirteenth Century. Bombay.
- - . 1981. Supplement to Elliot and Dowson's History of India, Vol. 2: The Ghaznavids and the
Ghurids. New Delhi.
RAVERTY, H. G. 1881-3. Notes on Afghanistan and Baluchistan. London, repr. Quetta, 1976.2
vols.
TOGAN, A. Z. V., 1936. Die Schwerter der Germanen nach arabischen Berichten der 9.-11. Jts.
ZDMG, Vol. 90, pp. 19-37.
VERCELLIN, G. 1976. The Identification ofFiruzkilh: a Conclusive Proof. EW, N. S. Vol. 26,pp.
337-40.

CHAPTER 9
ANON. 1937. Hudad al-calam. The 'Regions of the World', a Persian Geography 372 A.H. 982
A.D. English tr. and commentary V. Minorsky. London.
BANG, W.; GABAIN, A. von. 1929. Tiirkische Turfan-Texte. Vol. 2. SPAW, pp. 411-30.
BARTHOLD, W. 1963. Kirgizy. Istoricheskiyocherk. Sochineniya. Vol. 2,1, pp. 471-543. Moscow.
BAZIN, L. 1991. Les systemes chronologiques dans le monde turc ancien. Bibliotheca Orientalis
Hungarica XXXIV. Budapest/Paris.
BOMBACI, A. 1970. On the Ancient Turkic Title Eltabar. In: Proceedings of the IXth Meeting of
the Permanent International Altaistic Conference, Naples, pp. 1-66.
BRETSCHNEIDER, E. 1910. Mediaeval Researches in Eastern Asiatic Sources. London. 2 vols.
CHAVANNES, E. 1903. Documents sur les Tou-kiue (Tures) occidentaux. St Petersburg.
CHAVANNES, E.; PELLIOT, P. 1911-12. Un traite manicheen retrouve en Chine. JA, 1911,Vol. 2,
pp. 499-617; 1912, Vol. 1, pp. 99-392. Also published separately, Paris, 1912.
CZEGLEDY, K. 1983. From East to West. The Age of Nomadic Migrations in Eurasia. AEMAe,
Vol. 3, pp. 25-125.
DROMPP, M. R. 1986. The Writings of Li Te-yii as Sources for the History of the T'ang-Inner
Asian Relations. Ph.D. diss. Indiana University, Bloomington.
ECSEDY, H. 1982. Uigurs and Tibetans in P'ei-ting. AOHung, Vol. 17, pp. 83-104.
GABAIN, A. VON. 1973. Das Leben im Uigurischen Konigreich von Qoco
(850-I 250).Wiesbaden.

449
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

GENG SHIMIN. 1984. A Study of Two Uighur Contracts of the Yuan Dynasty (1271-1368). Zen-
tralasiatische Studien, Vol.17,pp. 7-18.
GENG SHIMIN; HAMILTON, J. 1981. L'inscription ou"igoure de la stele commemorative des Iduq
de Qoco. Turcica, Vol. 13, pp. 10-54.
GENG SHIMIN; KLIMKEIT, H.-J. 1988. Das Zusammentreffen mit Maitreya. Teil I: Text, Uber-
setzung und Kommentar. Teil II: Faksimiles und Indices. Wiesbaden.
HAMILTON, J. R. 1955. Les Ouigours a I'epoque des Cinq Dynasties, d'apres les documents
chinois. Bibliotheque de l'Institut des Hautes Etudes Chinoise, Vol. 10. Paris.
- - . 1962. Toquz-Oyuz et On-Uyyur. JA, pp. 23-63.
- - . 1966. Manuscrits ouigours de Touen-houang. Vols. 1-2. Paris.
HANSEN, 0. 1930. Zur sogdischen Inschrift auf dem dreisprachigen Denkmal von Karabalgasun.
JSFOu, Vol. 44, 3.
JAGCHID, S. 1989. The 'Uighur Horses' of the T'ang Dynasty. In: W. Reissig and K. Sagaster
(eds.), Gedanke und Wirkung. Festschrift zum 90. Geburtstag von Nikolaus Poppe, pp. 175-88.
Wiesbaden.
KHUDJAKOV, J. S. 1980. Vooruzhenie eniseeyskikh kirgizov. Novosibirsk.
- - . 1982. Kirgizy na Tabate. Novosibirsk.
- - . 1990. Pamyatniki uygurskoy kultury v Mongolii. In: V. E. Larichev (ed.), Tsentral'nayaAziya
I sosednie territorii v srednie veka. Novosibirsk.
KLSELEV, S. V. 1951. Drevnyaya istoriya Yuzhnoy Sibiri. 2nd ed. Moscow.
KLYASHTORNY, S. G. 1982. The Terkhin Inscription. AOHung, Vol. 37, pp. 355-66.
- - . 1985. The Tes Inscription of the Uighur Bogti Qaghan. AOHung, Vol. 39, pp. 137-56.
KLYASHTORNY, S. G.; Livshits, V. A. 1971. Une inscription inectite turque et sogdienne. La stele
de Sevrey (Gobi meridional). JA, pp. 11-20.
KYZLASOV, L. R. 1960. Tashtykskaya epokha v istorii khakassko-minusinskoy kotloviny. Moscow.
- - . 1964. Tuva v sostave gosudarstva drevnikh kirgyzov. Istoriya Tuvy. Vol. 1, pp. 138-64,
390-2. Moscow.
- - . 1969. Istoriya Tuvy v srednie veka. Moscow.
LE COQ, A. VON. 1922. Ttirkische Manichaica aus Chotscho III. APAW, Vol. 3, pp. 1-49.
LIGETI, L. 1971. Apropos du 'Rapport sur les rois demeurant dans le nord'. In: Etudes tibetaines
dediees ala memoire de Marcelle Lalou, pp. 166-89. Paris.
LIU, YL-TANG. 1971. A Study on the Uighur Horses [in Chinese]. Bulletin of the Institute of
China Border Area Studies, Vol. 2, pp. 155-208.
MACKERRAS, C. 1969. Sino-Uighur Diplomatic and Trade Contacts. CAJ, Vol. 13, pp. 215-40.
- - . 1972. The Uighur Empire According to the T'ang Dynastic Histories. A Study in Sino-Uighur
Relations 744-840. 2nd ed. Australian National University Asian Publications Series, 3.Can-
berra.
- - . 1990. The Uighurs. In: D. Sinor (ed.), The Cambridge History of Early Inner Asia, pp.
317-42, 484-5. Cambridge.
MAENCHEN-HELFEN, 0. 1939. The Ting-ling. HJAS, Vol. 4, pp. 77-86.
- - . 1951. Manichaeans in Siberia. In: Semitic and Oriental Studies ... Presented to William
Popper. University of California Publications in Semitic Philology, 9, pp. 311-26. Berkeley,
Calif.
MALOY, S. E. 1951. Pamyatniki drevnetyurkskoy pis'mennosti. Moscow/Leningrad.
- - . 1959. Pamyatniki drevnetyurkskoy pis'mennosti Mongolii i Kirgizii. Moscow/Leningrad.
MARQUART, J. 1912. uwainl's Bericht tiber die Bekehrung der Uiguren. SPAW, Vol. 25, pp.
486-502.

450
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

MARTINEZ, A. P. 1982. Gardizl's Two Chapters on the Turks. AEMAe, Vol. 2, pp. 109-217.
MINORSKY, V. 1942. Shara/ al-Zaman Tahir Marvazf on China, the Turks and India. London.
- - . 1948. Tamim ibn Bahr's Journey to the Uyghurs. BSOAS, Vol. 12, pp. 275-305.
MORIYASU, T. 1980. La nouvelle interpretation des mots Hor et Ho-yo-hor dans le ms.
TIBeTAITAINLLIOT 1283. AOHung, Vol. 34, pp. 171-84.
- - . 1981. Qui des Ouigours ou des Tibetains ont gagne en 789-792 aBesbaliq ? JA, pp. 193-205.
- - . 1991. A Study on the History ofUighur Manicheism. Research on Some Manichaean Mate-
rials and Their Historical Background. Memoirs of the Faculty of Letters, Osaka University,
Vol. 31-2.
MULLER, F. W. K., 1909. Ein iranisches Sprachdenkmal aus der nordlichen Mongolei. SPMW,
Vol. 1, pp. 726-30.
- - . 1910. Uigurica. II. APAW, Vol. 3, pp. 1-110.
- - . 1912. Der Hofstaat eines Uigurenkonigs. In: Festschrift Vilhelm Thomsen zur Vollendung des
siebzigsten Lebenjahres, pp. 207-13. Leipzig.
- - . 1915. Zwei Pfahlinschriften aus den Turfanfunden. APAW, Vol. 3.
- - . 1920. Uigurische Glossen. Ostasiatische Zeitschrift, Vol. 8, pp. 310-24.
PELLIOT, P. 1912. Les noms tibetains des T'ou-yu-houen et des Ouigours. JA, Vol. 2, pp. 520-3.
PULLEYBLANK, E. G. 1952. A Sogdian Colony in Inner Mongolia. TP, Vol. 41, pp. 317-56.
- - . 1956. Some Remarks on the Toquzoghuz Problem. UAJ, Vol. 28, pp. 35-42.
- - . 1960. The Name of the Kirghiz. CAJ, Vol. 34, pp. 98-119.
RAMSTEDT, G. J. 1913. Zwei uigurische Runeninschriften in der Nord Mongolei. JSFOu,Vol. 30,
3.
ROMODIN, V. A. (ed.). 1973. Materialy po istorii kirgizov i Kirgizii, Vol. 1.
SAVINOV, D. G. 1984. Narody Yuzhnoy Sibiri v drevne-tyurkskuyu epokhu. Leningrad.
SCHLEGEL, G. 1896. Die chinesische Inschrift auf dem uigurischen Denkmal in Kara Balgassun.
MSFOu, Vol. 9.
SENGA, T. 1990. The Toquz Oghuz Problem and the Origin of the Khazars. Journal of Asian
History, Vol. 24, pp. 57-69.
TEKIN, T. 1983. The Tariat (Terkhin) Inscription. AOHung., Vol. 37, pp. 43-68.
TIKHONOV, D. I. 1966. Khozyaystvo i obshchestvennyi stray Uygurskogo gosudarstva X-XN vv.
Moscow/Leningrad.
YAMADA, N. 1993. Sammlung uigurischer Kontrakte. Ed. Juten Oda, P. Zieme, Hiroshi Umemura
and Takao Moriyasu. Osaka. 3 vols
ZIEME, P. 1973. Die Uiguren und ihre Beziehungen zu China. CAJ, Vol. 17, pp. 282-93.
- - . 1992. Religion und Gesellschaft im Uigurischen Konigreich von Qoco. Kolophone und Stifter
des alttiirkischen buddhistischen Schrifttums aus Zentralasien. Wiesbaden.
Literature in Japanese
ABE, T. 1955. Nishi Uiguru kokushi no kenkyu [Historical Study of the Western Uighurs]. Kyoto.
OKAZAKI, S. 1972. Studies on the Ancient History of the Tungus. Oriental History Series, Vol.
27. Kyoto.
TORU, H. 1931. Seiiki bunmeishi gairon [An Historical Outline of the Western Region]. Tokyo.
- - . 1946. Seiiki bunkashi [A Cultural History of the Western Region]. Tokyo.
UMEMURA, H. 1977a. Uighur Documents with a Forfeiture Clause: Where and When They Were
Written. The Toyo Gakuho [Journal of the Research Dept. of the Toyo Bunko], Vol. 58, 3-4,pp.
1-40.
- - . l977b. Official Powers in Uighuristan of the 13th Century. The Toyo Gakuho [Journal of the
Research Dept. of the Toyo Bunko], Vol. 59, 1-2, pp. 1-31.

