Buddhism in Central Asia by B.N. Puri

BUDDHISM IN
CENTRAL ASIA
B.N. PURI
BUDDHISM IN CENTRAL ASIA
BUDDHIST TRADITION SERIES
Edited by
ALEX WAYMAN
Editorial Advisory Board

J.W. DE JONG
KATSUMI MIMAKI
CHR. LINDTNER
MICHAEL HAHN
LOKESH CHANDRA
ERNST STEINKELLNER
VOLUME 4
BUDDHISM
IN CENTRAL ASIA
B.N. Puri
MOTILAL BANARSIDASS PUBLISHERS

PRIVATE LIMITED • DELHI
First Edition: Delhi, 1987
Reprint: Delhi, 1993, 1996
© MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED

All Rights Reserved
ISBN: 81-208-0372-8
Also available at:

MOTILAL BANARSIDASS
41 U.A. Bungalow Road, Jawahar Nagar, Delhi 110 007
120 Royapettah High Road, Mylapore, Madras 600 004
16 St. Mark's Road, Bangalore 560 001
8 Camac Street, Calcutta 700 017
Ashok Rajpath, Patna 800 004
Chowk, Varanasi 221 001
PRINTED IN INDIA
BY JAINENDRA PRAKASH JAIN AT SHRI JAINENDRA PRESS.

A-45 NARAINA, PHASE I, NEW DELHI 110028
AND PUBLISHED BY NARENDRA PRAKASH JAIN FOR
' MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED,
BUNGALOW ROAD, DELHI 110007
FOREWORD
Central Asia is still an enormously important area of the world

in a political sense, its vitality unabated. Buddhist texts were
certainly disseminated into the Khotan area during the time of the
‘Old Silk Road’ (100 b .c . - a . d . 200). This road went from China
to the Oriental Roman empire through Central Asia and had a
branch extending down into Northwest India. A variety of dialects
and dialect mixtures were current in Central Asia. Buddhism
spread from India by way of the trade routes, in China starting
with Tunhuang at the Western gateway. After the downfall of the
Han in a .d . 220 Buddhism rapidly advanced and from about
a .d . 300 had penetrated the high gentry clans in Northern China.
When the Tibetan king Sron-btsan sgam-po (b. a .d . 569) was
converted to Buddhism by two Buddhist princesses, one Nepalese,
the other Chinese, whom he married, he sent a mission to India
(possibly. Kashmir) to create a Tibetan alphabet. Then translations
were begun in the seventh century, at first from both Sanskrit and
Chinese, by end of that century just from Sanskrit.
B. N. Puri’s work on Buddhism in Central Asia recounts these
fascinating events. He has spared no inspection of previous scho
larly work for his coverage of the main facts. Whether it be the
history, the literature, realities of life, or the art, Puri maintains
a firm control of the relevant supporting treatises. Students of
Central Asia should welcome this addition to the topic and the
bibliographical introduction. There are of course many specialized
works on particular aspects of Central Asia but Puri’s broad
coverage is probably unique. Sixteen tastefully chosen plates add
an artistic touch to this valuable addition for the Buddhist
Traditions series.
A lex W a ym an
PREFACE
The study of Buddhism in Centra! Asia cannot be carried

out in isolation, since it is related to several factors. Its in
troduction there was the result of religious missions as also of
peaceful international relations. Traditional accounts, no doubt,
suggest that Indian colonists settled in Khotan during the reion
of Asoka. This may not be accepted as final since no precise
date can be fixed for the introduction of Buddhism in the Tarim
basin. It is, however, evident that the religion of the Tathagata
was flourishing in Central Asia about the time of the Christian
era. It was from there that Buddhism spread to China not later
than the middle of the first century A.D. The enormous breadth
of the landscape provided by Buddhism and Buddhist savants
in Central Asia from China to the frontiers of Persia, during,
the course o f a long period of nearly a millennium or more,
with various races contributing to its growth from the Yuehchi
and the Kusanas to the Uighurs, therefore, demands its
comprehensive study. This study is to be undertaken against
the background of the geographical area—its configuration
and peoples—nomadic and sedentary, as also its location as
the meeting ground of the Orient and the Occident. It was a
two-way traffic with the role of Central Asian peoples in
other countries, and that of others in different parts of this
region.
The political history ofCentral Asia has to be studied in details
since royal patronage to the creed of the Buddha was always
forthcoming, as recorded by the Chinese pilgrims who passed
through their kingdoms. Those lying on the Northern Route—
Badakshan, Kashgar and Kucha—were great centres of the
Sarvastivadin school, while Mahayanism dominated in Khotan
and Yarkand. This demarcation of distinct representation
symbolised two religious currents passing over these areas. The
spread of the Sarvastivadin school o f Buddhism is no doubt
viii Preface
connected with the growth of the Kusana empire, maybe pre

ceding the conversion of Kaniska as supposed by some scholars.
Mahayanism subsequently became more popular. The two
schools were like coaches provided for travelling the same road
to salvation, gradually absorbing the traffic awaiting the final
journey. Names of Hindu rulers in the north and a long list of
Vijaya monarchs of Khotan, available from the Kharosthi re
cords which also mention purely Indian as also mixed names
of donors and administrators, suggest the Indian way of life
being adopted in the countries of the Tarim basin.
Those contributing towards the expansion of Buddhism in
Central Asia, as also providing new dimensions included savants
o f different nationalities, besides the Indian ones. These com
prised Tokharian, Parthian, Sogdian and Yuehchi scholars,
some of whom were from royal families. The names of
D harmagupta from Kashghar, Suryabhadra and Suryasena
from Chokkuka—Karghalik-Yarkand, Dharmsk$ema, Siksa-
nanda from Khotan, and, above all, Kumarajiva of Kucha an -
his contemporaries—Dharmamitra, Buddhayasa, Buddhabhadra
and many others, are notable for translating Buddhist texts into
Chinese. The Buddhist literature discovered in Central Asia is
equally abundant and, like its architecture, represents severa}
periods, catering to the needs of both the schools. The fragments
of the Sanskrit Agamas from Turfan, Tun-huang, and in the
Khotan district those of the dramas and Kavyas of Asvaghosa
from Turfan, the Pratimoksa of the Sarvastivadins from Kucha
and numerous versions of the anthology called Dharmapada
or Udana, with extracts in Tokharian and Sanskrit, may be
noted in this context. The newer stratum of literature consists
of Mahayanist sutras and includes Prajnaparamitd, ‘the Lotus
of True Law’—Saddharmapuntjarika and Suvarnaprabhasa Sutra
—translated into Uighur and Iranian oriental, and to a still
later period the dhdranis or magical formulae which have been
found in great numbers. Turkish sutras discovered at Turfan
contain a discourse of the Buddha to the merchants Trapusa
and Bhallika. Buddhist texts in Kuchean, Tokharian, Sogdian
and Bactrian Greek have also been traced. A survey of the
Buddhist literature and the contribution of the Buddhist
savants have been made separately. The relation of Buddhism
Preface ix
to other religions, especially Manichaeism and Nestorian

Christianity was in tune with the eclectic and tolerant spirit
of the peoples of Central Asia. Tibetan Lamaism and its
acceptance in Mongolia suggests that Buddhism provided a
creed acceptable in different forms to superstitious, emotional
and metaphysical minds.
The study of material culture in Central Asia provides an
insight into cultural integration and people’s life, manifested
through different facets—with joint family ties, position of
women, items of food and food habits, dress and ornaments,
pastime and recreations, agricultural and pastoral economy,
trade and transport and several other items related to the
socio-economic life. It is interesting to find Buddhist monks
owning land and slaves, and participating in the material life
of the time. Buddhism, thus, provided a living and a changing
stream of thought adaptable to men of different emotional
backgrounds. New forms of Buddhism providing a moral ideal
and not personal perfection or individual salvation were evolved
with Buddhas and Bodhisattvas serving as angels of mercy,
peace and knowledge. These are manifest in Central Asian art.
These Bodhisattvas are supposed to have indefinitely postponed
their nirvapa process for the sake of alleviating the sufferings of
mankind, while faith in a Buddha, especially in Amitabha
could secure rebirth in his paradise. *
Factors accounting for this change in the gospel of Buddhism
were both internal or Indian, and external. The Indian factor,
involving development of both Brahmanism and Buddhism in
a parallel way with more areas of similarities in thought and
action, seems to have emanated from Taxila—Taksasila, the
great centre of learning, as also from Kashmir which provided
the largest number of Buddhist savants to Central Asia, and
from thence to China. Foreign influence was the product of all
those who contributed to Buddhism after accepting it in their
way of life. Greeks, Parthians, Sakas and Ku$anas were greatly
responsible for stimulation to Buddhist mythology and imagery.
When Buddhism passed into the hands of those foreigners,
especially Greeks, who were accustomed to Greek statuary,
the desire to venerate Buddhist personalities, especially the
figures of Buddha and the Bodhisattvas, took a definite shape
x Preface
on the western model without alienating Indian characteristic

features.
The Central Asian artists—some being imported from the
region of Gandhara and including Greeks, like Titus at Miran,
were inspired by Greek tradition as modified in Rome. Native
tradition, however, accepted and adopted these artistic influences
with discrimination. With the expansion of Buddhism in Central
Asia and setting up of stupas and free-standing, shrines, the
demand for artists and sculptors shot up. Commercial enterprise
on the trade routes with the patronage and contribution of
traders and merchants provided incentives for artists who receiv
ed handsome payments for their service. Well-defined styles of
Central Asian paintings are the result of contributions and im
pacts of different art traditions—Indian, Persian and Chinese.
The common factor, however, is the Buddha and his legend.
Diversities in design and treatment did not rule out an identical
mannerism in the same composition at different places. A study
of the Central Asian Centres of art with particular reference to
paintings at Buddhist Cave Shrines, lying on the Northern as
well as the Southern routes, would not fail to reveal, in a way,
the competitive spirit of the votaries of the Buddha in venerat
ing their lord. They displayed their artistic talents in projecting
new forms ofûddhist ideals for the masses.
A comprehensive study of Buddhism in Central Asia is, thus,
attempted in this work, taking into account all the factors and
forces responsible for its introduction, prosperity, and eventual
decay and decline. Nature no doubt provided complete shelter
to Buddhist art treasures and polyglot libraries which were seal
ed off to protect them from vandalists and marauders. These
lay enveloped till the spade of explorers and archaeologists
could reveal the mysterious past of Buddhism in this vast area
of the Tarim basin studded with sites and spots which are
separated by considerable distances, and which were once humm
ing with the spiritual and temporal activities of the people
wedded to the religion of the Buddha.
The present study fulfils the demand of the Indian Council of
Historical Research, New Delhi, which p-ovided me a fellow
ship as also contingency expenses towards its successful com
Preface xi
pletion. I am grateful to the Council.
This work is dedicated to the memory of my younger
daughter Taruna, who left us over seven years back in a road
accident. Her memory is ever fresh and green and is a source
of constant incentive for me to work with dedication.
B. N . P uri
CONTENTS
I. INTRODUCTION 1
Role of Central Asian Peoples 7
Ancient Routes 17
Buddhist Finds—Literary Texts and Monuments 24
II. EARLY HISTORY OF CENTRAL ASIA 30

The Political States of Central Asia 45
Kashgar 46
Khotan 52
Northern Route Sates 68
Kucha . 69
Agnidesa or Karasahr 74
Kao-chang or Turfan 77
The Regality and Buddhism in the Northern States 80
III. BUDDHISM AND BUDDHIST SAVANTS OF 86

CENTRAL ASIA
Buddhism in Afghanistan, Bactria and Parthia 89

Buddhism and the Southern States 104
Buddhism and Buddhist Scholars in the Northern States 114 '
Kumarajiva—His Life and Contributions 116
Kumarajiva and His Contemporaries 121
Later Buddhist Savants 125
Buddhism and Other Religions in Central Asia 129
Brahmanism 130
Manichaeism, Nestorian Christianity and Buddhism 136
Some New Trends in Buddhism 141
Tibetan Buddhism 147
Buddhism in Mongolia and Tibet 157
XIV Contents
IV. LANGUAGE AND LITERATURE 174

Languages 179
Scripts 185
Canonical Literature 189
Local Translations 210
Tibetan Buddhist Literature 216
V. MATERIAL CULTURE 225

Cultural Integration 227
Family Life 232
Food and Food Habits 236
Dress and Ornaments 237
Pastime and Recreations 241
Agricultural Economy 243
Handicrafts and other Industries 246
National Economy and Medium of
Exchange and Barter 248
Labour and Transport 250
Administration & Rural Economy 252
VI. THE ART OF CENTRAL ASIA 255

Miran 260
The Khotan Complex 268
The Northern Schools 274
Kara-shahr 282
The Turfan Group 284
Tun-huang 290
Soviet Central Asia 292
Afghanistan 297
Fondukistan _ 302
Begram 303
Hadda 304
Gandhara Region 305
Tibetan Art 307
Contents xv
VII. THESUMMING-UP 317

S elect B ib liog raphy : B ooks 339
P apers 341
I ndex 347
PLATES (after p. 352)
I. Buddha with Six Monks

II. Bodhisattva Seated on a Throne with a Devotee
III. Two Girl Worshippers
IV. Hariti
V. Bust of Buddha
VI. Worshipper or Indra
VII. The Trimurti Divinity with a Small Seated Buddha
VIII. Cowherd Listening to the Sermon
IX. Head of Mahakasyapa
X. Group of Swimmers
XI. Goddess and Celestial Musicians
XII. Buddha and Praying Monk
XIII. Worshipping Bodhisattva
XIV. Buddha under a Canopy
XV. An Uighurian Prince
XVI. Bodhisattva
CHAPTER I
INTRODUCTION
The term Central Asia is supposed to connote the Tarim Basin,

with the inclusion of neighbouring regions such as the Oxus
region and Badakshan. This basin is a depression, surrounded
on three sides by high mountains. Oh the north it is bounded by
the Tien-Shan ‘the Celestial Mountain’, while the snowy Kiun-
Lun on the south separates it from the Tibetan part of Central
Asia. To the west, the Pamirs, the ‘Imaos’ of the ancients, join
the Tien-Shan to the Hindu Kush, giving rise to the head-waters
of the Oxus on its western flank. On the east, the barrier dividing
it from China is relatively low. The water of the entire area is
discharged through the many branched Tarim river into Lake
Lob-Nor—only a flooded morass. The basin is a desert with
occasional oases lying chiefly near its edges. There might have
been more fertile portions in the past, but this remote and lonely
region has only provided interest for exploration among archaeo
logists and explorers of different nationalities within the last
hundred years or more. Its complete isolation from oceanic in
fluences as also its geographical insularity have no doubt contri
buted to evolving its own cultural pattern with the contribution
of warring tribes of this as well as the neighbouring areas in the
past. While in the north the taiga, the Siberian forest zone serves
as a barrier for any communication, in the south an almost un
broken chain of mountain ranges, nearly four thousand miles or
six thousand four hundred kilometres long running from China
to the Black Sea restricts any access in the direction of the South-
East Asia, the Indian Sub-continent and the Middle East. Only
certain sections of this long chain of mountain ranges—the
Hindu Kush, the Paropamisus and the Elburz have never restric
ted the movements of peoples in either direction. The two plate
2 Buddhism in Central Asia
aux—Tibet on the south, enclosed by the Himalayas, and Iran,

flanked on the south-east by the Kirthar and Sulaiman ranges,
and on the south-west by the Zagros, have provided historical
links with Central Asia proper.
The eastern and western limits of Central Asia are not properly
defined. In the east, the Great Wall of China could provide an
approximate line, while in the west the grasslands of the Ukraine
extending as far as Rumania and Hungary supposedly provide
both a geographical and historical extension of the Central Asian
steppe zone.1 This might be an exaggerated concept of Central
Asian western limit, but it is now accepted that the vast region in
its geographical and political aspects need not extend beyond
Iran, especially its eastern part. The physical features of Central
Asia with its predominant steppe and desert area, include some
of the highest mountain ranges in the world as also depressions
like those around Turfan in Sinkiang and the area north-east of
the Caspian with extreme rise of temperature. This last region of
Central Asia lying approximately between latitude 35°N and
55°N could be divided in a convenient manner into a northern
and a southern zone separated by a line along the Syr-Darya and
the Tien-Shan. The northern zone with its sufficient moisture,
provided extensive grazing ground for pastoral peoples, while
the southern one was primarily an urban area and oasis society
with extensive cultivation of land through skilful application of
irrigational techniques for agricultural purposes. Shortage of
water no doubt required hard and laborious efforts in storage of
rain water into reservoirs. The clash between the northern no
mads and the settled and domesticated peoples of the southern
region was natural. The nomadic peoples north of the Tien-
Shan from time to time plundered and occupied the oases of the
1. G. Hambly. Central Asia (London, 1969)—henceforth Hambly, p. 2.

In contemporary Soviet usage the geographical term ‘Central Asia’ has
rather different connotation from the term as it is normally used in western
Europe. It includes the territory of the Uzbek, Tadzhik, Turkmenian and
Kirghiz Republics and the southern part of the Kazakh Republic, but ex
cludes eastern Turkestan, Mongolia and Tibet which'are covered by the
western European term (Aleksandr Belenitsky : Central Asia translated by
James Hogarth—London (date not mentioned)—henceforth Belenitsky,
p. 15.
Introduction 3
Tarim Basin without making their occupation a permanent

feature.
The configuration of the mountain ranges has no doubt exer
cised control over the movements of peoples of Central Asia
from one region to the other. So also the influence of the de
serts has been profound in this direction. Geographical factors,
thus, tended to segregate civilizations bordering on its periphe
ries—Indian, Iranian and Chinese in the past. The ancient cara
van routes provided not only mutual information but also
established some sort of limited but important contact between
the two extreme ends—China and Iran and the western world.
Of course, in terms of commerce and cultural achievements there
have been only some important parts of Central Asia, as for
instance the one between the Amu-Darya (Oxus) and the Syr-
Darya (Jaxartes), and the area known as Khurasan to the Arabs,
which lay to the south of the Amu-Darya and extended south
west as far as the Iranian Dasht-i-Kavir. Between the middle
reaches of the Amu-Darya and the Syr-Darya lay the country
known to the Greeks as Transoxania, and to the Arabs as
Mawarannahr, with Bukhara and Samarkand as its most impor
tant urban centres in the medieval period. The latter has no
doubt an old history of its own, tracing its antiquity to the
period of Alexander’s invasion, or even earlier. There were many
other centres of importance—political and cultural—in the medi
eval period. Khwarazm on the lower reaches of the Amu-Darya,
Shash, north-east from Mawarannahr beyond the Syr-Darya
comprising the country round Tashkent, were other areas of
importance in the medieval period. For the earlier period one
has to follow the ancient caravan routes linking China with the
west, which passed through many centres. These caravan trade
routes are also described as ‘Silk Routes’ or ‘Silk Roads’, trans
gressing the Central Asian forbidden no-man’s land. While geo
graphical factors provided insularity and security to the Central
Asian peoples, with the Lop and Gobi deserts in the east, the
long chains of the Tien-Shan and the Kun-Luns to thenorthand
south respectively and the Pamirs to the west linked to the Kun-
Luns by the Karakoram ranges, man triumphed over natural
barriers and successfully crossed the hurdles lying in his way.
Buddhism in Central Asia
Silk, one of China’s chief mercantile commodities became the

key word along the general trade and transport routes. It was
along these routes that not only traders and merchants moved
from one direction to the other, but Chinese pilgrims and Indian
Buddhist savants covered long distances to satisfy their intellec
tual curiosity as also to convey the message of the Tathagata in
the language of the peoples of those areas where these scholars
were invited or where they finally settled down. The expansion
of Buddhism in Central Asia is closely linked with the silk trade
routes and the centres associated with these communication lines
as also the settlements and the sylvain retreats nearby for meet
ing the seclusive requirements of the Buddhist monks. The finds
of monuments associated with Buddhism no doubt confirm this
affirmation. A study of Buddhism in Central Asia would demand
fuller reference to the land and its peoples, as also to its physical
and political geography in depth, and also to the nature of
migratory movements of these peoples and its impact on the
areas lying with the peripheries of Central Asia.
This interior portion of Asia is at present divided politically
into three parts : a Soviet one, a Chinese one and Mongolia. The
Soviet part of Central Asia was called Turan, now referred to as
Russian or Western Turkestan. It includes the present territory
of Uzbek, Tadzhik, Turkmanian and Kirghiz Republics and the
southern part of the Kazakh Republic. The Chinese part, histori
cally known as Chinese Turkestan, is now represented by Sinki-
ang-Uighur Autonomous Region. It is bordered on the north
east by the Mongolian Peoples’ Republic, on the south-west by
Kashmir and a narrow strip of Afghanistan, and on the west
and north by the Central Asian Republics of the Soviet Union.
Sinkiang, physiographically, consists of two mountain-ringed
basins from the east-west treading Tien-Shan. The Dzungarian
basin in the north, with an elevation of 600 to 1500 ft., receiving
summer rainfall, makes it an ideal grazing land, while the Tarim
basin with an elevation of 2500 to 3000 ft., more arid than
Dzungaria, has the sandy Takla Makan desert in the centre and
the salt lake and marsh land of Lop Nor, at the eastern end.
Thus, Central Asia, as a whole, is a land of sharp geographical
contrasts with the greater part of the area occupied by high
mountain systems or great deserts, unfavourable for human
Introduction 5
settlement. Many of its river valleys, however, with their fertile

loess soil, have been occupied by settled population from very
early times.
It has, no doubt, been established by archaeological investi
gations and explorations that man appeared in Central Asia as
early as the Palaeolithic period and continues to be in occupa
tion of some part or the other since then, and has been in com
munication with peoples of the Orient and the Occident. The
continental forces have never been lull and the role of the peo
ples of Central Asia has always been significant in the history of
other countries as also in the export of tribal culture in those
areas. It is proposed2 that the wild horse was first domesticated
on the steppes of Central Asia and that it was from this region
that the horse-culture—the use of the horse for driving and later
on for riding—gradually spread to other parts of the world.
Many objects associated with horses, the saddle and the stirrup
of the later times, had their origin in Central Asia. As horse
riding became very common, so also the custom of wearing
trousers spread from Central Asia to other parts of Asia and
Europe. Another product closely associated with horse riding,
namely boots, gradually replaced the slippers and sandals, which
were universally worn. At a later time the Central Asian people
also initiated the custom of putting heels on boots and shoes. As
such, the contribution of Central Asia to the diffusion of culture
has been significant and the role of this region is considered still
more important in the transmission of cultural traits from one
part of the world to another.
The exchange of cultural traits between East and West also
took place by way of Central Asia. New inventions, ideas,
2. McGovern : Early Empires o f Central Asia, Chapel Hill, 1939

—henceforth—McGovern, p. 2. Horses are known to have existed sporadi
cally at a very early time in different parts of the old world. They were
certainly known to the vase-painters of Anau, of Tripolye, and of North-
Western China, but among these peoples horses remained relatively un
important adjuncts. With the northern nomads, on the other hand, the
horse played an all important role, and the spread of these Nomads into
other regions is always associated with the spread o f horses and horseman
ship {op. cit, p. 36). For the close association of the horse with the Aryans,
see Childe : The Aryans, p. 83.
manners and customs were transmitted from one region to an

other transgressing political boundaries. Constant exchanges bet
ween East and West were provided through this region of the
world—from Europe or the Near East to the East, India and
China, as also in the reverse direction. These exchanges of cultu
ral traits started long before the period of written history. Exca
vations at Anau3 in Central Asia suggest that the ancient civili
zation here provided a link between the early Near Eastern and
Far Eastern civilization. In historical times, too, Alexander’s
invasion provided cultural stimulus to the Far East through the
peoples of Central Asia. The emergence of the well-known
Greek school of art, also called Greco-Buddhist Art, exerted a
widespread influence in different parts of Turkestan and even
tually reached China where it revolutionised the indigenous
school of painting and sculpture.4 It is proposed that the impor
tation of Greco-Roman glass through Central Asia into the
celestial Empire of China had a profound influence on local
workmanship, and indirectly contributed to the invention of
Chinese porcelain. It was also through Central Asia that China
imported from Iran alfalfa and grape vine, and that country also
contributed towards the development of Chinese armour as also
its military strategy. Manicheanism, a third century Persian reli
gion and for a long time a rival of Christianity in Central Asia,
could also secure a foothold in China. Central Asia also estab-
3. For the excavations at Anau, see R. Pumpley : Explorations in
Turkestan {Washington, 1903-04) and also his Pre-historic Anau (Washing
ton, 1905-08).
4. Bachhofer in his work *A Short History o f Chinese Art' (London,
1944)—henceforth—Bachhofer= Chinese Art refers to the existence of Bud
dhist sculpture of the Khotanese type in northern China. According to him,
rarely in the history of sculpture can the origin o f certain forms be so
exactly located as in the case of the large standing Maitreya in bronze, now
in the Metropolitan Museum, New York. They had been taken over lock,
stock and barrel from the clay and mud statues of Kucha in Central Asia.
This particular style of representing a garment had developed there as an
imitation of the mature stucco sculpture of the Kushanas in north-western
India and Afghanistan. Chinese Buddhists and sculptors in the second half
of the fifth century looked upon them as Buddhist statutory par excellence,-
they took them as models and endeavoured to copy them as faithfully as
possible. (O p.cit, pp. 65-66). See also Waldschmidt, E; Gandhara, Kutscha,
Turfan (Leipzig, 1925).
Introduction 7
lished communication between India and China, and it was from

here that Buddhism spread to China.5 Many of the early mission
aries carrying the message of the Buddha to the land of Confu
cius were not from India but native scholars of Central Asia.
China, in return, also transmitted cultural traits of its own to
the Near and Middle East.6 Many features of her civilization were
passed on the same channel. Peach and apricot, as also ginger
and tea and several other items were indigenous Chinese pro
ducts which were introduced into the western world through
Central Asia. It was in the time of Augustus that Chinese silk
reached Rome, and for several centuries the import of silk from
China was a regular feature of the commercial activity of the
Roman Empire- The Great Silk Road from China to Rome pass
ed through Central Asia, and the control of this Road, occasion
ally became a subject of dispute leading to important political
changes. It is also suggested that the art of paper making dis
covered by the Chinese in the second century A.D. was carried
through Ceniral Asia to the Arabs in the eighth century, and the
European method of paper making is merely a copy of the old
Chinese craftsmanship. The issue of paper currency, well-known
in .China, was taken to Persia during the thirteenth century by
the Mongols, a Central Asian people, whose extensive empire
covered both China and Persia. Several other instances could be
quoted like the art of printing of books, known in China as
early as the beginning of the tenth century, which were taken
from the Chinese and then developed in the West.
Role o f Central Asian Peoples
While Central Asia provided cultural stimuli and also served
as a transmitting centre in communication between the Far East
5. The story of the first contact between India and China is mixed up
with legend. It is proposed by Bagchi that Buddhist missionaries from
India made their first appearance in the Chinese capital as early as 217 B.C.
under the Tsin dynasty. This story has no historical confirmation. It is,
however, known through an undisputed authority that in the year 2 B.C.
Buddhist texts and images were first presented to the Chinese court by the
Yueh-chi rulers. Buddhist missionaries certainly did not arrive in China
before the year A.D. 65. (Bagchi : India and China, Greenwood Press, West
Port, Connecticut, 1975, p. 6).
6. For a fuller account—see McGovern : Op. cit.
Oriental and the Occidental world, the role of the peoples of this
region in the history of Asia and Europe also deserves consider
ation. In fact, this participation has been the result of a long
series of migrations and invasions carried out by the warring
tribes claiming Central Asia as their homeland through the ages.
From earliest times Central Asia had at least two distinct and
separate racial and linguistic groups, called the Scythians and
the Huns. The former are supposed to belong to the ‘Great White
race’, speaking an Aryan or Indo-European language. The Huns,
however, belonged to the ‘Great Yellow race’ with a good deal
of Mongoloid blood, and speaking a language different from the
former and generally termed as Turanian or Ural-Altaic. The
Scythian or Indo-European element in the population of Central
Asia was dominant in the region now called Turkestan and it was
here that the members of this group, pressed by political as well
as economic forces, spread outward in all directions.7 The
Scythian migrations had greater impact upon the regions to the
south of Turkestan, sweeping over the Persian Plateau and into
North-Western India where they settled down permanently, shak
ing off their nomadic norms and habits.Those Scythians left in
Turkestan, however, retained their nomadic habits and were not
slow to invade their kinsmen to the south, leading to permanent
conquest and occupation. The Scythian group in Parthia ruled
there for nearly five centuries, establishing political control over
the agricultural population of the Persian Plateau. The position
in North-West India (now Pakistan) was, however, different with
several Scythian groups ruling one after another.
The Hunnish group, consisting of various tribes generally
spoken of as Hiung-nu, a term given by the Chinese to their
adversaries, dominated that region of Central Asia, called
Mongolia. The unified Hunnish empire posed danger to the
security of China. Earlier, the Great Wall of China was built
in 214 B.C. to'keep the Hunnish nomads out of bounds from the
fertile plains of the Yellow River. The Chinese no doubt manag
ed to prevent the Huns from securing a permanent footing in
China till the close of the third century A.D. The centuries
7. McGovern : Op. cit, pp. 6flf.

Introduction 9
following, however, witnessed the turn of fortunes8 with the

Huns becoming masters of all Northern China. TheHiung-nu or
Huns also moved westward and dislodged the Yueh-chi from
their homeland who, in turn, clashed with the Wu-Suns and
overpowered them. The latter with the support of the Hiung-
nu had another clash with the Yueh-chi, compelling them to
leave westwards, and occupy the territory of the Sakas, called
Ta-hia. After settling down, they were no longer nomads and had
divided their territory into five principalities, of which the Kuei-
Shuang was the most important. The capital of the kingdom of
Ta-Yueh-chi was Kien-chi (Lan-chu) with Ki-pin lying on its
southern frontier. In the Annals of the Later Han Dynasty, the
Yueh-chi capital is Lan-shi, the old capital in Badakshan, and
it also refers to the integration of five principalities by the Kuei-
shuang prince Kieou-Stsieou-Kio (Kujula Kadphises) who styl
ed himself as king.
The movements of these wild tribes and their occupation of
Bactria is also noticed by classical Greek and Latin writers.
Strabo mentions a Saka conquest of Bactria9 where the Greek
kings were ousted by Scythian nomads, and some of these noma
dic tribes are mentioned by him, notably the Asioi, Pasianoi,
Tocharoi and Sakarauloi. Trogus also provides an account in
the 41st Book, while dealing with theestablishment of an empire
in Bactria byDiodotus about the middle of the 3rd century B.C.,
of the way the Scythian tribes, the Saraucae and the Asiani took
possession of Bactria and Sogdiana. In this connection, according
8. While Turkestan was ruled by the Scythians, Mongolia in this

period was dominated by various Hunnish tribes, generally spoken of as
Hiung-nu, a term given by the Chinese. The Great Wall o f China was built
in 214 B.C. to keep the Hunnish nomads out of the fertile plains of the
yellow river. Shortly afterwards the Hunnish inhabitants of Mongolia inte
grated themselves to form a single empire which in one form or another
lasted for over three hundred years (B.C. 209—A.D. 160). McGovern :
Op. cit, p. 18.
9. The references to the Yueh-chi and their relations with the Kuei-
shuang-Kusanas, as recorded in the classical Greek and Latin sources, are
recorded in full and considered in detail in my work India under the Kusd-
nas, pp. 5ff. and note pp. lOff.—henceforth Puri—Kufdnas with comprehen
sive bibliographical details. The political history of this Central Asian dy
nasty in India is discussed in detail in this work.
to Justin, the Bactrians lost both their empire and their freedom
being harassed by the Sogdians, the Drangae and the Arii, and
were finally oppressed by the Parthians. Further information
relating to the Asiani becoming kings of the Tocharians and of
the annihilation of the Saraucae is provided by Trogus in the
‘Prologus’ of the 42nd book. The Asiani are identified by Sten
Konow with the Yueh-chi of the Chinese Annals. According to
the late Norwegian Professor, the Tocharians were well-settled
in and to the east of Bactria, when the Yueh-chi became their
masters. The relations of the Kusanaswith the Yueh-chi, either
as one of the five Yab-gou or tribe of the big pastoral race, or as
one of those five clans or kingdoms which became dependent on
the Ta-Yueh-chi after their conquest of the Ta-hia is evident
from the Chinese annals. The capital of the Yueh-chi became
the old Ta-hia capital Lan-shiin Badakshan which remained
their stronghold down to the fifth century A.D. The Yueh-chi
occupied the whole of Ta-hia country in the period of the Hou-
Han-Shu. According to Chavannes, henceforth they are Ta-hia.
The Kusanas supposed to be an important branch of the
Yueh-chi finally succeeded in consolidating their hold and estab
lished a vast empire from the southern parts of Central Asia,
including Afghanistan to Bihar in India, and from Kashmir to
Sind (now in Pakistan) in the south-west. It was in fact the
Ku§anas who were instrumental in bringing out an integration of
peoples of different nationalities into a single political fabric. A
detailed study of the Kusana history in the context of political
unification as also their ^contribution to Buddhism, would, how
ever, be made later on. In this context of the role of the Central
Asian people, it may be pointed out that the Huns who had
pushed out or absorbed the Scythians or Indo-European peoples,
slowly moved westward. By the fifth century A.D. they had
complete control over all parts of Turkistan, and in a clash with
their immediate neighbours the Sassanids, they defeated them
and forced them to pay tribute. The Huns moved further west
and set up their kingdom in the middle Danube basin10 and
10. McGovern : Op. clt, p. 385. The centre of the Hunnish kingdom in
this period was the middle Danube basin, corresponding to the Austro-
Hungarian domain of the nineteenth century.
Introduction 11
extended their tentacles in different directions, dominating

over other tribes.
References might as well be made of the Hephthalite Huns,
also called the White Huns11 who originally lived in the Altai
region and are mentioned by the Persian historians as Hayathe-
lite and the Chinese as Ye-ta or Hephtha, and in Indian records
as Huna. In the beginning of the 5th century they were an
important people who owed allegiance to another nomadic tribe
of Turkish origin called Juan-Juans in Mongolia. In the second
quarter of the same century they started spreading westwards
and conquered the entire steppe area upto the Aral. Their terri
tory included the valley of the Hi up to the Balkash, the valley
of the Issiq-kul. The steppes of Chu and Chao, and the valley
of the Jaxartes upto Aral, they occupied Sogdiana and Tokha-
restan by A.D. 440, and probably Balkh followed suit about the
same time. Akhsunwar of the White Huns invaded Khorasan in
484 and killed king Peroz. In the South they clashed with a
branch of the Kusanas, known as Kidara Kusanas, named after
their king Kidara, ruling in Tokharestan. The latter crossed the
Hindu Kush and settled down in the Kabul valley, after ousting
the later Kusanas from that region. The Hephthalites, in the
meantime, consolidated their position in Tokharestan and with
their conquest of Talekhar, Mervard, Herat and later on of
Gandhara extended their dominion further South-east. Subse
quently they crossed the Indus and invaded India. At that time
the Hun empire had reached its zenith, including the entire
steppe from the upper Yulduz (north of Karasahr) to the Aral
and Sogdiana, Merv, Eastern Iran, Afghanistan and the Punjab.
The role of the Central Asian peoples in the centuries follow
ing is equally notable. In the sixth century, the Huns were suc
ceeded by the Turks as overlords of Turkestan. The Turks,
closely affiliated to the Huns, both in race and language, proved
more dangerous to their neighbours than their predecessors.
For some time they made occasional inroads into Persian terri
tory, but finally the Seljuk Turks overran the whole of Persia and
were soon masters of all the countries in the Near East. The
11. For a succinct account o f the Hephthalite Huns, see—Bagchi :
India and Central Asia, Calcutta, 1955, pp. 9fF—henceforth—Bagchi—
Central Asia.
Turks are mentioned in the Chinese accounts as Tu-Kiu i.e.

Turkut, meaning ‘strong’, and according to Bagchi,12 they are
known in Indian literature as Turuska. They were descendants
of the ancient Hiung-nu race and were living towards the begin
ning of the sixth century A.D. in the region of Altai mountains
as dependents of the Juan-Juans. The internecine struggle among
the ruling class chieftains paved the way for the emergence of
the Turks to absolute power and alienation with the former
overlords. The Turks too were divided into Eastern and Western
groups.
Reference might as well be made to the Parthians andSogdians
and their interest in Buddhism and its expansion. While Par
thian relations with India could be traced to the Achaemenian
times, and Parthian satrapies were established in North-West
India (now Pakistan) before the Christian era, these did not
result in the inflow of Buddhist culture to Iran—the region occu
pied by the Parthians. It was the Tokharians who seem to have
transmitted Buddhism in that region and attracted Parthians to
the religion of the Tathagata. The Parthian monks studied Bud
dhism and the original Buddhist texts in the Buddhist centres of
Tokharestan, like Balkh before proceeding to China, where these
monks are distinguished by the prefix An (Ngan), named after
their country An-She (Arsak). A Parthian prince is said to have
appeared in the western country of China with a load of Bud
dhist texts in the year 148 A.D.—the date of the commencement
of the war and the downfall of the Arsacidan dynasty. The
contribution of the Parthian monks towards Buddhist textual
translations need not be assessed here, though the flow of such
monks from Parthia to China could be traced in the 3rd and 4th
centuries A.D.13
12. The role of the Turks is also assessed by Bagchi in the chapter on
‘Nomadic Movements in Central Asia’ in his book India and Central Asia,
Op. cit, pp. lOff).
13. A reference to the contributions by Parthian scholars is made by
Bagchi : Op. cit, pp. 36ff. The Parthian prince visiting China is known to
the Chinese as Ngan She-Kao or Lokottama. He abdicated the throne in
favour of his uncle and became a Buddhist monk at an early age. He left
for China and reached Lo-yang in A.D. 144 where he settled down in the
monastery of Po-ma-Sse or ‘The While Horse Monastery’ built for the
first two Indian monks, Dharmaratna and Kasyapa Matanga .
Introduction 13
Besides the Parthians, the Sogdians are also worth recording

in this context. Ancient Sogdiana was situated to the north of
Tokharestan with its centre at Samarkand. The earliest reference
to the country and its people could be traced to the Achaeme-
nian records and the Greek historian Herodotus. The Behistun
inscription describes them as Suguda, while the Greek historian
calls them Sogdoi with Xorasmioi, Areioi (Haraiva), Parthoi,
Gandarioi and Dadixai as their neighbours. The Achaemenian
inscriptions, however, mention Parthians, Arienas, Sogdians,
Xorazmians, Bactrians, Zarankes, Sakas and Gandarians. Sog
dians, primarily traders, had gone to different parts of Central
Asia and established colonies, and they were in contact with
Buddhism and its culture in other parts of Central Asia. Sogdian
monks also played their part in the transmission of Buddhist
culture to China,14 and were noted by the prefix K’ang, after the
ancient name of Sogdiana in Chinese Kang-Kiu.
The Tibetans and the Mongols are the two other important
people who effectively contributed towards Buddhism and its
expansion in Central Asia in the early and later medieval times
respectively. The Tibetans, a Mongoloid people, inhabited the
Tibetan plateau and lived in isolation for a long time, with noth
ing known about their history prior to the seventh century A.D.
It is generally accepted that Buddhism was first preached in Tibet
at the instance of King Srong-tsan-gam-po, who ascended the
throne in A.D. 629. The Tibetans, like the Chinese, had heard
something about it either from India or Khotan in Central Asia,
before they invited Indian preachers to visit them. According to
tradition the Tibetan ruler Thonmi Sambhota was sent to India
for evolving Tibetan script which was borrowed from the Indian
Brahmi script of the Gupta15 period. The reign of this ruler esta-
14. Some Sogdian monks actually belonged to the school of Ngan-
She-Kao. The earliest reference is (hat of the Sogdian collaborator with the
Parthian Ngan She-Kao, named Yen-Fo-tiao (Buddhadeva). Sogdians were
in close contact with India where they were known as Sulika. The Sogdian
monks equally contributed in the transmission of Buddhist culture to China.
Their names are distinguished in Chinese by the prefix Kang, as the ancient
name of Sogdiana in Chinese was Kang-kiu. (Bagchi : Op. cit, p. 39).
15. See Hocrnle : Manuscript Remains found in East Turkestan, 1916,
pp. xvii ff and Franke : Epigraphica Indica (El XI. pp. 266fF) and on the
other side Laufer in JAOS. 1918, pp. 34ff. According to Eliot, there is a
blished the foundations of a civilization conducive to the efflore

scence of Buddhism as also the unification of Central Tibet. The
monarchy also provided the necessary leadership during the period
of expansionism, with the Tibetans coming into conflict with
most of their neighbours, particularly China under the Tang
dynasty (A.D. 618-907). The Mahayana Buddhism brought
to Tibet during the seventh and eighth centuries, overladen with
a growth of Tantric occultism, soon absorbed elements of Bon
faith, resulting in a synthesis of the two with lofty metaphysical
speculation flourishing alongside with gross superstition. Padma-
sdmbhava summoned fromNalanda—the main object of venera
tion of the Nying-ma-pa or Red Hat Sect—gave to Tibetan Bud
dhism its distinctly Tantric character. The later history of Tibe
tan Buddhism and the contribution of Tibetans in the political
and cultural history of Central Asia need not be considered here,
and could receive exhaustive and fuller treatment later on.
The Mongols—said to be members of one small tribe living
south-east of Lake Baikal did not make any contribution to the
political and cultural history of Central Asia before the time of
Chingiz Khan in the twelfth century A.D. At the beginning of
this century three important tribes dominated the region, now
called Mongolia—the Tatars, the Karaits and the Naimans. Living
in close proximity to the Chinese frontier, they were most affected
by Chinese culture.16 Most of the inhabitants of Mongolia were
considerable difference between the printed and cursive forms of the Tibetan
alphabet. Is it possible that they have different origins and that the former
came from Bengal, the latter from Khotan. A number of Papers contri
buted by Sarat Chandra Das in the ‘Journal of the Asiatic Society of
Bengal’ between 1881-82 provide comprehensive information as also the
role of Indian scholars who were invited there for tne propagation of Bud
dhism in that country. (See Das : Indian Pandits in the Land o f Snow, Cal
cutta, 2nd edition, 1965).
16. According to Hambly, the influence of China on these tribes
whether direct or indirect, depended partly upon their respective military
strength at a particular time and partly upon the receptivity of different
tribes to the blandishments of Chinese civilization. It is suggested by him
that the Mongolian tribes probably acquired more knowledge of Chinese
civilization from the dynasties of nomadic origin—the Khitans or Liao
(947-1125) and the Jurchids or Kin (1122-1234) ruling over northern China,
than from the Chinese direct. In general, it was the tribes in closest proxi
Introduction 15
Shamanists, with the Shaman (boge) exercising a dominant role.

Tribal chieftains were called Khans and the ruler of a tribal con
federacy had the title of Khagan. It is also suggested17 that a
new pattern of social relationship was emerging—a sort of noma
dic feudalism in the twelfth century which provided the social
and military basis for Chingiz Khan’s conquests. It was from
Tibet that Buddhism reached Mongolia in the thirteenth century
under the ChingizKhanids18 who were eclectic in religious mat
ters. Chingiz Khan is said to have corresponded with the famous
Sakya Pandita (1182-1251), the abbot of the Sakya monastery,
which had been founded in the second half of the eleventh
century and was a great centre of learning. Tibetan-Mongolian
relations were established on the political and religious fronts
with the nomination of Sakya Pandita as viceroy of Tibet by
the Mongol prince Godan, an authority which was passed on to
his nephew Phagwa who exercised great influence over Qublai.
It was during this period that the Mongols first became
familiar with Tibetan Buddhism. Phagwa also provided the
Mongols with an alternative script to the Uighur earlier adopted
by Chingiz Khan.
The role of the peoples of Central Asia in the history and cul
ture of that region as also in that of the neighbouring ones has
been very conspicuous and effective. The Sakas and the Huns,
followed by the Turks closely affiliated with the latter in race and
language more or less shaped the destiny of the countries con
quered by them. The Turks proved more dangerous than their
mity to the Chinese frontier who were most affected by Chinese culture,
their chieftains proudly accepting such Chinese titles as Wang and Tai-tsi,
(Op. cit, pp. 86-87).
17. ibid. p. 88. The origin of this body of feudal lords seems to have
been the personal following of Chingiz Khan at the outset of his career, who
helped him in asserting his supremacy over neighbouring rivals. The
Mongol word for retinue—nokod—plural of nokor (a companion) is suppos
ed to carry distinct feudal and heroic overtones.
18. Hambly : Op. cit, pp. 246-247. Similarity between Tibetan and
Mongolian Lanviism is very close and there seems to be no difference bet
ween the two in deities, doctrines or observances. Mongolian Lamas imi
tate the usuages of Tibet, study these when they can, and recite their scrip
tures in Tibetan, though there are translations of these scriptures in their
own language. (Eliot : Op. cit , p. 401).
predecessors. They made occasional inroads into Persian territory,

but in 1040 the Seljuk Turksoverran the wholeof Persia and were
soon masters of all the countries in the Near East. Later on, with
the decay of their power, their place as overlords was taken by
the Mongols early in the thirteenth century, to be followed later
on by the Mongol-Turkish hordes led by Timur. The Huns who had
created havoc in Europe, with Bleda and Attila noted for their
empire as also their vandalism, like Toramana and Mihirakula
in Northern India more or less at the same time, were finally
absorbed in the local population. The Turanian Turks, a few
centuries later on their conversion to Islam, were not slow to in
vade India for booty and eventually for settlement. The Mughals—
more conspicuous than the Mongols—werejprominent among the
foreign Muslim dynasties who ruled in India. It may, thus, be
suggested that the Hunnish or Turanian invasions had permanent
impacts on the historical development of the other peoples of
Asia as also of Europe. According to Mc-Govern19, among the
most important factors in the shaping of the European world,
the background of all the stupendous events—the fall of the Roman
Empire, the pushing of the Germanic peoples into the west of
Europe, the introduction of the Slavic people into Central and
Southern Europe, the Renaissance and the revival of the ancient
classics in Western Europe, and finally the voyages leading to
the discovery of the New World, lurked the hordes of Central
Asia. In case of each of these events it can be shown that one of
the most, if not the most, important of the immediate condition
ing causes was the invasion and conquest of some portion of
Europe by peoples of Central Asiatic origin.
It must as well be understood that prior to the fall of Constan
tinople, there was an active trade between Europe on the one side
and China and India on the other, and the fluctuations in this
oriental trade had much to account for the rise and fall of several
of the great European cities. The Europeans could secure fine
silks from China, spices from the East and cotton prints and
precious stones from India. While the Greeks in the old days had
a virtual monopoly of the oriental trade, followed by the Romans
through their cities of Venice and Genoa, in the thirteenth cen
19. Op. clt, p. 11.

Introduction 17
tury, Italian merchants including the famous Marco Polo, had

made their way all over China and India, carrying tales about
the great wealth of these countries. The Turkish conquest of
Constantinople put a brake on the overland communication bet
ween Europe and the distant East, reducing movement of Asiatic
merchandise, once a mighty stream, to a negligible trickle.
Ancient Routes :20

While the role of the geography and peoples of Central Asia
in the political and cultural history of that region and the coun-
20. Detailed information about the routes is provided in Chinese
accounts. From very early times important trade routes passed through the
Tarim basin from the frontiers o f China to the west. These are also known
as the Silk-routes during the first few centuries o f the Christian era. Two
routes passed through the Tarim basin from the frontiers of China upto
Balkh. These routes were also used by the Buddhist monks and savants for
dissemination of Buddhist culture and thought to the States of Eastern
Turkestan and to China. Tun-huang, like Purusapura(Peshawar), situated on
the highway was a meeting place of foreigners, and had become a great
centre of Buddhist art and culture. The southern route from here passed by
the Gate o f Yang-Kuan, and proceeding westward it reached the country of
Shan-shan (to the south of Lob Nor). From there it went along the course
o f the river Tarim upto So-Kiu (Yarkand) and crossing the Pamir reached
the country of Yueh-chi (Balkh) and Parthia (Ngan-si). The northern route
passed by Kiue-she (Turfan), the ancient capital of the kingdom of Lou-lan.
Following the Tarim right up to the west to Shu-lei (Kashgar), it continued
across the Pamir (Kizil) upto Ta-wan (Fergana), Kang-Kiu(Sogdiana) and
other countries in the valley of the Oxus. (Bagchi : Op. cit, p. 16).
Fa-hien notices in detail the route to India with the principal localities
lying on it—from the province of Kanu-su westwards, Lan-chou, Leang-
chou, Kan-chou, Su-chou—and Tun-huang and Shan-shan to the south of
Lop-Nor. The outward journey lay through the countries of Yen-ki (Kara-
sahr), Yu-tien (Khotan) and Kie-cha(Kashgar). Passing by To-li (Darel in
Dardistan) and then crossing the mountains one reached the valley of Gilgit
finally leading to the region of the Indus. Song-yun, however, followed the
southern route upto the Pamir region—Tash-kurghan (Tsiu-mo) to Pa*ho
(Wakhan) and passing by Po-che (the mountainous region to the north o f
Chitral to She-ni, directing southwards to Udyana in the valley of Swat
and then to Gandhara (Peshawar).
Hsuan-tsang taking the northern route from Kan-su to Kao-chang (Yark-
hoto near Turfan), visited the countries of A-ki-ni (Karasahr), Kiu-che
(Kucha), Po-lo-kia(Yaka-aryk), to the south of the Tien-shan. Crossing it by
the Bedal pass, he passed by Sogdiana and crossing the ‘Iron Gates’, to the
tries in the periphery has been assessed rather summarily, it is

necessary to take into account the routes and the regions lying on
the international line or lines of communication connecting
China with India and the Near East, and finally the western
world. The ancient route started from the capital Chang-an (pre
sent Sian) in the province of Shensi and crossed the Gobi desert
to the oasis of Tun-huang where, approaching the Taklamakan
desert, it bifurcated into two : the northern route passed through
Hami, Turfan, Karashahr, Kucha, Aksu, Tumshuk and Kashgar
to Samarkand and the southern route traversed via Miran,
Cherchen, Keriya, Khotan and Yarkand to Herat and Kabul.
These overland routes opened since the Han times, passed through
Central Asia—called the Innermost Heart of Asia, receiving the
currents of life and civilization from different quarters, and also
serving as an intermediary through which different cross currents
passed. As pointed out earlier, the Tien-Shan or Celestial Moun
tains in the north and the Kun-lun ranges in the south, bounded
by the Nan-Shan in the east and the Pamir in the west provided
flow of the rivers—the Kashgaria and the Yarkand—powerful at
the source and gradually diminishing in volume in the proximity of
the desert of Taklamakan. It is along the basin of these rivers that
flourishing colonies of peoples from different directions had been
established. It was in the first century A.D. that Buddhism was
taken to these countries, and peoples from Kashmir and North-
West India proceeded to the region of Khotan and Kashgar and
set up small colonies* 21 with kings claiming descent from Indian
south of Kesch (Sahar-i-sabz) he reached the country of Tokharestan with

its capital Huo (Kunduz) to the south of the Oxus. He descended by the
pass of Bamiyan to the valley of Kapisa. On the return journey twenty
years later, the pilgrim followed the southern route from KapiSa after cross
ing the Hindu Kush by the valley of Panjshir, reaching Kunduz, Balakshan
(Pa-to-chuang-na), Yung-po-kien (Yamgan) and Hun-to-lo (Kandut). Cross
ing the Pamir, he visited the countries of Tash-Kurgan (Kie-pan-to), Ki-She
(Kasghar), Che-Kiu-kia (Karghalik) and Kiu-sa-tan-na (Khotan) from
where he followed the usual route by the south of Lop-nor to Chang-ngan,
the capital. (Bagchi : Op. cit, p. 17).
21. The colonies that flourished in the southern part of this region
were from west to east—Cokkuka in the region of modern Yarkand, Saila-
desa in that of Kashgar, Khotamna—Godana in the region of Khotan and
Calmadana in the region of Cherchen. The Kharo$lhi documents from
Introduction 19
royal families. The routes starting from North-West India in those

days passed by Hadda and Nagarahara (Jalalabad), and reached
Bamiyan before crossing the Hindu Kush. It had grown into an
important centre of Buddhist culture and was a halting place for
the Indian monks proceeding to Central Asia and China, and
also attracted traders and pilgrims from all quarters. Further to
the north of this region beyond the hill was Bactriana (modern
Balkh)—the Bahlika of the Indian and Fo-ho of the Chinese. Its
local culture had blending of Hellenic and Indian influence with
the Iranian one in its sub-stratum.
Buddhism is supposed to have been introduced in this region
in the first century A.D. or even earlier in the time of Demetrius
and Menander.22 The political changes in the country from the
Greeks to the Sakas, from the latter to the Yueh-chi-Kusanas,
and finally to the Hephthalite Huns, in no way affected the
allegiance of the people of this region towards Buddhism. In
the time of Hsuan-tsang, in the seventh century A.D., the great
Buddhist establishment of Balkh, known as the Nava-Samgha-
rama, was a great centre of Buddhist learning with a continuing
chain of commentators of the Buddhist canon. Bactria was, in
fact, the meeting place of two different roads leading to Central
Asia and China. The northern one passed through ancient Sog-
diana, crossing theJaxartes, reaching Tashkent and then through
the passes of Tien-Shan reached Uch-Turfan. The southern
route, rather the shorter one, and used mostly by the travellers
proceeding to China, passed through the country of the
Niya, Endere and Lou-lan mention Indian names with rulers assuming
Indian titles devaputra and the official documents commencing with the
formula : mahdniava maltdray a lihati. A comprehensive study of Indian
culture in Central Asia, based on these inscriptions is made in my book in
Hindi entitled Madhya Asia me Bharatiya Sanskriti (Lucknow, 1981). For
the study of the original documents, see Boyer, Rapson and Senart : Kharo-
sthi Inscriptions discovered by Sir Aurel Stein in Chinese Turkestan—Text
and Translation, Oxford, 1928.
22. See Bagchi paper on Krimisa and Demetrius published in Indian
Historical Quarterly; XXII, pp. 81 If, in which he refers to this Indo-Greek
ruler’s interest in Buddhism. As regards Menander, his general-successor,
the Milindapanho—discourse with Nagasena, the famous Buddhist scholar
is a conclusive proof of it. See Lamotte: Ilistoire dit Bouddhisme Ihdien—
Louvain, 1958, pp. 461-469.
Tokharians, the Tu-ho-lo of the Chinese, near Badakshan, and

over the difficult passes of the Pamirs reached the plains at
Kashgar. Here it was joined with a still shorter route passing
through the Gilgit and the Yasin valleys uptoTashkurgan.
The importance of Kashgar both from the point of view of
commercial activity and that of Buddhist expansion was great.
The place, because of its geographical location, provided relief
and hospitality to the travallers after an arduousjourney through
the hills. Numerous monasteries had also come up, and by the
middle of the seventh century A.D. the Chinese pilgrim Hsuen-
tsang could count these by the hundreds. From this place again
two routes extended upto the borders of China—one known as
the southern route went along the fringes of the Tarim basin
with a number of prosperous states growing up along this
passage. These included, beyond Kashgar, Yarkand, Khotan,
Niya and also a number of other sites like Dandanulik, Endere
and Miran. These centres played an important role in the de
velopment of cultural relations between China and the western
countries, and were equally notable for Buddhist monasteries as
centres of learning, revealing deep Buddhist influence on local
culture. This whole region primarily constituted an Iranian zone,
as the documents found by the explorers and archaeologists in
this area are in Kharosthi. Khotan-figures prominently in ancient
records23 and was known to the Chinese writers as Yu-tien,
colonised in the time of Asoka with the blinded prince Kuiiala
being set up as the ruler of this newly founded kingdom. The
Gomati vihara here—the premier Buddhist establishment—was
noted for its learned savants who also wrote canonical texts,
thus contributing to the development of Buddhist literature.
23. Khotan (Sanskrit and Prakrit Kustana or Kustanaka; Khotamina,
Khodana or Khotana; Chinese Chien-tun, Chu-sa-tan-na, Chu-tan, Ho-tien,
Huan-na, Huo-tan, Yo-tien and Yu-tun, Manchu—Ho-thian, Mongol—Hu-
t ’an, O-duan, Wa-duan, Wu-duan; Tibetan Li-yul, U-then [Ho-then]) has a
long history. The information from the Chinese Annals on the history of this
place confirms the Tibetan account. A local dynasty which used the title of
Vijaya had been ruling in Khotan since very early times. This dynasty was
of Indian origin, as is evident from the names of its members. Buddhism was
introduced in Khotan during the reign of king Vijaya Sarhbhava, grandson
of Kustana, its founder. (For a comprehensive account of Khotan—in all
aspects—see Stein : Ancient Khotan, Oxford, 1907.)
Introduction 21
The countries along the northern route from Kashgar to the

Chinese frontier were equally important from the commercial
as well as the cultural viewpoints. These centres disseminated
India culture and Buddhism in Central Asia as well as in China.
The notable ones were Bharuka—Chinese Po-lu-Kia near Uch-
Turfan, Kuchi or Kucha—Chinese Kiu-tse, modern Kuchar,
Agnidesa, Chinese Yen-ki, modern Karasahr and Turfan—
Chinese Kao-chang. The people and language of the countries
or centres on the northern routes were different from those of
the south; but both had common affinities in the form of Bud
dhist religion and culture. Kucha or Kuchi like its counterpart
Khotan on the southern route was the most important24 kingdom
and its ancient rulers had Indian names like Suvarnapuspa,
Haradeva, Suvarnadeva etc. The Buddhist monks of this place
were well-conversant with Sanskrit, as is confirmed by the finds
of manuscripts in Sanskrit and bilingual documents in Kuchean
Sanskrit. In the words of the Chinese pilgrim, it was not only a
centre of Buddhist studies but it also catered towards the ex
pansion of Buddhism in China through its famous teachers and
savants, who translated Buddhist texts as also gave discourses.
The noted Buddhist scholar Kumarajiva25 belonged to this place.
24. Kucha has all along been one of the most important territories in
the Tarim basin, in many ways, according to Stein, ‘a worthy pendant of
Khotan owing to its geographical position and the role it has played in
Buddhist art and civilization’. The Chinese annals provide exhaustive inform
ation about this place which was utilized by S. Levi in connection with his
paper entitled Le Tokharian B. Langue de Kouteha, J.A. 1913, Sept. Oct.,
pp. 323-80. See Stein : Serindia (Oxford, 1921, p. 1238). The name of
Kuchean king of the Tang period had been Sanskritised. According to Bagchi,
from the old Kuchean documents we get the name of king Suarnage (Suvarna-
datta), that of his father So-fa-pu-kiue (Svarna), and brother Ho-li-pu-she-pi
(Haripuspa). The Chinese pilgrim mentions a former ruler of Kucha named
‘Gold Flower’ (Suvarpa-pu§pa)—(Bagchi : Op. cit, p. 79).
25. Kumarajiva’s life history is important in several respects. His father
came from India and he himself went as a youth to study in Kipin (Kashmir)
and then returned to Kucha, his mother Jiva’s home. Living in this remote
corner of Central Asia he was recognised as an encyclopaedia of Indian
learning including a knowledge of the Vedas and ‘heretical Sastras’. After
his return to Kucha he was converted to Mahayanism. It was from Kucha
that he was taken to China where he had a distinguished career as a translator,
and brought China into intellectual touch with India. (For an account of
Kumarajiva see Hastings : Encyclopaedia o f Religion and Ethics, Edinburgh,
The grottos in the hills in the neighbourhood of Kuchar met

the requirements of Buddhist monks for meditation and quiet
stay, like those at Bamiyan in Afghanistan, and the ones in the
western Ghats in India. Agnidesa*26—in the region ofKarasahr—
the next place on the northern route was linked with Kuchi-
Kucha in culture, race and language, and common Buddhist
religion. The finds of literary texts and art objects in this region
testify to the active role of this place and its scholars in the
propagation of Buddhism. Turfan, the next stage on the nor
thern route further towards the east, was a centre of Central
Asian as well as of Chinese cultural impacts. It has preserved
relics ofBuddhist civilization and contributed in full towards the
propagation of Buddhism in China.
The Northern as well as the Southern routes met on the
Chinese frontier at a place called Yu-men-Kuan or the ‘Jade
Gate’27, close to the famous Tun-huang—noted for its thousand
grottos—one of the famous centres of Buddhist learning. In
the hills nearby were the caves carved out between the fifth and
the eighth centuries A.D. for the use of Buddhist monks pro
ceeding to China; and these also served as the meeting place of
Buddhist scholars coming from various countries—Persia,
Bactria, India, Sogdiana, Khotan, Kucha and other places—for
holding discussions and translating sacred text into Chinese.
Situated on the cross road of the two main highways, Tun-
huang had also its strategic importance, besides cultural and
commercial one. From the first century of the Christian era
1926, Vol. VII, p. 701a; Eliot : Hinduism and Buddhism, III, p. 203ff; S.C.
Das : Indian Pandits in the Land o f Snow : Op. cit, pp. 33 ff.)
26. AgnideSa or Karasahr, according to Bagchi, was closely connected
with Kucha. The Chinese accounts contain some references to the political
condition of Karasahr from the Han times to the end of the Tang period.
Its political history and relations with China are traced by Bagchi (Op. cit,
pp. 75 ff).
27. The location of the Jade Gate barrier is discussed by Aurel Stein
in Chapter XIX of his Serindia. Its history is also traced. It is said to have
been established in c. 96 B.C. With a rapid increase in the diplomatic relations
of China with the west and in trade also, it was necessary to safeguard the
passage of envoys and of caravans, and to assure supplies for them en route.
‘It is stated that military posts were established from place to place from
Tun-huang westwards to the Salt Marsh’ (Serindia, p. 728).
Introduction 23
onwards, its importance28 increased and it also attracted forei

gners. While the Buddhist pilgrims from the second century
onwards sought shelter here on their way to China, some Indian
families settled down here in the third century A.D. and the
place hubbed with activities—secular and religious. The finds of
a large number of manuscripts by the archaeologists testify to
the importance of Tun-huang as an active centre of learning.
These manuscripts are in several languages—Chinese, Sanskrit,
Tibetan, Syriac, Khotanese etc. suggesting that it was a meeting
place of Buddhist scholars from China, India and the Near
East.
The trade routes in Central Asia and the peoples living at
important centres were equally instrumental in carrying cultural
traits from one end to another. The Tarim basin people were in
touch with Bactria and the regions conquered by Alexander and
through them with western art and thought. Its inhabitants
included not only Iranian tribes but also speakers of a language
which is classed as Indo-Aryan. From the dawn of history down
to the middle ages, warlike nomads were continually passing
through this region of Central Asia. The Sakas, Yueh-chi and
the Huns of the ancient period and the Turks and Mongols of
the Medieval one had the same peculiarity. They communicated
and transported ideas and cultural features from one region to
the other. The peoples receiving these were equally interested
and receptive. The finds of numerous manuscripts in different
languages suggest the cosmopolitan nature of the peoples of
different parts of Central Asia. It might be interesting to note
that 20,000 manuscripts were discovered in Tun-huang29 alone,
where they had been walled up for 900 years for protection
against invaders. These are in Brahrai. Kharostbl, Persian,
Tibetan, Turki, Uighur and Tokharian scripts and languages,
28. Tun-huang was famous among the four commands of Ho-hsi or
western Kansu, side by side with Liang-chou, Kan-chou and Su-chou. It
derived its importance from the great advantages which its geographical
position and resources offered, and which, according to Stein, are easy to
recognize even now when the line of the great Central-Asian route finally
shifted northward. The Chinese realized the value which Tun-huang possessed
for them at the time of their first advance into the Tarim basin. (Serindia,
p. 581.)
29. Eliot. III. op. cit, p. 189.
and point to the cosmopolitan community of peoples that lived

in these Central Asian kingdoms. The art of Central Asia, as
depicted in the frescoes at Tun-huang and other places of
Central Asia is expressive of the unique synthesis of religions and
cultures of peoples. The languages recorded in the manuscripts
and inscriptions are numerous, pointing to the number of
tongues simultaneously in use for popular or learned discourses.
Besides the great polyglot libraries like Tun-huang, even a small
one at Toyug30 contained Indian, Manichaean, Syriac, Sogdian,
Uighur and Chinese books. The writing materials include im
ported palm leaves, leather and paper, the last one being in use
from the first century A.D. onwards. These survived through
the ages because of the dry atmosphere.
Buddhist Finds—Literary Texts and Monuments3132

Besides numerous Sanskrit writings dealing with religious or
quasi-religious subjects like medicine and grammar, compara
tively modern Mahayanist literature is traced in abundance.
Portions of lost Sanskrit canons, corresponding to the Pali
ones, provide the original text for translation into Chinese.
Portions of a Sanskrit grammar have also been found near
Turfan, thereby suggesting that Sanskrit was probably under
stood in polite and learned society. According to the testimony
of Fa-hien, the monks of Central Asia were all students of the
language of India, and the same point was recorded about the
Buddhist monks of Kucha by Hsuang-tsang. Some palm leaves
from Mingoi contain fragmentsof two Buddhist religious dramas,
one being the Sdriputraprakarana32 of Asvaghosa, written in
30. Eliot : Op. cit, III. p. 189. At Toyuk Stein picked up a considerable
number of torn fragments from Chinese Buddhist SQtra rolls in a debris-
strewn ravine where they had been thrown out from shrines above in the
course of some previous exploration. (Serindia, 1167n).
31. A comprehensive study of the manuscript remains from Central
Asian sites as also remnants of Buddhist monuments in this region is reserved
for separate treatment in two to three chapters in this work. Here only a casual
reference might be made in the context of importance of such from religious
and literary viewpoints and the achievements of different missions in this
vast region.
32. Remains of the works of the two great Buddhist poets Asvagho§a
and Matriceta were discovered amongst the Central Asian finds. Fragments
Introduction 25
the script of the time of Kaniska. It is the oldest known Sanskrit

manuscript as well as the oldest specimen of Indian dramatic
art. Besides these, the Prakrit version of the Dhammapada in
Kharosthi characters discovered by Dutreuil de Rhines mission
near Khotan33 with numerous documents in this language and
alphabet are more or less of the same time as the period of the
Kusanas in India.
Central Asia is also supposed to provide two new languages
written in a special variety of the Brahmi script called Nordarisch,
probably the language of the Sakas or of the Kusanas. Its basis
is Iranian but it is strongly influenced by Indian idioms.34 It is
supposed to have been spoken principally in the southern part
of the Tarim basin, and many translations of Mahayanist works
such as the Suvarnaprabhasa, Vajracchedikd and Aparimitdyus-
Sutras were made into it35. The other new language was that of
the Tokharas or Indo-Scythians, spoken primarily on its northern
edge, and was called the language of the Kucha or Kuchanese.
It is supposed to exist in two different dialects whose geogra
phical distribution is uncertain, but numerous official documents
dated in the first half of the seventh century show that it was
the spoken language of Kucha and Turfan. It was also a literary
language and the translations discovered in this one include
those of the Dhammapada and Vinaya. This language spoken by
the early and perhaps original people of Kucha is supposed to
belong to the Aryan family, closely related to the western than
the eastern branches. It is classed in the Indo-Iranian group.
Besides the Nordarish language mentioned above, written in
Brahmi, remnants of three other Iranian languages are traced
of the works of the former were brought by the German mission from the
Turfan region. This drama—not traceable elsewhere—was discovered in the
Turfan region only in fragments. The work is in Sanskrit but there are dialogues
in Prakrit, older than those used in Sanskrit dramas. (See Keith : Sanskrit
Drama, Oxford, 1924, pp. 80 ff.)
33. Eliot : Op. cit, p. 190. The Central Asian finds include the Dhamma
pada in Prakrit and the Udanavarga in Sanskrit, the former from the region
of Khotan and written in Kharosthi script of about the 3rd century. See
Senart : Les ms Kharotfhi du Dhammapada JA. 1898 II, p. 193.
34. Eliot : Op. cit, p. 190. See Luders : ‘Die Sakas und Die Nordarische
Spraehe'—quoted by Eliot; see also Bagchi : Op. cit, pp. 98 ff.
35. Hoernle : JRAS, 1910, pp. 837 ff and 1283 ff.
in Central Asia. These are written in an alphabet of Aramaic

origin.36 Two of them apparently represent the speech of the
south-western Persia under the Sassanids, and of north-western
Persia under the same rulers. Manichaean texts are preserved in
both these languages. Sogdian—the third language—has a more
varied literary content offering Buddhist, Manichaean and
Christian texts. Originally the language of the region round
Samarkand, it acquired an international character being used
by the merchants throughout the Tarim basin, and it spread
even to China.
Buddhist literature from Central Asia is not confined to
languages and scripts noticed above. It is available in other
alphabets and dialects as well. A Turkish dialect written in the
Uighur alphabet, derived from the Syrian, was, like Sogdian,
extensively used for Buddhist, Manichaean and Christian litera
ture. Uighur37 represented the literary form of the various
Turkish idioms spoken north and south of the Tien-Shan. It
was used considerably for Buddhist literature when the Uighur
supplanted the Tibetan power in the Tarim basin about A.D.
860 and founded a kingdom which extended upto China and
lasted for quite sometime. It is suggested by Eliot that Sutras
in Uighur were printed at Peking in 1330, and Uighur manu
scripts of the time of Kang-Hsi (1662-1723) were reported from
a monastery near Suchow.
Besides these languages, the Tibetans who ruled in Central
Asia in the Tarim basin from the middle of the eighth until the
middle of the ninth century, also enriched Buddhist literature
with their language. A large number of Tibetan manuscripts
were found in the region of Khotan, Miran and Tun-huang. The
Tibetan influence, however, seems less conspicuous in Turfan
to the north. The documents discovered comprise many official
and business papers as also Buddhist translations.38 They are
considerably important for the history of the Tibetan language
36. Eliot : Op. cit, p. 191.