451
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

Literature in Chinese
BAI BIN. 1988. Biography of Yuan Hao. Changchun.
- - . 1989. Studies on the History of Dang Xiang. Changchun.
CHEN BINGYING. 1985. Studies on the Cultural Remains of the Western Hsia Dynasty. Yinchuan.

FENG JIASHENG et al. (eds.). 1958. Weiwuerzu shiliao Jianbian [A Concise Compilation of
Historical Materials on the Uighurs]. Beijing.
GENG SHIMIN (tr.). 1981. Yutian heng, xi-yu wenhua shi [A Cultural History of the Western
Region]. Urumchi.
- - . 1983. Weiwuerzu gudai wenhua he wenxian gailun [An Introduction to the Ancient Culture
and Literature of the Uighurs]. Urumchi.
LI FANWEN. 1983. Symposium on Studies in the Western Hsia Dynasty. Yinchuan.
LI LUZHI. 1975. History of the Western Hsia Dynasty. Datong.
LI TAO. 1881. Sequel to the Annals of Chinese History, Published by the Zhejang Publishing House
in the Seventh Year of the reign of the Emperor Guang Xu.
LI WEI. 1989. Historical Studies on the Western Hsia Dynasty. Yinchuan.
SHI JINBO. 1986. The Culture of the Western Hsia Dynasty. Changchun.
- - . 1988. A Concise History of the Buddhism of the Western Hsia Dynasty. Yinchuan.
TAI XIZHANG. 1988. Chronicle of the Western Hsia Dynasty. Checked and revised by Luo
Maokun. Yinchuan.
WANG CHENG. 1787. A Concise History of the Eastern Capital, Published in the Year of Yimou
of the Reign of the Emperor Qian Long.
Wu TIANXI. 1980. Manuscripts on the History on the Western Hsia Dynasty. Chengdu.
ZENG GONG. 1784. Long Pi!]g's Collection, Published in the Forty-Seventh Year of the Reign of
the Emperor Kang Xi.
ZHONG KAN; WU FENGYUN; LI FANWEN. 1979. A Concise History of the Western Hsia
Dynasty. Yinchuan.

CHAPTER10
ABU 'L-FADL cALLAMI. 1897-1921. Akbar-nama. English tr. H. Beveridge. Calcutta. 3 vols.
AZIZUDDIN, M. M. 1987. Tarrkh-i Chitral. Rawalpindi.
BEG, Q. 1962. Tarrkh-i cahd-i catiq-i riyasat-i Hunza. Rawalpindi.
BIDDULPH, J. 1977. Tribes of the Hindoo Koosh. Repr. Karachi.
DANI, A. H. 1987. Shah Ra'is Khan's History ofGilgit. Islamabad/Gilgit.
- - . 1989a. History of the Northern Areas of Pakistan. Islamabad.
- - . 1989b. Islamic Architecture. The Wooden Style of Northern Pakistan. Islamabad.
DREW, F. 1877. The Northern Barrier of India. London.
FRANCKE, A. H. 1907. A History of Western Tibet. London.
GHAFOOR, M.A. 1965. Two Lost Inscriptions Relating to the Arab Conquest of Kabul and the
North-West Region of West Pakistan. Ancient Pakistan, Vol. 2. Peshawar.
HASAN MOHIBBUL. 1959. Kashmir under the Sultans. Calcutta.
HASHMATULLAH KHAN. 1942. Tarrkh-i Jamma. 2nd ed. Lahore.
JETTMAR, K. 1979. Bolor. A Contribution to the Political and Ethnic Geography of Northern
Pakistan. Journal of Central Asia, Vol. 2. Karachi.
LEITNER, G. W. 1868-73. Results of a Tour in Dardistan, Kashmir, Little Tibet, Ladak,
etc.London.

452
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

LORIMER, D. L. R., 1980. Materialen zur Ethnografie von Dardistan (Pakistan) aus den nachge-
lassenen Aufzeichungen von D. R. Lorimer. Teil V. Hunza. Graz.
MAQBUL AHMAD, S.; RAJA BANO. 1984. Historical Geography of Kashmir. New Delhi.
MIRZA GHULAM MURTAZA. 1962. Tar!kh-i Chitral. Peshawar.
MIRZA MUHAMMAD HAYDAR DUGHLAT. 1895. Tarr:kh-i Rashr:dr:. Ed. and English tr. N.
Elias and E. Denison Ross, A History of the Moghuls of Central Asia. London.
PARMU, R. K. 1967. A History of Muslim Rule in Kashmir. New Delhi.
SACHAU, E. 1910. Alberuni's India. London. 2 vols.
STEIN, Sir Aurel. 1899. The Ancient Geography of Kashmir. JASB.
--1907. Ancient Khotan. Oxford.
--1922. A Chinese Expedition Across the Pamirs and Hindu Kush A.D. 747. GJ, Vol. 59.
SUFI, G. M. D., 1949. Kashir. A History of Kashmir. Lahore.
THOMAS, F. W. 1935. Tibetan Literary Texts and Documents concerning Chinese
Turkestan.London.
VIGNE, G. T. 1842. Travels in Kashmir, Ladakh and Ishkardo. London.

CHAPTER 11
BOSWORTH, C. E. 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual.
Edinburgh/New York.
BRETSCHNEIDER, E. 1911. Mediaeval Researches from Eastern Asiatic Sources. London. 2 vols.
BUELL, P. D. 1979. Sino-Khitan Administration in Mongol Bukhara. Journal of Asian History,
Vol. 13, pp. 121-51.
- - . 1992. Early Mongol Expansion in Western Siberia and Turkestan (1207-1219): A Recon-
struction. CAJ, Vol. 36, pp. 1-32.
- - . 1993. Yeh-Hi A-hai, Yeh-Hi T'u-hua. In: In the Service of the Khan: Eminent Personalities of
the Early Mongol-Yuan Period (I 300-I 300), pp. 112-21. Ed. I. de Rachewiltz et al. Wiesbaden.
EI2 , arts. Ilek-Khans or Karakhanids; Kara Khitay.
FRANKE, H. 1990. The Forest Peoples of Manchuria; Kitans and Jurchens. In: The Cambridge
History of Early Inner Asia, pp. 400-23, 492-4. Ed. D. Sinor. Cambridge.
HOLMGREN, J. 1986. Yeh-Hi, Yao-lien and Ta-ho: Views of the Hereditary Prerogative in Early
Khitan Leadership. Papers on Far Eastern History, Vol. 34, pp. 37-81. Canberra.
JUWAYNI, cATA-MALIK. 1958. Tar!kh-i Jahan-gushay. English tr. J. A. Boyle, The History of
the World-Conqueror. Manchester. 2 vols.
KARA, G. 1986-87. On the Khitan Writing Systems. Mongolian Studies, Vol. 10, pp. 19-24.
Bloomington, Ind.
- - . 1996. The Kitan Scripts. In: P. T. Daniels and W. Bright (eds.). The World's Writing Systems,
pp. 230-5. Oxford.
LIGETI, L. 1970. Le tabgatch, un dialecte de la langue sien-pi. In: Mongolian Studies, pp. 265-308.
Ed. L. Ligeti. Budapest.
MINORSKY, V. 1942. Shara/ al-Zaman Tahir Marvazr: on China, the Turks and India. London.
RACHEWILTZ, I. DE. 1962. The Hsi-yu lu by Yeh-Hi Ch'u-ts'ai. Monumenta Serica, Vol. 21, pp.
1-128. Sankt Augustin.
- - . 1993. Yeh-Hi Ch'u-ts'ai. In: In the Service of the Khan Eminent Personalities of the Early
M ongol-YUan Period (I 300-I 300), pp. 136-72. Ed. I. de Rachewiltz et al. Wiesbaden.
SINOR, D. 1995. Western Information on the Kitans and Some Related Questions. JAOS, Vol. 115,
pp. 262-9.

453
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

S1EIN, R. 1940. Leao-tche. TP, Vol. 35, pp. 1-154.

WALEY, A. 1931. The Travels of an Alchemist. London.
WITIFOGEL, K. A.; FENG, CHIA-SHENG. 1949. History of Chinese Society. Liao (907-1125).
Transactions of the American Philosophical Society, N. S. Vol. 36. Philadelphia.

CHAPTER12
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
CLEAVES, F. W. 1952. The Sino-Mongolian Inscription of 1346. HJAS, Vol. 15.
- - . 1982. The Secret History of the Mongols. Vol. 1: Translation. Cambridge, Mass.
EI2 arts. Cingiz-Khan; Cingizids; Mongols.
ENDICOTT-WEST, E. 1989. Mongolien Rule in China. Local Administration in the Yuan Dynasty.
Harvard-Yenching Institute Monograph Series, Vol. 29. Cambridge, Mass.
GROUSSET, R. 1952. L'empire des steppes. Attila, Gengis-Khan, Tamerlan. 4th ed. Paris. English
tr. Naomi Walford, The Empire of the Steppes. A History of Central Asia. New Brunswick, N.
J., 1970.
ISHJAMTS, N. 1974. Mongold negdsen tar baiguulagdaj feodalism bUrelden togtson an [The
Foundation of the United State and the Formation of Feudalism in Mongolia]. Ulan Bator.
JAGCHID, S.; HYER, P. 1979. Mongolia's Culture and Society. Boulder, Colo./Folkstone, Kent.
JUWAYNI, cATA-MALIK. 1958. Tarrkh-i Jahan-gushay. English tr. J. A. Boyle, The History of
the World-Conqueror. Manchester. 2 vols.
LATTIMORE, 0. 1963a. Chingis Khan and the Mongol Conquests. Scientific American, Vol. 209,
No. 2.
- - . 1963b. The Geography ofChingis Khan. GJ, Vol. 129, No. 1, pp. 1-7.
LIGETI, L. (ed.). 1972. Monuments en ecriture Phagspa. Pieces de chancellerie en transcription.
Budapest.
LUBSAN-DANZAN. 1937. Altan tobci. Ulan Bator.
MORGAN, D. 0. 1986. The Mongols. Oxford.
PETRUSHEVSKY, I. P. 1968. The Socio-economic Condition of Iran under the II-Khans. In: The
Cambridge History of Iran, Vol. 5, pp. 483-537.
RACHEWILTZ, I. DE. 1989. The Title Chinggis Qan/Qaghan Re-examined. In: Gedanke und
Wirkung. Festschrift zum 90. Geburtstag von Nikolaus Poppe. Wiesbaden.
RASHID AL-DIN. 1952. Jamie al-tawarrkh. Russian tr. A. Khetagurov, Sbornik letopisei, Vol. 1,
pt. 1. Moscow/Leningrad.
RICCI, A. 1931. The Travels of Marco Polo. English tr. London.
SHIRENDEV, B.; NATSAGDORJ, S.; BIRA, S.; et al. 1966. BUg Nairamdah Mongol Ard Ulsiyin
TUUh [The History of the Mongolian People's Republic], Vol. 1. Ulan Bator.
VLADIMIRTSOV, B. 1938. Le regime social des Mongols. French tr. M. Carsow. Paris.

CHAPTER13
BARTHOLD, W. 1935. ZwolfVorlesungen Uber die Geschichte Mittelasiens. German tr. T. Menzel.
Berlin. French tr. M. Donskis, Histoire des Tures d'Asie Centrale. Paris, 1945.
- - . 1958-62. Four Studies on the History of Central Asia. English tr. V. and T. Minorsky. Leiden.
3 vols.
- - . 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr. T. Minorsky,
ed. with further addenda and corrigenda by C. E. Bosworth. London.

454
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

BOSWORIB, C. E. 1996. The New IslamicDynasties. A Chronological and Genealogical Manual.

Edinburgh/New York.
BOYLE, J. A. 1971. The Successors of Genghis Khan. Tr. from the Persian of Rashid al-Din.New
York.
EI2 , arts. Caghatay
-
Khan: Caghatay
-
Khanate.
Elr, art. Central Asia. v. In the Mongol and Timurid Periods.
GROUSSET, R. 1952. L'Empire des steppes. Attila, Gengis-Khan, Tamerlan. 4th ed. Ch. 2. Paris.
English tr. Naomi Walford, The Empire of the Steppes. A History of Central Asia. Part II, Chs.
8-11. New Brunswick, N. J., 1970.
HAIDAR, M. 1984. The Mongol Traditions and their Survival in Central Asia, XIVth/XVth Cen-
turies. CAJ, Vol. 28, pp. 57-79.
HAMBLY, G. (ed.). 1966. Zentralasien. Frankfurt. English tr. Central Asia. London, 1969.
MANZ, B. F. 1983. The Ulus Chaghatay before and after Temtir's Rise to Power. The Transforma-
tion from Tribal Confederation to Army of Conquest. CAJ, Vol. 27, pp. 79-100.
SPULER, B. 1966. Geschichte Mittleasiens seit dem Auftreten der TUrken. In: K. Jettmar et al.
(eds.), Handbuch del Orientalistik. I. V. 5. pp. 123-310. Leiden. English tr. F. R. C. Bagley, The
Muslim World. A Historical Survey, Vol. 3, pp. 219-59. Leiden, 1969.

CHAPTER14
cABDU'L-HALIM. 1961. History of the Lodi Sultans of Delhi and Agra. Dacca.
AHMAD, A. 1964. Studies in Islamic Culture in the Indian Environment. Oxford.
AHMAD, M. A. 1987. Political History and Institutions of the Early Turkish Empire of Delhi
(/ 206-90). Revised ed. Lahore.
AMIR KHUSRAU. 1931. Khaza'in al-futuh. English tr. M. Habib. Madras.
BARAN!, DIYA' AL-DIN. 1862. Tar!kh-i Flruz-Shah!. Ed. Sayyid Ahmad Khan. Calcutta.
- - . 1972. Fatawa-yi Jahandarr:. Ed. Afsar Saleem Khan. Lahore.
BARTHOLD, W. 1968. Turkestan down to the Mongol Invasion, 3rd ed., with an additional eh. tr.
T. Minorsky, ed. with further addenda and corrigenda by C. E. Bosworth. London.
BASHAM, A. L. 1985. The Wonder that was India. London.
AL-BIRONI. 1910. India. English tr. E. Sachau. London. 2 vols.
BOSWORIB, C. E. 1973. The Ghaznavids. Their Empire in Afghanistan and Eastern Iran
998:I040. 2nd ed. Beirut.
- - . 1977. The Later Ghaznavids, Splendour and Decay. The Dynasty in Afghanistan and North-
ern India I 040-I I 86. Edinburgh.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual.
Edinburgh/New York.
BROWN, C. J. 1922. The Coins of India. Calcutta.
DIGBY, S. 1971. War Horse and Elephant in the Delhi Sultanate. A Study of Military Supplies.
Oxford.
2
EI , arts. Dihll Sultanate; Ghaznawids; Ghi1rids; Hind. IV. History; Khaldjis; LOdrs; Sayyids; Surrs;
Tughlukids;
-
Djawnpur; Gudjarat;
- -
Macbar; Malwa; Sharkls.
FIRISHTA. 1829. Tar!kh-i Firishta. Abridged English tr. John Briggs, History of the Rise of the
Mahomedan Power in India, till the Year A.D. I6/2. Vol. 1. London.
HABIB, M. 1951. Sultan Mahmud ofGhaznin. 2nd ed. Delhi.
- - . 1974. Politics and Society During the Early Medieval Period (Collected Works). Ed. K. A.
Nizami. Aligarh.

455
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

HABIB, M.; KHAN, A. S. n.d. The Political Theory of the Delhi Sultanate. Delhi.
HABIB, M.; NIZAMI, K. A. 1970. A Comprehensive History of India, Vol. 5: The Delhi Sultanate
( A.D. 1206-I 526), Chs. 2-4. Delhi.
HABIB ULLAH, A. B. M. 1945. The Foundation of Muslim Rule in India. Lahore.
HARDY, P. 1960. Historians of Medieval India. Studies of Inda-Muslim Historical Writing. Lon-
don.
HUSAIN, AGHA MAHDI. 1938. Rise and Fall of Muhammad bin Tughluq. London.
- - . 1963. Tughluq Dynasty. Calcutta.
IBN BATTUTA. 1971. Travels. English tr. H. A. R. Gibb. Vol. 3. Cambridge.
c1sAMI. 1938. Futuh al-salatrn. Ed. Agha Mahdi Husain. Agra.
JACKSON, P. 1975. The Mongols and the Delhi Sultanate in the Reign of Muhammad Tughluq
(1325-51). CAJ, Vol. 19, pp. 118-57.
- - . 1986. The Problem of a Vast Military Encampment. In: R. E. Frykenberg (ed.), Delhi through
the Ages. Delhi.
LAL, K. S. 1963. Twilight of the Sultanate. A Political, Social and Cultural History of the Sultanate
of Delhi from the Invasion of Timur to the Conquest of Babur I 398-I 526. London.
- - . 1980. History of the Khaljis. Revised ed. Delhi.
MAJUMDAR, R. C. (General Ed.). 1955-60. The History and Culture of the Indian People,Vol.
4: The Age of Imperial Kanauj; Vol. 5: The Struggle for Empire; Vol. 6: The Delhi Sultanate.
Bombay.
MORELAND, W. 1929. The Agrarian System of Mortem India. Cambridge.
MUJUMDAR, B. P. 1960. The Socio-Economie History of Northern India (Ilth and !2th Cen-
turies). Calcutta.
NAZIM, M. 1931. The Life and Times of Sulan Mahmud ofGhazna. Cambridge.
NIZAMI, K. A. 1974. Some Aspects of Religion and Politics in India during the Thirteenth Century.
2nd ed. Delhi.
- - . 1985. State and Culture in Medieval India. New Delhi.
QURESHI, I. H. 1958. Administration ofthe Sultanate of Delhi. 4th ed. Lahore.
RAICHAUDHURI, T.; HABIB, I. (eds.). 1982. The Cambridge Economic History of India. Vol. 1.
Cambridge.
RIAZUL ISLAM. 1989. Aspects ofcAla ud-Din Khalji's Economic Policy and its Impact. Journal
of Central Asia, Vol. 12. Islamabad.
ROEMER, H. H. 1986. Timilr in Iran. In: The Cambridge History of Iran, Vol.6, pp. 42-97.
SHARMA, R. S. 1965. Indian Feudalism, c. 300-I200. Calcutta.
- - . 1983. Material Culture and Social Formations in Ancient India. Delhi.
SIDDIQUI, I. H. 1992. Social Mobility in the Delhi Sultanate. In: Irfan Habib (ed.), Medieval
India. Vol. 1: Researches in the History of India I200-I7)0, pp. 22-48. Delhi.
SIRHINDI, Y. A. 1931. Tarr:kh-i Mubarak-Shahi. Ed. M. Hidayat Husain. Calcutta.
WRIGHT, H. N. 1936. The Coinage and Metrology of the Sultans of Dehl!. Delhi.

CHAPTER15
cALLAMI, abu 'L-FADL. 1939-49. A'r:n-iAkbari. English tr. H. Blochmann and H. S. Jarrett. 2nd
ed. Calcutta.
ANON. 1937. Hudad az-cazam. The 'Regions of the World', a Persian Geography 372 A.H. 982
A.D. English tr. and commentary V. Minorsky. London.

456
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

BALOCH, N. A. 1954. Chronology of the Samma Rulers of Sind. In: Proceedings of the Meeting
of the Second Session of the Pakistan Historical and Archives Commission. Peshawar.
- - . (ed.). 1983. Fathnamah-i Sind. Being the Original Record of the Arab Conquests of Sind.
Islamabad.
AL-BIRUNI. 1910. India. English tr. E. Sachau. London. 2 vols.
BOSWORIB, C. E. 1973. The Ghaznavids: Their Empire in Afghanistan and Eastern Iran
998-I040. 2nd ed. Beirut.
- - . 1977. The Later Ghaznavids, Splendour and Decay. The Dynasty in Afghanistan and North-
ern India I 040-I I 86. Edinburgh.
- - . 1994. Rulers of Makran and Qusdar in the Early Islamic Period. Studia Iranica Vol.23, pp.
199-209.
- - . 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual. Edin-
burgh/New York.
DUGHLAT, MIRZA MUHAMMAD HAIDAR. 1895. A History of the Moghuls of Central Asia.
The Tarikh-i-Rash!d!. English tr. N. Elias and E. Denison Ross.
DUTT, J.C. 1879-98. Kings of Kashmira. (English tr. of the chronicles of Kalhana, Jonaraja and
Srivara and Shuka.) Calcutta. 3 vols.
EI , arts. BalUcistan; Hind. IV. History; Kashmir; Kilat; Kusdar; Makran; Multan; Sind; EI2 Suppl.
2

arts. Bayhakl Sayyids; Caks.