37. Eliot : Op. cit, p. 192.
38. Ibid, p. 193. The list of manuscripts in Sanskrit, Khotanese and
Kuchean prepared by A.F. Rudof Hoernle, and a note on Stein’s collection
of Tibetan documents from Chinese Turkestan figure as Appendix F and G in
Serindia, pp. 1432-1459.
Introduction 27
as well as Buddhism. There have not been any Tibetan trans

lations of Manichaean or Christian texts.
The role of Central Asia and its intercourse with China is also
evident from a large number of Chinese texts—both religious
and secular. Some documents of the Tang dynasty are Mani
chaean with an admixture of Buddhist and Taoist ideas. A
series of dated documents ranging from 98 B.C. —A.D. 133
have also been recovered from the old military frontier near
Tun-huang. The Buddhist monuments of Central Asia com
prising of stupas, caves and covered buildings used as temples
or viharas point to the propagation and prosperity of the
religion of the Tathagata in this area. Representations of Hindu
deities no doubt have been found, but Hinduism is not supposed
to have existed there as an independent separate religious unit
but only as a part of or rather a contributory factor to the
religious ethos. The general scheme and style of the caves in
the Tarim basin as also at Tun-huang on the frontier of China
are similar to those in India. The ornamented figures, as sculp
tural pieces, add to the beauty and grandeur of the caves.
These are in stucco. The walls at Tun-huang and Tarim have
also fresco paintings reminding one of Ajanta frescoes. Temples
and caves were sometimes combined, as for instance at Bazaklik
many edifices were erected bn a terrace in front of a series of
caves in a mountain corner. The commonest type of temples was
a hall with a cellar at its further end, with a passage behind
for circumambulation. Occasionally side rooms were added to
the halls.
Some buildings had also stupas nearby, either in combination
or independently, the bestpreserved being the stupa of Rawak.39
It is set in a quadrangle bounded by a wall which was orna
mented on both its inner and outer face by a series of gigantic
statues in coloured stucco. A rectangular base provides the
setting of the dome. It is disposed of in three stories and is said
to characterize all the stupas of Turkestan as well as those of
the Kabul valley and the nearby regions. In its architectural
setting, the dome provides a synthesis of Indian (Gandhara)
and Persian elements. There are, however, some innovations
39. Stein : Ancient Khotan. Op. cit, p. 483 fT.

here, as for instance, some of the caves at Ming-oi have dome

like roofs ornamented with a pattern composed of squares with
in squares, set at an angle with each other.40
The frescoes on the walls of caves and buildings and paintings
on silk paper provide evidence of other antiquities related to
Buddhism in Central Asia. The influence of Gandhara is notice
able in architecture, sculpture as well as painting. The oldest
works are described simply as Gandharan, but later on there is
some development both in technique and in mythology, repre
senting Indian Buddhist art as modified by local painters and
sculptors. Thus, the frescoes in Turfan are Indian in drapery
and composition, but the faces are eastern Asiatic, sometimes
representing people with red hair and blue eyes. These testify to
the influence of Eastern art on the ideas and designs of Indian
Buddhism.41 Persian influence is also noticed in many paintings.
Aurel Stein refers to the Bodhisattva figures from Khotan—one
being of the familiar Indian type, and the other suggesting at
first sight a miniature of some Persian prince, black bearded
and putting on high boots. Currents of art and civilization
flowing from neighbouring and distant regions seem to have met
and mingled here, providing a cosmopolitan nature of Buddhist
art in Central Asia.42 The Iranian link in the chain connecting
the Greco-Buddhist art of the extreme North-West India with
the Buddhist art of Central Asia and the Far East, is recorded
by Stein on the basis of the finds of a Buddhist monastery in
the terminal marshes of the Persian province of Seistan. The
Indian Buddhist element dominates everywhere.
This introduction should provide a background for a proper
study of Buddhism in Central Asia. The geographical factors
with the nature of the peoples of different regions—nomadic as
40. Eliot : Op. cit, p. 194, quoting Grunwedel : Buddh Kulstatten,
pp. 129-130 and plate and Foucher : L'Art Greco-Bouddhique, p. 145.
41. Eliot : Op. cit, p. 195. According to Bussagli, in the first stylistic
phase of painting at Turfan, there was a slow decline of the classical Iranian
and Indian characteristics which had penetrated into the region and been
re-elaborated in the Kuchean manner. These gave way before Chinese in
fluence, which sometimes managed to modify even the common motifs of
Buddhist iconography. (Paintings o f Central Asia, Geneva, 1963)
42. Eliot : Op. cit, p. 195, quoting Ancient Khotan—Op. cit, Vol. II
plates lx and lxi.
Introduction 29
well as pastoral—receptive to external influences but equally

dominating and aggressive—present a picture of external chal
lenges and internal responses. In this context, the international
trade and its routes facilitated the movement of ideas from one
part to another. In the process of cultural communication be
tween the Orient and the Occident, the role of Central Asia was
significant and equally substantial. The process of introduction
of Buddhism in Central Asia was the work of missionaries,
savants, merchants and traders and banished princes. The Chi
nese sources, however, point to royal invitations extended to
Indian Buddhist scholars to visit their country and propagate
the message of the Tathagata. These people were from the Bud
dhist religious establishments in Central Asia itself, or from the
country of the Sakyamuni and passed through Central Asia. Tra
dition, no doubt, connects Buddhism in Khotan with Kunala,
the banished Mauryan prince from Taxila in the last years of
Asoka, or might be even a little earlier. A number of Indian
colonies had come up in Central Asia as recorded by the Chinese
pilgrims. While the earlier picture of Buddhism and expansion of
Indian culture in Central Asia might not be very distinct and
sharp, that of the later times is more pronounced, vivid and clear,
as is evident from the finds of Buddhist manuscripts as also
monuments and sculptures. Patronage to the Buddhist scholars
was extended by the local rulers whose historical account no
doubt demands some study.
CHAPTER II
EARLY HISTORY OF CENTRAL ASIA
The early history of Central Asia is closely related to the

movement of its peoples sharply divided and concentrated in two
regions, named after their traits—the nomadic and the sedentary.
The former comprises North Steppes from South Russia to Man
churia, while the region to the south includes the oases of Tur
kestan occupied by the sedentary peoples. The nomadic region
itself is further divided into two : one from South Russia to the
valley of the Yenisei occupied in the past by the nomadic hordes
of Aryan stock called Scythians, the other one in the east includ
ing outer and Inner Mongolia, Manchuria and further east
occupied by Turco-Mongol hordes. In the south the land between
the Jaxartes (Syr) and the Oxus (Amu) was occupied by the
peoples of Iranian origin, while the Eastern Turkestan from the
Pamirs up to the frontiers of China was occupied by the Aryan
speaking people of different affiliations. The movements of the
nomads of the steppes either on the west or the east affected the
sedentary life in the south through the centuries. It is really the
history of this southern region with the movements of different
peoples affected by the onslaught of foreigners or the nomads of
Central Asian origin that needs to be told on the basis of the
available evidence, however scanty, whether archaeological or
accounts of travellers or second-hand source material.
It is in the seventh century B.C. that history opens in the
steppes of the Caspian sea, with a large-scale nomad migration
in progress. The powerful-Massagatae Confederacy had pushed
westwards across the Volga the peoples who became later famous
\as the Scythians—a term given to various tribes of Indo-European
origin, speaking either Iranian or other Indo-European dialects.
They are supposed to be the nomadic section of the Iranians
and others left behind in the north steppes who had refused to
adopt the culture of the sedentary conquerors. They are known
Early History o f Central Asia 31
in Persia as Saka and in India as Saka.1 TheSakas in old Persian

inscription2 include three different tribes of the Scythians: the
Saka Haumvarka—the Sakas of Persian and Indian literature
settled in the region of Fergana and extended up to Kashgar;
the Saka Tigrakhaude—or Sakas with pointed caps who spread
towards the Aral and occupied the lower valley of the Jaxartes,
and &aka Taradraya or Scythians who lived beyond the sea in
south Russia. Besides the Sakas, other tribes noticed in old
Achaemenian inscriptions and/or in Herodotus’ Historica are
Sarmatians, probably, neighbours of the Scythians of South
Russia, Massagetes from Mausyagata—fishermen speaking Ira-
1. The Sakas are frequently mentioned in Indian literature. They are
spoken of as belonging to the barbarous peoples who will rule in the Kaliyuga,
or as degraded K$atriyas. They are mentioned together along with other
north-western peoples, such as Kambojas and Yavanas. The word Saka is
the same name which is handed down in Iranian and classical sources as Saka 7
Sai or Sak of Chinese annals. There is a general consensus of opinion to the
effect that the Sakas were foreigners and Iranians. The history of the Sakas
can be reconstructed on the basis of classical, Iranian and Chinese sources.
The stray references found in classical literature give us some idea about the
homeland of the Sakai. Herodotus mentions them together with the Bactrians,
(1.153; VII-64; IX. 113) and with India (VII.9; III.93) as forming the
fifteenth nomos of the Persian empire together with the Kaspioi. Arrian
connects them with Bactrians and Sogdians (Anabasis, III.8.3; VII-10.5)
while Strabo (XI.8.2) and Pliny locate them beyond the Jaxartes. Strabo
further states that the ancient historiographers of the Hellenes called the tribes
beyond the Caspian Sea partly Sakai, partly Massagetai. They were mostly
nomads and had spread over a large territory. Sten Konow suggests that the
old home of the Sakai was in the Pamir country to the north of the Hindu
Kush and east of Bactria, and that Saka tribes were further considered to
exist to the east of the Caspian Sea and beyond the Jaxarte. (Corpus 1ns-
criptionum Indicarum—henceforth CII.II(l) Calcutta, 1929, pp. xvi IT.)
2. In the Behistun inscription 1.6 Darius mentions the countries which
he inherited, and Saka here comes between Bactria, Sogdiana and Gandhara
on on>, side, and Thatagush, Arachosia and the Makas on the other. In ii.2
Saka is mentioned among the provinces which revolted while Darius was
in Babylon, after Parthia, Margiana and the Thatagush. They are also
mentioned in the Persepolis Inscription, as also in the Naksh-i-Rustam one,
which mentions some individual Saka tribes : the Tigra Khauda, the
Haumvarka and those beyond the sea (taradraya or paradraya) (ibid, p.xviii).
In Darius’s Suez inscription the hieroglyphic text renders the first as ‘Sacae
of the marshes’ (presumably those on the shore of the Aral sea) and the second
as “Sacae of the plains’ (G. Posencr, La premiere domination perse en Egypte,
Cairo, 1936, p. 183, quoted by Hambly : Central Asia, p. 25 and note).
nian language who lived near the Aral, Arinaspes and Issidones
—who lived in the east in the northern steppes—the former pro
bably Iranian and lover of horses, and the latter of a different
race.
The old Persian inscriptions provide information relating to
the Achaemenian domination over Central Asia and its states,
while Herodotus notices their annual assessments of tribute to the
Persian treasury.3 The Achaemenian control over these provinces
was no doubt fully effective during the reign of Darius I. His
Susa building inscription records that gold for the work on the
palace was obtained from Bactria, lapis-lazuli and carnelian
from Sogdiana, and turquoise from Chorasmia. Ivory came from
India and Ethiopia and also from the province of Arachosia.
The Achaemenians are said to be the intermediaries who trans
mitted the irrigation techniques of Babylonian civilization to
Central Asia. During the reign of Xerxes (486-465 B.C.) the
Central Asian contingent formed part of the Persian army invad
ing Greece in 480 B.C. The Bactrians and Assyrian Sacae (old
Persian Haumvarga) were under the command of Hystaspes,
son of king Darius and Queen Atossa.4 There is probably no
evidence regarding the nomadic tribe living beyond the Achae-
mcnian northern frontiers.
A significant role of Central Asia and its peoples, however,
comes in view from 330 B.C. when Alexander passing through the
Caspian Gates expanded Greek rule across the Trans-Oxus
region in the province of Sogdiana and Bactria. There was not
much of resistance and the Macadonian ruler founded a number
of cities named after him5 in this region, which remained key-
3. Herodotus (III.91 ff) quotes the annual assessments of tributes to
the Persian treasury which varied between 170 and 600 talents. That to be
paid by the Sacae was 250. The tribute was paid in Silver bullion. (Hambly :
Op. city p. 22; 317 n. 15)
4. Herodotus. VII. 66. There are references to other provinces and the
contingents provided by them, as also their commanders. (Quoted by Hambly:
Op. cit, p. 24)
5. The compaign of Alexander is also noted for the foundation of cities
which became famous later on in history. These were named after him and
include Alexandria in Ariana, modern Herat; Alexandria Prophthasia in
Drangiana (location uncertain); Alexandria in Arachosia (placed by Tarn
in Ghazni -Greeks in Bactria and India—henceforth Tarn, Cambridge, 1938,
points in Asia for centuries. These were strongly garrisoned for

keeping a firm hold on the land routes. Internal dissentions,
army revolts and finally Alexander’s death left his eastern empire
in a state of dissipation. Seleucus, however, attempted the reuni
fication of his Master’s eastern provinces with Bactria and Pra-
thia forming part of his empire, only to become independent
later on in the time of Antiochos III, under Diodotus and
Arsaces respectively.8 There is, however, no ancient narrative
about the tale of this Greek kingdom-of Bactria—that of Euthy-
demus and his successors, Demetrius and Menander, and later on
of Eucratides and his family, subsequently followed by the emer
gence of Antialkidas whose ambassador Heliocles recorded his
dedication to the Indian god Visnu at Besnagar (Vidisa) in Madhya
Pradesh, India.
While the Bactrians and Indo-Parthians were involved in
struggles, storm clouds, gathering along the Central Asian
frontier of Jaxartes, posed threats to these political kingdoms.
New pressures in this river’s steppe were the outcome of fast
eastern nomadic unrests along the frontiers of China consequent
to the pressure from the warring tribe of Hiung-nu. They had
reached the height of their power in Mongolia in the third
century B.C. and posed great threat to the rulers of northern
China. The great Wall*67 was constructed to ward off their attacks.
p. 470); Alexandria of the Caucasus probably located on the site of the

medieval city of Parvan, at Jebel Suraj on the Salang; and the short-lived
Alexandria on the Jaxartes (Hambly: Op. cit, pp. 207-29).
6. Bactria was the rich country between the Hindu Kush and the Oxus,
corresponding in large measure to Northern Afghanistan. Beyond it, between
the Oxus and the Jaxartes (Syr Darya) lay Sogdiana (Bukhara). Parthia, an
ancient land, corresponding roughly to the modern province of Khurasan
in Iran, became independent under Arsaces. (For a detailed history of these
two kingdoms—see Cambridge History o f India—CHI, Vol. I, Chapter XVII,
pp. 427 ff.)
7. In 214 B.C. the Emperor Shih Huang-ti of the Chin dynasty linked up
the defensive lines by which the feudal kingdoms of the North had endeavoured
to protect themselves against the inroads of the Hiung-nu or Huns and thereby
first created the famous ‘Great Wall’. At first it extended from Shan-hai Kuan,
on the Gulf of Liao-tung, westwards as far as Lin-tas, corresponding to the
present prefecture of Min, in the extreme south of Kan-su and about 110
miles south of Lan-chou. It was not until a century later that the ‘Great Wall’
The clash between the Hiung-nu and the Yueh-chi, pastoral

nomads led to the movement of the latter down westwards,
clashing with the Wu-sun, another nomadic tribe. In the final
confrontation between the Yueh-chi and Wu-sun, aided and
supported by the traditional warring Hiung-nu, the former lost
ground and were driven further west on the heels of the Sacae
(Sakas) whom they had earlier dislodged from their home near
Issyk-kul. These two powerful hordes—the Sacae and the Yueh-
chi were poised against ~the Greco-Bactrian frontier of the
Jaxartes. According to Strabo,8 the nomads who became the
most famous were those who took away Bactria from the
Greeks—the Asii or Asiani, the Tochari, and the Sacaraucae,
who set out from the far bank of the Jaxartes, adjoining the
Sacae and the Sogdiani which the Sacae had occupied. The
Prolugues9 of Pompeius Trogus also refers to the seizure of
Bactria and Sogdiana by the Scythian tribes of the Saraucae.
The Asiani, according to the next Prologue, became kings of
the Tochari and the Saraucae (Sacaraucas) were destroyed. The
displaced nomad groups soon after-wards overran the Greco-
Bactrian kingdoms. The invasion, according to Tarn10, took
place after 141 B.C. A wave of nomad invaders is said to have
burst into Parthia in 129 B.C. It is also proposed that the
region of the Helmand Lake (now called Hamun) ceased to be
known as Drangiana and came to be called Sakastan (Seistan).
was extended to the north-west, while earlier this wall catered for consoli
dated defence; later on its purpose was offensive aiming at expansion into
Central Asia (Stein : Seriudia, Vol. II, pp. 722-23).
8. Strabo mentions a Saka conquest of Bactria (XI ■8,4) where the Greek
kings were ousted by Scythian nomads, and some of these nomadic tribes
are enumerated by him, notably, the Asioi, Pasianoi, Tocharoi and Sakarauloi
(cf. XI.8.2). Sakas were thus instrumental in overthrowing the Greek empire
in Bactria and some of these seem to be called Sakarauloi (cf. Ptolemy Saga-
raukai, VI.14.14) who could be identified with the Sai-wang. According to
F.W. Thomas, the Saraucae or Sacaraucae started from the country north
of Parthia and between the Caspian and the Aral Sea (JRAS, 1906, p. 186).
9. Trogus deals with the establishment of an empire in Bactria by Dio-
dotus in the 41st book, and this event took place about the middle of the
third century B.C. (Sten Konow : Op. cit, p. xxi).
10. Greco-Bactrian Rule in India, 2nd Edition, Op. cit, p. 280. See also
Puri : India under the Kushanas, Bombay, 1965, pp. 5 ff and notes pp. 10 ff.
It is concluded that the §akas during the first century B.C.

passing through the Herat gap, established themselves in the
former Drangianaand, continuing north-eastwards into Arachosia
and reaching Indus, followed upstream until they reached
Taxila on the one hand, and Saurastra and Ujjain on the other
hand.
The first century B.C. is regarded as the epoch of the £aka
empire in India and Arachosia. The extreme mobility of the
Saka forces seems to have been an important factor in their
overrunning such a wide region. The history of the Sakas in the
North-west seems to have begun with Maues,11 figuring at
Taxila as the sovereign ruler at the expense of the Indo-Greeks.
An analysis of the coins and other finds, no doubt, suggests fluc
tuating Saka-Indo-Greek fortunes with the former under Maues
being ousted; and the Indo-Greeks finally settled in the Kabul
valley and Gandhara region. Later on, Azes I seems to have put
an end to the Greek dynasties and asserted his paramountcy;
and probably started an era of his own. The Sakas in North- _
West India and Western India seem to have played an important
role in political history and also made their contributions in the
cultural ethos of the country of their adoption.
The other participants of the great migratory movement in
Central Asia were the Tochari and the Asiani who were living
on the north bank of the Oxus, and to the east of the Saka
advance line. The advance of the Tochari is suggested by the
association of their name with the area called Tukharistan
centring round Qunduz and Baghlan, close to the upper Oxus.
Ptolemy in his Geography calls them a ‘great tribe’ with their
advance guards pushing southwards towards the Hindukush
passes. The identification of Asiani has been a matter of dispute
11. It is proposed by A.K. Narain that Maues led a separate Saka group
directly to Taxila from the north, passing from Khotan over the Pamirs and
Indus Kohistan—an arduous route called that of the ‘Hanging Pass’ by
Chinese chronicles ( The Indo-Greeks, Oxford, p. 136). Hambly considers
this suggestion fantastic. He proposes that Maues began his career as a
commander of Saca mercenaries in the service of the late Indo-Greek kings.
With the Indo-Greek princes divided and Saca invaders at the gates, such a
personage would be well placed to assume sovereign power (Op. cit, p. 41).
According to Sten-Konow, he was not the ‘first Saka emperor in India, (Op.
cit, p. xj).
amongst scholars.12 The Tocnarians—Sanskrit, Tusaras—are

equated with the Kusanas who consolidated their hold over the
Kabul valley under their leader Kujula Kadphises.13 The early
history of the Kusanas or Kwei-shuang, a yabqou of the Yueh-
chi, is recorded in the well-known passage of the Hou-Han-shu
or ‘Annals of the Later Han Dynasty’. A reference to the re
lations between the Kusanas and the Yueh-chi and the later
history of the former has already been made, but the historical
aspect demands fuller consideration in the context of their
contribution to political integration of Northern India, including
present Pakistan with South-west Central Asian territories. Their
empire extended in the east probably as far as Khotan and
certainly Bactria, their earlier hold according to the Chinese
Annals. The first ruler Kiu-tsiu-kio who had attacked the other
four hi-hou, styling himself as king of his kingdom called Kwel-
Shuang, later on invaded An-si (Parthia) and seized the territory
of Kao-fu (Kabul). He triumphed over Pu-ta and Ki-pin and
possessed these kingdoms. He died as an octogenarian and was
succeeded by his son Yen-kao-chin, who, in turn, conquered
India (Tien-Shu).
Kujula Kadphises is mentioned as a prince in the Takht-i-
12. The identity of Asiani, according to Hambly, presents some problems,

though their historical role is, however, clear. They were the group who
became king of the Tochari. Though the identity of the names is unlikely,
their activity coincides with that of a clan-group later on famous as the
Kushanas. They were regarded as historically equivalent by Tarn (Op. cit,
287, 533); lately endorsed by Sinor (Introduction a l ’etude del’ Eurasie Centrale,
233) that the Asiani were identical with the Wu-Sun (Hambly: Op. cit,
p. 42).
13. The advance of Tochari is suggested by the attachemnt of their
name to the district of Tokharistan, which centres round Qunduz and Baghlan
close to the Upper Oxus. Ptolemy calls them a great tribe in his Geography
<VI, 11,6). Early in the first century B.C. their advance guards seem to have
pressed southwards towards the passes of the Hindu Kush. But it was not
until the opening decades of the Christian era that they achieved final unity
under the leadership of the Kushan clan, and secured the Kabul valley. The
history of the Kushana rulers Kujula Kadphises and his son Vima Kadphise
(Yen-Kao-chen) is mentioned in the Chinese annals of the Later Han Dynasty
(Hau-Han-Shu)—(Hambly : Op. cit, pp. 42-43). The Chinese account is
noticed by Sten Konow (Op. cit, Ivi, lxii) and is discussed in detail in my
Kusharias—Op. cit, Chapter II, pp. 12 ff).
Bahi inscription14 of Gondophernes. About two decades later

this Kusana is a full-fledged monarch (maharaja)and in another
fourteen years in the year 136 of the old era he is great king,
king of kings and son of God’.15 The Kusana empire founded by
this ruler expanded on both sides of the Hindukush and became
the most influential civilizing force in Central Asia. The Kusanas
also provided the means and resources to stimulate Chinese
trade and to form a bridge between the civilizations of India
and China. The political unification of peoples and tribes with a
different ethnic background, language, culture and religion was
the significant contribution of this period. This mighty Kusana
empire stretched from the Aral Sea to the Arabian Sea and
heralded an era of political stability and unity.
While Bactria was the original nucleus and centre of the
Kusana kingdom, the vast empire of these Central Asian peoples
included Northern India as far as Bihar in the east, and Sind
and Baluchistan, now in Pakistan in the south-west. In the
north and north-east it included Kashmir and extended upto
Khotan, while in the north-west Bactria and Parthia were parts
of it. The Greek traditions in Bactrian culture were preserved
and further assimilated under the Kusarias who also looked to
Hellenic, Iranian and Indian ideas and icons to form a totality
of ethnic-cultural and socio-political phenomena. This is evident
from the portrayal of divinities drawn from different pantheons16
and portrayed on their coins. Buddhism was patronised by the
great Kusana ruler Kaniska in whose time the fourth Buddhist
Council was held at Kalakavana, either in Kashmir or in
Jalandhar.17 The Great Silk Route, the first transcendental trade
and diplomatic road in the history of mankind was laid along the
kingdom of the Kusanas from China to the Mediterranean
Roman empire.18 The entire eastern Section of the Great Silk
14. Sten Konow : CII. 11(1), pp. 57 ff.

15. Ref. Panjtar inscription (122) ibid, pp. 67 ff; and Taxila Silver Scroll
inscription (136) ibid pp. 70 ff.
16. Ref. Puri : Kushanas, pp. 218 ff.
17. Kern : Manual o f Buddhism, Reprint, Delhi, 1968, pp. 121 ff.
18. See ‘Along the Ancient Silk Routes—Introduction by Herbert Hertel,
New York, 1982, p. 18. See also Gafurov ; ‘Kushan Civilisation and World
Culture’ in Central Asia in the Kushan Period, p. 76.
38 • Buddhism in Central Asia
Route was plied by the Sogdian merchants who founded their

colonies and settlements in Central Asia and established trade
and cultural relations with peoples of different nationalities.
The co-existence of various ethnic-cultural traditions and
different religious systems and creeds was remarkably signi
ficant in this period. It was noted for the spirit of tolerance and
understanding involving the spirit of peaceful co-existence. As
has been suggested, images of divinities associated with Indian,
Iranian and Hellenistic pantheons were portrayed on the coins of
Kaniska and Huviska.19 These included Mithra embodying
justice, the goddess Ordokhsh signifying fertility, Vertragna, the
mighty God of War, Siva, Buddha, Helios, Selena and Sarapis
and many others. The syncretism of the Kusapa pantheon on
their coins reflects the ethnic and cultural heterogeneity of the
vast empire’s population, its close association, reciprocal in
fluences and mutual enrichment of peoples of different back
grounds—social and religious—who constituted the Kusana
state.
Despite the eclectic approach of Kaniska and his successors,
the great Kusana monarch was kind towards Buddhism. His
patronage of the fourth Buddhist Council is already recorded.
A Kharosthi record20, probably of the year 1, from Shah-ji-
Kidheri(near Peshawar) records his association with the founda
tion of the great stupa at the site where the famous casket was
also found. This was the gift of the emperor Kaniska to
the monastery in Kaniskapura. The Buddhist Council held in the
time of this ruler prepared commentaries on canonical texts.
One of these—the Mahavibhasa—was jointly compiled by Par§va
and Vasumitra. Buddhism seems to have made considerable
progress not only in India but also in Central Asia in this period.
Spreading across the Hindukush along the trade route to China,
its contributions in this period were tremendous in the form of
great Buddhist monuments at Bamiyan, Surkh-Kotal and
Adjina-Tepe in Tajikistan. The knowledge of Indian Kharosthi
script is revealed from the documents from Niya not far from
19. Rosenfield, J.M. : The Dynastic Arts o f the Kushanas, California,

1967, pp. 69 ff; Puri : Op. cit, pp. 218ff.
20. CII.II(l) pp. 135 flf.
Khotan, and such Buddhist scriptures as the Gandhari Dharma-

pada.2i
The period of the Kusanas, corresponding with that of
Hadrian and his successors in Rome, was one of great pros
perity in the ancient world. In fact, this could be due to the
great commercial activity for the overland silk trade between
China and Rome in the first two centuries of the Christian era,
and the Kusanas and the people under their rule seem to have
participated in it. At the time of Parthian hostility, the Kusanas
could divert the caravans southwards from Balkh to the Indus
Delta for export of commercial goods by sea. Items of export
from China included primarily silk, with Indians supplying
spices, exquisite ivory etc; and, in return, Rome sent manufac
tured things like woollen tapestries, engraved gems, figurines,
metal ware and magnificent glassware. A long list of items of
export and import from and to Indian ports on the western
coast is provided by Ptolemy, as is also furnished by the
Periplus,22 Excavations at Begram in Afghanistan have also
revealed valuable information in this direction, and many other
sites are equally helpful in supplying minor finds.
The Kusana period in Central Asian history is equally not
able for artistic activity. Mahayana Buddhism became wide
spread in this area and with it grew great religious awakening,
as also literary and artistic contributions of the people having
faith in the religion of the Tathagata. The spread of Buddhism
did not mean the annihilation or assimilation of local creeds
21. Hambly : Op. cit, p. 48.
22. Sec. 41, 59. Majumdar : Classical Accounts o f India, Calcutta, p.
290 ff. The material from these classical works has been utilised by several
authors. See Warmington : The Commerce between the Roman World and
India, Cambridge 1928; K.A. Nilakantha Sastri : A Comprehensive History
o f India, Calcutta, 1957, Chapter XV, p. 430; Puri : Op. cit, pp. 116 ff.
Sewell on the basis of the study of Roman coins found in Southern India
came to the conclusion that the culmination of trade between India and Rome
had reached during the time of Augustus, and it continued upto A.D. 65, the
time of Nero. He points out that the paucity of funds of Roman coins in
Southern India was the consequence of the change of the social conditions
of Rome itself rather than to any political reason. This contention of Sewell
is challenged by Warrington who denies loss of trade {Op. cit, p. 89). For
the Gandhari Dharmpada see J. Brough’s book under this name, London
1962.
and traditions. It was an intricate process of mutual influence,

a creative adoption of the new doctrine with such modifications
as were found necessary under the impact of local traditions.
The creative approach to Buddhism was reflected in Buddhist
temple architecture. Its layout and arrangement did not as a
rule mean copying the Buddhist prototype of the pre-Kusana or
early Kusana Indian counterpart, but it resembles the temples
typical of the Near and Middle East, with their enclosed sanc
tuary surrounded by narrow processional corridors, as revealed
from excavations at Kara-Tepe. This local strain in Central
Asian Buddhism is said to have survived until a later period, as
is evident from the excavations at Adjina Tepe, a Buddhist
monastery in southern Tajikistan.
It is proposed by Gafurov23 that the Bactrian-Tukharian
school which became mature on the local central Asian tradi
tions of Kusana art, contributed in no less a manner to the
Buddhist art of Central Asia. It was from this region that
Buddhism, its art and literature was carried to China, Japan and
Korea. The Chinese writings provide names of Buddhist savants
from Bactria, Sogdia and Parthia contributing in terms of theo
logical treatises and translations of original Buddhist texts into
Chinese. Regular contacts were established between Asia and
China in the Kusana period.
The Ku$anas in Central Asia were in turn supplanted by the
Sassanians from Persia. The founder of this dynasty was
Ardashir I who defeated and slew the Parthian emperor Artavan
V and declared himself paramount ruler of Persia—modern
Iran. According to Al-Tabari, he waged a war in the east of
Iran, occupied Sistan, Abar Shab (modern Nishapur), Meru,
Balkh and Khwarzim and received the homage of the Kusanas
in A.D. 227. The Sassanian conquest is recorded in the inscrip
tion of Shapur I (A.D. 240-72) at Naqsh-i-Rustam, near Perse-
polis. This record, drafted in three languages—Pahlavi, Parthian
and Greek—mentions the lists of provinces of the Sassanian
empire in about A.D. 260. The victorious Persian army is said
to have seized Peshawar, occupied the Indus country and
pushing north, crossed the Hindukush, conquered Bactria,
23. Central Asia in the Kushan Period, Vol. I, Moscow, 1974, pp. 71 flf.
crossed over the Oxus and entered Samarkand and Tashkent.

The Kusana dynasty founded by the great Kaniska was deposed
and replaced by another line of princes who recognized the
suzerainty of the Persians and ruled over a considerably reduced
area.24 The Sassanian-Kusana political relations were, however,
disturbed when the latter helped the king’s brother, who held the
important post of viceroy of Seistan, in trying to seize the throne
in the time of Babram II (A.D. 276-93). .The feudal aristocracy
was always a threat to the Sassanian monarchy. Bactria, Sogdiana
and Gandhara, however, remained under governors of the
Sassanian royal house.25
At the other end in Central Asia, by the fourth century A.D.,
the nomad empire of the Hiung-nu in Mongolia had long been
divided in two—the, northern and southern portions—with
incessant clash between the two groups. In A.D. 311 the
southern section of the Hiung-nu captured and burnt Lo-yang,26
the capital of the northern portion, where they set up a dynasty
which survived for about four decades till its destruction by a
renegade of the same race in A.D. 350. The northern section
had to face another catastrophe in being driven away from the
vicinity of Lake Baikal by their rivals, the Hsien-pi. They finally
emerged upon the Jaxartes steppe to the north of Sogdiana.
The various sections of the Hiung-nu from A.D. 350 onwards
invaded the eastern provinces of the Sassanian empire where
they came to be known as the Chionites.27 The first clash bet
24. Ghirshman, R. : Iran (Pelicans). (Reprint 1978, p. 292.)

25. Hambly : Op. cit, p. 51. Their coins have been described by Herzfeld
— The Kushans-Sassanian Coins—(Memoirs of the Archaeological Survey
of India, XXXVIII) Calcutta, 1930, and analysed by A.D.H. Bivar. ‘The
Kushans-Sassanian Coin Series, Journal Numismatic Society of India (JNSI)
XVIII, 1956, 13-42.
26. McGovern : Op. cit, pp. 307 ff. Lo-yang was famous amongst the
Romans as Sera Metropolis, the terminus of the overland Silk Route. The
ensuing disturbances along the land routes further west are recorded in the
Sogdian Ancient Letters. (W.B. Hennings, ‘The date of the Sogdian ancient
letters.’ BSOAS.XII, 1948.603; Hambly : Op. cit, p. 53.) •
27. According to Ghirshman, the Chionites (a term which he understood
to include the Kidarites) were not distinct from the Hephthalites who play an
important part in the history of the fifth century A.D. McGovern and Enoki
consider the Hephthalites to be fresh arrivals who descended on Bactria early
ween Iran and the new power in the east, the kingdom of the
Chionite—Hephthalites seems to have occurred under Yazdgard
I. According to Ghirshman,28 after they had been settled by
Shapur II on Kusana territory with the title of ‘Confederates’,
the Hephthalites succeeded in evicting the Little Kusanas and
formed a powerful kingdom which, towards the beginning of the
fifth century, took advantage of Indian weakness to expand on
both sides of the Hindukush, and also posed a threat to
India.
The internal situation of the Sassanian empire further de
teriorated under Peroz (A.D. 459-84). His wars against the
Hephthalites brought the country'to the verge of disaster; and
in his ill-fated enterprise of attacking his adversaries, he paid
with his life. During the reigns of the four- successors of Peroz—
a period of over half a century—the Hephthalite king not only
exacted a heavy annual tribute in cash, but also intervened in
Iranian domestic affairs.29 The.Hephthalites continued to pose
threat to Iran and its security even after the restoration of their
protege Qubad to the Sassanian throne in A.D. 488-89. It was
only after A.D. 557 that with the help of the Turks called in
western sources as Sinjibu or Silzibul, Khosrau Anoshirvan
(A.D. 531-79) was able to crush the Hephthalites and their land
was partitioned along the line of the Oxus.
It was during the predominance of these Hephthalites in
Bactria in the fifth and early sixth centuries A.D. that they also
undertook a series of incursions into the Punjab and shook the
foundations of the Gupta empire. By A.D. 570 the Huna chief
Toramana had set up his rule over a big chunk of land in
Northern India from Kashmir to Madhya Pradesh (Malwa'
region). The Huna rule beginning with Toramana did not end
with his son Mihirakula who in A.D. 525 was repulsed by a
confederacy of Indian princes. Their hold in Kashmir lasted
longer. Lakhana and Khingila in the second part of the sixth
century A.D. are supposed to have ruled at Kabul or at Gardiz,
in the fifth century A.D. and drove the Kidarites southwards (quoted by
Hambly : Op. cit. p. 55).
28. Op. cit, p. 298.
29. Ghirshman : Op. cit, p. 299; Hambly : Op. cit, p. 56.
the latter for at least eight years, as is evident from his record.30
The Huns are mentioned as anti-Buddhists and were responsible
for the destruction of Buddhist monastery., in India. Their rule
in Bactria and other parts of Afghanistan as also in North-west
India (now Pakistan) and Kashmir, had left their homeland
open to other political forces. The Hsien-pifor a time dominated
the steppes of Mongolia, but by the sixth century A.D. a group
known as the Juan-juan, identified with the Avans of the later
times in Europe, gained ascendancy. The rise of the Turks result
ed in the expulsion of the Juan-juan dynasty from the Mongo
lian steppe and its final extinction by A.D. 552.
The founder of the Turkish empire was the chief called Tu-
men (in Chinese sources) and Bumin in the Turkish inscriptions,
with his residence at Aq-Dagh to the north of Kucha. Soon the
Turkish realm extended westwards as far as Oxus and Caspian
sea. Sinjibu (Silzibul), brother of Tu-men in alliance with
Khosrau I Anoshirvan of Iran was responsible for the destruc
tion of the Hephthalite kingdom, and after division set up a
common frontier with Sassanian Iran.31 Istemi, the brother of
Tu-men, same as Sinjibu (Silzibul) died in A.D. 576 leaving
strong Turkish influence in Sogdiana. The Turks were, however,
divided into Eastern and Western; and both made nominal
submission to the Tang dynasty of China in A.D. 630 and 659
respectively. In Mongolia a new empire of the Eastern (Blue)
Turks was established32 in A.D. 682. Qapghan of this dynasty
subjugated the Kirghiz and Turgesh in the west and reached the
Iron Gates in Sogdiana. Between A.D. 699 and 711, the
Khanate of the Eastern Turks included that of the Western one
as well. The Arab expansion with the conquest of Sogdiana, now
named Mawarannahr, and their clash with the Iranian ruler
which lead to the death ofYazdagird III (A.D. 632-51), finally
poised the Arab armies on the banks of the Oxus against the
Turks for the possession of the provinces to the north of the
river.
30. El. XXXV. pp. 45-7.

31. Hambly : Op. cit, p. 59.
32. Giraud : L'Empire des Turcs Celestes, and D. Sinor : Introduction
a I’etude de I’Eurasie Centrale—quoted by Hambly : Op. cit, 323 notes.
At the eastern extremity of the Turkish world in Mongolia,

a new state of the Uighurs33 was formed at the expense of the
t Eastern Turks with its capital at Ordu-Balish in A.D. 744. The
trilingual inscription of Qara Balgasun, the present name of the
old capital records the conversion of the rulers of the Uighur
Kaghanate to the Manichaean religion.34 The text of this
record is in Chinese, Sogdian and Turkish. Sogdian was origi
nally the language round Samarkand, but acquired an interna
tional character as it was used by merchants through the Tarim
basin and spread even to China. As a more varied literary con
tent, this language has provided Buddhist, Manichaean and
Christian texts. That explains its use in this record. The Turkish
dialect written in the Uighur alphabet was extensively used for
Buddhist and other literatures. It may be mentioned that the
name Uighur is more correctly applied to the alphabet than to
the language which appears to have been the literary form of
the various Turkish idioms spoken north and south of theTien-
shan. The use of this dialect for Buddhist literature spread
considerably when the Uighurs supplanted the rule of the
Tibetans in the Tarim basin in about A.D. 860 and founded a
kingdom themselves.35 This might have been a later phase since
the first Uighur empire lasted until A.D. 840 when a sudden
rising of the Kirghiz tribes along the river Yenisei led to the
destruction of its capital and the dispersal of its various tribes.
Certain groups of survivors migrated south-westwards, and
established themselves in the oases of the Tarim basin in
Sinkiang.
33. A Turkic-speaking people of interior Asia are scattered for the most
part in western China in the Sinkiang Uighur Autonomous Region, while a
small number occupy Central Asian parts of the Soviet Union. These Uighurs
located in various parts of Mongolia and western China during the 9th to
13th centuries, eventually moved westwards {Encyclopedia Britannica—
Chicago 1972, Vol. 22 pp. 467-68). The earlier historical account is provided
by Hambly {Op. cit, p. 60) based on Hamilton’s study {Les Ouighours a
I'epoque des Cinq Dynasties).
34. E. Chavannes and P. Pelliot—‘Un traite Manicheen retrouve
Chine’. Journal Asiatique J.A. 1913. 177. (Hambly : Op. cit, p. 324, n.32.)
35. Eliot : Hinduism and Buddhism—Vol. Ill, p. 192. This kingdom is
said to extend into China and lasted long, for Sutras in Uighur were printed
at Peking in 1330 and Uighur manuscripts copied in the reign of Kan-Hsi
(1662-1723) are reported from a monastery near Suchow.
The Political States of Central Asia

Reference has no doubt been made to the different tribes
playing an active role in the history of Central Asia as also in
the carving out of big empires. The political condition of the
country is clearly described by Hsuan-tsang in the middle of
the seventh century A.D. The Chinese pilgrim divides thecountry
into two zones—the western and the eastern. The western zone
consisted of the sixteen states from Ta-mi (Termez) to Kien-chin
(Gaz) covering the whole area from the Iron Pass (Derbent) to
Gaz, the most fertile zone of the Oxus valley. The eastern zone
consisted of the other states, beginning, with An-ta-lo-Fo
(Anderab), which were mostly mountainous and forming part
of the ancient Tu-ho-lo (Tukhara) country. These states were
under Turkish domination, while the western ones were preserv
ing some sort of autonomy. These might have been the original
administrative divisions36 of Tokharistan under the Yueh-chi.
In a broader political perspective one need not enumerate these,
but concentrate on the bigger states which had separate iden
tity in the vast Tarim basin, and lacked political unity. These
units centering round cities or groups of towns, divided by
deserts lived their own civic life and had considerable indepen
dence under local rulers, although the Chinese, Turks or Tibe
tans stationed their troops there and appointed residents to super
vise the collection of tribute. The chief of these cities or oases
were Kashgar in the west; Kucha, Karashahr, Turfan and Hami
successively to the north-east, and Yarkand Khotan and Miran
36. S. Beal : Buddhist Records o f the Western World, London 1906,
Vol. I, p. 37 ff. The pilgrim informs us that at that time the royal race had been
extinct for many centuries, and in his time each of the 27 States, divided by
natural boundaries, was under a chief who was ruling independently. The
country as a whole was dependent on the Turks, and was divided into two
zones—the western and the eastern. The former consisted of the sixteen states
from Ta-mi (Termez) to Kie-Chin (Gaz), the most fertile zone of the Oxus
valley. He visited these states on way to India. The eastern zone consisted of
the other states, mostly mountainous, beginning with An-ta-lo-fo (Anderab)
which he visited in the course of his return journey from India towards
Kashghar. All these states are described by the pilgrim as ‘the ancient Tu-ho-lo
(Tukhara) country—probably under the Turkish domination (See also Vol.
II, pp. 286 ff). Bagchi enumerates these kingdoms and records their identi
fication (India and Central Asia, Op. cit, pp. 25 ff).
Buddhism in Central Asia
to the south-east. Pelliot also notices37 a Sogdian colony to the

outh of Lob Nov, which according to him, might have had
much to do with the transmission of Buddhism and Nestorianism
to China. The historical account based on Chinese sources of
these states might be traced with corroboration from archaeo
logical evidence, with particular reference to Buddhism and its
expansion in these areas.
Kashgar3839
The earliest account of Kashgar could be traced in the
‘Annals of the Former Han Dynasty’. The Chinese writers re
produce the name as Ch’ia-sha,Chich-cha, etc., but also call the
region Su-le, Shu-le, or She-le. After the mission of Chang-Chien
(c. 139-127 B.C.), there was expanding Chinese knowledge of the
western Regions to the Oxus and the confines of Persia. Trade
with Bactria andSogdiana grew rapidly and Kashgar, which was
a convenient emporium, became a Chinese protected state in
the first century B.C. With the relaxation of the Chinese hold
about the time of the Christian era, it was subdued by the
neighbouring kingdom of Khotan. The conquests of Pan-ch’ao
restored Chinese supremacy, but early in the second century
A.D. the Yueh-chih interfered in the political affairs of this
place and placed their protege on the throne.33 It is proposed
that the interference of the Yueh-chih in the affairs of Kashgar
in c. 120 A.D. resulted in the introduction of Buddhism into
that territory.40 This assumption, according to Stein, would
37. Eliot : Op. cit, 200 n; also J.A. January, 1916, pp. 111-123.
38. For a comprehensive account of Kashgar, see Aurel Stein : Ancient
Khotan (Oxford, 1907) Chapter III, pp. 47 ff. Stein utilizes O. Franke’s paper
‘Kaschgar and die Kharosthi’; Sitzungsber der Kon preuss. Akad der fV/ssen-
schaften, Berlin 1903, pp. 184 ff for various Chinese designations of Kashgar.
39. Stein : Khotan, p. 55. For a summary of the Chinese records con
cerning the history of Eastern Turkestan during the first century of our era,
see Richthofen : China, i, pp. 468 (T. See also Spect ‘Etudes Sur L Asie
Centrale ii in JA. 1897, p. 97.
40. Stein : Op. cit, p. 56 and note 23. Franke calls attention to a Tibetan
text translated by Rockhill embodying traditions of Khotan or Li-Yul which
mentions that a princess of Ga-hyag, who bacame the wife of king Vijayasimha
of Khotan, helped to spread Buddhism in Shu-Iik. (See Rockhill : Life o f
the Buddha, p. 240.)
agree with the tradition recorded by Hsuan-tsang which makes

the princely hostage from the States east of the Tsung-ling,
including Sha-Ia or Kashgar, reside in a Buddhist Convent,41
and connects their allegiance with the reign of Kaniska, the
great patron of Buddhism. He presumes that it is more probable
that the Buddhist Church and its establishment, whatever its
period, was the result of impact from Baktra and not Khotan,
as Mahayanism was the prevailing form of Buddhism in Khotan
while Hinayanism was prevalent in Kashgar.
The rapid decay of Chinese power in Central Asia commenced
under Emperor An-Ti (A.D. 107-125) and for nearly five
hundred years the Chinese sources provide only scanty inform
ation about the political conditions of Kashgar and eastern
Turkestan. The revolt of the Uighur tribes in the region of
Turfan and Hami threatened the Chinese domination in the
Tarim basin at its north-eastern end about the same time that
Kashgar fell under Chinese influence. Emperor Wu-Ti (A.D.
265-290) appears42 to have made efforts to re-establish Chinese
influence in the south of the Tarim basin, but the dynasties
following in quick succession until the advent of the Tang
(A.D. 618) were too weak to resume a policy of conquest be
yond Sha-chou—the western-most part of Kan-su. There is no
information about Kashgar during the subsequent centuries
when probably it was part of the Khotanese kingdom. In the
reign of Wen-Cheng-Ti43 (A.D. 452-466) we hear of an envoy
sent by the king of Su-le to the Imperial court to present a sacred
relic, the reputed dress of Buddha which proved incombustible.
Early in the following century, Kashgar figures among the
numerous territories of Eastern Turkestan which, according to
the Annals of the Liang Dynasty and the Pei-Shih (a work
published about A.D. 644), was subject to the Yeh-ta or Heph-
thalites, who had founded a powerful empire in the Oxus Basin.
They carried their conquests down to Gandhara and beyond the
Indus in the south, and as far as Khotan and Karasahr in the
east.44 The Hephthalites, in turn, succumbed to the attacks of
41. Si-Yu-Ki trans. Beal. i. p. 57.
42. Stein : Op. cit, p. 57.
43. ibid, p. 58.
44. The information about the Hephthalites and their conquests is avail-
the western Turks (called Tu-chuch by the Chinese) and all the
territories north of the Oxus then passed under the domination
of the western Turks, who exercised their sway from their
encampments in the valleys of Tien-Shan north of Kucha and
Kashgar. According to the Chinese accounts, the subject states
were left in charge of their hereditary local rulers, each under the
control of a Turkish Tudun who watched over the collection of
the tribute.*45
The reassertion of Chinese influence begins from A.D. 630
when they subjugated the Northern Turks while the western
ones were dissipated by internal feuds after the murder of Tung
Shin-hu Kagan who had accorded a grand reception to the
Chinese pilgrim Hsuan-tsang in A.D. 630. The Imperial Chinese
army in 640 had occupied Kao-chang or Turfan, and the
Protectorate of An-hsi was established in that territory. The
king of Karashahr, who in 632 sent an embassy to China, but
later on had retracted, was vanquished and carried off as a
prisoner. The Chinese conquest ofthe Tarim basin was complete
with the defeat of the Western Turks in 658, and their suzerainty
was re-established over Eastern Turkestan. The whole of this
region was divided into four administrative divisions called Four
Garrisons.46 The Chinese conquest found Kashgar and other
territories of Eastern Turkestan under the rule of the indigenous
princes, and left them undisturbed after accepting their submis
sion. The notice on Su-le contained in the Tang Annals mentions
able from the account of Sung-Yun who in A.D. 520 visited both the Yeh-ta
seats in Badakshan and the king (Mihirakula) representing their power in
Gandhara, distinctly attests Khotan as the eastern limit of the vast dominion
tributory to them. The Pei-Shih which derives its notices about the Heph-
thalites from Sung-Yun’s missions, mentions Sha-li (Kashgar) along with
Sogdiana, Khotan, Bokhara and over thirty "smaller states as among the
western countries, subject to them (Stein : Op. cit, p. 58, n. 5).
45. Stein : Khotan, p. 58 and n. 8.
46. According to the passage from the Tang Annals, the protectorate
was intended to govern Yu-tien (Khotan), Suei-shin (Tokmak) and Su-le,
the whole of these territories (including Kucha itself) being thenceforth known
as ‘Four Garrisons’. There can be no doubt that this term included all
Eastern Turkestan, and not merely the territories enumerated as seats of the
‘Four Garrisons (Stein : Op. cit, p. 60).
the despatch of the first embassy of the king of Kashgar in the

year 635,and the second four years later as tokens of submission.
The Chinese authority and its continuance depended upon their
power to enforce it. The authority over Kashgar and the rest of
the ‘Four Garrisons’ did not last long. In the year 662, a rebel
lion broke out among the western Turks and the Chinese army
suffered humiliation at the hands of the Tibetans who, after their
conquest of the Kuku-Nor region (663 A.D.), rapidly rose to be
formidable rivals of the Chinese power in Central Asia. They
made themselves masters of the ‘Four Garrisons’. From 692 till
the re-establishment of the Protectorate of An-hsi at Kucha and
the reconquest of the ‘Four Garrisons’, the Chinese had to face
two powerful enemies, the Arabs in the west and the Tibetans
in the south: The struggle against both continued during the
reign of Emperor Hsuan-tsung (713-762 A.D.), who had to
adopt a purely defensive policy. In the year 714, the Chinese
pursuing a more active foreign policy initiated by Emperor
Hsuan-tsung recovered their supremacy over the western Turks,
and were then face to face with the Arabs.47 The Tibetan
inroads were continuing into the southern portion of the Tarim
basin, but Chinese diplomacy succeeded in checking them. The
suzerainty of China was acknowledged by all the States threat
ened by the Arabs in the west from Kashmir to the Oxus and
Jaxartes. Kao-Hsien-chih, a general of Korean origin specially
appointed by Emperor Hsuan-tsung to take charge of the
campaign against the Tibetans in the year 747 to wrest Little
Po-lu (modern Gilsib) after marching over the Pamirs and
Hindukush from Kashgar, suffered a crushing defeat in 751 from
which the Chinese never recovered.48
A succession of disasters occurring about the same time were
contributory factors to the decline of the Chinese prestige in
Central Asia. The Tibetans who had a hand in fomenting in
ternal troubles in China, availed of the opportunity, and between
758-59 they gradually overran the regions of Ho and Lung,
corresponding to the present province of Kan-su and the ex
treme west of Shan-si, and all direct communication between
China and the Protectorates of An-hsi and Pei-ting was inter-
48. ibid, p. 63.
rupted.49 The Chinese governors, however, managed to maintain

their authority as heads of the ‘Four Garrisons’ and of Pei-ting
(near Guchen) over the territories under their charge, and
succeeded in sending envoys to the Imperial Court through the
friendly territory of the Uighurs. InA .D . 791, the Protectorate
of Pei-ting was taken by the Tibetans, and from that time on
wards nothing more is heard of An-hsi or the ‘Four Garrisons’.
The Eastern Turkestan disappears from the accounts of the
Annalists of the Tang dynasty, and its history is shrouded
under obscurity for more than a century. The Tibetans domi
nated over the Tarim basin, and posed as dangerous neighbours
to their old friends and allies, the Arabs, in the region of the
upper Oxus. Between 860 and 873 A.D. their supremacy was
broken by the Uighurs who established a powerful kingdom
comprising the region once ruled from Pei-ting and extending
westward as far as Ak-su. Khotan apparently became indepen
dent. The rest of the territories once comprised in the ‘Four
Garrisons’ became subject to Turkish princes of the Karluk
tribe, residing at Bala Saghun, near Lake Issik-kul. Between
A.D. 926-41 occasional missions were sent to the Chinese court.
Very soon afterwards, the ruler holding the territories from the
Issik-kul to Kashgar, was converted to Islam. As Satok Boghra
Khan he established link with the west and the later history of
Kashgar falls beyond the scope of this work.
The information relating to Buddhism in Kashgar, forming
the subject of enquiry, is no doubt supplied by the Chinese
pilgrims who visited Kashgar on their way to or from China.50
The general description of Kashgar and of its people in the
Tang Annals is substantially the same as found in Hsuen-tsang’s61
49. Ibid; Bushell : Early History o f Tibet, p. 41.
50. From the time of the Former Han Dynasty when the States of
Central Asia were first opened upto the political interference of China, down
to the Tang Period the region of the present Kashgar was generally known
by the name of Su-le or Shu-le. The name Sha-le is given to Kashgar in the
accounts of the journeys of the pilgrims Sun-Yun, Kumarajiva (400 A.D.),
Fa-Yung (420 A.D.), Dharmagupta (c. 593-595 A.D.) and Wu-Kung; the
latter’s itinerary distinctly records the identical application of Sha-le and
Sh-li (Stein : Ancient Khotan, p. 48).
51. Stein : Op. cit, p. 66 ff; Beal : Op. cit, Vol. I, p. 19. The notices
of Kashgar during the Tang period are collected by Stein in his work on
Ancient Khotan (pp. 65-70).
Hsi-Yu-chi. The name of the ruling family (Pei) is actually found

in the imperial decree conferring in A.D. 728 the royal title on
An-chih, chief of Su-le. The title A-mo-chih is attested by the
royal decree and was also shared by the rulers of Khotan in the
eighth century A.D. Chih-meng (A.D. 404) saw at Chi-sha
(Kashgar), besides Buddha’s alms bowl, also his spittoon, which
he describes as being made of a stone of variegated colour. The
biography of Kumarajiva records a visit of this savant to Sha-le
or Kashgar about A.D. 400 and specially mentions that he placed
on his head the alms bowl (patra) of Buddha which was believed
to possess the miraculous quality of changing its weight. The
Chinese monk Chih-meng who proceeded to India via Lop-nor
and Khotan in the year 404 witnessed the identical miracle when
handling Buddha’s alms-bowl which was shown to him in the
kingdom of Chia-sha. Fa-hien (Fa-hsienJ also describes his stay
at Chien-cha (Kashgar), being invited there to participate in the
great quinquennial assembly {pafica-pari§ad), and the royal
hospitality accorded to him and other monks (iramanas) parti
cipating in it. The pilgrim mentions a spittoon which belonged to
Buddha, *made of stone, and in colour like his alms-bowl. This
alms-bowl was seen by Fa-hien in Purushpur or Peshawar,
where it was a chief object of pious worship, and is described
by him ‘as of various colours, black predominating, with the
seams that show its four-fold composition distinctly marked.’
Chi-sha, like so many other places, also boasted of a tooth of
Buddha, for which the people, according to the pilgrim, had set
up a stupa. Connected with the stupa were more than a thousand
monks and their disciples, all students of the Hinayana.62
Hsuen-tsang, the next Chinese pilgrim, provides some details
about Chi-sha or Kashgar, with the uncomplimentary character
of the people, but praise for the textile productions of the place.
The written characters in his time were an Indian type, in all
probability a variety of the Brahmi script, but the language and
pronunciation were quite different- from other countries. Bud
dhism was at that time in a flourishing condition, taking into
consideration the number of its followers and their zeal for the
religion of Buddha. He records several hundreds of Sangharamas52

with some ten thousand followers, studying the Little Vehicle
and belonging to the Sarvastivadin school. Further, without under
standing the principles, the monks recite many religious chants
wading through the three Pi\akas and the Vibhasa Two other
pilgrims Dharmacandra of India and Wu-kung of China
passed through Kashgar, the former on his way back from China
and the latter on his way to Gandhara. The former died at
Khotan, while the latter has left valuable evidence relating to
Chinese control over the territory of the ‘Four Garrisons’. The
extension of Chinese power westwards, thus, was helpful not
only in fostering intercourse with India—the home of Buddhism,
but also with its centres in Central Asia. It also simultaneously
profited both Christianity53 and Zoroastrianism for missionary
activities in this part of Central Asia as also in China itself.
Stein records545 a number of stupas close to Kashgar—the
stupas of Kurghan-Tim and Kizul-Debe, as also the ruins near
Khan-vi, including ruins ofTopa-Tim and Maurn-Tim stupa,
but antiquities found in these are scarce.
Khotan56
The early history of Khotan and its importance in Central
Asia are not matters of conjecture. Chinese, Tibetan and Arch
aeological sources provide ample information on both these
points. Yu-tien, Yu-tun, Kiu-tan, Huo-tan etc. are the Chinese
transcription of the original name Godana or Khotana. Hsuan-
53. Op. city p. 71. Marco Polo when passing here on his way to China
(c. 1273-1274) notes of ‘Carcar’ that ‘there are in the country many Nestorian
Christians who have churches of their own’. In Yarkand too he found Nesto
rian and Jacobite Christians. (Yule : Marco Polo A. pp. 182, 187.)
54. Op. cit, pp. 73 If.
55. For a detailed account of Khotan, its early history and archaeological
finds, see A. Stein : Ancient Khotan—Op. cit, pp. 151 IT. He also quotes
A Remusat : Histoire de la Ville de Khotan, Paris, 1820; Stein : Sand-buried
Ruins o f Khotan, Personal Narrative o f a Journey o f archaeological and geo
graphical exploration in Chinese Turkestan, London, 1903. Extracts from
Tibetan Accounts of Khotan based on Rockhill’s Life o f the Buddha, pp. 230 ff,
and the works there quoted, as also Notes from the Gokfihga Vyakarana—also
figure in the form of Appendix E, communicated and annotated by F.W.
Thomas (Stein : Op. cit, pp. 581 -85). The present account is based on Stein’s
work, of course, with proper references and scrutiny to the sources quoted by
him, as could be available.
tsang mentions the name Chu-sa-tan-na corresponding to the

Sanskrit Kustana—‘breast of the earth’—which is also noticed
in Kharo$thi documents of the third century A.D. from Niya
Site. These in their Prakrit queerly mixed with Sanskrit phrases,
employ the form Kustana or Kustanaka along with the far more
frequent one Khotamna. The pilgrim also mentions the name
formerly used Yu-tien by which Khotan is invariably designated
in all the Chinese dynastic histories from the period of the
former Han down to that of the Mings. The Tibetan notices of
Khotan history called ‘The Annals of Li-yul’ also provide infor
mation about the foundation of the Khotan kingdom, as also
about the origin of various Buddhist sanctuaries. These also
provide a long string of royal names with occasional reference
to pious foundations and doctrinal matters. The regal names are
all Indian and formed with Vijaya as the pre-fix. The Chinese
Annals and Wu-kung (Chinese pilgrim c. 786 A.D.) mention
Wei-chih as the family name, which might be a rendering of
Vijaya. The identity of Li-yul and Khotan first correctly indi
cated by Wassilieff is not questioned. The foundation of Khotan
is traced to the God Vaisravana (Pi-sha-men) or Kubera. The
Tibetan Annals make an exiled son of ASoka (Kustana) ulti
mately establishing his kingdom in Khotan, while it is not
possible to fix any date for the supposed Indian immigration.
Both Hsuan-tsang and the Tibetan ‘Annals of Li-yul’, however,
agree that it could be roughly placed before the introduction of
Buddhism into Khotan, which took place in the reign of Vijaya-
sambhava, 170 years after the establishment of the kingdom.5657
According to the Chinese pilgrim, Arhat Vairocana from Kashmir
is said to have first preached the Law in Khotan and had also
set up the earliest Buddhist convent there. It is no doubt pre
sumed67 on the basis of the Kunala legend that there must have
been communication between Buddhist Khotan and the Taxila
region, through Kashmir or Gandhara. A settlement of immi
56. Rockhill : Life o f the Buddha Op. cit, p. 234. Stein : Op. cit, p. 160.
57. According to the account uniformly told in the Hsi-Yu-chi as well
as in the ‘Life’, there stood to the south of the royal city at a distance of 10 li
a large convent built by an ancient king in honour of the Arhat Vairocana
(Pi-lu-che-na) to whom was attributed the introduction of Buddhism into
Khotan. (Op. cit, p. 231).
grants from the extreme north-west of India, as assumed in the

Khotan tradition, seems confirmed by Stein,68 who was frequ
ently struck by a certain curious resemblance in general appear
ance of features between the Khotanese and the Kashmiris.
According to the Annals of the First Han Dynasty, the first
embassy from the Yu-tien was received in China in the reign of
Emperor Wu-ti5859 (140-87 B.C.). Khotan in this period was
a small state with a population of 19,300 consisting of 3,300
families, and included 2400 soldiers. The other small states in its
neighbourhood were then independent. During the second quar
ter of the 1st century A.D., Yu-tien rose to political importance
for the Chinese. However, towards the end of the reign of
Emperor Kuang-Wu-ti (25-27 A.D.), king Yu-lin of Khotan had
become subject to the powerful king of So-che in the territory
of Yarkand but soon after during the period 58-73 A.D., a
Khotanese general named Hsiu-mo-pa revolted and asserted his
independence. His nephew and successor Kuang-te conquered
Cokkuka and made Khotan so powerful that thirteen states to
the north-west as far as Kashgar acknowledged his sovereignty.
Khotan and Shan-shan (near about Lop nor) are described as
the two territories controlling the southern routes leading to
China. Kho'tan, however, submitted to Pan chao, the general
issimo of the Imperial Chinese force. The Chinese power in the
Tarim basin, however, was steadily waning during the second
century A.D. Events recorded for the years 151-152 A.D. provide
a glimpse of the modest limits of Chinese control at \u-tien.
According to the Tibetan Annals,60 the successor of Vijayasimha,
16th in descent from Kustana, allied himself with king Kanika
58. Stein : Op. cit, p. 165.

59. Remusat : Ville de Khotan, pp. 1 ff; Stein : Op. cit, p. 166.
60. According to Sten Konow, citing the annals of Li country (Thomas :
Tibetan Literary Texts etc., Op. cit, London, 1935, p. 119) originally king
Kani§ka and the king of the Gu-zan and the Li ruler, king Vijayakirti, and
others led an army into India and captured a city named Soked (evidently
Saketa). Similarly the translation of Kumaralata’s Kalpanamanditika by
Kumarajiva (c. 405 A.D.) states that Chen-tan old pronunciation, according
to Karlgran No. 1194 and Tsien-de’an 967, Kani$ka conquered Tung Tien-chu
i.e. Eastern India (Levi J.A. IX. viii. 1896, p. 447; IA. XXXII, 1903, p. 385;
Sten Konow : IHQ. XIV, p. 149).
of Guzan (Kani§ka of Kusapa family) and helped him in his con

quest of Eastern India as far as Soked (Saketa).
Khotan’s relations with China were not smooth during this
period. There was a revolt under the leadership of the local gover
nor Shu-po against the Chinese, resulting in the killing of the Chi
nese commander but there was no retaliation from the Chinese
side. Khotan, however, sent embassies to China between 202 and
220. During the epoch of the three kingdoms (A.D. 220-264) Yo-
tien (Khotan) appears to have been under a powerful ruler, and
subjugated the neighbouring states of Yu-ni (Kemiya), Su-le
(Kashgar) and Yung-lu, but it did send another embassy to China
in 222. During the period of the Tsin dynasties (A.D. 265-419),
the Chinese influence seems to have been only casual and super
ficial, if any. Fa-hien’s reference61 to Khotan, where he reached
in A.D. 400 after a difficult journey from Kucha is no doubt
reproduced in Pien-i-tien which furnishes an acccount of Khotan
for the period of the Northern Wei Dynasty (A.D. 386-534). The
Chinese pilgrim found Buddhism in Khotan in a very flourish
ing condition and describes the glories of its monastic establish
ments in some detail. The monks numbered several thousands,
most of them being students of the Mahayana. There were hos
pitable arrangements in the Sangharamas for the reception of
travelling monks, and he notices the custom of erecting small
stupas in front of each dwelling family. The Gomati monas
tery, the residence of the pilgrim and his companions, alone
contained 3000 monks of the Mahayana school. He also
refers to Buddhist celebrations with the taking out of the images
of the fourteen great monasteries, more than thirty cubits high.
The religious festival commencing in the fourth month lasted
for fourteen days with the king and queen participating in it and
offering their obeissance. The pilgrim closes his account with
reference to another great shrine, known as ‘the King’s New
Monastery’.
In 445 Khotan was invaded by Mu-li-yen, the chief ofTu-yuk-
hun,62 resulting in the death of the king and the ravaging of the
61. Fahien’s Travels trans. Legge, pp. 16-20; Stein: Op. cit, p. 169.
62. This chief of the Tu-Yuk-hun, named Mu-liyen was driven from the
Tangut country by a Chinese army, and he had to take refuge westwards. He
country. Relations with China were, however, maintained with

the despatch of embassies in A.D. 457, 466, 467 and 468. To
wards the close of the reign of Hsien-Wen-ti (c. 470) an envoy
called Su-mu-chieh arrived from Khotan to seek help against
Juan-juan which was, however, denied. During the reign of king
Vijaya Sangrama, Khotan was invaded by A-no-so, resulting in
devastation and burning of monasteries. The Khotanese also
retaliated later on, and seem to have recovered very soon.
Normal embassies were sent to China a number of times
between 509 and 541, and the presents included a Buddha
statue of Jade carved in foreign lands. In 519 Sung-Yun reached
Khotan from the direction of Shan-shan and refers to the first
stupa of Khotan—erected miraculously by Vairocana. Both Sung-
Yun and the Pei-shih agree in enumerating Khotan among the
numerous states of the Tarim and Oxus basins which at the time
of the former’s journey acknowledged the sovereignty of the
White Huns. This dependence on the basis of the reference
from the Annals of the Liang dynasty continued during
the whole period of this dynasty (A.D. 502-556). The pilgrim
Jinagupta from India, who in A.D. 555 passed through Yu-tien
on his way to China, just records this fact.*63 During the epoch of
the Sui (A.D. 581-618), the king of Khotan was Wang with his
title Pei-She-pi-lien (Vijaya). According to the Tang Annals as
well, the family of the ruling dynasty was called Wei-che (Vi
jaya), the title borne by members as well, some of whose names
are also mentioned. Reference is made to some rulers who sent
embassies to China, like Wu-mi in 632. Fu-tu-Sin of Khotan
went to China to receive a mandate personally from the Emperor
in 643-649, followed by another Fu-tu-hiong in 674-675 to pay
homage along with his retinues. The names of several Khota
nese rulers are also mentioned in Annals, like Fu-tu-king, Fu-
tu-ta, Fu-tu-koei and Fu-tu-sheng, the last one dying in China
in 757. At that time his brother Vijaya-Yao was ruling the
is said to have killed the king of Khotan and to have effected great carnage
(Remusat : Op. cit, pp. 18, 21; Voyage : de Song Yun, p. 16n; Stein: Op.
cit, p. 170).
63. Stein. Op. cit, p. 172. For the life history of this prince of Gandhara
who lived in the city of Fou-lieou chafou lo (Puruâpura) see Bagchi, Le Canon
Bouddhiane en Chine—henceforth Bagchi, Le Canon—Paris, 1927, pp. 276 ff.
country. He ruled till 786 when Wu-kong visited the country.64

The Chinese supremacy in Khotan ended about 791, although
several missions were sent to China between 940-966.65 Khotan
resisted for long against the Arab invaders but ultimately lost the
ground and by the year 1000 the Muslim rule was established in
Khotan.66
While the political history of Khotan and its relations with
China have been sketched, reference might be made to its impor
tance as a Buddhist centre as also the archaeological finds in the
neighbouring areas constituting the old kingdom of Khotan. The
introduction of Buddhism during the reign of king Vijaya Sam-
bhava, the grandson of Kustana, has already been recorded, as
also the reference made to the monk Arya Vairocana who was
supposed to be an incarnation of Maitreya and for whom king
Vijaya Sambhava built the great monastery of Bar-ma, the first
in Khotan. Fuller information about this event is provided by
Song-yun and Hsuan-tsang. The former mentions the name of
the monastery as Tsan-mo, the Tsar-ma of the Tibetan sources,
identified by Stein67 with Chalma-kazan in the neighbourhood
of Yotkan. Three Arhats from India—Buddhaduta, Khagata and
64. See Chavannes and S. Levi—L ’ltineraire d ’Ou-kong, p. 27, cited by
Stein—Op. cit, p. 177.
65. See Stein—Op. cit, p. 178. Remusat- Ville de Khotan, pp. 74-81.
The renewed diplomatic relations were revived, it seems, to secure help against
the Tibetans. The Chang-Kuang-yeh mission to Khotan was a step in this
direction. The renewed presentations of tribute which are recorded for the next
few decades, however, limited the practical advantages which Khotan could
derive from the distant empire (ibid, p. 179). In 969, the king Nan-tsung-
chang is named as the sender of a mission conducted by Chih-mo-shan, and
accompanied by one of the Buddhist monks who had previously visited the
Imperial court. In 971 it was again a Buddhist priest (Chi-hsiang) who brought
a letter from the king of Khotan, offering to send in tribute a dancing elephant
which he had captured in a war against the kingdom of Kashgar (Remusat.
Op. cit, p. 86; Stein. Op. cit, p. 180).
66. The Pien-i-tien’s extracts notice an embassy from Khotan (Yi-tien)
sent in the year 1009 and sent by the king or hei-han of that territory and that
the ambassador bringing the tribute was a hui-hu i.e. probably a Muhammadan
Turk, called Lo-ssu-wen. This title is a transcription of the Turkish title
Khakan (Khan). The Mohammadan historians furnish one definite fact, viz.
that in 1006 Khotan was held by Yusuf Qadr Khan (Stein—Op. cit, p. 180).
67. Op. cit, p. 232.
Khagadrod settled at Aryastana (Sthana) of Hgeu-to-san, the

vihara built by king Vijayavirya68 Three generations later, the
Chinese princess Pu-nye-shar (Pune§vara), who was the queen
of king Vijayajaya and responsible for introducing seri-culture69
inKhotan, built two monasteries for Kalyanamitra Arya Samgha-
ghosa who had gone from India. One of the king’s three sons,
Dharmananda became a monk and came to India. He belonged
to the Mahasanghika school of Buddhism and built eight mona
steries for it. The number of monastries kept on increasing and
at the time of Hsuan-tsang’s visit there '-'ere in the capital over
100 monasteries with 5000 monks who were all followers of
Mahayanism. The pilgrim also mentions70 some of the impor
tant monasteries, besides the oldest one of Tsar-ma : Ti-ko-po-
Fa-na, Sha-mo-no, GoSimga and Mo-She. The first one probably
Durgha bhavana, contained an image of Buddha, mysteriously
brought to Khotan and is the same'as Bha-va-na of the Tibetan
Annals. The monastery of Sha-mo-no was an important convent
with a stupa nearly 100 feet high. This seems identical with
Sum-na of the Tibetan text, built by king Vijayasimha for Anan-
dasena, the king of Kashgar after his defeat and subsequent
conversion to Buddhism. Its ruins, according to Stein, are
located at the village of Somiya in the neighbourhood of Yotkan.
68. Ibid, p. 186, 223.
69. A painted panel discovered by Aurel Stein from Dandan-Uilliq
presents a spirited picture of the Chinese princess in this act of offering pro
tection to a basketful of unpierced cocoons. An attendant pointing to the
princess’s head dress recalls her beneficent smuggling by which Khotan was
supposed to have obtained its first silk worms, while another attendant
engaged at a loom or silk weaving implement symbolizes the industry which
was founded at the princess’s initiative. A divine figure seated in the back
ground is supposed to represent the genius presiding over the silk worms.
(Stein. Op. cit, p. 230. See also Chapter IX, Section V in this work.) This may
be compared with the version provided by the Annals of Li-yul (Rockhill.
Op. cit, pp. 238 ff). It is connected with the legend of the So-mo-je Convent
and Vijayajaya’s association with it. It seems he was hostile to the raising of
silk worms till the queen Princess Pu-nye-shar, the daughter of the ruler of
China, could exhibit the utility of this industry. The king called from India
the Bhik?u Sahghagho$a and made him his spiritual adviser, and to atone for
his wickedness, he built the Po-tarya and Ma-dza caityas and a great vihara
or, the caitya and the great vihara of Ma-dza. (Stein—op. cit, p. 230.)
70. Beal : Op. cit., I, pp. 10 ff.
The monastery of Mo-she was built by the Chinese queen of a

former ruler (Vijayajaya of the Tibetan text), and was the same
as Maza, built in commemoration of the successful introduction
of sericulture. Aurel Stein locates its ruins at Kum-i-shahidan.
The famous GoSrnga or Gosirsa monastery was built on the
slopes of the Go^rnga Hill. Its inmates were all Mahayanists
in the time of Hsuan-tsang. The monastery is also mentioned in
the Siiryagarbha-sutraf1, a Buddhist canonical text which was
translated into Chinese between 589 and 619 by Narendra YaSas
in a list of holy spots (pifhas) sanctified by the presence of a
Bodhisattva. The site has been located in a two-storeyed cave in
the Kohmali hill on the bank of the Kara-kash. A fragmentary
manuscript of Prakrit Dhammapada in Kharosthi was procured
here by Duttrevil de Rhins.7172 Reference to the Gomati monas
tery, as described by Fa-hien, is already noticed. The pilgrim
also refers to the regulated life of the monks in that monastery.
Khotan at the height of its power extend^! upto Ni-jang .
(present Niya site) in the east and to So-kiu (Cokkuka) in the
west. It originally included four states Jong-lu-yu-mi, Kiu-le,
Pi-shan and Yu-tien (Khotan) proper. Later on the number was
raised to six—Ilchi or Khotan, Yuruna-kash, Kara-kash, Khira,
Keriya and Lo-la-Sung.73 Explorations and excavations have not
only identified the ancient capital of the country, located at the
village of Yotkan, but have also revealed several other sites in
the periphery of Yotkan. The ancient capital before the Chris
tian era continued to be so till the end of the Buddhist period
71. Stein. Op. cit, p. 186,190. The Chinese translation of the SQryagarbha-
sOtra made by Narendrayasas between the years 589 and 619 A.D. in a list
of holy spots (pitha) sanctified by the presence of a Bodhisattva mentions the
residence and caitya of the saint Chu-mo-so-lo-hsiang or Goma Salagandha
near Mount Niu-tou (Oxhead : Go§ir?a) on the steeply-scarped bank of the
river in Yu-tien (cf. Levi. “Notes Chinoises sur 1’Inde.” IV. pp. 31, 40 quoted
by Stein. Op. cit, p. 186n. 10).
72. Stein. Op. cit, p. 188; cf. Grenard. Mission de Rhins, pp. 142 Sq.;
Senart. Les Fragments Dutrevil de Rhins. J.A. 1898. Sept-Oct. cited by Stein.
73. The place Lo-la-sung, according to Stein, is of doubtful identity.
The Hsi-yu-tu-chih, a modern Chinese account quoted by Chavannes gives
Thakkaga as the name of the sixth place. The term ‘six cities’ of the documents
might be compared with the references made in the Tang annals to the five
districts or towns dependent on Yu-tien (Stein, op. cit. p. 268).
(beginning of the eleventh century). A number of earlier

Buddhist sites excavated in the desert include Dandan Uiliq (the
house with ivory), identified with Li-hsieh74 which was probably
the Chinese garrison headquarters till the end of the eighth
century, when it was abandoned by the Chinese. The finds here
include a large number of stucco images and relics from various
sites, as also frescoes with Brahmi inscriptions. The manuscripts
and other records are mostly in Brahmi and Chinese. The
manuscripts include fragments of two canonical works of Maha-
yana school of Buddhism—Prajhdpdramitd and Vajracchedika75
in Sanskrit. Brahmi records, in languages other than Sanskrit,
are mostly in ancient Khotanese, an Eastern Iranian language.
The Chinese records are largely official ones of the military
officers of Li-hsie, besides deeds of private transactions. Buddhist
finds, similar to those of Dandan-Uiliq, have also been discovered
from ruins of two stupas at Rawak, 7 miles to the north of
Dandan-Uiliq. The Buddhist images found in these areas seem
closely related to the Gandhara school.
Pi-mo76 and Ni-jang77, identified by Stein with Uzun-tati and
Niya respectively, were other Buddhist sites mentioned by Hsuan
tsang. The pilgrim refers to a sandalwood image of Buddha
supposed to have been made by king Udayana of KoSambi in
the life time of the Tathagata. Ruins of old Buddhist stupas
have been found at the old site. Niya was also an important
Buddhist centre in the time of Hsuan-tsang, but its ancient parts
were probably destroyed by the erosion of the desert in his time.
Explorations in these old sites have secured a large number of
74. Ibid, p. 267.

75. Stein Op. cit, pp. 257,258,295. The fragments of the first work belong,
according to Hoernle, to the eighteenth and nineteenth chapters of a Maha-
yana text, apparently some kind of Prajndpdramitd written in Gupta characters
of the seventh or eighth century on leaves of very large size, about 18 by 17
inches. The size of the second work is 14^ by 3 inches, with six lines on each
side of Brahmi characters of the upright Gupta type. Vajracchedika is a famous
treatise of the Mahayana school.
76. Stein identifies it with Marco Polo’s Pein (op. cit, p. 285) and modern
Uzan-Tati (ibid, p. 462).
77. Stein (op. cit, p. 311) identifies it with Niya which may well be said
to have retained for long the character of a frontier station as noted by Huen-
tsang.
Kharosthi documents, mostly tablets on wood. These documents

are in Kharosthi and Prakrit dialect of North-Western India in
the Kusana period.78 The kings in these documents bear the
title devaputra, while the names of persons occurring in them are
purely Indian like Bhima, Bangusena, Nandasena, Sitaka, Upa-
jiva,etc., as also local adaptations of Indian names, like Angaca,
Cuvatylina, Phummasena, Piteya, Saiighita etc. There are some
non-Indian names as well—Lipeya, Opgeya, Limira etc. These
records shed light on the material culture and life of the people
in that region during the period of the records.79
Khotan became an active centre of Buddhist studies. In A.D.
259, a Chinese monk Chu-she-hing80 came to Khotan for the
study of Buddhism. He is a fairly well-known figure in the early
history of Buddhism in China being the compiler of the first
catalogues of Buddhist texts translated into Chinese. She-hing
as a serious scholar collected 90 bundles of original Buddhist
texts which he could send to China with his disciple Fu-ji-tan
(Punyadhana). He stayed on in Khotan and died at the age of 80.
These texts sent home by Chu-she-hing were partly translated
by a Khotanese Buddhist scholar Moksala81 who had gone
there in 291. His assistant was an Indian monk, named &ukla-
78. The tablets, though written by many different hands, shared through
out the characteristic peculiarities of that type of Kharosthi writing which
is exhibited by inscriptions of the extreme North-West of India during the
Kusana or Indo-Scythian rule, falling within the first three centuries of our
era (Stein. Op. cit, p. 320).
79. These Kharosthi documents from Niya and other sites in Chinese
Turkestan have been collected and published without translation by Boyer,
Rapson and Senart (Oxford. 1921-29), and are translated by T. Burrow,
Cambridge 1936. The information deduced from these records forms an
interesting piece of separate study.
80. Bagchi, India & Central Asia (Calcutta. 1955) pp. 59fl. The name is
transcribed as Ko-shu-lan, an upasaka of Indian descent, who was born in
China, and translated 2 works in 5 fasciculi, under the reign of Hwvi-ti A.D.
290-306. His translations were lost in A.D. 730. (Nanjio : A Catalogue o f
the Chinese Buddhist Tripitaka, Oxford, 1882, Japanese Reprint, p. 394.
Appendix II, 27.
81. Wu-lo-kha or Wu-kha-lo i.e. Moksala was a $ramana of Yu-then
i.e. Kusutana (Khotan), who together with Ku-sho-lan translated one sGtra
in A.D. 291. (PancavimSati-sahasrika prajhapdramita (ibid. ii. 26).
ratna.02 The texts translated were Pahcavimsati-Sahasrika-Pra-

jhaparamita, Vinayakirtinirdeia and Surahgama-sutra—all Maha
yanist canonical texts. In the beginning of the 5th century,
Dharmaksema,83 a Mahayanist Buddhist scholar from Magadha
was working in Leang-chou. He had an incomplete manuscript
of the Mahaparinirvana-stitra which he had brought from India.
For the restoration of the text, he had to go to Khotan probably
in 412 or 413 and on recovering the second part of it, he returned
to China. The text in 33 chapters was finally translated between
414 and 421. A pupil of Dharmak?ema, Tsiu-Kiu-king-sheng,
from a noble family of Leang-chou, also went to Khotan to
study Mahayana Buddhism. This he did at the Gomativihara
under an Indian teacher Buddhasena,84known as ‘the lion among
the savants’ in all countries of the west. He brought back with
him Mahayana texts on ‘Dhyana’. References to subsequent visits
of Chinese monks of Leang-chou to Khotan, for the collection
of Buddhist manuscripts for translation are fairly well-known.
Some visited in 439, while earlier Fa-ling had brought the
famous manuscript Avatamsaka-sutra which was translated by
Buddhabhadra85 in 418, and that of Saddharma-punfarlka was
brought from Khotan by a monk named Fa-hien in 475 and
translated by Dharmamati86 in 490. Khotan also transmitted
Buddhism through its savants to China. §iksananda, one of the
greatest among the Buddhist scholars of Khotan, reached China
in 695 and worked there till his death in 710. His translations
of 19 texts include his magnum opus, Mahavaipulya or the
Avatamsaka-sutra87 in 80 chapters.
82. Bagchi, Op. cit, p. 60. For Vimala-kirti-nirdeSa and translations,
see Nanjio : Op. cit, Nos. 144, 145, 146, 147, 149, 181.
83. Bagchi, ibid. He belonged to Central India and was a follower of
Mahayana Buddhism. He went to China through the Central Asian route
reaching Leang-chou in the beginning of the 5th century. He translated 25
texts into Chinese. While attempting to go back to Khotan in 433 he was
murdered in the way (Bagchi : India and China, New York 1951) p. 209.
84. Bagchi, Le Canon Bouddhique, p. 222. Known to the Chinese as
Fo-to-se-na, Buddhasena was an enthusiastic Mahayanist, and called the lion
(simha—Che-tseu) among the learned.
85. Bagchi, ibid, pp. 341 ff. Transcribed in Chinese as Fo-to-po-to-lo
translated as ‘intelligent sage’.
86. Bagchi, ibid, p. 409—translated in Chinese as Tan-mo-mo-ti.
87. Bagchi, India & China, Op. cit, p. 140.
The southern route from Niya onwards passed through

several important places which are mentioned by Hsuan-tsang
for their political and cultural importance. According to the
pilgrim, the country to the east of Niya was all desert. About
400 li (about 67 miles or 90 K.M.) to the east was the old Tu-
ho-Io country which was an important territory in early times
being situated in the deltaic region of the Tarim river. Its arable
area was eroded by the desert and the city fell in ruins. It has
been identified with the actual Endere site.88 The extent of the
area over which pottery debris could be traced suggest a relative
ly large settlement. Several fragments of Salistambha-sutra on
paper written in upright Gupta characters of the 7th or 8th
century were found in the ruins here.89 This is probably a
Buddhist canonical work of the Dharanl class.
Stucco relief fragments, as well as fresco paintings on stucco,
were also found on excavations in the Temple area. An interest
ing painted panel in chapel E. 11 shows seated GaneSa90 with a
(yellow) arm, wrist and ankle, ornaments (yellow), tiger skin
dhoti (yellow), tight paijamas (dark brown), feet bare, rest of
flesh pink with red outlines and rosary of white dots. The
archaeological finds from Endere have brought to light Kharo
sthi tablets and manuscripts of the type found in Niya. Ruins of
a Buddhist stupa also suggest that Buddhism was established
here and was prospering before its final abandonment.
Hsuan-tsang also mentions further to the east, Che-mo-to-
na91 and Tsiu-Umo, same as Chu-mo, at a distance of 600 li
from Niya and 1000 li from Niya respectively. It is identified
with Calmadana of the Kharosthi documents, and present
Cherchen.92 The petty records in Kharosthi provide information
relating to material culture, as also to Buddhism. A rectangular
tablet dated in the ninth year of king Jitroghavarman relates to
a transaction by a certain Buddhaghosa, apparently the slave of
88. Stein, Khotan, Op. cit, p. 435.