HABIB, M.; NIZAMI, K. A. 1970. A Comprehensive History of India, Vol. 5: The Delhi Sultanate
( A.D. 1206-I 526). Delhi.
HASAN, MOHIBBUL. 1959. Kashmir under the Sultans. Calcutta.
KALHANA. 1900. Rajataranginf. English tr. M.A. Stein. London.
KHAN BALOCH, M. S. 1958. History of Baloch Race and Balochistan. Quetta.
KHUHRO, H. (ed.). 1981. Sind through the Centuries. Proceedings of an International Seminar
Held in Karachi in Spring I 975. Karachi.
AL-KOFI, cALI B. HAMID. 1960-2. Chach-nama. English tr. K. Mirza. Karachi.
LAMBRICK, H. T. 1964. Sind. A General Introduction. Hyderabad, Sind.
LONGWORIB, M. (tr.). 1954. The Baloch Race. London.
MACLEAN, D. N. 1989. Religion and Society in Arab Sind. Leiden.
NAZIM, M. 1931. The Life and Times of Sulan Mahdotmad ofGhazna. Cambridge.
PARMU, R. K. 1969. A History of Muslim Rule in Kashmir. New Delhi.
RAFIQ!, A. Q. 1972. Sufism in Kashmir from the Fourteenth to the Sixteenth Century. Delhi.
STERN, S. M. 1949. Ismacm Propaganda and Fatimid Rule in Sind. IC, Vol.23, pp. 298-307.
SUFI, G. M. D. 1948-9. Kashir. A History of Kashmir. Lahore. 2 vols.
WINK, A. 1990. Al-Hind. The Making of the Indo-Islamic World, Vol. 1: Early Medieval India and
the Expansion of Islam 7th-llth Centuries. Leiden.
YULE, H.; CORDIER, H. 1903. The Travels of Marco Polo. 3rd ed. London.

CHAPTER16
1. Sources
CLAVIJO, DON RUY GONZALEZ. 1881. Dnevnik puteshestviya ko dvoru Timura v Samarkand,
I 403-I 406. Spanish text and Russian tr. I. I. Sreznevskiy. St Peters burg.
- - . 1928. Embassy to Tamerlane /403-I406. English tr. G. Le Strange. London.
GHIYATH AL-DIN cALI. 1915. Kitab-i Raznama-yi ghazawat-i Hindustan. St Petersburg.

457
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

NIZAM AL-DIN SHAMI. 1937-56. Zafar-nama. Ed. F. Tauer as Histoire des conquetes de Tamer-
lane intitulee Zafamama par Nizamuddrn Sam!. Prague. 2 vols.
SHARAF AL-DIN YAZDI 1972. Zafar-nama. Facsimile ed. A. Urunbaev. Tashkent. Also ed.
Tehran, 1366/1987.
THOMAS OF METSOPc. 1957. The History of Timur Lang (in Armenian). Istoriya Timur-Lanka.
Russian tr. Baku.
2. Studies
BARTHOLD, W. 1935. ZwolfVorlesungen Uber die Geschichte Mittelasiens. German tr. T. Menzel.
Berlin. French tr. M. Donskis, Histoire des Tures d'Asie Centrale. Paris, 1945.
BARTOL'D, V. V. 1914. K istorii orosheniya Turkestana. St Petersburg.
- - . 1918. Ulugbek i ego vremya. Petrograd. English tr. V. and T. Minorsky, in: Four Studies on
the History of Central Asia, Vol. 2: Ulugh-Beg. Leiden, 1958.
BOSWORTH, C. E. 1996. The New Islamic Dynasties. A Chronological and Genealogical Manual.
Edinburgh/New York.
CHEKHOVICH, 0. D. 1951. Vakfnyi dokument vremen Timura iz Kollektsii Samarkandskogo
Muzeya. Epigrafika Vostoka, Vol. 4.
2
EI , arts. Karts; Sarbadars; Soyurghal; Timur.
Elr, arts. Central Asia, V, In the Mongol and Timurid Periods.
FRAGNER, B. 1986. Social and Internal Economic Affairs. In: The Cambridge History of Iran,
Vol. 6, pp. 491-567.
GAFUROV, B. G. 1972. Tadzhiki. Drevneyshaya, drevnyaya i srednevokovaya istoriya. Moscow.
GROUSSET, R. 1952. L'empire des steppes. Attila, Gengis-Khan, Tamerlan. 4th ed. Paris. English
tr. Naomi Walford, The Empire of the Steppes. A History of Central Asia. New Brunswick, N.
J., 1970.
HABIB, M.; KHAN, A. S. n.d. The Political Theory of the Delhi Sultanate. Delhi.
MAKHMUDOV, N. 1966. Zemledelie i agramye otnoshenniya v Sredney Azii v XIV-XV vv.
Dushanbe.
MANZ, B. FORBES. 1983. The Ulus Chaghatay before and after Temtir's Rise to Power. The
Transformation from Tribal Confederation to Army Conquest. CAJ, Vol. 27, pp. 79-100.
- - . 1989. The Rise and Rule ofTamerlane. Cambridge.
MASSON, V. M.; ROMODIN, V. A. 1964. Istoriya Afganistana, Vol. 1. Moscow.
MUKMINOVA, R. G. 1969. Remeslo i torgovlya (Samarkand kontsa XIV-XV vv.). In: Istoriya
Samarkanda, Vol. 1, pp. 195-212. Tashkent.
NASONOV, A. N. 1940. Mongoly i Rus'. Moscow/Leningrad.
PETROSYAN, Y. A. 1990. Osmanskaya imperiya: mogushchestvo i gibe!'. Moscow.
PETRUSHEVSKY, I. P. 1949. K istorii instituta soyurgala. Sovetskoye Vostokovedenie, Vol. 6.
--.1960. Zemledelie i agramye otnosheniya v Irane XIII-XNvv. Moscow/Leningrad.
ROEMER, H. R. 1986. The Jalayirids, Muzaffarid and Sarbadars, and Timur in Iran. In: The Cam-
bridge History of Iran, Vol. 6, pp. 1-97.
SMITH, J.M. Jr. 1970. The History of the Sarbadar Dynasty /336-/38I A.D. and Its Sources. The
Hague.
STROEVA, L. V. 1949. Serbedary Samarkanda. Uchenye Zapiski Leningradskogo Gosu-
darstvennogo Universiteta, no. 98. Leningrad.
UMNYAKOV, I. I. 1969. Vneshnepoliticheskie svyazi Samarkanda s gosudarstvami Zapadnoy
Evropy. Istoriya Samarkanda, Vol. 1. Tashkent.

458
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

CHAPTER17
1. Sources
cABD AL-RAZZAQ SAMARQANDI. 1941-9. Matlac al-sacdayn. Ed. Muhammad Shaff.
Lahore. Partial French tr. E. Quatremere, Notice de l'ouvrage persan qui a pour litre: Matla-
assadein, in Notices et Extraits de la Bibliotheque du Roi, XIV/1, Paris, 1843.
ABU 'L-GHAZI BAHADUR KHAN. 1871-4. Shajarat al-Atrak. Chaghatay Turkish text and
French tr. Baron J. J.P. Desmaisons, Histoire des Mogols et des Tatars. Paris. Repr. Amsterdam,
1970.
cALI SHIR NAWA'I. 1948. Mahbab al-qulab. Moscow/Leningrad.
BABUR. 1905. Babur-nama. Ed. A. S. Beveridge. Leiden/London. English tr. A. S. Beveridge, The
Babur-nama in English. London, 1922.
BINA'I HARAWI. Shaybanr:-nama. Ms. Tashkent, Uzbekistan State Library no. 1235.
FADL ALLAH B. ROZBIHAN KHUNJI. 1976. Mihman-nama-yi Bukhara. Ed. and Russian tr. R.
P. Jalilova. Moscow. Partial German tr. U. Ott, Transoxanie und Turkestan zu Beginn des 16.
Jahrhunderts. Freiburg, 1973.
JENKINSON, A. 1886. Early Voyages and Travels to Russia and Persia. Hakluyt Society, lst
Series, vols. LXXII-LXXIII. London.
KHWANDAMIR. 1857. Habr:b al-siyar. Bombay.
SALIH, M. 1908. Shaybanr:-nama. St Petersburg.
2. Studies
AKHMEDOV, B. A. 1965. Ulugbek i politicheskaya zhizn' Maverannakhra pervoy poloviny XV
v. In: Jz istorii epokhi Ulugbeka. Tashkent.
BARTHOLD, W. 1945. ZwolfVorlesungen Uber die Geschichte Mittelasiens. German tr. T. Menzel,
Berlin. French tr. M. Donskis, Histoire des Tures d'Asie Centrale. Paris, 1945.
BARTOL'D, V. V. 1918. Ulugbek i ego vremya. Petrograd. English tr. V. and T. Minorsky. In: Four
Studies on the History of Central Asia, Vol. 2: Ulugh-Beg, Leiden, 1958.
- - . 1928. Mir Ali-Shir i politicheskaya zhizn'. Moscow/Leningrad. German tr. W. Hinz, HeraJ
unter Husein Baiqara dem Timuriden. Leipzig, 1938.
BOSWORTH, C. E. 1996. The New Islamic Dynasties. A Chronological and Geneaological Man-
ual. Edinburgh/New York.
CHEKHOVICH, 0. D. (ed.). 1970. Samarkandskie dokumenty XV-XVI vv. o vladeniyakh Khodzhi
Akhrara v Sredney Azii i Afganistane. Moscow.
DAVIDOVICH, E. A. 1965. Materiali dlya kharakteristiki denozhnoy reformy Ulugbeka. In: Iz
istorii epokhi Ulugbeka. Tashkent.
EI2 , arts. ShibanI Khan; Shibanids; Timurids.
Elr, art. Central Asia. V. In the Mongol and Timurid Periods.
FRAGNER, B. 1986. Social and Internal Economic Affairs. In: The Cambridge History of Iran.
Vol. 6.
GROUSSET, R. 1952. L'Empire des steppes. Attila, Gengis-Khan, Tamerlan. 4th ed. Paris. English
tr. Naomi Walford, The Empire of the Steppes. A History of Central Asia. New Brunswick, N.
J. 1970.
IVANOV, P. P. 1945. Khozyaystvo dzhuybarskikh sheykhov. Moscow/Leningrad.
MANZ, B. FORBES. 1983. The Ulus Chaghatay before and after Temtir's Rise to Power. The
Transformation from Tribal Confederation to Army of Conquest. CAJ, Vol. 27, pp. 79-100.
MUKMINOVA, R. G. 1966. K istorii agramykh otnosheniy v Uzbekistane XVI v. Po materialam
'Vakf-name '. Tashkent.

459
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

- - . 1976. Ocherki po istorii remesla v Samarkande i Bukhare v XVI veke. Tashkent.

PE1RUSHEVSKIY, I. P. 1960. Zemledelie i agrarnye otnoshenya v Irane XIII-XN veka.
Moscow/Leningrad.
ROEMER, H. R. 1986. The Successors ofTimilr, and The Ttirkmen Dynasties. In: The Cambridge
History of Iran. Vol. 6, pp. 98-188.