89. ibid, p. 439.
90. Ibid, p. 442.
91. Ibid, p.435.
92. Ibid, p. 311 n. 7. Stein surmises that by Calmadana is meant the same
locality which Hsuan-tsang calls Chh-mo-fo-na and which, being placed ten
marches to the east of Niya, manifestly corresponds to the present charchan.
the 6ramana Anandasena, concerning some household goods,

pawned or perhaps taken in pledge. The prosperity of Buddhism
as the widely spread, if not actually the prevailing, religion in
the territory is well brought out by frequent references to Srama-
nas, and enumeration of many sacred categories of Buddhist
heaven.93 In Song-Yun’s time (beginning of the 6th century),
there were a hundred families94956'in the city of Cherchen. In this
town there were representations of a Buddha and Bodhisattva
in pure Chinese style which had been brought by Lu-Kuang to
wards the end of the 4th century A.D. in the course of his
expedition in the Tarim basin. In Hsuan-tsang’s time the place
was abandoned but soon afterwards it was re-occupied by the
Chinese army under the Tang, and its name was changed from
Tsiu-mo to Po-hsien about A.D. 674-676. Recent explorations
have exposed the ruins of a Buddhist stupa thus testifying its
association with Buddhism.
Other important towns lay to the north-east of Cherchen.
This, according to the Hsuan-tsang include Na-fo-po which in
ancient times was called Lou-lan, called in the Han Annals
Shan-shan with two cities—the old one called Yu-ni and the new
one Yi-hsiun. The Tibetan documents found here give the
names ‘Great Nob’ arid ‘Little Nob’ respectively. The Chinese
name Na-Fo-po seems identical with Nob of the Tibetan docu
ments and Lop of modern times.05 Shan-shan replaces the old
name Lou-lan, a Chinese transcription of the original name
Kroraina or Krorayina of the Kharosthi documents, The loca
tion of Lop on the oldest route between Tun-huang and the
western countries (like Niya, Khotan etc) along the foot of the
Kuruknag and the Lop desert, adds to its importance.06 A fairly
93. Stein, Op. cit, pp. 363, 367.

94. Stein, Op. cit, p. 436 n. 5. Sun of Yun travelling in 519 A .D. describes
Tso-mo which, according to Chavannes, is identical with Chu-mo (Tsiu-mo)
of Hsuan-tsang and the Tang Annals as a town inhabited by a hundred
families only.
95. Khotan, Op. cit, p. 435.
96. This route retained its importance for traffic even during those times
when Chinese political power had ceased to assert itself in the western region.
The Chinese general and diplomat Pei-Chu has provided important description
of the three routes to the west compiled about A.D. 607. According to him,
detailed account of ancient Kroraina (Lou-lan, Shan-Shan) is

provided by the Chinese Annals.97 The name Lou-lan under the
former Hans, was changed by the Chinese to Shan-Shan from
77 B.C. The neighbouring territories like Tsu-mo (Cnerchen),
Hsiao-yian (on the road to Kucha in the north) and Ching-
chuch (Niya) were all dependencies of the state of Shan-Shan. A
military colony was actually established here to control the road
to Karasahr and Kucha in the north and to fortify Shan-Shan
against the Hiung-nus. The kings of various western countries
came to offer submission. Despite the change of the name to
Shan-Shan, the easternmost part of the countty continued to
bear the old name of Lou-lan (Kroraina).
Kroraina was a stronghold of Buddhism and was visited by
Fa-hien on way from the frontier territory of Tun-huang, cover
ing the distance in seventeen days.98 The ruler of this place pro
fessed Buddhism. There were 4000 monks, all followers of
Hinayana in the country in his time. The common people, how
ever, were not so strict in the observance of Buddhist rules of
conduct, despite their professing Buddhism. Archaeological
evidence proves, according to Stein, that the old settlement north
of the Lop-Nor marshes was by that time abandoned and Shan-
Shan of Fa-hien is actually represented by the remains of Miran
and Charkhlik. Besides remains of stupas99 and Buddhist shrines
a large number of Kharosthi documents100 were also found on
wood and paper. The use of Kharosthi—the same Indian script
found in the records of the Niya site—in the Lop region for
administration and business purposes poses interesting problems.
While at Khotan its use might have been the result of early
immigration from India as an important element in the local popu
lation, its popularity so far away to the east at the very threshold
of China, could be due to the spread of Buddhism. The missionaries
might have carried with them the language and script prevalent
in the extreme north-west of India for common use throughout
the southern route passed through Shan-Shan, south of Lop-nor, on to Yu-

tien or Khotan (Richthofen, China I, p. 530, note. Stein, Serindia, p. 323).
97. Stein, Serindia, pp. 325 ff.
98. Ibid, p. 324.
99. Ibid, pp. 389 ff.
100. Ibid, Sec. IX, pp. 413 ff.
the Tarim basin. The other possibility, suggested by Aurel Stein,

is the temporary extension of the Indo-Scythian power from
across the Pamirs. Some light is shed on this point by the
Buddhist traditions in China. These documents show the closest
agreement with those from the Niya site in character, language,
phraseology and other aspects. Style, phonetics and spellings
suggest that identical standards were followed from Khotan to
Lop during the period of the records. In both the documents
there are numerous names of unmistakably Buddhist or Indian
derivation,101 such as Anandasena, Bhatisame, Bhimaya,
Budhamitra, Dhamnapala, Komudvali, Purnadeva, Caraka,
Sujada, Vasudeva occurring side by side with those of local origin
such as Cauleya, Cuvalayina, Kapgeya, Kalpisa, Kipsa, Kitsai-
tsa, Lampurta, Maldraya, Porbhaya, Pulkaya, Signaya, Tasuca,
Tameca, Varpeya. The common official titles in KharosthI
records from both the sites are Cojhbo, Gusura, Kori and
Vasu.
In this context reference might as well be made to notices in
the Chinese historical records which shed light on the character
of the ruined settlement of the Lou-lan site. The change of the
name from Lou-lan to Shan-Shan from 77 B.C. onwards has
already been mentioned, as also the latcf events relating to pro
posed military colony and subsequent history of Chinese occu
pation and submission of the kings of Kucha and other western
territories recorded. The fuller history and references in later
Chinese annals now need be told. The epoch of the three king
doms (A.D. 221-65) coincides with the period of the extant
documents and other remains of the Lou-lan site date. The Wei-
lo, a work composed by Yu-huan between A.D. 239-65,
furnishes information102 relating to the ancient route through
Lou-lan, starting from Tun-huang and Yu-men-kuan (Jade Gate
barrier). A later Chinese text, the commentary on the Shui-ching
composed by Li-lao^yuan before his death in A.D. 527, also
mentions Lou-lan and the ancient geography of the Lop country,
as also the kingdom of Mo-Shan same as Shan-Shan through
which the river Ho (i.e. River of the North)—rivers of Kash-
101. Ibid, p. 414.
102. Ibid, p.418.
ghar and Yarkand passed. This river in its course eastwards

passed through the town of Chu-pin, identified by Aurel Stein
with Ying-pan, where remains of Buddhist sites were found.
There is no trace of any Chinese record relating to the military
colony established at Lou-lan of a date later than the time to
which Li-tao-yuan’s notice refers. It is proposed103 that the
abandonment of the Chinese station at Lou-lan took place some
time during the fourth century A.D.
The site of Miran, explored and excavated by Aurel Stein-
points to its strategic importance and also as a centre of
Buddhism. Situated to the south of Lop-nor, it was of great
importance to the Tibetans who in all probability built the fort
in the eighth century. The Tibetan domination in the Tarim
basin extended from the downfall of Chinese power westwards.101
The excavations did not reveal the slightest scrap of Chinese
writing while the Tibetan documents were found in abundance.
This little oasis formed the key to the direct routes which led
from the southern oases of the Tarim basin to the easternmost
part of this basin and from Central Tibet and Lhasa across the
high plateaus and ranges of the Kun-lun to this point. The
Tibetans tried to keep a firm hold upon Miran as long as their
political and military ambition was focussed towards Eastern
Turkestan.105 With the disappearance of Tibetan power in this
region, Miran sank into insignificance. Excavations have reveal
ed ruins of ancient Buddhist shrines with damaged stucco sculp
tures showing good modelling and proportions, influenced by
the Greco-Buddhist Art of Gandhara. Several colossal heads,
all displaying with equal clearness features of Greco-Buddhist
style, suggest that the outer wall was lined by huge seated figures
of Buddha or Bodhisattvas in niches of which these heads formed
part. Stein discovered in front of and close to the base of one
statue a fragment of a palm-leaf manuscript about six inches
long written in Sanskrit with early Gupta Brahmi characters.
These frescoes of Miran form a class by themselves and are con
sidered the earliest, and are also supposed to be a product of the
103. Ibid, p. 426.
104. Stein. Ruins o f Desert Cathay, London, 1912, p. 449—henceforth
Ruins.
105. Stein. Serindia, p. 476.
Gandhara school.106 In one of the paintings depicting a scene

from the Visvantara Jataka, the painter’s name Tita—a Prakrit
form of Titus—is written on the leg of an elephant. A detailed
study of the theme and technique of Miran paintings as also of
finds in excavations is reserved for later consideration.
Representations of winged angels as in early Christian art
point to western influence. The silk banners discovered in Miran
contain Kharosthi inscriptions, which reveal donation of such
banners to the Buddhist shrine. Most of the donor’s names
appear to be Indian like Asaga, Caraka, Caroka (Caruka),
Samanay (Sramanaka), though a few Iranian ones like Friyana,
Firina and Mitraka are also mentioned. Miran seems to have
formed part of Kroraina107 (old Lou-lan, Shan-Shan) and the
frescoes, banners and other antiquities bearing Kharosthi inscrip
tions belong to about the fourth century A.D. The Kharosthi
documents mention five kings of a dynasty which reigned in
Kroraina in the 3rd century A.D. These are : Pepiya, Tajaka,
Amgoka, Mahiri and Vasmana—the third and fourth probably
had a longer reign. The rulers also used titles borrowed from
the Kusanas—Maharaya, Rayatiraya, Devaputra (Maharaja,
Rajadhiraja, Devaputra) and Dharmiya, Mahamta etc.
Northern Route States
The States on the Northern Route were equally important for
their strategic location as also for their political and cultural
history. They were culturally integrated, and they actively parti
cipated in the diffusion and expansion of Buddhist culture. The
Chinese pilgrim Hsuan-tsang mentions them according to their
106. See Busagli. Painting o f Central Asia, p. 20. According to the Italian
Professor, Miran represents the beginning of Painting in Central Asia.
Although the date of these wall paintings cannot be fixed with any precision,
some clues are provided by the writing of the Kharosthi inscriptions. (Ibid,
p. 21)
107. According to Stein, if the site of Miran was known to the Tibetans
as ‘Great Nob’, it appears very probable that by ‘Little Nob’ they meant
Charklik. This distinction would closely correspond to that which the Han
Annals indicate between the two main places of Shan-Shan or Lou-lan. Yu-ni,
‘the old town’ to the east, and I-hsun or ‘the New Town’, these two being
now represented, I believe, by the sites of Miran and Charklik respectively.
(Ruins. I. p. 449)
location from west to east : Po-lu-chia, Kiu-che, A-ki-ni and

Kao-chang. Po-lu-chia in earlier Chinese texts is mentioned as
Ki-me and Ku-ine. Some Sanskrit documents of Central Asia
mention it as Bharuka and it is identified with modern Aksu.108
The former Han Annals place it correctly to the west of Kuei-
tzu or Kucha at a distance of 670 Ii or 110 miles. The documents
from Central Asia mention it as Kuchi. These two kingdoms
had many things in common like customs, literature and
language. The third kingdom A-ki-ni is mentioned in earlier
Chinese accounts as Yen-ki, Wu-ki and Wu-yi—all connected
with Sanskrit Agni or its derivatives. The Sanskrit documents
mention it as Agni and its rulers as Agni Maharaja. It is identi
fied with modern Karasahr. These kingdoms had separate
identity but seem to have common history, especially with
regard to their relations with China—political and cultural, and
in the context of Buddhism.
Ak-su—the Ku-mo of Han Annals, is equated with the little
kingdom of Po-lu-chia by Hsuan-tsang, whose description is
reproduced in the Tang Annals without adding more than the
identity of the ‘Little Kingdom’ with Ku-mo or Chi-mo. The
area of the kingdom is recorded as about 600 li from east to west
and 300 li from north to south, and the size of its capital as 5
to 6 li in circuit. According to the pilgrim, in general charac
teristics this country and its people resembled Chiu-tzu (Kucha)
and its people, but the spoken language differed a little. There
were some tens of monasteries with above a thousand
brethren.109 The fine cloth and serge of the district were esteem
ed by the neighbouring countries. Despite the material advance
ment and its geographical location—as the meeting ground of
trade route between Karasahr and Kashgar and that to the great
fertile valley, north of Tien-Shan, connecting the Tarim basin
with the Hi valley and the trade emporium of Kulja, Aksu was
far less important than Kucha in resources as also as a centre of
Buddhist activities.
Kucha has at all the times been one of the most important
territories in the Tarim basin owing to its geographical position
108. Stein. Serindia, pp. 1297 IT.
109. Watters. Op. cit, 1. p. 64.
and the role it has played in Buddhist art and civilization. It is

supposed to be a worthy pendant of Khotan. Its language too
has engaged the attention of scholars.110 During the Han times
Kiu-she was divided by the Tien-shan into two zones —one was
called Kiu-she anterior or inner, and the other Kiu-she posterior
or outer. The name of the country was changed into Kao-chang
from the beginning of the fourth century. The anterior part of
Kiu-she is identified with modern Turfan, and the other one with
Guchen. The history of the four northern states—Aksu, Kucha,
Karasahr and Turfan—can be traced from the Chinese sources,
which no doubt speak in greater detail about Kucha. It was the
most powerful of the four states and played the same great role
in the North as Khotan did in the South in the diffusion of
Indian culture and Buddhism. The Chineres sources record its
political relations with China as also the activities of the Buddhist
savants of this place and their contribution to Chinese
Buddhist literature.
According to the Annals of the former Han dynasty, the
capital of the king of Kucha is called the city of Yen. In the
south the country borders on Tsing-tsiue, in the south-east,
Tsiu-mo, in the south west, Yu-mi, in the north Wu-sun and in
the west Ku-mo. It was brought into contact with China in the
reign ofWu-ti (40-87 B.C.)and it became a place of considerable
importance, because of its position at the junction of the western
trade routes leaving Kashgar and Aulieata respectively. Kucha
accepted the Chinese civilization but it had its periods of
troubles as well involving conflict with Imperial China. It is not
until the western Tsin dynasty that it figures as a seat of
Buddhism. The Tsin annals (265-317 A.D.) point to its enclosure
by a triple wall and its containing a thousand stupas and
Buddhist temples as well as a magnificent palace for the ruler.111
This suggests that by that time Buddhism was already establis
hed, but there is no evidence regarding the date of its introduc
tion. In 383, Emperor Fu-chien of the Tsin dynasty sent
110. Levi. JA. 1913, pp. 311 ff.

111. For an account of the political history of the Kingdom of Kieou-
tseu (Koutcha), see Appendix A, Part ii of Chavannes article ‘Les Docu
ments sur Bois de Niya’ in Stein’s Ancient Khotan, p. 544.
Luk-uang to subdue Kucha.112The expedition resulted not only in

political success but also in the loss of the famous Buddhist
savant Kumarajlva who was taken as a captive to China. Lu-
kuang subsequently became ruler of the State known as southern
Liang with Kumarajlva acting as his adviser.
The Chinese relations with the States of the Tarim basin
could be traced to the time of Emperor Wu-ti (140-87 B.C.) of
the Han dynasty. Dated records beginning with the year 98 B.C.
testify to the presence of a 'Chinese garrison near the modern
Tun-huang.113 Earlier, an envoy named Chang-chien was sent to
the Yueh-chi soliciting their cooperation against of the warring
and growing power of the Hiung-nu which, however, proved
infructuous. It only helped in revealing the importance of the
nations on the Oxus which were in touch with India on the one
hand and the mysterious one on the other. The Chinese were,
therefore, interested in keeping the international trade route leading
westwards from the extremity of the modern Kansu province
open. The main hinderanceto this intercourse with the west was
the hostility of the wild tribes pillaging caravans and blocking the
route. The Chinese, therefore, aimed at creating a wedge between
two tribes with their alliance with the one against the other.
The reign of Emperor Ming-ti is important as the traditional
date for the introduction of Buddhism as also for the victorious
compaigns of the famous general Pan-chao, whose career lasted
from A.D. 73-102. The Chinese activity ceased for quite some time
after him. The relations with Kucha in the centuries preceding
and following the Christian era were at times hostile and at
others friendly.114A Wu-sun princess, connected with the Chinese
imperial family on the mother’s side, was married to Kiang-pin,
the king of Kucha who went in 65 B.C. to China to pay his
homage, and this bond of kinship continued in the time ofChen-
gi (32-1 B.C.), his son. After the fall of the Hans in A.D. 24,
112. Eliot. Hinduism and Buddhism. Op. cit, p. 203. The circumstances
which provoked this expedition are not very clear. It was escorted by the king
of Turfan and other small potentates who were the vassals of the Tsin and
also on bad terms with Kucha. They probably asked Fo-chien for assistance
in subduing their rival which he was delighted to give (ibid. n.3).
113. Ibid, p. 197.
114. See Bagchi—India and Central Asia. Op. cit, pp. 70 ff.
the Hiung-nu dominated the Tarim basin and the small states,
excluding Kucha, became dependencies of the Hiung-nu, Interstate
rivalries and confrontations only helped foreign powers—the
Hiung-nu br the Chinese in keeping their control over this area
and its political activities. In A.D. 73 Kien was made king of
Kucha by Hiung-nu who attacked Su-le (Kashgar) and conque
red it. He appointed Tou-!i,a man of Kucha, as its king, only
to be defeated and taken prisoner a year later by Pan-chao, the
Chinese General, who set up Chong, the nephew of the last king
as his protege on the Kashgar throne. After the death of Ming
in A.D. 75, Kucha and Aksu attacked Kashgar. This was
followed by Pan-Chao’s arrival and a joint force of Kashgar,
Sogdiana, Khotan headed by him attacked Aksu and She-chang
(LJch Turfan)and defeated them. Kucha was tinally robbed of
its independence after a decisive battle in A.D. 88115116, and a later
attempt to regain it with the help of the Yueh-chi forces failed
to retrieve the situation. Po-pa was set up as king of Kucha
with a Chinese governor-general to look after the local affairs and
maintain peace. Subsequent attempts against the Chinese proved
abortive.
In A.D. 124 the Kucha king Po-ying along with rulers of Aksu
and Uch-Turfan helped the Chinese against the Hiung-nus116
who were finally defeated. Kucha continued to maintain friendly
relations with China and in 224 sent an ambassador with presents
to the Chinese court. At that time Aksu and Turfan were de
pendencies of Kucha. This process was followed by the des
patch of a Kuchean prince in 285 to enter the imperial service,
115. The Chinese power in the Tarim basin under the great general
Pan-Chao reasserted itself with the conquest of Khotan and Kashgar about
A.D. 74, followed by extending influence over other territories. In A.D. 88
Pan-Chao succeeded in subjecting So-Che (Yarkand) in spite of the help of
Kucha, which too was obliged to make its submission along with other
territories on the northern ruin of the Tarim basin. (Stein. Serindia. I. p. 83)
116. Chavannes. Toung-pao, 1906, p. 252 quoted by Stein—Op. cit,
p. 332. It is said that after Pan-yung’s arrival at Lou-lan in February A.D.
124, the king Shan-Shan was awarded for his submission by new honours.
The kings of Kucha, Aksu, and Uch-Turfan came to offer their allegiance.
Taking the numerous force brought by them, Pan-Yung then moved upon
Turfan and after inflicting a defeat on the Hsiung-nu or Hunus, established
a military colony at Lukchun in the Turfan depression.
and by embassies in the first quarter of the 4th century.

Enmity between Kucha and Karasahr resulting in the killing of
Po-Shan the Kuchean king, and subjugation of his kingdom
was followed some years later by the reverse process of murder
of Long-Huei. The Karasahr ruler of the Tarim basin, finally
involved China in the political embroil of the region. Reference
has already been made to the mission under General Lu-kuang
which invaded Kucha in 382 and subjugated it.
During the Wei period (386-534) Kucha-Chinese relations
were again strained and another punitive expedition was sent in
448. After this, a long series of tribute bearing missions were
sent first to the court of Wei, and afterwards to the Liang, Chou
and Sui. King Su-fa Pu-Kiu and Su-fa sent missions to China,
the latter in 630. It was about this time that Hsuan-tsang passed
through Kucha and gave an account of the kingdom and its
rulers who were of the Kiu-che (Kuchean ) race. He refers117 to
the use of modified Indian alphabets, and dwells on the many
monasteries and great images, the quinquennial assemblies and
religious processions. There were more than 100 monasteries
and upwards of 5000 brethern who all followed the Sarvastivada
school and the ‘gradual teaching’, probably meaning HInayana
as opposed to the sudden illumination caused by the Malia-
yanist revelation. The monasteries were centres of learning.
The Kuchean-Sino relations entered a phase of bitterness and
hostility, once again, in 648 when an expedition was sentagainst
Kucha because of its help in the revolt of Karasahr against the
Chinese authority. The Kuchean army led by Kie-lie-tien was
totally defeated, resulting in the capture of king Ho-li Pu-sho-pi,
minister Na-li and the general. They were sent to China. In 658
Kucha was made the seat of government for the territory known
as the ‘Four Garrisons,'118 but the Chinese protege of the old
house continued to rule. Su-ki, son of Ho-li-pu-She-pi, appoint
ed king by an imperial mandate, sent an embassy to China in
674 while Yen-tien-tie personally came to China in 692 to pay
personal homage to the emperor. It was about this time that the
117. Beal. Op. cit. I. pp. 19 fT.
118. By the term ‘Four Garrisons’, the territories of Kashgar, Khotan,
Kucha and Karasahr, then occupied by the Chinese forces, are meant (Stein.
Khotan. 7n.).
I
Tibetans drove away the Chinese from the southern part of the
Tarim basin, and extended their conquests as as far as Karasahr.
The Chinese headquarters were removed to Kucha from Turfan.
During the next century Kucha sent several missions to the
Imperial Chinese Court, and, about 788 was visited by Wu-
Kung,119 who found Buddhism and Music flourishing here. He
mentions an Abbot who spoke with equal fluency the language
of the country, Chinese and Sanskrit. At that time Po-Hoan
was the king of Kucha. The following period is a blank in the
history of this kingdom as also in its relations with China. It is
only in the eleventh century that reference is made to the mission
going to China from this place. These were under the Uighurs,
but Buddhism was not extinct even at that time. In 1096 the
envoy presented a jade Buddha to the Emperor. The new rulers
of the Uighur stock took the title of ‘Lion-king’.
Agnidesa or Karasahr
The kingdom of Karasahr was politically connected with
Kucha and very often they joined hands in resisting the Chinese
aggression for retaining their independence. A Chinese tablet120
(N.XV.93) mentions Shan-Shan along with Karasahr, Kucha
and Kashgar as subject to an unnamed native ruler. It seems to
belong to the lime of Wu-ti (A.D. 265-290), the first emperor
of the western Chin dynasty. The tablet probably refers to Lung-
Hui, king of Karasahr who about the close of that reign esta
blished his hegemony over the whole of the Tarim basin.121
Some documents from the Lou-lan site also provide interesting
information about the affairs of Yen-chi or Karasahr. One such
document122 (No. 934) reports political events in which Tsang,
King of Yen-chi, was implicated. It also mentions Kucha. In
another document (935) there is reference to a declaration of
war. Karasahr offered easy access and was a convenient gate for
Hun inroads into the Tarim basin.123 The Huns also made an
119. Eliot. Hinduism and Buddhism, p. 205.

120. Stein. Serindia, Op. cit, p. 329.
121. Cf. Chavvannes note entitled Les Documents sur Bois De Niya—
Ancient Khotan, pp. 537 ff; also Royaume : De Yen A7(Karachar) pp. 542 ff.
122. Serindia, p. 413.
123. ibid, p. 338.
attempt in 104 B.C. to cut off the return of an expedition to Ta-

yuan (Farghana) by a force of cavalry posted in Lou-lan.124 The
Chinese accounts contain some references to the political con
ditions of Karasahr from the Han times to the end of the Tang.
During the first expansion of the Chinese supremacy under
Emperor Wu-ti (140-87 B.C.), Karasahr recognised Chinese
suzerainty by sending an ambassador to the Imperial Court. The
relations with China, however, vacillated between friendship and
hostility. Karasahr threw off the Chinese domination towards
the end of the first century B.C., but later on submitted to China
along with the other states. Trouble started again with Shuen,
the king of Karasahr, who revolted against Chinese authority
and killed the Chinese governor and his retinue in A.D. 75. The
Pan-Chao expedition reduced these states, once again, to the
Chinese vassalage with the imposition of Yuang-Mong,125 an
officer of Karasahr in the Chinese Court, being placed on the
throne of his country in A.D- 94. He, in turn, also revolted and
was defeated by the Chinese force in 127, He then sent his own
son to the Chinese court with presents. Since then Karasahr
maintained friendly relations with China for some time. An em
bassy was sent to China in 225, and later on in 285, king Long-
An sent his son to be enrolled in the Imperial Guard. Another
son of Long-An, Huei, was successful against various states of
the Tarim basin and became master of Kucha for some time. His
son Hi was defeated by the Chinese in 345, and several embassies
were sent to China from Karasahr in 437, 439 and 448, but the
Chinese sent an expeditionary force due to the vacillating policy
of Karasahr. Its ruler Kiu-She-pei-na fled to Kucha. The history
of the country is blank for want of evidence for a century or more.
There are references to embassies sent to China from Karasahr
in 564 and 606, the second time by Long-Tu-ki, the then
ruler. The Chinese pilgrim Hsuan-tsang visiting this place in 629
provides good information about the religion and culture of this
place, but says practically nothing about the political condi
tion,126 except referring to the king of the country, a native of
the place. There were some ten or more Sangharamas with two
124. Ibid.
125. Bagchi. India and Central Asia. Op. cit, p. 76.
126. Beal. Buddhist Records. Op. cit. I. p. 17 ff.
thousand priests or so, belonging the Little Vehicle of the school

of the Sarvastivadins127 (Shvvo-Yis-tsai-yu-po). According to the
pilgrim, the doctrine of the Sutras and the requirements of the
Vinaya are in agreement with those of India, as also the written
characters might be with a little difference. The professors of
religion read their books and observed the rules and regulations
with purity and strictness.
The Annals of the Tang dynasty128 mention Karasahr several
times. In 632 king Long-Tu-ki-chi sent an embassy to the Im
perial Court. He was equally helpful in getting a shorter route
to China opened through the desert by isolating Turfan which
was gradually passing under the influence of the Uighurs, and was
interfering in the normal relations between China and the Tarim
desert. The king of Turfan was defeated by the Chinese army
in 640 and was forced to release those inhabitants of Karasahr
who were held as prisoners. Soon afterwards Karasahr fell under
Uighur influence with a treaty of alliance with this tribe, and
alienated its relations with China. The younger brothers of the
king named Hie-pi, She-hu and Li-po-chun. however, entered,
into league with the Chinese to defeat and dislodge Long-Tu-
ki-che. The last one was appointed king of Karasahr replacing
his elder brother who was taken as a captive to China. This
position did not last long. The Uighurs invaded Karasahr once
again, and deposed Li-op-chun. A ding-dong struggle for the
throne continued with the Chinese and Kucha interfering in the ;
127. According to Fa-hien who visited Yen-chi or Wu-ti, as he calls it,
about A.D. 400 from Shan-Shan, has little more to tell us than that there were
four thousand monks, students of the Hinayana in the territory (Legge.
Travais o f Fahien, pp. 14 Sq.). For the name Wu-i (also written Wu-chi) in
other Buddhist texts cf Watters. Yuan Chwang, i. p. 46; Chavannes—Toung-
pao, 1905, p. 564 n. 2. Wu-king, who stayed at Karasahr about A.D. 788,
also calls the town Wu-chi. (J.A. 1895, Sept.-Oct. p. 364).
128. In a very long notice which the Tang Annals devote to Yen-chi
(Kara-shahr) and its afTairs, it is specially pointed out that the territory has
always been subject to the western Turks. The geographical position of Kara-
shahr no doubt added to its strategic importance, and from A.D. 719, Yen-chi
is reckoned as one of the ‘Four Garrisons’ assuring the Chinese hold over
Eastern Turkestan. These Annals estimate the number of households at four
thousand and the number of soldiers at two thousand. Reference is also made
to brisk trade in fish and salt (Stein. Serindia, p. 1181). See also Bagchi, Op. cit,
pp. 76-77, for the relations with China.
affairs of this kingdom. Tu-Ki-che who was a hostage in China

was sent back to his kingdom in 648. Relations between Karasahr
and China continued on a cordial and stable plane with the
former sending embassies between 742-755 after which the
Tibetans conquered Karasahr, thus snapping all relations with
China.
Kao-chang or Turfan129
Turfan is an oasis containing the ruins of several cities and
probably different sites represent the ruins of capitals at different
periods. The name Turfan appears to be modern. The Ming
Annals state that this city lies in the land of ancient Che-shih
(or Ku-shih) called Kao-chang in the time of the Sui. This
name was abolished by the Tang but restored by the Sung. The
main city now known as Chotscho seems to be identical with
Kao-Chang130 and Idiqutshari. It is called by the Mohammadans
Apsus or Ephesus, connected with an ancient sacred site renamed
‘the cave of the seven sleepers’. Extensive remains found in the
oasis include works in Sanskrit, Chinese and various Iranian
and Turkish idioms and also in the two dialects of the Tokharian
language. According to the Chinese sources,131 in the Han
period there was a kingdom, called Ku-Shih with two capitals.
It was destroyed in 60 B.C. by the Chinese general Cheng-Chi
and eight small principalities were formed in its place. Earlier
in the time of Wu-ti (160-108 B.C.) the southern part of the
territory, which is Turfan, had submitted to China, but the
northern one, Guchen, was under the domination of the Hiung-
nus, who were completely defeated by the Chinese in A.D. 89.
This did not mean the end of hostility. War between the two
was frequent and Turfan had to seek Chinese help in A.D. 96
129. The political history of Kao-Chang or Turfan is summarily described

in Bagchi’s India and Central Asia (Op. cit), pp. 77-79. Aurel Stein provides
‘Glimpses of Turfan Ruins’ in his earlier work, *Ruins o f Desert Cathay’ Op.
cit, Vol. II, pp. 353 ff. For Stein’s work in Lou-lan and Turfan areas, see his
monumental work—Innermost Asia.
130. Eliot. Hinduism and Buddhism. Op. cit, p. 205. Pelliot suggests that
Chotscho or Qoco is the Turkish equivalent of Kao-Chang in Tang pronun
ciation, the nasal being omitted (JA. 1912.1 p. 579 quoted by Eliot, ibid, n.2).
131. Eliot, ibid, p. 206 quoting Chavannes. Tou-Kiue Occidentavy, p. 101.
when the army of Guchen attacked Turfan in the time of Wei-

pei-ta. The trouble from the other side continued thereafter as
well, because the two kingdoms were allied with two other war
ring powers. In A.D. 280, the king of Turfan sent his son to the
Chinese couit. From the beginning of the 4th century, the name
of the southern part of the territory was changed to Kao-chang by
the Chinese, and that constituted actual Turfan. In the fourth and
fifth centuries A.D., Turfan had some relationship with two ephe
meral states of Kansu—theHou-liang and Pei-liang, the former
founded by Lu-Kuang, the general who attacked Kucha. The
other one was the creation of Men-haun of the Chu-chu tribe,
related to the Hiung-nu. This tribe took a good deal of territory
from the Hou-tiang in Turkestan as well as in Kansu. Its leader
Men-haun devoted his later years to literature and Buddhism.
Finally, the kingdom was conquered by the Wei dynasty in 439
and then its two members tried their fortune in Turfan where
they ruled for twenty years. An-chou, the second one died in
480 and nine years after his death a temple to Maitreya was
dedicated in his honour with a long inscription in Chinese.
Another line of Chinese rulers,132 with the family name of
Chiu, established itself at Kao-chang in A.D. 507, and one of
them married a Chinese princess. Turfan continued to pay hom
age to the Tang dynasty but later on missions from this place
stopped going to China causing suspicion and alarm. The Chinese
forces, therefore, destoryed the kingdom in 640, as is recorded
by Hsuan-tsang133 who earlier on his outward transit visit had
good reception from the king of Kao-chang. The political and
cultural life of the kingdom, however, was not disturbed by the
event. It was conquered later on by the Uighurs at an uncertain
date. They had no doubt established themselves there in the
132. Bagchi mentions the names of the Kiu (Chiu) dynasty as known
from different sources. These are Kiu-Kiu (497-520), Kiu-Kuang (521-30),
Kiu-Kien (531-47), Kiu-Hiuen-li (548-54), Kiu-Meon (555-60), Kiu-Han-Ku
(561-601); Kiu-Po-Ya (602-623), Kiu-Wen-tai (624-40) and Kiu-Che Mou
(640-?). In 640 Turfan was brought under direct Chinese control.
133. In his biography there is a description of his reception by the king
of Kao-Chang on his outward journey. But in the account of his travels written
after his return he speaks of the city as no longer existent. (For the account
of king’s reception, see Beal : ‘The Life o f H iuen-tsangLondon, 1911, p. 37.)
eighth and ninth centuries. About 750, their Khan adopted

Manichaeism as the state religion. The finds of manuscripts in
Sogdian134 and other Persian dialects at Turfan suggest its close
connection with the west. Tibetan influence also affected Turfan
in the eighth and ninth centuries since many Tibetan documents
have been found here. About 843, this Uighur kingdom was
destroyed by the Kirghiz.135 Buddhist priests seem to have been
massacred, as indicated by vaults filled with skeletons
still wearing fragments of the monastic robe. Buddhism was not
extinguished by this event, but lingered on here longer than in
other areas of the Tarim basin. Even in 1420, the people of
Turfan were Buddhists and according to the Ming Annals, there
were more Buddhist temples than dwelling houses at Huo-chou
(or Kara-Khoja).
Among the states of the Tarim basin whose history and per
sonal relations have been sketched, Kucha seems to have played
the most important role for nearly a thousand years. In their
relations with China, while Turfan was nearest to bear the Chinese
onslaught, Kucha seems to have offered stiff resistance. The
Kucheans and their closeallies Aksu and Uch-Turfan had many
things in common, besides Buddhism, customs, literature and
language. The Kuchean rulers are called ‘Po’ meaning white in
Chinese, thereby pointing to their fair complexion. The names of
the Kuchean rulers in the Tang period are also Sanskritised.
The old Kuchean documents mention king Swarnate (Suvarna-
datta), transcribed into Chinese as Su-fa-lie and Ho-ti-pu-she-pi
(Haripuspa). The Chinese pilgrim Hsuan-tsang also noticed
Gold Flower (Skt. Suvarnapuspa), a former ruler of Kucha. The
names of individuals in ancient documents are also Sanskritised
such as Wiryamitre (Viryainitia), Wiryascne (Viryasena), Jnana-
sena, Moksacandra etc.136 This phenomenon must have been
the result of Buddhist influence as also of Indian culture. There
might have been Indian immigrants or their descendants in this
part as well like those in Khotan.
134. Stein. Ruins o f Desert Cathay. Op. cit. 11. p. 359.

135. Eliot. Op. cit. HI, p. 207.
136. Bagchi. Op. cit, p. 77.
The Regality & Buddhism in the Northern States

The exact date of the introduction of Buddhism in the Tarim
basin kingdom is not known, but it could not have been later
than the end of the 1st century A.D. A fuller picture of Buddhism
in Kucha is available from about the third century onwards.
According to the annals of the Tsin dynasty (265-316) there
were a thousand Buddhist stupas and temples in Kucha. Po-yen,137138
probably a member ofthe royal family, became a Buddhist monk
and went to China in 256-60, and translated 6 Buddhist texts
into Chinese in the famous Buddhist temple of Po-ma-sse at Lo-
yang. Another Kuchean Po-£rimiira13S went to China during
307-12 and then moved southwards He translated Buddhist
texts between 335-342. Another Kuchean prince named Po-yen
had gone to Leang-chou in 273. He was a Buddhist scholar of
repute but no information is available regarding his literary
contributions. A good bit of information is available about
Buddhism in Kucha in the fourth century A.D. In one of the
Chinese texts139 of this period, reference is made to Kucha as a
Buddhist city and the ruler’s palace appearing like a monastery
with standing images of Buddha carved in stone. The number of
Buddhist monasteries and monks residing there are also record
ed, as for instance Ta-mu had 170 monks. Po-shan hill also called
Che-hu-li had 50 or 60 monks and the new monastery of the king
ofWen-Su (Uch-Turfan) contained 7o monks. They were all
under the direction of Fu-tu-she-mi (Buddhasvamin), a scholar
and follower of the Agamas (Hinayana) school. His talented
pupil Kiu-kiu-lo (Kumara), however, became a Mahayanist.
Buddhist nuns too had their establishments at Kucha. Thus, the
A-li (Aranyaka) monastery had 180 nuns, Liun-jo-kan had 50
137. Po-Yen, a Sramana of the western region, translated some Sutras

in the White Horse monastery at Lo-yan in A.D. 257. (Nanjio—Catalogue.
Op. cit, Appendix II No. 16, p. 387)
138. Poh-Sh-le-mi-to-lo-ye-£rimitra literally meaning a lucky friend
was a sramana of the western region who was an heir-apparent of a king of
the country, but gave up his realm to his younger brother, and became a
monk. He came to China in the Yun-Kia period A.D. 307-312 under the
western Tsin dynasty. (Nanjio—Op. cit, ibid No. 36, p. 398)
and A-li-po had 30 nuns. These nunneries were under the direc
tion of Buddhasvamin, and the members were all drawn from
royal families and households of nobles of the countries to the
east of the Pamirs i.e. of the Tarim basin. The identity of Ku-
mara with the famous savant Kumarajiva140 is certain, and his
life history and contributions to Buddhism demand fuller and
separate study in the text chapter covering ‘Buddhism and Bud
dhist savants of Central Asia’. The monastery where Jiva, mother
of Kumarajiva, retired as a nun was called Tsio-li about 40 li to
the north of Kucha. It was here that Jiva learnt the language of
India, might be Sanskrit. Kumarajiva was at that time seven
years old; and the monastic environment had great impact on
his personality as also acquisition of knowledge and breadth of
learning. He was responsible for introducing Mahayanism in the
countries of the Tarim basin and also in China, rather in an
authoritative manner. He was one of the greatest exponents of
this school of Buddhism and also of the Madhyamika philosophy.
His compatriot Vimalaksa141 later on joined Kumarajiva in China
in 404.
Another scholar from India, Dharmagupta,142 visited Kucha
towards the end of the sixth century and stayed there in the
king’s monastery for two years enjoying the patronage of the
ruler who too was a great believer in Mahayanism. He also
taught various Sastras in Kucha, including logic {tarkasastra)
before leaving for China. Hsuan-tsang’s description of Kucha
and its Buddhist monasteries has already been recorded. The
140. Kumarajiva—called Thun-Sheu— meaning ‘boy age’ ‘a longevity’—
was an Indian Sramana whose forefathers were successively ministers of the
country. A fuller account of Kumarajiva’s life and literary contributions is
provided by Nanjio (Op. cit, II. 59, 406 ff) and also in Encyclopedia o f Religion
and Ethics.
l4 j. Vimalak$a, translated in Chinese Wu-Keu-Yen meaning ‘without
dirt-eye’ was a Sramana of Kublia (Kabul), who was a great teacher of Vinaya
in Kwei-tsz i.e. Kharakar or Kuke where Kumarajiva was one of his disciples.
He arrived in China in A.D. 406. (Nanjio— Op. cit No. ii, 44, p. 400)
142. Ta-mo-Kiu-to i.e. Dharmagupta whose name is translated Fa-mi
literally ‘law-secret’ or ‘law-repository’ was a sramana of the Lo-lo country or
State of southern India (Nanjio—Op. cit. ii. 131. p. 434). Dharmagupta passed
through Kucha about 584 and has referred to the king favouring Mahayanism
(Levi. J.A. 1913. II, p. 348 quoted by Eliot Op. cit, p. 204).
Chao-hu-li (same as Tsio-li of other accounts), about 40 li to the

north of Kucha where Jiva also studied, contained a beautiful
image of Buddha. The Chinese pilgrim saw two standing images
of Buddha about 90 feet high outside the western gate of the city.
The quinquennial assemblies used to be held in front of these
images. Hsuan-tsang mentions143 another famous monastery
called A-She-li-ni which was nearby, and was the meeting place
of Buddhists from all countries. The chief priest of this monastery
was Mo-cha-kiu-to (Mok$agupta) who commanded respect from
all for erudite learning and ability. He had travelled in India for
twenty years and had also studied there* According to his version
given to the pilgrim, the Buddhist libraries of Kucha were well
equipped and contained valuable works such as Sariiyukta-hridaya,
Abhidharma-Kosa and Vibhasa. Hsuna-tsang also describes the
annual car festival when images of Buddha used to be carried in
procession, with thousands of people participating in it, including
the kings and nobles, who had great respect for Buddha and the
Buddhist savants. Religious fasts were also observed by one and
all and there were discourses on sacred teachings. Buddhism
flourished in Kucha till the eighth century. When Wu-kong144
visited Kucha in 751, he met a Buddhist teacher named Wu-ti-ti-
su-yu (Utpal-tsirawne—Skt. Utpalavirya). He knew a number of
languages—those of India, the countries of the Tarim basin and
also China. At the request of Wu-kong, he translated Dasabala-
sutra and two other works in collaboration with a Khotanese
monk, Siladharma. The pilgrim also furnishes information relat
ing to Buddhism in the two states of Karasahr and Aksu* He
mentions about 10 monasteries and 2,000 monks in Karasahr,
143. For Hsuen tsang’s description see Beal. Op. cit, Vol. I, pp. 19 ff.
144. The last Chinese traveller to come to India in the Tang period,
Wu-Kong left China on an official mission to escort the Indian ambassador,
who had come to China from the kingdom of K&pisa. He passed through
Kucha, Kashgar, Shignan in the Pamirs and Wakhan, and reached the Indus
by the route of Yasin and Gilgit valley. He first came to Uddiyana and Gan-
dh&ra which were under the kingdom of KapiSa in this period. Here he was
converted to Buddhism. A short account of his travels has been preserved
(Bagchi. India and China, Op. cit., p. 78). This person Wu-Kong was also
known as Dharmadhatu who remained some time in India, took the vows
and ultimately returned to China with many books and relics (Eliot. Op. cit.
III. 262).
all belonging to the Hinayana school of the Sarvastivadins. The

number of monasteries was the same in Ak-Su but the monks there
were only a thousand. The monks of both the States as well as
of Turfan were dependent on Kucha for religious leadership in
the early period, but towards the end of the 5th century, with
the establishment of the Chinese dynasty, the Chinese influence
was more apparent. Buddhism continued to be the religion of the
people till the middle of the seventh century in Turfan as record
ed by Hsuan-tsang and enjoyed the hospitality of king Wen-tai in
630. There he came into contact with a Chinese Buddhist scholar
who had studied at Chang-ngan. Hsuan-tsang stayed at Turfan
and discoursed on the doctrine of the Prajnaparamita. The Chi
nese inscriptions of this period from Turfan provide good bit of
information about the influence of Buddhism on the life of the
people.
Buddhism flourished under the Uighurs as well. The Turks
were equally familiar with Buddhism in Tokharestan and some
of the rulers showed great interest in it. The Uighur Turks
inherited that tradition and assimilated a good deal of Buddhist
culture from the people of the northern area of the Tarim basin
of Turfan, Karasahr, Kucha and Aksu, with whom they were in
close contact in the 7th century. With the foundation of the new
empire with Turfan as its metropolis in the 9th century, the old
civilisation of Kucha-Karasahr had no doubt disappeared but
the Uighurs had assimilated much of it. Buddhism continued to
have its due place under the Uighurs despite the state patronage
accorded to Manichaeism in this period. A large number of
Tokharian Buddhist texts were translated into Uighur, forming
the oldest literature of the Turks. Most of the Buddhist sites in
Turfan belong to the Uighur period (750-550) of Central Asian
history. Bezeklik contains the largest Buddhist sites of Turfan.
There are cave temples and beautiful frescoes and numerous
Buddhist shrines. The Buddhist sites of the Tarim basin (espe
cially Bezaklik) are replete with stucco figures and frescoes,
suggesting the Far Eastern expression of the Greco-Buddhist
art of Gandhara. The ancient sites in Kucha such as Kizil,
Kumtura, Duldur, Akhur etc. contain vestiges of the ancient
Buddhist art rather of a mixed character.
In a review of the political history of Central Asia in the
pre-Muslim period it may be suggested that the region lacked

indigenous national unity though sometimes the Tarim basin
was united under foreign rule. Cities or groups of towns, sepa
rated by deserts had their own way of life with political inde
pendence under native rulers. Chinese, Turks or Tibetans
stationed troops in these places and appointed residents to
supervise the collection of tribute. The main cities or oases were
Kashgar in the west, Kucha, Karasahr, Turfan (Idiqutshahri,
Chotsche) and Hami lying successfully to the north-east, and
Khotan and Miran to the south-east. Inter-State confrontations
were frequent though confined to neighbouring ones, and the
interference of foreigners was not ruled out. Economic and
political factors resulting in the clash of tribes and the displace
ment of the vanquished one necessitated migratory movements.
These took the form of political upheavals and consequent
formation of new kingdoms and empire. The roles of the Hiung-
nu, and Yueh-chi are notable in this direction. The Chinese
participation in the political affairs of Central Asia and its state
was fairly active all through this period except when confronted
with superior force or dissipated by internal troubles. The
Chinese were equally interested in their trade with the west and
wanted to have friendly relations with the States through which
the route passed. Their concept of friendship was determined
by the despatch of missions to the Imperial court with tribute
and presents. Sometimes even royal princes were left there for
Imperial service. The states were no doubt vulnerable to external
pressures and there were cases of proteges of foreign powers
ruling there. Thus, we find Kashgar becoming a Chinese pro
tected state in the first century B.C. and on the relaxation of
Chinese hold about the time of the Christian era, it was subdued
by the neighbouring kingdom of Khotan. Chinese supremacy
was restored with the conquest of Pan-chao, but the Yueh-chi
interfered in the internal politics of Kashgar and placed a prince
of their choice on the throne. Kaniska is also credited with the
conquest of Kashgar and Khotan, and Hsuan-tsang refers to
the princes of this state being kept as hostages. The Hephtha-
lities, in turn, incorporated Kashgar in their dominions. The
reference to the Indian rulers of Khotan with the prefix added
to their names suggests Indian colonisation which is recorded in
the Tibetantraditions. The history of the other kingdoms further

east in the Tarim basin was in no way different except that those
nearer to the Chinese border had greater control of the latter.
Turfan in this sense was more receptive to Chinese influence—
political as well as cultural. Kucha and the nearer states had
much in common.
In this context the role of Buddhism and the part played by
the royalty in its propagation are equally important. In fact,
the religion of the Tathagata was a cultural integrating force in
Central Asia, despite the fact that according to Hsuan-tsang’s
religious conspectus of these regions, Kashgar, Osh and Kucha
belonged to the small vehicle of the Sarvastivadins, while
Yarkand and Khotan mainly Mahayanist. The small vehicle also
flourished at Balkh and at Bamiyan. In Kapisa the Great
Vehicle was predominant and there were many Hindu sects. The
royalty played an active part in the propagation of Buddhism
and also participated in Buddhist cultural activities. In Kucha,
the royal palace appeared like a monastery with figures of
Buddha everywhere in Hsuan-tsang’s time. The Tibetans and
Uighurs were no exception in their patronage of Buddhism. The
contribution of Buddhist savants of Central Asia in that area as
also in China where they translated many canonical texts into
Chinese demands proper study in a separate chapter.
CHAPTER III
BUDDHISM AND BUDDHIST SAVANTS OF

CENTRAL ASIA
A comprehensive study of Buddhism in Central Asia involving

its introduction, expansion and branching off into several sects
presupposes an investigation into the source material—literary,
epigraphic, travel accounts as well as monumental remains of
ancient Buddhist sites in this vast region. It is not a uniform
phenomenon as different areas of Central Asia received the
message of Tathagata in different periods and, of course, through
manifold channels—royal emissaries, individual scholars and
even the trading and mercantile class who might have contributed
materially towards its expansion. The zeal and curiosity of the
indigenous element might as well have been a contributory factor
in the expansion of Buddhism which, in a couple of centuries,
became very popular not only in Central Asia but also in China,
and thence eastwards in Korea and Japan. A long history of
Buddhism and its expansion, however, demands regional study—
namely that of the western region comprising Afghanistan,
Bactria and Parthia, the central zone covering the present
Russian Turkestan, and the eastern one assimilating both the
northern and southern wings of Chinese Turkestan. Central Asia
in general played an active role through its savants in the trans
lation of Buddhist sacred texts into Chinese as also in other
languages of the countries of its adoption. A long list of such
scholars who worked in this direction is available from the
catalogue of Chinese Buddhist Tripitakas in different periods.
A study of the list would bring out the fruitful contribution of
Buddhist savants from other parts of Asia including India in
different centuries. Thus, according to the list of Nanjio,1 among
1. See : A Catalogue o f the Chinese Buddhist Tripitaka, Oxford (1882),
Appendix II, pp. 379 ff. In his Handbook o f Chinese Buddhism, Earnest J.
Eitel refers to the contributions of different scholars in this context (Reprint—
Buddhism and Buddhist Savants o f Central Asia 87
72 translators who worked between 67 and 420 there were 15

Indians, 7 Yueh-chi, 5 Parthians, 7 Kubhans, 21 from the western
countries and 17 Chinese; among them 22 worked in the south,
of whom 5 were Chinese. Among 43 workers who worked
between 420 and 550, 14 were Indians, 10 from Kubha and other
western countries, 4 Sinhalese and Indo-Chinese, and 4 of un
certain origin. Among them 27 worked in the south. A close
study and scrutiny of the list would reveal the contribution of
Parthians and these from Kubha (Kabul) as eminent Buddhist
scholars who went to China for Buddhism—its propagation
through translation of sacred texts. The western countries
include Central Asia and comprised of both the modern sectors
of Russian and Chinese Turkestans. The two areas, however,
demand separate treatment because of political considerations as
also their separate archaeological investigations.
The history of Buddhism in Central Asia is also related to the
political history of different regions and the hold of different
political powers in later times, like that of the Chinese, Tibetans
and Uighurs. The languages of these people have also left literary
remains in Central Asia, which no doubt call for separate
treatment. A brief notice of the languages represented in the
manuscripts and inscriptions discovered will reveal many influ
ences at work in Central Asia, and its importance as a receiving
and distributing centre. Numerous Sanskrit writings have been
found, all dealing with religious or quasi-religious subjects, as
medicine and grammar. Relatively modern Mahayanist literature2
Cosmos Publications, New Delhi, 1981; original edition 1888). This calculation
is recorded by B.A. Litvinsky in his booklet entitled ‘Outline History of Buddh
ism in Central Asia’ prepared and circulated at the Diushandbe Session
(1968) of the International Conference on the History, Archaeology and
Culture of Central Asia in the Rushan Period, p. 13. Brough, J. suggests that
certain Yueh-chih translators may have come not from the land of the Great
Yueh-chih, but from the milieu of the Eastern Turkestan ‘Little Yueh-chih’
(quoted note 63, Op. cit, p. 83).
2. Central Asia, according to Eliot, has astonished the learned world
with two new languages, both written in a special variety of the Brahmi alpha
bet called Central Asian Gupta. One is called Nordarish, regarded by some
as the language of the Sakas, and by others as the language of the Ku$anas
and of Kani§ka’s empire. The Mahayanist literature translated into this
language include the Suvarriaprabhasa Vajracchedika and Aparimatyus-sutras.
It appears to have been spoken principally in the southern part of the Tarim
is abundant. Unknown forms of Prakrit are not unknown, while

two new languages, written in Central Asian Gupta script,
provide translations of Mahayanist literature. Three other
Iranian languages, all written in an alphabet of Aramani origin
have left literary remains in Central Asia. Of the three, Iranian
language, Sogdian with a more varied literary content also otters
Buddhist texts, besides later Manichaen and Christian ones. It
was originally the language of the region round Samarkand, but
was used by merchants throughout the Tarim basin and spread
even to China.*3
As the Tibetans were the dominating power in the Tarim basin
at least from the middle of the eighth until the middle of the
ninth century, it is not surprising to record finds of Tibetan
manuscripts in the regions of Khotan, Miran and Tun-huang. In
Turfan, traces of Tibetan influence are fewer though not entirely
absent. The documents include Buddhist translations4 besides
numerous official and business papers. At that time Buddhism
seems to have shared with the Bon religion the allegiance
of the Tibetans. Of course, there are no Manichaean or Chris
tian translations in Tibetan. A large number of Chinese texts—
both religious and secular—from the principal centres of Central
Asia, including a series of dated documents ranging from 98
B.C. to 153 A.D. point to intercourse between China and
Central Asia at this period. Some documents of the Tang dynasty
are Manichaean with an admixture of Buddhist and Taoist ideas.5
Basin (Hoernle in JRAS, 1910, pp. 837 ff and 1383 ff; 1911 pp. 202 ff, 447ff.)
The other language spoken principally on its northern edge is called Tokharian,
associated with the Tokharas or Indo-Scyths. According to Eliot, it is safer
of it as the language of Kucha or Kuchanese. It exists in two different dialects
known as A and B whose geographical distribution is uncertain. Numerous
official documents dated in the first half of the seventh century show that it
was the ordinary speech of Kucha and Turfan. It was also a literary language
and among the many translations discovered are versions in it of the Dharma-
pada and Vinaya. (Eliot. Hinduism and Buddhism—Op. cit III, pp. 190-91 with
references to many articles including that of Sylvain Levi—‘Tokharian B,
Langue de Koutche’ JA. 1913. II. p. 311)
3. Eliot : Op. cit, pp. 191-92.
4. Aurel Stein. Ancient Khotan—Op. cit, Appendix B—entitled Tibetan
Manuscripts and Scraffiti discovered by M.A. Stein at Endere—edited by
L.D. Barnett and A.H. France.
5. Eliot : Op. cit, p. 193. See also Chavannes—Les documents Chinois
decouverts par Aurel Stein* 1913; also Appendix A entitled ‘Chinese Docu-
Besides these languages and dialects used as mediums of Buddhist

literature in Central Asia, another dialect Uighur8 was used for
Buddhistliterature.lt spread considerably when the Uighurs
broke the power of Tibet in the Tarim basin about 860 and found
ed a kingdom themselves. It extended into China and lasted
long.
Buddhism in Afghanistan, Bactria and Parthia
Monumental remains, as also reference to Buddhist savants in
Chinese Buddhist literature point to the flourishing state of Bud
dhism in this part of Central Asia; and Alberuni67, ‘too, does not
fail to record this fact. In his words, ‘in former times, Khorasan,
Persia, Iraq, Mosul, the country upto the frontier of Syria, was
Buddhistic, but then Zarathustra went forth from Adharbayjan
and preached Magism in Balkh (Baktre)’. His doctrine came into
favour with king Gushtasp, and his son Isfendiyad spread the
new faith both in East and West, both by force and by trea
ties. He founded five temples through his whole empire, from the
frontiers of China to those of the Greek Empire. The succeeding
kings made their religion (i.e. Zoroastrianism) the obligatory
state religion of Persia and Iraq. In consequence, the Buddhists
were banished from those countries, and had to emigrate to the
countries east of Balkh. He also refers to the finds, of Buddhist
idols, and their remnants from India, China and among the
Toghuzghus. The people of Khorasan call them ‘Shamanan’ or
(Sanskrit, Sramana) while the shrine known as vihara could be
seen in the border district of Khorasan adjoining India. This is
the impression about Buddhism and its monuments, as recorded
by the talented Khoresmian Alberuni (973-1048), the author of
Tahkik-i-Hind (Inquiry into India).
ments from the sites of Dandan Uiliq, Niya and Endere’ translated and
annotated by Eduvard Chavannes in Ancient Khotan—Op. cit, pp. 521 if.
6. The name Uigur is perhaps more correctly applied to the alphabet
than the language which appears to have been the literary form of the various
Turkish idioms spoken north and south of Tienshan. The use of Uigur in
China is evident from the printing of Sutras in Uigur at Peking in 1330, and
Uigur manuscripts were copied in the reign of Kang Hsi (1662-1773) (Eliot.
Op. cit, p. 192).
7. ‘Alberuni’s India’—Translated by F.C. Sachau, Vol. I. London, 1888,
p. 21.
The history of Buddhism in this part of Central Asia might

be dated-from the time of the Indo-Bactrians, although the tea
chings of the Lord might have found their way even earlier. In
this context, the Kandahar bilingual edict8 of ASoka is rightly
considered as an eloquent testimonial to the extension of Bud
dhism in the direction of Central Asia. The finds of other inscrip
tions in Afghanistan suggest that this part of Central Asia had
come within the range of Mauryan cultural activities, it being a
part of the empire of ASoka. Buddhism as such must have had
a substantial following in the Kandahar area of what is now
southern Afghanistan by the middle of the 3rd century B.C. It is
proposed by Bagchi9 that Buddhism was introduced in Balkh in
the time of Asoka who, no doubt, speaks of his efforts to in
troduce his dhamma among the peoples of Gandhara, Kamboja
and Yona. The three peoples were neighbours, Kambojas were
probably a branch of the Tukhara people, while the Yonas were
no doubt the Bactrian Greeks. In this context a legend recorded
by Hsuan-tsang10 refers to the first two lay disciples of Buddha,
Trapusa and Bhallika as responsible for introducing Buddhism
in that country. Originally these two were merchants of the king
dom of Balhika, as the name Bhalluka or Bhallika probably
suggests the association of one with that country. They had gone
to India for trade and happened to be at Bodhgaya when the
Buddha had just attained his enlightenment. Offering him cakes
and honey out of their provision they were also ordained by the
Tathagata as his first two disciples. The progress of Buddhism to
the north of Afghanistan is also borne out by a KharosthI ins
cription on a clay object recovered from the Begram excavations
8. J. Harmata. ‘Sino-Indica’—‘Acta Antiqua’, Vol. XII, No. 1-2, Buda

pest, 1964, p. 4.
9. India and Central Asia—Op. eft, p. 32. Foucher, however, holds that
Buddhism did not appear in Bactria until the end of the 1st or the 2nd century
A.D. (La Vielle route de Vlnde de Bactres a Taxila, Vol. II, Paris, 1947, pp.
280-84). Litvinsky, however, considers Foucher’s view to be erroneous (Out
line History o f Buddhism in Central Asia, Dushanbe, 1968, p. 77 n.15).
10. T. Watters. 'On Yuan-Chwang’s Travels in India' (Photoprint edition,
New Delhi, 1973, Vol. I, 112; II 131).
in the first layer, placed between the third and second centuries
B.C. recording a Buddhist name.11
The foundation of the Greco-Bactrian state no doubt contri
buted to the expansion of Buddhism in this region in the first
two centuries preceding the Christian era. Both Demetrius
and Menander were interested in Buddhism.12 The former is
supposed to have attacked India to punish the Brahmin Sunga
ruler for his anti-Buddhist activities and persecution of Buddhist
monks, while the latter seems to have embraced the religion of the
Tathagata after his discourse with the Buddhist philosopher
Nagasena. The Milindapaiiho or ‘Discourses with Menander’ is the
theme of the great work. This is both in Pali and also in Chinese
translation.13 Its composition might have been of a later date,
probably after Menander’s death. It is now generally accepted
that this Indo-Greek ruler was a devotee of the Tathagata, and
certain symbols like the stupa on the coins of Agathocles, as
also the legends on the coins of Menander are in agreement with
the theory of infiltration of Buddhism in the realm of the Indo-
Greek or Bactrian rulers.14 It seems fairly evident that the
Greco-Bactrian kingdom unified in one political state the north
Indian regions, Afghanistan and several parts of western Turkes
tan. The political atmosphere was, thus, congenial for the
Buddhist missionaries as also for the local followers in taose
11. J. Harmata. Sino-Indica ‘Ariana Antiqua’, Vol. XII, No. 1-2, Buda
pest, 1964, pp. 4-5.
12. L. Levi. LeBouddhisme et les Grecs—Revue de VHistoire de Religions,
XXIII, 1891 pp. 436-49; reprinted in Memorial Sylvain Levi, pp. 204
ff. The Questions o f King Milinda SBE, XXXV; P. Denievitte. Les versions
Chinoises de Milindapahho BEFCO, XXIV, pp. 168 ff.
13. The Chinese version of the Milindapahha, written several centuries
later, describes Alasanda as 2,000 yojana from Sagala instead of 200(Demie-
ville—Op. cit; cf. Pelliot. J.A. 1914, Pt. II, pp. 413-19; Levi. IHQ. 1936,
p. 126).
14. The Pali texts represent the Greeks as taking part in missionary
activities. We are told that after the conversion of Yavana (Greek) country
to Buddhism, Moggoliputta Tissa went there and selected an elder Dharma-
rak$ita for missionery work. He was then sent to Aparantaka where he made
a large number of converts. The Greeks were also represented by this Elder
Mahadharmaraksita at the Great StQpa ceremony in Ceylon, initiated by
Dutthagamini in the middle of the second century B.C. (Levi. Op. cit).
areas to convey the message of the Lord outside the boundaries

of this kingdom.
Further information regarding the association of Indo-Greek
rulers as also the Yonas with Buddhism between the second and
the first century B.C. is also available from other sources. Two
KharosthI inscriptions15—one from the ancient country of Udyana
(Swat valley) and the other from Bajaur (south-east of Jalalabad)
record dedications by a Greek officer named Meridarkh Theo-
dorus who enshrined the relics of the Lord Buddha, and the
pious act of one Viyaka-mitra—an apracharaja (Skt. apratyag-
raja) respectively- The former was an officer of rank and the latter
was a vassal—both probably under Menander or his successor. It
is also proposed that the wheel on some coins of Menander is
connected with Buddhism.16 Theassociation of these Indo-Greek
rulers—Demetrius and Menander—was not a formality but an
article of faith. According to a tradition cited by Taranath, Mena
nder’s association with Buddhism is independent of the Milinda-
pahho. He refers to king Minara in the land of the Tukharas,
who is identified with Menander by Lassen.17 Further, theassocia*-
tion of Greeks with Buddhism in the north-west area in the
early first two centuries preceding the Christian era, is also evinc
ed from the Ceylonese chronicles. It is said that when Duttaga-
mini founded the Great stupa (Mahathupa) in Ceylon, sometime
in the middle of the second century B.C. he invited Buddhist
teachers from various countries. The elder Mahadharmarak$ita
represented the Greeks who came from the city of Alasanda.18
15. Sten Konow. Corpus Inscriptionum Indicarm (CII) II (i),
These are discussed by K.A. Nilakantha Sastri: A Comprehensive History
o f India, p. 219.
16. J. Marshall : Taxila, 3 Vols, Cambridge, 1951, Vol. I, pp. 33-34;
Narain : Indo-Greeks—Op. cit, p. 98. According to Tarn, the idea that
Menander ever became a Buddhist in the sense of entering the order may be
dismissed at once (Op. cit, p. 268).
17. This tradition as recorded by Taranath, refers to king Minnara in the
land of the Tukharas, and he is identified by Lassen with Menander (Narain,
98 n).
18. According to the Mahavamsa, the Ceylonese king Du((agamini whose
regnal years are assigned by scholars between 101 -77 B.C. marked the laying
of ‘Great Stupa’ by a huge celebration attended by a large number of bhiksus
from many foreign lands (XXIX-29). These included wise Mahadeva from
Pallavabhoga with 460,000 bhik$us, and Yona mahadhammarakhita with
This place is mentioned in the Milindapahho as a dvipa at a dis

tance of,200 yojanas from Sakala (Sialkot)and it was in the village
of Kalasi in this dvipa that Menander was born. Both Alasandas
seem identical and are supposed to be located at Charukar
between Panjshiv and Kabul rivers.19 Tarn, however, identifies
the Alasanda of the Milindapahho with Alexandria of the Caucasus,
and not with Alexandria in Egypt, as presumed by some French
scholars. Buddhism, thus, appears to have flourished in Bactria
and the neighbouring regions, and its savants could attract the
attention of those interested in Tathagata and his religion even
in far off places like Ceylon. Coedes also records a local tradition
connecting Menander with the origin of the famous statue of
Buddha in Indo-China, the statue of Buddha of the Emerald
produced by supernatural magical power by Menander’s teacher
Nagasena-
•Besides Bactria and its rulers, who it seems experienced the
earliest infiltration of Buddhism in their areas, the Yueh-chi
equated with the Tusaras and their later homeland Tokharistan
also had the impact of this religious discipline more or less at the
same time- The first inflow of Buddhist culture to China was
from Tokharestan- It was in the year 2 B.C. that the Chinese
ambassador Tsing-Kiang received Buddhist texts as presents to
the Chinese court.20 The passage recording this incident in the
‘Wei-Lio’, a third century source, has received different inter
pretations by the sinalogists. While the date is accepted as
2 B.C-, it is suggested that the Yueh-chi crown prince acquaint
ed the visitors with Buddhism. Some scholars, even question the
historical authenticity of this account, but the role of the Yueh-
^chi in popularising Buddhism in China is not disputed. This
30,000 bhik$us from Alasanda, the city of the Yonas (ibid, XXIX, 38-39).
While the number of guests seems to be exaggerated, the participation of
these foreigners might be accepted.
19. ‘Quarterly Journal of the Mythic Society’ (QJMS. LXXI, p. 180;
Tarn : Op. cit, p. 141). The Mahavamsa calls it a city of the Yonas (Yona-
nagaralasanda—XXIX, 37.). According to the Milindapahho, the town is
mentioned as situated on an island in the Indus (III.7.4).
20. Chavannes. Les documents Chinois decouverte par Aurel Stein, 1973,
Introduction. The earliest documents are of 98 B.C. See Eliot : Hinduism
and Buddhism—Op. cit. Ill, p. 245.
must have been at the time of the composition of the ‘Wei-lo’ in

the third century A.D. or again the early fifth century, when the
commentaries on it appeared. The role of the people and savants
of Tokharistan in the dissemination and elaboration of the
Buddhist religion in the early centuries of the Christian era is also
evident from other sources.
The famous scholar Ghosaka, born in Tokharistan, played a
prominent part at the fourth Buddhist Council in Purushpura.
He was the author of the commentary on the Abhidharma-
Vibhasa which was compiled there. This theologian also compos
ed an original treatise on Abhidharma—the Abhidharmdmrita—
which is preserved in the Chinese translation of the third century,
being one of the most clear expositions of the Abhidharma doct
rines. Ghosaka had returned to Turkestan and was associated
with the Vaibhasika school, later on divided into branch schools.
One of such branch was connected with the country of Balhika
or Balkh and was called the western Vaibhasika school. The tra
dition of this school could be traced to Ghosaka. The impor
tance of this school in Tokharistan is revealed from the first
translation of its treatises into the Tokharian language by a local
monk named Dharmamitra of Tarmita or Termez. The Vaibha
sika school, a branch of the Sarvastivada sect of Buddhism was
widespread in western Turkistan, although some important ele
ments in the Vaibhasika doctrine are supposed to have brought
it closer to the Mahayana school, paving the way for its intro
duction and flourishing nature in Khotan. Balhika or Balkh was
a great centre of Buddhist studies even in later times. The Nava-
sangharama of this place, according to the Chinese pilgrim
Hsuan-tsang, was the only Buddhist establishment north of the
Hindukush in which there were a constant succession of Masters
who were commentators of the canon. The importance of the
Vaibhasika school of the Tukhara country is also evident from
the finds of Tokharian literature to which fuller reference would
be made later on. Aryacandra, who first translated the Maitreya-
samiti into the Tokharian language, was a Vaibha?ika.
The role of the Tokharian Buddhist monks in the propagation
of Buddhism in China was consistent. The Chinese literature
distinguishes these monks by prefixing the word Che (from
Yueh-che) to their names. It is said that the first two mission
aries to China in A.D. 68, Kasyapa Matanga and Dharmaraksa,

were met by the Chinese ambassadors in the country of the
Yueh-chis.21 The Chinese refer to a monk called Chu-Fan-lau
who came from Central Asia and found some difficulty in obtain
ing permission to leave his country. He therefore followed shortly
afterwards. Both were installed at Loyang, the capital of the
dynasty, in the White Horse Monastery, so called because the
foreign monks rode on white horses or used them for carrying
books. The story does not state that there was no Buddhism in
China before A.D. 62. On the contrary it only implies that it
was not sufficiently conspicuous to the Emperor. It is very likely
that Buddhism entered China half a century earlier. While poli
tical relation and the despatch of a Chinese mission to the Cen
tral Asian power of the Yueh-chi (129-119 B.C.) under Chang-
Chian could be traced to the second century B.C., and the
documents discovered by Aurel Stein in Central Asia confirm
China’s communition with this part of Asia, there is no reference
to Buddhism in any of these source materials. The notice in the
Wei-lue is the earlier reference to Buddhism in B.C. as pointed
out earlier. The Later Han Annals also intimate that in A.D.
65, the Prince of Chu was a Buddhist and that there were Srama-
nas and upasakas in his territory.
Tokharistan seems to have played a consistent and steady role
in propagating Buddhism in China in the centuries following. A
monk of rare learning named Lokasema of Tukhara origin went
to Loyang in A.D. 147 and translated there some of the most
21. The traditional date for the introduction of Buddhism is A.D. 62,
when the chronicles tell how the Emperor Ming-Ti of the Later Han Dynasty
dreamt of seeing a golden man fly into his palace, and how his courtiers
suggested that the figure was Fo-to or Buddha, an Indian God. This was
followed by the despatch of an embassy in 65 to the kingdom of Ta-Yueh-
chih or India with instructions to bring back Buddhist scriptures and priests.
On the return journey it was accompanied by a monk called Kasyapa Matahga,
a native of Centra! India, followed later on by another person Chu-Fa-lan
from Central Asia. The story does not state that there was no Buddhism in
China before 62 A.D. According to Wei-Iueh, which gives a brief account of
the Buddha’s birth in the year 2 B.C., an ambassador sent by the Emperor
Ai to the court of the Yueh-chih was instructed in Buddhism by order of their
kings. Also in 65 A.D. The Prince of Chu was a Buddhist and there were
sramanas and upasakas in his territory. (For the account of the Chinese
mission, see—BEFCO, 1910; also Eliot : Op. cit III, 244-45.)
important texts of the Buddhist canon into Chinese. He worked

at translations till A.D. 186. There are twelve translations ascrib
ed to him.22 Towards the end of the same century (190-220 A.D.),
one of his young disciples, Che-kien, also, of Tukhara origin, was
working in North China, but had to leave on account of politi
cal trouble and settled down at Nanking where he worked till the
middle of the 3rd century A.D. He translated over a hundred
Buddhist texts—yet another Buddhist monk named Dharmaraksa
(Chinese name Fa-hu) was born of a Tukhara family. He had
settled down in Tun-huang towards the middle of the 3rd century
A.D. and after travelling far and wide in Central Asia, had pick
ed up 36 different languages. He went to China in A.D. 284 and
worked there till 313. He translated nearly two hundred Buddhist
texts into Chinese, out of which 90 are still existing. Another
Tukhara monk named She-lun came to China in 373 and trans
lated four works into Chinese. The last Tukhara monk to go to
China, as recorded in the Chinese works, was Dharmanandi23
who went there in 384 and translated a number of works into
Chinese.
While Tokharistan seems to have provided a number of Bud
dhist monks who went to China and translated canonical works
into Chinese, Parthia and Kabul (Kubba) also furnished a num
ber of Buddhist scholars for the cause of Buddhism in China.
That is possible only if the home country itself was in a flouri
shing state. While reference has been made to the region of
Kabul forming part of the territory of Indo-Greeks, Parthia and
Drangiana demand consideration as centres of Buddhist activi
ties in the early centuries of the Christian era. Reference has no
doubt been made to Yona Mahadhammarakhita with 30,000
bhiksus participating in the great celebrations at the laying of
the ‘Great Stupa’ in the time of king Dutta-gamini of Ceylon.
The Mahavamsa24 also refers to the visit of the wise Mahadeva
from Pallavabhogga with 46,000 bhik$u monks. There might be
exaggeration in the number of foreign participants, and the
Buddhist chronicle of the sixth century A.D. must have based its
22. Nanjio. Catalogue—Op. cit, ii. 3. pp. 381-82.

23. Nanjio—Op. cit, ii, 57, pp. 403-04.
24. W. Geiger. Mahavamsa—Op. cit, XXIX, 29.
information on older material—primarily local chronicles, no

longer available. The authenticity of the information according
to Geiger25 need not be questioned. The Pallavas are associated
with Pahlavas i.e. the Parthians. Geiger takes it as a Persian
name. As regards the wise Mahadeva and his bhoggam, he
might have been from a satrapy (bhoga) of the Parthians and not
Parthia itself. As such, this Ceylonese account could refer to
Parthian Buddhism in the first century B.C. According to -
Bagchi,26 real interest in Buddhism is shown by the Parthians
only after its establishment in Tocharestan under the Kusanas.
The Tokharians must have transmitted Buddhism to them.
While there is no trace of old Parthian translations of Buddhist
texts, the Chinese accounts do refer to a number of Iranian
Buddhist scholars who went to China in the 2nd and third cen
turies A.D. and collaborated in the translation of Buddhist texts
into Chinese. In the absence of any centre for Buddhist studies
in Iran, it is suggested27 that the Parthian monks had studied
Buddhism and Buddhist texts in original at the Buddhist centres
of Tokharestan before proceeding to China. It may, however, be
presumed on the basis of reference28 to thePaithian monk An-
Shib-kao, who visited China in the second century A.D. that
Parthia had already a long tradition of Buddhism. It is suggest
ed29 by G.A. Koshelenko that the Parthians made their acquain
tance with Buddhism not later than the beginning of our era,
and he also refers to the appearance of Buddhism in Margiana
in the first century A.D. This could have been possible through
the south-eastern territories of the Parthian kingdom of the 1st
century B.C. at the latest.
25. ibid, pp. ix-x.

26. India and China, p. 30.
27. ibid, p. 37.
28. Nanjio. Op. cit, II.4, pp. 381-82. An-shi-Kao was a prince royal of
the country of An-Si (Eastern Persia or Parthia or Arsak) which character
is affixed to the names of other translators of the same country living in China,
as their surname eg. An-Huen. When An-shi-Kao’s father died, he gave up
his kingdom to his uncle and became a Sramana. He came to China in A.D.
148 and worked at translations till A.D. 170.
29. ‘The Beginning of Buddhism in Margiana’ ‘Ariana Antiqua’ (Hungary),
XIV, fasc. 1-2, Budapest, 1966, p. 180.
The names of Parthian Buddhist monks in Chinese are distin

guished by the prefix An (Ngan) from the'old Chinese name of
Parthia An-She (Arsak), a name given to the country in the
period of the Arsacid dynasty. The Chinese historian mentions
the Parthian prince Ngan-She-kao (same as An-Sheh-kao)
or Lokottama who appeared in the western frontier country
of China with a burden of Buddhist texts, after the fall of the
Arsacidan dynasty in A.D. 148* He is said to have left his throne
in favour of his uncle, and left the family to become a Buddhist
monk at an early age. He worked at translations till A.D. 170.
He personally translated into Chinese more than a hundred
Buddhist texts of which 55 are recorded by Nanjio*30 Many of
these texts are extracts from the Buddhist Agamas, generally
illustrating the fundamental doctrines of Buddhism. Another
Parthian scholar, named Ngan-Hiuan,31 came to Loyang as a
merchant. He received royal favour in the form of military
service with the designation ‘Chief of the Cavalry’. He soon gave
up his official position and became a Buddhist monk—an upasaka
of An-hsi* He collaborated with Yen-Fo-thiao,32 a Sramana (or
an upasaka) of Lin-hwai in China and translated two works at
Lo-yang in A.D. 181. Besides these Parthians, there were some
Sogdian monks belonging to the school of the royal prince
Ngan-She-kao who laid the foundation of a school for systematic
interpretation of Buddhism to the Chinese. The first Chinese
scholar Yen-Fo-tiao33 (Buddhadeva) was an assistant of An-
Huen (Ngan-Hiuan). He learnt Sanskrit, the original language
of the Buddhist texts from Central Asia, and could recite the
whole of the Pratimok$a. He was given the Sanskrit name
Buddhadeva, and accorded the title of ‘Acarya’. Later on
several other Parthian monks of minor importance went to
China during the third and the fourth centuries A.D. They
equally contributed to the spread of Buddhism in China as also
to the work of translating texts into Chinese language.
The Sogdians, originally from their centre at Samarkand to the
30. Nanjio. II. 4, pp. 381-2.

31. ibid, II, 6, p. 383.
32. ibid. II, 9, p. 384.
33. Bagchi. La Canon Bouddhique en Chine—Op. cit, pp. 48-50.
north of Tokharestan were famous traders, who had their

colonies in different parts of Central Asia and had come into
association with Buddhism and Buddhist culture. The Sogdian
monks also cuntributed towards the transmission of Buddhist
culture and religious thought in China. Their names are distin
guished by the prefix Kang, as the ancient name of Sogdiana in
China was Kang-kiu. Some of these monks had collaborated
with Ngan-She-Kao. An illustrious Sogdian monk who worked
in South China in the 3rd century was Seng-Hui.34 His ancestors
had at first settled down in India. His father was a merchant
aud he had to stay in Tonkin (Kiao-Che) where Seng-hin was
born. After his father’s death, this young boy left the world and
became a monk. He soon proceeded to Nanking where he built
a monastery and founded a Buddhist school. He was the first to
introduce Buddhism in Southern China. Nanjio, however,
refers35 to his administrative background, as the eldest son of
the prime minister of the country of Khan-Ku i.e. Kambo or
Ulterior Tibet or Kamboja, whose family continuously resided
in India. He came to the capital of the kingdom of Wu in A.D.
241. In A.D- 247 he had the Kien-Ku-Sh or Kien-Ku monastery
built by order of Sun-Khuen, the first sovereign of the Wu-
dynasty who named the plsce Fo-tho-li or the Buddha village.
In A.D. 251 he began his translation work, continuing'it for nearly
thirty years till his death in A.D 280- He translated nearly a
dozen Buddhist texts into Chinese-
While the evidence from the Chinese sources records the con
tribution of Buddhist monks and savants from different nationa
lities towards the dissemination of Buddhism and translation
of canonical Buddhist texts, epigraphic, numismatic and art and
archaeological evidences might as well be quoted in this context.
The oldest record36 found in Tadjikistan from the site of Darshai
in the west Pamirs by A. N. Bernahtam mentions Narayana ‘be
victorious’ in Kharosthi script of the end of the second or the
beginning of the first century B. C. While Narayana is another
34. Bagchi. India and Central Asia—Op. cit, pp. 39; Toung-Pao, 1909,
199.
35. Catalogue—Op. cit, II, 21, p. 39.
36. J. Harmata. ‘The oldest Kharo$thi inscription in Inner Asia’ ‘Acta
Orientalia’, Vol. XIX, N o ., Budapest, 1966, pp. 1 -12.
name of Vispu, and this inscription like that of Heliodorus at

Vidisa could point to the acceptance of Bhagavatism37 by forei
gners, here it is in the context of reverence to Buddha Narayaria.
Buddha is noticed in the Khotanese-§aka documents from
Eastern Turkestan, and ‘Narayana’ the deva occurs in the Bud
dhist Sogdian documents.38
The spread of Buddhism in Central Asia was no doubt stimu
lated by the formation of the Kusana empire, which at its zenith
seems to have overshadowed the Greco-Bactrian kingdom. The
first ruler Kujula Kadphises styles himself on his coins as ‘stead
fast in true law’ (Sacadhrama-thitasa =>Satyadharma sthitasya)
and the figure of Buddha is depicted on his coins. His successor
appears to be Saivite, while Kaniska, the great Kusana ruler, was
definitely a Buddhist though publicly he professed Eclecticism.
Buddha figures prominently on his coins with the legend
CAKAMANO-BOYDO—BUDDHA SAKYAMUNI. Kaniska’s
name is also associated with a vihara or monastery set up for the
Sarvastivadins in an inscription dated in the year 1 from Shah ji
Ki Dheri39 near Peshawar. Reference has already been made to
the Buddhist theologian Ghosaka, born in Tokharistan, who was
one of the leading figures at the Buddhist Council held in Kaniska’s
time at Purusapura, who belonged to the Vaibhasika school, a
branch of the Sarvastivadins. The finds of potsherds from Kara-
Tepe, with inscriptions in Brahmi point to the spread of the
Sarvastivada school of Buddhism in the present Russian or
Western Turkestan in the time of Kaniska.40 The same site also
yielded inscribed potsherds in Kharosthi reflecting the teachings
of the Mahasanghika school. The reference to pioneer Buddhist
savants instrumental in translating Buddhist canons into Chinese,
hailing from the country of the Yueh-chi indirectly points to the
kingdom of the Kusapas as their homeland. It is suggested that
37. According to Dey, S.K., NarSyaija is connected with the cult of the
Sun (Bhagavatism and Sun-worship—Bulletin o f the School o f Oriental and
African Studies, BSOAS, 1931, Vol. VI, Pt. 3)
38. J. Harmata : Op. cit, p. 6.
39. Sten Konow : Corpus Inscriptionum Indicarum (CII) Vol. II (I)
pp. 134 IT.
40. Litvinsky, B.A. Outline History o f Buddhism in Central Asia,
Diushanbe, 1968, p. 10.
Buddhism also consolidated its position in Margiana in the first

century A.D. and it reached Eastern Turkestan through the
western part of it, and from there it went to China. Zurcher
enacts41 this process between the middle of the 1st century B.C.
and middle of the 1st century A.D. Among the early Buddhist
missionaries to China (Lo-Yang) were two Parthians (An-shib
Kao and An-Hsuan), three Yueh-chih (Chih-lou-chia-Chien—
Lokasena, Chih-vao and Chih-liang), two Sogdians (Kang-Meng.
hsiang and Kang-Seng-hui). Reference has already been made to
the life and literary activities of these savants.
At the beginning of the third century, the Sogdian monks again
figure as translators and this process continued throughout the
third century largely due to the efforts of those whose ancestors
had emigrated from western Turkestan. One such scholar was
Chih-Chien,42 also known as Chih-Yueh, grandson of a native of
the land of the Yueh-chi who had settled at Loyang. Another
was the Sogdian Kang-Seng-hui43 whose merchant father had
settled down in Loyang. The famous scholar Yueh-chi Dharma-
raksa43 who worked between A.D. 266 and 309 belonged to Tun-
huang where his ancestors had settled down earlier. Among his
disciples was a Sogdian. Reference is also made to the Parthian
An-Fa-chin engaged in translations of canonical works between
A.D. 281 and 306. Another scholar who had lost touch with the
country of his forefathers was Kang-Seng-yuan44, a Sogdian born
41. E. Zurcher. The Buddhist Conquest o f China. The Spread and Adapt
ation o f Buddhism in Early Medieval China, Vol. I (text) Leiden, 1956,
pp. 22-33. (quoted by Litvinsky : Op. cit, p. 10, also note 43 p. 81). •
42. Chih-Chien or K’Khien, who had the liberal appellation Kun-min,
and also another Cognomen Yueh, was an upasaka of the country of Yueh-K’
who came to China towards the end of the Eastern Han dynasty which came
to an end in A.D. 220. Afterwards he took refuge in the kingdom of Wu, where
he was appointed as a professor by Sun-Knien, the first sovereign of the Wu
dynasty. He translated numerous works in A.D. 223 -253. (49 Sutras according
to San-Kwhan fasc. 1, fol. 9 BC). Nanjio : Op. cit, 11, 18, p. 388.
43. Nanjio considers Khan San-hwui, as an Indian Sramana who was
the eldest son of the prime minister of the country of Khan-Ku i.e. Kambu or
ulterior Tibet, or Kamboja whose family was continuously resident in India.
He came to the capital of the kingdom of Wu in A.D. 241. In A.D. 251 he
began his work of translation and died in A.D. 2S0. (Nanjio : Op. cit, II,
21, p. 390).
44. Ku Than mo-Io-Kha (i.e. Dharmarak$a whose name is translated
in a foreign land. At the end of the 4th century one Dharmanan-

din from Tukharistan (Tu-ho-lo) came to China. He stayed
there between A.D. 384 and 391 and translated five works in this
period, including very important Hinayana ones.45 It has been
calculated that among the translators engaged in rendering the
Buddhist scriptures into Chinese, some time before the end of
the Western Chin dynasty (A.D. 265-316), there were six or
seven of Chinese and six of Indian origin as well as sixteen be
longing to various Central Asian nationalities—six Yueh-chi,
fourParthians, three Sogdians, two Kucheans and oneKhotanese.
These appear to be only approximate figures, pointing to the
interest of Central Asian Buddhist scholars in the propagation
of Buddhism in China, as also its flourishing state in Central
Asia. There are also indications of Central Asian monks engaged
in Buddhist activities in north-western India and in the Taxila
region in the Ku$ana Period. The earliest reference is to a certain
Bactrian (Bahlikena), a resident of the town of Noacha or
Noachea (unidentified) in the Taxila inscription46 of the year
136 of the unknown era (probably of 57 B.C.). It is recorded
on a silver scroll in a vessel containing a small golden casket in
which there were fragments of bones. It invokes wishes for the
bestowal of health on the great king, the king of kings, the son
of God, the Kusane, who is generally identified with Kujula
Kadphises. The contents of the inscription show that the Bac-
trians acted as zealots of Buddhist religion in India. In this
context it might be necessary to record the inscriptions relating
to Buddhism—its differentschools and protoganists in the Kusana
period.
The existence of the two important schools of Buddhism,
Fa-hu, literally ‘law protection’, was a Sramana whose family resided in the
Thu-Kwan district (the western extreme of the Great Wall in Kan-Shuh in
Nan-Si-Keu (China). He was a descendant of a man of the country of Yueh-
chi. He became a disciple of the foreign Sramana Ku-Kao-tso. He went to
the western regions with his teacher, and was well acquainted with thirty-six
languages or dialects. In A.D. 266 he came to Lo-yan where he worked at
translations till A.D. 313 or 317 and died afterwards in his seventy-eighth
year. He is supposed to have translated 90 works (Nanjio : Op. cit, pp. 391-
92).
45. Nanjio : Op. cit, II, 57, pp. 404-05.
46. Corpus Inscriptionum Indicarum (CII.II(i)) Op. cit, pp. 70 ff.
namely that of the Sarvastivadins and of the Mahasanghikas is

noticed in the Brahmiand Kharosthi records of the Kusana period.
An earlier record from Mathura—the Lion capital inscription of
the time of Sotfasa,47 however, points to rivalry between the two
schools, and it also specifies the location of the headquarters of
the Sarvastivadins at Nagarahara in Jalalabad (Afghanistan).
This record is important from the point of view of history of
Buddhism, as also for the conflict between the two schools. It
refers to the import of dialectician (Khalula), the Sarvastivadin
monk Budhila from Nagara to teach the foremost Mahasanghi-
kas the truth. The Sarvastivadins are also noticed in the Kalwan
Copper plate inscription48 of the year 134, probably of the old
era. It records the enshrinement of relics in the stupa shrine by
Chandrabhl, the female worshipper in acceptance of the Sar-
vastivada for the attainment of nirvana. The inscriptions of the
Kusana period connected with this school are the famous
Taxila casket of the year I, Zeda of the year 11 and Kurram of
the year 20 of Kanjska’s era.49 The donors are Hipea Dhia
(Zeda) and Yola Mira Shahi (undated Tordher-Thal valley,
Loralai Baluchistan) who appear to be foreigners. These records
from the Kharosthi region suggest Sarvastivadin establishments
in Afghanistan, West Punjab and Sindh. Mathura, of course,
was the important centre of the Sarvastivadins who were facing
the stiff opposition of the Mahasanghikas there and had therefore
to seek the help of their fellow brother dialectician at the other
end of the Kusana empire.
The Mahasanghikas, too, had their establishment in Afgha
nistan, as is evident from the Wardhak inscription dated in the
year 51 ofthetim eofH uviska.lt refers to the deposit of the
relics of Lord Sakyamuni in the Vagramarega vihara, in posses
sion of the.Mahasanghika teachers. There is no other Kharosthi
record of this school, although quite a few from Mathura50 are
associated with it. It appears that Afghanistan had several centres
of Buddhism and Buddhist activities in the Kusana period, as
47. ibid, pp. 30 fT.
48. Journal of the Royal Asiatic Society (JRAS), 1932, pp. 949 IT.
49. Puri, B.N. India Under the Kushdnas (IUK) pp. 141 ff for consoildated
references and comments.
50. ibid, p. 142.
is evident from these records in Kharosthl from that country.

This is supported by reference to several monks from the Kabul
(Kubha) region who went to China and translated Buddhist
canonical works into Chinese, of course later on in the begin
ning of the fourth century A.D. This could not have been possible
except for a strong base of Buddhism in the Kabul area, esta
blished during the period of the Kusanas. These included Khu-
than San-Kie-ti pho61 i.e. Gautama Sanghadeva (A.D. 383),
Pi-mo-lo-kha i.e. Vimalaksa515253 a great teacher of Vinaya,
with Kumarajiva as one of his disciples, who arrived in China
in A.D. 406; San-kie-poh-khan63 i.e. Sanghabhuti who
translated 3 works in A.D. 381-385; and Fu-zo-tu-lo i.e.
Punyatrata54 who arrived in China in the Hun-sh period A.D.
399-415 and was a collaborator with Kumarajiva; Than-mo-ye-
sho i.e. Dharmayasas55 who translated 2 or 3 works in A.D.
407-415.
Buddhism and the Southern States
While Western Turkestan, Bactria and Parthia as also Afgha
nistan and Kashmir contributed a lot towards the spreading
and development of Buddhism during the first few centuries of the
Christian era, the area now forming part of Chinese Turkestan
—Kashgar, Yarkand, Khothan and various sites of Kroraina—
, was equally humming with Buddhist activities. The States on the
51. Nanjio. Op. cit, II, 39, p. 399. He was a Sramana of the country of
Kc-pin i.e. Kubha. In A.D. 383 he arrived at Khan-an, then the capital of the
Former Tshin dynasty of the Fu family where he translated two works. Bet
ween A.D. 391 -398 he translated five other works in two different places.
52. ibid. II. 44, p. 400. He was a Sramana of Kubha (Kabul) and a great
teacher of Vinaya in Kwei-tsz i.e. Kharakas or Kuke, where Kumarajiva was
one of his disciples. Afterwards in A.D. 406 he arrived in China and was
respected by his former disciple Kumarajiva who was flourishing there. After
the latter’s death, which happened between 409 and 415, Vimalaksa went
southward and translated 2 works. He died at the age of seventy-seven.
53. ibid, II, 54, p. 404. A Sramapa of Kubha (Kabul) who translated 3
works in 27 or 37 fascicule between A.D. 381-85.
54. ibid. II. 60, p. 408. He was a Sramana of Kubha (Kabul) who arrived
in China in the Hun-Sh period (A.D. 399-415) and in A.D. 404 he, together
with Kumarajiva, translated one work in 58 fasciculi (Sarvastivadavinaya).
55. ibid, II, 62, p. 408. He also belonged to the country of Kubha and
translated 2 o r '3 works in A.D. 407-415.
northern route comprising of Kucha and Turfan seem to have

been under the spell of a separate wave of Buddhism, and, as
such, demand independent study in this context. Kashgar also
included Kie-pan-to (Sarikol)56 and Wu-sa (Yangi-Hissa) as
Hsuan-tsang speaks of linguistic affinity of these states in his
time. Ho-pan-to, the name given by Song-yun-to, is the same as
Kie-pan-to of the later Chinese pilgrim. The TangAnnals'accord
it several other names : Han-to, Ko-kuan-to etc. besides Ho-
pan-to. Its capital on the river site (Hsi-to) identified with the
Yarkand Darya is located at modern Tashkurghan which con
tains ruins of earlier times. In Hsuan-tsang’s57 time there were
more than 10 monasteries in the capital with 500 monks be
longing to the Sarvastivada school. The ruler at that time pat
ronised Buddhism and was a cultured scholar. A famous scholar
ofTaksasila, Kumaralata58 was brought here and he was the
founder of the Sautrantika school of Buddhism, and a contem
porary of Asvaghosa, Nagarjuna and Aryadeva. He is said to
have written many treatises, but only one book the Kalpanaman-
ditika59 a recast of the Sutrdlamkara of Asvaghosa, has been
partially discovered from Chinese Turkestan. The local ruler had
built a monastery for this Buddhist scholar which existed in the
seventh century A.D.In the next country of Wu-sha which lay
56. The early Chinese accounts of Sarikol are recorded by Aurel Stein
in his Ancient K/iotan, Vol. I, pp. 27 IT. The Chinese transcriptions variously
render the name as Chieh-pan-to, Han-pan-to etc. The Tang Annals also
mention several names Ho-pan-to, Han-to, Ko-Koan-tan or Ko-lo-to. The
Annals also notice the identical appearance and language of the people of this
region with those of Yo-tien or Khotan. For Hsuen-tsang’s account, see Beal :
Op. cit, ii, p. 298, also Life (translated by Beal : Op. cit, p. 196).
57. Beal : Op. cit, II, p. 298.
58. Khotan, p. 37. The venerable Kumaralata (labdha) was a native of
Taksasila. From his childhood he showed a rare intelligence, and in early
life gave up the world. (Beal : Op. cit, ii, p. 302). Hsuan-tsang then records
the spiritual excellences which made Kumaralabdha renowned as a great
teacher in the north, like Aivaghosa in the east, Deva in the south, and
Nagarjuna in the west. The king of this country (Chieh-pan-to), therefore,
having heard of the honourable one and his great qualities raised an army,
made his troops attack Tak$asila and carried him off by force. He then built
this convent and thus manifested the admiration with which he (Kumara-
Jabdha) inspired him.
59. See. H. Luders Bruchstucke der Kalpanamanditikd des Kumaralata
Leipzig, 1926; Levi. Journal Asiatitjue (JA), ccxi, 1927, pp. 95 IT.
on way to Kashgar on the eastern slopes of the Tsung-ling

(Pamirs), Hsuan-tsang noticed60 10 monasteries with nearly 1000
monks, all belonging to the Sarvastivada school. This country
is identified with Yangi-Hissar in the extreme north-west of the
Yarkand oasis.
Kashgar, besides of its strategic position, had played a signi
ficant role in the transmission of culture to the northern and
southern states of the Tarim basin. It is mentioned by several
names61 in the Dynastic Annals of the Chinese, while the
pilgrim to India calls it Kia-she, Kie-sha (Hsuan-tsang) or
Kie-cha (Fa-hien). The place is supposed to be known to
India as Khasa or Khasya and its script Khasyalipi is men
tioned in the Lalitavistara (2nd or 3rd century A.D.), while
Ptolemy calls it Kashaozi. The history of this kingdom has been
recorded earlier. The position of Buddhism in Kashgar62 is
60. Beal : Op. cit, ii, p. 304; for the identification on Wu-Sha, see Stein :
Ancient Khotan, pp. 42.
61. From the time of the Former Han Dynasty, when the States of Central
Asia were first opened upto the political influence of China down to the Tang
period, the region of the present Kashgar was generally known by the name
Su-lc or Shu-le. Cf. accounts of pilgrims Sung-Yun, Kumarajiva (c. 400 A.D.),
Fa-Yung (420 A.D.), Dharmagupta (c. 593-95 A.D.) and Wu-King. Hsuan-
tsang and the Tang Annals record the name ‘Chia-sha (transcribed as Kia-cha).
According to - Levi, Shu-le—Chia-lo-shu-tan-le in reality a transcription
of Kharo$thi was an ancient name of Kashgar from which the Kharo§{hi
script received its designation (Bulletin de I'Ecole d ’Extreme Orient
(BEFFO. ii, p. 246 sq.); Stein—Ancient Khotan, Op. cit, pp. 48-49).
62. Leggc. Fa-hien (translation) p. 23. According to a statement of
Klaproth, gathered apparently from Chinese sources, the interference of the
Yueh-chih in the affairs of Kashgar, towards A.D. 120, resulted in the intro
duction of Buddhism into that territory. Buddhism no doubt flourished in
the Yueh-chi dominions on both sides of the Hindukush. It is proposed
by Stein that the prolonged sojourn of the Kashgar prince, subsequently
elevated to the throne, might have facilitated the spread of Buddhist propo-
ganda in that part of the Tarim basin. This assumtion agrees with the tradition
recorded by Hsuan-tsang making the princely hostages from the states east
of the Tsung-king, including Sha-le or Kashgar reside in a Buddhist Convent
and connects their stay with the reign of Kaniska, the renowned patron of
Buddhism (Ancient Khotan. Op. cit, p. 56). Further, to whatever period
the first establishment of the Buddhist church in Kashgar may prove to belong,
it is far more probable that it was brought from the side of Baktria than from
that of Khotan.
recorded by Fa-hienwho visited the country towards the closing

years of the 4th century. He found the religion of the Tathagata in
a very prosperous condition, and he also witnessed the quinqu
ennial assembly (pahcavarsika). In his time, Kashgar had more
than two thousand monks, followers of Hinayana probably of
the Sarvastivada school, following canonical rules strictly. The
impressions of Fa-hien about Buddhism in Kashgar and the
'pageantry are confirmed by other Chinese pilgrims visiting this
country shortly afterwards: Che-mong (404), Fa-yong and Tao-
yo (420).6364 The number of Buddhist monks had swelled to
10,000 in Hsuan-tsang’s time84 without any deviation from
their sectarian allegiance. They were all followers of the Sar
vastivada school, and many of them could recite the Buddhist
texts including the entire Tripitaka and the Vibhdsa—all in
Sanskrit. The place was also visited by Wu-kong who travelled
back to China about A.D. 186.
Among the scholars from India or other parts of Central
Asia .were BuddhayaSa of Kashmir and Kumarajiva, born in
Kucha, who had his education in Kashmir and on way back he
stayed there for nearly a year. Buddhayasa was of great help to
Kumarajiva in his academic pursuits at Kashgar; and he stayed
on there even after the departure of Kumarajiva, exercising great
influence on the local ruler, named Pu-tu. His son Ta-mo-fo-to
(Dharmaputra) was greatly influenced by Buddhayasa’s friendly
approach. He maintained close relations with the regal family as
also with his contemporary Kumarajiva who later joined in
China. He had persuaded the local ruler to send an army to
Kucha against Chinese aggression in 382, more for the sake of
security of his friend Kumarajiva, and personally accompanied
the force which reached there rather late after the fall of Kucha.
The bond of academic union between these two Buddhist savants
of Central Asia was so close that they joined hands in trans
63. Bagchi : India and Central Asia, Op. cit, p. 46.

64. Beal : Op. cit, ii, p. 307. According to the Chinese pilgrims the
writing (writing character) is modelled on that of India, while language and
pronunciation are different from that of other countries. They have a sincere
faith in the religion of Buddha, and give themselves earnestly to the practice
of it.
lating several canonical works in Chinese.05 As teacher of

Kumarajiva,he was given the epithetTa-pi-po-cha (mahavibhasa).
He translated four works into Chinese.
Another Buddhist scholar at Kashgar was Dharmacandra,60
originally from Magadha who had gone to China in A.D. 730
from Kucha at the invitation of the Chinese ambassador. On
his return journey from China in 741, he passed through
Kashgar where he had to stay back on account of political
trouble ahead of the route at Shughnan. Finally, he settled down
at Khotan where he died two years later in 743 and a stupa was
erected in his memory. Only one work Prajhaparamitd-hridaya
is attributed to him. A number of old Buddhist stupas found
at various sites around Kashgar are suggestive of the importance
of this place as a centre of Buddhist activities and its scholars.
The country of Che-kiu-kia, accorded different name in the
Chinese sources and identified with Cokkuka in the Cenral
Asian documents,6567 presently Karghahalik-Yarkand, was an
important Buddhist settlement. According to the Chinese
pilgrim Hsuan-tsang,07 the people here were earnest Buddhists
with tens of monasteries and more than 100 monks, all followers
of Mahayana school of Buddhism. The Mahayana. tradition here
could be traced to the fourth century A.D. as two princes of
Cokkuka, Suryabhadra and Suryasena had gone to Kashgar to
receive initiation from Kumarajiva and to study Mahayana texts
with him. It is quite likely that these princes belonged to an
Indian ruling dynasty which had earlier settled down there.
Khotan and its origin and history have been recorded earlier-
As regards the introduction of Buddhism here, it was done in
the time of Vijayasambhava who was grandson of Kustana
with the monk Vairocana, an incarnation of Maitreya coming
here- He was responsible for bringing the relics of Buddha
from Kashmir. The Khotanese ruler Vijayasambhava built for
Vairocana the great monastery of Tsar-ma,68 the first one in
65. Bagchi : Le Canon Bouddhique en Chine—Op' cit, pp. 200 ff; Also
A. Stein : Ancient Khotan, I, pp. 47-48. Bagchi records the collaboration of
these two savants in the translation of several works, (pp. 202-3).
66. Bagchi—ibid, pp. 565 ff.
67. Bagchi—India and Central Asia—Op. cit, p. 48 ff.
68. A. Stein—Ancient Khotan—Op. cit, p. 232.
Khotan, identified by Aurel Stein with Chehna-Kazan near

Yotkan. A number of other monasteries were set up during the
reigns of subsequent rulers. Three Arhats from India—Buddha-
data, Khagataand Khagadrod—went to Khotan, and a later king
Vijayavirya built two monasteries for them. This was followed
later on by another two built by the Chinese, princess Punesvar,69
queen of king Vijayajaya, in honour of Kalyanamitra who had
gone there from India. The ruler’s eldest sonDharmanandawho
became a Buddhist monk was a follower of the Mahasanghika
school and built several monasteries which kept on swelling in
subsequent generations. In the time of Hsuan-tsang there were
100 monasteries in the capital with 5000 monks who were all
followers of Mahayana school of Buddhism. The pilgrim also
refers70 to some of the important monasteries like those of Ti-
ko-po-fa-na, Sha-mo-no, Gosringa and Mo-she. Earlier, Fa-hien
refers 71 to the Gomati monastery to which 3000 monks were
attached, and they were all followers of Mahayanism. The
Gosringa monastery built later on was supposed to be a spot
formerly associated with the visit of Buddha. This monastery is
also mentioned in the Suryagarbha-sutra,72 a Buddhist canonical
text translated into Chinese between A.D. 589 and 619.
Khotan, it seems, became an active centre of Buddhist studies
69. According to the ‘Annals of Li-yul’ (Rockhill : Life o f the Buddha,
pp. 238 ff), Vijayasimha, the third successor of Vijayavirya married Princess
Pu-nye-shar, the daughter of the ruler of China. She was responsible for
raiming silk worms much against'the wishes of her husband, who later on
repented for his deeds. He called from India the Bhik?u Sanghagho$a and
made him his spiritual adviser, and to atone for his wickedness he built the
Po-ta-rya and Ma-dza caityas and a great vihara. (Stein : Ancient Khotan,
p. 230.)
70. Beal. Buddhist Records o f the Western World. Op. cit, ii, pp. 309 ff.
For an account of these monasteries, see p. 313 ff.
71. Fa-hien’s account of Khotan is recorded in his Travels (Legge.
Translation, pp. 16-20). The monks in Khotan were mostly students of the
Mahayana. The pilgrim specially refers to the hospitable arrangements made
in the Sangharama for the reception of travelling monks, and notices the
custom of erecting small stupas in front of each family’s dwellings.
72. The Chinese translation of the Suryagarbha-sutra made by Narendra-
yaias between the years 589 and 619 A.D. in a list of holy places (pitha) sancii-
fied by the presence of a Bodhisattva (Stein : Ancient Khotan. Op. cit,
p. 186).
and activities. In the year 259, a Chinese monk named Chu-she-

hing came to Khotan for the study of Buddhism, first hand
under good teachers. In a few years time he collected 9000 bun
dles of original Buddhist texts and could manage to send these
home through his disciple Fu-jo-tan (Punyadhana ?). He died
in Khotan at the age of 80. These were translated by a Khota-
nese Buddhist scholar Moksala73 who went to China in 291. He
was assisted by an Indian monk, probably Sukta-ratna. The
translated texts were Pahca-\i\hsati-sdhasrikd-Prajhdpdramitd,
Vimalakirtinirdesa and Surahganasiitra—all Mahayana canonical
works. It appears from this account, as also from the reference
to translation that Khotan was famous as a centre of Mahayana
Buddhistic studies in the third century A.D. Fa-hien’s account
quoted earlier not only confirms this contention but equally
points to the prosperity of this school of Buddhism in his time
with 3000 monks in Gomati monastery alone living a regulated
ordained life. Dharmaksema7475of Magadha working in Liang-
chou had to come to Khotan for tracing the lost portion of his
incomplete manuscript of Mahaparinirvana-sutra which he trans
lated into Chinese. His pupil Tsiu-kiu-king Sheng, a noble man
of Liang-chou also came to Khotan, settled down in the Gomati
mahavihara and studied Mahayana Buddhism under an Indian
scholar Buddhasena76 who was noted for his academic attain
ments in all the countries of the west and was called S/ie-tseu
(Simha)—the lion of learning. On his return to China King
Sheng translated the texts on Dhyana which he had studied in
Khotan and had brought with him.'
A little later eight Chinese monks of Liang-chou came to
Khotan in 439 in search of Buddhist texts. Their visit coincided
73. For an account of Mok$ala—see Bagchi : Le Canon Bouddhique
en Chine—op. cit, pp. 119-120; Nanjio : Catalogue, Op. cit, ii, 26, p. 394.
He was a Sramaria of Yu-then i.e. Kusutana (Khotan) who together with
Ku-Shu-lan, an upasaka of Indian descent born in China, translated one
Sutra in A.D. 291 (Pahcavimsati-sahasrikd-prajhparamitd). According to
Eital, he was also the author of a new alphabet for the transliteration of
Sanskrit (A Sanskrit-Chinese Dictionary, Op. cit, p. 100 B).
74. Bagchi : Le Canon Bouddhique en Chine, pp. 212-221.
75. Buddhasena (Fo-to-Sien) was probably the master of Buddhajur
who came to China from Kashmir in 423 A.D. and worked at Nanking under
the Sung (Bagchi—ibid, p. 179 n).
with the Pancavarsika assembly, and they attended the discourses

of a few teachers and recorded some texts from dictation which
they took back to China. This was not the solitary instance. A
number of important Buddhist texts were taken to China from
Central Asia A monk named Fa-ling took a manuscript of the
famous Avatamsaka-sutra which was translated by Buddha-
bhadra76778 in 418. He was a contemporary of Kumarajiva as well
as of Fa-hien and cooperated with them in translation work. He
met Kumarajiva in China and whenever the latter had any
doubts, the former was always consulted for an explanation. He
made some translations with Fa-hien. Between A.D. 398-21 he
translated 13 or 15 works. He died in A.D. 429 at the age of 71.
Namjio quotes 7 v'orks of this savant—all Mahayanist. This
process of transmission of Buddhist thought and literature from
Khotan—a great centre of Mahayana Buddhism—to China con
tinued later on. A monk named Fa-hien brought from Khotan
the manuscript of Saddharmapundarika which was also translat
ed by Dharmamali,77 a sramana of western origin, in 490. Among
the Buddhist scholars of Khotan Siksananda was the foremost
who reached China in 695 and worked there till his death in 710.
He translated in all 19 texts, the most voluminous one being the
Mahavaipulya or the Avatamsaka-sutra78 in 80 chapters.
Literary evidence from the Chinese source material, the finds
of Buddhist monuments—stupas and sculptural figures, stucco
statues of Buddha or Bodhisattvas as also painted panels from
nearby sites do provide indication of the efflorescence of Bud
dhism in this part of Central Asia which was strong enough to
transmit its offshoots in the form of its scholars and literature
to China. It is proposed by Aurel Stein,79 that Indian elements—
the use of Sanskrit and Brahmi script preceded Buddhism in
Khotan. Even before the introduction of Buddhism Khotan’s
76. Bagchi—Le Canon Bouddhique en Chine—Op. cit, pp. 341 -46; Nanjio
Catalogue, Op. cit, ii, 42, p. 399-400. His life history and achievements are
recorded by Bagchi, Op. cit.
77. Bagchi. ibid, p. 409. He was a Sramana who was invited by Cha-men
Fa-hien to the monastery of Wa Kouan-sse in the capital for translating
sacred texts. Naniio : Catalogue, Op. cit, ii, 96, p. 42*
78. Bagchi : India and Central Asia, p. 61.
79. Ancient Khotan. Op. cit, pp. 163-64.
population had fusion of a strong Indian element and the cul

tural influences accompanying it. The finds of documents from
Niya in Kharosthi—a script of the north-western India and
written in Prakrit—tends to signify a later under-current of
Buddhist influence.
Besides Khotan and the neighbouring areas, a number of old
Buddhist sites excavated in the desert at Dandan-Uiliq have re
vealed stucco images and reliefs. Frescoes with Brahmi inscrip
tions, manuscripts and other records mostly in Brahmi and
Chinese are fairly interesting pieces of evidence in the context of
Buddhism in that region. The Brahmi manuscripts include can
onical works of Mahayanism—The Prajhdparamitd and Vajra-
cchedika80 in Sanskrit. The Rawak stupa81 at a distance of only
7 miles to the north of Dandan-Uiliq has exposed Buddhist
images mostly of the Gandhara school. Ruins of Buddhist
stupas have also been found at other sites. Niya was an impor
tant Buddhist centre in the time of Hsuan-tsang. A large num
ber of Kharosthi documents82 from this site shed light on the
material culture as also on administrative setup. The documents
are in Kharosthi script of the 3rd century and their language is
Prakrit allied to the one in the north-western Frontier of India
in the Kusana period* Later on, Kharosthi was ousted by Brahmi,
probably with the introduction of Sarvastivada Buddhism from
Kashmir. Both the schools of Buddhism were introduced pro
bably in the 4th century. Kumarajiva studied the Tripitaka of
the Mahayana school and its* Vibhasa in Kashgar—the gateway
to Khotan. Reference has already been made to the country of
Tu-ho-lo about 400 li to the east of Niya, represented, accord
80. Ref. find of Prajna-Paramita MS. at Dandan-Uiliq in Stein :• Ancient
Khotan, pp. 257-58. Hoernle’s analysis proves that the fragments belong to
the eighteenth and nineteenth chapters of a Mahayana text, written in Gupta
characters of the seventh or eighth century. As regards the Vajracchedika a
famous treatise of the Mahayana school, first edited by Max MOiler in the
Anecdota Oxoniensia, fifteen out of twenty-four folia have been found intact
(ibid p. 258).
81. Stein, A. Ancient Khotan. Op. cit.pp. 483 ff.
82. Stein, A. Serindia, Op. cit, I, pp. 433 fF. These documents have been
edited by Boyer, Rapson and Senart and published under the title ‘Kharosthi
Documents discovered by Sir Aurel Stein in Chinese Turkestan; Text and
Translation (Oxford, 1928).
ing to Aurel Stein,83 with the modern site of Endere. It seems to

have been connected with the Little Yueh-chi, probably of some
earlier time. The Chinese pilgrim calls it the ancient Tu-ho-lo
(Tukhara) country- Ruins of a Buddhist stupa as also the finds
of Kharosthi tablets and manuscripts similar to ones found at
Niya, suggest that Buddhism flourished here. This place had
close relations and contacts with Tsiu-mo or Tso-mo84of Hsuan-
tsang, identified with Calmadana, now represented with Char-
chen.
Further to its north-east was Na-fo-po,85 in ancient times
called Lou-lan in the region of Lobnor, situated on the oldest
route connecting China with Central Asia. In the Han Annals,
the name of the country is Shan-Shan. Loulan was the
Chinese transcription of the original name Krorania or Kroran-
jina of the Kharosthi documents.86 It was a stronghold of Bud
dhism and Indian culture. According to Fa-hien,87 there were
4000 monks here, all followers of Hinayana in his time. The
Kharosthi documents discovered at various sites of Kroraina are
written in Prakrit, as was also current in Niya, and was the offi
cial language of the country. The names recorded in these docu
ments appear to be of Indian derivation as well as of local
origin.88 Among the important sites is Miran—famous for its
school of painting as also for its sculptures—mainly in Gandhara
style. The subject matter of painting is Buddhist and the style is
linked with that of Gandhara.
83. Serindia, I, p. 286. For Hsuan-tsang’s reference, see Julien : Memoirs,

II, p. 247; Walters : Yuan Chwang, ii, p. 304; also Stein : Ancient Khotan,
J, p. 435. Stein also refers to Henry Yule’s Notes on Hsuen-tsang’s Account
of the Principalities o f Tokharistan. J RAS. N. S. VI> p. 95.
84. Julien, Memoirs, i p. 247; Walters : Yuan Chwang, i, p. 304; Stein ;
Ancient Khotan, i. p. 435 note 9; Serindia, I, p. 298.
85. Stein : Serindia, I, p. 321; See Julien : Memoirs, ii, p. 247; Beal,
ii, p. 325.
86. See Inscriptions No. 754, 922 and 907, noticed by Aurel Stein in
Serindia, I, p. 413.
87. Legge. Travels o f Fa-hien, pp. 11 flf.
88. The Indian names are Anandasena, Bhaflsama, Budhamitra, Kumud-
vati, Vasudeva etc. Others of local origin are Cauleya, Kapgeya, Kalpisa,
Kitsaitsa, Pulkaya, Varpeya etc. (Serindia, I, p. 414).
Buddhism and Buddhist scholars in the Northern States

The political history of the States on the northern route—
Akshu, Kucha, Agnidesa or Karasahr, Kao-chang or Turfan,
has already been recorded in the previous chapter. Attention
could now be confined only to Buddhism and its propagators in
these areas rather than to political figures and contacts. It has
been proposed earlier that the people of the four states of Akshu,
Kucha, Karasahr and Turfan were of one racial stock in ancient
times, with a common language with slight dialectical differences
of a local nature and a common culture. The language is termed
as Kuchean. The specimens of this language were discovered
from various sites in Kucha and Karasahr, with a large number
of mural inscriptions in this language. Specimens of ancient
Tokharian have also been discovered in the neighbourhood of
Karasahr and in Turfan, probably as a result of the immigra
tion of Buddhist priests in this area after the Arab conquest of
Tokharestan. The Tokharestan texts were translated in Uighur,
the language of the Turks who had set up a new empire with
Turfan as the centre in the 9th century after the Arab conquest
of their homeland. As such both Kuchean and Uighur texts are
available in this area.
The exact date of introduction of Buddhism in Kucha is not
known but it could not have been later than the end of the 1st
century A.D. allowing it to take deep roots before blossoming in
full form about the third century A.D. According to the Annals
of the Tsin dynasty (265-316), there were nearly one thousand
Buddhist stupas and temples in Kucha at that time.89 The Bud
dhist monks from Kucha started moving to China from this
period, and the pioneer monk from the royal family Po-Yen90
89. As regards the early history of Kucha, it is mentioned in the Han
Annals and when brought into contact with China, in the reign of Wu-ti
(140-87 B.C.), its importance was realised because of its geographical location.
It is, however, described as a seat of Buddhism only in the time of the western
Tsin dynasty. According to the Tsin annals, it was enclosed by a triple wall
and contained a thousand stQpas and Buddhist temples as well as a magni
ficent palace for the king. This does not, however, suggest any date for the
introduction of Buddhism in that area. (Chavannes in Stein’s Ancient Khotan,
p. 544.)
90. For an account of the contributions of this Sramana of the western
region who translated some Sotras in the White House Monastery at Lo-yan
came to Chinese capital in 256-260. He translated six Buddhist

texts into Chinese in 258 at the famous Buddhist temple of Po-
ma-sse at Loyang. The larger version of Sukhamtl-vyuha was
made by him. Another scholar from Kucha Po-£rimitra91 went
to China during the period 307-312. The political troubles there
compelled him to move to South China where he translated
three Buddhist texts. Another Kuchean prince Po-Yen went to
Liang-chou. Though a scholar of repute and conversant with a
number of foreign languages he did not leave any translated
canonical work.
The next century—fourth—is noted for great Buddhist activity
in Kucha which was more or less a Buddhist city with the palace
of the ruler looking like a Buddhist monastery with standing
images of Buddha carved in stone. The city had a large number
of monasteries, some special ones being founded by the kings.
The monastery called Ta-mu had 170 monks, while that on the
Po-Shan hill in the north, called Che-hu-li had 50 or 60 monks.
The new one of the king of Wen-Su (Uch-Turfan) had 70 monks.
All these monasteries were controlled by Fu-to-she-mi (Buddha-
svamin). Some information is available about the working of
these monasteries as also on the activities of the monks from a
Chinese text.82 The head ‘Buddhasvamin* was a great scholar, a
follower of the Agamas—the Hinayana, while his brilliant disci
ple Kiu-kiu-lo (Kumara) probably Kuraarajlva of great fame in
his time was a Mahayanist. There were separate establishments
in A.D. 257, see Bagchi,: Le Canon Bouddhique en Chine, pp. 79 ff, as also
Nanjio : Catalogue, ii, 16, p. 387. According to Sylvain Levi, the royal
dynasty of Kucha from the first century A.D. was known as Po (J.A., 1913
p. 322), and the rulers have this prefix added to their names, suchasPo-ying
(124 A.D.), Po-Chang (285 A.D.) etc.
91. Po-Shi-li-mi-to-lo i.e. Srimitra—literally meaning ‘lucky friend’ was
a Sramana of western origin who was the heir apparent of a king of that
country. He, however, gave up his kingdom to his younger brother and
became a Sramana. He came to China in the Yun-Kia period A.D. 307-12,
under the western Tsin dynasty and translated 3 works at Kin-khan (Nan
king) under the reign of Yuen-ti, A.D. 317-322 and died at the age of eighty
in the Hhien-Khan period, A.D. 335-342. The works are Mahahhishek-
ardddhidhdrani-sutra, Mahdmayuri-Vidydrdgniand in two Faseimulae (Nanjio:
Catalogue, ii, 36, pp. 397-98).
92. Bagchi : India and Central Asia, p. 80. See also his Le Canon de
Bouddhique, p. 155.
for the nuns at the monasteries of A-li (Avanyaka), Liun-jo-kan

and A-li-po with 50 and 30 nuns respectively. These nunneries
were also under the control of the head of Buddhist establish
ments, Buddhasvamin. The nuns, mostly from the regal and
noble families, led a strict disciplined life, observing as many as
five hundred rules of conduct. The life and activities of Kumara—
Kumarajiva who played an important role in the dissemination
and expansion of Buddhist ideals and canons both in Central
Asia as well as in China, no doubt call for special reference and
recording.
Kumarajiva—his life and contributions93

The name Kumarajiva is sometimes transcribed as Kiu-mo-
lo-che and at others as Kiu-mo-to-tche-po with the Chinese
translation Tang-cheu. He hailed from a noble family of
ministers of state in India. His grandfather Ta-to had a great
reputation while his father Kiu-mo-yen was equally intelligent.
Kumarayana, as he was called, renounced his claim to hereditary
succession in favour of his relations and joined the Buddhist
order and left his home in search of learning and enlightenment.
Crossing the difficult routes of the Pamirs he came to Kucha
where he was appointed purohita (Kouo-che)—the royal priest.
The sister of the king—a beautiful princess of 20, discarding
other proposals, fell in love with this young Indian and they
were married. Kumarajiva, taking the name after his father and
also mother (Jiva), was the issue of this union. Soon after Jiva
took holy orders and joined a nunnery. She retired to the Tsio-
li monastery about 40 li to the north of Kucha and there learnt
the language of India. Kumarajiva at that time was seven years
old, but so sharp was his intellect that he committed to memory
the sacred texts, and could recite a number of Sutras as also a
93. For an account of Kumarajiva’s life history and his contributions to
Buddhist literature and canons, see, Nanjio : Catalogue, ii, 59, pp. 406-408;
Bagchi : Le Canon Bouddhique en Chine, pp. 178-200; and also Encyclopaedia
o f Religion and Ethics, Vol. VII, p. 701 a; Sylvain Levi : JA. 1913, p. 335-338;
Pelliot : Toung Pao, 1912, p. 392, in which he placed this savant between
344 and 413 A.D.; P. Wieger : Histoire des Croyances religieuses en Chine
etc. (1922) p. 416 in which is given a resume of the notices of Tsin Chou on
Kumarajiva.
thousand Gathas. This was followed by learning Abhidharma.

On attaining the age of nine, his mother took him to Kashmir
for better education under Bandhudatta, Pan-teou-ta-to, the
renowned scholar and a relation of the king. The young scholar
studied with him Dirghagama and Madhyagama and could
discuss with the heretics. After three years of stay, Jiva along
with her son Kumarajiva decided to return to Kucha. Crossing
the mountains and passing the kingdom of the Yueh-chi (Tokha-
restan) they reached Sha-le (Kashgar). An arhat in the way
noticing the young boy predicted great future for him who, like
Upagupta, would initiate a large number of people into the
canons of Buddhism. Arriving at Kashgar, he stayed there for a
year and studied the whole of Abhidharmapitaka with the
Kashmirian scholar BuddhayaSa who was then in Kashgar. He
also studied the four Vedas, five sciences, Brahmanical sacred
texts as also astronomy, the Satasdstra and the Mddhyamaka-
siistra etc. during his year’s stay here* It was also the period of
his ordaining two distinguished persons—the two sons of king
Tsan-kiun, the son of the king of So-kiu (Chokkuka). These
two princes were Suryasama (Siuli-yeso-mo) and Suryabhadra
(Siuli-ye-poto).
Leaving Kashgar, the mother Jiva and her son Kumarajiva
then twelve years old and an erudite scholar, reached Wen-Su
(Uch-Turfan) which was then the northern limit of Kucha. Here
he defeated a Taoist teacher in discussions and this spread his
fame. The king of Kucha, Po-Shui personally approached him
at Uch-Turfan for taking him to his place. Kumarajiva was by
that time well-versed in canonical literature, and he explained
Mahasannipdta and the Mahdvaipulya sutras to the daughter of
the king A-Kie-ye-mo-ti who had become a nun. He himself had
not taken holy orders till the age of 20 when he was fully ordain
ed at the king’s palace. At that time there were more than 10,000
monks in Kucha, He was provided accommodation in the new
monastery built by kingPo-Shun. This young scholar was equally
interested in exploring ancient manuscripts and he was fortunate
enough to trace the Pancavimsati-sdhasrikd Prajhdpdramitd in
the old palace. His academic exercises were not at a standstill.
Discourses and discussions with other scholars were part of his
routine. He explained the Mahayana Sutras in the great mona
stery of Tsio-li, and encouraged contacts with outside scholars*

One such was Vimalaksa from Kashmir who was received by this
Kuchean savant and in turn the Indian scholar helped his host
in the study of the Sarvastivada Vinayapitaka. Kumarajlva seems
to have alienated his filial relations, and one gathers that Jiva
left for Kashmir to spend the rest of her life there.
The political expedition of the Chinese force under Lu-Kuang
guided by the king of Kiu-che and that of Yar-khoto ran
sacked Kucha and carried numerous prisoners including Kumara
jlva. This savant’s merits and talents were recognised by this
general who had at first ill-treated him. He stayed at Liang-chou
till 401 and then at the request of Emperor Yao-Chang and
his successor Yao-Hing, Kumarajlva after great hesitation finally
left for Chang-ngan. He was warmly received by the new ruler
who appointed him his purohita—priest-counsellor (Kouo-che).
As his adviser and functionary, he exercised great power and
influence on the ruler. He became a great admirer of triratna
(San-pao) and asked him to arrange conferences in the royal
palace as also to translate sacred texts. He was offered coopera
tion and collaboration from several distinguished monks in the
task of editing, collecting and translating Buddhist Sanskrit texts.
These included Seng-che, Seng-kien, Fa-kiu, Tao-liou, Tao-
heng, Tao-piao, Seng-jovei, Seng-tchao etc. A little over eight
hundred monks attended the assembly at Chang-ngan addressed
by Kumarajlva.
Kumarajlva was well-versed with Chinese and the studies cover
ed by him were vast* He had studied ancient works, some
obscure and erroneous too. He had to make comparative studies
of such works along with the monks from India (Tien-tchou) and
those fromlndo-Scythia(Yue-tche). He was able to remove doubts
and dispel erroneous interpretations of monks from all quarters,
including those of Che-Tao-chong of Long-kouang, Houei-yuan
of Lou-chang. A monk of great intelligence and clear thinking
Seng-jovei was always withhim serving as his secretary and posed
doubts and differences in texts before him. Explaining the differ
ences between the literatures of India and China, Kumarajlva
referred to literary composition, mixture of rhythmical phases
with music, the universal songs of dedication (vandana), and
music always accompanying the visits of the images of Buddha.
While translating the Sanskrit texts into Chinese he refers to the

loss of literary charm in translated texts.
Kumarajiva’s stay at Chang-ngan was till his death in 413.
There he translated 98 works, characterising his stay there as the
glorious epoch in the history of Chinese Buddhism. Kumarajiva’s
Sarvastivadin guru Vimalaksa who was staying in Kucha all
this time, joined him in A.D. 404. He was followed a year
later by another Indian scholar named Dharmamitra who had
come to Kucha for a few years’ stay and then he came to the
Chinese capital in 424 when Kumarajiva was no more. Other
scholars who had earlier come in contact with him and later on
joined him were the Kashmirian scholar Ruddhayasa under whom
Kumarajiva had studied in Kashgar, and Buddhabhadra another
scholar from Kashmir.
Several interesting details about Kumarajiva’s life and activi
ties are recorded about him. It is suggested that two years after
his mother, who had already become a nun, brought her son to
Kubha (Kabul), the young monk became the disciple of the
famous priest, named Vandhudatta, a cousin of the king of that
place. In his twelfth year when Kumarajiva was brought back to
Karasahr by his mother, another Arhat told the mother that ‘she
shouldcarefullyguard this 3ramana against disorder till his thirty-
fifth year when he would propagate the law of Buddha; but if he
could not keep moral precepts (£ila) he would not be more
than a clever and skilful priest. Kumarajiva’s study of the Sar-
vastivada Vinaya under Vimalaksa, and later on that of Maha-
yanism with Suryasena, led him to exclaim that ‘his formerstudy
of the Hinayana was like that of comparing a copper ore with
the excellent golden one of which he was formerly unaware.’From
that time he devoted himself completely to the propagation of
the Mahayana,and he could convert his former teacher Vandhu
datta as well. The Chinese general Lu-Kuang taking Kumarajiva
with him compelled this young monk, then not yet thirty-five,
to sleep with a daughter ofthe former Kucha king.
Kumarajiva stayed with Lu-Kuang in Liang-chu, China till
A.D. 401 and on the twentieth day of the twelfth month of the
same year he arrived at Chan-ang and was greatly welcomed by
Yao-Hsin, the second ruler ofthe Later Tshin dynasty. Between
A.D. 402-412 he translated numerous works, and also wrote a

treatise and some verses in Chinese. He is said to have had Chinese
priests as his disciples, more than three thousand in number,
among whom there were about ten great ones who wrote several
works. Kumarajiva died in the Hun-sh period between A.D.399-
415, but the exact date is uncertain. Different traditions provide
varying data between A.D. 405 and A.D. 411. There are, however,
some of his translations of a much a later date.
Kumarajiva was responsible for introducing Mahayana in the
countries of the Tarim basin and alsoin China in a more respon
sible and authoritative manner. He was one of the greatest
exponents of this school of Buddhism and also of the Madhya-
mika philosophy. He introduced texts relating to these systems
into Chinese through translations, as also through interpretations
sought by many Mahayana teachers of this period. His contem
poraries included many savants of Kucha like Vimalaksa who
was staying there all this time and joined Kumarajiva in China
in 401. Another scholar who was of Indian origin, Dharmamitra,
came to Kucha and stayed there for a few years before leaving
for Tun-huang in China, finally reaching the capital in 424.
Buddhayasa, thefamousscholar of Kashmir under whom Kumara
jiva had studied in Kashmir, was not only a contemporary but
alsoa personal friend ofKumarajiva. He also joined him in China.
Buddhabhadra was another Kashmiri scholar who joined Kuma
rajiva first in Kucha, and later on in the Chinese capital for a
number of years. There might have been many more contempo
raries of eminence. Kucha and its Buddhist savants deserve
special attention for their contribution to Buddhism and its
canonical texts with numerous recensions. As an example, the
Chinese translation of the Candraprablia-sutra, a part of the
Mahasannipata we are told of 99 manifestations of Buddha while
Aksu had 24. According to this text, Kucha had her divine
protectors, such as the constellation Sravana.
Kumarajiva is said to have introduced a new alphabet and
translated some 50 works including Pancaviriisati-sahasrika
Prajhdparamitd, Prajhaparamitahridaya-sutra, Saddharmapunda-
rika, Sukhdvativyuha, Sarvastivada-pratimoksa, Mahaprajhapara-
mitdSutra-sastra, ‘Life of the Bodhisattva AsvaghosaVLife of the
Bodhisattva Nagarjuna’,and ‘Life of the Bodhisattva Deva’. The
translations ascribed to him are ranked as classical Chinese, and

his translation of the ‘Lotus of the True Law’(Saddharmapunja-
rika) remains the most valued and revered of the Chinese Bud
dhist scriptures. Kumarajiva was a monk but he is said to have
led rather an irregular life. Yet his talents were so appreciated
and his fame so high that his patron and the people honoured
him despite his neglect of Buddhist discipline and in spite of
attacks heaped upon him.
Kumarajiva and his contemporaries

Kumarajiva was not only a scholar; he was an institution in
the true sense who drew votaries to his shrine of his learning
both in Kucha as also in the Chinese capital where he lived till
the end of his life. Reference might be made to the lives and
activities of those mentioned earlier, namely, Vimalaksa, Dhar-
mamitra, Buddhayasa and Buddhabhadra. The first one Vimala
ksa,94 name translated as Wu-keu-yen meaning ‘without-dirt-eye’
was a Sramana of Kubha (Kabul). He was a great teacher of
Vinaya in Karasahr where Kumarajiva was one of his disciples.
When Kucha was invaded by Lu-Kuang in A.D. 383 he escaped
to the western country. Later on, he went to China in the year
406 and was cordially received by his pupil Kumarajiva, but
after the latter's death in 413 he left for the south and arrived at
Cha-tchouen (Ting-yen) and settled down in the monastery of
Tche-cheu-kien. With his sound knowledge he completed the
Vinaya of the Sarvastivadins which Kumarajiva had at first
undertaken but could not complete. Amongst his collaborators
was Huei-kouang, a monk of profound learning. Vimalaksa died
at the age of 77 years in the Kien-sse monastery. His two works
are Sarvastivdda-Vinayanidiuia which exists, constituting a part
of Dasddhydya- Vinaya, and another one which is lost.
Dharmamitra,95originally from Kashmir came to Kucha where
he was accorded due respect and he stayed in the capital for a
couple of years. Then he wanted to leave for the East but the
•
94. For an account of Vimalaksa’s life and his contributions, see Bagchi :
Le Canon Bouddhique, pp. 338-339; Nanjio’s Catalogue, ii, 44, p. 400.
95. See Bagchi : Op. cit, pp. 388-391; Nanjio : Catalogue, ii, 75, pp. 414.
king was opposed to it. He had, therefore, to move in secret for

Liang-chou from where he left for the South. He worked in
China from A.D. 424 till 441 and died in his eighty-seventh year
in A.D. 442. Ten works are ascribed to him, five of which are
said to have been lost. Those works preserved in the Oxford
collection and quoted by Nanjio include Akasagarbha-bodhisattva-
dharanl-sutra, as also Dhydna-sutra, Sarvadharma-raga sutra, Stri-
vivarta-vyakarana-sutra and Samantabhadra-bodhisattva-dhyana
caryadharma-sutra.
Buddhayasa,96 translated in Chinese as Kio-ming, was from
a Brahmin family of Kashmir or Kubha, according to Nanjio.
His father, not believing in the religion of the Tathagata, had
insulted a monk and he had to pay for this sin in the form of
physical ailments. The son, however, was attached to a monk at
the age of 13 and he made a serious study living in a monastery.
At the age of 19, he could recite millions of words of Hinayana
and Mahayana texts. A proud young man, as he was of his learn
ing, he was not very popular and did not join holy orders till
the age of 27. Being not satisfied with his knowledge he left
Kashmir for the kingdom ofCha-le (Kashgar) where the Crown
prince Dharmaputra appreciated his talents and invited him to
stay in the palace. At this time Kumarajiva had also come there
and he studied with him before returning to Kucha with his
mother. The invasion of Lu-Kuang, the Chinese general, the fall
of Kucha and the taking of Kumarajiva as a prisoner to China
made YaSa desperate and uncomfortable. Later on he went to
Kucha and from there left for China to join Kumarajiva,
ignoring the advice of the ruler and quitting secretly. At Chang-
ngan, where he settled down with his former pupil, he collabora
ted with him, and translated four works into Chinese including
the Dirghdgama and Dharma-guptaka-vinaya between 410 and
413 A.D. He subsequently returned to Kashmir.
Buddhabhadra97 was a contemporary of Kumarajiva. He was
born at Nagaraharaand claimed direct descent from Amritodana,
the uncle of Buddha. He became an orphan at an early age,
having lost his parents, and was admitted to the Buddhist order.
96. Bagchi : Op. cit, pp. 200-204; Nanjio : Op. cit, ii, 61, p. 408.
97. Bagchi: Le Canon, pp. 341-346; Nanjio: Catalogue, ii, 42, pp.
399-400.
He completed his studies at the age of seventeen and went to

Kashmir along with his companion Sanghadatta but did not stay
there for long. He was interested in carrying the message of
Buddha to different countries and also to study their customs.
At that time Che-yen who was travelling in India with Fa-hien
came to Kashmir and desired a scholar of repute to accompany
them to China. Buddhabhadra availed of this opportunity and
left with Che-yen on foot, not by the Central Asian route but
probably through Burma to Tonkin from where they sailed by
boat to China. When Kumarajiva learnt about his arrival in
China and working at Chang-ngan, he personally went there to
meet him. Buddhabhadra was a scholar of repute and was equally
proud of his learning. He was invited by the Chinese Emperor
to go to the capital and translate Buddhist texts there. He went
to Nanking in 421 where he remained till his death in 429. This
Indian scholar, whose name literally means ‘intelligence-wise’,
translated between A.D. 398-421 13 or 15 works at twodifferent
places, namely at Lu mountain and Kien-khan, the capital.
Kumarajiva acknowledged his superiority and whenever he had
doubts, he always sought Buddhabhadra’s explanation. He also
collaborated with Fa-hien in the translation of Mahasahghika
Vinaya which he had brought from Pataliputra. The general
plan of the work is similar to that of other Vinayapitakas but it
is much richer in cultural contents.
While some reference has been made earlier to Sanghabhuti,
Gautama Sanghadeva and Puniyatrata who went to China by
the end of the fourth century A.D., their contribution to
Buddhism and its literature, as also collaboration with Kumara
jiva no doubt call for detailed reference to their lives and acti
vities. Sanghabhuti989 from Kashmir had reached the northern
capital in A.D. 381. He was welcomed by some of the leading
Buddhist savants of China; and at their request he translated a
number of texts into Chinese, including a major one—an ex
haustive commentary on the Vinaya-Pitaka of the Sarvastivada
school. He was there till A.D. 384. Another scholar from
Kashmir—Gautama Sanghadeva,89 came to Chang-ngan in 384.
98. Bagchi : Le Canon, pp. 160-161; Nanjio ii. 54, pp. 404.
99. Bagchi : Op cit, 161-162; Nanjio : Op. cit, ii, 39, p. 399.
A profound scholar and a born teacher, he specialized in the

metaphysical literature of Buddhism (Abhidharma). He remained
in the north for a few years and availed of the opportunity in
picking up Chinese language. In 391 Sanghadeva went to China
where a strong Buddhist school had been founded by the Sogdian
monk Senghui. He was also invited at this time by a Chinese
Buddhist scholar Hui-yuan to his institution at Lu-Shan. Here
too Sanghadeva left his impress translating a few Sanskrit texts.
From Lu-shan he went to Nanking in 397 where- he impressed
the official circle, and one of the functionaries had a monastery
built for him. With the assistance of his Chinese friends and
followers from Kashmir he translated a number of important
Buddhist texts into Chinese, probably staying there till his
death.
Punyatrata100 and his pupil Dharmayasa,101both from Kashmir,
also went to China and were associated with the translation of
a number of important Buddhist texts of the Sarvastivada school.
Punyatrata collaborated with Kumarajiva in China. His pupil
Dharmayasa, who had joined his master at the age of 14, soon
attained great name and fame for his erudition. He left Kashmir
at the age of thirty and having travelled in various parts of
Central Asia reached China sometime in the last quarter of the
fourth century A.D. He stayed there till 453, spending a number
of years in the south as well, translating a number of works
with his associates. Later on he returned to Central Asia, and
finally was back home in Kashmir. It is said about Punyatrata that
he came to China towards the end of the fourth century and
collaborated with Kumarajiva in A.D. 404. Probably both had
met earlier in Kucha and Punyatrata’s visit to China might have
been in response of Kumarajiva’s invitation to his friend and
past colleague for future collaboration. /
Among the Buddhist savants of the first quarter of the fifth
century A.D., more or less contemporaries of Kumarajiva, were
Buddhajiva102—also from Kashmir—who had come to South China
reaching Nanking in A.D. 423. His association and collaboration
with Fa-hien need not be repeated here. It is fairly certain that
100. Bagchi : Op tit, pp. 176-177.
101. Bagchi : ibid, pp. 174-176; Nanjio, ii, 62, p. 408.
102. Bagchi : Op. cit, p. 363-364; Nanjio : Op. tit, ii, 73, p. 414.
he translated a couple of Fa-hien’s manuscripts brought from

India. He probably stayed in China till his death. He is called
Ko-to-she translated as Kiao-she in Chinese literature. A great
teacher of Vinaya in Kashmir, and a follower of the Mahisa-
saka School, he translated three works of the same school bet
ween A.D. 423-424. These include Mahisasaka-vinaya and
Pratimoksa of the Mahisasakas.
Gunavarman103 was another great scholar—a prince of the royal
family of Kashmir—who reached China (Nanking) a few years
later by the sea route. Both his father Sarighananda and grand
father Haribhadra were banished from their kingdom—the latter
for his oppressive rule and the former for his father's lapses. This
young prince left the house at the age of twenty and lived as a
Buddhist monk. Mastering the Buddhist canon in all its sections,
and the agamas this young scholar, rejecting the offer of his
paternal kingdom, left his country for Ceylon from where he
subsequently went to Java. He was able to convert the Royal
family to Buddhism. His name and fame attracted the attention
of the Chinese Emperor, who personally invited him, and
Gunavarman reached Nanking in A.D. 431. During his short
one year’s stay at the Jetavana monastery he was able to trans
late eleven works into Chinese.
An inmate in the same Jetavana monastery was another
scholar from Kashmir, named Dharmamitra,104, who collabora
ted with him. His contributions are no doubt recorded
earlier.
Later Buddhist Savants
The contribution of Buddhist savants from Central Asia to
the history of Buddhism—its thought and literature—can be
assessed in terms of their translation of texts in China. In the
absence of detailed information about the life and activities of
these scholars, one could only trace their origin, the period of stay
in China as also their literary contributions in the form of trans
lation of earlier texts. Dharmaksema105 (Chinese Tan-mo-chian
103. Bagchi: ibid, 370-375; Hastings: Encyclopedia o f Religion & Ethics,
VIII, 701-6; Eliot : Hinduism and Buddhism, III, pp. 176-177.
104. Bagchi : Op. cit, pp. 388-391; Nanjio, ii, 75, pp. 414.
105. Bagch: Op. cit, pp. 212-221.
or Tan-wu-Chian) translated as Fa-Feng—‘land prosperity’

—originally from Central India, was a follower of Mahayana
Buddhism and at first went to Kashmir—then a great seat of Bud
dhist learning. From there he went to China through Central Asia
reaching Liang-Chou in the beginning of the 5th century. He
had to stay at Ku-tsang—then the capital of an independent
kingdom—and translated 25 texts into Chinese. He wanted to
returned to Khotan in A.D.433 much against the wishes of the
ruler. While attempting to do so he was killed in the way. He is
also attributed the translation of the famous Kavya of ASvaghosa
—Buddhacarita (Fo-Pen-hing-king) in five chapters. The incom
plete manuscript of the Mahaparinirvanasutra—which he had
brought from India—was translated by him into Chinese. The
rest of the manuscript which he wanted to get in his second visit
to Khotan, actually cost him his life.
About the same time a pupijl of Dharmaksema Tsiu-kiu-
Kingsheng106—a noble man of Liang-chou—went to Khotan—
then a great centre of Mahayana studies—and studied the texts
of this Buddhist school with an Indian scholar named Buddha-
sena who was a great Mahayana teacher, called She-tseu (Sirhha)
in all the countries of the west. Kingsheng on his return to China
translated the texts which he had studied in Khotan and had
brought with him. Khotan’s reputation as a Buddhist acade
mic centre with its famous savants attracted many Chinese monks.
Among them were eight monks from Liang-chou who came there
in 439 in search of Buddhist texts. At that time the quinquennial
assembly was being held there. These monks took dictation of
texts which they carried with them to China. A Chinese monk
named Fa-ling107 brought from Khotan a manuscript of the
famous Avatamsaka-sutra which was translated by Buddhabhadra
106. Bagchi : Op. cit, pp. 221 -223 ; 394-398.
107. Fa-ling was a disciple of Hui-Yuan who was a great Sanskrit scholar,
who did not translate any text into Chinese, but devoted a greater portion of his
time to the work of organization. It was at his suggestion that the whole of
the Sarvastivada-vinaya was translated into Chinese. He sent a number of
scholars to KumUrajiva to be trained by him, and a batch of disciples, Fa
tsing, Fa-ling and others to the desert and snowy mountains (Central Asia)
in search of Sanskrit manuscripts. After several years they came back with
Sanskrit texts which were later translated into Chinese. (Bagchi : India &
China, p. 100)
in 418. Another monk named Fa-hien108 brought from Khotan

in 475 the manuscript of the Saddharma-puniarika which was
translated by Dharmamati into Chinese in 490.
Khotan continued to play an important role in the transmission
of Buddhism to China in the Tang period (A.D. 618-907). One
of the greatest scholars of this period was Siksananda109 who
went to China from Khotan and worked there till his death in
710. He translated in all 19 texts, the magnum opus being the
Mahavaipulya or the Avatamsaka-sutras, in 80 chapters. The
original text named Buddhdvatanisaka-mahu-vaipulyasutra, which
was earlier translated into Chinese in the beginning of the fifth
century by the Kashmirian scholar Buddhabhadra —Siksananda
translated it a second time in the seventh century A.D. Another
scholar from Khotan was Devaprajna110 or Devend rajnana. He
went to the Chinese capital in A.D. 689 and in all translated
six works. According to Nanjio,111 eight works are ascribed to
him, including a part on the practice of compassion in the
BuddhavatamSaka-vaipulyasutra, a part on the Acintya-vi$aya
and Sarvabuddhahgavati-dharani. Shih-Kiyen112, a son of the
king of Kustana (Khotan) was another dignitary, who became
a&ramana in A.D. 707. He went to China as a hostage. In 721
he translated four works into Chinese, including Sutra on the
good law determining the obstacle of Karman, Anantamukha-
sddhaka-dhdrani-sutra on the lion king SudarSana’s cutting his
flesh to feed others, and a collection of important accounts
taken from several sutras on the practice of a Bodhisattva who
practices the Mahayana.
There are references to several other Buddhist scholars from
other parts of Central Asia who went to China and translated
Buddhist texts. These include Mitrasanta113—translated Tsiu-
108. Bagchi : Le Canon, p. 409.
109. ibid, p. 517. This name is faithfully transcribed in Chinese as Cheu-
nan-to. He was originally from the country of Yu-tien (Khotan) situated to
the north of Tsong-Iing (Pamirs).
110. Bagchi : Le Canon, pp. 514-16. He was a Sramapa of Khotan.
111. Catalogue, ii. 143, p. 439.
112. Nanjio : Catalogue, Appendix II, 152, p. 443. Shib-K-yen’s original
surname and cognomen according to Nanjio were Yu-Kh’lo.
113. ibid, II, 147, p. 440.
yiu, literally ‘calm friend’. He was a sramana of the country of