CHAPTER18
1. Sources
cABD AL-HAQQ DIHLAWI. 1892. Akhbar al-akhyar. Delhi.
AMIR KHWURD. 1885. Siyar al-awliya'. Delhi.
cATTAR, FARID AL-DIN. 1297/1879. Mantiq al-tayr. Bombay. English tr. C. S. Nott, The Con-
ference of the Birds. London, 1954.
BARAN!, DIYA' AL-DIN. 1860-2. Tar!kh-i Flruz-Shahi. Calcutta.
BENJAMIN OF TUDELA. 1841. The Itinerary of Benjamin of Tudela. Text and English tr. A.
Asher. London.
FAKHR AL-DIN B. KASHIFI AL-WAciz. 1911. Rashahat cayn al-hayat. Lucknow.
HASAN MIYAN, SHAH. 1911. Tadhkirat Hadrat Abu 'l-Najlb cAbd al-Qahir al-Suhraward!.
Luc know.
JAMI. 1915. Nafahat al-uns. Lucknow.
KAY KAWDS B. ISKANDAR. 1951. Qabas-nama. English tr. R. Levy, A Mirror for Princes. The
Qabas Nama. London.
NAJM AL- DIN KUBRA. 1957. Fawa'ih al-Jamal wa-fawatih al-jalal. Ed. F. Meier. Wiesbaden.
NIZAM AL-DIN AWLIYA'. 1992. Fawa'ib al-fu'ad. English tr. B. Lawrence, Nizam al-Din
Awliya. Morals of the Heart. New York, 1992.
SHARAF AL-DIN YAZDI 1885-8. Zafar-nama. Calcutta. 2 vols.
2. Studies
AMORETTI, B. S. 1986. Religion in the Timurid and Safavid Periods. In: The Cambridge History
of Iran, Vol. 6, pp. 610-55.
BAUSANI, A. 1968. Religion under the Mongols. In: The Cambridge History of Iran, Vol. 5, pp.
538-49.
BOSWORTH, C. E. 1963. The Ghaznavids: Their Empire in Afghanistan and Eastern Iran
994-I040. Edinburgh; 2nd. ed., Beirut, 1973.
- - . 1984. An Early Persian Sufi. Shaykh Abu Sacid of Mayhanah. In: Logos Islamikos. Studia
Islamica in Honorem Georgii Michaelis Wickens. Ed. R. M. Savory and D. Agius. Toronto, pp.
79-96.
- - . 1996. The New Islamic Dynasties. A Chonological and Geneaological Manual. Edin-
burgh/New York.
EI2, art. Ahmad YasawI.
NASR, S. H. (ed.). 1991. Islamic Spirituality. Vol. 2, pp. 6-25. New York.
NICHOLSON, R. A. 1921. Studies in Islamic Mysticism. Cambridge.
NIZAMI, K. A. 1950. The Shattari Saints and their Attitude towards the State. Medieval India
Quarterly, Vol. 1, pp. 56-70.
- - . 1957. The Suhrawardi Silsila and its Influence on Medieval Indian Politics. Medieval India
Quarterly, Vol. 3, pp. 109-49.
- - . 1961. Religion and Politics in India during the Thirteenth Century. Bombay.
- - . 1985. State and Culture in Medieval India. Delhi.

460
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

- - . 1989. Akbar and Religion. Delhi.

- - . 1992. The Life and Times of Shaikh Nizam-u 'd-Din Auliya. Delhi.
QAMARUDDIN. 1985. The Mahdavi Movement in India. Delhi.
1RIMINGHAM, J. S. 1971. The Sufi Orders in Islam. Oxford.

CHAPTER19
Part One
As noted in the text, material on food and clothing in the Islamic East and Central Asia can only be
pieced together from mentions in the historical and geographic works of medieval Islamic authors
such as NarshakhI, Abu '1-Fadl Bayhaqr, GardizI, Juwaynr, Rashid al-Din, etc. There are no specific
studies. In general, however, reference may be made to the chapters 'The Standard of Living' and
'Land Products' in A. Mez, The Renaissance of Islam, tr. S. Khuda Bukhsh, Patna, 1937, and to
sections on food and clothing in B. Spuler, Die Mongolen in Iran, Leipzig, 1939, pp. 442-6, and
idem, Iran in friih-islamischer Zeit, Wiesbaden, 1952, pp. 508-18, whilst the EI2 articles Ghidha
(M. Rodinson) and Libas. iii. Iran (Y. K. and N. A. Stillman) and 'Libas. iv. Turkey' (T. Majda)
provide detailed information about types of food and cooking and about different garments and
headgear in medieval Islamic times.
Part Two
FLETCHER, G. F. 1968. China and Central Asia, 1368-1884. In: J. K. Fairbank (ed.). The Chinese
World Order. Traditional China's Foreign Relations, pp. 357-404. Cambridge, Mass.
JAGCHID, S. 1972. Peace, War and Trade between the North Asian Nomadic Peoples and the
Agricultural Chinese. Taipei.
JAGCHID, S.; HEYER, P. 1979. Mongolia's Culture and Society. Boulder, Colo./Folke-stone,
Kent.
JOHN OF PLANO CARPINI. 1955. Historia Mongalorum quos nos Tartaros appelamus. In: C.
Dawson(ed.), The Mongol Mission. Narratives and Letters of the Franciscan Missionaries in
Mongolia and China in the Thirteenth and Fourteenth Centuries. London/New York.
KISELEV, S. V. 1965. Drevnemongol'skie goroda. Moscow.
LATTIMORE, 0. 1940. Inner Asian Frontiers of China. New York.
NATSAGDORJ, S. 1978. Mongolynfeodalizmyn undsen zamnal. Ulan Bator.
WILLIAM OF RUBRUCK. 1990. The Mission, of Friar William of Rubruck. His Journey to the
Court of the Great Khan Mongke I253-I255. Tr. P. Jackson. Commentary and notes P. Jackson
and D. 0. Morgan. The Hakluyt Society. London.
WITFOGEL, K. A.; FENG CHIA-SHENG. 1949. History of Chinese Society: Liao (907-//25).
Philadelphia.
(VARIOUS CON1RIBUTORS). 1963. Istoriya Khalkhi [History of the Khalkha]. Ulan Bator.
(VARIOUS CON1RIBUTORS). 1966. Istoriya MNR [History of the Mongolian People's Repub-
lic], Vol. 1. Ulan Bator.

CHAPTER20
ABDUR RAHMAN, 1979. The Last Two Dynasties of the Sahrs (An Analysis of their History,
Archaeology and Palaeography). Islamabad.
ALTEKAR, A. S. 1948. A Bull and Horseman Type of Coin of the Abbasid Caliph al-Muqtadir
Billah Jafar. Journal of the Numismatic Society of India, Vol. 8, pp. 75-8.
BAYHAQI, ABU 'L-FADL. 1945. Tarrkh-i Mascadr. Ed. Q. Ghan! and A. Fayyad. Tehran. Russian
tr. A. K. Arends, Istoriya Ma~uda I030-I04I, 2nd ed. Moscow, 1969.

461
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

BIVAR, A. D. H. 1974. A Mongol Invasion Horde from Eastern Afghanistan. In: D. Kouymjian
(ed.), Near Eastern Numismatics, Iconography, Epigraphy and History. Studies in Honor of
George C. Miles, pp. 369-81. Beirut.
DANI, A. H. 1955. Lakor, Qa-Gaur or Nagor. Journal of the Numismatic Society of India, Vol. 16,
Part 2.
DARWISH TAJ AL-DIN HASAN WALi 1965. Isnad-i astana. Ed. M. T. Danish-Pazhiih. Tehran.
DAVIDOVICH, E. A. 1960. Iz oblasti denezhnovo obrashcheniya v Sredney Azii XI-XII vv.
Numivnatika i epigrajika, Vol. 2, pp. 92-117. Moscow.
- - . 1961. Kanibadamskiy klad karakhanidskikh monet (K kharakteristike obrashcheniya med-
nykh poserebrennykh dirkhemov v kontse XII-nachale XIII v.). Sovetskaya Arkheologiya, Vol.
1, pp. 186-200. Moscow.
- - . 1966. Denezhnoe obrashchenie v Maverannakhre pri Samanidakh. Numizmatika i Epigrajika,
Vol. 6, pp. 103-34. Moscow.
- - . 1972. Denezhnoe khozyastvo Sredney Azii posle mongol'skovo zavoevaniya i refonna
MaS:ud-beka (XIIIv.), p. 174. Moscow.
- - . 1979. Klady drevnikh i srednevekovykh monet Tajikistana, p. 462. Moscow.
- - . 1980. Zoloto v denezhnom khozyaystve Sredney Azii IX-XVIII vv. In: Blizhniy i Sredniy
Vostok. Tovarno-denezhnye otnosheniya pri feodalivne, pp. 55-69. Moscow.
- - . 1983. Iztoriya denezhnovo obrashcheniya srednevekovoy Sredney Azii (mednye monety XV
-pervoy chetverti XVI vv. v Maverannakhre), p. 360. Moscow.
EHRENKREUTZ, A. S. 1963. Studies in the Monetary History of the Near East in the Middle
Ages, II. The Standard of Fineness of Western and Eastern Dinars before the Crusades. Journal
of the Economic and Social History of the Orient, Vol. 6, Part 3, pp. 243-77.
IBN HAWQAL. 1938. Kitab Surat al-ard. Ed. J. H. Kramers. Leiden. French tr. J. H. Kramers and
G. Wiet, Configuration de la. terre, Paris, 1964. 2 vols.
IBN KHURRADADHBIH. 1889. Kitab al-Masalik wa 'l-mamalik. Ed. M. J. de Goeje. Leiden.
AL-ISTAKHRI. 1927. Kitab al-Masalik wa 'l-mamalik. Ed. M. J. de Goeje. 2nd ed. Leiden.
JUWAYNI, c ATA-MALIK. 1958. Tar!kh-i Jahan Gushay. English tr. J. A. Boyle, The History of
the World-Conqueror. Manchester. 2 vols.
KHADR, M. 1967. Deux actes de waqf d'un Qaraxxxanide d' Asie Centrale. Avec une introduction
par Cl. Cahen. JA, Vol. 255, pp. 305-34.
KHAN, A. N. 1990. Al-Mansurah. A Forgotten Arab Metropolis in Pakistan. Karachi.
KHAN, F. A. 1969. Banbhore. 3rd ed. Karachi.
LANE-POOLE, R. 1876. The Coins of the Mohammadan Dynasties in the British Museum. Vol. 2.
London.
LOWICK, N. M. 1975. An Early Tenth-Century Hoard from Isfahan. NC, 7th Series, Vol. 15, pp.
110-54.
- - . 1977. On the Dating of Samanid Outsize Dirhams. Spinks Numismatic Circular, pp. 204-7.
London.
MASSON, M. E. 1957. Istoricheskiy etyud po numizmatike Djagataidov (Po povodu Talasskovo
klada monet XIV v.). Trudy Sredneaziatskovo gosudarstvo universiteta. Novaya seriya. Vol. III,
pp. 41-108. Tashkent.
MITCHINER, M. 1973. The Multiple Dirhems of Medieval Afghanistan. Sanderstead, Surrey.
NARSHAKHI. 1973. Tar!kh-i Bukhara. Ed. Mudarris RadawI. Tehran 1351/1973. English tr. and
commentary, R. N. Frye, The History of Bukhara. Cambridge, Mass, 1954.
PERVI, T. N. 1969. Coins of the Early Muslim Period from Banbhore. Pakistan Archaeology, No.
6,pp.117-81.