Tu-kwa-lo i.e. Tukhara who translated one work in A.D. 705,
named Vimalasuddha-prabhasa mahadharani'. Ratnacinta114
(Chinese O-ni-panna) translated >as ‘Jewel thinking’ was a
Sramana of Kia-shi-mi-lo i.e. Kasmira. He translated seven
works and died in A.D. 721 attaining the age of more than a
hundred years. His works include Pratibimbdbhishiktagurjia-sutra^
Amoghapasa-hridaya-mantraraga-sutra, Sutra on the Dharani-
riddhimantra of great freedom to be obtained by one who wishes
for it. Another sramana of Kia-Si-mi-lo or Kasmira was Thien-
si-tsai—literally ‘heaven or god’ ( = deva) ‘stopping misfortune’,
who arrived in China in A.D. 980 and worked at translations for
twenty years. In A.D. 982, he received the title Min-kiao-ta-sh and
died in A.D. 1000. Eighteen works are assigned to him which in
clude Ghanavyfdia-sutra, Sukarma-du/jkarmaphala-vise$aiia-sutra
etc. A Sramana from Kubha (Kabul) named Pan-jo116 or Prajna,
translated as Che-hui, was educated in the Buddhist lore in
Kashmir. He went to China in A.D. 781 and settled down at
Chang-ngan in 810. He translated eight works into Chinese which
include Buddhdvatamsaka-Vaipulya-sutra. A more prolific scholar
was Danapala116—‘the gift protector’—She-hu in Chinese. He
was a Buddhist scholar of Uddiyana in north-western India and
arrived in China in A.D. 980. He worked there at translations for
some years. Two years later he received from the Chinese em
peror the title of Hien-kiaoa-ta-sh. In the Nanjio’s catalogue,
111 works are ascribed to him, including Mahasahasra-pramar-
dana-sutra, Mahdydna-ratnacandra-kumara-pariprikkha-sutra,
Mahdycina-cintyarddhivifaya-sutra etc.
Tibet too was not quiet in this direction, and there is a reference
to a monk Pa-ho-sz-pa117 or Pa-sz-pa—Bashpa—a §ramaya of
the country of Tu-po (Tibet)who was the Ti-sh, literajly ‘emperor’s
teacher’. He translated one work in A.D. 1271. He died in A.D.
1280 in. his forty-second year. It is said that he had become a
confidential adviser of Kublai Khan during the latter’s career of
conquest in China. In A.D. 1260 he was'named Kwo-sh—
114. ibid, II, 148, p. 440.
115. Nanjio : Op. cit, II, 156, p. 448.
116. ibid, II, 161, pp. 453-454.
117. ibid, II, 169, pp. 457-458.
‘Preceptor or Hierarch of the State’, and recognised as the head of

the Buddhist Church. In A.D. 1269 he constructed an alphabetic
system for the Mongol language which was utilised for writing.
In reward for his services he received the exalted title of ‘Ta-
pao-fa-wan’ or ‘Prince of the Great and Precious Law (of
Buddha)’. His solitary contribution is Mulasarvastivada-nikaya-
pra vragyopasampada-karmavaka.
Buddhism and other Religions in Central Asia
The study of Buddhism in Central Asia is recorded on the

basis of the accounts of Chinese pilgrims who passed through
this vast region and have noticed the number of monasteries as
also that of monks. The Chinese source material is equally rich
in providing information on the lives and activities of these
Buddhist savants of Central Asia who went to China to translate
sacred texts. The archaeological finds of stupas and sculptures
as also paintings equally testify to the flourishing state of
Buddhism—both Hinayana and Mahayana. It is equally propo
sed that there were two streams of Buddhism and there were
running contacts between India and the Oxus region. The use of
Prakrit and of various Iranian idioms point to actual coloniz
ation from these two quarters. According to the Chinese pilgrim,
Shan-Shan (near Lob-nor), Turfan, Kucha and Kashgar were
HInayanist, whereas Yarkand and Khotan were Mahayanist
centres. The terminus a quo of Buddhist art, religion and
philosophy could be any time in the first century A.D., though
traditional accounts, especially in relation to Khotan could place
this event a couple of centuries earlier. The earliest Chinese
notice of the Buddhism in Kashgar and Kucha date from 400
(Fa-hien) and the third century (Annals of the Tsin 265-317)
respectively. In Turfan the first definite Buddhist record is the
dedication of a temple to Maitreya in 469 and probably the
history of Buddhism in this region was similar to that in Kucha.
Khotan was probably the starting point of Buddhist activity in
Central Asia, and it was carried from there to China. This
Mahayanist Buddhism of Khotan was a separate stream and,
according to Hsuan-tsang, it came from Kashmir. While the
Mahayana school was predominant in Khotan, it was not to the

exclusion of the other school. The Chinese pilgrim Fa-hien
(c. 400) found the Hinayana prevalent in Shan-Shan, Kucha,
Kashgar, Osh, Udyana and Gandhara. Hsuan-tsang also notices
its presence in Balkh, Bamiyan and Persia. Both the schools
co-existed everywhere though disputations amongst their
votaries were not unknown. An interesting passage in the ‘Life
of Hsuan-tsang’118 relates his dispute with a Mahayanist scholar
in Kucha which was a centre of the Hinayanists. This monk
held that the books called Tsa-hsin Chu-She and Pi-Sha were
enough for salutation(apparently referring to Samyuktabhidharma-
hridaya (Nanjio 1287), Abhidharmakosa (ibid 1267), Abhidharma-
vibhasa (ibid 1264)and Yogikarabhumi(Nanjio 1170). Hedenoun-
ced the Yogasastra as heretical to the great indignation of the
pilgrim, who no doubt accepted this work, its importance
being revealed by Maitreya to Asaiiga.
Brahmanism
Besides Buddhism with its two most important schools that
of the Hinayana and Mahayana, there are traces of other re
ligions and cultures mingling in the Tarim basin and the lands
of the Oxus. It is quite likely that besides Hinduism, Zoroas
trianism and Christianity as also Manichaeism not only flourish
ed but also interacted with their impact on Buddhism. The
available evidence sheds light on the co-existence of these re
ligions as also the mutual impact of Buddhism and Hinduism
and of the former with the other two. Hinduism—in its two
facets—Vaisnavism and 6aivism with its gods and goddesses
also came to be known in Central Asia quite early. The famous
record from Vidisa119 inscribed on the column refers to Helio-
dorus, son of Dion, as a worshipper of Bhagavata. He had come
there from the court of Antialkidas of Taxila, during the four
teenth year of the reign of the Indian king KaSiputra Bhaga-
bhadra. This ruler of Taxila could be placed sometime in the
last quarter of the second or the first quarter of the first century
B.C. The Krisna legend seems to have travelled outside India
during the second century B.C. according to Zenob’s story of
118. S. Beal : Life o f Hiuen-Tsiang (London. 1911), p. 39.
119. Luders : List of Inscriptions (EI.X. Appendix)
the Indians in Armenia. This story refers to two Indian chiefs

called Gisane (Kisane) and Demeter (Teraeter) having sought
shelter with Valarashak or Valarsaces of Armenia (c. B.C. 149-
127 B.C.). Later on, they were put to death, but their sons lived
there and erected two temples of their gods (Gisane and Demeter).
These were later razed to the ground by Saint Gregory despite
stiff resistance.120 It is suggested that both were Brahmanical
deities—either Krisna and Balarama or some solar ones.
The Narayana cult, probably associated with Vaisnavism, how
ever, had not so uncomfortable an end—in another region of
Central Asia. A fragment of an inscription discovered by A.N.
Bernshtam in Tadjikistan in 1956 refers to the cult of Narayana
—‘Narayana be victorious’—jayato—as the inscription suggests.
On palaeographic grounds the inscription can be attributed to
the second-first century B.C. Harmata considers121 the identity of
this Narayana at great length in his long paper. In case this
Narayana is equated with Vi$nu, then the record points to Brah
manism preceding Buddhism in Central Asia. On the other
hand, Narayana, the Buddha, is mentioned in the Khotanese,
Saka documents in Eastern Turkestan, and Narayana, the deva,
occurs in the Buddhist Sogdian documents.122 In a Tun-huang
painting Narayana on Garudais shown as attendant of Avaloki-
tesvara.123 This vahana Garuda is closely associated with Visjju
and it is likely that the artist had the Brahmanical deity in mind
while painting his subject.
Reference to Vi$nu worship along with that of Siva and Surya
is evident from the famous Nicolo seal which was first noticed
by Cunningham.124 According to him, the device consists of the
four-armed Visnu with a devotee standing by his side with folded
hands showing obeisance. There is also an inscription, correctly
120. JRAS. 1904, p. 330.
121. Acta Orientalia, Hungary, Tome XIX, 1966, pp. 1-32.
122. For the Saka and Sogdian data, Harmata refers to Bailey, BSOAS.
X. 1942, pp. 909 and 914; & E. Benveniste : Vessantara Jataka (Paris 1916),
pp. 58-59 and Texts Sogdiens (Paris 1940), p. 107.
123. Arthur Waley : A Catalogue o f Paintings recovered from Tun-huang
(1931), p. 54. See also P. Banerji’s Paper entitled ‘Hindu deities in Central
Asia’ in India's Contribution to World Thought and Culture (Vivekanand
Memorial Volume, Madras, 1970, pp. 281-287.)
124. Numismatic Chronicle (N.C.) 1893, pp. 126-127, PI. X, Fig. 2.
132 Buddhism in Centrat Asia
deciphered by Ghirshman.125126 This record in Tocharian language

mentions the name of Mihira, Visnu and Siva. The devotee,
according to the late French scholar, is not the Kusapa king
Huvi$ka, but some known Hephthalite chief, and the use of the
Tocharian language and script suggests that a composite cult
of Siva, Vi?pu and Surya was popular with certain peoples of
Central Asia about 500 A.D.
Definite Vaisnavite influence is also traced in a Buddha image
from Balawaste126 in the Domko region on the southern silk
route. This image, now in the National Museum as part of the
Stein collection of wall paintings, is assigned about the eighth
century A.D. The body and arm of the figure are covered with
symbols or devices including the Srivatsa, diamonds, mandara
as churning rod, horse Uccaihsrava (suggestive of the story
of the churning of the oceans), the sun, the moon, vajras, books,
triangle and circles—all mystic elements depicting the Vihariipa
aspect of the Buddha on the model of the Kri?na ViSvarupa of
the Gita. This painting is suggestive of the cosmic aspect of the
Buddha, and not the Sakyamuni.
Saivism, however, seems to be more popular in Central Asia
with its amalgam with Buddhism effected by Asanga, the well-
known philosopher of about A.D. 400. The Saiva pantheon was
accepted by the Buddhists, and it held the imagination of the
Central Asian peoples for a long time and over a wider area. It
is well-known that the figure of Siva appears on certain coins of
Gondopheres, Maves and others. An elaborate Siva pantheon
is represented on the coins of the Kusana rulers with Wima
calling himself MaheSvara—a devotee of MaheSa.Nana is identi
fied with Amba or Uma as consort of Siva.127 Several Saiva
125. Les Chionites : Hepthalites, pp. 55-58, Fig. 65'and PI. VII.
126. Bussagli : Central Asian Paintings, PI.; Banerji : Op. cit, p. 282.
127. For a study of deities on Kusana coins, see Puri: Kufanas, Appendix
A, p. 213 ff. The epithet ‘deva’ applied to Gondophemes on coins is signi
ficant, likely to mean Siva and no other god. Huang-tsang in his Si-yuki also
refers to a Siva temple outside the gate of the city of Pu§kalavati, simply, as
a Deva temple in the seventh century. An important Saiva image is the so-called
TrimQrti with Siva as the central Figure (c. third century A.D.) from Akhun
Dheri near Charsadda, now in the Peshawar Museum. (Baneiji : Op. cit,
p. 283).
sculptures and Brahmanical figures are found in Afghanistan128

like the head of Durga overcoming Mahisasura. They are all of
the seventh or eighth century A.D. probably later of the time of
the Hindu Sahi rulers of Afghanistan. The inscribed Mahavinaya-
ka or Gane§a with Ordhvamedhra—phallus erectus—clad in a
tiger skin—now in the Kabul Museum, is another Brahmanical
divinity.
The popularity of Saivism extended to Sogdiana and Eastern
Turkestan, as is evident from a fragmentof a wallpainting from
Penjikent129 on the river Zervashan. In this Siva is portrayed with
a sacred halo and a decorated sacred thread ( Yajhopavita) and
stands in alitjha pose. He is clad in a tiger skin while his atten
dants are clad in Sogdian dress, ornamented long coat with open
collar.
Certain finds in Dandan-Uiliq and neighbouring sites of the
Taklamakan desert suggest extension of Saivism to Eastern or
Chinese Turkestan. A very interesting representation of Siva
on a painted wooden panel (D. VII.6) shows him three-faced
and four-armed seated cross-legged on a cushioned seat supported
by two couchant bulls. The rich diadem on the central head with
the side ornament resembling a half moon, the third eye on the
forehead, the tiger skin forming the loindoth (dhoti), and finally
the bull as his vahana, are all symbolical of theBrahmanic God
Siva.130 Another panel from Dandan-Uiliq depicts a four-armed
Trimurti form of Siva and his Sakti131 kneeling on her right
thigh. The god sits cross-legged, clothed in tight fitting long-sleev
ed white vest. Traces of the third eye are visible. Siva is four
armed with a massive armlet on the left arm, the other resting
on thigh and holding vajra. The lower right hand is thrown
round the neck of Sakti- The deity has a long neck-cord tied
into bows at intervals. His hair is long and wavy and the crown
is of Iranian type. The face to the left is fearful and the third
eye is clearly visible. The Sakti dressed in a long stola with
128. See Benjamin Rowland : Ancient Art from Afghanistan, pp. 107-108.
129. Banerji : Artibus Asiae. XXXI. 1969, also Vivekanand Volume,
Op. cit, p. 284.
130. Stein : Ancient Khotan, I, p. 279.
131. Dandan Uliq panel No. D.8—ibid. p. 261, pi. LXII. It is now in the
British Museum. (Banerji : Op. cit, pi. 52).
tight sleeves to wrist holds a cup in her right hand, which she
is handing to the deity. She has well-drawn eyebrows—highly
arched and long eyes, and is white-complexioned.
Another Trimurti figure from Eastern or Chinese Turkestan,
now in the National Museum, also points to the popularity
of Saivism in that area. The deity sits full face with head
slightly turned to left. The other two heads project either side
from behind ears. The central face has a third eye in the fore
head and a long thin moustache. The eyes are heavy-lidded and
dreamy. A skull is shown on the head against the back top
knot. The deity has four arms, the two upraised ones areholding
the sun to the left and the moon to right, a pomegranate in the
lower right hand, and the left one resting on the thigh, probably
grasping a vajra (?). The conspicuous £aiva features are the
third eye on the central head, the skull on the headdress and
the three heads with four arms.132
Besides Siva and GaneSa, Kumara, Kartikeya, Brahma, Indra,
the sun, the moon and the lokapalas—all connected with
Brahmanism, figure in Central Asian art. Brahma is noticed,
according to Coomarswamy,133 on the caves of Kucha area. A
fragment of wall painting from Balawaste, now in the National
Museum,134 is supposed to be that of Indra. The figure is either
kneeling or sitting cross-legged, his body leaning forward and
head tilting back. The eyes are downcast and hands folded and
uplifted to neck-level. The head is covered with a close fitting
cap with a head band in dark pink studded with pearls. The
face is Indian and the figure puts on various types of ornaments
and a mukta-yajnopavita. The presence of the eye on the hand
identifies the figure Indra.
The Tun-huang cave paintings depict Gane£a, Kumara, the
sun and the moon. Ganesa was no doubt very popular in Khotan
as seems evident from a number of bronze tablets and painted
wooden panels discovered by Stein at Endere, now in the British
132. Andrews : Catalogue o f wall Paintings from Ancient Shrines in

Central Asia. Bal. 0200, p. 87.
133. History o f Indian and Indonesian Art, London, 1927, p. 150 n.
134. Andrews : Catalogue, p. 13. See also Indo-Asian Culture, XVII. No.
4, pp. 14 ff.
Museum.135 This four-armed god with an elephant head is shown

seated, wearing tiger-skin, dhoti and tight trouser. He has a
rosary of dots, and holds a bowl or fruits or sweetballs, a spear
head, a turnip and an axe. Dandan-Uiliq has also certain wooden
panels depicting Ganesa,136 and so also in Khadlik where he is
shown seated on an open lotus. A fine figure of Gariesa in cave
285 of Tun-huang137 is equally interesting. Kartikeya is re
presented in a wall painting from Bezaklik,138 seated on a bird,
with one leg hanging down and the other tucked up in front.
Mahakala and Garuda are shown carrying away the nectar
after the famous sea churning. Mahakala's image with a demon’s
head seated on a yak-tail (Nandi), and Lokapalas13” fully
absorbed in Buddhism also figure in paintings in Central Asia.
The four well-known Lokapalas—Dhritarastra, Virudhaka,
Virupaksa and Vaisravana were assimilated in Buddhist art and
each was provided with a direction as its guardian. These
protectors of the four quarters are shown in Central Asian art
as warrior kings gorgeously dressed with armours, sometimes
accompanied by Yaksas or demons.
Besides Brahmanical divinities depicted in Central Asian art,
and mythological legends faithfully brought out in artistic
presentations, the Rama legend as also the other epic stories
were also known to the Central Asian people in ancient times.
The story of Ramayana in Khotan legend is somewhat different,
but the reference to Rama, Laksmana, Sita and DaSaratha and
also ParaSurama is significant. Prof. Bailey has gone140 into the
question of Ramayana in Central Asia thoroughly. Scholars who
have delved into Central Asian studies also took notice of many
names of epic heroes—Bhima, Arjuna and several others as pro
per names in Kharosthl inscriptions. There are scores of Brah-
135. Stein : Ancient Khotan, pp. 431, 442.
136. ibid, pp. 292 ff.
137. Basil Gray : Buddhist Cave Paintings at Tun-huang, pp. 1 -20.
138. Andrews : Catalogue o f Wall-Paintings (Op. cit), pp. 35-36.
139. Bussagli : Central Asian Paintings (Op. cit); also Stein : Ancient
Khotan, I, pp. 494 ff.
140. See his paper ‘The Rama Story in Khotanese’ (Journal of the
American Oriental Society. JAOS—LIX, pp. 460-8; also Bulletin of the
School of Oriental and African Studies—BSOAS, 1949-50, pp. 121-39; 920-
938.)
manical names in such records which might as well point to

Brahmanism flourishing in Central Asia where these records
were found. It is equally significant that the Yueh-chi and their
descendants or associates, the Kusanas, should have so soon
accepted Saivism—Wima Kadphises calls himself Mahesvara
or a devotee of Mahesa or Siva while Siva and his Nandi figure
on the Kusana coins all through from this Kusana ruler on
wards. Skanda, Kumara and VaiSakha are also depicted on the
coins of Huviska. Gane§a, another Brahmanical deity—son of
Siva—became very popular in Central Asia141 as also in Mon
golia, China and Japan, as the remover of obstacles ( Vighna
naiaka). One might as well refer to the famous Mathura inscrip
tion142 of the year 28 of Huviska recording dedication of 550
Puranas each into the two guilds of samitakara and dhahgika—
wheat flour and corn dealers respectively, for the exclusive bene
fit of Brahmins. The donor hailed from Wakhan or Badakshan.
This religious impulse appears to have been deep-rooted.
The available evidence—artistic and epigraphic—no doubt
suggests that Brahmanism—both Vaisnavism and Saivism with
their divinities, epic heroes and legends, found their way to
Khotan and other parts of Central Asia where Buddhism, of
course, was the dominating religious force. Brahmanism came
closer to Buddhism and several Brahmanical divinities figure
independently. Lokapalas or guardians of the quarters were
easily incorporated and assimilated in Buddhism, especially
Mahayanism. With the passage of time it came closer to &aiv-
ism and Vaispavism—presenting the best of the Indian religious
systems and the composite nature of Indian culture.
Manichaeism, Nestorian Christianity and Buddhism
Manichaeism, the religion of Mani143 or Manes, is one of
those systems, usually classed together under the name of Gros-
141. See Getty : Ganesa, Oxford 1936, p. 40; Stein : Ancient Khotan,
p. 221; Shivarama Murti : Ganeia’, also M.K. Dhavalekar: *Ganesa beyond
the Ind'an Frontiers' (Vivekanand Volume : Op. cit, pp. 1-15).
142. EL XXI. pp. 55 ff.
143. Mani is said to be of noble birth, and like Zoroaster, the Buddha
and Jesus, claimed to have been sent by God to fulfil what had been previously
revealed. He preached a new universal religion which like Christianity em
ticism. This religion arose in Babylonia about the middle of the

third century A.D. and during many generations exercised great
influence both in the East and in the West. Of course, very little
of the Manichaean literature has survived. The fragments of
manuscripts concerning this religion were discovered in Central
Asia144 thus confirming its existence in that area. The informa
tion about it being meagre, it is difficult to provide a fuller
picture. Reference might as well be made to the writings of non-
Manichaean authors most of whom wrote their accounts in a
hostile manner. By far the most celebrated of the western
authorities on Manichaeism is Augustine who was for nine
years (A.D. 373-382) a professor of Manichaean thought. He
represents Faustus as one of the ablest and the most influential
among the Manichaeans, but it is doubtful if he could read the
sacred books of his religion in the original Aramaic. The
Mohammadan literature relating to this religion does not begin
before the 9th century A.D. According to Al-Biruni,145 Mani
himself says that ‘wisdom and deeds have always from time to
time been brought to mankind by the messengers of God. So
in one age they have been brought by the messenger called
Buddha to India, in another by Zaradusht (Zoroaster) to Persia,
in another by Jesus to the West. Thereupon this revelation has
come down, the prophecy in this last age, through one Mani,
the messenger of the God of truth to Babylonia.’ Such was
braced all races and conditions of men. His doctrines were derived from the
cults of Babylonia and Iran and were influenced by Buddhism and Christianity.
Ghirshman : Iran (Pelicans), p. 315.
144. Encyclopedia o f Religion and Ethics (ERE) VIII, pp. 394. These frag
ments are written in various languages (Persian, Turkish and Chinese). They
have been published by F.W.K. Muller, A. Von Le Coq, C. Salemann,
Chavannes and Pelliot. (Consolidated references provided in ERE, Vol. VIII,
394, Col. i, n. 1). The discovery of the Turkish version of the Khuastuanift— a
Manichaean Confession prayer amidst Chinese Buddhist texts and monastic
records from the Polyglot Library in the Turfan area is very interesting. It
shows how easy it was for Mani’s Church in Central Asia to share the same
sacred site with Buddhist cult (Stein : Serindia, Vol. II, p. 819 and 921). The
Chinese Manichaean texts were found at Chien-fo-tung. A Manichaean work
in Chinese has been translated and annotated by Chavannes and Pelliot
(JA. November-December, 1911, pp. 499-617),
145. Stein : Serindia, p. 819.
the claim put forward by Mani. The extreme simplicity of their

cult, and in particular their abhorrence of idolatry, seems to have
saved them from molestation. It is difficult to fix the exact
date of introduction of Manichaeism into Central Asia and
China. The finds in a polyglot library146 at Tun-huang included
a remarkable manuscript exhibiting a third variety of the
Syriac script transplanted to Central Asia, and the one which
discovered at the ruined sites of Turfan first revealed as pecu
liar to Manichaean writings. It was an excellently preserved
narrow roll of paper about 15' long containing the beautifully
written and almost complete text of a Manichaean confession
Prayer147—a Turkish version of the K/ivastuaniff. The discovery
amidst Chinese Buddhist texts and monastic record of this
Manichaean relic was interesting. The Turfan excavations and
finds point to the existence of Mani’s church in Central Asia,
sharing the same sacred site with the Buddhist cult, and with
Christian worship too, though remaining for centuries a
formidable rival to both of them.
The presence of the Turkish-speaking Manichaeans at Tun-
huang is attested by another important find from the walled-up
chapel—the perfectly preserved small book in Turkish ‘Runic’
script. The discoveries at Turfan sites furnish abundant proof
for the existence of Manichaean and Buddhist worship existing
side by side among a population which had come relatively
early under Turkish influence as well as domination. Mani
chaean propaganda seems to have secured a firm foothold in
that part of Central Asia as well as in China in Tang times.
This is also confirmed by the finds of Chinese Manichaean texts
found at Chien-Fo-tung. Pelliot discovered a fragment of a
Chinese treatise setting forth points of Manichaean doctrine.148
146. Quoted by E.R. Bevan in his article in the Encyclopedia o f Religion

and Ethics, p. 396, Col. r, n. 3. He refers to Alberuni’s book on Chronology
(E. Sachau : Leipzig, 1878 and translated by him into English— The Chro
nology o f Ancient Nations, London, 1879).
147. For an edition and annotated translation of this text, see Von
Lecoq’s paper 'Dr Steins Turkish Khuastuanift from Tunhuang', JRAS, 1911,
pp. 277-314.
148. See Chavannes and Pelliot ‘Un traite Manicheen retrouve en Chine'—
JA. 1911. 1913, which is a most valuable contribution to our knowledge of
The admixture of Buddhism in Manichaeism is equally re

markable. The discoveries made in Central Asia seem to support
the Chinese edict of 739 accusing Manichaeism of falsely taking
the name of Buddhism and deceiving the people. This is not
surprising since Mani is said to have taught, as pointed out by
Al-Biruni, that Zoroaster, Buddha and Christ had preceded him
as apostles and in Buddhist countries his followers naturally
adopted words and symbols familiar to the people. Thus, Mani-
chaean deities are represented like Bodhisattvas sitting cross-
legged on a lotus. Mani receives the epithet Ju-lai or Tathagata
as in Amida’s Paradise. There are holy trees bearing flowers
which enclose beings styled Buddha, and the construction and
phraseology of Manichaean books resemble those of a Buddhist
Sutra. The patrons of Manichaeism in Central Asia were the
Uighurs who were established there in the eighth and ninth
centuries A.D. and around 750 A.D. their Khan adopted Mani
chaeism as the State religion.149 The many manuscripts in Sog-
dian and other Persian dialects found at Turfan show that it had
an old and close connection with the west. Mani’s teachings
seem to be influential in Central Asia, but not before A.D. 700.
Nestorian Christianity also existed in the Tarim basin and was
quite prominent in the seventh century A.D. This is in agree
ment with the record of its introduction into China by A-lo-pen
Manichaeism in Central Asia and China. (Eliot : Hinduism and Buddhism,

III, p. 216 n. 3)
149. The most influential cultural event in the history of the Uighur
Khanate was the conversion of the rulers to the Manichaean religion under
the third Kaghan in A.D. 762. This event is recorded in the trilingual ins
cription of Qara Balgasun (E. Chavannes and P. Pelliot : Un traite Mani-
cheen retrouve en Chine. Journal Asiatique, 1913, 177). The texts are in
Chinese, Sogdian and Turkish, of which only the first is satisfactorily preserved.
Apparently it was as a result of the Uighur occupation of the Chinese silk-
route terminus of Lo-Yang that the Kaghan was brought into contact with
Manichaean missionaries who had been established in China since A.D. 694.
The new creed—a syncretistic one, including elements of Zoroastrian, Christian
and Buddhist origin, was founded in Mesopotamia by its prophet Mani soon
after the rise of the Sassanian Empire in Iran which took place in A.D. 224.
The new creed penetrated early into Khurasan and Sogdiana under the
leadership of the apostle Mar Ammo. It was carried by its devotees of Sogdian
nationality and by tradition traders along the routes to China. (Hambay :
Central Asia, Op. cit, pp. 60-61)
in A.D. 635, almost simultaneous with Zoroastrianism. Frag

ments of the New Testament have been found at Turfan mostly
belonging to the ninth century, but one is of the fifth century.
The most interesting document relating to the history of this
faith is the Nestorian stone,15015bearing a long inscription partly
in Chinese and partly in Syriac, composed by a foreign priest
called Adam or in Chinese King-Tsing. It provides a long
account of the doctrines and history of Nestorianism. This
inscription also contains many Buddhist phrases, such as Seng
and Ssu for Christian priests and monasteries. It also omits
references to the crucification and simply speaks of the creation
that God arranged the cardinal points in the shape of a cross.
It reviews in some detail the life and activities of Christ. It is
suggested that the motive for omission must be the feeling that
redemption by his death was not an acceptable doctrine, and
the Nestorians as also the Jesuits were unwilling to give publicity
to the crucification. It is equally interesting to find that king
Tsing consorted with Buddhist priests and even set about trans
lating a sutra from the Hu language. Takakusu161 quotes a
passage from one of the catalogues of the Japanese Tripitaka
which states that he was a Persian and collaborated with a monk
of JCapiSa called Prajna.
On the basis of manuscript finds, there is clear evidence of
the co-existence of Buddhism and Christianity, and friendly
relations between Buddhist and Christian priests in China. It
is possible that in Western China and Central Asia Buddhism,
Taoism, Manichaeism, Nestorianism and Zoroastrianism all
borrowed from one another, though definite proof for this is
wanting. Buddhism, no doubt, was in strength and the most
important of all other religions. Its religious texts were translat
ed much earlier. The happy land sutra (Sukhavativyuha) and
Prajha-paramita (Nanjio 23, 5) were translated before A.D. 200,
and portions of the ‘Avatamsaka’ and ‘Lotus’ (Nanjio 100, 103,
138) before A.D. 300. The principal doctrines of Mahayana
Buddhism must have been known in Khotan much before the
150. H avert: La Stele Chrestienne de Singan Fu in varietes sinologues,

pp. 7, 12 and 20 quoted by Eliot in reference.
151. I-tsing, pp. 169, 223.
presence of Christianity there.152 So also were Turfan and

Kucha centres of Hinayana Buddhism.
The association and contribution of Zoroastrianism might
have contributed to the development and transformation of
Buddhism since the two were certainly in contact with each
other rather closely. Kaniska, the great Kusana ruler, as also a
patron of Buddhism, no doubt, portrayed many Zoroastrian
deities153 on his coins, and these are more copious and frequent
than tlie figure of the Buddha. This could be possible only if
Zoroastrianism was flourishing in his realm. According to the
information available from the Chinese sources, the two religions
co-existed at Khotan and Kashgar. The Tang Annals describe
not only Buddhism as flourishing in Khotan—one of the Four
Garrisons—but also the ‘cult of the Celestial god* by which the
Zoroastrian religion is to be understood.16415An official Chinese
document dating from the year 768 A.D., which was excavated
at Dandan-Uiliq, refers to the capital Kia-che (Kashgar) and
its people. It mentions worship of the ‘god of heaven’ which
according to Chavannes, recognizes a reference to the Zoroas
trian cult.165 Some hostile references to Buddhism are equally
traced in Persian scriptures with Buiti and Gaotema called the
heretic.156157
Some New Trends in Buddhism
As a result of mingling of different religious ideas, the co
existence of denominational institutions and their patriarchs in
Central Asia, mutual influence in a spirit of toleration and
assimilation is understandable. Buddhism, as such, is supposed
to have amalgamated with Zoroastrianism or Christianity, Confu
cianism and Taoism. Thus, the dated inscription167 of the temple
erected in Turfan, A.D. 469,isan amalgam of Chinese ideas—both
152. Eliot : Hinduism and Buddhism, Op. cit, IE, p. 218.
153. Stein : Zoroastrian deities on the Kusana coins, I.A.
154. Stein : Ancient Khotan, (Op. cit, p. 172.
155. ibid, p. 155.
156. Sacred books o f the East : SBE, Vol. IV, pp. 145, 209; XXIII,
p. 184; Eliot : Op. cit, p. 218.
157. Eliot : Op. cit, p. 216. An Chou, the prince to whose memory the
temple was dedicated, seems to be regarded as a manifestation of Maitreya.
Confucian and Taoist—with Indian ones. This record is in honour

of Maitreya, a Bodhisattva, known to the Hinayana but regard
ed in that area not merely as the future Buddha but as an
active and benevolent deity manifesting himself in many forms.
According to a tradition, the works of Asanga were revelations
made by him. This record also mentions Akasagarbha and the
Dharmakaya, and equally speaks of heaven (tien) as appointing
princes, and of the universal law (tao). The Central Asian
Buddhism while borrowing many personages frorfi the
Hindu pantheon, has also many others—Buddhas and Bodhi-
sattvas such as Amitabha, Avalokita, ManiuSri and Ksiti-
garbha—without clear antecedents in India. These might have
been the outcomes of other external factors. The most important
Buddhist divinity is Amitabha. He does not figure in the earlier
art and literature of Indian Buddhism. In the earlier part of the
Saddharmapundarika—‘The Lotus of the True Law’—he is just
mentioned whhout any special importance attached to him.
The Mahxyana-sraddhotpada-sdstraUB—‘The Awakening of
Faith’—ascribed to Asvaghosa—though not certain—mentions
Amitabha towards the end.'Reference is made by Asvaghosa to
a sutra (probably the Amitayus-sutra or Sukhavativyuha) on
Buddha Amitayusor Amitabha and his Buddhaksetra Sukhavati.
Further Amitabha’s paradise is just mentioned in the Mahayana-
sutrdlamkara159 of Asanga. Amidst these cursory and meagre
notices in Indian literature, and the nameless Buddha figures in
the Gandhara region providing uncertain inference about their
identity with Amitabha, it is quite possible that Amitabha15819160 and
158. Nanjio : Catalogue No. 1249, p. 274. This work composed by the
Bodhisattva A$vagho$a was translated by £ik$ananda, A.D. 695-700 of the
Than dynasty.
159. XII, p. 23; Eliot : Op. cit, p. 219. See Nanjio No. 1190, p. 262.
This work composed by the Bodhisattva Asanga was translated by Prabhakara-
mitra, A.D. 630-633 of the Than dynasty, A.D. 618-907.
160. Lokesh Chandra in his paper entitled ‘Iranian Elements in the
Formation of Tantrik Buddhism’ presented to the Symposium on ‘The Silk
Route and the Diamond Path’ held on 7, 8 November, 1982 at the University
of California suggests that the cult of Amitabha represents transcedental
tendencies in Buddhism. Sakyamuni, the Man, was replaced by Amitabha.
His historic Enlightenment was transcended into Supreme Enlightenment
whose illumination became the new dynamised centre reflected in the new
his concept were evolved in Central Asia where his worship

flourished in the early centuries of the Christian era. Two trans
lations of the principal Amidist scriptures were made into Chinese
in the second century A.D. and four in the third, all by native
scholars of Central Asia.
According to the Tibetan historian of Buddhism,161 Taranath,
Amitabha’s worship could be traced back to Saraha or Rahul-
bhadra, a great magician, and reputed to be the teacher of
Nagarjuna, who saw Amitabha in the land of Dhingkota and
died with his face turned towards Sukhavati. The name Saraha
does not sound Indian, probably a Sudra represented in Tibetan
scrolls with a beard and top knot and holding an arrow162 in his
hand. Thus, the first person whom tradition connects with the
worship of Amitabha was of low caste and bore a foreign name. He
saw the deity in a foreign country, and was represented as totally
unlike a Buddhist monk. While it cannot be proved that he came
from the lands of the Oxus or Turkestan, there seems little diffi
culty, according to Eliot, in accepting Zoroastrian influence on
this cult or worship.163 The main principles of Amidist doctrine
are that there is a paradise of light belonging to a benevolent
deity and those good men invoking his name would be led to
that region. The highest heaven (following after the paradises
of good thoughts, good words and good deeds) is called ‘Bound
less Light’ or ‘Endless Light’.164 In this connection reference
might be made to this region and its master, Ahura Mazda, who
are constantly spoken of in terms implying radiance and glory.
It is also a land of song, like that of Amitabha’s paradise
Tathagata Amitabha, with Sakyamuni changing into Amitabha, the attendant

acolytes also changed names. Brahma became Avalokitesvara and Sakra
became Mahasthamaprapta (circulated paper, p. 55). Further, according to
Lokesh Chandra, the development of triads Amitabha, flanked by AvalokiteS-
vara and Mahasthamaprapta, was influenced by the traditions of Mithra
flanked by his companions Rasnu (the god of Justice) and Sraosa (the god of
obedience) (ibid, p. 58).
161. Translation Schiefner, pp. 93, 105 and 303.
162. Eliot : Op. cit, p. 219.
163. ibid, p. 220.
164. Zenda-Avesta-trans. Dermesteter, Names. Sacred Books of the
East (SBE), Vols. IV, p. 239; XX11I, pp. 317, 344.
re-echoing with music and pleasant sounds.16516This paradise could
be reached and Ahura Mazda and the Archangels would show
the way to the pious*160 Further, it is also said in the Zoroastrian
text that whoever recites the ‘Ahuna-variya formula’, his soul
would be led by Ahura Mazda to ‘the lights of heaven*’167 The
repetition of Ahura Mazda’s name is repeatedly reported to be
efficacious enough to lead the person to paradise. It is proposed
by Eliot,168 that the chief features of Amitabha’s paradise are
Persian, only his method of instituting it by making a vow is
Buddhist. While numerous paradises are the outcome of Indian
imagination, and the early Buddhist legend tells of the Tushita
heaven, the Sukhavatl is unlike these early Buddhist abodes of
bliss. It appears suddenly in the history of Buddhism as some
thing exotic, ‘grafted cleverly on the parent trunk, but sometimes
overgrowing it’. Eliot equally poses the question of tracing
connection between Sukhavatl and the land of Saukavastan
governed by an immortal ruler and located by the Bundehist bet
ween Turkestan and Chinistan. While there is no etymological
relationship, it is likely that Saukavasta, being well-known as a
land of the blessed, might have influenced the choice of a signi
ficant Sanskrit word with a similar sound.169
This Zoroastrian influence is traced by this British diplomat
scholar even in the concept of Avalokita170 who is also connected
165. Eliot : Op. cit, p. 220. It may also be noticed that Ameretat, the
Archangel of immortality, presides over vegetation and that Amida’s pardise
is full of flowers.
166. SBE, XXIII, pp. 355-7.
167. Saddharmapuncfarika, SBE, XXI, p. 261.
168. Op. cit, III, p. 220.
169. ibid, p. 221 n.
170. Avalokita is considered not only ‘a great god’ but a ‘good provi
dence’. The compound Avalokitesvara may mean either ‘the lord of what
we see’, i.e., of the present world, or ‘the lord of the view’ or ‘the lord whom
we see’, the ‘lord revealed’. He is placed along with a certain number of
companions—the sons of Buddha—Avalokita Samantabhadra (the wholly
auspicious), Manjugho§a (lovely voice=Manju$ri), K?itigarbha (earth womb)
and Vajrin (thunderball holder = Vajrapani). He is a Buddhist Siva in visible
form, while Amitabha is the Siva Brahman. He is an ascetic, a magician and
also a saviour; from his fingers flow rivers which cool the hells and feed the
pretas (ghosts), terrifying all the demons. He is the refuge, Buddha-Dharma
and Sangha all in one (ERE. II, pp. 256-261 n).
with Amitabha’s paradise. His figure assumes distinct and

conspicuous proportions in India at a fairly early date, and there
is no ground for tracing his origin or association with Central
Asia. Later works, no doubt, describe him as the spiritual son or
reflex of Amitabha. This aspect recalls the Iranian idea of the
Fravashi defined as ‘a spriritual being conceived as a part of a
man’s personality but existing before he is born and independent
of him belonging to divine beings.171 The relationship between a
Dhyani Buddha and his Bodhisattva is compared with the Zoro-
astrian doctrine of the Fravashi.
Traces of Central Asian origin or affiliation of other Bodhisat-
tvas is also noticed by scholars. Thus,Sylvain Levi suggests172 the
Tokharian origin of Bodhisattva ManjuSri. According to Eliot,173
his worship at Wu-tai-shan in Shan-Si is ancient andlater Indian
tradition connects him with China, while local traditions also
associate him with Nepal, Tibet andKhotan and he is sometimes
represented as the first teacher of civilization or religion. His
Central Asian origin could be probable but not very certain. E.
Huber174 was the first to observe that the canon of one of the
Buddhist schools of the Little Vehicle or Hinayana contained
traditions foreign to India, as for example, the legend about
Khotan, and he wondered, therefore, whether this canon had
not been considerably augmented and modified in Tukharistan
itself. According to Louis de La Vallee Poussin,175 it is now
certain that Serindia—from the Pamir Mountains to the Great
Wall and later China itself collaborated in the development of
Buddhism. The story of Manjusri is interesting. According to the
Chinese pilgrim (I-tsing) ,178 he dwells in China. He is represented
in the miniatures of the Nepalese manuscripts as god worshipped
in China and seems to have come from there to Nepal. Like the
majority of Buddhist gods he is represented under various aspects
171. Eliot : Op. cit, III, p. 221.
172. Journal Asiatique (JA) 1912-1-p. 622; also Levi \L e Nepal pp.
330 ff.
173. Op. cit, III, p. 221.
174. Etudes de litterature Bouddhique, VIII. (Bulletin Ecole France de
l’extreme orient, VI (1906) 385, quoted in ERE. VIII, p. 4069.)
175. ERE. VIII, p. 4069.
176. A Record o f the Buddhist Religion, trans. J. Takakusu, Oxford,
1896, p. 169.
—aBodhisattva in the Mahayana or Great Vehicle and the prin

cipal ‘exponent’ of the Supreme Being in the ‘Tantric Vehicle’.
The cult of Manju$ri originating in Iftdia took a peculiar deve
lopment in China. As a Tantric God, half a dozen Tantras (Kan-
jur) bear his name; among them is the List of the true Names
of Manjusri-Jnana-sattva. Magical rituals are devoted to Maft-
ju$ri.177
The Bodhisattva Ksitigarbha also receives similar treatment.
He appears to be known but not very prominently in India in
the fourth century A.D., but by the seventh century, if not ear
lier, his cult was flourishing in China and subsequently he became
a popular deity second only to Kuan-yin in the Far East. This
popularity, according to Eliot, was connected with his gradual
transformation into a god of the dead. But it is uncertain if he
was prominent first in Central Asia or in China. He was known
as Ti-tsang in Chinese and as Jizo in Japanese. The devotion of
the Chinese to their dead is suggestive of his great position among
them. As a guide to the next world he has a parallel in a similar
position with the Zoroastrian angel Srosh. Ksitigarbha is always
clearly distinguished by the shaven head of the monk and the
barred or mottled mantle, the mendicant’s garment. He has gene
rally a flask in his left hand and elsewhere holds a more familiar
emblem of the flaming jewel.178 Ksitigarbha was accepted by the
Manichaeans as one of the ‘Envoys of Light,179 thereby suggest
ing impact or action of Buddhism on Manichaenism. Ksiti-
garbha’s paintings are interesting and important in respect of
their iconography and artistic value. He stands as a possible rival
to Avalokitesvara in popularity among the Bodhisattvas of
the Buddhist pantheon of the Far East. He is one of
the Eight ( Great Bodhisattvas. Through innumerable incarna
tions he has been working for the salvation of living beings,
and is especially honoured as the breaker of the power of
hell. With his pilgrim’s staff he strikes upon the doors of
hell and opens them, and with the lustrous pearl that he carries
he illuminates its darkness.180 The Central Asian paintings from
Tun-huang exhibit several aspects of his character. An analysis
177. ERE, Vol. VIII, p. 406.
178. Stein : Serindia, Vol. II, p. 364, n 16.
179. Eliot : Op. cit, p. 221, JA. 1911, II, p. 549.
180. Stein : Serindia, Op. cit, p. 965.
of the character of Central Asian paintings with references to

thematic and stylistic aspects and the role of the Bodhisattvas
and their mandalas is reserved for consideration in another con
text in the chapter on Central Asian Art.
Central Asia was the earliest and, on the whole, the principal
source of Chinese Buddhism. Reference has already been made
to the Buddhist savants and monks who went eastwards from
Central Asia to preach and translate the sacred scriptures into
the language of the country. This aroused in the intellectual class
a curiosity to proceed to the land of Buddha’s origin—India—
in search of truth and also to collect sacred texts which could be
carried home for translation, thereby creating an atmosphere
of Buddhism and Buddhist learning in China. Tibet and
Mongolia, however, form independent areas of study of Buddhism,
for synthesising local esoteric elements in Buddhism providing
new colour as also in its expansion in the new form in areas
coming under political influence of the Tibetans.
Tibetan Buddhism
The introduction of Buddhism181 into Tibet is generally asso
ciated with the famous ruler of this land of snow-king Srong-
tsan-gam-po, who ascended the throne in A.D. 629, a contem
porary of king Harsa and the Tang Emperor Tai-Tsung. The
two queens of the Tibetan ruler—a Chinese princess and a
Nepalese one were instrumental in persuading their husband to
introduce Buddhism in Tibet as a necessary part of civilization.
181. The early history of Buddhism in Tibet is shrouded in nebulous
legends. The missionaries sent by king Asoka might have touched the Tibetan
borders, but it is more than a hundred years later that there is legending
account of the establishment of a Buddhist temple on the Tibetan side of the
mountain range. This legend and the story of the miraculous descent of four
caskets containing Buddhist treasures, in the fourth century A.D. might be
suggestive of Tibet’s contact with Buddhist mission. Definitive evidence for
the introduction of Buddhism in Tibet is the 7th century in the time of Srong-
btsan-Gam-po, who was married to a Chinese princess Wen-Ch’eng in A.D.
641. She brought with her Buddhist statues and books and probably some
priests and, thus, established a firm footing for Buddhism in Tibet. The
Nepalese princess, the ruler’s second wife, was instrumental in introducing
the occult worship of the Buddhico-Hinau goddess Tara. This event determined
to a great extent the nature of Buddhism destined to prevail in Tibet. {En
cyclopedia o f Religion and Ethics, (ERE), VIII, p. 702 a.)
The borrowing of the Tibetan alphabet from Indian Brahml,

the fruition of the Sambhota mission, are well-known. According
to Eliot,182 recent investigations, however, have advanced the
theory that the Tibetan letters are derived from the alphabet of
Indian origin used in Khotan and that Sambhota made its acqu
aintance in Kashmir. The reign of Srong-tsan-gam-po and his
two queens, regarded as the first patrons of Lamaism and wor
shipped as incarnations of Avalokita and Tara, no doubt laid
the foundation of Buddhism and civilization associated with it
in Tibet. After his death in 650 there is little information about
Buddhism here for some decades.
About 705 king Khri-gtsug-Ide-btsan took some interest in
Buddhism, building a few monasteries, causing translations to
be made and also inviting monks from Khotan, as recorded in
the edict of 783 preserved in Lhasa. His zeal thus, paved the
ground for further progress in Buddhism with the invitation
extended to Padmasambhava by Khri-Sron-Ide-btsan and the
arrival of this savant in Tibet.183 This invitation seemed to have
been extended at the instance of Sangharaksita said to be the
teacher of the ruler, though there was opposition by. Chinese
bonzes, while politically Tibetan influence had penetrated much
beyond its natural frontiers with its predominance in the Tarim
basin and rule over parts of Ssu-chvan and Yunnan; on the
religious plane the Chinese influence consequent to the ruler’s
relation with that country was prominent. At this time Amogha,184
182. See Noernle MS Remains found in E. Turkestan, 1916, pp. xvii ff
and Francke : E l.X I. pp. 266 ff; Contra see Laufer : JAOS. 1918, pp. 34 ff.
According to Eliot, there is considerable difference between the printed and
cursive forms of the Tibetan alphabet. He poses the question : Is it possible
that they have different origins and that the former came from Bengal, the
latter from Khotan ? According to F.W. Thomas, the mission of Sambhota
is made to precede the King’s code of laws and his Nepal and Chinese
marriages. The incentive to the acquisition of a script is said to follow the
matrimonial alliance with Nepal which cannot be dated long after A.D. 634,
while the king longed for a Chinese princess later on which materialised in
640. It was from Nepal that the Tibetan script was borrowed, since after the
alliance with Nepal there was no need to go further in quest of a script.
(Festscriff Zur Feier des zoo jahriegen Bestchens der Akademie der Wissen
Schaffen in Gottingen, 1951, pp. 146 ff.)
183. See Waddel. JRAS, 1909, p. 931.
184. Amogha was a native of Ceylon (or, according to others, of Northern
a Ceylonese monk, was flourishing in China and he bad intro

duced the Mahayana system or Chen-Yen in that country. This
form of corrupt Buddhism was brought to Tibet where with its
pliant nature amalgamated easily with local observance and
demonolatry.
Padmakara or Padmasambhava, the Indian savant, belonged
to then popular ritualistic and mystic yoga school at Nalanda,
and he was skilled in Buddhist spells (dharani). Originally he
was a native of Udyana (later known as Swat and Kafiristan)and
he arrived in Tibet in 747 with several other Indian monks who
were induced to settle in the country. Padmakara established the
first monastery at Sam-yas in A.D. 749 on the left bank of the
Brahmaputra river, about 30 miles to the south-east of Lhasa,
and installed Santaraksita185 as its abbot, with seven Tibetan
novices as the nucleus of the order. Padmasambhava ‘the lotus-
born’ or ‘the teacher treasure’, the founder of the order of Bud
dhist monks in Tibet, was a believer in Tantric mysticism with
its prayers to various Buddhist gods and goddesses. While no
India) who arrived in China with his teacher Vajrabodhi. After the latter’s
death he revisited India and Ceylon in search of books and came back in 746.
He wished to return to his own country, but was refused permission and so
stayed in China until his death in 774. He received the title of K'tsan literally
meaning ‘wisdom-repository’ which name is translated into Prajnakosa.
According to Nanjio, he was allowed to go back to his own country in A.D.
749, but when he arrived at the South-sea district, he was ordered to stay in
China by the Imperial command. In A .D. 756, he was called back to the capital.
According to his version as recorded in his memorandum presented to the
Emperor he translated 77 works. He died in 774 in his seventieth year, and
received posthumous title of ‘great-eloquence’ correctwide-wisdom. He was
held in high veneration at the court of successive sovereigns of the Than
dynasty. (Nanjio : Op. cit, II, 155, pp. 444-447; Eitel : Op. cit, p. 8a; Eliot:
Op. cit, pp. 39, 264, 327).
185. Santarak$ita, a native of Gaur, who was the High Priest of the
monastery of Nalanda, was invited by king Thi-Srong-deu-tsang. He was
received by the Tibetans with all the honour due to his position as the spiritual
teacher of the king of Magadha and he was named Acarya Bodhisattva. He
was appointed as the High Priest of Tibet and under his direction, Buddhist
monachism was introduced in Tibet. This came to be known as Lamaism.
While Santaraksita attended to the moral and disciplinary part of the church,
his eminent co-adjutor Padmasambhava, took charge of the Tantric part of
the Buddhist liturgy. (S.C. Das : Indian Pandits in the Land o f Snow, Calcutta
(Reprint 1965), p. 51).
canonical translations are ascribed to him, he is the reputed

author of several manuals of worship (sadhanas) and propitia
tion of deities by means of the repetition of spells (dharani) like
the Brahmanical mantras. Padmakara or Padmasambhava186
also initiated an era of great literary activity and scholarship with
the translation of the Buddhist canon from Sanskrit. Several
Tibetan scholars were sent to India to learn Sanskrit and Bud
dhist philosophy. These Tibetan translations of the Buddhist
canonical texts preserve with remarkable accuracy the Indian
texts, of which most of the originals have been lost in India.
The institution of the local order on these Indian lives was
opposed by Chinese Buddhists in Tibet under a Mahayana monk
named Hwa-shang (corresponding to Sanskrit upadhyaya or
master). These Chinese, who were itinerant priests, moving
from one place to another, were defeated in argument by the
Indian Kamalasila and expelled from the country, leaving the
Indian system to be developed undisturbed. Many monasteries
and Buddhist temples were established all over the country and
Buddhism became the state religion of the land. During the
reign of Khri-Sron-Ide-btsan and the visit of Padmasambhava
(which began in A.D. 747 according to the traditional chrono
logy) the number of translations began to increase. Two works
ascribed to the king and one to the Saint are included in the
canon, but the most prolific writer and translator of the period
was Kamalasila,187 the hero of the intellectual bout between the
186. ERE. VIII, p. 745 a; Eliot : Op. cit, pp. 249 ff. Padmasambhava
was one of the most celebrated exponents of Tantric Buddhism and in Tibet
is often called simply the teacher (Guru or Mahac&rya). His portraits represent
him as a man of strongly marked and rather angry features, totally unlike a
conventional monk. The account of his life, as read in Tibet, appears to be
rather fantastic. Padmasambhava is not celibate but is accompanied by female
companions. In Tibet where the older religion consisted of defensive warfare
against the attacks of evil spirits, he assumes the character of a victorious
exorcist, subduing demons. He preached a non-celibate and magical form
of Buddhism, ready to amalgamate with local superstitions and needing new
revelations for its justification. (Eliot : Op. cit, pp. 349-50)
187. Kamalasila, the great Buddhist philosopher of Magadha was invited
to Tibet by king Khril-Sron-Ide-tsan at Santaraksita’s initiative to participate
in the debate between the Indian Pandits and the Chinese ho-shang. After
Kamalasila came to Tibet, a grand philosophical debate was organised
between the Chinese priests and the followers of £antarak$ita with the ruler
Indian Pandits and the Chinese bonzes. Seventeen of the origi

nal works of this Buddhist savant from India are preserved in
the Tanjur and he translated part of the Ratnakuta. This great
period of translation commencing from the arrival of Padmakara,
Padmasambhava with the active participation of Kamalasila and
others is rightly called the Augustan age of Tibet. A solid foun
dation was laid by composing two dictionaries containing a
collection of Sanskrit terms.
Padmasambhava was one of the greatest exponents of Tantric
Buddhism and was known as the Guru or Mahacarya in Tibet.
He is represented as a man of strongly marked and rather angry
look, totally unlike a conventional monk. A popular account of
his life is no doubt available and might contain some grains of
truth, but it is difficult to account for the later events in his life.
It is said that when he had finished his work in Tibet he vanished
from there rather miraculously. His Interest in Tantric ideas was
probably the result of his association with Bengal which was no
doubt a centre of Tantrism. This took a more active and virulent
form in Tibet where he is not only associated with female com
panions, but also assumes the character of a victorious exorcist
against the attacks of evil spirits. Padmasambhava, there
fore, preached a non-celibate and magical form of Buddhism*188
amalgamating local superstitions and needing new revelations
as the judge. Bu-Stan provides details of this debate. In the presence of the
assembled court, Kamalasila came out victorious over the Chinese sage and
was placed at the head of the metaphysical branch of the Buddhist church.
Ho-shang, the Chinese contender in the debate was ordered to leave the
country. (See, Das : Op. cit, pp. 51-52; Alaka Chattopadhyaya : Atifa and
Tibet. Calcutta. 1967. pp. 228 ff.)
188. According to Waddel, there is no certain evidence regarding the
character of his teaching. There is no certain evidence that it was of the
flagrantly magical and necromantic type ascribed to him in the indigenous
works on the subject, which are mostly late compositions of the 14th century
onwards—when works of a similar nature were being issued by the Mahayana
Buddhists in India ascribing precisely similar ritualistic spells to Buddha
himself. From the high literary attainments of his contemporary pupils it
seems probable that his teaching was more or less orthodox Indian Buddhism
of the Mahayana type, and of the Middle Path School (Madhyamika) to
which he reputedly belonged, and that it afterwards became degraded in the
hands of the converts from the indigenous Shamanistic Bar religion. (En
cyclopedia o f Religion and Ethics (ERE), Vol. IX, p. 591 a.I
for its justification. Reference has already been made to the

monastery built by him at Samye and its abbot Santaraksita,
who is said to have laid the foundation of the order of Lamas.
Samye became a great literary centre where many translations
were made. A monk from Kasmir named Vairocana was one of
the best translators.
The opposition to the new Buddhist order in Tibet both from
the priests of the old native religion and also from Chinese
Buddhists189 might as well have accounted for the absence of
Buddhist personal names and direct reference to Buddhism in
Tibet in numerous Tibetan documents discovered in the Tarim
basin. Buddhist priests (bon-da) are occasionally mentioned,
but the title ‘Lama’ is not found. The writers of these docu
ments no doubt seem to be familiar with the usages of the
Bon-po religion and there are notices of religious struggles
as well. This picture of Buddhism in Tibet from the time of
Padmasambhava onwards could no doubt be visualised from
the indigenous sources alone. When Padmasambhava vanish
ed from Tibet, he left behind twenty-five disciples to pro
pagate his teachings. They were no doubt sorcerers and the
new religion flourished till the time of Ralpachan, the grandson
of Khri-Sron-Ide-btsan. Monasteries multiplied with state en
dowments attached to them. They had the right to collect tithes
and each monk was assigned a small revenue derived from five
tenants. There was reorganization of the hierarchy.190 Many
translators were engaged in this period and a considerable part
of the Buddhist canon was then rendered into Tibetan. The
ruler’s patronage of Buddhism, however, alienated disgruntled
elements resulting in his murder191at the instigation ofhis brother
189. Numerous Tibetan documents discovered in the Tarim basin date
from this period. The absence of Buddhist personal names in them as also the
rarity of direct reference to Buddhism indicate that though known in Tibet,
it was not yet prominent. Buddhist priests (ban-de) are occasionally mentioned
but the title Lama has not been found. According to Eliot, the usages of the
Bon-po religion seem familiar to the writers and there are allusions to religious
struggles. (Eliot : Op. cit, p. 351. See also JRAS, 1914, pp. 37-59).
190. Rockhill : Life o f the Buddha, p. 225.
191. Various dates are given for his death, ranging from 833 to 902
(Rockhill : Op. cit, p. 225; Bushell : JRAS, 1880, pp. 440). According to
Eliot, the treaty of 82 was made in his reign (Op. cit, p. 351 n).
and successor Lang-dar-ma. He attempted to extirpate Lamaism

by destroying monasteries, burning books and driving
Indian monks out of Tibet. This process of persecution lasted
for three years, finally leading to the murder of the anti-Buddhist
king by a Lama. Political confusion, however, soon followed
with the disintegration of the united kingdom, divided among
clans and chieftains, and the collapse of Tibetan power in the
Tarim basin. It is suggested that during this period of political
turmoil and instability Buddhism had almost disappeared in
Tibet, and the silence of ecclesiastical history during the tenth
century confirms the gravity of the catastrophe. On the other
hand, the preservation of translations made in the ninth century
is suggestive of the existence and functioning of monasteries, for
instance at Samye.
According to the chronicles,192 foreign monks arrived in
Tibet at the beginning of the eleventh century. This period of
revival of Buddhist learning in Tibet is considered a separate or
later one, phyi-dar in contrast to the earlier diffusion called sna-dar.
The chief savants in the new diffusion were La-chen-Lo-chen, the
royal Lama Yes’es and Atisa or Dipankara Srijnana. Lo-chen
was from Kashmir which provided many other Lamas who were
engaged in Tibet. The debased Tantrism passing onas Buddhism
was instrumental in the despatch of young Lamas to India for
instructions as also invitation to learned monks at centres of
Buddhist learning. The initiative in this direction was taken by
Yesesttod,193 a king or chieftain of mNa-ris in Western Tibet.
Atisa,194 who came to Tibet under this programme was from
Bengal, ordained at Odantapuri and studied in Burma or
Suvarnadvipa. Subsequently he was appointed head of the
monastery of Vikramasila. He stayed in Tibet for fifteen years.
He reformed Tibetan Buddhism. With a view to introducing
pure Buddhist monachism in Tibet, he selected intelligent lads,
192. Eliot : Op. cit, p. 352.
193. Lha Lama Yes’ehod, king of Tibet, was a devout Buddhist. He
ruled peacefully for many years. About the year 1025 he founded the monastery
of Thoding in Purang. (S.C. Das : Indian Pandits in the Land o f Snow, Rep.
1965, p. 54)
194. For the Life of Dipankara Srijfiana Atisa, See Das : Op. cit, pp. 53ff;
also Alaka Chattopadhyaya : Atisa and Tibet—Op. cit also Encyclopedia o f
Religion and Ethics, Vol. II, p. 1976.
each ten years old, and carefully trained them up in Tibetan

and finally admitted them into the sacerdotal order with the
consent of their parents. Not satisfied with the Buddhist teachers
of Tibet, whose cults had become greatly debased by the admix
ture of Tantric and Bon mysticism, he sent these young monks
to Kashmir and Magadha and other places of India. There pure
Buddhism still prevailed, and they could study the philosophy
of Ananda Garbha of Kashmir and Vinaya code of monastic
discipline. These young monks were also instructed to invite to
Tibet, if possible, the renowned Kashmirian Pandit Ratnavajra
and the Buddhist hierarch of Magadha and other holy men and
savants. Out of the 21 Tibetan monks sent to India, Rinchhen
Zan-po, the great Lochava and Leqspahi-Serab, were the only
survivors who returned to Tibet after. successfully completing
their stay and study in India. The Lochavas had heard of
Dipankara 3rijnana who then occupied the highest position
among the Buddhist scholars of Magadha. He was the second
Sarvajna of the school of the 500 Arhats which was com
monly called the Mahasanghikas. On their report the king
of Tibet ordered Rgya-tsar-gru-Senge to proceed to Vikrama-
Sila, with an invitation from him along with a large piece of bar
gold as a present from the sovereign. The Indian Buddhist
savant, however, refused to proceed to Tibet for quite some
time. Lochava was sent again and he was finally successful in
persuading the Vikramasila Abbot to visit Tibet in 1038. He
remained there until his death fifteen years later. He introduced a
new calendar195 and inaugurated the second period of Tibetan
Buddhism which is marked by the rise of successive sects
in the form of reforms.
While stressing the importance of philosophical wisdom,
Ati$a Dipankara felt that the most urgent task at that time in
Tibet was that of moral reform—a sine-qua-non for any religion.
He, therefore, laid great emphasis on moral reform in the Bodhi-
patha-pradipa. Prajha or the highest wisdom, according to him,
is that knowledge which is the realisation of the intrinsic nature
195. The Tibetan system of computing time is based on cycles of sixty
years beginning in 1027, and not in 1026. In many dates there is an error of a
year. Pelliot : JA, 1913.1. p. 633; Laufer : Toung Pao, 1913, p. 569; Eliot :
Op. cit, 353 n.
of the void {Svabhava-sunyata) of everything. He, however,

insisted on establishing it on the secure basis of right conduct
or upaya. Wisdom (prajha) and right conduct (upaya) are
indispensable to each other,' and none can be ignored. He
stressed on the supreme importance of moral conduct for a true
Buddhist. He did not denounce Tantricism outright, but prescri
bed as preconditions of the Tantric initiations a correct insight
into the real teachings of the Tantras, an extreme purity of
moral conduct and correct guidance from an ideal teacher. He,
therefore, concentrated on the reform of Tantric practices in
Tibet. He retained the form of the Tantra but infused it with
the content of classical Buddhism. His own knowledge of the
Tantras was too imposing to be questioned even by the
staunchest supporter of Tantrism. He was no doubt eager to
preach in Tibet the Mahayana ideal of universal emancipation
and the Madhyamika philosophy called Sunya-vada.
AtiSa Dipankara introduced a new calendar and inaugurated
the second period of Buddhism, with reforms introduced in the
religion of the Tathagata based on purity of conduct and moral
disciplines, without wiping out completely the local influence
based on superstitions and tantric rites. He professed the
Kalacakra ‘the wheel of death’. This symbolised the extreme
Tantric phase introducing around the 10th century A.D., a
rampant demonolatry with exacting priestly rites into a religion
which in its origin was largely a protest against worship and
ritual of every kind.196 According to Eliot,197 the Kalacakra
tantra was introduced in 965 A.D. from Sambhala, a mysterious
country in Central Asia. This system is said to be Vaisnavite
rather than Saivite. It especially patronizes the cult of the
mystic Buddhas such as Kalacakra and Heruka, all of whom
appear to be regarded as forms of Adi-Buddha or the primor
dial Buddha essence. Waddel however, presumes198 that
196. Encyclopedia o f Religion and Ethics (ERE), Vol. IV, p. 5926.
197. Op. cit, p. 386. In Tibetan the Kalacakra is known as Dus-Kyi-
hkhor-lo and in Mongol Tsagun-Kurdun. It is also equated with Vajrayana.
According to Eliot, the system has some connection with the Turkish cycle
of twelve animals used for expressing dates. (See Laufer : Toung Pao, 1907,
p. 402).
198. ERE. IV. p. 5926.
the majority of these demons were monstrous ‘king devils’

of the most hideous Saivite type with their equally repulsive
spouses. The chief were Vajra-bhairava, Samvara, Hayagriva
and Guhyakata. Their function was to be tutelaries (Yidam) to
guard their human votary against the attacks of minor demons,
while they acted as beneficiaries of their votaries. Some of them
were also selected as defenders of the faith (dltarmapala)
and also as guardians of particular monasteries and sects. The
Kalacakra is described as the latest and most corrupt form of
Indian Buddhism, but it was doubtless superior in discipline
and coherency to the native superstitions mixed with debased
tantrism which it replaced.
The great fame acquired by Dipankara in Tibet was due to
several external factors, besides his personal scholarship and
erudition. These include the patronage extended to him by the
highly enlightened ruling family of Western Tibet, followed by
his extensive tour of Tibet inspiring the nobles of the country
and fostering a revival of Buddhism, as also his confrontation
with Rin-chen-pzan-po,199 the great Tantric scholar, much too
senior to him. He humbled the pride of his adversary and there
by established his scholastic supremacy. His composition of the
Bodhi-patha-pradipa proved crucial not only for his successful
career in Tibet, but also for the subsequent history of Tibetan
Buddhism. This work, though extremely brief, consisting of
only thirty-six verses, its impact on Tibetan Buddhism was all
the more great. It contains in brief the essence of the whole
sutras and tantras paving the way for the spread of Buddhism.
This was followed by another big work called Bodhi-marga-
pradipa-panjikd—the author’s commentary on his previous one.
The original was intended to be a manifesto of Buddhist reform
in Tibet.
Atisa Dipankara’s stay in Tibet was equally responsible for
the assemblage of monks at a grand convocation and learned
199. Rin-Chen-bzan-po is said to be the first Tibetan scholar of Buddhism
in the real sense. At the age of thirteen he was ordained by the Upadhyaya
Ye-Ses-bzan-po. As a youngman he went to Kashmir and there studied
numerous treatises in the Mantrayana. His great fame as a scholar rested
primarily on his knowledge of the Tantras and none could match him in
scholarship. When Atisa arrived in Tibet, this scholar was in his eighty-fifth
year. (Alaka Chattopadhyaya : Op. cit, pp. 339 ff.) *
translations of Sanskrit works into Tibetan as also for setting

up of substantial structures to house monastic communities. The
translators included Ssan-Kaar-Lochava, Rva-Lochava, Nau-
Lochou, Lodan Serab, etc. The sage Marpa, Mila-Gonpo, and
the famous Pandit Sakya 6ri of Kashmir, besides many other
Indian Pandits were exponents of Buddhism in the following
century. The monasteries became the largest and safest buildings
in the land, possessing the strength of walls and inviolability.
The most important was the Sakya monastery which had abbots
of royal blood but not celibate. This monastery received royal
benefactions. In the reign of Tagpa-de, the ninth in descent from
Tse-de, an image of Maitreya Buddha costing 12000 Bot-shad or
a million and half of rupees was set up.200 He also got an image
of ManjuSri made with seven bre—a measure equal to a tenth
fraction of the English bushel—made. His son Asode, a greater
Buddhist than himself, annually sent offerings and presents to
the Vajrasanaat Bodh-Gaya (Dorje-dan) which was continued
even after his death. His grandson Ananmal prepared a complete
set of the Kangyur, written on golden tablets. His son in turn
put the golden dome over the great temple of Buddha at Lhasa
and set up the image of the God at an immense cost. The latter’s
grandson was initiated into Buddhism by the Sakyapa Lamas
and subsequently became king. This dynasty of ecclesiastical
statesmen practically ruled Tibet at a critical period marked by
the conquests of Chingiz Khan and the rise of the Mongol
Empire.
Buddhism in Mongolia and Tibet
There is no evidence that Chingiz favoured Buddhism in a
special way. He believed in one king and one God, and thought
of religions not as incompatible systems but as different methods
of worship with their prayers in different languages. While
there is no proof of the early Mongols invading or conquering
Central Tibet, it is known that Kublai subdued the eastern
provinces and through the Lamaist hierarchy established special
relations between his dynasty and Tibet. This relationship began
in the time of his predecessor with the summoning of the head
Lama of the Sakya monastery—commonly called Sakya Pandit
200. S.C. Das : JRASB. 1881. p. 238.
or Sa-Skya-pan-cen to the Mongol Court in 1246-8; and he

cured the Emperor of an illness. The favourable impression
created by this great Lama—a man of great learning and in
fluence—no doubt facilitated the achievements of his nephew
and successor known as Bashpa or Paqspa. The Mongol Court
already favourably impressed by Tibetan Lamas, became in
terested in Buddhism. The Emperor had a feeling that the
intellectual calibre of the Mongols and the Tibetan was identical
and it was politically easy as also expedient to conciliate the
uncanny spiritual potentates ruling in a land which was no doubt
difficult to conquer. He, therefore, summoned the abbot of
Sakya to China in 1261 and was initiated by him into the my
steries of Lamaism. The role of Paqspa in this context is re
corded by the Mongol historian Sanang Setsen,201 while there
are certain traditional accounts as well which suggest a greater
degree of Saivism incorporated in the religion first taught to the
Mongols.
It is said that before Paqspa’s birth the god Gape£a had an
eye on the land of Tibet and he helped Paqspa at the Chinese
court with Mahakala helping him with mystery which he im
ported to Kublai. It is called Hevajravasita—a magical formula
which compels the obedience of spirits or natural forces. It is
associated with Hevajra—the same as Heruka—conceived as
manifestations of Buddhas and Bodhisattvas, made for a special
purpose and closely corresponding to the manifestations of §iva.
Paqspa received the title o f ‘Kuo-Shib’ o r‘instructor of the nation’
and was made the head of all Buddhists, Lamaists and others.
He was recognised as the head of the Church in Tibet, and a
tributary ruler with a lay council to assist him in the government
run through three heads of provinces.
Paqspa was also charged by the emperor to provide the
Mongols with an alphabet202 as well as a religion* He used a square
201. See Schmidt : Mongol History o f Senong Setsen—quoted by Eliot :

Op. cit, p. 355. Paqspa or Bashpa—a title equivalent to Arya in Sanskrit,
was the nephew and successor of Sakya-Pandit (or Sa-Skya-pan-cen). He
made great achievements in Mongolia.
202. The»Mongolian alphabet was received from the Uigurs, a Turki
people, in the 13th century. The latter took it from the Sogdians, an Iranian
people. The Sogdian alphabet was an adaptation of a Semetic alphabet
Buddhism and Buddhist Savants of Central Asia 159
form of the Tibetan letters written not in horizontal but in

vertical lines. This experiment of introducing Tibetan script fell
into disuse and was replaced later on after Paqspa’s death by an
enlarged and modified form of the Uighur alphabet. This was
already employed by 6akya-Pandita for writing Mongol. Its
definitive form for that purpose was elaborated by the Lama
Chos-Kyi-hod-zer in the time of Kublai’s successor.
This alphabet, supposed to be of Aramaic origin had already
been utilized by Buddhists for writing religious works. Its appli
cation to Mongol is supposed to be merely an extension of its
general currency in Asia. Paqspa also supervised the preparation
of a new edition of the Tripitaka in Chinese, not in Mongol;
and its learned editors were scholars well-up in Sanskrit, Chinese,
Tibetan and Uighur. This comprehensive work also records re
ferences to Chinese texts in Tibetan, wherever available. The
Tibetan influence also extended in the field of fine arts and
appointed a Lama called Aniko, skilled in both sculpture and
painting as the head of the bureau of fine arts.*203 The direct
influence of Lamaism in Mongolia through parts of northern
China, however, seemed superficial and temporary. It appears
that the first growth of Mongolian Buddhism as part of a politi
cal system died down with the end of the Yuan dynasty.204 The
Strong Lamaist influence was partly, if not wholly, responsible
for this political debacle. The subservience to the clergy and
extravagant expenditure on religious buildings and ceremonies
depleted the royal chest. The role of the later Lamas after the
departure of Paqspa, holding a high position at the Peking Court
and exercising considerable influence might as well have added to
the political malaise. Chos-kyi-hod-zer and Gyun-ston-rdo-rje-
(Aramaic). The Mongolian alphabet was written vertically downward.
(.Encyclopedia Britannica, Chicago, 1972, Vol. XV, p. 734.)
203. Eliot : Op. cit, p. 356.

204. The fall of the Mongol empire in China in 1368 led to the weakening
of the Tibetan Buddhist faith throughout Mongolia. The old popular religion
of Shamanism gained ground again and Taoist beliefs spread to Mongolia.
The reformed version of Tibetan Buddhism, the Geng-pa was introduced
by Neyica toyin (1557-1653)—the first Buddhist missionary to eastern Mon
gols, and the first Jaya Pandit (1599-1662)—the two most prominent re
presentatives of the new west Mongolian Buddhism .{Encyclopedia Britannica'.
Op. cit, XV, p.730 a.)
dpal were two such figures. The latter was a great exponent of
the Kalacakra system and also the teacher of the great historian
Bu-ston who is said to have arranged the Tibetan canon.
Despite royal favours heaped upon priests and monasteries,
Buddhism does not appear to have flourished in Tibet during the
fourteenth century. From 1270 to 1340 the abbots of Sakya were
rulers of both church and state. All the abbots were appointed
or invested by the Emperor and their power declined with the
Yuan dynasty. Mutual conflicts were not unknown. In 1320 the
Sakya abbots burnt the rival monastery of Dikung; but in due
course other monasteries increased in importance and a chief
known as Phagmodu succeeded after many years of fighting in
founding a lay dynasty which ruled over parts of Tibet until the
17th century.
The Ming dynasty supplanting the Yuan in 1368 was not
pro-Buddhist. Its rulers had no preference for Lamaism but were
equally anxious to maintain good relations with Tibet treating
it a friendly but vassal state. They recognised the dynasty of
Phagmodu and also the abbots of eight monasteries, of course
with and implication of suzerainty. The primacy of the &akya
monastery, atone time a reality, was reduced to only one among
several great monasteries. The advent of the Ming dynasty also
coincided with the birth of the reform movement leader Tsong-
Kha-pa in the district of Amdo on the western frontiers of the
Chinese province of Kansu. He absorbed instructions from many
teachers and as a youth went to Tibet where he studied at Sakya,
Dikung and finally at Lhasa. Noticing the discrepancy between
Lamaism in theory and in practice, he decided to undertake the
work of reform, which became visible in the Geluga—the sect
presided over by the Grand Lama. It acquired paramount im
portance in both ecclesiastical and secular matter and came to be
known as the ‘Established Church of Tibet,’ also conveniently
called the Yellow Church.
Tsong-Kha-pa’s reforms were on two lines- Firstly, he stres
sed on strict monastic discipline, insisting on celibacy and
frequent services of prayer; secondly he greatly reduced the
Tantric and magical elements in Lamaism. An effective organi
zation was set up to perpetuate these principles. The great
monastery of Gandan near Lhasa was set up by him and he
became its first abbot. Three others at Sera and Depung near
Lhasa and Tashilhunpo were also set up in his life time. Tsong-
kha-pa205 seems to have ruled by virtue of his personal authority,
and after his death in 1417 his nephew and successor Geden-
dub claimed the said right. The Lamas had gained considerable
prestige and the Ming Emperors utilized their position for gaining
political influence in Tibet. The Kanjur was printed in China in
1410.
The ecclesiastical and political hierarchy was vested in
the Grand Lajna, the abbot of the Tashil-hun-po monastery,
residing at Lhasa. The theory of successive incarnations which
is the characteristic of Lamaism was developed and defined.
Two ideas were combined in it—the first being the appearance
of divine persons in human form, and secondly the real conti
nuity of life in a school, sect or church. Accordingly, a great
teacher is reborn in the successive occupants of his chair. The
hereditary soul is identified with a Buddha or Bodhisattva, as in
the great incarnations of Lhasa and Tashilhunpo. These
incarnations are not confined to the great Lamas of Tibet. The
heads of most large monasteries in Mongolia also claim to be
living Buddha. A long record of the Lamas of Tibet need not
be mentioned here, but reference might be made to their associ
ation with the Mongols in ecclesiastical and political matters.
The Mongols were no doubt converted to Buddhism when their
capital was at Peking and mainly affected these Mongols residing
in China. When the Yuan -dynasty was dethroned with the Mon
gols driven back to wild regions, they relapsed into their original
superstitions and beliefs. About 1570 AltanKhagan, the powerful
chief of the Turned became acquainted with Tibet through some
Lamas captured in a border fray and taken to his court. He
thought it politically expedient to invite the Grand Lama to his
court.206 The Lama set out on his travels with great pomp and
205. Tsau-Kha-pa, the founder of the Dge-lugs-pa sect, was in fact the
same (i.e. the incarnation of) Jo-bo-rje (AtiSa). In the eyes of the common
people he appeared to receive the upade&a or margakrana of the bka-gdams
from mahaupadhyaya Nam-kha-rgyal-mutshan and Chos-skyabs-bzan-po.
He removed the dirt of doubt and distortions and made changes in Jo-bo’s
upadesa in course of time. (Alaka Chattopadhyaya : Op. cit, p. 12).
206. The visits of the third Dalai Lama in 1557 to Turned and Ordos and
in 1587 to the Kharatsin tribe led to the mass conversion of these Mongol
appeared to the bewildered Mongols in the guise of Avalokita

with four arms and the imprint of his horse’s hoofs showed the
six mystic syllable om mani padme hum. A great congregation was
held near Lake Kokonor. The Lama bestowed on the Khagan
high sounding titles and himself received the epithet ‘Dalai’ or
‘Talai’, the Mongol word for ‘sea’, signifying vast extent and pro
fundity. This is the origin of the name Dalai Lama. The hierarchy
was divided into four classes corresponding to four ranks of
Mongol nobles. The use of meat was restricted and the custom
of killing men and horses at funerals forbidden. The observance
of Buddhist festivals was madecompulsory, and native gods were
more or less assimilated in the new pantheon. The Grand Lama
specially recommended to the Mongols the worship of the Blue
Mahakala, a six-armed representation of Siva standing on a figure
of Ganesa. He left with the Mongols a priest who was considered
an incarnation of Manjusri, and a temple and monastery were
built for him in Kuku-Khota. The Grand Lama returned to Tibet
but made a second trip to Mongolia after the death of Altan
Khagan in 1583, to ensure allegiance of the new chiefs. He also
received an embassy from the Chinese Emperor Wan-li as also
the honorific titles similar to those received earlier by Pagspa
from Kublai. This move was politically motivated to neutralise
the Tibetan-Mongolian alliance.
This Grand Lama died at the age of forty-seven and was born
in the Mongol royal house, as a great-grandson of Altan Khagan.
Till the age of 14 he lived in Mongolia and then moved to Lhasa.
A Lama was appointed as his Vicar andPruniate of all Mongolia
with residence at Kuren or Ugra. About this time the Emperor
of China issued a decree that these hierarchies must be reborn in
Tibet and not reappear in a Mongol family for fear of too close
unionism of religion and patriotism. The church of Lhasa became
tribes. The old popular religion of Shamanism was suppressed, its ideas being
incorporated in the doctrine of popular Buddhism. The foundation of many
Buddhist monasteries; the conversion of many members of the Mongol
nobility to religion and their entry into the monastic life; the formation of
an educated class able to read and write and to translate into Mongol from
Tibetan and Sanskrit; and the translation of religious works into Mongol
and the subsequent creation of a written secular literature in the Mongolian
language are the peculiar features of this period. (Encyclopedia Britannica,
1972, Vol. XV, p. 731 d.)
a powerful religious and political force with the ascending influ

ence of the Grand or Dalai Lama. Lozang, the fifth Grand Lama,
established his temporal and ecclesiastical sovereignty over Tibet.
He built the Potala and his dealings with the Mongols and the
Emperor of China are of considerable importance for political
relations with the neighbouring states. China was equally anxious
to enter into friendly relations with Tibet with a view to contain
Mongol pressure on its borders. Their negotiations finally cul
minated in the visit of Lo-Zang to Peking in 1652-3 and he was
treated as an independent sovereign and received from a long
title contained in the phrase ‘Self-existent Buddha’, ‘Universal
Ruler of the Buddhist faith’. The later history of Tibet—ecclesi
astical and political—is modern and could be studied in the con
temporary context. One need, however, take fuller notice of
canons connected with Lamaism as also the different sects of
Buddhist groups.
Lamaism is defined as a mixture of late Indian Buddhism
which itself is a mixture of Buddhism and Hinduism—with vari
ous Tibeton practices and beliefs. These include, in the main,
demonophobia and the worship of human beings as incarnate
deities. Religion and magic are expected to protect against fierce
and terrible spirits in the atmosphere. Even benevolent deities
assume a terrible form in order to strike fear into evil spirits and
keep them away. The worship of incarnate deities common in
eastern Asia acquires a new dimension and extent in Tibet- An
ancient inscription207 applies to the kings of Tibet the word
hphrul meaning the transformation of a deity in a human form-
This term was applicable to the Grand Lamas and the Yellow
Church officially recognized the Emperor of China as an incar
nation of ManjuSri. India provided terrible deities, like Kali
and her attendant fiends, and also the concept of divine incarna
tion in human form. It is, therefore, proposed that Tibetan Bud
dhism is not so much as amalgam, as a phase of medieval Hindu
religion disproportionately developed in some directions, with
the Lamas acquiring the same status as the Brahmanas serving as
intermediaries between gods and men. In Tibet Buddhism acquir
ed more life and character than it had in Bengal. The native
207. Waddell : JRAS, 1909, p. 941; Eliot : Op. cit, p. 383.
character had something monstrous and fantastic with features

similar to the oldest form of Tibetan religion called Bon or Pan.
Books and images introduced from India did not assume any
national character prior to 747. The arrival of Padmasambhava
in that year marks the first phase of Lamaism. The Nying-ma-pa
or old school claims to represent his teaching. He brought with
him the late formoflndian Buddhism called Mantrayana, closely
allied to the Chen-Yen of China, and transported to Japan under
the name of Shingon and also to Java. Padmasambhava’s teach
ings included toleration and incorporation of local cults, free
use of spells (dharani) and magical figures {manialas) for sub
duing demons and acquiring supernatural powers, the belief in
assuming divine form through such methods. The worship of
Amitabha, among other deities, and a belief in his paradise, presen
tation of offerings and the performance of sacrifices on behalf of
the departed souls and finally the worship of departed and per
haps of living teachers. According to Griinwedel,208 the later
corruptions of Buddhism in northern India, Tibet and Central
Asia are connected with the personage known as the eighty-four
Mahasiddhas or great magicians.
The next phase of Buddhist activity in Tibet with the visit of
AtiSa in the eleventh century A.D. and some others, is associated
with the Kalacakra system also known as the Vajrayana. While
a legend209 credits the Sakyamuni promulgating this system in
Orissa (Dhanya-Kafaka) and that Sucandra, king of Sambhala,
writing the Kalacakratantra in a prophetic spirit, Tibetan autho
rities state its introduction in Nalanda by a Pandit called Tsilu
or Chilo. It was accepted by Narotapa, then the head of the
University and from there it spread to Tibet. Its promulgation
is also ascribed to a personage called SiddhaPito., Since it means
Islam and Mohammad it is perhaps connected, according to
Eliot,210 with anti-Mohammadan movement which looked to
Kalki,m the future incarnation of Visgu as their Messiah. The
mythology of this school is Vi?nuite, not Saivite. It may as well
208. Mythologic des Buddhismus, p. 40; Eliot : Op. cit, p. 385.
209. JASB, 1882, p. 225, quoted by Eliot Op. cit, p. 386.
210. Eliot : Op. cit, p. 387.
211. See KalkipurSpa; Vi§pupurapa IV. XXIV; Bhag. PurSr.a XII.
ii. 18; Eliot : ibid.
be mentioned that the Pancaratra system which had some con

nection with Kashmir, lays stress on the wheel or discus (cakra
or sudarsana) of Visnu, which is said to be the support of the
universe and the manifestation of creative will. The Kalacakra
is mentioned as a special form of this cosmic wheel having six
spokes. The Buddhist Kalacakra stresses on an AdirBuddha or
primordial Buddha God from whom all other Buddhas are
derived. Under the influence of Kalacakra the Lamas did not
worship one Supreme God but they identified with the Adi-
Buddha some particular deity, like Samantabhadra, who usually
ranks as a Bodhisattva. It also adopted ail the extravagances of
the Tantras and provided the principal Buddhas and Bodhi-
sattvas with spouses, including one for the Adi-Buddha him
self.212
In the last phase of reformation Tsong-Kha-pa disciplined
the priesthood and enabled the heads of the Church to rule
Tibet- It no doubt imported new ideas but emphasised the wor
ship of Avalokita as the patron of Tibet and systematized the
existing beliefs about reincarnation. It created a powerful hier
archy and restricted Tantrism without abolishing it. The Tibe
tans also elongated the number of deities from the Nepalese
Buddhist original pantheon. A single deity was represented in
many forms and aspects for purposes of art and popular worship.
The adoration of saints and their images is also more developed
than in Nepal. The later doctrines of Indian Mahayanism are
naturally prominent. The three bodies of the Buddha are well-
known and also the series of fine celestial Buddhas with corres
ponding Bodhisattvas and other manifestations. In fact, Lama-
ism213 accepted the whole host of Indian Buddhas and Bodhi
sattvas with additions of its own. Among the Buddhas those
most worshipped are Amitabha, Sakya and Bhaisajyaguru or
the Buddha of Healing ; among the Bodhisattvas the popular
ones are Avalokita, Maitreya and Manjusri. The peculiarities
of Tibetan Buddhism lie in the Tantric phase which is eschewed
212. W addell: Buddhism, p. 131.
213. For a comprehensive study of Lamaism, see the Bibliography pro- •
vided in L.A. Waddell’s article on the subject in the Encyclopedia o f Religion
and Ethics, Vol. VII, pp. 784 ff. See also Eliot : Hinduism and Buddhism,
Vol. Ill, Chapter LII relating to the Doctrines of Lamaism, pp. 382, ff.
elsewhere. All the deities turning familiar spirits are coerced by

spells. Great prominence is given to goddesses either as the
counterparts of male deities or as independent. These deities
appear in various forms—as mild, angry or fiendish and even
the benevolent ones are not immune from furious frenzy. The
tutelary deities of individuals are called Yi-dam like the Ista-
devatas of the Hindus. The most efficacious tutelaries are tantric
forms of the Dhyani Buddhas, especially Vajrasattva, Vajra-
dhara and Amitayus.
Another class of tutelages consists of the so-called Buddhas,
accompanied by Saktis and are terrific in form. Some of them
such as Mahakala and Samvara seem to reveal their identity
from their name while there are others like Hevajra, Buddha-
kapala and Yamantaka who have grotesque features. Several of
these are well-known figures in Hindu mythology. Hayagriva,
the horse-necked god, appears to be connected with Visiju rather
than Siva- The Mongols regard him as the protector of horses.
Other gods include Yama, the Indian god of the dead, Maha
kala, the form of Siva who inspired Paqspa to convert Kublai
Khan, Lha-mo the goddess Devi the spouse of Siva and Cam-
Sran, a war god, a Tibetan form of Karttikeya. Other deities
frequently included in this group are Yamantaka, Kubera or
VaiSravaga, and a deity called white Brahma (Thsangs-pa-
akarpo). The Lamaist books mention numerous other Indian
divinities such as Brahma, the thirtythree Devas, the kings of
the four quarters etc. as also Nagas, Yak§as and Raksasas,
rather a part of the old folklore of Tibet. The great goddess
Tara is described in Lamaist theology as the spouse of Avalokita.
Originally benevolent and depicted with the attributes of Laksmi
she is transformed into various terrible shapes.214 Twentyone
Taras are often enumerated. Amongst them the Green Tara is
214. Tara is a goddess of India, Nepal and the Lamaist Church and al
most unknown in China and Japan. Tara means ‘a star’ as also ‘she who causes
to cross’, that is who saves life and its troubles. It is not known if the name
was first used by Buddhists or Brahmanas but after the seventh century, there
was a tendency to give Tara the epithets bestowed on the Saktis of Siva and
assimilate her to these goddesses. Thus, in the list of her 108 names she is
described among other more amiable attributes as terrible, furious, the slayer
of evil beings, the destroyer and Kali. (Eliot : Op. cit, Vol. II, 18-19)
Buddhism and Buddhist Savants o f Cental Asia 167
the commonest form in Tibet while the White Tara is the favou
rite of the Mongols. The goddess Marici is often depicted with
Tara.
Reference might as well be made to the use of praying wheels
and the famous formula Om mani padme hum.215 The origin of
both seems obscure. They were unknown in India. The total
absence of praying wheel in India as well as in the ruined cities
of Central Asia negates their origin or association with both
these places. The praying formula appears to be a Dharani and
seems to have come to Tibet with the first introduction of
Buddhism. The first and last words are mystic syllable Mani
Padme which are generally interpreted to mean the jewel in the
lotus. The formula was originally an invocation of the Sakti
under the title of Manipadma, although it is considered by the
Tibetans as an address to Avalokita. It is even suggested that its
prominence might be due to Manichaean influence, but in the
absence of the formula being in use in the Tarim basin, this
suggestion is negatived.
There is no reference to sects216 in Tibetan Buddhism prior
to Lang-darma’s217 persecution in the 9th century nor till more
than a century and a half later. The sectarian movement seems
to coincide with the visit of the great Indian Buddhist monk
Atisa in 1038. Atisa while clinging to Yoga and theistic Tantri-
215. Many scholars think that the formula Om manipadme hum, which
is supposed to be addressed to Avalokita, is readily an invocation to a form
of Sakti called Manipadma. (Eliot : op. cit, Vol. II, p. 17 n, cf. ERE, Vol. II,
p. 260 and JA. IX. 192). Waddell quotes a similar spell known in both Tibet
and Japan, but addressed to Vairocana Om Amogha-Vairocanamahamudra
manipadma-jvalapravarthtaya hurii—Buddhism, Op. cit, p. 149; Eliot : Op. cit,
p. 395 n.
216. For a study of sects in Tibetan Buddhism, see Encyclopedia o f
Religion and Ethics, Vol. VII, pp. 787-788; also Eliot Op. cit, Vol. Ill, chapter
LIII on ‘Sects’ in Tibetan Buddhism pp. 397 ff.
217. The last and perhaps the worst of Tibetan monarchs, Landarma,
commenced his reign by persecuting the Buddhist whom he considered his
mortal enemies. He was joined in his wicked plans of persecution by his prime
minister Batagna (Sbas-Stag-Sans). He reviled the first Chinese princess,
wife of king Sron-tsan-Gampo as an evil goblin (a Yak$ini) who had brought
the image of Sakya Muni into Tibet. It was for that inauspicious image that
the Tibetan kings were short-lived. The country was infested with maladies
and often visited by famines and wars. (Das : JASB, 1881, p. 229).
cism, a t once started a reformation movement on the lines of

the higher Indian Mahayanism enforcing celibacy and high
standard of morality and denouncing the Bon rites which had
crept into some of the priestly practices of the Buddhist monks.
Atisa identified himself with the Ka-dan or ‘those bounded by
the orders’. This sect three and a half centuries later in Tsong-
Kha-pa’s hands became less ascetic and more ritualistic under
the title of Ge-lug or ‘Virtuous Order’, the Yellow Hals.218
While.Atisa or the ‘Lord’ (Jo-bo-rje) was the great reformer
of Tibetan Buddhism, other parallel reformations were
initiated by his pupils. These were Kar-gyu and Sas-kya sects
which were directly based upon Atisa’s teaching to a great extent.
These two sects may be considered as less austere in their stress
on morality and discipline.
The remaining members of the Buddhist Church, bereft of
the best and most intellectual members, were called the ‘Old’
or Nying-ma, still adhering to the old corrupt practices. These
Nying-ma Lamas began to discover hidden revelations
(ter-ma), or fictitious gospels, ascribed to Guru Padmakara,
authorising these practices. The revelations treat mainly of Bon
rites and they prescribe forms of worship mostly of the Buddhist
model.These gospels formed the starting point for further sub
division of the semi-reformed and the old unreformed sects. The
distinctions between the various sects are partly theistic and
partly ritualistic. None of them relate to the personality or
doctrine of the historical Buddha as is totally accepted by all.
The differences are confined to the personality and title of the
primordial deity or Adibuddha, special source of divine inspira
tion and the transmitters, meditative system of mystical insight,
special tantra-revelation, personal tutelary (yi-dam) or Saivite
Indian protective demon and the guardian demon Dharmapala,
sometimes of the Tibetan type.
The Ge-lug or dominant yellow Hats have as their primordial
deity Vajradhara (holder of the thunderbolt) and they derive
their divine inspiration from Maitreya, the next coming Buddha,
as revealed through the succession of Indian saints from Asanga
down to Atisa, and then through the Tibetan saints Bronton,
218. Waddell : ERE, Vol. VII, p. 787.

Atisa’s pupil downwards to Tsong-nha-pa. Their tutelary Indian

demon (Yi-dam) is the fearful thunderbolt (Vajrabhairava)—
Tib Dorje-jig-je, supported by Sam vara and Guhyakala; and
their guardian demon (Dharmapala) is the six-armed lord
(Gon-po) or the horse-necked (Hayagriva— Tib-Tam-ch’en)
both of them Indian. The yellow-robe Lamas carried a begging
bowl which was soon dropped out of use. The ‘yellow Hat’
distinguished them from the other sects who wore red hats in
sharp contrast to the black caps of the Bon priests.
The next sect Kar-gyu219 was founded in the later half
of the 11th century by the Tibetan monk Mar-pa who had
visited India. Its distinct features are its hermitage, practice of
meditation in caves and other solitary places, and certain other
peculiarities. Its primordial Buddha is also Vajradhara and its
tutelary Samvara. Its mystical insight is Mahamudra (pyag-
rgya-chen) of the Middle Path, its Tantra Sun-Kar-bsdus, its
guardian ‘the lord of the black cloak (Bar-bag)’, its hat has a
frontal badge, to symbolise that meditation with crossed knees
is its special feature. With these is associated a stricter
observance of the Indian monastic rules. The strict hermit
features rendered it rather unattractive and several sub-sects
arose out of it which dispensed with the necessity of hermit
ages.
The third great reformed sect is the Sas-Kiya220 or Sa-Kya,
named after the monastery of that place, founded in A.D. 1071.
It became under imperial Chinese patronage the first great
hierarchy in Tibet and in 1251 attained for a time the temporal
sovereignty, until eclipsed by its later rival, the Ge-lug sect. Its
special source of inspiration is the Bodhisattva Manjusri,
through the Indian saints from Nagarjuna to Vasuputra (Vasu-
bandhu). Its mystic insight is the deep path (gambhira-darsana),
its tutelary Vajra-phurpa, and its guardians are ‘the tent-lord’
and ‘the presence-lord’ (Gon-po-Zhab). The Sakya sect connected
with the great monastery of the same name had acquired great
political importance from 1270 to 1340 with its abbots as the
rulers of Tibet. The historian Taranatha belonged to one of its
sub-sects and about 1600 settled in Mongolia with a monastery
219. ibid, p. 788.
220. ibid.
founded at Ugra. The main distinction is between the Gelugpa

or yellow Church and all other sects. The use of the yellow cap
and the veneration paid to Tsong-Kha-pa’s image are its external
manifestations.
As Buddhism was carried from Tibet to Mongolia, there seems
no difference between Tibetan and Mongolian Lamaism in
deities, doctrines or observances.221 Mongolian Lamas imitate
the usages of Tibet, study there when they can and recite their
services in Tibetan, although they have translations of their
scriptures in their own language. It is proposed by some scholars
that Lamaism was equally responsible for the political and
military decadence of the Mongols with the substitution of
priestly for warlike ideals. There may be many other factors
for this decadence. An analogy could be cited of other tribes
like the Turks and Tartars equally dissipating and collapsing.
The role of religion in this direction could be only marginal.
The Mongols were not only interested in Buddhism, as they
continue to be, they are equally anxious to frequently use
Sanskrit words, such as Manjusri or slightly modified forms
such as a Dara, Maidari (Tara, Maitreya) as pointed out by
Eliot.
In a long review of the history of the expansion of Buddhism
in Central Asia, one is likely to come across certain interesting
features. Different areas of this vast region received the message
of the Tathagata in different periods, and, of course, through
manifold channels—emissaries, scholars and the trading class.
The zeal and curiosity of the native elements was also a con
tributory factor to the expansion of Buddhism in those areas
of Central Asia. A regional study of Buddhist activity in these
parts with particular reference to the monastic establishment,
schools, scholars and their contributions in the translation of
Buddhist sacred texts has, no doubt, been attempted. With the
western region comprising Afghanistan, Bactria and Parthia, the
Central one including the present Russsian Turkestan, and the
Eastern one assimilating both the northern and southern wings
of Chinese Turkestan, and Tibet at the one end and Mongolia
at the other, the stage of Buddhist activities was set for a proper
221. Eliot : Op. cit, p. 401.

display of this religion as also for a fuller projection of the per

sonalities of the actors. The centres of Buddhist learning in
Kashmir as also in Afghanistan acted as feeders of Buddhist
monks and missionaries, and similar ones at Nalanda with
Vikramasila and their eminent savants equally contributed towards
the expansion of Buddhism in Tibet and through it in Mongolia.
The history of Buddhism in the western region of Central
Asia might be dated from the time of the Indo-Bactrians,
though the teachings of the Lord might have found their way
there even earlier. In this context the Kandahar bilingual edict
of Asoka is considered as an eloquent testimony to the exten
sion of Buddhism in the direction of Central Asia. The progress
of Buddhism to the north of Afghanistan is also evident from a
Kharosthi inscription on a clay object recording a Buddhist
name. It is from the first layer of the Begram excavations dated
between the third and second centuries B.C. Further, the as
sociation of Demetrius and Menander with Buddhism is well-
known, as also the contribution of the Kusana ruler. Kaniska.
In his time the famous scholar Ghosaka, born in Tokharistan,
played a prominent role in the deliberation of the
Fourth Buddhist Council. The Vaibhasika school, traced to
Ghosaka, had its treatises first translated into the Tokharian
language by a monk named Dharmamitra. In this context the
role of the Tokharian Buddhist monks in the propagation of
Buddhism was consistent with the pioneer work undertaken
by Lokasena, a monk of rare learning who went to Loyang
in China in A.D. 147 and translated many works. Parthian
and Kubhan scholars also distinguished themselves, the
former being signified with the prefix An. A Parthian
Mahadharmaraksita is said to have participated in the great
celebrations in the time of Duttagamini in Ceylon.
The Sogdians, originally from Samarkand, had their colonies
in different parts of Central Asia and their monks were equally
interested in the profession and propagation of Buddhism.
Senghui, a Sogdian, was the first to introduce Buddhism in
Southern China. The southern States of Central Asia experien
ced active Buddhist activities in the first few centuries of the
Christian era. Of course, those lying bn the southern silk trade
route, Kashgar, Yarkand, Khotan and other sites of Kroraina,

were predominently Mahayanist, while those on the northern
one comprising Kucha and Turfan were centres of Sarvastiva-
dins. This regional predominance of different schools could be
suggestive of two separate waves of Buddhist activities in Central
Asia. The Chinese pilgrims have recorded the number of mona
steries of different schools and the places visited by them.
The band of Central Asian Buddhist scholars, from Kumara-
jlva onwards who contributed towards the dissemination of
Buddhist religion and thought in Central Asia and also in
China through expositions and translations of sacred texts,
comprised scores of such galaxy of savants. Their list, fortu
nately, is available in the catalogue of Chinese Buddhist texts.
Each century produced its own set of scholars who tried to
emulate the examples of their predecessors. Buddhabhadra,
Sahghabhuti, Gautama Sanghadeva and Pupyatrata might be
quoted as some of the best specimens of Central Asian Buddhist
scholars, whose life and academic activities are properly record
ed and assessed.
The study of Central Asian Buddhism and its savants would
no doubt be incomplete without reference to Tibet and Mongolia
which continue to be strongly Buddhists. Their native and indi
genous beliefs and practices were assimilated in Buddhism in
such a fine manner that the original ethos is not completely
obliterated. Buddhism became the religion of the people, thanks
to Indian savants like Padmasarabhava, Kamalasila, Ati$a
Dlpankar and the contribution of host's of Indian Buddhist
Pandits in the land of snow. Mongolia accepted Tibetan Lamaism
in its totality. Their religion—hierarchic in character and equally
symbolising control over political affairs—is defined as a mixture
of late Indian Buddhism and Hinduism with various Tibetan
practices and beliefs incorporated in it. The position of the Dalai
Lama with his ecclesiastical prestige and political importance as a
temporal sovereign of Tibet and the fiction of continued existence,
was a stabilising factor in both countries. The Tibetan Buddhists
were no doubt divided into three sects, but the Yellow Hat ones
predominate.
A study of Buddhism in Central Asia in proper historical per
spective from the early centuries of the Christian era and the
period slightly preceding it, to about the seventeenth century
Buddhism and Buddhist Savants of Central Asia 173
A.D., has no doubt been made with particular reference to men
and matters connected with it. The finds of Buddhist manuscripts
as also objects of art and artistic importance no doubt call for
separate studies. These could as well be made in the context of
different areas and schools of religion as also of art. The material
culture, associated with Buddhism and its believers—the lay
members—who played an active role in their socio-economic
activities, as also contributed to the development of their per
sonality in an organized and effective manner, demands inde
pendent and separate study.
CHAPTER IV
LANGUAGE AND LITERATURE
The manuscript finds in Central Asia by different missions

engaged in activities of exploration and excavation in the course
of a fairly diversified and long period1 reveal certain interesting
features worth noticing. Firstly, these are primarily concerned
with Buddhism—both of the Sarvastivadin as well as of the Maha-
sahghika (Mahayana) schools, as also of the later ones in their
more evolved esoteric forms. These manuscripts include frag
ments not only of Sanskrit texts,butalso of translations of Indian
1. The history of exploration of Central Asia, particularly Chinese
Turkestan is traced in full by J.K. Dabbs in his book History o f the Discovery
and Exploration o f Chinese Turkestan, London, 1971. Earlier, Kali Das Nag
in his work ‘Greater India', Bombay, 1969, described in brief explorations in
Central Asia, beginning from A. Regal, a German Botanist in the service of
Russia, in the oasis of Turfan in 1879. The acquisition in 1891 of the famous
birch-bark codex by Col. Bower caused a great sensation amongst the Indo
logists. It is now preserved in the Bodlein Library at Oxford. It contains seven
texts of which three have medicinal contents. It is said to belong to the second
half of the 4th century A.D. (JASB, 1891, pp. 79 flf). Several manuscripts of a
still earlier period were found in later explorations, like the dramatic fragments
of Asvagho$a, collected by the German mission and the manuscript of the
Udanavarga, a Sanskrit version of the Dhammapada, brought by the French
mission. Both are written in quasi-Kusana character of the 2nd century. There
was a regular campaign among scholars of different nationalities for the
collection of manuscripts at government representative’s levels as well as at
private ones. These were done through agents. The British collection was
catalogued by Rudolf Hoernle, who could publish only the first volume
entitled ''Manuscript Remains o f Buddhist Literature found in Eastern Turkestan,
London, 1916. In this context the names of Dutrevil de Rhines of the French
mission in 1892, the founder of the birch-bark manuscript of the Prakrit
Dhammapada at a place not far from Khotan, Aurel Stein, the Great Hungarian
explorer, leader of several Indian missions, and A. Von Le Cog of the German
expedition may be recorded for their pioneering work in the fields of explor
ation and excavation of ancient sites in Central Asia, as also for the discovery
and acquisition of manuscripts including those from the famous Tun-huang
library.
Language and Literature 175
works in Central Asian languages, like Kuchean and Kho-

tanese, named after the regions where these were spoken. A brief
catalogue of the languages represented in the manuscripts and in
scriptions suggests that many influences were at work in Central
Asia which was notable as a receiving and distributing centre.
As such, the number of tongues simultaneously in use for popu
lar or learned purposes was fairly large. This might be evident
from the huge collection of manuscripts in different languages
from Tun-huang as also from a small one at Toyog, represent
ing Indian, Manichaean, Syriac, Sogdian, Uighur and Chinese
books. The writing material includes imported palm leaves,
birch bark, plates of wood or bamboo, leather and paper, in use
from the first century A.D. onwards.
It appears that with the establishment of the centralised auth
ority of the Kusanas over this vast area of Central Asia touch
ing Meru, Tashkentand Kashgar with base in the Northern Indian
Sub-continent, cultural, commercial and religious contacts were
closely established between one region and the other. In this
period Buddhism became very popular in Central Asia, now re
presented by the Soviet and Chinese Turkestan, and with it were
introduced Indian thoughts, languages and literatures, as also the
writing media—first Kharosthi, widely prevalent in the North-Wes
tern part of the Indian Sub-continent and lateron Brahmias well.
The Indian impact on the literary horizon of Central Asia was
heightened by the contributions of Indian Buddhist scholars to
that region and further to China, as also by the visits of savants
from these countries to India in search of manuscripts as also for
satisfying their religious urge. This age of Buddhist intellectual
curiosity lasted for several hundreds of years with the contribu
tions of men of different nationalities—all wedded to the religion
of the Tathagata—trying to translate the life activities and teach
ings of the Lord. The code of discipline and conduct introduced
by him as well as the philosophical and mysterious element in
Buddhism, passed into practice as also in their languages.
In this context it may be mentioned that the translators were
probably not unfamiliar with the original Pali texts although they
indented on the Buddhist Sanskrit ones. The Sarvastivada school
of the Hinayanists had its adherents especially in Kashmir and
Gandhara, from where this school spread to Central Asia, Tibet
and China. It had a Sanskrit canon of its own.2 In wordings

and in the arrangement of the texts, this Sanskrit canon evinces
great similarity to the Pali canon, of course, with many points of
difference as well. This might be due to both having a common
source, probably the lost Magadhi one, as proposed by Winter-
nitz, from which first the Pali canon branched off in one part of
India, and then, later on, the Sanskrit canon in another one.
Here reference might be made to fragments of the Pratimoksa-
sutra3 of the Sarvastivadins, as well as other texts of the Vinaya-
pitaka of the Sanskrit Buddhist texts, which were found in Cent
ral Asia (Eastern Turkestan), and a few in Nepal as well. The
Chinese and Tibetan translations are no doubt helpful in recons
tructing the Pratimok$a-sutra. The Sanskrit texts of the Sarva
stivadins and the Vinayas of the Mahisasakas, Dharmaguptas and
Mahasaiighikas, evidently show manifold difference and diver
gences in separate details from the Pali Canon and from one an
other, though the original stock of rules is one and the same.
Numerous Sanskrit writings have been found, all dealing with
religious or quasi-religious subjects, such as medicine and gram-
2. According to Winternitz, though no complete copy of this canon has
come down to us, we know it firstly from many fragments, large and small,
which have been discovered among the manuscripts and block-prints brought
from Eastern Turkestan by Stein, Grunwedel, Von Le Cog, Pelliot and others;
secondly from quotations in other Buddhist Sanskrit works, such as Maha-
vastu, Divydvadana and Lalita-vistara, and lastly from Chinese and Tibetan
translations. The principal texts of the Mula-Sarvastivadin canon were trans
lated from Sanskrit into Chinese by the Chinese pilgrim I-tsing in theyears
700-712. There are, however, translations of single texts dating from the
middle of the 2nd century onwards, and these were adherents of the Sarvasti-
vada in India, as early as the 2nd century B.C. Winternitz: History o f Indian
Literature (HIL), Vol. II, p.232 with references to papers of other scholars
on the subject.
3. The Sanskrit text of this work is edited by Finot on the basis of the
fragments of manuscripts found by Pelliot in Central Asia and it is translated
into French, with the Chinese version of Kum&rajiva, by Huber (JA.Ser.II,
t. 11, 1913, pp. 465-558). A fragment from Stein’s collection is translated by
Louis de La Val6e Poussin in JRAS, 1913, p. 843; a fragment in the Kuchean
language by S. Levi in J.A., S. 1 0 1. XIX, 1919, pp. 101 ff; JRAS, 1913, pp.
109 ff; Hoemle, Manuscript Remains (pp. cit) pp. 357 ff. Waldeschmidt in his
Bruchsticke des Bhiksuni-Pratimoksa der Sarvastivadins, Leipzig, 1926, has
edited the fragments of manuscripts of the Pr§timok$a of the nuns, which
were found at Turfan, comparing these with the corresponding texts of all
the other schools. (Winternitz : Op. cit, p. 233 and n).
mar. While the Central Asian manuscript finds are mostly

Mahayanist, greater interest attaches to portions of an otherwise
lost canon which agree in substance though not verbally with the
corresponding passages in the Pali canon, and is apparently the
original text from which much of the ChineseTripitaka was trans
lated. The published manuscripts4from Central AsiaincludeSutras
from the Sariiyukta and Ekottara Agamas, a considerable part of
the Dharmapada, and the Pratimoksa of the Sarvastivada school.
Interest in Sanskrit—its use and popularity—among the monks
of Central Asia is also recorded by Fa-Hsien,5 who states that the
monks of Central Asia were all students of the language of India.5
The other Chinese pilgrim Hsuan-chuang passing through Kucha in
the seventh century confirms the statement of his predecessor.8
Portions of a Sanskrit grammar were found near Turfan, and it is
4. The published material includes fragments of the Sanskrit canon

discovered among the Manuscripts found in Eastern Turkestan, and edited
by Pischel (SBA 1904, pp. 808 f, 1138 ff; Levi in Toung Pao, S.2, Vol. V,
1904, p. 1297 ff; JA. S 101XVI, 1910, pp. 433 ff, 450 ff; JRAS, 1911, p. 764 ff;
L de La Vallee Poussin in JRAS, 1911, p. 772; 1912, p. 1063 ff; 1913, p. 569 ff;
Hoernle : Manuscript Remains, pp. 16-52). A fragment of the Srong-SOtras
of the Samyukta-Agama (cf. Samyuttanikaya 22, 49 f = Vol. Ill, p.48 ff’) has
been edited by La Vallee Poussin from Mss of Cecil Bendall’s collection in
JRAS 375 ff. Quotations from the Samyutta and Ekottara Agamas have been
traced by S. Levi in the commentary on Asanga’s Mahayana-SQtralamkara.
(Winternitz ; Op. cit, p. 234, n. 3). According to Winternitz, who records
points of agreements and divergences, it may be suggested that the Sanskrit
Agamas and the Pali Nikayas were compiled from the same material, but were
arranged in different ways in the different schools (ibid, p. 235).
5. Beal : Buddhist Records o f the Western World, Vol. I, p. xxiv. Des
cribing the country of Shen-Shen (Lou-lan), the king of this place is said to
honour the law (of Buddha). There are some 4000 priests, all of the Little
Vehicle belief (learning). The laity and the sramapas of this country wholly
practise the religion of India. From this proceeding westward, the countries
passed through are all alike in this respect, only the people differ in their
language (Hu words). The professed disciples of Buddha, however, all use
Indian books and the Indian language.
6. ibid, p. 17. Noticing the country of Kiu-chi (Kuche), this Chinese
pilgrim refers to a hundred convents (sahgharamas) here with five thousand
and more disciples, all these belonging to the Little Vehicle of the school of
the Sarvastivadins (Shwo-Yih-tsai-Yu-pu). Their doctrine (teaching of the
SQtras) and their rules of discipline (principles of the Vinaya) are like those
of India, and those who read them use the same (originals).
suggested7 that in the earlier period Sanskrit was probably under

stood in polite and learned society. This is also evident from the
finds at Ming-oi of some palm leaves containing fragments of
two Buddhist religious dramas, one of which is the Sariputra-
prakarana of Asvaghosa.8 The handwriting seems coeval with
the period of Kaniska, thus pointing to the oldest specimen of
Indian dramatic art as also of the antiquity of Sanskrit manus
cripts in Central Asia. The dramas are no doubt written like the
Indian classical ones in Sanskrit and also various forms of Pra
krit. Another Prakrit text found in Central Asia is the version
of the Dharmapada written in Kharosthi letters. It was disco
vered by the Dutrevil de Rhins mission near Khotan.9 The same
region also provided numerous official documents in Prakrit writ
ten in Kharosthi.
7. Eliot : Hinduism and Buddhism (Eliot : Op. cit), Vol. Ill, p. 190. The
Kuchean and Tokharian fragments—all translations of Sanskrit Buddhist
texts—contain some bilingual pieces with original Sanskrit by the side of
translations. These bilingual texts must have been used as handbooks for
teaching Sanskrit to local Buddhists. A Buddhist scholar of Kucha named
Li-yen compiled in the 7th century a Sanskrit lexicon which is preserved in
the Chinese Tripi(aka. The lexicon modelled after Sanskrit kosas, also
contains some words of Central Asian origin (Bagchi. India and Central Asia,
pp. 106, 111); also Bagchi. Deux Lexiques Sanskrit Chionis, Vols. I and II
(Paris, 1929-37; B.N. Mookerji in the Cultural Heritage o f India, Vol V,
pp. 710-11).
8. Among the works of Asvaghosa, a fragment of the Buddha-caritra
brought from the Turfan region shows that the text was studied by the Buddhist
monks of Central Asia. A drama entitled Sdriputra-prakarana discoveied
in the same region, and not known from other sources, is also ascribed to
Asvagho$a. Only portions of the original work have been found, but they
clearly point to its remarkable character. It is the oldest Indian drama known
to us. It is in Sanskrit but there are dialogues in Prakrit. See Luders : Bruck-
stucke buddhistisclar Draman (1911); Keith : The Sanskrit Drama (London,
Rep. 1959) pp. 80 ff.
9. This important manuscript in three small oblong birch-bark leaves
was acquired in the spring of 1892 by the French traveller Dutrevil de Rhins
in Komari Mazar in the valley of the Karakash Daray, 21 Kilometres (about
13 miles) from Khotan and was sent to France along with other finds. The
text was published by M. Senart with palaeographic and philological notes
in the ‘Journal Asiatique’ for the year 1898 (IX e Serie, tome XVI, pp. 193 flf).
Several scholars made important contributions on its date and place of origin.
According to Buhler, it belonged to the first century A.D. and it was brought
from India to Chinese Turkestan by a Buddhist monk. (On the origin o f the
Languages
It is equally evident from the Central Asian literary finds that
in the early centuries of the Christian era two distinct languages
prevailed—one spoken in the north and the other in the south.
These are differently named by scholars, as for instance, the nor
thern one is named ‘Tokhari’ by Muller, and the Southern one
‘Nordarisch’ by Leumann, and ‘Saka' by Liiders. These have
not found favour with scholars, and it would be preferable to
accept the suggestions of Sylvain Levi and Sten Konow to name
these according to the centres or capitals of the territories with
which the northern and southern ones were associated, namely
Kucha or Kuchar and Khotan respectively.10 Kuchean is sup
posed to be an Indo-European language of extremely early affinities
with the two great western and eastern groups. Spoken princi
pally on the northern edge of the Tarim basin, it is also called
Tokharian which name implies its association with the Tokharas
or Indo-Scythians.11 There is, however, no proof of this and it
Indian Brdhml Script, Strassburg, 1898, p. 122.) Jules Bloch on philological
grounds places its origin in the N.W. part of India (JA. Xe Serie, tome XIX,
pp. 331 IT). Sten Konow suggests its composition in a dialect of North-Western
India, but was written down in Khotan where it was discovered (Festschrifb
Windisch. Leipzig, 1914, pp. 85 ff). The manuscript has also been edited by
B.M. Barua and S.N. Mitra, with adjustments and notes (Calcutta, 1921).
The Kharo$thi Manuscript of the Prakrit Dhammapada figures as a section
in Kalidas Nag’s work (Greater India, op. cit, with fuller references, pp. 245-
247).
10. A.F.R. Hoernle in the General Introduction to his work on Manu
script Remains o f Buddhist Literature found in Eastern Turkestan discusses
this question in detail, taking into account the views of other scholars on the
subject with proper references. Eliot in his work (op. cit, p. 191) as well records
the views of Luders {Die Sakas and die Nordarische sprache' Sitzungsber de
Kon Preuss Akad. 1913); Sten Konow. Goffing Gel Anz. 1912, pp. 551 IT;
and Hoernle in JRAS, 1910, pp. 837 ff; and 1283 ff: 1911 pp. 202 ff; 447 ff.
11. P.C. Bagchi notices in detail Kuchean or western Arsi—a forgotten
language of the Indo-European family in his India and Central Asia (op. cit,
pp. 152 ff) with Sylvain Levi's work entitled Fragments des Texts Koutcheans
including a section in Hoernle's work (op. cit) pp. 357 ff and some fragments
belonging to the collection of the Pelliot's Mission, some to the Stein, and a
few others to the Russian Mission. Of the two dialects marked A and B, the
first has been deciphered and interpreted from the manuscripts of the German
collection of Grunwedel and Von Le Cog by Sieg and Siegling. The second
dialect has been deciphered and interpreted by S. Levi from the French
is safer to speak of it as the language of Kucha or Kuchanese.

It exists in two different dialects called A and B, of course, with
uncertain distribution. Numerous official documents dated in
the first half of the seventh century suggest that it was the ordi
nary speech of Kucha and Turfan. As a literary language it
was in use for many translations, including versions of the
Dharmapada and Vinaya. 'As pointed out earlier, it belongs to
the Aryan family and is related more clearly to the western than
the eastern branch, showing affinities to Latin, Greek, Celtic, Sla
vonic and Armenian.12
The other language associated with the southern region of the
Tarim basin, sometimes called Nordarisch and regarded by some
scholars as the language of the Sakas, is designated the Khotanese
language. The manuscript remains in this language suggests
two stages in its evolution—the earlier and the later. In the earlier
one it is associated with the Buddhist Canonical literature13 while
the later stage is confined to the official documents of the eighth
century. This divergence could be explained with reference to
the introduction of Buddhism into this part of Central Asia from
north-western India as early as the beginning of the Christian era.
This was soon followed by the translations of the principal Bud
dhist Canonical texts from the original Sanskrit, or, as supposed
by Hoernle, from the Indian vernacular of those days into the
language of the native people of Khotan. The process continued
from time to time with the growth of Sanskrit Buddhist literature.
collection along with those collected by other Missions. The documents written
in A dialect came from the region of Karasahr and its neighbourhood whereas
those in the B dialect were from the region of Kucha and its adjoining places.
According fo Bagchi, it is risky to connect the two dialects with the Indo-
Scythians (op. cit, p. 125). »
12. Eliot, (op. cit) p.191. While the numerous papers on this language
are naturally quickly superseded, the two important ones : Sieg and Siegling
4Tokharisch’, 'Die Sprache der Indoskythen’ (Sitzungsber—der Berl.. Ak.
wiss, 1908 p. 815) and that by Sylvain Levi ‘Tokharian B, Langue de Koutcha’
J.A. 1913, II, p. 311 may be mentioned, according to Eliot.
13. Hoernle, Op. cit, p. xii. According to this scholar, Khotanese must
be classed with the Iranian languages. Its study has been facilitated through
the recovery of several complete texts by Aurel Stein in the immured library
of Tun-huang. This language continued to flourish as a spoken one in the terri
tory of Khotan as late as the eighth century A.D.
According to Eliot,14 the basis of the Khotanese language is Ira

nian but strongly influenced by Indian idioms. Many translations
of Mahayanist literature, as for instance, Suvarnaprabhasa, Vajrac-
chedika and Aparamitayus-sutras were made into it. This lan
guage appears to have been in use as a spoken one principally in
the southern part of the Tarim basin.
Besides these two languages in use in Central Asia and con
nected with the Buddhist literature, three other Iranian languages
have left literary remains in Central Asia. These are all written
in an alphabet of Aramaic origin, unlike the Brahmi and Kha-
rosthi scripts associated with the earlier two. Two of these Ira
nian languages are supposed to represent the speech of south
western Persia under the Sassanids and of north-western Persia
under the Arsacides, as suggested by Eliot.1516* Both these langu
ages have preserved only Manichaean texts, but the third one,
called Sogdian, offers a more variegated literary contentwith Bud
dhist, Manichaean and Christian texts. It was originally the
language of the region around Samarkand but soon acquired an
international character.18 It was used by merchants throughout
the Tarim basin and from there it spread even to China.
14. Eliot. Op. cit, p. 191 n. 1. See Luders ‘Die Sakas und die Nordarische
Sprache’ Sitzungsber der Kon Preuss Akad, 1913; Konow. Gotting Gel. Anz,
1912, pp. 551 ff.
15. Eliot. Op. cit, p. 191. According to Aurel Stein, the languages rep
resented in these Brahmi texts from the walled-up board are mainly Sanskrit
and that ‘unknown’ tongue of Iranian type for which the term’ Khotanese’
now recommended by Sten Know and Hoernle, appears the most convenient
provisional designation (Serindia. Op. cit, Vol. II, p. 814, 9i4; 111, p. 1289).
This Iranian-Unknown so-called North-Aryan language is traced in MSS
from Khadlik {ibid, Vol. I, pp. 155, 158, 164); on wooden tablets from Mazar-
toghrak, 205 Sq-
16. Aurel Stein discusses the language and script of the early Sogdian
documents from Tun-huang. He suggests that from the outward appearance
of these strange documents, it appears that the writing was in the same un
known script, resembling early Aramaic, like the one noticed on a small piece
of paper from the Loulan site (LA. VI.ii.0104). This Semitic script found on
the border of China might probably have been used for an Iranian language
{Serindia, op. cit. II. p. 675). The original homeland of this Sogdian language
seems to be Scythia infra Irnaon of Ptolemy of the first century of Christ, and
it was probably carried by traders to the Tun-huang region. Further, Stein
quotes Robert Gauthiot’s views that the language of the documents was an
early form of that Sogdian, first revealed in Buddhist manuscripts recovered
A Turkish dialect written in Uighur alphabet, derived from the

Syriac, was (like Sogdian) widely used for Buddhist, Manichaean
and Christian literature.*17 Uigur, named after the alphabet, ap
pears to have been the literary form of the various Turkish idioms
spoken north and south of the Tien-Shan. The use of this dialect
for Buddhist literature gained momentum when the Uighurs
supplanted the Tibetan power in the Tarim basin about 860 and
founded a kingdom of their own there. This language was in
use in China as well, with the Sutras printed in it in Peking in
1330 and several manuscripts were also copied in this language
much later on.
The Tibetans with their hold over the Tarim-basin from
at least the middle of the eighth until a century later, were not slow
in providing Buddhist canonical literature in their language
Tibetan manuscripts18 have been found in the regions of Khotan,
Miran and Tun-huang. Traces of Tibetan influence in Turfan
are fewer though not completely absent. The Tibetan documents
from Turfan. Their writing was shown to represent a cursive Aramaic, inter
mediate between the Aramaic proper and the Sogdian script from which the
Uigur script had been evolved (Serindia.ll.p. 675).
17. The name Uighur (Uighur) is perhaps more correctly applied to the
alphabet than the language which appears to have been the literary form of
various Turkish idioms. A variety of this alphabet written in vertical columns
is still used, according to Eliot (op. cit. III. 192) in some parts of Kansu where
a Turkish dialect is spoken, and this appears to have been introduced into
Khotan only after the Moslem conquest. Several Buddhist manuscripts in
Uigur were recovered from many sites in Central Asia, including Turfan sites
(Stein : Innermost Asia—op. cit, 1147), and Khara-Khoto (ibid) 1049. A
translation of a portion of the Saddharma-pundarika was found at Turfan
(Eliot. Op. cit. III.p. 192n; Winternitz. Op. cit 304 n) and also of DiSa-
svastika-sutra from the same area, Turfan (ibid, 384 n). Fragments of
Uighuric texts were edited by W.K. Muller in his Uigurica (Abhandlungen
der Berliner Akademie der Wissen Schaften ABA. 1908, pp. 10 ff; and Luders :
Sitzunqsberichte der Preussischen Akademieder Wissenchaften—S.B.A. 1914,
p. 99—quoted by Winternitz. Op. cit, p. 304 n.).
18. The abundance of Tibetan texts in the walled-up Hoard at Tun-huang
suggest that Buddhism there must have been subject to an even more powerful
influence from the South during a certain period (Stein, Serindia.II.p.816).
Pothi’s Manuscripts in Tibetan were recovered from M ran (ibid.1.p.462),
Chen-Fo-tung (ibid.II. pp. 816, 823,919 Ser.). See Notes on the Tibetan Manu
scripts by F.W. Thomas—Appendix R. Innermost Asia (op.cit, pp. 1084-86),
Serindia, pp. 1410-7. See also Stein. Khotan—op. cit—Appendix B and
Franke. JRAS. 1914, p. 37).
discovered are anterior to the ninth century and comprise numer

ous official and business papers as well as Buddhist translations.
These are important for the history of the Tibetan language as
also for the association of Tibetans with Buddhism and their
allegiance to the Tathagata, without completely alienating them
selves from the traditional religious ethos—the Bon religion of
the homeland.
A large number of Chinese texts—both religious and secular—
have also been recovered from the principal religious centres.19
These are interesting for providing political information, parti
cularly relating to old military outposts near Tun-huang, and in
tercourse between Central Asia and China between 98 B.C. and
A.D. 153. Some documents of the Tang dynasty are Manichae-
an in character with an admixture of Buddhist and Taoist ideas.20
The use of Gandhari Prakrit21 written in Kharosthi seems to
19. For the finds of Chinese manuscripts from Central Asian sites,
see Stein : Innermost Asia, p. 358 (Chen-Fo-ting); p. 449 (Khara-Khoto);
p. 636 (Murtuk) pp. 647, 652, 665, 689 (Astana). See also E. Chavannes
‘Chinese inscriptions and Records—translated and annoted’—Appendix A.
Serindia.III. pp. 1329-1339; Pelliot. BEFEO. 1908. pp. 508 ff; Chavannes :
Documents Chinois decuverts par Aurel Stein. 1913 (quoted by Eliot. Op. cit.
III. 197).
20. See Chavannes and Pelliot ‘Trait Manichaen’ in J.A. 1911 and 1913,
Eliot. Op. cit, p. 193 n. The discoveries at Turfan sites have furnished ample
reason for belief that Manichaen and Buddhist worship had existed there
probably side by side among a population which had come relatively earlier
than Turkish domination as well as racial influence. Pelliot discovered a
fragment of a Chinese treatise manifestly setting forth points of Manichaean
doctrine {Serindia—Op. cit.III.p 922; See Pelliot. BEFFO. VIII. p. 518; also
Chavannes— Pelliot—‘Untraite Manichaen retrouve’ en Chine’. JA novembre-
decembre. 1911. pp. 499-617).
21. This form of Prakrit, now called by scholars as Gandhari Prakrit,
agrees closely with the language of the post-Asokan Kharo§thi inscriptions
of the north-western part of the Indian sub-continent (including Gandhara).
It differ from other varieties of Prakrit according to the degree of modifications
in its inflectional system. It was subjected to two foreign influences viz.
Iranian and that of the native language of Krorania. Loan words from several
non-Indian languages like Iranian (including Sogdian), Greek, Tibetan etc.
may be traced in Gandhari Prakrit. Several innovations were made in the
Kharos(hi script in writing Gandhari Prakrit, and so also was the pronunciation
of Prakrit words affected by the Phonetic structure of the Koraini; (See
J. Brough : The Gandhari Dhammapada, London 1961; H.W. Bailey :
Gandhari, BSOAS. Vol. XI, part 4, pp. 764ff; Mookerji. B.N. : The Cultural
Heritage o f India, Vol. V, Section on Central Asia Literature, pp. 703 ff).
mark the beginning of the introduction of Buddhism and Bud

dhist literature in Central Asia. This is evident from the finds
of manuscripts as also of numerous inscriptions—all in KharosthI—
which seems to have been introduced in Central Asia earlier than
BrahmI. According to H.W. Bailey,22 the name Gandharl has
a sufficiently wide scope and includes the Buddhist literary text.
The Dharmapada found in Khotan, written likewise in Kharos-
thi, the KharosthI documents on wood, leather and silk from
Niya (Cadota) on the border of the ancient kingdom of Khotan
representing the official language of thecapital Krorayina of the
Shan-Shan kingdom; and one document, no. 661, dated in the
eign of the Khotanama/idraya raydtirayahinajhadeva vijida sithha.
With this more copious material must be grouped the scattered
traces of the some Middle Indian dialects in Khotanese, Tibetan,
Agnean, Kuchean, the earlier Buddhist translations, as in parti
cular in the Dirghama of the Dharmaguptaka sect and the re
mains inSogdian, Uighur, Turkish and in Mongol (in living use),
and also in Manchu texts. It appears that this Gandharian Pra
krit text played an active role in the dissemination of Buddhism
and its literature in Central Asiaand China. The same medium
language and script seems to have been employed in the north
western parts of India for recording donations and dedications
for the faith of the Tathagata.
Besides the Prakrit recension of the Dharmapada23 in Gand-
hari, of which fragments were found in Khotan in 1892 and 1897,
there are traces of a fairly good literature in this language and
script in Central Asia. Some literary pieces are noticed in a few
22. BSOAS.XI.iv. p. 764.

23. J. Brough : The Gandharl Dhammapada (Op. cit). The Dhammapada
texts in their various recensions were very popular in Central Asia and China.
Besides the Prakrit one, and the text of Udanavarga written in Sanskrit, a
collection akin to the Dhammapda, in some places more extensive, there are
four separate Chinese translations. These are based on four different recensions
of the text, three of which had been carried to China' from Central Asia. They
are Fa-Kiu-King-Dharmapada-sutra translated in 224 A.D.; Fa-Kiu-pi-yu-king-
Dharmapada-avadaha-sQtra translated between 290 and 306 A.D.; (iii) Chu-
yao-king translated in 398-99 A.D. The first two translations were probably
based on a text similar to that of the Prakrit Dhammapada (Bagchi : India
and Central Asia—Op. cit, p. 99).
Kharosthi records,24 and one such record mentions the subjects

of study like grammar, music, astronomy and science of poetics.
This language seems to be known to early translators of Buddhist
texts into Chinese. The earliest Chinese versions of the Sukha-
vativyuha (c 3rd century A.D.) shows influence of Gandhari. It
is also suggested that the Buddhist texts like the Dirghagama,
rendered into Chinese in A.D 413, were translated from Gand
hari Prakrit. This rendering might be due to the association of
the earlier translators with the Gandhara region.25 It is calculated
from the catalogue of Chinese translations of Buddhist canonical
works that until A.D. 316 among the translators were six Yueh-
chis, four An-hsin (Parthians), three Sogdians and six Indians,
who should have been familiar with the Gandharian Prakrit.
The practice of writing Gandhari Prakrit in KharosthI seems to
have continued till the sixth century A.D.26when Brahml became
popular in Khotan—the centre of Mahayanism and seems to.
have replaced KharosthI.
Scripts
The introduction of the Brahml script in Central Asia and its
association with Buddhism might be dated earlier than KharosthI27.
Brahml inscriptions—donative and didactive—have been found
24. Rapson, Senart and Boyer—Kharosthi Inscriptions Discovered by
Sir Aurel Stein in Chinese Turkestan (Oxford 1920) Nos. 501, 516 and 204,
514.
25. Mookerji—Op. cit; cf. BSOAS. XII. p. 764; XVIII. p. 609; XXIV.
p. 527 and Asia Major (old Series) Vol. II, pp. 270-7.
26. While the terminus ad quem for the use of writing Gandhari Prakrit
in Kharo?thi may be placed in the sixth century A.D. the earliest use of
Prakrit in Soviet Central Asia is suggested by a Kharosthi record from
Tajikistan, dated c. first century B.C. Reference might as well be made to
Kharosthi inscriptions from Wardhak and Qunduz in Afghanistan and at
Fayaz-Tepe and Kara-Tcpe near Termez in Tajikistan for the use of Kha-
rosttni script and the Prarit language for the dissemination of Buddhism in
those areas. (Corpus Inscriptionum Indicarum, CII. Vol. II (1) p. 170; B.A.
Litvinsky. Op. cit, p. 8; BEFEO. Vol. LXI (1974) pp. 22 and 58; Mookerji.
Op. cit, pp. 76-77 n).
27. This statement might be true in the context of Kucha and Karasahr
regions. In the southern one, Brahml supplanted Kharosthi. The finds of
Asvaghosa’s dramas in Brahml script from Turfan area suggest its terminus
a quo. The manuscript is written in Kushan character and may be placed on
palaeographic groups about the middle of the second century A.D. There
at Kara Tepe Soviet Central Asia.* 28 The script was used for
recording Sanskrit texts and three varieties of its alphabets are
traced. The first is a purely Indian variety of the time of the
Kusanas and the Guptas. The second and third ones are develop
ment of the later Gupta scripts. The purely Indian Brahmi script
is found in various Sanskrit manuscripts—the earliest being the
fragments of the dramas of Asvaghosa. On paleographic grounds
it may be placed about the middle of the second century A*D.
The manuscript containing fragments of the Kalpanamantfitika
of Kumaralata, written in Gupta characters, seems to be of the
first half of the fourth century A.D. The famous Bower manu
script also belongs to the fourth or the fifth century A.D.29
The other two distinct scripts used in the manuscripts from
Central Asia are technically known as the ‘slanting* Gupta script
and the upright Gupta script, associated with the Kuchean or
Tokharian and Khotan region respectively* According to Hoer-
nle,30 the northern or Kuchean Brahmi script, noted for its
upright ductus, was imported from India through immigrants.
In the hands probably of the natives of the country, the upright
type of Indian Gupta developed a more or less starting ductus.
This northern or Kuchean slanting type of Gupta script must
have originated at a very early period since some of the frag
ments exhibiting it were found along with those having the true
upright Gupta of the fourth or fifth century A.D.31 The deve
are fragments of a birch-bark manuscript in the French collection, written in

a quasi-Kushan character, and may be dated about the end of the second
century A.D. (Ghoshal : Greater-India—Op. cit, p. 255).
28. B. Y. Stavisky. Novuie Nahodici na Kam-Tepe V Staram Termez
(Moscow. 1975) p. 70; Mookerji, B.N. Op. cit, p. 707. Of the two Indian
languages, Prakrit and Sanskrit, known in Central Asia in the first millennium
A.D. Sanskrit was written in Brahmi. An isolated inscription containing the
whole text of the Pratitya-samutpada-sutra discovered from Tun-huang area
is also in this early Brahmi (Bagchi : Op. cit, p. 93).
29. The manuscript containing fragments of the Kalpanamantjlitika of
Kumaralata—edited and published by Luders has been placed in the first
half of the fourth century A.D. The Bower manuscript is also supposed to
belong to the fourth or fifth century A.D. (Ghoshal : Op. cit, p. 258).
30. Manuscript Remains (Op. cit) p. xiii. This is evident, according to
Hoernle, from the Vinaya fragment, No. 149, found in the vicinity of Bai,
west of Kuchar in the northern area of Eastern Turkestan.
31. Hoernle : Op. cit, p. xiii n. 14.
lopment of the southern, or Khotanese type of Gupta script

did not commence quite as early.32 In that part of Eastern
Turkestan, the slanting type was never in use. The Indian up
right Gupta script prevailed. Later on, gradually some of its
letters, notably those for the initial vowels or vocalic radicals
were modified in a calligraphic and a cursive form. The former
type was in use for writing the sacred books of the Buddhist
canon, and the latter one for literary works of non-canonical
nature and more commonly for the writing of public and private
letters and documents. The area covered by the Brahmi script for
canonical purposes was extensive. Meru seems to have been the
western-most point since it yielded a Sanskrit manuscript.33 In
the north-eastern region, Brahmi was regularly used for writing
not only Sanskrit, but also Kuchean, Agnean (language of
ancient Agnidesa or the Karashar area) and Saka-Khotanese in
the southern part of Central Asia. Brahmi was sometimes used
along with other scripts on the same leaves of manuscripts, as
for instance, some paper leaves from Khara-Khota, Turfan and
Mazar Taqh bear Chinese or Uighur with internear Brahmi
writing. The relationship of the Brahmi script with the Chinese
language is traced in a manuscript containing a Buddhist
Chinese text written in Brahmi characters similar to those used
in Saka-Khotanese documents.34
32. The use of Brahmi in the Khotan area seems to be later than in the
northern zone. Of course, sporadic use of Brahmi in Southern Chinese Central
Asia in the early centuries of the Christian era could be traced in three lives
in Brahmi characters of the Ku$ana period recorded on a wooden tablet with
a Kharosthi inscription on the other side. (Stein : Ancient Khotan, Vol. I
(Oxford 1907 p. 369.) The testimonies of Sung-yun as well as of Hsuan-tsang
suggest that Brahmi seems to be popular in Khotan in the sixth/seventh century
A.D. Saka (including Khotanese) manuscripts, written in Brahmi, are datable
between the seventh and the tenth centuries A.D. There are no doubt a few
traces of earlier use of Brahmi. A mutilated folio and a fragment of a Palm-
leaf pot hi in Sanskrit from Miran (No. M.l 1.0011) are written in the upright
Brahmi script of the early Gupta age. The pothi seems to be a part of a
glammatical text. Hoernle dates these to c. A.D. 400 (Serindia. Op. cit, p. 489).
It is, however, evident that Kharo§thI was in use in Khotan and Shan-Shan
etc. before Brahmi became popular (Bagchi : Op. cit, 93).
33. B.A. Litvinsky : Op. cit, p. 65; Journal o f the Asiatic Society (JAS),
Vol. XI (1969).
34. Stein : Innermost Asia (op. cit) Vol. Ill, pi. cxxv; Serindia (Op. cit)
pi. cii.
Sanskrit in Central Asia

The role of Sanskrit in the propagation and preservation of
Buddhism and its canonical works can well be ascertained from
the finds of Central Asian Buddhist manuscripts and documents
in this language in different parts of that region. It is not con
nected with any particular Buddhist school, like the association
of Pali with the Theravadins and their canonical works, but it is
patronised by the Sarvastivadins, Mahasanghikas as well as the
Dharmaguptikas. It was no doubt introduced into Central Asia
by the Sarvastivadin school of the Hinayana sect which had its
adherents more especially in Kashmir and Gandhara. This school
had a Sanskrit canon of its own. While no complete copy of this
canon has come down to us, its existence is confirmed by many
fragments, large and small, discovered among the manuscripts
and block prints brought from Eastern Turkestan by Stein,
Grunwedel, Von Le Coq, Pelliot and others.35 Quotations from
this religious text can as well be traced in other Buddhist Sanskrit
works, such as Mahdvastu, Divyavaddna, and Lalitavistara, and
also from Chinese and Tibetan translations.36 The principal
texts of the Mulasarvastivada canon were translated from Sans
krit into Chinese by the Chinese pilgrim I-tsing between A.D.
700-712.37 The sacred literature of this school of Buddhism in
Sanskrit was studied by scholars of several Buddhist kingdoms
35. cf. W. Geiger ‘Die archaeologischen und literarischen Funde in

Chinesisch Turkestan und ihre Bedeitung Fur die Orientalistische Wissen-
schaft’ (Rede bcin Antritt des Prorektorats) 1912; H. Oldenburg in Nachrichten
Vonder Kgl. Gesellschaft der missenchaften Gottingen—NGGW 1912,171 ff;
H. Luders ‘Uber die literarischen Funde Von Osturkestan in Sitzunqs-berichte
der Preussischen Akademie der Wissenschaften in Berlin—SBA; A.F.R.
Hoemle : Manuscript Remains o f the Buddhist Literature found in Eastern
Turkestan—Op. cit., I, Oxford, 1916; Winternitz : Op. cit, II, p. 232 n. 1.
See also Stein : Serindia and Inner Asia (Op. cit)—relevant references to
literary works in the Index, as also the Appendices.
36. cf. Oldenburg in ZD MG 52, 1898, 654 ff; 662 ff; M. Anesaki in
Transactions of International Congress of Orientalists (O.C.) Hamburg, 1902,
p. 61; JR AS. 1901, pp. 895 ff; Winternitz : Op. cit, p. 232 n. 2.
37. J. Takakusu. A Record o f the Buddhist Religion by I-tsing, translated,
p. xxxvii, cf. Ed. Huber in BEFEO VI, 1906, pp. I IT.; S. Levf in ‘Toung Pao’
Ser. 2, t. V. 1904, pp. 297 ff.
such as Kucha and Agnidesa in Central Asia, as recorded by

Hsuan-tsang.38 It is called the language of India. The Mahayana
school of Buddhism introduced into Khotan, Kashgar and
Kucha in the fourth century has also its literature in Sanskrit.
A Kharo?thi inscription39 discovered by Stein in southern
Sinkiang and datable to third/fourth century A.D. actually
refers to Mahayana. Thus, Sanskrit occupied a pre-eminent
position in respect of both the schools of Buddhism in
Central Asia. It is mentioned as arsa = arya in a Tokharian
manuscript.40 It was methodically taught in Central Asia in the
monastic schools of Kucha as also in other places. Some bilin
gual documents containing Sanskrit texts and their Kuchean or
Agnean or (Saka) Khotanese versions have also been found.
Finds of manuscripts of dictionaries, such as a Sanskrit-Tok-
harian vocabulary (c. A.D. 700) in the Kucha area and a
Sanskrit-Chinese lexicon undertaken by a Kuchean scholar in
the seventh century A.D. fully testify to the extensive and ela
borate study of Sanskrit in Central Asia.41 Its study by local
scholars enabled them to translate Sanskrit Buddhist texts into
regional languages. In this context reference might as well be
made to the discovery of several manuscripts of the Kdtantra-
vyakarana or Sanskrit grammar according to the school of
Katantra.
Canonical Literature
The Central Asian manuscript finds have clearly established
the use of Sanskrit language by the Buddhists, thus providing
authentic canonical literature. Pali is not the sole repository of
Buddhist thought, nor are the Theravadins its custodian. Even
the conservative school of the Hinayanists in Central Asia has
its sacred works in Sanskrit. A complete Tripitaka or the three
baskets is no doubt lost, but Central Asian fragments of the
38. Thomas Watbers : Op. cit, I, pp. 53, 60; Bagchi : Op. cit, p. 94.
39. Boyer, Rapson and Senarb : Kharo$thi Inscriptions discovered by
Aurel Stein—Op. cit (No. 390.
40. Mookerji ‘Central Asia’ in Cultural Heritage o f India, Vol. V, Op. cit,
p. 711; quoting BSOAS. VII. p. 895.
41. Bagchi : India and Central Asia—Op cit, pp. 106, 111. The Sino-
Sanskrit lexicons were apparently meant for the Chinese scholars or those
conversant with Chinese who were interested in learning Sanskrit.
190 Buddhism in Central A sia
canonical text belong to this lost Sanskrit Tripitaka—consisting

of the Sutrapitaka, Vinayapitaka and Abhidharmapitaka• The
first one has this difference with the corresponding Pali Nikayas
that it contains only four Agamas in contrast to five in Pali. A
study of these Agamas (A-han in Chinese) literature is possible
with the help of translations done by the Buddhist scholars of
Central Asian origin42 in Chinese or by the local scholars in
China. The four Agamas, mentioned earlier, were translated
into Chinese at different dates between the fourth and-sixth
centuries. Thus, the Dirghagama was translated by BuddhayaSas
in A.D. 412-13, the Madhyamagama by Saiighadeva earlier in
A.D. 397-98, Samyuktdgama by Gurtabhadra in A.D. 420-7,
and Ekottaragamaby Dharmanandi in A.D. 384-85. Incident
ally all the translators hailed from Kashmir which was a strong
centre of Buddhist learning43 and its scholars. The collections
of the Agamas seem to have completed in Kashmir before the
fourth century and these were deeply studied there before being
carried to Central Asia and China along with the Sarvastivada
school of Buddhism towards the end of the fourth century A.D.
The Central Asian manuscripts belong to the same period. The
Sutrapitaka—the Sanskrit Tripitaka canon of the Sarvastivada
school—being a collection of Agamas,44—has left fragments of
Sutras of at least three Agamas as following.
42. For the relation of the Chinese translations to the Pali Tripitaka
and to a Sanskrit canon now preserved only in a fragmentary state, see inter
alia Nanjio Catalogue—Op cit, pp. 127 ff. In class I—covering O-Iian-pu or
Agamclar—No. 542—noticing Madhyamagama-sutra and its translators—
Dharmanandi (A.D. 384-391) and Gautama Saiighadeva (A.D. 397-98)
No. 543. Ekottardgama-siitra by Dharmanandi (A.D. 384-85); No. 544—
Samyuktdgama-siitra by Ganabhadra (A.D. 420-479); No. 545—Dirghagama-
siitra by Buddhayasa together with Ku-Fo-nien A.D. 412-413.
43. The importance of Kashmir as a great centre of Buddhist learning is
an accepted fact. The Mulasarvastivadin Vinaya contained many legends
having reference to the conversion of Kashmir and North-Western India to
Buddhism, cf. Przyluski and S. Levi in JA.S.II t. IV. p. 493 flf; Winternitz :
Op. cit, p. 233 and n. 2; also Bagchi : Op. cit, p. 95. See also Bagclii. Le Canon
—Op. cit, p. 200.
44. The fragments of the Agamas are probably the oldest of the canonical
texts discovered in Central Asia. These were carried there by monks and traders
from Kashmir and Gandhara. The Agamas were Buddha ‘vacanas’ or
‘sayings’ of the Buddha. Their selection and division, as also modification
Dirghagama*45 (Ta-o-han in Chinese): The Sutras forming part

of this Agama and recovered from Central Asia include Samgiti
sutra,which enumerates different dharmas into ten classes accord
ing to the number of items constituting such dharmas- It is in
agreement more with the Samgiti-sutra of the Dirghagama in
Chinese translation than with the Samgitisuttanta in the Pali
Digha-nikaya. So also another Sutra of this Agama—the A\a-
ndtiya-sutra—is very much different from the Pali Atanafiyasut-
tanta, while a similar text under the title Mahasamaya-sutra is
noticed in Chinese Dirghagama. The Afandtiya-sutra might be
an elaboration of this text.
Madhyamagama in the Central Asian manuscript finds46include
jjpali-sutra—different from the corresponding Pali text, but in
agreement with the Chinese translation and Sukra-sutra—deal
ing with the doctrine of Karma retribution.47 This fragment
and enlargement of the Sutras followed later on. The compilation of the
Agamas might have been done in the time of Kaniska, a great patron of Bud
dhism. These differ from the Pali Nikayas of which they are not the exact
copies. While missing in India or in Nepal, a complete set of the Agamas is
available only in Chinese, being translations from the Central Asian Sanskrit
texts by the Buddhist scholars in China who had immigrated there from
Kashmir through Central Asia. A detailed account of the Agama translations
is given by Nanjio in his Catalogue (op. cit) pp. 127 ff; see also Bagchi : La
Canon Bouddhique en Chine (Op. cit) under individual translators. Kshanika
Saha notices Agama literature in China, Tibetan translation of the Agama
literature and early and later Buddhist texts in her work Buddhism in Central
Asia (Op. cit, pp. 351 f).
45. The Dirghagama SOtra of the Hinayana school included in Hoernle’s
Manuscr pt Remains (pp. 16 ff) are Sahgiti-sutra and Atanatiya-sutra—which
does not figure in the Chinese Dirghagama (Bagchi : Op. cit p. 95; K. Saha—
Op. cit, p. 47).
46. For a detailed reference to the contents of Madhyagama-Sutra in
its Chinese translation see Nanjio— Op. cit, No. 542, pp. 127 ff; No. 556,
p. 140—by Than-Kwo (Dharmapala) together with Khan-Man-Sian, A.D.
207. This is said to be an extract from a full text of the Dirghagama (No.
545).
47. See Nanjio : Op. cit No. 544, p. 135. It was translated by Gunabhadra
of the earlier Sun dynasty. Two other translations are noticed in the same
Catalogue (Nos. 546 and 547). Both are extracts from a full text as that of
No. 544. The names of the translators are lost, though one was done under
the three Tshin dynasties, A.D. 350-431 (No. 546), and the other ones earlier
under the Wei and Wu dynasties (A.D. 220-280). See also Bagchi : Op. cit,
pp.95-96; Saha:Op. cit, pp. 36-37. In the Sarhyuktagama the division into
agrees with the corresponding text in Chinese Madhyamagama,

but differs from the Pali one—more elaborate and with a different
title—Chulla-Kammavibhahga. The Sanskrit text was popular in
Central Asia with its translation in Kuchean language and so
was the case in China with its four separate translations.
The Samyuktdgama has a quite good many sutras left in
Central Asian manuscript finds. These include Pravarana-sutra
agreeing with the Chinese translation of Gunavarman. Two
other translations of the text were made in Chinese by Dharma-
k§ema (266-317 A.D.) and Fa-hien (Dharmabhadra) in the 10th
century. The text is in general agreement with the Pali Vangisa-
therasamyutta of the Samyutta-nikaya. The next sutra—Candro-
pama too, is in general agreement with Pali Samyutta (third
sutta of the Kassapa Section) but agrees more closely with the
Chinese translation. Sakti-sutra and Nidana-sutra are also in
agreement with their Pali counterparts—Pali Satti-sutta of the
Opamma section of Samyuttanikaya, and Nidana-samyutta of the
Samyutta-mk&ya. (XIII. 15) respectively. Both are also translated
into Chinese, while there is also an independent translation of
the Nidana-sutra which deals with the chain of causation and
with the means of its destruction. The other sutras of this
Samyuktdgama include Kokanada-sutra, Anathapindaka-sutra,
Dlrghakanfha-su tra, Sarabha-sutra, Parivraydka-sthavira-siitra
and Brahmana-Satyamsutra. These correspond to their Pali
counterparts and are as well translated into Chinese.
Ekottaragama and its fragments are no doubt missing in the
Central Asian literary finds, but the Chinese Ekottaragama48 was
first translated by Dharmanandin of the Tukhara region. He
went to Chan-an in A.D. 384 and completed the translation of48
‘vargas’ and ‘samyuktas’ is totally different and there are Samyuktas in the
‘Pali which are missing in the Chinese version and vice-versa (Winternitz:
Op. cit, p. 235 and n). The Sanskrit fragments of the Samyuktagama found
near Turfan contain parts of six sQtras arranged in the same order as in the
Chinese translation and are apparently the original from which it was made.
(Eliot : Op.c it, p. 297).
48. Nanjio : Op. cit, No. 543 ff. It was translated by Dharmanandi A.D.
384; but it was lost already in A.D. 730, while a later translation in existence
is said to have been made by Gautama Prajnaruci A.D. 397. See Bagchi : Op.
cit, pp. 93, 94, 95; Saha : Op. cit, pp. 35, 38,41, 42.
this Agama with the assistance of two Chinese monks. The trans
lation was revised by Sanghadeva who hailed from Kubha. In
the Sanskrit of canon an Ekottaragama corresponds to the
Anguttara-nikaya. Literally meaning ‘yet one more’, it is
synonymous with Anguttara. The absence of finds of fragments
of this Agama in Central Asia need not preclude us from pre
suming its literary existence in that region when there are positive
evidences of its translation in Chinese, where it is divided into 51
sections containing 454 sutras as against the Anguttaranikaya in
Pali with eleven nipatas i-e. sections containing 169 Sutras.
The canonical rules embodied in the Vinaya-pitaka of the
Sarvastivada school seem to have been scrupulously followed.
Fragments of texts of this work have been recovered from the
northern parts of Eastern Turkestan. Written in Sanskrit, a
complete text of the Pratimoksa-sutra49 belonging to Sarvastivada
Vinaya was discovered by the French Mission in the ruins of
Duldur-Aqur at Kucha. Other texts of the Vinayapitaka of the
Sarvastivadins have also been found in Central Asia, and a few
in Nepal too. It is possible to reconstruct the Pratimoksa-Sutra,
from Chinese and Tibetan translations.4950 Both the Sanskrit
canons of the Sarvastivadins and the Vinayas of the Mahisasa-
kas, Dharmaguptas and Mahasanghikas, reveal manifold
differences in separate details from the Pali canon and from one
another, though the original stock of rules is one and the same.
49. Bagchi : Op. cit, p. 96; See Nanjio : Op. cit Nos. 1110 (p. 245) and
1160 (p. 255); Eliot p. 214; 323. The ten primary commandments are called
Pratimoksa and he who breaks them is Parajika, who automatically leaves the
road leading to Buddhahood and is condemned to a long series of inferior
births. For Central Asian finds of Pratimok$a fragments and their translation
in European languages—see reference No. 3 (Op. cit) of this chapter with
fuller information as recorded by Winternitz. S. Levi has edited the Kuchean
fragments of Pratimoksa Prayakcittika and Pratidesaniya—with a French
translation in Hoernle’s Manuscript Remains, pp. 357 ff.
50. The Pratimoksa-vinaya (or Stitra ?) was translated by Gautama
Prajfiaruli A D . 543 of the Eastern Wei dynasty, A.D. 534-550. (Nanjio 1108,
p. 245). The MOla-Sarvastivadin (nikaya) Vinaya (or Pratimoksa). Stitra was
translated by I-tsing A.D. 710. It agrees with the Tibetan ‘K-Yuen-lu’ (Nanjio.
No. 1110, p. 245). There was also an earlier translation by Kumarajiva c.
A.D. 404. (Nanjio 1160 p.255). In Tibetan there is a literal translation of the
whole of the Vinayapitaka of the MQla Sarvastivadins (See Banerji ; Sarvasti
vada Literature—quoted by K. Saha : Op. cit p. 65).
The Mula-Sarvastivadin Vinaya contained many legends referring

to the conversion of Kashmir and North-western India to Bud
dhism. The Vinayapitaka also includes the siksas and sahgha-
karmas corresponding to the Pali sukkhas and kapimavacas, and
fragments of these have been found in Central Asia. Reverting
back to the text of the Pratimok$asutray one could trace agree
ments with its Pali counterpart. The text contains all the sections
such as Nidana, Parajika, SanghadiSesa etc. and agrees literally
with its Chinese translation by Kumarajlva in A.D. 404. In this
context reference might as well be made to the other canonical
work of the Sarvastivadins—the Bhikfunipratimokfa51 of which
fragments were discovered in the Kucha region by both the
French and the German missions.
Another important text entitled Mahaparinirvanasutra, belong
ing to theMula-Sarvastivada Vinaya and written in Sanskrit, has
been restored from a number of fragments discovered from the
ruins of Sorcuq near Turfan, and of Tumshuq near Maralbashi
by the German Mission. Parallel texts were also discovered
from the ruins of Kizil near Kucha. The Mahaparinirvanasutra
is an important piece of Pali Suttapitaka> but here it figures in
the Vinayapitaka and agrees with the Mahavagga of the Pali
literature. The Chinese Tripitaka contains ten different transla
tions of the Mahaparinirvanasutra. Three of these belong to the
Hinayana52 and seven to the Mahayana with the external form
51. Bagchi : Op. cit, p. 96; Saha : Op. cit 65. The Bhikfuni-prdtimokfa-
siitra forms one of the six parts of the Tibetan ‘Dulva’. This fourth part of
Bhikfurit-Prdtimokfa-sutra agrees substantially with Pali Bhikhuni Palimokkha
SQtta with which there is close agreement of the Central Asian Bhikfuni-
prPtimok$a-sutra, edited by Louis de la Vailed Poussin and Waldschinidt.
(Saha : p. 66 quoting Banerji). For the Chinese translation of this work see
Nanjio : Op. cit No. 1149 (p. 53); and No. 1160 (p. 255).
52. Bagchi : Op. cit, pp. 96-97. The Mahaparinirvanasutra corresponds
to Mahaparinibbdnasutta or ‘Great lecture on the Complete Nirvana’, a
continuous account of the last days of Buddha. It supplies the earliest begin
nings of a biography of the Tathagata. The Chinese translations contain
speeches which the Buddha is supposed to have uttered prior to his death
(parinirvdna)—Wintemitz : Op. cit 236 and n. For the Chinese translations
of the *Mahaparinirvana of the Hinayana’, see Nanjio No. 118fp. 40 translated
by Fa-hian of the Eastern Tsin dynasty A.D. 317-420; and No. 119 (ibid)
also translated by the same author in the same period. The third one of this
as the only common factor and contains the speeches which the
Buddha is supposed to have uttered prior to his death. On a
comparison of the Pali, Tibetan and Chinese versions with the
Central Asian one, it is suggested that it was the choice of the
Buddhist sects to place it either in the Sutrapitaka or in the
Vinayapitaka. Mula-Sarvastivadins and the Chinese Buddhist
scholars included the whole text in the Vinayapitaka including
the account of the two councils originally forming the conclud
ing part of this Sutra. «
The Central Asian finds also include Abhidharmapitaka53 frag
ments. This pitaka consists of seven texts which are preserved in
Chinese translations. A fragment of the original Sankrit text of
the Sahgltiparyaya was found in the Turfan oases area, and
some were also discovered in the caves of Bamiyan in Afghani
stan. This text corresponding to Safigiti-sutta appears among the
Sarvastivadins as a book of the Abhidharmapitaka. The script
of this fragment is slanting Gupta which was in use in the nor
thern part of Eastern Turkestan.
Fragments of a few sutras belonging to the Sutrapitaka of
Hinayana, but not strictly included in any of the Agamas are
also identified.5354 These include Da§abalasutra, Mahavadanasutra
and Saptabuddhaka etc. The Dasabalasutra agrees partially with
the Pali Dasakanipata of the Anguttara-nikaya, and the Dasaka
section of Chinese translation of Ekottaragama. It was a very
popular text of the Sarvastivadins. A separate translation of
this text was made at Kucha towards the end of the eighth cen
tury A.D. by the Chinese envoy Wu-kong in collaboration with
a Kuchean monk named W-ti-ti-si-yu.
school No. 552 (Nanjio: Op. cit, p.139) was translated by Po-Fa-tsu, A .D .
290-306, of the Western-Tsin dynasty A.D. 265-316. This is an earlier trans
lation and it agrees with the Tibetan, K’yuen-lu.
53. Bagchi : Op. cit, p. 97. A text corresponding to the Sangitisutta
entitled Sangitiparyaya appears among the Sarvastivadins as a book of the
Abhidharmapitaka of J. Takakusu. Pali Text Society (PTS) 1905— p. 99ff;
Winternitz : Op. cit, p. 44, n. 2. See also Nanjio : Op. cit No. 1276 p. 281.
The work entitled ‘Abhidharma-sangiti-parydyapdda' (Sastra) was composed
by the venerable Sariputra. According to Yasomitra’s Abhidharma-kosa-
vydkhya, its author is Mahakausthila. This is the first of the six Pada works
of the Sarvastivadins.
Another important canonical text is the Udanavarga?5—a com

plete text in Sanskrit found by Pelliot and Grunwedel from Eas
tern Turkestan. Stein could get only its fragments. The text is a
collection similar to the Prakrit Dhammapada found in the
Khotan area.56 The title literally means a collection of Udanas
or the utterances of the Buddha. It was earlier known through
its translation into Tibetan and Chinese. It is a work of the
Sarvastivada school of Buddhism. It is divided into 33 chapters
of varying lengths, each bearing a separate title as we find in the
Dhammapada—beginning with ‘Anityavarga’ or the chapter on
Impermanency, and ending with ‘Brahmana-varga’. The work
contains about 1000 verses. Its author is Dharmatrata, said to
be the maternal uncle of Vasumitra. The popularity of this work
is evident from its translation in the past into several languages.
It was first known through its Tibetan translation done by an
Indian named Vidyakaraprabha or Vidyaprabhakara in colla
boration with a Tibetan scholar named Lotsava Rin-chen-mchog.
Apart from the Chinese and Tibetan translations, the work was
also translated into other Central Asian languages. The frag
ments of this work found by Stein are in the Central Asian form
of Brahmi as was prevalent in Kucha about the seventh century
A.D.
55. Fragments of a Sanskrit anthology have been found in Central
Asia which were at first regarded as passages from recensions of the Dhamma
pada, but which in reality belong to the Udanavarga, known to us from a
Tibetan translation. It was compiled by Dharmatrata, who, according to
Taranatha, is supposed to have lived in Kanaka's days (Wintemitz. Op. cit
p. 237). These fragments of the Udanavarga from finds in Central Asia were
published by R. Pischel, ‘Die Turfan-Rezensinen des Dhammapada’ SBA.
1908 p. 968 fT.; Sylvain Levi and Louis de la Vallee Poussin in JAS. t. 10, t.
XVI. 1910 p. 444 ff; t. XVII, 1911, 431 ff; t. XIX, 1912, p. 311 ff; JRAS.
1911, p. 758 ff; 1912, p. 355 ff. The Tibetan version of the Udanavarga was
translated by W.W. Rockhill, London, 1883 and the Tibetan text was published
by H. Beckh, Berlin 1911.
56. Ancient Khotan, Oxford, 1907, I, p. 88; S. Konow (Fest-schrift Win-
disch, p. 85 ff) suggests that though this anthology was composed in a dialect
originating in the north-west of India, it was written in the neighbourhood
of Khotan. Fragments of this anthology have come down to us in a few leaves
of a manuscript written in the Kharo§{hi script which M. Petroffsky and
J.L. Dutreuil de Rhins had found at Khotan (for reference see Winternitz.
Op. cit, p. 238 n). See also Ghoshal. Op. cit, pp. 260 ff in which N.P. Chakra-
varti traces the history of the compilation of the Sanskrit Udanavarga.
The finds of Mahayana Sutras and fragments as also pothis

represent a new stratum in the history of Sanskrit Buddhist
literature in Central Asia. The Mahayana school is not supposed
to possess a canon of its own, since it does not represent one
unified sect. It is doubtful if the Buddhist council held in the
time of Kaniska could establish any canon, and even if so, in
which language and by which sect.67 A Chinese text translated
by Hsuan-tsang mentions a Bodhisattvapitaka consisting of a
long list of Mahayana texts, a Vinayapitaka and an Abhi-
dharmapitaka, and the same text also enumerates a lengthy list
of Mahayanasutras.575859 The so-called nine Dharmas or texts in
Nepal are not the canon of any sect, but a series of books
which were compiled at different times and belonged to different
sects. The title of these nine books are Asfasdhasrika-Prajnd-
Paramita, Saddharma-Pundarika, Suvarna-Prabhasa, Gan$a-
vyuha, Tathagataguhyaka or Tathagatagunajnana, Samddhirdja
and Dasabhiimlsvara. All these works are also called ‘Vaipulya-
Sutras’. Saddharmapun$arika59 or the ‘Lotus of the true Law’
is the most important as well as the most popular Mahayanist
work, highly respected in China and Japan. According to
Nanjio, there were eight or nine translations of this text into
Chinese, of which only three are available. The earliest (A.D.
286) is that of Dharmaraksa, a Yueh-chi scholar from Kan-su.
57. Winternitz. Op. cit, pp. 294 ff. According to Takakusu (JRAS.
1905 p. 414 f), the Council dealt with Hinayana and not the Mahayana. It
is proposed that the object of the alleged Council of Kani$ka was not to es
tablish a canon, but to collect explanations (Commentaries). (Winternitz.
294 n. 2).
58. See Levi and Chavannes in JA. 1916.S.11 1. VIII. p. 5 ff. The Buddhist
dictionary Maha-vyatapatti (Bibl. Buddh. XIII) p. 65 mentions 105 separate
Mah&yana texts, No. 12 of which is a ‘Bodhisattva-Pitaka’. It is also quoted
in the tfikfa-samuccaya pp. 190 and 311.
59. The Sanskrit text has been edited by Kern and Nanjio in Bibliotheca
Buddhica, X, translated by Burnouf (Le Lotus de la bonne Loi) 1852, and by
Kern (Saddharmapundarika) in the Sacred Book of the East, Vol. XXI, Oxford.
There are several translations in Chinese. Nanjio. Op. cit. No. 134 p. 44;
Translated by Kumarajiva A.D. 384-417; No. 135 p. 44 (translator’s name
lost); No. 138 p. 45 translated by Ku-Fa-hu (Dharmarak$a) of the western
Tsin dynasty, A.D. 265-316; No. 139 p. 45 translated by Jnangupta and
Dharmagupta A.D. 601 of the Sui dynasty A.D. 589-618, with an introductory
preface.
The next (400-2) is that of Kumarajiva, the famous Buddhist
scholar monk of Kucha, who was taken as a prisoner to China,
where he translated several Buddhist works into Chinese. The
third translation (A.D. 601) is of JnanaguptaandDharmagupta—
the two Indian Buddhist scholars who followed the Nepalese
manuscripts. It was also translated into Tibetan. A number of
fragments of this work were found at different sites in Central
Asia.6061In Hoemle’s ‘Manuscript Remains o f Buddhist Literature’
there are fragments of the manuscripts of this text found at
Khadlik and edited by F. W. Thomas (one) and Luders (two).
It purports to be a discourse delivered by Sakyamuni at the
Vulture Peak (Griddhakuta) to an assemblage of Bodhisattvas.
The ‘Lotus’ clearly affirms the multiplicity of vehicles, or various
ways of teaching the law, and also the eternity of the Buddha.
The fragments of the Saddharmapundarika found in Eastern
Turkestan and presenting a text diverging from that of the
Nepalese manuscripts, suggest that there were two recensions of
the work.
The next Mahayanist work in importance and in popularity is
the Prajhaparamita61 or ‘transcendental knowledge’. This gene
ric name is given to a whole literature consisting of treatises on
the doctrine of Sunyata, which vary greatly in length. They are
60. See Stein. Inside Asia p. 1018; No. 0153 from Domoko near Khotan;
Fragments from Khadlik {Serindia I.p. 163) from Shrine F.12 in the Khotan
region (Serindia p. 1254) and from the ruined shrines at Khadlik (Serindia
pp. 1433 ff). Several manuscripts and fragments of this works are noticed
in Appendix F. by A.F. Rudolf Hoernle, containing the Inventory List of
Manuscripts in Sanskrit, Khotanese and Kuchean.
61. For the finds of Manuscript fragments of this Buddhist Mahayanist
text, see Stein : Serindia, 814, 914, 1432-50; Khotanese 1454; Tibetan 1470
Sq. & Chinese 163 Sq, 687, 914, 925; Prajna-pdramitd means both the per
fection of ‘wisdom’ and the writings treating it. ‘Prajfia’ not only means know
ledge of the absolute truth, that is to say of SOnyata or the Void, but is regarded
as an ontological principle synonymous with Bodhi and Dharmakaya. The
Buddhas not only possess this knowledge in the ordinary sense but they are
the knowledge manifest in human form, and ‘Prajfia’ is often personified as a
goddess (Eliot. Op. cit. III. p. 52). According to Aneski, the innermost qualities
of Buddhahood can be sought nowhere else than in the profound abyss of
the prajfia. (ERE. IV. p. 837). The six paramitas include behaviour (sila),
contemplation (samadhi), wisdom (prajha), deliverance (vimukti) and the real
ization of the knowledge leading to it ( Vimukti-jnana-darsana).
classed as Sutras, being described as discourses delivered by the