462
ISBN 978-92-3-103467-1 Contents BIBLIOGRAPHY AND REFERENCES

RODGERS, C. J. 1894. Catalogue of the Coins Collected by Charles J. Rodgers, in the Lahore
Museum, Part 2. Calcutta.
SELLWOOD, D. 1980. The Striking of Samanid Double Dirhams. In: D. M. Metcalf and W. A.
Oddy, Metallurgy in Numismatics, Vol. 1, pp. 176-7. London.
SINGHAL, C. R. 1952. Bibliography of Indian Coins, Part 2. Muhammadan and Later Series.
Numismatic Society of India. Bombay.
THOMAS, E. 1848a. On the Coins of the Hindu Kings of Kabul. JRAS, Vol. 9, pp. 177-88.
- - . 1848b. On the Coins of the Kings of Ghaznf. JRAS, Vol. 9, pp. 267-386.
- - . 1860. Complementary Contributions to the Series of the Coins of the Kings ofGhaznf JRAS,
Vol. 17, pp. 138-208.
TYE, R.; TYE, M. 1995. Jitals. A Catalogue and Account of the Coin Denomination of Daily Use
in Medieval Afghanistan and North West India. Isle of South Uist, Scotland.
VASMER, R. 1930. Tanka, Tenka, Tenga. In: Worterbuch der Miinzkunde. Repr. Berlin, 1970, pp.
680-1.
WALKER, J. 1941. A Catalogue of the Arab-Sassanian Coins in the British Museum, Vol. 1: A
Catalogue of the Arab-Sassanian Coins. London.
WHITTAL, MAJOR J.M. 1921. Coins of Muhammad bin Tughluq. In: Numismatic Supplement,
25 (22a) to the JASB. Calcutta.
WRIGHT, H. N. 1907. Catalogue of the Coins in the Indian Museum, Vol. 2. Calcutta/Oxford.
- - . 1936. The Coinage and Metrology of the Sultans of Delhi. Delhi.
YAQUT. 1866-73. Mucjam al-buldan. Ed. F. Wtistenfeld as Yacut's geographisches Worterbuch.
Leipzig.

463
ISBN 978-92-3-103467-1 Contents GLOSSARY

GLOSSARY

cajam (non-Arabs) fatwa (legal opinion)

alba (in Mongolian, tribute) fiqh (Islamic law)
alchi (lower rank) futuwwa (chivalry)
c alim see c ulama' ghazal (lyric poetry)
amatya (minister) ghazl (fighter for the faith)
c amid (governor, secretary) ghulam, pl. ghilman (slave soldier)
c amid al-mulk (chief secretary) ghulaman-i saray (palace guards)
card (military review) ghulat (radical Shi cites)
carid-i lashkar (inspector-general ofthe army) hadrth (Muslim tradition)
ata (father) hajib (chamberlain)
atabeg (tutor) hakim (governor)
aymak (district in Mongolia) hisba (market control)
cayyar (band of urban rowdies andvigilantes) il (large Turco-Mongol tribal grouping)
barat (assignment of revenue) ins ha' (correspondence, secretaryship)
barda (slave captive, prisoner of war) iqtac (land grant, revenue assignment)
barld (courier and postal service) jaglr (assignment of land in Mughal India)
bata 'I (taxation taken as a 3e of theproduce of jihad (holy war)
the land in India) jizya (poll tax on non-Muslims)
b1gar (corvee labour) kachar (porter)
borgah (headquarters of the Mongol army) kafir (infidel, unbeliever)
balak (Turkish or Turco-Persian clan) kalantar (in Persia, head of the
dac I (propagandist) localcommunity)
dargah (royal court) kalima (the word of god)
daraghan (Mongol tax-collector, governor) karlz see qanat
dibqan (landowner) karkhanah (workshop)
d1wan (council, government office;collection katib al-insha' (chief secretary)
of poems) kesik, pl. kesigten (protective guard
d1wan-i acza (Supreme Ministry) inMongolia)
d1wan-i card (department of the army) khalrja (deputy)
d1wan-i barld (postal service andintelligence khalisa (state-administered lands)
department) khanaqah (convent or retreat house for
d1wan-i wizarat (revenue department) dervishes)
dadl(in Turco-Mongol usage, capitation tax) kharaj (land tax)
farsakh (measure of length, a league = approx. khass (crown domains)
6km) khatrb (preacher at the Muslim worship)
fat'h-nama (announcement of victory) khutba (Friday worship oration)

464
ISBN 978-92-3-103467-1 Contents GLOSSARY

kiiriyen (or 'circle', community of clansin ra'rs al-ru'asa' (head of of the townnotables)
Mongolia) rasa'il see risala
kurultay (meeting of the military chiefs and ribat (fortified outpost, caravanserai;hospice)
nobles of the tribes among the Mongols and risala, pl. rasa 'il (epistle)
Turco-Mongols) rubac f (quatrain)
kilt (superior rural class in India) rukn al-dawla (Pillar of the State)
laqab, pl. alqab (honorific title or nickname) sadr, pl. sudar (lit. 'eminence' leader of the
madhhab (law school) Muslim religious classes)
madrasa (Islamic college) saha (banker)
malik al-mashriq wa 'l-magbrib (King of the samac (mystical songs)
East and West) sarraf (money-changer)
manshur (patent of authority) sayyid (descendant of the Prophet)
mard-b!garf (corvee) sharf c a (divinely ordained Islamic legalcultic
mashayikh (Sufi spiritual leaders) system)
mawal! amfr al-mu 'minfn (Clients of the shihna (military governor)
Commander of the Faithful) shOlen (in Turkish, feast)
mazar (burial shrine) silsila (chain of spiritual authority)
milk (private property) sipahsalar (commander-in-chief)
mrrza (prince, in Persian and Perso-Turkish siyasa (secular law)
usage) soyurghal (land grant, lit. 'gift')
mukas (pl. of maks; non-Qur'anic taxes) al-sultan al-ac zam (Supreme Sultan)
mulk (power, authority, dominion al-sultan al-muc azzam (Exalted Sultan)
andpossession) sultan al-salatfn (sultan of sultans)
muqtac (holder of an iqtac, hence ta-chiang-chiin (in Chinese, Grand General)
localgovernor) ta-hsiang (in Chinese, Grand Minister)
murabitan (dwellers in ribats and frontier tamgha (tax levied on craft and trade)
fighters) tarrqa (Sufi order)
mushrif (overseer, inspector) tughra. (official insignum on documents)
na'ib-i mamlakat (viceroy) tughra'r (chief secretary, literally, 'affixer of
namazgah (area for ceremonial acts ofworship) the tughra')
naqqash (artist, painter) culama', sing. cazim (scholars learned in the
nasab (good birth) Islamic legal and theological sciences)
nawbat (military band salute) ulus (domains in Mongolia and the Mongol
noyan (commander in the Mongol army) conquered lands)
ordo, urdu (military camp) wakrl (intendant)
pargana (group of villages in India) walayat (spiritual jurisdiction of a Sufi leader
p!r (Sufi mystical guide) or saint)
p!shtaq (fore-porch) wall al-cahd (heir apparent)
qadr (judge) waqf (endowment for pious purposes)
qanat (= karfz, subterranean irrigationchannel) wathfqa (legal deed)
qarshi (palace, in Mongolian) wazfr (vizier)
qas!da (ode or eulogistic poem) yarghuchi (in Mongolian, Supreme Justice)
qasrm amrr al-mu'minrn (Partner of the Yeke yasa (among the Mongols, The Great
Commander of the Faithful) Law)
rabad (suburb) Yiian (in Chinese, The Origin)
ra'is, pl. ru'asa' (headman, town chief, mayor) Zakat (alms-tax)

465
ISBN 978-92-3-103467-1 Contents INDEX

INDEX

Abfi Hashim cAbd Allah, 51 Ambaqay, 251

Abu Hafs, 100, 270 Amir Husayn, 324, 325 , 327
Abfi Muslim, 30, 32- 34, 36- 39, 41 , 43 , 50, cAmr b. Layth, 387
52- 56 Amul, 29, 34, 91 , 378
Abaqa, 268, 386 Anandpal, 303
cABBASIDS, 27 Andarkot, 314
cABBASIDS, 30 Andijan, 268, 270
cAbbasids, 32, 33, 35 , 38-40, 43 , 45 , 52, 53 , Andkhud, 191 , 382
55 , 57, 59, 60, 62, 63 , 65 , 298, 299, cArabshah, 325
302, 368, 370, 390 Arslan Hajib, 185
cAbbasiyya, 52, 54, 63 Arslan Khan, 209
cAbd Allah b. Mucawiya, 31 , 34 Ashacira, 58
cAbd Allah b. Mucawiya, 52 cAwfi, 192, 305
Abiward, 91 , 372, 377, 384 Azariqa, 50
CAfifi, 289 Azerbaijan, 23 , 48, 54, 56, 57, 62, 336, 338,
Afrasiyab, 93 341 , 347, 349
Afrighid, 28
Ahmad b. cAbd Allah b. Salam, 369
Ahmad b. Khidr, 138 Babur, 190, 296, 308, 320, 374, 381
Ahmad BiharI, 378 Barbak Shah, 305
Akbar the Great, 291 Bayazid I, 338, 345 , 346
Akhsikath, 132, 134, 135 , 144, 147 Bayazid Jalayir, 271
Ala Shan, 213 Bada'un, 292- 295
Alamut, 65 Badhghis, 53 , 56
cAlI b. AbI Talib, 49, 51 Baghatur, 251 , 253
cAlib. Musa, 127, 129, 130 Baghdad, 40, 41 , 43 , 56, 63 , 64, 91 , 186, 275 ,
cAlI b. cfsa b. Mahan, 44 298, 338, 368- 370, 373, 376, 384,
cAlI Kufi, 305 386, 389
cAlI Tegin, 132, 134, 135 , 155 Bahadur Khan NuhanI, 296
cAlids, 35 , 58- 60, 128, 132, 134, 135 , 390 Bahlill LOdI, 290, 293
Almalik, 268, 270, 271 Bahram Shah of Ghazna, 186
Alp Bilge Kaghan, 197 Bahrabad, 375
Alptegin, 103 Bahrain, 62
Alughu, 268 Bakhar, 306, 308