Buddha on the vulture peak (Griddhakuta) at Rajagriha. The
Prajnaparamitas belong to the earliest Mahayana-sutras regarded
with greatest reverance and equally of the greatest importance
from the point of view of the history of religion. Being in the
nature of philosophical treatises, they talk of six perfections
(Paramitas) of a Bodhisattva, finally aiming at the highest one
called ‘wisdom’. This wisdom consists of the knowledge of
‘Sunyata’, ‘emptiness’ i.e. the unsubstantiality of phenomena—
all objects being endowed with a conditional or relative existence-
Since a Prajnaparamita was already translated into Chinese62 as
early as A. D. 179. This class of Mahayanist literature seems to
be the earliest one. Presented in a dialogue form, similar to
those of the Pali Suttas, the Buddha called Bhagavan, ‘theLord’,
generally appears in conversation with one of his disciples, espe
cially Subhuti. In other Mahayana-sutras Buddha usually talks
to a Bodhisattva.
A considerable number of Prajna-Pararaita of all texts were
already in existence in India and their number increased even
more in China and Tibet. Hsuan-Tsang translated 12 different
Prajnd-Pdramita-sutras in his magnum opus, Mahd-prajnd-para-
mita-sutra,63the longest is that of 10,000 slokas or verses and
the shortest that of 150. The following have come down to us in
Sanskrit : Satasahasrika-prajna-paramitas of 1,00,000 slokas,
62. Amongst the Sutras of the Mahayana, the Paramita class called Pan-
zo-pu, is the foremost. The earliest translation of the ‘Dasasahasrikd-prajna-
paramita' was done by Leu-Ki-Khan (Lokarak$a?) of the Eastern Han dynasty,
(Nanjio No. 5 p. 4) followed by several others including Kumarjiva (ibid)
No. 6 p.5) in A.D. 408, Dharmapriya together with Ku-Fo-Knien and others,
A.D. 382 (ibid No. 7 p. 5) and then repeatedly till the 10th century (Nanjio
Nos. 927, 935, 988, 1033, 1034).
63. See Nanjio. ii. No. 133 pp. 435-37 noticing the Chinese translations
of Buddhist Canonical works by Hsuan-Tsang who received his ordination at
Chan-tu in A.D. 622 and died in A.D. 664 in his sixty-fifth year. According
to Nanjio, there are 75 works of this Chinese Sramana of which he gives the
list that of Mahaprajnapdramitasutra (No. 1). According to Eliot, there are
ten collections of Prajnaparamita, besides excerpts sometimes described as
substantive works. See Wallesser. Prajnaparamita in Quellen der Religion
geschite, pp. 15 ff. SBE. XLIX. Nanjio Catalogue Nos. 1 -20 and Rajendra Lai
Mitra’s Nepalese Buddhist Literature, pp. 177 ff. See also Winternitz. Op. cit,
p. 314 and 315 and n. 1.
Pahcavimsatisahasrikd—of 25,000, Asfasahasrika of 8,000; Sar-

dhadvisdhasrika of 2,500, Saptasatikd of 700, Vajracchedika
Paramita, ‘the diamond cutter Prajna-paramita cutting as sharp
as a diamond; the Alpasdra-prajha-pdramitd and Prajha-pararhi-
tahrdaya Sutras are only used as protecting magic formulae
(Dharanis).
In Central Asia a few fragments of the Prajhaparamitas have
been found. Those of the Atfadasasahassrika prajna-paramita6465
comprise of nine folios found at Khadlik. The leaves belong to
four different manuscripts of the seventh-eighth centuries A.D.
Another fragment of a Prajnaparamita manuscript consisting of
48 leaves from Central Asia is in upright Gupta characters and
is supposed to be earlier than the sixth century A.D. Its langu
age is Sanskrit. Despite the numerous epistemological and
metaphysical discussions on Sunyata, its religious character is
also emphasised in several chapters. Great merit is supposed to
be acquired by the hearing and understanding of it, and the
learning and teaching of the Prajna-paramita is praised repeated
ly and extraordinarily.
The manuscript of Vajracchedikd-prajhd-paramita65discovered
by Stein is written on nineteen folios. In this work, too, we come
across the same paradoxes as in the Astasdhasrikaprajha-para-
mita and the longer Prajha-paramitds. The work seems to be an
64. See Central Asian fragments of ‘AâdaSa-sahasrika’ and of an
unidentified text edited by Sten Konow in Memoirs of the Archaeological
Survey of India No. 69. Calcutta. 1942. cf. the Sanskrit edition of the work
edited by Rajendra Lai Mitra—Bibliotheca Indica, Calcutta* 1888. According
to Winternitz, the *A}(a-sdhasrika-prajn<i-paramita is the earliest of these
writings which was, on the one hand, expanded into the larger works, and
the contents of which were, on the other hand, condensed in the shorter texts.
(Op. cit 316 and n 6 for the references).
65. Ancient Khotan—Op. cit, Vol. I p. 258; Hoernle. Op. cit p. 176 for
the Chinese translation see Nanjio Op. cit No. 10 p. 5 translated by Kumara-
jiva; Bodhiruci (A.D. 508-535 in China) No. 11 p. 6; Hsuan-Chuang No. 13
p. 6; I-tsing No. 14 p. 6; and Dharmagupta No. 15 p. 5; cf. also Vajracchedika-
sutra Sastra, a commentary on Nos. 10-15 compiled by the Bodhisattva
Asahga and translated by Dharmagupta A.D. 590-616 of the Sui dynasty
(No. 1167 p.257), and a commentary on No. 1167 under the same title compiled
by the Bodhisattva Vasubandhu, translated by Bodhiruci, A.D. 509 (ibid No.
1168 p. 257). The Khotanese texts from Central Asia include Vajracchedika
(Hoernle. Op. cit p. 214).
abridgement of the exposition of the doctrine of Sunyata pro

pounded in the ‘Hundred-thousand prajna-paramita*. It has a
commentary translated into Chinese by Dharmagupta (A.D.
590-616) and Vasubandhu, brother of Asanga, a convert from
Hlnayana to Mahayanism is said to have written a commentary
on this canonical work, as well as on Saddharmapuniarika and
Mahaparinirvana-sutra as a penance for his previous calumniation
of the Mahayana school of Buddhism. According to chapter 13
of the Vajracchedika, Buddha is made to say words to the effect
that one who selects merely one four-footed verse here from
this sermon and explains ‘it to others’ acquires greater religious
merit than a person who might sacrifice his life day by day for
aeons.66
Suvarnaprabhasa-sutra—Literally meaning the ‘Splendour of
Gold’, this later Mahayana Sutra contains partly philosophical
and ethical matter and many legends. While there are many
beautiful passages such as the confession of sins and the praise
of love (maitri) towards all beings, the doctrine of Sunyata is
developed in a full chapter (VII). Its major portion consists of
a glorification of the Suvarnaprabhdsa itself. The work enjoys
great reputation and popularity in all countries professing
Mahayana Buddhism. In Central Asia, too, fragments of this
work were discovered.67 Its Chinese translators were Dharma-
ksema (413-33), Paramartha and his pupils (552-557) and I-
tsing (703). There are also fragments of Uiguric texts68—
66. See Walleser. Prajha-paramita p. 146 f, 149 IT. In a fragment of the
Adhyardhakatika-prajhdpdramita from Khotan (Leumann ‘Zur nordarischen
Sprache und Litteratur’ p. 89) it is even said that he who ‘even while still in
the womb’ hears this section of the Prajha-paramita, is freed from all attacks
and dangers, and never goes to hell or is reborn in any other evil form of
existence (Winternitz, p. 320 n).
67. Hoernle. Op. cit p. 108 edited by Sarat Chandra Das and Sarat
Chandra Sastri, Calcutta, 1898 (Buddhist Text Society of India). See also
Rajcndra Lai Mitra—Nepal Buddhist Literature pp. 241 ff; cf. Burnouf.
Introduction pp. 471 ff, 490; Bendall. Catalogue p. 12 f; M. Anesaki in ERE p.
839; Winternitz. Op. cit p. 339 n; for the Chinese translations of this Buddhist
text, see Nanjio Op. cit No. 126 p. 41 by I-tsing which agrees with the Tibetan;
No. 127 p. 42 by Dharmarak§a which is an earlier and incomplete translation
of No. 126, though the most popular one in China.
68. An Uiguric translation appeared in Bibliotheca Buddhica in 1914.
For fragments of Uiguric texts see F.W.K. Muller. ‘Uigurica’ in ABA. 1908,
Ratnadhvaja-sutra,*69, Ratnarasi-sutra,70 Bhadrapala-sutra7172.

The fragments of these Sutras are all in upright Gupta charac
ters. The text of the first one is identified by Watanabe with
Ratnadhvaja of the Mahasannipata Sutra. It was translated into
Chinese by Dharmaraksa, a native of Central India between
A.D. 414-21 under the Northern-Lian dynasty. The next one—
Ratnarasi-sutra is also in upright Gupta characters. Its Tibetan
version is to be found in the Bkah-hgyur. The Sutra was trans
lated into Chinese in A.D. 397-439. Passages from the Sutra,
outside this fragment are cited in the Siksasamuccaya of Santi-
deva. The Bhadrapala-sutra also in upright Gupta characters
has been identified by Watanabe. Jnangupta translated it into
Chinese. Besides these there is a fragment of Candragarbhd72 of
the Mahasannipata class.
Fragments of Dharanis : Dharanis or ‘protective spells’ con
stitute a large and important part of Mahayanist literature.
According to the Saddharmapun<j.arika,73the dharanis are taught
for the protection, safety and shelter of the preachers. The
p. 10 ff; Luders. SBA. 1914 p. 99; Winternitz p. 341 n. This work was trans
lated from the Indian language into Chinese, and from the Chinese into
Turkish (Winternitz. Op. cit p. 341 n).
69. Hoernle. Manuscript Remains—Op. cit p. 100; Stein. Serindia, III,
p. 1432. The text has been identified by Watanabe as from the second chapter
of the second part of Ratnadhvaja of the Mahasannipata-sutra (Nanjio No.
84; ZDMG. LXII p. 100). It was translated into Chinese by Dharmaraksa
(Nanjio App. II. No. 67).
70. Hoernle. Op. cit p. 116; Serindia. III. p. 1439. The Sutra was trans
lated into Chinese in A.D. 397-439—Nanjio No. 23 (44) p. 19.
71. Hoernle. Op. cit p. 88. See Nanjio. Op. cit No. 75 p. 31. It was trans
lated by Jnangupta, Dharmagupta and others of the Sui dynasty, A.D. 589-
618.
72. Hoernle. Op. cit p. 103. According to Bagchi, the Suryagarbha and
the Candragarbha are two important SQtras of the ‘Mahasannipata’ collection.
They are preserved in Chinese translation of about the middle of the sixth
century by Na'rendra Ya§as (op. cit p. 109).
73. Kern. Op. cit p. 99. See Burnouf. Introduction pp. 466, 482 ff;
Wassiljew. Der Buddhismus pp. 153 ff; 193 ff, La Valee Poussin. Buddhisme
Belgium 1898 p. 119 ff; J. Haver. Die Dharani im nordlichen Buddhismus
Tubingen 1927; Tuchi. IHQ. IV. 1928 pp. 553 ff; Winternitz pp. 380 ff and
notes for references. See also Waddel ‘The Dharani Cult’ in ‘Ostasiat Zisft’
1912 pp. 155 ff. Formulae or spells called ‘Dharani’ are said to possess a
mysterious efficiency and potency.
protective and salutary magical power of dharani is primarily

due to some piece of wisdom contained in it, and not to any
occult mystical significances of the words and syllables, though
these Dharanis include magic words {mantrapaddni) of this kind/
The Prajna-Paramita-Sutras in their shortest form were used
as Dharanis. In this context special mention might be made of
the Prajhd-Paramita-hjrdaya Sutras.™ These sutras teach the
heart {hpdaya) of the Prajna-Paramita i.e. ‘the Mantra which
alleviates all pain’, contain the perfection of wisdom. Many
Prajfia-Paramita texts appear in the Tibetan Kanjur. There are
Mahayana Sutras which are only recommendations of Dharanis.
Thus, in the Aparamitayuh-sutra15 in Sanskrit and old KLhotanese,
as also in Chinese and Tibetan translations there is nothing more
than the glorification of a Dharani. The growth of the Dharani
literature might be placed between the fourth and eighth century
A.D. In the early Mahayana text Saddharmapundarika there are
a few Dharanis for recitation ensuring protection against evils.
In the twenty-first chapter of this work, Bhaisyaraja Bodhisattva
approaches the teacher for a protective spell needed for the
Dharmabhasakas engaged in handing down the texts of this
canonical work. In the Suvarnaprabhasa sutra (chapters XI, XII)
there is a section exhorting the gods and demons to protect
those reading and writing the sutra from any harm.
Among the Dharani fragments found in Central Asia, mention
maybe made of the Anantamukhadharani76and Nilakanthadharani11
74. Edited by F. Max Muller and B. Nanjio in ‘Anecdota Oxoniensia’
Aryan Series, Vol. I, Part III, 1884, and translated by Max Muller in SBE
Vol. 49. Part II, p. 145 fF; translated from the Tibetan by L. Feer in ‘Annals
du Musie Guimet’ (Paris) V, 176 ff; Winternitz. Op. cit 316 n. 5.
75. Serindia p. 914; Hoernle. Op. cit pp. 289 ff. It is a Dharani’ which
has long been known to exist in Sanskrit manuscripts and in Tibetan. A
complete manuscript of this version was found by Stein in the cave temples
at the Halls of the Thousand Buddhas which also yielded the manuscript of
the ‘ Vajracchedika. Sten Konow provides the Sanskrit text and Tibetan version
as also the Khotanese one.
76. Hoernle. Op. cit p. 86. See Nanjio. Op. cit Nos. 353 -360 pp. 89-90.
The earliest one (No' 353) is by C’Chien of the Wu-dynasty, A.D. 222-80.
77. The fragment of Nilakanthadharani’ brought from Central Asia by
Aurel Stein, with the Sanskrit text in Brahmi script and in Sogdian transcription
has been edited by La Vallee Poussin and R. Gauthiot in JRAS 1912 pp. 629 ff.
(Saha. Op. cit p. 96). See Nanjio. Op. cit Nos. 318,319. pp. 81-82.
and Sitatapatrd Mahapratyahgiradhdranl78. The first • text

has been identified be Watanabe as part of the Anantamukha-
dharani,o\' which the Chinese Tripitaka includes eight translations,
theearliest being by C’chien, whose date is A.D. 222-80. Thefrag-
nient of Nilakanthadharani, brought from Central Asia by Stein,
is in Sanskrit written in Brahmi script and in Sogdian transcrip
tion. This Dharani was popular in China between A.D. 650-
750. The fragment of Mahdpratyahgiradharani contains only a
small portion of it and also a series of epithets of the goddess
Tara. It is written in upright Gupta characters similar to those
of the Saddharma-punârika manuscripts. The whole of the
Sanskrit text of this Dharani was transcribed in Chinese charac
ters by the famous mystic teacher Amoghavajra (A.D. 704-774).
Another manuscript of this Dharani in the peculiar corrupt
Sanskrit current in certain parts of Eastern Turkestan, was
recovered from the Temple Library near Tun-huang. There are
also Nepalese manuscripts of this Dharani. Fragments of another
Dharan\—Suramgama-samadhi—are also noticed in Hoernle’s
manuscripts.7980 In fact, the text of this manuscript is the
conclusion of a Surdmgama-samddhi-sutra .followed by a
Dharani.
Non-Canonical Texts : Besides religious texts in Sanskrit
Central Asian finds include non-canonical ones as well. These
include poetical works of the two great Buddhist poets—
Asvaghosa and Matriceta. The German mission brought
fragments of the works of this contemporary of Kaniska.
ASvaghosa was a poet, philosopher and dramatist of
excellence. His two long Kavyas—Buddhacarita80 and Saundara-
78. Aurel Stein. Serindia p. 918; Hoernle. Op. cit pp. 52 ff. These are the
Nepalese manuscripts Nos. 61 and 77 of the Asiatic Society, Bengal Collection,
as also the Cambridge Collection (Cat. pp. 63, 68). There is also the Roll.
Ch. 0041 from the Temple Library near Tun-huang in the Stein Collection
which is noticed in the Journal of the Royal Asiatic Society 1911 pp. 460 ff.,
containing this Dharani in the peculiar corrupt Sanskrit. There are further
two Chinese translations, one of which is noticed in Nanjio (No. 1016 p. 223).
There is reference to an earlier translation by Amoghavajra A.D. 746-771.
79. Hoernle. Op. cit p. 125. It was translated into Chinese in A.D.
384-417. Nanjio No. 399 p. 98. Hoernle also quotes the Tibetan version.
80. A fragment of the ‘Buddhacarita’ brought from Turfan region shows
that the text was studied by the Buddhist monks of Central Asia (Bagchi.
nandail are well-known in original. A third one Sdriputrapra-

karana*818283, the earliest specimen of Sanskrit drama was discover
ed in the Turfan region. Only portions of the original work
have been found. As the oldest Indian drama known to
us it presupposes a great development of Indian dramatic art.
This work is in Sanskrit, but the dialogues are in Prakrit—older
than the ones used in classical Sanskrit dramas. ASvaghosa is
credited with the authorshipof another work, the Sutralankara6*
preserved in Chinese translation of Kumarajiva, as also a philo
sophical work entitled Sraddhotapada-Sastra which, too, has been
preserved in Chinese translation.84 The German mission discove
red another poetical work in Sanskrit, the Sutralankara of
Op. cit p. 100). See Serindia I p. 163; III, 1437. See Winternitz. Op. cit pp.
258-265. This is the Fo-Sho-hing-tsan-king, translated from Sanskrit into
Chinese between A.D. 414 and 421 by Dharmarak$a and translated into
English by Cowell in SBE Vol. 49. There is also a translation by E.H. Johnston.
Oxford. 1936; see also M. Aneski in ERE. II. 1909 p. 159 ff.
81. N o fragment of this work is found in Central Asia nor is it translated
into Chinese. It was discovered and edited by Haraprasad Shastri Bibl. Ind.
Calcutta 1910; critically edited and translated into English by E.H. Johnston,
Oxford 1928, 1932; which gives full bibliography.
82. Ref. N o.8. The discovery of fragments of manuscripts on palm-leaf,
of great antiquity at Turfan, has revealed the existence of at least three dramas.
One, of course, is Sariputraprakarana of Asvagho?a, son of Suvarnak?i. It
gives also the fuller title Saradvatiputra-prakarana, and the number of acts
is nine. The same manuscript which contains portions of this drama of Aiva-
gho$a has also fragments of two other dramas. There is no evidence of their
authorship, but they appear in the same manuscript as the work of Aivagho§a,
and they display the same general appearance as the work of that writer. It
is more probable that they are contributions of Asvagho$a rather than of some
unknown contemporary (Keith. Sanskrit Drama O.U.P. 1959 Reprint pp.
80 ff).
83. Fragments of this work were discovered in Central Asia. Its author
ship is controversial. The Chinese translation of the work made by Kumarajiva
about A.D. 405 assigns it to A£vagho$a (Nanjio No. 1182 p. 261), but the
finds of the fragments of the same work in Sanskrit in Central Asia and identi
fied by Luders accords it to Kumaralata, probably a junior contemporary of
Asvagho$a. Some scholars hold that A$vagho$a was the real author, and
Kumaralata only refashioned it. According to S.N. Dasgupta, Asvaghoâ
had nothing to do with its composition (History o f Sanskrit Literature,
Calcutta, Reprint 1977, p. 72 and n. 4.
84. Nanjio No. 1249 p. 274 translated by Sikhanand A.D. 695-700, and
earlier by Paramartha (No. 1250 ibid).
ASvaghosa as known from the Chinese translation of the work

by Kumarajiva. The colophon of the Turfan manuscript, how
ever, ascribes the work to Kumaralata, a well-known Buddhist
scholar. According to Luders who translated this work, its name
is Kalpanamanditika65 and its author is Kumaralata. According
to him, the Chinese tradition ascribed the work to ASvaghosa
by mistake. Kumaralata’s name is transcribed in Chinese as
Ku-mo-lo-to meaning ‘tong-show’ ‘boy received’. The Tibetan
tradition associates him with the school of the Sautrantikas.
Luders, however, advanced the hypothesis that ASvaghosa did
actually write a work entitled Sutralahkara, which was not
translated into Chinese but soon got lost, and was subsequently
confused with the Kalpanamanditika.
The Kalpanamanditika is a collection of pious legends on the
pattern of the Jatakas and Avadanas, and they are told in prose
and verse in the style of ornate poetry. It contains old legends and
parables, while others breathe the spirit of the Mahayana or at
least reveal Buddha worship. In this work itself, the teachers of
the Sarvastivadins are honoured and many of the stories in the
Kalpanamanditika are taken from the canon of the Sarvasti
vadins.8586*88 King Kaniska, as a ruler of the past, appears in two
narratives. It mentions the two epics—Mahabharata and
Ramayaija, and confutes the philosophical doctrines of Samkhya
and Vaisesika, and the religious views of the Brahmapas and
Jains, and also contains all kinds of reference to script, art and
painting. Kumaralata is further described in Chinese tradition
as a ‘Master of Comparisons’ (dristdnta) and as a ‘founder of
comparison’ (Ddrstantika). As such the principal work of
85. Luders. Bruchsticke der Kalpanamanditika des Kumaralata (Kon.

Preuss Turfan-Expeditions Kleinere Sanskrit Texte II). Leipzig 1926. Both
‘Kalpanamanditika’ and ‘Kalpanalankrta’ appear as titles in the Colophons
(Winternitz. Op. cit 267 & n). For the French translation of this work in
Chinese see Sutralankdra tranduit en francens sur la version Chinoise de
Kumarajiva par Ed. Huber, Paris 1908. cf. La Vallee Poussin in Le Museon
N.S.X. 1909, 86 ff.
86. Winternitz p. 269 and n. 1; cf. S. Levi in JA. 1968. S.10. t. XII. 91 ff;
184; Huber (in BFEO) 4,1904, pp. 709 ff has traced three stories in the Divya-
vadSna. The Sautrantikas originated with the Sarvastivadins and that explains
why there is no contradiction for Kumaralata, as a Sautrantika to honour
the Sarvastivada teachers (Luders. Op. cit, p. 22).
Kumaralata is mentioned in the Chinese texts as Yo-man-lun—

Dris\dnta-mald (pankti) 3astra. Levi, therefore, expresses the
opinion that the name of the Turfan text was really Dristanta-
pahkti (as found in the Colophon) and that Kalpandmantfitika
is its adjective. This text is considered by Levi to be a new edition
of the Siitrdlafikara of Asvaghosa expanded by the addition of
moral lessons and apologues in the form of examples (dristanta)
according to the practice of the Darstantika school. The Turfan
text, therefore, represents partially the Sutralankara of Asva
ghosa.87
Matriceta8889is another Buddhist scholar of eminence whose
hymns—the fragments of the poems—were brought to light in
Central Asia both in Sanskrit original as well as in Tokharian
translation- Like Kumaralata’s work—the Kalpanamantfitika—
some poems by Matriceta—have likewise been ascribed to Asva
ghosa in Tibet. It is proposed by Winternitz that Matriceta who
wrote the ‘Maharaja Kanika-lekha’, which has come down to us
in Tibetan language was an old contemporary of Asvaghosa. His
most famous hymns are the Catuh-sataka-stotra—‘the hymn of the
Four Hundred Verses’ Satapaficasatika** ‘the hymn of the
87. Bagchi. Op. cit p. 101; cf. Levi in JA. 1896, S. 9, t. VUI, pp. 444 ff.
88. For the life and works of contributions, see, F.W. Thomas’ article on
‘Matricheta’ in the Encyclopedia o f Religion and Ethics Vol. VIII, pp. 495-97.
Matricheta, the Buddhist author, is identified by Taranatha with Asvaghosa,
while a much older writer, the Chinese traveller I-tsing (2nd half of 7th cent.
A.D.) plainly distinguishes the two. The sole reason for the identification is
the fact that both writers stood in relation to Kaniska. A£vagho$a was no
doubt a figure at the court of Kaniska, where the epistle from Matricheta
declining upon grounds of old age and sickness, to visit the king, make the
possibility of their identification untenable. See Winternitz. Op. cit p. 270
& n. 2.
89. Hoernle. Manuscript Remains, p. 58 ff and 75 ff. The Satapancasatika-
Stotra folios were found in three different localities (Jigdaiik-Bai, Tun-huang
and Khora), as well as in different sizes. Besides introductory remarks of
Hoernle regarding this work and this author, see the comprehensive Intro
duction to the $ atapahcasatika—Sanskrit Text, Tibetan translation and
commentary and Chinese translation by D.R. Shaekleton Bailey, Cambridge,
1951. The Catuhsataka or Hymn of 400 verses is also noticed by Hoernle
Op. cit pp. 75 ff. Three fragmentary folios of this work came from two different
localities—Khora near Karashahr and Jigdalik near Bai. The two from Jig-
dalik have the same author and the same number of lines in a page. The three
one hundred and fifty verses’. Fragments of both these were

found amongst Central Asian manuscripts. According to I-tsing,*90
the Chinese pilgrim scholar of the seventh century, ‘it is delight
ful to- hear both the Hymns recited in the assembly of monks’.
These charming compositions are equal in beauty to the heavenly
flowers, and the high principles which they contain rival in dig
nity the lofty peaks of a mountain. He is admired by later poets
who imitate his style considering him as the father of literature.
His poems were part of the Syllabi for both the Mahayanist and
Hinayanist monks when they were initiated. Besides Asanga and
Vasubandhu, who admired him, Dignaga—the ‘Bodhisattva Jina’
compiled a hymn of 300 verses, known as ‘mixed Hymn of
Praise’in praise of the first work—‘Hymn of 150 verses’, whichwas
also translated by I-tsing into Chinese. Manuscripts of the Sans
krit original of the Satapahcasatika-stotra were recovered from
various sites in Central Asia such as Jigdalik-Bai, Tun-huang
and Khora, while Tokharian translations of the texts have also
been brought by the German expedition from Turfan. Frag
ments of the other works of Matriceta, the Catuhsataka-stotra
have also been recovered from Central Asia. There is no Chinese
translation of the text but the Tibetan translation gives the name
of the work as Varnanarha-varnana-stotra. Fragments of Sanskrit
manuscript were discovered from Khora (near Karasahr) and
Jigdalik(near Bai). The colophon of the chapters give the full
title of the work as Varnanarha-varnana-Buddha-stotra-catuhSa-
takam9192
Among other works of a literary nature and non-canonical in
character may be mentioned Aryasuras Jatakamald 92 Its manus
fragments are written in the slanting type of the Gupta script. I-tsing to whom
both this and the hymn of 150 (Satapaficasatika) appear to have been well-
known, translated only the latter into Chinese (No. 1456 p. 321) and not the
former.
90. For I-tsing’s praise of the work see Takakusu—Op. cit pp 156 f, 166.
92. Edited by H. Kern in Harvard Oriental Series I, 1891; translated by
J. S. Speyer in Sacred Books of the Buddhists Vol. I 1895, Oldenburg; JRAS.
1893 p. 308 ff. A. Gawronski, Studies about the Sanskrit Buddhist Literature,
Krakou, 1919, pp. 40 ff. Winternitz. Op. cit p. 273 n for other references.
I-tsing praises the Jatakamala (or Jatakamalas) among the works which were
particularly popular.
cripts have been found at Murtuq and Toyoq. It resembles in

style the Kalpanamanditika. This garland of Jatakas as the
generic name Jatakamala suggests, does not include or invent new
stories, but retells the old legends in ornate, elegant language in
the Kavya style—lofty and refined and more artistic than artificial.
The work contains 34 Jatakas like the 35 Jatakas of the Cariya-
pitaka, intended to illustrate the ‘Paramitas’ or ‘perfections of
a Bodhisattva’. The boundlessness of the Bodhisattva is also
glorified in many narratives.
Along with the Kavyas, two works on Sanskrit metrics and
grammar Chandoviciti and Katantra are supposed93 to have been
in use in the northern area of Chinese Turkestan. It is proposed
on the basis of Sanskrit literary compositions in official donatory
records from Kucha and Agnide£a of about the seventh/eighth
century A.D. that Sanskrit works were composed in Central
Asia. The Jatakastava preserved in a Khotanese metrical trans
lation by Vidyasila seems to have been written in Sanskrit as in
dicated by the Syntax. Its Khotanese translation was done in
the second half of the tenth century in the Sainanya monastery
in Khotan.94 In this context a Kharosthi inscription refers to
the study of grammar, poetics and astronomy. Another inscrip
tion95 notices the names of twelve naksatras named after twelve
animals, also recorded in a portion of the Mahasannipdta-sutra,
translated in the second century A.D. The text is supposed to
have originated, at least partly, in Central Asia and appears to be
a fragment of an astrological treatise written in a more or less
barbarous Sanskrit.96 It is an unidentified fragment and the text
written in a curiously debased dialect, is astrological. Its name is
Rishi Kharusta, said to be the reputed inventor of the Kharosthi
script.
The literary remains brought from the ancient sites include a
large number of fragments of Sanskrit medical texts, testifying to
the conveyance of Indian medical literature in Central Asia. The
93. Mookerji. Op. cit p. 714.

94. ibid.
95. Kharosthi Inscriptions (Ed. Boyer-et-others) (Op. cit) No. 514.
96. Hoernle. Op. cit pp. 121 ff.
earliest discovery was by Col. Bower97 of three different medi

cal texts from an old stupa near Kucha. One of these texts,
named after him, deals with the origin, use and efficacy of garlic
which is said to cure many diseases as also extending the longe
vity upto 100 years. The text also deals with digestion, an elixir
for longevity, correct mixing of ingredients, other medicines,
lotion and ointment for eyes etc- The second text contains 14
medical formulae for external and internal use, while the third one
—the largest one—called Navanitaka (cream) contains an abstract
of the earlier medical literature, including those of Agnidega, Bheda,
Jatikarria, K?arapani, ParaSara and Susruta. The text deals with
the preparation of powder, decoctions, oils and also with injec
tions, elixirs, aphrodisiacs, nursing of children etc. The instruc
tions at the beginning of the text enjoin that it should not be
given to anyone who has no son, nor to anyone who has no
brother, nor should it be taught to anyone who has no disciple.
The last part suggests provision for medical instructions. There
is also a treatise on a charm against snake-bite. The Indian
medical texts were also translated into Kuchean and Khotanese
languages, thereby pointing to the prevalence of Indian system
of medicine and its popularity.
Local Translations
The northern and southern regions—with their separate langu
ages—Kuchean and Khotanese—have also provided local transla
tions of Buddhist canonical and non-canonical texts in these
languages. The language of Kucha—known to the Uighur Turks
as Kusana, was spoken in the region from Aksu (ancient Bharuka)
to Turfan (ancient Agnide§a). The French Mission discovered
literary remains in different parts of this area. A few frag
ments in this language were also recovered by Aurel Stein, and
some others by the German and Russian explorers. The Kuchean
texts of the German collections have not been published, but
fragments of the Stein and Russian Collections were published
along with the French one by Sylvain Levi and later on by Jean
97. For an account of the discovery of this manuscript and its contents
in a summarised form see Ghoshal. Op. cit p. 266; Bagchi. Op. cit, p. 104.
Filliozat.98 Fragments of texts in the other dialect, called To-

kharian, were discovered by the German mission in Karasharand
Turfan region. The spoken language, however, was Kuchean.
It is suggested" that the Tokharian literature was probably deve
loped in some other area most likely in Tokharestanand the texts
were brought to this region during the Uighur period. The major
part of the German collections is from Shorchuq in Karasahrand
the rest from the sites of Bazaklik, Murtuq, Sangim and Idikut-
shari at Turfan.
The Kuchean and Tokharian fragments are all translations of
Sanskrit Buddhist texts with some bilingual extracts in original
Sanskrit as well. These pieces were helpful in interpreting the
Kuchean language extracts and in the identification of the frag
ments. These fragments include Kuchean translation of Sarvasti-
vada-Vinaya like Pratimoksa,100 Prdyascittika and Pratidesani-
y a f01 These canonical rules in the language of use were meant for
those monks who could not follow the Sanskrit texts. The bilin
gual texts might as well have been used for teaching Sanskrit
to the members of the Buddhist order in that region. The Kuch
ean translations of Udanavarga,102 Udanastotra and Uddndlankara
were very popular with the Buddhists. Fragments of Kuchean
translation of a very extensive work calledKarmavibhariga—a text
98. Textes Koutcheens : Fragments de Textes Kouctcheens de Medicine
et de Magie—Paris 1948.
99. Bagchi. Op. cit p. 106.
100. Hoernle. Op. cit p. 357; Stein. Ancient Khotan I p. 483 edited with
translation in French by S. Levi. The very earliest Buddhist literature must
have included a Pratimoksa. It is known in a Sanskrit version and in one
Tibetan and four Chinese versions. The Pratimoksa is to be recited twice a
month in an assembly of at least four monks, who confess their sins to each
other before the ceremony. This disciplinary code is at the same time a for
mulary of confession (Kern : Manual o f Buddhism—Reprint—Delhi 1968
p. 74); cf. the Chinese translation of the Pratimoksa of the Sarvastivadins
by Kumarajiva about A.D. 404. (Nanjio No. 1160).
101. Hoernle. Op.cit p. 365 with a French translation of both the frag
ments by S. Levi.
102. Fragments of the Udanavarga from finds in Central Asia have been
published by R. Pischel ‘Die Turfan-Resensionen des Dhammapada’. SBA.
1903 p. 96S IT; S. Levi and La Vallee Poussin JA.S.10 t. XV. 1910 p. 444 ff;
t. XVII, 1911,431 IT; t. XIX. 1912 p. 31 ff; JRAS. 191 l.p . 758 ff; 1912 p. 335
ff; Ref No. 55.
on the retribution of acts (Karma) were in use by the Buddhist

priests, as a handy text expounding the doctrine of transmigra
tion.
The non-canonical translations in the Kuchean language in
clude those of medical and Tantric Sanskrit texts. One such is
Yogasataka, a medical work in about one hundred verses. It is
an excellent abridgement of standard medical texts and a con
venient medical handbook which could be used with profit and
by practitioners. While Caraka and Susruta Sarhhitas are not
available in Kuchean translation (even in fragments), there are
no doubt fragments of local translations of medical texts from
Pelliot, Weber and Stein collections. Sanskrit names of drugs
are transliterated with phonetic changes which could be easily
adopted in local language.103 Besides Kuchean, fragments of
translations of Buddhist texts in the other dialect called Tokha-
rian have also been traced though not closely studied. They
contain bilingual texts and include such popular texts as the
Udanavarga.
Buddhist texts were also translated into theKhotanese language,
called by some as the Saka language. It continued to be in use
even as late as the seventh-eighth centuries when there was
no trace of the Sakas in Central Asia. Khotanese—as this
language is designated—is associated with the translation
of Mahayana Sanskrit Buddhist104 texts. These translations
include those of the Suvarnaprabhasa-sutra, Vajracchedika,
Aparimitayus-sutra,105 Bhadracaryadesand, Jatakastava and
Maitreya-Samiti-nataka. The non-canonical texts include
translation of two Indian medical texts : the Siddhasara
and Jivakapustaka.106 The former is attributed to one
103. Bagchi. Op. cit p. 106.
104. H.W. Bailey : Khotanese Texts, pp. 94 IT; Bagchi. Op. cit p. 107.
105. Stein. Serindia—Op. cit—II p. 914.
106. Bagchi. Op. cit 107. The name Jivaka appears in the Sanskrit portion
of a manuscript of medical treatise, written in a form of crude Sanskrit, and
also in £aka Khotanese. Fragments of this manuscript were found in the
cave of Thousand Buddhists near Tun-huang. On the folios available each
phrase of words in Sanskrit is followed by a full translation in Saka-Khotanese.
There are some medical formulaeas well, which are not traced elsewhere.
Neither the name of the text nor that of the author is known (Mookerji. Op.
cit p. 715 quoting R.G. Bhandarkar Commemoration Volume, Poona 1917
pp. 415 ff).
Ravigupta. The rendering is based on the Tibetan transla

tion and contains portions concerning tantra, dravya, arsa, bha-
gandara, panduroga, mutrakricchra, netra-roga etc. The original
of the other text, the Jivakapustaka, however, is not traced, but
it must have been in Sanskrit as the interlinear Sanskrit verses
show.
There are also Buddhist texts in Sogdian107 language which was
spoken not only in ancient Sogdiana (north of Tokharestan with
its centre at Samarkand), but also in other parts of Eastern
Turkestan, which had colonies of Sogdian merchants. It was
fairly popular in that part of Central Asia, and was used by the
Buddhist priests and later on in the 9th and 10th centuries by the
Manichaean priests as well. The Sogdian translations of Bud
dhist texts include those of Dirghanakha-sutra, Vassantara-Jdtaka,
Vimalakirtinirdesa, Dhvana-sutra, Dhutasutra, Nilakantha-
dharani108 and Padmacintamani-dliarani-sutra. The translation
of Buddhist Sanskrit texts into the major languages of Central
Asia is evident from Uighur manuscripts109 as well besides those
mentioned earlier. Tokharian translations of Maitreya-samiti-
ndtaka, Suvarnaprabhasa-sutra, Jatakas, Sutra of Kalyanamkara
and Pdpamkara etc. were rendered into Uighur-Turkish. The
Turfan area provides fragments of three other manuscripts whose
colophons refer to a work called Dasakramapudarutanamal
( = Dasakarnuj-patlidvaddnanmla) , belonging to the Avadana
class. It is supposed110 to have been first imported into
Kucha or might have been composed there. It was translated
from WKW Kwys’n ( = Sanskrit) intoToyari or Toxari (Tokha
rian =Kuchean) and from that into the Uighur-Turkish language.
The Indian Buddhist texts are also supposed to be amplified
in Central Asia111 at the hands of local scholars who might have
107. Serindia. II. p. 920 IT.

108. Among the rolls with Sanskrit texts, NilakaVtliadharani with an
interlinear Sogdian version dcseives special mention since its publication by
Louis de la Vallee Poussin and Gauthiot. Sylvain Levi has suggested strong
reasons for placing the date of this bilingual manuscript between A.D. 650-
700 (JRAS. 1912 pp. 629 Sq. 1063 Sq.).
109. For the Vigur manuscript, see Serindia p. 925; Bagchi. Op. cit,
p. 108.
110. Mookerji—Op. cit, p. 716.
111. Central Asia is supposed to have provided new aspects to Buddhist
been Indian immigrants or had their education at centres of

learning in India. The Chinese evidence does provide some infor
mation on these points. The story about the search by Dharma-
raksa in Khotan for a complete text of Mahaparinirvana-sutra,
even though he had already found the text consisting of ten
chapters, suggests its amplication in Central Asia. Besides this
canonical work, according to the Chinese evidence, manuscripts
of the Mahasannipdta-sutra, Avatamsaka-sutra, Vaipulya-sutra,
Ratnakuta-sutra, Lahkavatara-sutra, Sariputra-dharanl, Mahapra-
jhaparamita, Astasahasrika-prajhaparamita etc. were preserved
in the Chokkuka area. The contents of the Chinese versions of
Candragarbha-sutra and Suryagarbha-sutra in the absence of the
Indian originals could be suggestive of their being recast in Ser-
india. This contention could as well be true in the case of the
other Chinese translations of Buddhist canonical works noticed
above.
The Buddhist literature in the course of its migration from
India to China assumed some new aspects in Central Asia. In
this direction the famous monasteries, their abbots and dons,
seem to have played an effective role for preservation and pro
pagation of Buddhist canonical literature as also its reconstitu
tion. They seem to have made active and conscious efforts to
naturalise the canon by introducing in it such elements as could
bear the stamp of Sea-Indian literary contribution. Thus the
Suryagarbha-sutra11213in one of its sections mentions Buddhist
Holy places which include besides those in India, Cinasthana
(China), Khasa (Kashgar ?), and GomalaSalagandha Caitya on
the Gosringa mountain of Khotan—one of the holiest places
visited by the Buddhist devotees. The fragmentary manuscript of
Prakrit Dharmapada in KharosthI writing was procured from the
Gosringa monastery site built on the slopes of the mountain of
this name. The Candragarbha-sutra112 also mentions the kingdoms
literature in course of its migration from India to China. These are discussed
in details by Bagchi (pp. 108 ff). Unfortunately references are not provided
by him.
112. See Nanjio—Op. cit. No. 62 p. 28. It was translated into Chinese
by Narendrayasas of the Sui dynasty A.D. 589-618.
113. See Nanjio—Op. cit No. 63 p. 29. Translated by the same person
Narendrayasas but under the Northern Tsi dynasty A.D. 550-577.
of Central Asia supposed to be visited by the Buddha. These in

clude Asoka, Darada, Khasa, Cokkuka, Shalei (Kashgar),
Khotan, Kucha, Bharuka, Hecyuka, Yi-ni(Agnidesa), Shan-Shan
(Kroraina) and Chinasthan in the first list. The second one enu
merates the number of incarnations of Buddha in various count
ries, including those outside India. Judging by the number of
these incarnations, Khotan, Kucha and China are given places
of the greatest honour in the Buddhist world.
A Buddhist scholar of Kucha named Lu-yen114 compiled in
the seventh century a Sanskrit-Chinese lexicon which has been
preserved in the Chinese Pitaka. This lexicon modelled on the
Sanskrit Kosas also contains words of Central Asian origin such
as kalam (pen), kakari (paper), makara (monkey), kavasi (san
dal), and names of places in Central Asia and China such as
Trusaka (Turk), Korttana (Khotan), Kuchnia (Kucha), Wu &
Shu (provinces of China), Kunudana (Khumdan i.e. Ch’ang-
ngan, Capital of China). There is also a Sanskrit-Khotanese text,
useful for literary and colloquial purposes.115 While its text in
Sanskrit is corrupt, it is much influenced by Khotanese pronun
ciation.
The development of Buddhist literature in Central Asia and
the contribution of its savants could not have been possible
without the patronage of its rulers, some of whom bore Indian
names like Vijayakirti, Vijayasambhava etc. of Khotan, Hari-
puspa, Suvarnapuspa, Suvarnadeva etc. of Kucha, Indrajuna,
Candrajuna etc. of Agnidesa. The Indian element in the local
population and in the royalty appears to be fairly conspicuous
and that accounts for the patronage accorded to Buddhism and
its literature in Central Asia. This could as well be ferreted out
from a study of the material culture of the Central Asians—an
area of detailed study in the next chapter.
Bactria in Northern Afghanistan, with its language written in
Greek script, was also the centre of Buddhism and this script was
used in the records of certain Buddhist monasteries. Sanskrit
seems to be understood in this part of Central Asia, as is evident
114. Bagchi—Le Canon Bouddhique en Chinese—Op. cit pp. 565-66;

cf. India and Central Asia, pp. 110-111.
115. Bagchi—ibid p. 112.
from the discovery of a fragment of the Sanskrit text of the

Sahgitaparyaya116 in a cave at Bamiyan. It is written in the
North Turkestanese script of Chinese Central Asia. While it
might have been imported there for use by the local monks, it
nevertheless points to the understanding of Sanskrit canonical
literature by the Buddhist monks of Bactria, who might not be
having any literature in^their language. Prakrit was used there
in the early centuries of the Christian era in Buddhist donative
records. In this context a record from the Tochi valley—now in
Pakistan—uses both Bactrian and Sanskrit.11617 A Bactrian inscrip
tion on a vessel found in the ruins of a monument near Termez
suggests philosophical or religious impact in that region. It is
translated as : ‘He who makes no distinction between his own
‘I* and that of others is on the proper road’. Like the Bactrians,
the Parthians, although followers of Mani, are also supposed to
have been influenced by Buddhist literature in their amulatic
text, with a list of Yaksas closely resembling those in the Bud
dhist dharani texts. It is claimed that there is evidence revealing
a strong influence of the literature of Northern Buddhism on
Manichaean-Parthian literature.118 A Manichaean-Parthian text
refers to Mani as the Buddha.
Tibetan Buddhist Literature
It would not be out of context to take notice of Buddhist
literature in Tibetan from Central Asia as well as from Tibet,
based on Sanskrit Buddhist literature, as also the contribution
of important Indian scholars to Tibetan Buddhist literature.
The two main divisions of Tibetan literature are Kanjur and
Tanjur.119 The first one includes works esteemed as canonical,
including Tantras. The second is composed of expositions of
scriptures and also of many treatises on such subjects as medi
cine, astronomy and grammar. The canonical part is smaller but
the commentaries and miscellaneous writings are more elaborate.
116. Mookerji. Op. cit p. 718.

117. ibid. p. 718.
118. BSOAS. XII. pp. 47-48.
119. The Tibetan orthography is bKah-hgyur (the translated command)
and bsTan-hgyur (the translated explanation). Various spellings are used by
Language, and Literature 217
The great literary epoch of Tibet was in the ninth century. The
Tibetan scriptures reflect the late Buddhism of Magadha when
the great books of the Hinayanist canon were neglected and a
new Tantric literature was flourishing with exuberance. That
accounts for the absence of the Abhidhamma works of the
Hinayana and none of the great Nikayakas. The whole collec
tion of the Kanjur is commonly divided into seven parts : The
Dulva,120 equivalent to the Vinaya, is stated to be the Mula-
European writers such as Kah-gyur, Kandijour, Bkahgyur etc. Weddel writes

Kah-gyur and Tan-gyur (Eliot. Op.cit III p. 372 ff and n .1). See also Waddell:
Buddhism and Lamaism o f Tibet—Reprint, New Delhi 1974, pp. 157 ff. While
the two terms are suggestive of two forms of Tibetan canonical Buddhist
literature—the first one including Tantras also, the second one is composed
of exegetical literature and also of many treatises on such subjects as medicine,
astronomy and grammar. This distinction seems to hold good on the whole,
yet it is not strictly observed. Thus, the work called Udana and corresponding
to the Dhammapada is found in both the ‘Kafijur’ and ‘Tanjur’ (Eliot. Op.
cit, III, p. 372 n 2). The Lamaist Scriptures are faithful translations from the
Sanskrit texts and a few also from the Chinese made mostly in the eighth
and ninth, and the eleventh to the thirteenth centuries A.D.
120. Eliot. Op. cit, p. 373 and note for references. The Dul-va (Skt.
Vinaya), or ‘Discipline’ is said to be compiled by Upali in thirteen volumes.
It deals with the religious discipline and education of those adopting the reli
gious life, and also contains Jatakas, VySkaranas, Sutras and Nidanas. It is
sub-divided into seven parts :—
(1) The basis of discipline or Education (dul-va-gzi-Skt. Vinaya-vastu)
in four volumes translated from the Sanskrit in the ninth century
by Sarvajnadeva and Dharmakara of Kashmir and Vidyakara-prabha
of India, assisted by Tibetan scholars.
(2) SOtra or Emancipation (‘So-Vor-tar-pai-mdo’ Skt. Praiimok$a-sutrdl
in 30 leaves.
(3) Explanation of Education (‘Dul-va-nam-par-byod’pa Skt. Vinaya-
Vibhaga) in four volumes. It also provides directions for dress and
etiquette.
(4) Emancipation for Nuns (‘dGe-Shon mahiso-Sar thar paimod’ Skt.
Bhiksuni-pratimoksa-sutra.) 36 leaves in the ninth volume.
(5) Explanation of the Discipline of the Nuns (Skt. Bhik. Vinaya-Vibhaga)
in preceding volume.
(6) ‘Miscellaneous Minutiae concerning Religious Discipline’ (Dul-va
pran-tseqs-Kyi-gzi, Skt. Vinaya-ksudraka vastu) in two volumes.
(7) The highest text book on Education (Dul-va gzunv-bla-ma— Vinaya-
uttara Granthd) in two volumes. (Waddel. Op. cit, p. 160).
For the analysis of the Dulva etc. four parts in Asiatic Researches, Vol. XX,
1836, by A. Cosund Korosi; Translated into French by Feer '’Annals dumusee'
Sarvastivada Vinaya and agrees with the Chinese translation of

Kumarajiva and to a great extent with the Sanskrit Pratimok$a
found at Kucha.12112The second division Ser-chin corresponds to
the Prajhapdramita122 containing its longer and shorter versions.
In this group the Vajracchedika is very popular. Phal-Chan,
equivalent to AvatariiSaka and dkon-brtseqs or Ratna-kuta are
the two other divisions. The former in its sub-divisions appears
as separate treatises in the Chinese Tripitaka called Hua-yeri.123
The next one agrees closely with the similar section of the
Chinese Tripitaka despite its shorter form.124 The other three
sections are mdo equivalent to Sutra,125126comprising important
works as the ‘Lalitavistara’, ‘Lankavatara’ and ‘Saddharma-
pundarika’; ‘Myang-hdas’ or Nirvana128—describing the death
of the Buddha and rGyud or Tantra,127 with many of its texts
Guimet loin zme, 1881; also Waddele *Tibetan Manuscripts and Books' in
Asiatic Quarterly. July 1912 pp. 80-113. Eliot. Op. cit III p.373n for other
references to several editions and translations of short treatises.
121. See Nanjio. Nos. 1115-1119, 1112, 1132. Rockhill. Pratimok?a-
siitra selon la version Tibetanie, 1884. Finot and Huber.
122. Transcendental wisdom (Skt. Prajna-paramita) in twenty-one volumes
contain speculative doctrines which the Mahayana school attributes to
Buddha’s revelations in his mythical discourses. There is no historical matter,
all is speculation and metaphysical (Waddell. Op. cit p. 161). Estimated by
the Tibetans to be the Abhidharma, it is said to have been first collected by
Kasyapa, representing the teachings delivered by the Buddha in his fifty-
first year (Eliot. Op. cit p. 374).
123. Waddel in Asiatic Quarterly 1912, XXXIV p. 98 renders the title as
‘Vata-sangha’ which according to Eliot probably represents Avatamsaka.
S.C. Das identifies Phal-Chen, Sde pa with Mahasanghika. Fecr notices its
forty-five sub-divisions (Eliot p. 374).
124. cf. Nanjio—Op. cit No. 1234 p. 271. It enumerates several qualities
and perfections of Buddha and his doctrine (Waddel Op. cit. p. 162).
125. Of the 270 works contained in this section about 90 are prinia facie
identical with works in the Ching division of the Chinese Tripitaka and pro
bably the identity of many others is obscured by slight changes of title (Eliot.
Op. cit, p. 374). The subject of the works is various. The greatest part of them
consists of moral and metaphysical doctrine of the Buddhist system, the legen
dary accounts of several individuals etc. etc. (Waddel. Op. cit p. 162).
126. It consists of only one work, corresponding to Nanjio 113 (Mafia-
parinirvdna-sutra) p. 39.
127. It consists of twentytwo volumes containing about 300 treatises,
out of which between thirty and forty are prima facie identical with those in
agreeing with treatises comprised in the Chinese Tripitaka.

Among the Tantras*128 may be mentioned the Mahdganapati-
tantra, the Mahdkala-tantra, the Arya-Manjusri-mula-tantra
and Srl-Guhyasamaja\ the more important being Paramadi-
buddha-uddhrita-Sri-kalacakra on which the Kalacakra129 system
is founded.
The Taiijur130 is a considerably larger collection than the
Kanjur consisting of 225 volumes as against a 100 or 108 of the
the Chinese Tripitaka. Many of these are Brahmanic in spirit rather than
Buddhist, like the Mahdganapdlitanlra and Mahdkdlatantra) (Eliot. Op. cit,
p. 375).
128. The idolatrous cult of female energies grafted upon the theistic
Mahayana and the pantheistic mysticism of Yoga equally resulted in the
evolution of Tantric Buddhist literature in Tibet. The difference between the
‘Dharanis and the Mantras’ belonging to the Tantras became more and more
obliterated and finally the Dharariis were completely supplanted by the
Mantras. In the Tibetan Kanjur the Dharanis are to be found both among the
Sutras (Mdo) and the Tantras (Rgyut). There are four classes of Buddhist
Tantras : Kriyd-Tantras, Carya-Tantras—teaching of the practical cult, Yoga-
Tantras—dealing with the practice of Yoga and Anuttarayoga-Tantras which
deal with higher Buddhism (Winternitz. Op. cit p. 389 and n. 1 for reference).
129. For the reference to Tibetan literary texts see Eliot Op. cit III
p. 375. The extreme development of the Tantric phase was reached with the
Kalacakra which, according to Waddell, is a coarse Tantric development
of the Adi-Buddha theory combined with the puerile mysticism of the Man-
trayana. It attempts to explain creation and the secret powers of nature by
the union of the terrible Kali, not only with the DhyanT Buddhas, but even
with the Adi-Buddha himself. The demoniacal ‘Buddhas’ created through
this union under the names of Kalacakra, Heruka, Achala, Vajra-Vairabha
etc. are credited with powers not inferior to those of the celestial Buddhas
themselves. They and their female spouses—ferocious and blood-thirsty—
are to be conciliated with offerings and Sacrifices, magical circles, special mantra-
charms etc. (Op. cit p. 131).
130. The Tibetan commentary or Tanjur is a great cyclopedic compila
tion of all sorts of literary works, written mostly by ancient Indian scholars
and some learned Tibetans in the first few centuries after the Introduction of
Buddhism into Tibet commencing with the seventh century A.D. Divided
into the rGyud and mDo (Tantra and Sutra) classes, the former covers tantrika
rituals and ceremonies in eighty-seven volumes, while the latter concentrates
on science and literature in one hundred and thirty-six volumes. One separate
volume contains hymns or praises on several deities and saints, and another
is the index for the whole. The first sixteen volumes of the mDo (Sotra) class
are all commerntaries on the Prajhd-pdramitd, followed by several volumes
explanatory of the Madhyamika philosophy (of Nagarjuna) founded on the
latter. It is known to contain a great deal of relatively late

Indian Buddhist works such as those of Asvaghosa, Nagarjuna,
Asanga, Vasubandhu and other Mahayanist scholars, and also
secular literature such as the Meghaduta of Kalidasa together
with a number of works on logic, rhetoric, grammar and medi
cine. While some treatises, such as the Udana (Dhammapada)
occur in both the collections, the Tanjur group is considered as
a thesaurus of exegetical and scientific literature. While grammar
and lexicography are helpful in understanding the scriptures, the
study of secular law demands an amplification of the Church
rules and code of conduct. Medicine is useful in establishing the
influence of the Lamas. History compiled by theologians suggests
that true faith is progressive and always triumphs. Art and ritual
are very close to each other. It is, therefore, incumbent on a
learned person to be familiar with these disciplines. That
accounts for all-pervasive character of the Tanjur literature.*131
Both the compilations have Tantric compilations of later times
closer to the Mahayana Sutras than the teachings of the Buddha.
Further, the great majority of works in both the collections
have the Sanskrit name first prefixed in transcription followed by
its translation. The Indian proper names are generally translated.
There are some translations from Pali and a few from Chinese.
One work is translated from the Bro-za language (perhaps from
Gilgit) and another from the language of Khotan. Some works
in the Kanjur literature have no Sanskrit titles and are perhaps
original works in Tibetan. The Tanjur literature contains many
such works.
Prajnaparamita. One volume contains the Tibeto-Sanskrit dictionary of

Buddhist terminology, the bye-brag-tu-rtogs—the Mahavyutpatti. Later
commentaries, such as the Bodhi-patha (in Mongolian Bodhi Mur) are also
included under this heading. Its contents include rhetoric, grammar, prosody,
mediaeval mechanics and alchemy (Waddell. Op. cit p. 165).
131. Huth’s analysis of Vols. 117-124 of the Tanjur shows that they
contain inter-alia eight works on Sanskrit literature and philosophy besides
the Meghaduta, nine on medicine and alchemy with commentaries, fourteen
on astrology and divination, three on chemistry (the composition of incense),
eight on gnomic poetry and ethics, one encyclopaedia, six lives of the saints,
six works on the Tibetan language and five on painting and fine art (Sitzungs-
ber Kon Preuss Akad Wiss. Berlin. 1855—quoted by Eliot p. 376 n. 3).
Both the Kanjur and Tanjur as a whole represent late Bud-,

dhism of Bengal and also resemble in arrangement with the
Chinese Tripitaka replacing the old Pitakas and Agamas. Being
later than the Chinese, the Tibetan Canon lacks the Abhidharma
but adds a large section of Tantras. Both the Tibetan and the
Chinese canons recognise the divisions known as ‘Prajnapara-
mita’, ‘Ratna-kuta’, ‘AvatamSaka’ and ‘Mahaparinirvana’ as
separate sections.132 The Tibetan translation of all this literature
falls into three periods: from the seventh century until the reign
of Ralpachan in the ninth; the reign of Ralpachan, described as
the Augustus age of Tibet, and some decades following the
arrival of Atlsa in 1038. The first period includes contributions
of Thonni Sanbhota who was sent to India in 616 and made
renderings of Karanda-vyiiha and other works. A Tibetan
manuscript of the Salistambha-sutra (not later than A.D. 740)
was discovered by Stein at Endere.133 Padmasambhava and
Kamalasila also contributed during this period. The former
translated three works while seven of the latter’s original
works are preserved in the Tanjur. He also translated a part of
the ‘Ratnakuta’. It was also the beginning of the great period
of translation to follow and two dictionaries containing a collec
tion of Sanskrit Buddhist terms were also composed, namely
the Mahavyutpatti and an abridgement.134
Ralpachan135 who ruled in the ninth century summoned from
Kashmir and other parts of India many Buddhist savants for
revising the then existing translations with the assistance of
132. Eliot. Op. cit p. 378.

133. Ancient Khotan pp. 426-9 and App. B; cf. Pelliot. BEFEO. 1908,
pp. 507 ff.
134. The Mahavyutapatti edited by Minayeff in Bibliothica Buddhita
and an abridgement. (Eliot 379 n. 2).
135. Ralpa-Chan who ruled in the ninth century is described as the
Augustus of Tibet. It is probable that at least half of the works now contained
in the Kanjur and Tanjur were translated or revised at this time and that the
additions made later were chiefly Tantras (rGyud). It is also probable that
many tantric translations ascribed to this epoch are really later. According
to Feer (Analyse p. 325), the Tibetan historians state that at this epoch kings
prohibited the translation of more than a few tantric works (Eliot. Op. cit
III p. 379 and n).
222 Buddhism in Ceniral Asia
native monks and also adding many more of their own. The
most prolific translator was Jinamitra, a Vaibhasika scholar
from Kashmir, who translated a large part of the Vinaya and
many sutras assisted by Ye’ses-sde and Dpal-brTsegs commonly
described as Lo-tsa-va. The death of Ralpachan was a great
blow to Buddhist activities for a century. The revival that
followed was distinctly Tantric with the arrival and contribution
of AtiSa138 from Vikramasila. During the eleventh century a great
number of Tantric works were translated* Atisa is credited with
the revision of many works in ihe Tantra section of the Kanjur
as also twenty others composed by him. Atisa’s disciple Bu-ston
is credited with the compilations of the Kanjur and Tafijur with
definitive arrangement. The Kanjur was later on translated into
Mongol by order of Khutuktu Khagan136137 (1604-1634).
In a broad survey of the Buddhist literature in Central Asia
and Tibet, based on the literary finds in Chinese Turkestan as
well as in the western region comprising of Soviet Central Asia
and Bactria, reference has been made to the contribution of
Indian scholars and savants who first carried the message of the
Tathagata. A brief notice of the languages represented in
manuscripts, fragments and inscriptions would no doubt provide
reference to many influences at work in Central Asia which
served both as a receiving as well as a distributing centre. The
number of tongues simultaneously in use for popular as also for
religious purposes was fairly large. The writing materials em
ployed included palm leaves, birchbark, plates of wood or bam
boo, leather and paper. Numerous Sanskrit writings all dealing
with religious or quasi-religious subjects such as medicine and
grammar—have been found amidst the literary relics. While the
Mahayanist literature is abundant, that relating to the Hinaya-
136. Atisa while clinging to Yoga and Tantrism, at once began a reform
ation on the lines of the pioneer Mahayana system by enforcing celibacy and
high morality, and by deprecating the general practice of diabolic arts. Per
haps the time was now ripe for the reform, as the Lamas had become a large
and influential body, and possessed a fairly full and scholarly translation of
the bulky Mahayana Canon and its Commentaries, which taught a doctrine
very different from that practised in Tibet (Waddell. Op. cit, p. 54).
137. Eliot. Op. cit p. 401.
Language, and Literature 223
nist school is not missing. Sutras from the Agamas, a consider

able portion of the Dharmapada and the Pratimoksa of the
Sarvastivadin school, fragments of two Buddhist religious
dramas—one such being Sariputra-prakarana of Asvaghosa, many
translations of Mahayanist literature such as the Suvarnapra-
bhdsa, Vajracchedikd and Aparimitayus-sutra found in the
southern part of the Tarim basin, are all suggestive of Sanskrit
being understood in polite and learned society in Central Asia.
Kuchean and Khotanese—the former in two dialects—were as
well employed for local translations written in Brahml and
Kharosthi scripts. While the use of Kharosthi could be traced
rather to earlier times in the southern Tarim area, Brahml domi
nates the literary scene for a longer period and on a wider
horizon.
The Buddhist literature in Central Asia, like its architecture,
represents several periods distinguished by their contents, scripts
and fine spots. Fragments of the Sanskrit Agamas found at
Turfan, Tun-huang and in the Khotan area represent the first
period, followed by fragments of the dramas and poems of ASva-
ghosa from Turfan, the Pratimoksa of the Sarvastivadins from
Kucha and numerous versions of the anthology called Dharma
pada or Udana. Its Prakrit version found in the neighbourhood
of Khotan and fragments in Tokharian and Sanskrit are all
supposed to represent the canon, as it existed in the epoch of
Kaniska representing its old stratum.
The new stratum includes Mahayanistic Sutras either written
or re-edited in Central Asia. The popularity of the Prajhd-
paramitd, the Saddharma-pundarika and the Suvarnaprabhasa is
evident from finds of their fragments or part manuscripts from
different areas. As Buddhism seems to have co-existed with
other systems, its literary impact on those religions might not
be denied. The construction and phraseology of Manichaen
books resemble those of a Buddhist Sutra. Some similarities
could be traced between Buddhism and Nestorian Christianity,
Taoism and even Zoroastrianism. A fuller study of this point
is of course beyond the purview of literary enquiry in this con
text.
Reference to Buddhism and its canonical literature in Tibet is
also traced here with a fuller review of the literary contribution
of Indian scholars to Buddhist literature in the ‘land of the snow.’