466
ISBN 978-92-3-103467-1 Contents INDEX

Balasaghun, 91 , 130, 131 , 133, 139, 242, 244, Chin state, 217, 239
245 , 247 Chinggis Khan, 141 , 208, 220, 240, 245- 247,
Balcamr, 103 250- 252, 254--261 , 263 , 269, 331 ,
BALUCHISTAN, 297 332, 334, 341 , 368, 387, 389
Baluchistan, 298, 300, 308, 309 Chishtiyya, 24, 371
Bal'ik:l'ig river, 198 Chitor, 278
Baraq Khan, 265 Chunyu, 219
Barchkent, 265 Cutch, 306
Barchuk Art Tegin, 208
Barlas, 271 , 272, 324, 325, 328 Dadil, 381
Barmakids, 386 Damaras, 312
Bashkirs, 247 Denkard, 370
Bawandids, 59 Dabusiyya, 30, 90, 91
Bay Bal'iq, 198 Dandanqan, 134
Bayan Chur, 197 Dangxiang, 213
Baydar, 268 Darband, 266
BayhaqI Sayyids, 319, 322 Darya Khan, 31 O
Baysunqur, 348 Dawlat Chak, 321
Begtuzun, 129 Dawlat Khan, 291
Benakat, 90 Daylaman, 63
Berke, 266, 268, 375 Daylamites, 59
Beshbalik, 43 , 208, 209, 211 Deccan, 23 , 274, 278, 286- 288, 290
Bhattavatara, 322 DELHI SULTANATE, 273 , 284
Bhikshana, 314 Delhi Sultanate, 275 , 277, 283 , 285 , 289- 291 ,
Bihafarldiyya, 53 293 , 294, 296, 305 , 306
Bilge Kaghan, 198, 201 , 236 Deming, 214, 215
Binakat, 146, 147, 265 Deogir, 278, 286
Binkath, 147 Dera Ghazi Khan, 309
Bodonchar, 260 Devasvamr, 313
Borjigids, 249, 252, 253, 256 Do'ab region, 287
Brahmans, 289, 312, 317, 321 Doda'rs, 309
Brahuis, 308 Dughlat clan, 324
Bumin Kaghan, 197 Duwa Khan, 268- 271 , 278
Buqu Khan, 233 Dwarsamudra, 278
Burqan Qaldun, 251
Buyids, 59, 60, 65 Farldun, 184
Farah Rud, 182
Carmathians, 60, 62, 188, 303 Faraj, 338
Chahamanas, 188, 190 Fat'h Khan, 319
Chach, 36, 45 , 88, 90, 91 , 129, 130, 147, 374 Fatimids, 301- 303
Chaghaniyan, 88, 98, 135 , 140, 189, 271 Ferghana, 28, 30, 36, 39, 43 , 45 , 88- 90, 101 ,
Chaghatay, 262, 265 , 266, 323, 324 128, 130, 133- 137, 140, 142, 146,
Chaghr'i Beg, 372 147, 149, 171 , 268- 271 , 347
Chak dynasty, 321 Firdawsr, 94, 97, 98, 100, 308, 322
Changshi, 271 Firdawsiyya, 375
Chapar, 268 Firishta, 311 , 312

467
ISBN 978-92-3-103467-1 Contents INDEX

Ganapatis, 278 Hurfifiyya, 349, 378

GardizI, 53 Husayn Bayqara, 310, 347
GardizI, 203 , 238 Husayn Shah, 293- 295
George IV Lashen, 258 Htilegti, 262, 267, 324
Georgia, 336, 338
Gharchistan, 61 , 87- 89, 182, 186, 187, 194, Ibadi yya, 51
373 Ibn Battuta, 283 , 287, 306, 375 , 387
Ghazan Khan, 313, 324, 368, 377, 386, 389 Ibn cArabi, 375
Ghazna, 85 , 183- 189, 191- 195 , 238 Ibn Fadl Allah, 284
GHAZNAVIDS, 102 Ibn Hawqal, 102, 299, 300, 302
Ghaznavids, 65 , 85 , 125, 129, 131 , 133, 134, Ibn Sina, 94, 96, 100, 372
144, 184, 186, 190, 194, 236, 301 Ibn Sumar, 303
Ghur, 87, 183- 187, 189- 195 , 293, 298, 334, 11 Khanids, 386
373 Ilaq, 87, 89, 129, 134, 147
GHURIDS, 182 Iltutmish, 193, 275 , 277, 279, 281 , 305 , 380
Ghurids, 184- 188, 190, 192- 195 , 301 , 379 Ilyas Khoja, 325
Golden Horde, 24, 266, 267, 269, 332, 333, Imamiyya, 48, 58, 60
336, 337, 340, 346, 375, 387 Indra, 211
Govinda, 188 Irtysh river, 265
Gurjaras, 274 Is'haq the Turk, 38, 55
Gwalior, 189, 292, 295 , 381 clsa b. Macdan, 300
clsamI, 285
HajjIKhan, 318 Isfijab, 127, 133, 148
Hashimiyya, 51 , 52, 54, 63 Ismacn b. Ahmad, 127
HamadanI Sayyids, 319 Istiya, 186
Hamza b. Adharak, 44, 50 Itawa, 292
Hamza b. cAlI, 303
Hamziyya, 50 Jam Firfiz, 308
Han shu, 204 Jam Tughluq Jfina, 307
Hanafis, 368 JamI, 371
Hansl, 188 Jfiriyya, 377
Harbiyya- Janahiyya, 52, 54 Jacfar b. AbI Talib, 52
Harbiyya-Janahiyya, 52 Jacobites, 29, 368, 370
Hari Rud, 88, 90, 182, 183, 371 Jahangir Magre, 319
Harsha, 274 J ahanshah Kara Koyunlu, 347
Hasan Dilawar Khan, 290 Jalal Khan, 295 , 308
Hasanids, 128, 130- 135 Jalayirids, 347
Haydar Shah, 318 Jam, 195
Helan Shan, 213 Jamal QarshI, 140
Hephthalites, 28 Jamal QarshI, 269
Hinayana, 207, 211 Jammu, 315, 317, 318
Hoysalas, 278 Jamuka, 240, 253, 254
Hsi Hsia, 213- 216, 219, 220, 265 Janahiyya, 52
hsing shen, 215 Jawnpur, 290, 292- 295 , 382
Hulbuk, 93 , 94 Jaysalmir, 278
Humayun, 295 , 296, 320, 381 Jebe, 246, 258

468
ISBN 978-92-3-103467-1 Contents INDEX

Jehol, 232 Kebek Khan, 267, 269, 271 , 324

Jibal, 61 , 62 Kerait, 240, 251 , 253 , 254
Jibra'n b. Yahya, 42 Kerulen, 266
Jiqian, 212, 215 Khabushan, 377
Ju dynasty, 207 Kham Sultan, 346
Jujia of Jincheng, 207 Khalaj, 183, 188
Jurchen, 215 , 217, 219, 239, 265 Khaljis, 275 , 277, 285 , 293
Jochi, 265 , 266, 332 Khanballk, 264
Kharijites, 31- 33, 35 , 44, 49, 52, 56, 64, 300
KajI Chak, 320 Khazaria, 92, 370
Kacba, 55 Khazars, 370
Kachh, 298 Khentei mountains, 251 , 258
Kad"ir b. Ibramm, 141 , 142 Khidashiyya, 53
Kalhana, 311 Khidr Khan Sayyid, 338
Kalka, 258 Khingan mountains, 233 , 256
Kalmati, 309 Khokars, 190, 292
Kalmucks, 247 Khubilay, 246
Kanawj, 274, 292, 302 Khudaydad, 347
KARA KHITAY, 232 Khujand, 26, 90, 94, 95 , 129, 131 , 134, 137,
Kara Khitay, 22, 138, 139, 141- 143, 146, 190, 139, 243 , 265 , 268, 270, 271
191 , 237, 239, 240, 242- 247, 257, KHURASAN, 47
259, 265 , 270 Khurasan, 29- 35 , 37, 39, 40, 42, 44, 45 , 48- 65 ,
Kara Koyunlu, 347 86, 88, 89, 91 , 93 , 95 , 97, 98, 101 ,
Karabagh, 336 129, 134, 138- 140, 183, 184, 190,
Karabalghasun, 198, 202, 205 , 236 192, 193, 298, 301 , 326, 334- 336,
KARAKHANIDS, 125 346- 349, 369, 373, 375- 377, 384,
Karakhanids, 65 , 85 , 88, 126, 127, 129- 148, 385 , 389
152, 208, 243 , 244, 270, 371 Khurramiyya, 48, 52- 57, 60, 61
Karashahr, 207, 211 Khusraw Malik, 187
Karluk, 28, 36, 39, 42, 43 , 139, 140, 184, 197, Khwajagan, 371
208, 242, 243 , 370 Khwandamir, 327
Karmina, 270, 271 Khwarazm, 22, 29, 30, 91 , 96, 97, 142, 144,
Karramiyya, 183, 195 , 373 266, 269, 271 , 333, 346, 368, 369,
Kashghar, 43 , 126, 129, 130, 132, 133, 138, 374, 384, 387, 389
139, 143, 242, 245 , 246, 265 , 270, Khwarazm Shahs, 28, 125, 141 , 189, 190, 193,
320, 332 257, 265
KASHMIR, 310 Kimek, 28
Kashmir, 286, 31 0--313, 315- 322 Kirman, 187, 301 , 308
Katahr, 292 Kish, 37, 41 , 90, 129, 140, 271 , 324, 327, 343
Kattakurgan, 92 Kishtwar, 312, 315
Kay Kawus, 376 KITAN, 232
Kay Qubad, 277 Kitan, 205 , 213, 215- 217, 233- 244, 247, 260,
Kaykhusraw, 272 261 , 391
Kaysaniyya, 48, 51 , 54, 57, 60 Kocho, 28, 208, 211 , 212, 236, 238, 241
Kazaghan, 271 Kota Ram, 314
Kazakhstan, 137, 205 Kuba, 134, 147, 270

469
ISBN 978-92-3-103467-1 Contents INDEX

Kubrawiyya, 371 , 375 Mascud Mirza, 292

Kufic script, 148 Mascud of Ghazna, 184, 387, 390
Kuhistan, 63 , 191 , 386 MASHKEY, 301
Kutlugh Bilge Ktil Kaghan, 197 Mashkey, 299, 300
Kutlugh Shah, 368 Mazandaran, 59, 326, 336, 347, 377
Kutlugh Timur, 324 Mazdakites, 29, 52, 54, 55
Ktichltig, 141 Meerut, 189, 338
Klpchak, 24, 28, 258, 340 Melkites, 29
Klzi1, 210, 347 Merkit, 251 , 253
Koksaray, 329, 344 Mewat, 292
KOl Tegin, 235 Ming dynasty, 220
Ktichltig, 142, 143, 245 , 246, 265 Moghulistan, 22, 271 , 272, 324, 325 , 327, 347
MONGOLS, 248
LasharI, 307, 308 Mongols, 22- 24, 60, 192, 194, 220, 241 , 246,
Lakhnawti, 190, 192 247, 249- 252, 254, 256- 258, 260,
Lakshminagar, 315 262, 264, 265 , 271 , 275 , 277- 279,
Lalitaditya, 315 305 , 307, 312, 324, 326, 327, 329,
Lanzhou, 213 331 , 333- 336, 368, 385 , 386, 389- 392
Lavanyas, 312 Mubarak Shah, 267, 268, 271
Li Chiang, 202 Mubarakabad, 292
Li Deming, 213 Mucin b. Ahmad, 300
Liao dynasty, 208 Mudar b. Raja', 301
Mughira b. Ahmad, 300
Mir cumar, 309 Mughals, 320, 321
Mir Gwahram, 309 Muhammad b. Tughluq, 285 , 288, 307
Miranshah, 346, 347, 351 Muhammad Bakhtiyar KhaljI, 189
Mirza Haydar, 318, 320- 322 Muhammad Beg, 307
Macbar, 285 , 286, 290 Muhammad Ghawth, 381
Maharaj, 300 MULTAN, 302, 304
Mahmud ofGhazna, 131- 133, 144, 238, 299, Multan, 298, 299, 302, 303, 305 , 337
303, 311 , 372 Muzaffar I, 292
Mahmud TarabI, 266, 270, 271 Muzaffarids, 336
Mahmud Yalavach, 266, 267, 269- 271 Mongke, 255 , 259, 264, 266- 270
Malik Ahmad Yattu, 318, 319
Malik Kaffir, 279
Malik Mubarak, 290 Nanak, 381
Malik Shah, 65 Nasiriyya, 60
Malwa, 189, 278, 290, 292- 295 , 380, 381 Nazuk Khan, 320
Mamai, 333 Nurbakhshiyya, 376
Mamluks, 287 Naiman, 141 , 245 , 246, 272
Maragha, 368 Naqshbandiyya, 24, 348, 371 , 374
Marco Polo, 23 , 262, 311 , 384 Narojaipal, 311
Marghinan, 134, 135 , 146, 147, 270 Nasa, 34, 44, 91 , 142
Marwar, 306 Nasrid, 187
Mascud b. Hasan, 139, 140 Nawruz, 304
Mascud Beg, 266, 267, 269, 270 Nestorian Church, 368, 369

470
ISBN 978-92-3-103467-1 Contents INDEX

Nishapur, 33, 34, 37, 53, 55, 58, 61 , 64, 89, 91 , Qusdar, 298, 299, 301 , 304
94, 183, 187, 191 , 335 , 372, 376, 377, Qutula, 251 , 254
379, 384, 386, 389
Nizamiyya, 373 RaI Magre, 316
Nizari Ismacilis, 62 Ramananda, 381
Nizhniy Novgorod, 337 Ramdas, 381
Northern Ch'i dynasty, 235 Rashtrakutas, 274
Rawandiyya, 38
Ogedey, 247, 259, 260, 262, 263, 265- 269, RudakI, 94, 97, 99
385- 387 Rajah Man Singh, 295
Oghul Qaymish, 267 Rajasthan, 292
Oghuz, 28, 140, 186, 187, 190, 192, 235 , 236, Rajputs, 274
304, 370, 376, 379, 387 Ramadan, 33
Oljeytu, 368 Rayy, 34, 37, 55 , 56, 58- 61 , 347, 375
Ordos, 213 Renxiao, 218, 219
Ostiaks, 247
Rinchana, 312
Oxus, 23 , 28- 30, 48, 85 , 87, 101 , 131 , 138,
Rinds, 309
191 , 245 , 341

Pit Muhammad, 337, 346 Sufi dynasty, 333

Palas, 274 Silha Bhatta, 317
Partha, 311 Sa'in, 381
Pahlavi, 199, 370 Sabzavar, 326, 335 , 348, 377
Pakpatan, 374 SacdI, 371
Panipat, 296 Saffarids, 63
Panjab, 188, 191 , 274, 277, 278, 285 , 286, Samanids, 60, 61 , 64, 85 , 87, 88, 92, 97, 103,
291- 296, 298, 308, 320, 380 127, 129, 143, 144, 147, 184
Panjikent, 97, 98 Samgramraja, 311
Pratiharas, 274 Sammas, 306, 308
Prithvi Raja, 188 Sanjar, 139, 142, 187, 270, 373, 387
Sarbadars, 326, 327, 334- 336, 377
Qadiriyya, 24, 371 Sayyid Baraka, 329
Qasimiyya, 60 Sayyid cAlIHamadanI, 316, 375
Qabacha, 275 , 304 Sayyid dynasty, 291 , 293
Qabul Khan, 251 , 253 Sayyid Khidr Khan, 291
Qara Htilegti, 267 Sayyid Muhammad b. cAbd Allah, 376
Qatwan steppe, 139, 243 Sebtiktegin, 144, 184, 195 , 301 , 303
Qaydu, 264, 268, 269, 271 Seljuqs, 65 , 138, 139, 144, 190
Qazvin, 336 Semirechye, 39, 126, 130, 245 , 246, 324
Qinghai, 214 Shir Shah Sur, 321
Qiuci, 206 Shaficis, 195
QubawI, 147 Shah Beg Arghun, 307
Qubilay Khan, 264, 386 Shah cAbd Allah ShattarI, 381
Qum, 58, 61 , 347 Shah Hasan Arghiln, 305
Qunduz, 270 Shah Husayn, 305
Qur'an, 31- 33, 90, 189, 289, 367, 375 , 378 Shah Mir, 310, 311 , 313- 315 , 321 , 322
Qurayshite, 299, 302 ShahRukh, 291 , 292, 325 , 336, 346- 348, 376

471
ISBN 978-92-3-103467-1 Contents INDEX

Shams Chak, 320 T'ang dynasty, 209, 235

Shanxi, 213 T'ien Shan, 23, 206, 209
SharqI sultans, 290 T'ung Hu, 232
Shaykh Ahmad Ata YasawI, 371 Tatar Khan, 305
Shaykh cAbd Allah ShattarI, 379 TazI Bhatt, 319
Shaykh cAla' I, 382 Tabgach, 232
Shaykh cUthman, 379 Tahirids, 57, 60, 63
Shaykh Dawud Sabzavarr, 335 Talaqan, 34, 61
Shaykh Habib, 378 Tang Taizong, 209
Shaykh Ismacn KubrawI, 318 Tangut Toba clan, 212
Shaykh Khallfa, 326, 334, 377 Tatars, 240, 247
Shaykhiyya-Jilriyya, 378 Tatatungha, 25 5
Shen Zong, 213 Tayang Khan, 245
Shiburghan, 324 Tekish, 141- 143, 187, 190
Shira Muren, 233 Thacaliba, 50
Shughnan, 85 , 89, 189 Thankar, 189
Sichuan, 212 TIMUR, 323
Sigi Qutuqu, 255 Timur, 256, 272, 289- 291 , 305 , 316, 317, 325 ,
Sikandar, 295 , 296 327- 345 , 347- 349, 373, 376, 385 , 390
Sikhs, 381 Timurids, 310, 337
Simnan, 34, 375 Tirmidh, 29
SIND, 297, 306 Toghri1 Khan, 253 , 254
Sind, 188, 274, 277, 286, 288, 298, 299, 302, Toktam'ish, 332, 333, 336, 337
304, 305 , 307, 310, 315, 380 Toluy, 262, 265- 267
Singu Sali, 211 Toluyids, 264
Sirajr, 304 Tonyuquq inscription, 236
Siwalik, 188 Toqashi, 267
Siwistan, 306 Trebizond, 338, 346
Siyawush, 98 Tughluq Timur, 325
Sogdians, 98, 199, 201 Tughluqids, 284, 292
Soyurghatmish, 328 Tukharistan, 34, 89, 129, 185 , 189
Srivara, 322 Tun Bagha Tarkhan, 199, 201
Suhrawardiyya, 24 Tunguz, 233 , 239
Suizhou, 214 Turkistan, 22, 23 , 28, 125 , 133, 134, 241 , 243,
Sulayman mountains, 309 257, 267, 269, 271 , 312, 324, 332,
Sulduz tribe, 324 347, 385 , 389
Sultan Husayn Bayqara, 354 Turkmens, 139, 325 , 347, 387
Sultan Shah, 187, 190, 294 Tuytihun state, 212
Sultan Shah LOdI, 294 Ttirgesh, 30, 36, 39
Sunbadh, 37, 55
Sung dynasty, 208, 209, 212, 235 Uchch, 188, 275 , 305, 380
Sunnis, 49, 113 Uighurs, 29, 43, 197- 199, 201- 203 , 205 , 207,
Syr Darya, 29, 30, 43, 135 , 147, 265 , 266, 268, 209- 211 , 213, 236- 238, 241 , 256,
333 259, 261
S'ignak, 265 Ulugh Beg, 270, 346
Stibetey, 258 Umayyads, 29, 31 , 35 , 40, 49, 52

472
ISBN 978-92-3-103467-1 Contents INDEX

Urns Khan, 332, 333 Yacqilb b. Layth, 50, 64, 184, 376, 387
Ush, 147, 270 Yadavas, 278
Usrushana, 87- 91 , 99, 101 Yaghlakar, 202
cuthman, 29 Yahya Qarabi, 335
Utrar, 142, 146, 148, 265 , 268- 270, 346 Yasa, 329, 341 , 348
Uzgend, 130, 133- 135 , 140, 142, 143, 147, 270 Yasawiyya, 24, 372
Yenisei, 205 , 236, 257
Vijayamandirgarh, 189 Yesti Mongke, 267
Vijayanagar, 290, 348 Yestigey, 251- 253
Voguls, 247 Yeti Su, 265 , 267
Yuan Hao, 215 , 217
Yuan empire, 264
Wakhan, 85
Wakhsh, 135 , 189 Zafar Khan, 290, 292
Wang Yande, 210 Zahak, 99, 184
Wayhind, 303 Zakariyya Razr, 95
White Horde, 268, 332, 333, 343 Zarafshan, 29, 88, 89, 93 , 101 , 387
William ofRubruck, 23 , 268 Zayd b. cAU, 59
Ziyad b. Salih, 36
Xianbei, 212 Zuljil, 312
Xining, 214 Zuzan, 53, 386

473
ISBN 978-92-3-103467-1 Contents History of civilizations of Central Asia

History of civilizations of Central Asia

Volume I

The dawn of civilization:

earliest times to 700 B. c.

Volume II

The development of sedentary and nomadic civilizations:

700 B. C. to A. D. 250

Volume Ill

The crossroads of civilizations:

A. D. 250 to 750

Volume IV

The age of achievement:

A. D. 750 to the end of the fifteenth century:
Part One
The historical, social and economic setting
Part Two
The achievements

Volume V

Development in contrast:
from the sixteenth to the mid-nineteenth century

Volume VI

Towards contemporary civilization:

from the mid-nineteenth century
to the present time

Cover: Figure of a horseman on the inner surface of a twelfth century Persian lustre ware bowl,
outside diameter 35 cm. Rayy or Kashan (Iran). Courtesy of M.I. Mochiri.

474