It is not possible to provide a more comprehensive and elaborate
study of Central Asian literature, especially with reference to
Buddhism, till the huge stock of manuscripts and fragments still
lying unexposed are studied and published. One need only wait
patiently and be content with the available material on the
subject.
CHAPTER V
MATERIAL CULTURE
Central Asia presents a picture of rich and assimilative cul

tural pattern. Its complete isolation from oceanic influence
resulting in extreme aridity, physical configuration of a vast
area covered by the deserts, and mountainous and hilly tracts
making communication terrain difficult, have no doubt con
tributed to the development of local cultures along the oases
fringing the Tarim Basin. Its two zones—sharply divided into
nomadic and sedentary ones—were occupied by peoples of
different ethnic origin and engaged in separate avocations. The
former included nomadic hordes of Aryan stock —the Scythians
at the one and the Turco-Mongol at the other end in the region
now known as Outer and Inner Mongolia and Manchuria. In
the South, the region between the Jaxartes and the Oxus was
occupied by the people of Iranian origin. The area known as
Eastern Turkestan from the Pamirs upto the frontiers of China
was under the occupation of Aryan speaking people of different
affiliations. Movements of the nomads of the steppes either in
the west or in the east affected the sedentary life of the people
in the south through the ages. This historical process, therefore,
accounts for the evolution and development of local cultures
along the routes of communication, with the impact and in
fluence of the stronger one on the weaker, and equally resulting
in the assimilation of the former in the local ethos. Cultural
adaptability was not an unknown phenomenon. It ensured man’s
security and survival in his new environment. With this diver
sity provided by geographical factors in the cultural history of
Central Asia, two things, however, contributed towards cultural
integration, namely Buddhism and the Silk trade. Trade con
tacts between China and the West through Central Asia seem
to have been established earlier than the introduction of Bud-
dhism into Central Asia and through it into China. Central Asia
played its role in the transmission of cultural traits from one
part of the old world to another. The horse culture of Central
Asia was, however, its own and the use of long boots, stirrups
as also fur was carried from Central Asia to other parts of the
world. New inventions, new ideas, new manners and customs
continued to spread from one direction to another through this
area. Alexander’s campaign also provided cultural stimuli to
this move, with the impact and influence of classical art, while
Graeco-Roman glass influenced Chinese craftsmanship. The
Chinese contribution was in terms of peach and apricot, carried
through Central Asia and Persia to the European world. The most
important product was the silk from China which was exported
through two routes in Central Asia—the northern one passing
through Turfan, Karashahr (old AgnideSa) and Kucha, and the
southern one through Miran, Niya, Khotan and Yarkand. The
terminal points of the two routes at the eastern end were Tun-
huang and at the western one Kashghar. Trade provided
stimulus and incentive to merchants of different personalities
for participation in it and settling down at vantage points on the
trade routes. This afforded opportunities not only for the
people of northern and southern regions in Central Asia but
also of other nationalities in fostering a bond of amity and
understanding.
Along with trade and commerce, missionary activity and cul
tural transmission followed Buddhism which spread through its
missionaries, and was successful in carrying the message of peace
and understanding along with the sermons and scriptures of the
Lord. Buddhism, and with it Indian culture, finally succeeded
in establishing its firm hold throughout Central Asia, as is evi
dent from archaeological finds and traces of hundreds of
Buddhist shrines, stupas and monasteries all along the two
routes. Reference has already been made to the finds of Buddhist
texts as well as to secular writings. Besides these, hundreds of
documents from Loulan, Endere and Niya on the southern
routes in Kharosthi shed light on different aspects of material
culture of this part of Central Asia. Excavations, explorations
and finds of literary texts and documents from sites on the
northern routes are equally helpful in presenting a faithful
picture of the cultural life of the people and impact of Indian
Material Culture 227
culture on the local population. The accounts of the Chinese

travellers who passed through Central Asia on their way to
India in search of manuscripts equally furnish interesting
details.
Cultural Integration
Both Buddhism and trade relations contributed towards the
integration of social forces in Central Asia to a considerable
extent. The plurality of local cultures consequent to geographi
cal factors did stay, but the plantation of Indian colonies and
that of others on the trade routes necessarily set the process of
fusion into operation. Buddhism and its introduction, however,
accelerated the process. There are traces of Indian names in
records and manuscripts both in the northern and in the southern
areas of exploration. In this context the role of the Buddhist
missionaries1 was effective in converting the local population
1. The role of the Buddhist missionaries in Central Asia and China is
traced in M. Anesaki’s article in the Encyclopedia o f Religion and Ethics, Vol.
X, pp.700-704. It is proposed that the spread of Buddhist missions in Central
Asia was close on the heels of Asoka’s missionary enterprises. Historical
records, however, agree in assigning to A.D. 67 the first official introduction of
Buddhism into China with the coming of Kasyata Matanya and Dharmaraksa
along with Buddhist statues and scriptures to China. Conversions en masse
are said to have taken place in 71, many nobles and Taoist priests being
among the converts. The new religion was received with open arms and heart;
the way must have been long prepared for it. There is a gap of about eighty
years between the mention of the first missionary and the advent of two other
monks, one of whom was Shib-Kao of Parthia who came to China in 148.
He is said to have been of royal blood and to have left his country because of
the fall or decline of his family. This is one of the evidences that Buddhism
had a strong foothold in Parthia and Central Asia. The Sogdians—a branch
of the ancient Iranian people living inSamarkand and Bokhara— wereequally
propagators of Buddhism in Central Asia and China, and Seng-hui (Sangha-
bhadra) was a great Sogdian Buddhist scholar. Reference has already been
made to the role of the Indian Buddhist scholars from Kashmir. Kumarajiva
son of Kumarayana and the Kuchean princess nun—Jiva symbolises the spirit
of cultural collaboration and synthesis between India and Central Asia. San-
ghabhQti, Gautama Sahghadeva, Punyatrata and his pupil DharmayaSas are
some of the other scholars who contributed to Buddhism and its cultural ethos
in Central Asia and from there to China. The glorious days of Buddhism
seem to have ended about the beginning of the eleventh century when the
religion of the Tathagata was in a tottering state in the land of its birth. Per
sonal mysticism had weakened the community and the brethren were no
to a considerable extent- The Chinese pilgrims who passed
through these routes between the fourth and the seventh cen
turies A.D. have noticed this fact and have recorded domination
of Buddhist thought and way of life, sometimes in totality at
some of the places. There are also references to rulers bearing
Indian names in both the sectors. The Tibetan Annals,*2 how
ever, record introduction of Buddhism in Khotan much earlier
than is generally supposed. These record foundation of this
kingdom to a son of Asoka—born there—Ku-stana. Buddhism
was first introduced in Khotan during the reign of Vijayasam-
bhava, the grandson of the founder. All the successive rulers in
the Annals have the names beginning with Vijaya, such as
Vijayavira, -jaya, -dharma, -simha, -klrti, -samgrama, -bala,
-vikrama and so on. The Annals also speak of the Buddhist
foundations attributed to various rulers.
longer bound by any unbreakable tie. The internationalism of Buddhism

had lost its raison-d’etre with the decline of Buddhism in India and Buddhist
missionary forces ceased to operate. (See also Bagchi—India and China—
Chapter II on Buddhist Missionaries of India to China, pp. 28 IT.)
2. The history of Khotan is traced inRockhills—Life o f the Buddha—
London 1884 pp. 230fF; by Abee Remusat in his Histoire de la ville de Khotan;
and also by A. Stein—in his Ancient Khotan (Oxford. 1907). The Tibetan
Annals include Li-yul-gyi-lo-rgyus-pas, Li-yul-lung bstan-pa, Dgra-bchom-
pa-dge-hdun-hphelgyi-lung-bstan-pa—quoted by Stein (and noticed by Bagchi
— Op. c/fpp.51 ff). According to Thomas, the first introduction of Buddhism
is attributed to a sramana by name Vairocana, who by a miracle was over
the king Vijaya-Sambhava and under whose influence the first monastery in
the country, that of Tsar-ma, was built. He agrees with Spect and Levi (JA.
1896-97.1. pp. 14 Sq.; II. pp. 166 Sq.) in rendering c. 60 B.C. for the first
appearance of the Buddhist religion in Tibet. He further poses the question :
Can one point to any specific matters of Indian culture which may have come
to Khotan during its pre-Buddhist Period ? He suggests that certain Greek
terms occurring in Shan-Shan inscriptions, viz. Sadera and trakhma—Stater
and drachme, barracks —parambula—may have been brought from the Indian
side. Further the division of the country into parishes (simd) and further into
hundreds (sata), as also the ornate, formal, epistolary style noticed in ins
criptions is certainly Indian, and might have developed into Mauryan
Chancellories. Further, the terms lekha—letter, lekha-hciraka—letter carrier,
pothi— book noticed in records are Indian and so also nagarka—the town—
mayor. (Indianism and its Expansion—Calcutta University Readership
Lectures—Calcutta. 1942. pp. 60-61).
In analogy to the Indian rulers at Khotan, those at Kucha3

on the northern route include Haripuspa, Suvarnapuspa,
Haradeva, Suvarnadeva and others. The famous Buddhist
savant of Central Asia Kumarajlva also belonged to this place.
His father Kumarayana, a great scholar, came to this place from
Kashmir and had married Jiva, a royal princess. Another Indian
colony on the northern route was at Karasahr—Karashar
(Agnidesa) with its rulers bearing Indian names like Indrarjuna
and Chandrajuna. Other important personalities noticed in
manuscripts or in fragments include Yasomitra, associated with
a Kuchan monastery and one of the writers of the medical texts
contained in the famous Bower manuscript.4 The Indian mis
sionaries included Gautama Sanghadeva who arrived from
Kabul in 383; Buddhayasa, a Kashmiri Brahmin whocollabora-
tcd with Kumarajiva; Buddhabhadra, a Sakya from Kapila-
vastu; Dharmaraksa, a Mahayanistic monk from Kashmir, who
arrived in Kucha and then went to the court of a local king of
Kan-su; Jivagupta (528-605), born of a Ksatriya family in
Gandhara; and Dharmagupta from South India who spent a
couple of years in Kashgar, Karashar and Turfan and finally
went to Lo-Yang in China. The process of Indian missionaries
coming to Central Asia and then moving on to China continued
for a long time.
The Sung Annals5 mention by name several Buddhist monks
3. The Chinese accounts name the ruling dynasty of Kucha as ‘Po’ mean
ing ‘white’—probably to distinguish the rulers and their family from others.
The old Kuchean documents mention the Sanskritised names of the rulers
in the Tang period, like king Swarnate (Suvarnadatta), transformed into
Chinese as Su-fa-tie who ruled in the first quarter of the 7th century. His
brother’s name was Ho-li-pu-she-pi (Haripu$pa), while that of his father
was Su-fa-pu-kiue (Svarna or Suvarna-). The names of private persons had
also been Sanskritised in the Karasahr area in this period, such as Wiryamitre
(Viryamitra), Wiryasene (Viryasena), Jiianasena, Moksacandra etc. (Bagclii—
India & Central Asia—Op. cit, p. 79).
4. The Bower Manuscript is supposed to be a very interesting document
of the Indian cultural heritage in Central Asia. It contains seven medical texts
written by four different persons, all obviously residents of a Kuchan monas
tery. It was in the memorial stQpa of its last owner, Yasomati by name.
According to Sten Konow, Sanskrit played the same role in Khotanese medi
cine as Latin in European. (Aalto : “On the spread of Central Asia in the
spread of Indian cultural Influence’— Vivekanand Volume, pp. 251-52).
5. Aalto : ibid, p. 254. A detailed account of the life and activities of the
Buddhist missionaries who went to China from India through Central Asia
who around 1000 A.D. arrived in Kaifeng either from India or
from Khotan, Turfan or Kucha, bringing with them holy relics
and sculptures. The Indian Buddhist scholars were not inactive
in the south-western sector of Western Asia. An Indian mission
group headed by Prabhakaramitra is said to have converted the
Yabu of the western Turks around A.D. 622- Earlier, according
to the Chinese sources, a Chinese monk had persuaded the
Quran of the Eastern Turks To-Po (572-591) that the greatness
of China was based on Buddhism. The Quran founded a saiigh-
aramaand received from the emperor Buddhist Sutras translated
into Turkic by Liu-Che-tsing.*6 Jivagupta, the famous Buddhist
scholar, lived for ten years in the Turki court until A.D. 575.
Later Bilga Quran around 720 planned to build a Buddhist
temple but he was dissuaded by his prime minister Tonyukuk
who pointed out that Buddhism did not promote warlike spirits
in the people.
These references to Indian regnal names and those of Buddhist
scholar missionaries in different sectors in Central Asia suggest
that Buddhism played quite an effective role in bringing about
cultural integration among the peoples of Central Asia. The
royal patronage extended to this religion and its propagation
did bring about cultural evolution in the life style of the people.
It was not a one-way traffic as would be evident from a study
of the data provided by the Kharosthi records from Lou-lan,
Endere and Niya, mentioning Indian names, mixed names and
indigenous ones7—all forming part of the native cultural ethos.
is recorded in the earlier chapter dealing with Buddhism and Buddhist Litera
ture.
6. BEFEO-1905. p. 253; JA. 1895. p. 355, quoted by Aalto—ibid.
p. 251.
7. Aurel Stein in his note on the Kharosthi documents from the Lou-lan
site, notices similarity of names with those in the Niya Series. Just as in the
Niya ones, numerous names of unmistakably Buddhist or Indian derivation,
such as Anandasena, Bhatisama, Bhimaya, Buddhamitra, Dhammapala,
Kumudvatl, Pumnadeva, Caraka, Rutra, Sujada, Vasudeva, are noticed side
by side with others which seem of local origin eg. Cauleya, Cuvalayina, Kap-
geya, Kalpisa, Kipsa, Kitsaitsa, Lampurta, Maldraya, Porbhaya, Pulkaya,
Signaya, Tasuca, Tameca, Varpeya. Besides names, the official titles of Cojhbo,
Gusura, Kori, Vasu are common to both Lou-lan and Niya records. Peculiar
ities of style, phonetics and spelling have no doubt as to identical standards
There was stamp of Indian thought and way of life quite per
ceptible from these records in matters of family life, position of
women, professional avocations, and other items of material
culture. The academic side and attainments as also the relations
between the individual and the State are some other areas
where one could trace Indian impact. The Indian names appear
ing in records are Anandamitra, Buddhamitra, Dharmapala,
Punyadeva, Vasudeva. Epic names like Arjuna, Bhimasena,
figure in several records and suggest familiarity with the Maha-
bharata story. The Ramayana lengend*8 in Khotan and its
popularity is equally well-known. The documents from Niya
and Lou-lan suggest the co-existence of Indians as well as the
natives in these areas. Thus, one finds names of unmistakably
Buddhist or Indian derivation, such as Bhatisama, Bhumaya,
Budharaitra, Dhammapala, Kumudvati, Pumnadeva, Caraka,
Rutra, Sujada, Vasudeva, occurring side by side with others,
seemingly of local origin, like Cauleya, Cuvalayina, Kapgeya,
Kalpisa, Kipsa,Kitsailsa, Lamputra, Maldraya, Porbhaya, Pul-
kaya, Signaya, Tasuka, Tameca, Varpeya. The official titles of
Cojhbo, Gusura, Kuri, Vasu are common to both Lou-lan and
Niya records. These are local ones in contrast to the Indian
titles like maharayasa rayatirasa Skt. maharajasa rajatirajasa,
and avijidasimhasya Skt. avijitasimhasya and devaputra.9 These
seem to be based on the appellations used by the Kusana
monarchs. The association of names—Indian and native—in
records suggest their active participation in administrative
functions. Thus, one inscription (318) mentions Cozbo Indra-
been followed by the chancelleries, from Khotan to Lop at the period to which
these records belong (Serirtdia.l.p.414).
8. The Ramayana legend in Khotan is discussed in detail by H.W. Bailey
in the Journal of the American Oriental Society (JAOS. LIX.pp 460-8) as
also in the Bulletin of the School of Oriental and African Studies (BSOAS)
1940-42 pp. 365-76; pp. 559-605.
9. See nos. 204,209.332. The term Cozbo occurs in the maximum number
of records—about forty. The imperial title—maharajasa rajatirajasa maha-
nuava rajadevaputrasa is accorded to Raja Tajaka (no. 422); and Raja Mihira
(500), and Raja Anguvaka carries only the titles of mahanuava and devaputra
(572). These may be compared with the same set of titles used by Kani?ka
and his successors.
sena and Kirtisama, with the ogus10 Asuraga, Piteya, Rohana,

Camkura, Purnadeva, and Cozbo Mitrapala who heard the case
of theft. Apparently the names are suggestive of both indigenous
people and Indians who must have settled down there for long,
participating together in sharing administrative responsibilities.
Another inscription (418) records a priest Sariputra receiving an
adopted child from Dewga Amto—named Sirsateyae—and
marrying her to the priest Buddhavarma in lawful marriage.
The daughter from this lady was given in marriage to the priest
Jivalo Athane. Reference to three generations of cross relations
in this record reveals many interesting features of social life.
These may now be viewed in the context of family life, marriage,
position of women, social gradations, pastime and recreations,
food, dress and ornaments, agriculture and industries, pastoral
avocations, weights and measures, communications, labour and
finally, a review of material culture.
Family Life
The family, as an institution, is found in all human societies.
There could be differences over its constitution and functions,
but its general concept is a widely accepted phenomenon. In the
modern western concept, a married couple and their own
children—socially recognised, form the family group. A joint
family, according to oriental standard and norms, however, con
sists of two or more lineally related kinsfolk of the same
sex, their spouses and offspring occupying a single home
stead and jointly subject to the same authority or single head.
It is a cohesive unit forming a solid group. The evidence provi
ded by the records from Central Asia is suggestive of the
traditional joint-family system prevailing in this area. The
family included father, mother, brothers and younger sisters
(Kudi)(Nos. 164, 195) with the headman exercising authority
over them (450, 562). Terms used are primarily Indian like
10. Ogus, Gusura, Kala and Cankura are prefixed with the names, pro
bably suggesting their position in administrative hierarchy. It is, however,
not possible to determine their social or administrative status, or assess their
position. The Cozbo Sanjaka in one record (272) is enjoined to finish his
administrative work at full speed, and anybody interfering in his functioning
was to be punished.
matu, madu for mother, putra and suta for son, pitumaha for
grand-father, prapotra, prautra for grandson; napata for
daughter’s son; hhrata, bhratu for brother; brhrata-putra
for brother’s son; jamdta for son-in-law, svasu for sister,
bhdryd for wife ;putri and dhitu for daughter. Kula and parivara,
suggesting family as a unit, are also noticed. One inscription
refers to Visatitaga Skt. Visa—suggesting a family. There is
evidence of familiarity and closeness among kinsmen and re
lations. Letters to near and dear ones11 are addressed in familiar
tones, informative in nature particularly concerning the family
problems. It was natural to be communicative with brothers on
domestic problems. A Buddhist monk addresses his brother
in an affectionate manner (646) without alienating himself
completely from his family. Two letters (499, 612) addressed to
friends are in sincere and well-wishing tone (kalyanakari),
soliciting an early reply. A lady communicates with her sister
(No. 316) pointing to literacy among upper strata of society and
addresses her as a pleasing personality (priyadarsini). Letters
are addressed to daughter and son-in-law (690), wife’s brother
(syala) (140, 475), conveying good and bad news.
The head of the family was enjoined to restrain the other
members from raising any settled issue in future, failing which
he had to renounce his control over them (No. 621). He had to
look after their safety and see them settled down comfortably
elsewhere in period of emergency or distress (No. 362). There
could be addition to the family through adoption as also
through purchase of slaves who were also a part of the family.
Children—both boys and girls (nos. 331, 542 etc)—could be
given in adoption after payment of proper consideration
(Kutichara) in cash or in kind (11, 31). Even elderly ladies
could be accepted in adoption (528) and provision was made
for their share in inheritance. The birth of a son was an occasion
for rejoicing in the family (no. 702), that of a girl, of course,
was not one of distress or depression, since she could as well be
helpful to the family in future. Several terms used for girls
11. The list of such letters is very long (106, 109, 435, 476, 512 etc.) and
primarily concerned with domestic affairs or personal matters involving the
communicating parties (nos. 139, 152, 157, 499, 519 etc). They are addressed
in a very affectionate manner, signifying close family ties.
and ladies suggest their position in society. A young girl is called

Kudi and when grown up is known as dhitu (279), in Sanskrit
duhitri or ghihare (no. 46) grihani. After marriage she is called
bharya and matu—Skt. matri when she has children. A sister is
known as Svasu—Skt. Svafru and maternal aunt—mahula—
Skt. matull. Certain other terms associated with ladies denote
their status and position in society like vesitfrl (709), probably
vesya—a public woman and khakhonistrl or khasorna (Nos. 58,
63, 248) identified with svasurani, aniti or anitd probably
synonymous with bharya and finally dasi (621) signifying
female slaves. Some insteresting information about the position
of ladies is available, as for instance, the father’s responsibility to
marry his daughter for which he could demand bride’s price
called lote, from the son-in-law.12 He could legally claim it, but
not if the girl had run away with her suitor (No. 621). As part
of the family, she had no doubt her position, but being treated
as the property of her husband she could be given in exchange,
or be a salable commodity and her value was determined on the
basis of her height and in cash or in kind (nos. 587, 437) etc.
As part of the sale transaction or gift (no. 380), or in exchange
(no. 551), as also for payment of debt (no. 114), she could be
transferred to the other party without hindrance. Women could
amass money over which they had full rights and on their death
it was to be shared equally among their children (no. 474). Girls
were generally married to persons of the same avocational
group. A Buddhist monk married his daughter to another monk
rather legally (no. 18, 474), but another such girl defying her
father eloped with a potter’s son (no. 621). Marital relations
among close relatives were not unknown and sometimes these
were on a mutual exchange basis (vinimaya).13 This was fairly
12. The demand for the bride’s price seems to have an accepted principle
in society. In the Ar?a form of marriage, mentioned in the Hindu Dharma-
Sastras, the bridegroom had to provide a pair of oxen. On the other hand, in
the Brahma form the father-in-law catered to the needs of the daughter and the
bridegroom. In China, the acceptance of any bride’s price was banned by the
Tang Emperor Kao-Sung, but if such money was accepted , it was to be spent
on the bride’s necessities alone. (For a detailed account of marriage in different
countries see Encyclopedia o f Religion and Ethics, Vol. 8 pp. 423 ff.)
13. In the first century A.D., A Huna noble Jiji had married the daughter
of Kanga-gu and in exchange had given his daughter to his father-in-law.
common among the Hunas probably more for political reasons.

Communication difficulties encouraged near distance or rather
marriages in the same locality with the mutual offering of girls in
exchange (nos. 279, 481). The status of a woman in the family
was rather limited. She could not participate in social and public
activities. Her husband had full right over her person—from use
of violence (no. 279) to pledging her or disposing her in satis
faction of debt clearance (no. 719). Her value was assessed in
terms ofrolls of silken cloth (no. 3), the price of a camel (nos.
209, 578), carpets of different sizes and even household goods
(no. 706). She along with her children could be the subject of
robbery and easily disposable (no. 415). Dissolution of marriage
(no. 34, 621) was possible in cases not covered by monetary
payment to the girl’s father. A term vivega14—meaning probably
‘separation’ is recorded in an inscription (no. 34). Ladies of the
upper strata in society seem to have been better off—educated
and capable of free communication with friends and relations
(no. 316). They could as well send presents. Administration no
doubt accepted their credentials and they could be called for
evidence in legal matters (nos. 3, 420).
Slaves—male and female—as well formed part of the family.
Several terms are used for such slaves like dasa (nos. 345, 491),
dasi (no. 621), dajha (no. 569), dhaja (no. 227), dhajhi (no. 39,
45) with the master known as bhatare (no. 147), bhataraga^ySkt.
bhatfaraka. A servant is called presi (no. 204) who seems to be
different from dasa or slave. Another term for a domestic emp
loyee is vathayaga— vadhaya (no. 118) equated with Khotan-
ese vaksayalb and Tokharian upasthayaka. Some records (no. 19,
54, 401) mention his scale of emoluments in terms of salary and
perquisites—parikraya bhojana(pacevara)—food, and clothes (co-*145
(McGovem : Early Empires o f Central Asia p.190). An inscription (no. 279)
notices a girl of Yave-avana being married to a native of Ajiyama—Avana,
on an exchange basis. A similar case is recorded in another inscription (no.
481). For discussion on the terms lote and mukesa, see BSOAS. 6. pp. 523 If.
14. Burrow. Language o f the Kharofthi documents from Chinese Turkestan,
Cambridge, 1937, p. 116.
15. Burrow derives Vatayaga or Vathayaga from upasyaka, and it was
accepted in Khotanese as vathiaya (Languages—Op. cit, p. 118). In Hindi it
is accepted as bhritya or anucara. Bailey equates Khotanese vadhaya with
vakfaya, or vaghayarha (BSOAS. Vol. II, p. 791).
daga), and he was required to look after the grazing cattle. The
slaves received only food and clothing. A record (no. 25) men
tions 3 milim rice and another (no. 470) 10 khi rice as the
remuneration for such a servant. Carriers on their back pritha-
bhdrika > Skt. pristhabhdrika (no.396) transported short distance
goods. The slaves and servants looked after the farms of their
masters as also their household. The slaves were bonded for a
specified period like ten years (no. 550) or twelve years (no. 364).
The breach of agreement was punishable (no. 764). Slaves could
even otherwise be punished for theft (no. 518) or stealing of clo
thes, cattles etc. (345,561). Sometimes the punishment inflicted
was exemplary and caused their death (no. 144). Abduction of
slaves was not an unknown phenomenon (no. 36, 324,491). One
who returned to his original master did not involve any payment
by the abductor to the former lord, otherwise money payment
(Jote) was necessary. Slaves were marketable as well as presen
table commodities (nos. 491, 324). A couple of inscriptions
record sale of girls and men ( prusdhaya) transferring ownership
as also the right to sell, pledge, exchange and present the slaves
(nos. 589, 590, 591). The master was liable for the lapses of his
slave, and for theft he had to make good the loss involved (nos.
345,561). The slaves could adopt someone only with the consent
of their master. The longer stay at a farm away from the mas
ters home occasionally generated a spirit of defiance and an urge
for independence in a slave. He could, however, purchase his
freedom by paying back lote and mukesi to his master (no. 585).
A benevolent master sometimes transferred his interest in the
farm in favour of his favourite slave (no. 36). This concession
enabled the slaves to build up some strength in terms of money
and cattles as also clothes (nos. 24, 327). The master could not
illegally appropriate his slave’s earned property (no. 24), nor
could this be executed in payment of debt (no. 49). The monks
also kept slaves to look after their farms and property interests
(no. 152) and there could be change of masters.
Food & Food Habits
.The records provide interesting information regarding items
of food and food habits. The village economy was fairly deve
loped and the agriculturists fairly experienced. Wheat, rice
and corn were produced as items for staple food. The term
generally used for these products is pacevara. Flour (ata-ata and
iaktu) (no. 359) and rice (dhanya) were commonly used. One
record (no- 359) notices provisions consisting of 2 milima, 15 khi
of meat and 5 khi of maka and one vest (Kavasti). Condiments
and saucy stuffs made food delicious with the addition of pepper
(marica), ginger (orakhing), arakhima of pepper ( pipali-pipala),
svaca and cardamons (susmela) (no. 702), milk, sugar (,sarkara)
and ghee (,ghridra) are also noticed in records (nos. 13, 15). The
nonvegetarian food was far nutritious than the vegetarian one
(no. 514). Jars of ghee and a hundred jars of oil are no match,
according to the author of this record, to a sixteenth of one piece
of meat. Food was meant to sustain one’s body and maintain
his existence (nisaganam) (nos. 478, 641). Provision was made
for soldiers in the capital in the form of corn and sheep (no.
478). Spirituous liquor was in common use as is recorded in
several inscriptions (nos. 175, 244, 317, 329, 343), and the state
administration realized cess on its production and sale. A record
(no. 175) refers to the supply of old wine to the ruler and the
people taking only three khi of ordinary drink. This might have
been the restricted limit. Another record mentions cultivation
of grape vineyard, but there is no reference to the wine shops
for ths supply of drinks or bars. The wine supplied to the ruler
had to be sealed to ensure its genuineness (no. 247).
Dress and Ornaments
The dress of the Central Asians conformed to the climatic
conditions, as also to the impact of Indian as well as Chinese cul
tures. Some information on this aspect of material culture is
provided by the accounts of the Chinese pilgrims who passed
through this region on their way to India. The data from the
records under consideration as also the sculptures and paintings
from different sites have much to reveal. These people, accord
ing to Hsuang-Chuang, used clothes of wood and fur, as also
silk and cotton.16 The ladies’ dress consisted of a trouser and a
petty-coat type of covering pamzavanta made of prigha—a kind
of silk (no. 316). Another inscription (no. 318) mentions an
. 16. Stein—Ancient Khotan {op. cit) p. 139.

embroidered vitfapa—a jacket made of white silk, a samimna, a

lyokmanaof many colours, a yellow coloured Kharavarana, a pato-
varnaga garment, golden dave, a varsara, fine hasta of woollen cloth
(Umna = Orna),bluedyed kigin. Cotaga or codaga was the coat or
the upper covering, since the term is used with reference to males
as well as females (no. 19, 506). Another inscription (505) men
tions cafoga which is another form of cotaga and is translated
as ‘cloth’ which was universally used. Several other terms sug
gestive of items of clothing are kavasi, kuhchali, kacavadha, candri.
Kavasi appears to be inner clothing—same as kavaji, while kahcull
could be equated with the Sanskrit kahculika meant to cover the
upper part of the ladies’ body. This was made of silk (kauseya) as
well as of wool (urna) and was dyed as well (no. 318). A small
jacket was known as kahcabandha (no. 149). This inscription
also makes reference to roughly woven clothes, woollen clothes,
two jackets, two sanstamni (?), two belts and three Chinese robes
in some context. The head dress is suggested by a term Clnabeda
(no. 353)—a kind of Chinese head covering. Two inscriptions
(nos. 272, 714) mention a term candri-kammata, which, accord
ing to Bailey,17 is suggestive of some sort of trouser made from a
sheet of cloth. Others take it to signify some tax or cess or jade.
The clna-clmara (no. 149) ->-Skt. Clna-clvara-is suggestive of
Chinese overcoat, although clvara is a term used for the upper
covering of Buddhist monk. Several other terms, probably con
nected with dress, are noticed in inscriptions (no. 714) like kaci-
kumufa Khotanese komadayi—a form of trouser; puchama (534)—
some sort of cotton night dress, pasamvamta (no. 534) made of
cloth; lastuga (nos. 144, 298, 566) made of many-coloured silk—
probably some kind of scarf. The Indian blanket (kambdla) is
noticed as loyi—Hindi Loyl, and also as kojava (no. 583, 593,
599). An inscription (660) refers to rolls of silk.
Some idea of the dress and its style can be had from the Cent
ral Asian paintings as well with the male and female donors and
devotees dressed in their traditional costumes and head dress.
Thus, in a painting of the Parinirvana—‘the dying Buddha scene’
at Bezeklik18—a Chinese dignitary wears the head dress of a judge,
17. BSOAS. Vol. XI. p. 793; cf. Burrow. Op. cit. p. 143.
18. F.H. Andrews. Wall Paintings from Ancient Shrines in Central Asia
(henceforth—Andrews. Paintings.) p. 75.
another Chinese has a broad-brimmed conical hat, while an

elderly person puts on a loosely tied white turban (pagrl). The
same scene depicts an unmistakable Persian figure with grotesque
ly long and bulbous nose, full beard and flowing moustache,
wearing the typical wing-shape Persian hat. Similarly (the figures)
in the pictorial remains from the caves of the thousand Buddhas,
the head gear and robes of the donors and donatries suggest
‘Chinese translation of the quasi-secular forces’.19 In a scene
from Kizil2.0 (7th century) showing the preaching Buddha (17,
850), a lady sits on a low chair, her hands in the anjali mudra.
She wears a white blouse with blue spots and a green under-gar
ment. The young monk behind her puts on his brown robe.
The main group in the foreground consists of a man laden with
jewellery, and wearing only short dhoti reaching to his thighs, and
to his right a woman dressed in a green jacket with fancy embroi
dery and.wide brown oversleeves under which green borders are
noticeable. Her blue speckled skirt reaches down to her ankles
and the upper part of the body is covered with a short piece of
white cloth. Similarly Vajrapapi from Kizil21 is shown elabora
tely draped in chains and strings of beads or metal disks. He
wears a brown skirt with ruffled green borders, and the ends of
long green scarves suspended from his head dress almost touch
the ground.
Portrayals of princes and their families are an interesting
feature of the art of Central Asia, and these also provide an
insight into the aristocratic life style. The finest painting of a
Uighurian prince22 on cloth, now in the Berlin Museum, depicts
him dressed in a fine robe, patterned with a large floral design.
It is a long-sleeved, round-necked garment reaching to the feet
with a belt fastened in front, with rectangular decorative pieces
containing a slit from which to hang straps with objects of daily
use. A slit in the side of the garment reveals a black knee-length
boot. On his head the prince puts on a three-pronged cap, held
by a strap under the chin, with a shoulder-length veil hanging
19. Stein— Serindia• Vol. II. p. 850.

20. Martin Lerner. Along the Ancient Silk Routes— Central Asian Art
(New York. 1982) p. 96.
21. ibid. p. 68.
22. Bussaghli. Paintings o f Central Asia, p. 106.
down behind. Along with an attractively curled moustache, he

has a beard covering his chin and framing his cheeks.
The Miran paintings—in all probability Indian in conception
and execution—depict Indian garments23 and bare feet. They
wear loin-cloths {dhoti) gathered about their legs, and stoles
draped over the shoulders. The drapery is shaded or contoured
with a suitable colour; red with black lines, green with dark
grey; white with pale grey, and yellow with red. The dhoti is
traced in the painting from Toyuk24 near Turfan in the northern
part of Central Asia. The costumes are of two kinds; one with
a long dhoti covering the legs to the ankles; the other short,
leaving the legs bare. A loin-cloth is usual and a kind of upavlta
or narrow shawl crosses from the left shoulder and passes round
the body below the left arm. A stole, passing over the shoulders
and winding round the arms, with freely flowing ends, seems
common to all. Most of the figures wear a simple headdress
(muku(a). In the Turfan area costumes are either the loose
draperies customary in India, or elaborately ‘tailored’ garments25
conforming to fashions of the intruding communities. The use
of sandals or shoes of more or less elaborate fashion is a notable
feature. The highly decorative shoes, an imbricated pattern in
red on a white ground, may be a form of plaiting or quilting,
are noticed in Bezeklik paintings.26
Ornamentation of the body for both male and female is notic
ed in paintings and a few Kharosthi inscriptions from Lou-lan,
Niya and Endere also mention some items. Thus, one record
(no. 566) mentions seveu-stringed pearl/ornament (mutilafa) and
one ear pendant (sudl). Reference to a silver ornament is record
ed in another inscription (no. 149). While the costumes are sim
ple and there is absence of jewellery in Miran paintings, in the
later ones, are noticed different kinds of jewellery in general use.
These include armlets, anklets, bracelets, ear-rings, garters,
mukuta, carcanet, armlets and bangles, adorning the male figures
along with pearl studded garlands or necklets. A prince in a
23. Andrews. Op. cit. p. XXII, also p. 8.

24. ibid, p. 42.
25. ibid, p. xxvi.
26. ibid, pp. 58, 60.
Bezeklik painting27 has the long, slit lobes of his ears carrying
gold rosettes from which hang bunches of five coloured beads.
His necklet is of chased gold and red in alternate sections, with
red and green beads. Beads in bunches are also attached to arm-
lets, and to ear-rings. The mukuta or tiara also presents a pro
minent look with its varied compositions—crown like, jewelled
and decked with flowers28 or composed of a group of three gold
bosses at the centre, one at each side and a red and green palmet-
tes above.29 Coiffure and head-dress are varied and significant at
Bezeklik, floral and pearl studded.30
Pastime and Recreations
A sophisticated life style enjoins provision for recreation in a
dignified manner. It could be outdoor games or hunting exer-
sions, rather more masculine exercises or indoor gatherings with
music, dancing and other items of recreations. A solitary
inscription (no. 13) enjoins stoppage of wounding mares and
horses in hunting. This negative approach in the form of issuing
injunctions against slaughter of useful animals could imply
people’s interest in hunting. The painting scenes,31 no doubt,
depict some of the animals, as for instance, lions, wild goats—
their horns as votive offerings, elephants, bulls (139), camels—
noticed in inscriptions as well. In the famous cowherd Nanda
scene from Kizil32, two cattle, one dark in colour and the other
white, lie on the ground, while in another painting of seated
Vajrapani from the same cave one notices two sharp-beaked
falcon-like birds. A hunting scene is depicted in a clay sealing
from Farhad — Beg-Yailaki with a man on horseback galloping
27. ibid, Bez. iii. A.B. p. 71.

28. ibid, pp. 31, 55, 58.
29. ibid, p. 72.
30. ibid, pp. 52, 84, 88. Stein notices the headdress of a princess in the
mural painting in Celia V at Miran. Her hair descends in black tresses below
the neck, with love-locks in front of the ears and two fringes crossing the fore
head. In another figure two strings of red beads crossing the hair obliquely
are fastened with a large circular jewel ornament above the middle of the
forehead. (Serindia, Vol. I p. 519)
31. Stein. Serindia, pp. 506, 891, 950, 943, 1024 Sq. etc. for reference to
these animals in painting.
32. Lerner. Along the Silk Route—Op. cit, p. 67; Bussagli. Op. cit p. 72
(HI.)
with a sword in his uplifted right hand. A lioness is shown

behind the horse and a goat or deer is shown running.33 The
swimmer’s scene34 from Kizil shows persons in some pool among
water lilies, while the middle one seems barely able to keep his <
head above water; the man on the right swims with powerful
strokes. His expression and gestures reveal great determination
and skill in swimming.
Khotan and Kucha were noted for their music, as is evident
from certain paintings from these regions, as also from the
account of the Chinese pilgrim Hsuan-tsang. This Chinese pil
grim records that the green-eyed people of Khotan were very
fond of music and dancing. In the famous ‘attainment of Pari-
nirvaija’ painting35 from Bezeklik, four musicians contribute to
the sounds of lamentation, with professional enthusiasm and
energy. The drummer holds the drum under his right arm and
beats it with a ball-headed, stick. To his left the cymbal player
clashes his box-like instrument above his head. Another, whose
head is missing, plays a flute with the left hand fingering the
notes while the last one—the white-haired biwa player uses the
classical plectrum. The animated poses of the legs suggest time
beating. The divine musician from Kizil38 suggests the continua
tion of the musical traditions at Kucha which was the centre of
Indian influenced music in Central Asia from where it spread to
China. It is reported that when Chinese rule was extended to
this region in 382, the General Protector Lu-Kuang brought
groups of performers—actors, musicians and dancers—back to
China as booty. Among the instruments used were the five
stringed lute, the harp, the mouth-organ, drums, cymbals, gongs,
and the four-stringed Kucha lute.37 In China one of the teachers
33. Stein. Serindia, p. 1257.
34. Lcrner. Op. cit. p. 75.
35. Andrews. Bez. xii. A.l.pl. XXVII. pp. 95 for a detailed account,
including reference to the Chinese pilgrim Hsuang Tsang’s account.
36. Lerner. Along the Silk Route—Op. cit. p. 93.
37. The musical traditions at Kucha are mentioned in early Chinese
records. Chang-Chien, whom emperor Wu-Ti sent to the West in 138 B.C.
is said to have brought back musical instruments and melodies from Kucha to
the capital Ch’ang-an (Liu-Mau-tsai. Kutscha and seine Beziehungen Zu
China—Wiesbaden 1969—quoted by Hertel—Along the Silk Route, p. 93).
of this lute was a brahmin, and another lute player Sujiva who
went to China in 568 bore the family name of Po of the royal
house of Kucha. It appears that along with Buddhism, Indian
music was brought to Kucha, an inference that is supported by
the many musical subjects depicted in the wall paintings. In this
scene, the fragment shows the head of a dark skinned divinity
playing the flute.
Agricultural Economy
The Central Asian socio-economic structure seems to be relat
ed to the land economy, cattle rearing and a few productive
avocations. The State was no doubt the owner of the land with
the upper class people (no. 120) enjoying it on payment of rent
in terms of produce. They cultivated this land with the help of
hired labourers and slaves. The existence of a slave-owning
society is fairly evident from numerous inscriptions recording
their status. Besides payment of salary, food and clothing were
also provided to the labourers (no. 25). Payment of wages was
necessary (no. 50) but there could be disputes as well. Slave
women too were not denied these perquisites though they
could be purchased. An inscription (no. 591) assesses a male
slave’s value (prusdhaya) in terms of a five year old camel, a
horse of the same age and 25 altga (not known). He could be
pledged, exchanged and utilised as desired by his master, but he
had certain rights as well, as for instance acquisition of property
through one’s savings (no. 671) and its alienation as well (no.
419). This right extended to female slaves as well. The use of
violence against working women (no. 20,29,53) and appropriation
of their land produce was not unknown (no. 36). The master
was not responsible for the debts of his slave (no. 24). There are
references to several kinds of lands. The arable one was called
mishi88 (no. 572) while the barren one was known as akri. Land
38. Misi, according to Aurel Stein, is some crop. He also notices the full
proprietory rights over this tnlsi— the full enjoyment of all its benefits in
whatever way the purchaser desires, whether for ploughing or sowing or for
giving to another as a gift or as a namanya (namaneya, tenancy ?). Further,
if at any subsequent time a vasu ageta (vasu, a common title; ageta also
apparently the title of some official) shall give any order concerning it, such
a verbal order shall be invalid at the king's court (Seritidia, Vol. I. p. 232).
left unploughed could be barren. The ploughing (krisitam—

krisnata) and sowing (bavitaga— vapanta) operations are re
corded (no. 320). While reaping and winnowing processes are
not mentioned in any record, the reference to barn-house
(gothada) in a record (no. 36) used for storage is interesting.
This belonged to a slave who had complained to the authorities
for theft. The term kurora (no. 514) is also noticed in connection
with land and is defined as ‘the earthern boundary line of the
ploughed land’. The amount of seed required in a measured
plot of land is also recorded in several inscriptions (nos. 422,
519, 580), and the price of the land was accordingly determined
and paid for in kind or in barter (no. 715). There is no reference
to cash payment. One inscription (no. 366) records collective
farming, since the land in dispute was returned to villagers.
Irrigation was equally important after the sowing of the seed,
and several records notice it. Tibetan documents from Chinese
Turkestan mention irrigation of fields by the labourer called
chun-pa. These documents also mention the ploughing of fields
and threshing of the grain, and announce punishment for those
who let the water dry up. An ordinary field labourer was called
zhing-pa,39 The wheat crop (goma—godhuma) needed at least
two-three waters (no. 72). Canals carried water to distant areas
urgently needing it and payment had to be made for it (no.
368). One inscription notices payment both for seeds as well as
for water (no. 160) which was not to be misappropriated or
allowed to go waste (no. 502). The land owner had to be careful
by not allowing the water to flood the adjacent land and the
farm house attached to it (nos. 47, 125). Seeds borrowed had to
be returned with interest (no. 140) or in double the quantity
(no. 142). Land disputes were fairly common (nos. 90,124).
There are also references to state farms (nos. 272, 278) and also
to the collection of revenue—one-third of the produce (no. 291)
through the state collecting agency (no. 198) every year (nos.
275, 295). There is a reference to interest on tax (no. 211).
Geographical factors with paucity of rainfall could no doubt
endanger agricultural economy resulting in famine, noticed in a
record (no. 581). The drought impelled its owner to sell his
39. Stein, ibid. p. 1464.

vineyard to the scribe just for a carpet (tavastaga) six feet long,
one kavajiy two sheep and one milima of corn. It appears to be
a case of exploitation in adverse circumstances. The ownership
of land was unfettered with the right to plough, to exchange, to
sell, to mortage it (nos. 586, 587). Slaves and women were not
denied this right to sell their land (nos. 574 & 677). There are
also references to land disputes which necessitated affirmation
on oath by witnesses on behalf of the contesting parties (nos. 90,
124). Sometimes land was put under the care of someone, pro
bably some minor official either in the case of disputed owner
ship or that of failure to pay state dues (no. 278).
There are also references to crops other than the cereal ones—
like growing of cotton, hemp, fruits and vegetables. Grape
vineyards were equally productive and remunerative. The major
portion of Central Asia being arid desert, it was only the hunted
area which could be utilised for cultivation and the steppes
could cater for grazing purposes. The Khotan area was produc
tive of cotton and hemp. Animal husbandry, closely connected
with pasture lands, equally engaged Central Asians in their eco
nomic activity. Among the animals some were useful for trans
port and others like sheep for the by-products. Camel or u{a is
mentioned in several records (nos. 4, 6, 10, 16 etc.). It was ex
tremely useful for transport of goods as also for conveyance of
traffic on hire (nos. 6, 16, 21). Heavy loading could sometimes
prove fatal for which the person engaging the animal on hire
was responsible (no. 52). Veterinary help was available during
transit (no. 40). Camels and horses were also used for military
transport (nos. 125, 367) and the state also looked after their
maintenance and upkeep. Camels also served as medium of
exchange and could be accepted in payment of taxes (nos. 589,
592, 715). Horses were used for ploughing as well as for trans
port and could be loaned as well (nos. 24, 119). They formed
items of presentation (no. 243) and were equally exchanged with
other animals. Sheep provided wool, while rams could be used
for carrying light goods (nos. 568, 633). Cows yielded milk as
well as ghee, and there are several references to presentation of
cows and also their sale (nos. 13, 122, 439, 514). The state cow-
pans were looked after by the keeper called gopalaka (no. 439).
ghrid—ghrita—ghee was made on a grand scale, probably under
state supervision. An inscription records the production of a

hundred jars (no. 574) which could be possible only in a co
operative or state enterprise, and its theft was not an unusual
feature (no. 15).
The grazing grounds called kabodha were outside the town
area and were generally enclosed (no. 392) while the open one
was called lathanam. One inscription (no. 55) records provision
to be made for the royal camel in all cities. This included fodder
and water. This might not have been on payment. The respon
sibility for the upkeep of animals was on their owners and
cruelty to animals was punishable. A record (no. 13) bans hunt
ing of the mares and horses. Cattle rearing seems to be encourag
ed because of the utility of animals in day to day life and eco
nomic necessities. These also provided avenues for ancilliary
avocations like those of blanket and carpet makers and of other
items. The sheep wool provided the raw material for rough
texture. Milk and its products like ghee were in great demand;
and this industry seems to have contributed quite a bit to the
economic life of the people. Hsuan-Tsang has referred to the
enormous cattle population in Khotan, especially camels.
Handicrafts and other Industries
There are references to several industrial arts and artists and
economic professions. The makers of gold and silver ornaments
were quite conspicuous (no. 578). Reference has no doubt been
made to different types of ornaments and beads found in Khotan
and at other places in Central Asia. The finds of wooden tab
lets—recorded and painted—are suggestive of workers in wood
as also painters and sculptors. Leather and its goods, especially
long boots, point to this professional class of workers. Weavers
and tailors looked after the cloth industry and the making of
garments. The silk industry was very prominent, and the famous
scene of a Chinese princess secretly bringing silk cocoons to this
region sheds light on the introduction40 and subsequent prospe
rity of this industry. One inscription notices coloured silk
40. Stein, Innermost Asia. pp. 471, 598; 232 Sq; The Chinese princess is
credited according to a legend reproduced by Hsuan-tsang, for the introduction
of sericulture into the kingdom of Khotan (Stein—Ancient Khotan. 1.pp.229
Sq; Beal. Si-Vu-Ki. II. pp. 318 Sq).
(no. 566) of which a lastuga-gown was made. The finds of

paintings on silk,41 as also silken fabrics with weaves and designs
and embroidered point to the skill of the silken weavers. This
industry has a long history and even the trade route between
China and Persia was known as the famous silk trade route.42
The local industries included manufacture of other necessary
items of use including utensils, shoes, comb and mirror which
are noticed in inscriptions and were also found in excavation.
Stein mentions several kinds of shoes made of leather, cotton,
wool and hemp—the prominent type being the long boots with
straps.43 The use of comb and mirror is evident from a record
(no. 566) as also from their finds of both.44 The Kharosthi
records mention a term tavastaga (no. 431) meaning'carpet*
while another term thavastaye (no. 714) is supposed to suggest a
cloth carpet.4546This part of Central Asia was noted for carpet
making40 and these were exported from there. Rolls of silk were
imported as well as exported (no. 660), with the Chinese traders
playing an important role (no. 35). Pafa is the word used for
silk which was used for making clothes, and its rolls also served
in barter transactions. One record (no. 3) values a woman’s price
in terms of 41 rolls of silk. Another record (no. 489) prescribes
the fine of a roll of silk as punishment for breach of monastic
order rules, like non-participation in the posatha or joining it in
the dress of a householder.
41. See Stein : Innermost Asia, pp. 471 ff; and also for references to
silken fabrics (ibid 232 Sq.), weaves and designs (ibid 673 Sq), and embroidery
(ibid 235, 489).
42. Ptolemy furnishes interesting information on the subject of silk trade
from China during the first century A.D. (Geographia. I. XI. 7; XII.8 Sq).
Stein records the importance of the route for trade during the centuries before
and after the beginning of the Christian era, when Baktra was a chief
emporium for the great silk trade passing from China to Persia and the
Mediterranean. It led from Kashgar to the Alai valley and thence down the
Kizil-su or Surkh-ab towards the Oxus. Nature favoured the use of this route,
since it crossed the watershed between the Tarim basin and the Oxus where
it was the lowest. (Stein. Innermost Asia. Vol. II. p. 848)
43. Serindia. II. pp. 704, 719, 813; Ancient Khotan, p. 297.
44. Serindia. pp. 943, 967, 721.
45. Bailey. BSOAS. II. p. 793.
46. McGovern. Early Empires—Op. cit, p. 53.
There were dealers in items of food and drinks. Several records

(nos. 175, 244, 317 etc.) notice wine and its trade with other
countries. The vineyards supplied grapes for processing and the
industry had the support of the administration. It was also
accepted by way of tax and collected by officials appointed for
this purpose (no. 206) who sometimes misappropriated the
accumulated tax (no. 272). There was a special department for
this old and new wine tax collection (no. 567). Another record
notices selling of wine (no. 247). As a medium of exchange
and barter, it could be used for payment of debt and also as the
price of some commodity (nos. 168, 244). Like the state distille
ries there were dairy farms as well which were looked after by
the administration. An inscription (no. 314) refers to state cows
probably suggesting public sector enterprises in ghee and oil.
An officer called ‘Satavida’ looked after this state enterprise and
he was expected to provide full account of its export. Sometimes
there was misappropriation of the state revenue. One record
(no. 621) notices kulala or kutala—a potter, who manufactured
clay utensils and tablets. Jars were made of clay and these were
used for storage purposes.
National Economy and Medium o f Exchange and Barter
The Central Asian economy seems to be properly planned,
with a fair amount of gold reserve. The yellow metal and its
sale is mentioned in a record (no. 140) and reference has been
made to the manufacture of gold ornaments. Another record
(no. 177) mentions presentation of gold. The reference to ‘Suva-
rna satera’ might be suggestive of gold coin equivalent in value
to Greek Strateros—Stater. While barter was fairly common,
the medium of exchange or even the collection of state dues could
be in the shape of animals and commodities. This might be a
primitive way of economic activity in the absence of coinage,
but it seems to have worked. One inscription (no. 586) records
despatch of cows in exchange for some commodities purchased.
Another record (no. 180) mentions 6 regal camels probably
received in trade transaction or in tax. There are many references
to land deals in terms of barter and exchange with productive
animals like cows, sheep, camels and horses. Some types of coins
were also in use as is evident from references in inscriptions
(nos. 43,324, 419, 431). These are sadera—stater, trasya, muli,

kampomasa, ghare, mamaka,47A sa/er and two trakhma—drachm
was the price paid for a slave (no. 324), while a single sater could
bring a vineyard (no. 419)—rather an unusual or a very cheap
deal, when another record (no. 702) notices sugar worth 4
sadera and pepper worth 2 trachma—drachme. These coins must
have reached Khotan area through trade transactions from the
western region. There is no reference to the ratio between gold
and silver coins but one of the Tibetan documents provides the
equation : \ zho of gold = 3 zho of silver.48 A Kharosthi record
(no. 43) mentions a term kampo along with the golden sadera,
probably suggestive of some kind of silver denomination. Several
records (nos. 500, 149) notice the term masa (masa)—masa of
the ancient Indian coinage which appears to be of a small de
nomination since it appears in big numbers as 8000, 2800, 2500
in inscriptions. Another term noticed in records is ghane
(no. 702) which might have been a quarter of drachm or four
times the value of a karsapana as supposed by some scholars.
Midi of the inscriptions (431, 579) might have been the local
form of mulya or value although its denominational value is
proposed as the twelfth part of a sadera-stater, since a carpet
of certain length is valued in both the denominations in two
records separately. Yet another term mamaka is supposed to
denote some coinage since Tita—Titus the painter was paid
300 in this denomination for his work.49 In the absence of
finds of local coinage, the information available from the Kharo
sthi records alone is considered in this context.
There are several terms relating to weights and measures
which are recorded in inscriptions (nos. 25, 98, 162). khi and
milima are the two weights which are supposed to be primarily
used for weighing purpose. Their ratio was 2 0 : 1. These were
47. These terms are considered by several scholars like Thomas, F.W.,
Sten Know, and H.W. Bailey in their papers in JRAS 1926 pp. 507 ff; ibid.
1924 pp. 671 IT; A.O. VI pp. 255 ff; BSOAS. 1948 pp. 128 ff; R.C. Agrawaia
has also contributed a few papers in this context. These are ‘Numismatic
Data in the Kharosthi documents from Chinese Turkestan’ JNSI, 1953, pp.
103 ff; ibid 1954 pp. 229 ff; and a study of weights and measures in the
Kharosthi Documents from Chinese Turkestan—JBRS. Vol. 38. pp. 365 ff.
48. Stein. Serindia. III. p. 1465 n. 6.
49. ibid. p. 530.
uniformly used for weighing purposes as also for transactions

relating to wine (no. 168, 175). One record (no. 175) notices
people taking 3 khi of wine, probably in illegal drinks. Another
one (no. 168) prescribes 30 khi of seed in a plot of land for
sowing purpose. Salary was also paid in this denomination (no.
210) and so also was tax paid accordingly. Even the load pre
scribed rather the maximum—to be carried by a camel—3
milima (no. 200). One record (no. 76) associates another word
‘casaga' with khi. It might be the local form of Sanskrit casaka.
The drinking bowl also used for measuring liquid things in
cluding ghee. Yet another terra sparna connotes measuring of
suki wine (no. 169). It was smaller than khi. Vacari is another
term mentioned in several records (nos. 159, 214, 295). It might
have been a kind of measuring pot, as proposed by Burrow.60
The prasta (no. 721) could be equated with the Sanskrit prastha.
Hasta—the term of measurement is noticed in several records
(nos. 581, 578, 83) and is generally used for carpet’s measure
ment. The height was measured in dithi (no. 187)-»-Skt. disti,
while another term kufhala is noticed (no. 327) along with misl
in land transaction (nos. 419, 582). These weights and mea
surements seem to be based on the Indian pattern while some
might have been derived from the current ones in other
neighbouring regions.
Labour and Transport
Labour poses a very important problem in the economic
activity of any country. The labourer has to be paid for his
services, otherwise associated with profits in an enterprise. The
payment of salary for work done seems to have been quite a
normal feature (no. 25) along with food and clothing (no. 50).
There could be wage dispute (no. 54). Labourers and slaves
formed the corps of workers (nos. 20, 29, 53) and they could
keep their earnings with them or invest these in land (no.
677), but exploitation of labour was not an unknown feature of
economic life and so was recourse to violence against women.
Labour was comparatively cheap and its mobility from one
place to another at the command of the master or in exchange50
50. Burrow. Translation—Op. cit p. 77.

is also recorded (nos. 491, 324). Bonded labour for a stipula

ted period of ten to twelve years (nos. 550,364) is also noticed.
Their purchase is recorded in inscriptions. A slaves price is
mentioned as 25 alta along with a five year old camel and a
horse of the same age (no. 591). There is no reference to skilled
labourer. An inscription (no. 272) prohibits immigration as
also over-crowding in the city, possibly in some period of
emergency.
Information about trade routes and medium of transport is
provided by the accounts of the Chinese travellers as well as
from inscriptions. The geography of Central Asia no doubt
forbids communication between China and the Western world
because of the Lop and Gobi deserts in the east, the long chains
of the Tien-shans and the Kun luns to the north and south;
and the Pamirs linked to the Kun-luns by the Karakoram
ranges in the south-west. These to-gether form a pair of
pincers around the heart of Chinese Central Asia. Further,
the sands of the Taklamakan desert with its deadly perils
were equally deterrents for traders and pilgrims. However,
the foot-hills of the mountain slopes in the north and
south provided the ground for traffic, facilitated by human
habitations along the course of the Tarim river. The famous
‘Silk Road’ or ‘Silk Route’ traversed this region of Central
Asia, and catered to the main traffic between China and the
West with several stopovers. The route started from the capital
Chang-an (modern Sian) in the province of Shansi and crossed
the Gobi desert to the oasis of Tun-huang from where it bifur
cated in two directions. The northern route passed through
Hami, Turfan, Karashahr, Kucha, Aksu, Tumshuk and Kashgar
to Samarkand. The southern one was via Miran, Cherchen,
Keriya, Khotan and Yarkand to Herat and Kabul.51 The means
of transport along these routes were camels and mules. There
are references to hiring of camels (nos. 52, 195). Any harm
done to this precious animal—rather the ship of the desert-
constituted an offence for which compensation had to be paid
to the owner. On a complaint by the owner, the culprit was sent
under escort (no 262). Another record (no. 195) awards com
51. Lerner. Along the Ancient Silk Routes—Op. cit p. 18.

pensation for the lost animal. The rates for hiring a camel for
transport seem to nave been fixed. The hiring of the animal
even by the State had to be done at regular rate (no. 272).
Escorts were provided on certain routes (no. 14). Another
record (no. 35) expresses regret that at that time the Chinese
traders in silk had not come there and so some dispute regard
ing advance on silk could not be settled in Court. Yet another
record (no. 223) notices the use of horses for transport. In this
case the mule was taken on hire since local arrangements could
not be made for transport. The states seem to have maintained
camels for transport (no. 248), and the local officials were
forbidden to use them for their families (no. 362
Administration and Rural Economy
The Kharosthi records as well provide a good deal of inform
ation relating to the administrative setup in these areas and the
officials associated with it. While it may not be necessary to
record the powers of the Cozbo—the ruling local authority and
his associates, in this context, it might be desirable to take into
account measures affecting economic activity as part of admini
stration. These include taxation and officers connected with
assessment and realization of rent and the role of officials in
regulating economic enterprises. The local official’s functions
included collection of land tax and other cesses in kind every
year. Assessment is recorded in several inscriptions (nos. 42, 57,
206, 275, 714 etc.) and the tax was paid in kind—grain, wine,
animal, ghee and other commodities (no. 714) which were
deposited in the State storage depots (nos. 279, 59, 291
etc.). The despatch of tax was not to be delayed although
pilfering of objects was not unknown (no. 567). Local cess
(dranga) was laid on articles of consumption. There are
several references of defaults (nos. 42,158,165) for which interest
was added to the arrears (no. 211). One inscription (no. 450)
records the case of a land holder who had not paid tax for four
years. It involved serious punishment with his land and house
being auctioned and he was required to work on the state farms
along with his wife and children. Two types of taxes—rottana
and churma—are also mentioned in this context. One can gather
from these records that tax collection agencies had to be firm
and strict and the defaulters had to undergo privation in the

form of services to be rendered at the state farms. There are also
instances of royal scribe (divira) purchasing such defaulters or
slaves farm or vineyard. Feudalistic tendencies and exploitation
of labour were normal features. The officials seem to have been
paid through land and its perquisites and it was quite natural to
make use of the available labour.
In a review of the material culture of Central Asia one notices
several features; Firstly the impact of the local population on
the foreigners who had settled down there as immigrants—
Indians, Iranians, Uighurians and Chinese, as also foreign in
fluence on the local ethos. The Indian element was forceful, as
is evident from numerous Indians, probably descendents of
Buddhist monks, traders and scholars as also some princes who
trekked into various parts of Central Asia, and set up some
kingdoms there. This is evident from the use of Indian regal
titles as also from the Indian administrative norms traced in
inscriptions. The association of some Indian names wjth the
local ones suggests understanding and spirit of co-existence bet
ween the natives and the immigrants. There are references to
Brahmins and also to upper class people—probably the ruling class
or those belonging to the administrative hierarchy. The impact
of Buddhism seems to have eliminated caste consciousness of
which there is hardly any trace. Of course, class consciousness
did exist. Native chiefs, feudal lords, affluent householders and
administrative officials constituted the upper and the middle
class, while slaves in general and other workers and artisans
belonged to the last group. The well born and the noble class
included natives as well as Indians, as is evident from the
names—Vasu Mogiya, Vasu Kekeya and Jeyaka (no. 588). The
exploitation of the slave class is evident from a record (no.
591) which notices the owner’s right to sell a purchased slave
just for a camel, a horse and 25 atga, as also to pledge him, to
exchange him, to give him to others as a present, and do what
ever he likes with him. The feudalistic society based on voc
ational status and economic division seems to have cast its
shadow on the Buddhist order, since one finds the Buddhist
monks owning lands as well as keeping slaves and also entering
into matrimony. Rural economy seem to have been shaped
accordingly, with the upper crust of society and the bureaucracy

owning land and farms, which they managed with the help of
labour and slaves, paid and provided for. Some slaves could
manage to set up their own farms and vineyards but these could
be purchased or appropriated in lieu of arrears of taxes by the scri
be or other tax-collecting officials. There wete also artisans and
workers who contributed to the national economy in their own
way. The international trade popularly called the silk trade
between China and the West had no doubt contributed to the
material culture of the people of Central Asia since it passed
through both'the Northern and the Southern areas. The material
finds at the places excavated and explored reveal a fairly good
standard of socio-economic life of the people dipped in native
traditions and cultured with the impact of Buddhist scholars
from India firmly pressed on the local ethos. The Chinese in
fluence is equally traceable in those parts which lie in close
proximity to China and is deeper in periods of its political
hegem onyinCentralAsia.lt is no doubt difficult to sift or
apportion the contributions—Indian, Iranian and Chinese as
also that of the later period—in the composite culture of the
peoples of Central Asia till about the tenth century, when Islam
had spread its influence and gradually changed the tenor and
temper of the people in these areas.
CHAPTER VI
THE ART OF CENTRAL ASIA 1
Central Asia provides the best example of a complex fusion

of peoples and ideas of different nationalities in the realms of
material culture and art. This vast area insular in nature, noted
for its virtually, unrelieved deserts and steppe, witnessed the
development of two types of human society—the nomadic one
1. The art of Central Asia is located at several centres on the northern
and southern silk trade routes. The peoples in these regions took over the art
forms of the great sedentary civilizations around them: those of the semi-
classical cultures of the East, of Iran, of Indo-Greek and Indo-Scythians, of
China, and of Gupta India. The borrowings, no doubt, vary in extent and
intensity from minute reminiscences of detail to essential concepts of style
and iconography. The Iranian element predominates in the western region,
while the Chinese influence in the figurative arts is prominent in the eastern
area of Central Asia nearer to or forming part of the Tang empire. The Indian
influence associated with Buddhism and Gupta art made itself felt in Khotan
and other places, varying, of course, in degree and dimension. Stylistic currents
emanating from Central Asia itself are supposed to be instrumental not only
in shaping Indo-Iranian trends of North-west India and Afghanistan,
considered the southernmost offshoot of Central Asian art. They also reacted
on the artistic evolution of Kashmir, on much of the ancient art of Tibet
(particularly painting) and even on Chinese Buddhist painting (Bussagli :
Op. cit, p. 16-17). Several scholars have made a general study of Central Asian
Art—sculptures and paintings, as’also of specific areas. These include :
Bussali, Mario : Painting o f Central Asia—translated from the Italian by
Lothian Small, Geneva, 1963; Talbot Rice, Tamara : Ancient Arts o f Central
Asia, London, 1965; Montebello, Phillippe de : Along the Ancient Silk
Routes—Central Asian Art from the West Berlin State Museums, New York,
1982; Rowland, B : The Wall Paintings o f India, Central Asia and Ceylon,
Boston, Mass. 1938; F.H. Andrews : Wall-Paintings from Ancient Shrines in
Central Asia recovered by Sir Aurel Stein, London, 1948; Nobuo Kumagai :
Central Asian Painting, Tokyo; Gray, B : Buddhist Cave Paintings at Tun-
huang, London, 1959; Hackin, J. : Buddhist Art in Central Asia, London,
1936. For a comprehensive bibliography on Central Asian art, see Hambi’s
Article on the subject in the Encyclopedia o f World Art, Vol. I, pp. 837-38.
in the north and the sedantary one in the south where there
was water and equal facilities for irrigation. In this settlement
complex the development of civilization was influenced by the
more vigorous cultures that flourished on the southern borders.
The cultures of the permanent settlements of the southern
region, vitalized by exchanges and contacts with the western
world—as also Persia, India and China, seem to exhibit a degree
of consistency in their artistic development, which those in the
northern areas, however, were influenced less by the major
Eurasian civilisations. Buddhism and its expansion from India
to Central Asia and the Far East, no doubt, provided the base
for the artistic activities in both the regions. It acted as a centri
petal force for devotees of different social groups in Central
Asia. They joined hands or acted independently in offerring
their services for the cause of the religion of the Tathagata.
Various episodes from the life of the Buddha 3akyamuni and
his previous incarnations were depicted on the walls of cave
shrines on the Indian models of Ajantaand Bagh as also theGan-
dhara pictorial art of Bamian, with the narration following the
ancient Buddhist texts. The Chinese influence was equally ac
cepted and the artists from that country were not inactive in their
devotion and service in this direction. Other external influences—
classical Greek and Roman, Persian and Sassanian—could as
well be traced in the pictorial art of Central Asia. While the
theme continues to be Buddhist, the actors in the pictorial
drama change with the brush of the painter, who fully makes
use of his imagination and background, as also his colour
scheme. In this context the earliest impact was with the
Gandhara art and its artists, who are supposed to have been
inspired by Greek traditions as modified in Rome. These artists
carried with them their pictorial and sculptural art to Central
Asia. The indigenous talent, however, accepted these artistic
influences with discrimination.
While it is difficult to suggest the terminus a quo of the wall
paintings in Central Asia, fragments of pictorial art still adhering
in patches on the walls or fallen and scattered in the accumu
lated dust and plaster on the floor of ruined shrines, suggest a
facile technique of long standing, but equally showing a varying
degree of skill in drawing. The decorator’s industry seemed to
be extensive with a prolonged period of evolution of art. The
The Art o f Central Asia 257
examples of painting in caves and free-standing shrines suggest

that the industry of painters was in a fairly advanced state and
equally productive with the cooperative efforts of local talents
as also experts, probably foreign ones. The caves afforded
greater protection from nature as well as from human aggres
sive activities. As such, these were best suited for the display
of Buddhist pictorial art. With the expansion of Buddhism and
the setting up of a large number of stupas and shrines, there was
a big demand for sculptors and painters to enrich these monu
ments with ornamental figures in stone or clay and artistic pictures
depicting the life and activities of the Sakyamuni as also those of
his previous lives as recorded in the Jatakas. The subject matter
provided by the Buddhist legendary accounts was later on
enriched with complex Hindu mythological elements, thus
expanding the scope and dimension of the artists’ plan and
pictorial display. The commercial enterprise on the trade routes
with the patronage and contribution of the traders and mer
chants was equally helpful in generating this move with the
services of roving artists easily available. It is equally likely that
painters sometimes imposed their creations on the earlier ones,
serving as palimpsests. The original one might have been better
or probably connected with the earlier school of the Hlnaya-
nists, and the second one gratified the donor suiting his personal
faith.
Central Asian paintings are rightly grouped under two groups,
purely on a geographical basis : those from sites to the south of
the great Taklamakan Desert comprise one group; while the
other one includes paintings from sites lying on the northern
route. Farhad-Beg Yailaki, Balawaste, Dandan-oilik, Khadlik
and Miran, on the southern route, and IChara Khoja, Toyuk,
Bezeklik, all in the Turfan area on the northern one, represent
primarily the centres of art related to the two groups. Tun-
Husang, the meeting pi ace of the two routes, is noted for the
caves of the Thousand Buddhas. It seems to have adopted
Miran’s style and iconography as the basis for its own school
of art whose artistic expression is in the Chinese manner. The
art of Tun-huang2, however, reflects the constant mixture of
2. For Tun-huang paintings, see ‘Essays on the Buddhist Paintings from
the caves of the Thousand Buddhas. Tun-huang’ by Raphael Petrucci and
influences of diverse origins. According to a seventh century

inscription on a stone tablet (beginning of the Tang dynasty),
the oldest Chen-Fo-tung monastery was consecrated by an
Indian monk in A.D. 366. It is even proposed3 that China may
have acquired her interest in the human figure from Miran,
where such renderings are particularly lovely, and that Tun-
huang may have owed its delight in landscape and its preoccu
pation with real and imaginary animals to China. The Tibetan
paintings of the Northern Wei and Sui periods, however, reveal
the forms and influences received from more western regions.
Well-defined styles of Central Asian paintings no doubt sug
gest Indian and Chinese inspirations. Indian influence, or rather,
qualities dominate in the south with Persian influence intruding,
while the Chinese influence is perceptible, in the north of course
with modification by Tibetan, Uighur and others. The common
factor in all the compositions—both in the North as well as in
the South, is the Buddhist legend—variously expressed. Hindu
and Tantric importations equally stimulated the imagination of
Chinese and Tibetan artists who were more concerned with the
stronger attraction of the decorative possibilities than with the
spiritual element which inspired the earlier Indian renderings. The
paintings are all linked to-getherbyacommon bond of one basic
religion—Buddhism. The manner of expression, however, differs
widely in style and treatment. This was due to the changing
political situations and the complex social milieu, with the domi
nant power changing hands from one race to another differing
Lawrence Bin Yon—Appendix E; Stein, A : Serindia, Vol. Ill, pp. 1392 ff;
Gray, B : Buddhist Paintings of Tun-huang, London, 1959; Pelliot, P : Les
Grottes de Touen-houang, 6 Vols. Paris, 1920-4; Stein, A : The Thousand
Buddhas, Ancient Buddhist Paintings from the Cave Temples o f Tun-huang,
London, 1921-22.
3. Op. cit, p. 214. Tun-huang is supposed to have adopted Miran’s style
and iconography as the basis for its own school of art. Miran became the
centre for influencing contemporary artists of Niya, Cherchen and Loulan.
The reference to an artist Titus working at Miran (Stein : Serindia, I. p. 538)
is interesting. This Roman Eurasian artist trained in the Hellenistic tradition
and impregnated with Buddhism was instrumental in the diffusion of the
Central Asian School at Miran. It is suggested by Talbot Rice, that Titus
must have been familiar with Byzantine painting and his use of chiaroscuro
also reflects Byzantine influence and so too perhaps do the large eyes.
in ideals and traditions in art as also varying in psychological

conceptions. That accounts for diversities in design and treat
ment of the paintings, although identical mannerisms and even
use of some compositions in places at considerable distance from
each other could be suggestive of the same artist and his troupe
of painters moving from one centre to another.
The art of Serindia, as the region came to be called by Aurel
Stein, which had developed amidst the political vicissitudes of a
number of centuries, roughly from the first of the Christian era
until the Muslim invasion, reveals varied influences4 that had
played upon it. In its earliest phase, supposed to be Greco-
Buddhist in form, it revealed Indian as also Persian influence
depending on the circumstances, and in a way reflected the art
of western Asia. Then it followed the Chinese influence, domi
nating throughout the Tang dynasty. Subsequently, western
elements seem to have influenced it. Later on, this art, cut off
from its Persian and Indian sources of inspiration consequent to
Islamic conquest, absorbed foreign influence only from China.
This would be evident from art centres nearer to China proper
under the impact and influence of Chinese artists. The southern
centres of the silk Route, passing through the Tarim Basin, in
habited over long periods, were later abandoned because of sud
den shifts in the course of the rivers. These included Miran,
Endere, Niya and those forming the Khotan group : Yotkan,
Rawak, Dandan-oilik and Ak-terak. The remnants of art in
these centres represent the earlier phases of the Central Asian
Art, and, therefore, deserve prior consideration.
4. Numerous artistic currents flowed along the caravan routes leading to
and from China. In addition to the Central Asian elements, the influence of
the art of the Gupta period also penetrated to the Taklamakan through Central
Afghanistan. Hellenism likewise reached the eastern cities both in the form
into which it had been fashioned in Gandhara and also directly from the
western world as suggested by Foucher. In the words of Aurel Stein, ‘the
paintings of Serindia reflect the Mediterranean delight in beauty, youth and
life’. Indo-Greek art with its classical elements is traced by Grunwedel both
at Kucha and in Khotan, and was even produced in Kashgaria as a centre
of Buddhist art and religion. As Buddhism advanced towards the Taklamakan,
Kashgaria with Yarkand and Khotan in the west, Tumsuk, Akshu and Kizil
in the north; Sorchuk and Karasahr in the east, and Miran and Cherchcn in
the south became important centres of Buddhist art and thought. These in
their turn transmitted their artistic style to religious centres. The Tun-huaug
Miran* :
The ruins of Miran, located south of Lop Nov in a region pro
bably supplied from Cherchen Darya, comprise a massive build
ing, square in plan. On a circular base in a round room inside
stands a stupa invisible from outside. As a result of the protec
tion of the sand, the ruins with wall paintings could be safely
preserved through the centuries. These paintings may be dated
from the end of the 3rd centurv or the beginning of the 4th by
comparison with other paintings from western Asia. Apart from
their artistic and technical interest, the Miran paintings suggest
familiarity with the Gandhara art. In fact, Miran is considered
as an outpost of Gandhara Art. The motive of the festooned
garland carried on the shoulders of amorini as also the winged
angel busts placed in the upper hollows of the festoon rendered
in a more interesting way appear to be in line, or rather an im
proved version of the same in Gandhara sculptures. According
to Aurel Stein,* 56 ‘the approach to purely classical design and co
louring was closer in these frescoes than in any work of ancient
pictorial art’. The painted dado of beautiful winged angels, as
‘classical representations of Cherubim, recall cherished scenes of
Christian imagery, most surprisingly on the walls of what was
beyond all doubt a Buddhist sanctuary’. This might be due to
the importation of classical artists. The name of one of the
painters Tita—perhaps the equivalent of Titus—very probably a
artists encountered in addition styles which, though evolved in the Khotanese
area, had undergone alterations in the course of migration. (Talbot Rice :
Op. cit, p. 180—adaptation)
5. The Miran paintings are first noticed and recorded by Aurel Stein in
his Ruins o f Desert Cathay, Vol. I, pp. 452 flf and later on in his Serindia,
Vol. I, pp. 497 fT. These are described in detail by F.H. Andrews in his
Catalogue—1934 as also in detail along with illustrations in his— Wall
Paintings from Ancient Shrines in Central Asia— 1948, XVIII-XX, 2-16;
Bussagli has a special chapter on Miran Paintings in his work on Paintings
o f Central Asia—Op. cit, pp. 19 ff. Miran’s art is supposed to have stemmed
from India and Gandhara. The reference to the artist Titus and his school
envisages closer connection of this art centre with the western art centres of
Antioch and Dura Europos. According to Talbot Rice, Miran probably
acquired the Greco-Roman elements in its art by direct contacts with the
West rather than indirectly (Op. cit, p. 214). Its style and iconography was
adopted as the basis for its own school of art at Tun-huang.
6. Ruins o f Desert Cathay—Op.cit, p. 458 flf.
Roman subject of Asian origin but Hellenistic training is record
ed in Kharo$thi on the leg of an elephant in the famous Ves-
santara Jataka scene. In fact during the third and fourth cen
turies A.D. the whole of Central Asia formed a peripheral area
within the Graeco-Roman sphere of influence. The classical
influence emanated from the semi-classical school of Gandhara
art. A single group of itinerant artists comprising the master
and his pupils seem to have moved about in these centres of Bud
dhist monastic establishment, and their services were availed of.
While most of the artists remain anonymous, this particular one
Tita actually reveals his personality and he also received payment
for his services. These artists exhibited classical elements not
only in the treatment of drapery, but also made skilful use of
chiaroscuro—treatment of light and shade—in their paintings.
Further, the wall decorations of Miran are considered as the most
extensive group of Gandhara paintings in view of some highly
significant features of their style and iconography. These are
linked up with a particular trend of Gandhara art rich in bril
liant elements. These may be recorded with reference to a few
specimens of Miran paintings.
The Buddha with six monks7 is a typical scene of Buddhist
iconography. On the left one notices the Master standing, dres
sed in a simple robe of dark red-brown colour—characteristic of
Indian tradition. The halo and the top-knob of hair, partly
broken, and the right hand raised in the pose of protection
(abhaya-mudra) confirm this figure to be the Buddha. His left
hand, held low in front probably supported his drapery (sanghatl).
Behind the teacher and to his left are six monks or arhats ranged
in two rows and wearing robes in a variety of bright colours.
The shaven head of monks is conspicuous. The monk on the
left end of the upper row carries a white fan, probably meant for
a yaktail or chauri traditionally associated with holy or regal per
sonalities. To the left of the saints, an elliptical mass probably
part of a tree, is studded with red and white flowers and poppy
like leaves on dark greyish-green ground. An upraised right
hand grasping a handful of white buds or flowers, is shown aga
7. Andrews : Opi cit„ p.3; See also Stein : Ruins—Op. cit, p. 470, pi. V.
inst this background suggesting the act of throwing flowers on the

distinguished guests. This fresco fragment is considered of great
value and interest for the artistic treatment in composition, design
and colouring. The details of the subject—no doubt Buddhist—
in its presentation point*to adaptation from classical models. The
head of Buddha is supposed to be of a type unmistakably Hel
lenistic, in spite of a slight Semitic touch in the nose and of comp
liance with Indian Buddhist convention in regard to the top-knot
and long pierced ears. The large straight eyes of the teacher and
disciples are equally striking.
Another fragment from Miran (III. 002) depicts a figure seated
on a throne. Stein takes8 it to be that of a teacher meant in all
probability for Gautama Bodhisattva. A dark red undergar
ment reaches from the hips to the ankles, and a buff-coloured
cloak is thrown over the left shoulder, leaving most of the upper
part of the body bare. The drapery is treated in a fashion, sup
posed to be unmistakably classical. The same dressing of the
folds is noticed in the smaller adoring figure to the right. He
appears to be a princely character clearly marked by a curious
white conical cap enriched with rings and two lunette-shaped red
flaps. This headdress, not traced anywhere in the Buddhist
sculpture of Gandhara or in any later Buddhist shrines of Eas
tern Turkestan, seems local in character. It recurs in a number
of the Miran frescoes. It is, however, proposed that it represe
nts a feature introduced for a time from one of the more
western territories like Bactria or Sogdiana. Of a second
adoring figure on the left, only parts of the knee and arm
are noticed. The big eyes, the drooping ones of the seated figure
and the gazing and attentive ones of the adoring one, and the
Indian loin-cloth dhoti and the scarf curving are suggestive of
Indian influence. The theme is unidentified but appears to be
of a religious nature, probably a saintly person or a religious
teacher advising the young prince who is fairly attentive with
folding hands. The faces are of rather Semitic type, with fine
straight-set eyes, arched eye-brows,0 well separated above the
nose; small but thick black moustache, carefully pointed and a
8. Stein : Cathay, p. 473.
9. Andrews : Op. cit, p. 3.
thin wavy lock of hair falling in front of each ear. The hands
are strong and broad with the thumb abducted and short finger
nails.
Another interesting fresco fragment (M. III. 0019) represents
heads of two worshippers—girls with folded hands, as they gaze,
with their wide eyes, towards the left. The faces are fair with
pink cheeks and are painted with definite chiaroscuro, the shades
being pearly-grey. The girl to the left is very young and in an
amusing mood, while that to the right appears older and more
sophisticated and stern in her look. The smiling lips of the
younger one are solid red, while the transverse wrinkle in the
necks of both suggest their superior birth. The black hair is
long and stylishly dressed, with tresses falling behind the ears to
the shoulders and wavy love-locks on the cheeks. The hair on
the forehead of the girl on the right, precisely and fancifully ar-r
ranged, suggest her gentle upbringing. White bands crown the
hair of the girls and their eyebrows are delicately arched, disting
uishing^ separated. The ears are not very much elongated.
The dress of the girls is stylish. The robe of the figure to the
left is yellow, ontlined with red V-shape opening at the neck;
that of the other one is similar in form, but light green outlined
with dark grey. Andrew considers them to be the two daughters
of Prince Vessantara, ready to participate in the drama of renun
ciation as narrated in this Jataka story.
Another interesting feature at Miran is a scheme of a heavy
floral garland carried on the shoulders of youthful supporters
placed at regular intervals, and undulating completely round the
shrine. A typical head and bust of a man or woman rises bet
ween supporters from each hollow followed by the downward
droop of the festoon. It is a motive widely used in Gandhara
sculpture.10 The supporters in the example at Miran are remini
scent of Italian amorini dipped in Eastern nuances. They are
sometimes clothed in Persian garments consisting of a narrow
10. It is suggested by Andrews that this motive does not seem to have
survived into later periods in India or Central Asia. Festoons of fruits and
flowers, depending from ox-skulls, were used to adorn temples in ancient
Rome as offerings; and the motive, in decoration, has persisted widely in
the west, but not quite in the form of a continuous garland carried by human
supporters. ( Wall Paintings, p. 10).
sleeved tunic and Phrygian cap, and sometimes only in a scanty

loin-cloth. But more remarkable still are the portraits filling in
succession the hollows of the undulating festoon. In each rises
the head and bust of a man or girl, presented in classical outlines
but with a freedom of individual expression. The types of men’s
head differ, some quite Roman in look, others with their peculiar
cut of hair and beard. A graceful girl is shown playing on a four
stringed mandoline.11 A wreath of white flowers is set on her
rich black hair, bound by a crimson ribbon and gathered in a
bunch behind the neck. A diadem made up of red beads and
pendant jewels stretches across the forehead. A crimson flower
or ornament of that shape hangs from each ear, before which a
curly love-lock descends. The full sensual lips are supposed to
harmonize with the elaborate adornment of this mature beauty.
Next to her to the right is a Phrygian-capped figure draped in gre
en carrying the festoon, and beyond this facing the girl is a beared
male bust, striking in features and dress. His heavy curled hair,
the moustache and the long beard clearly distinguish him from
other classical male faces seen elsewhere in the dado. Another
one12 shows a young Indian prince, clean shaven, except for a
curling moustache and cap similar to the one in the attitude of
worshipping the Buddha. The dreamy looking eyes with an ex
pression of softness are supposed to be typically Indian. The
characteristic headdress is also Indian, with two red-lined flaps
turned upwards over the forehead, and a white pugree-turban
wound round it. Its end is gathered behind into a sort of hood,
as seen in many Gandhara sculptures, representing prince Sid-
dhartha and other royal figures. A large ornament in the ear, a
broad jewelled band round the neck, and two heavy armlets over
the right wrist suggest his high rank. He holds some object, pro
bably a fruit in his hand. A cloak of light green is thrown over
the left shoulder, leaving the rest of the breast bare.
The next pair of busts13 appear to be strikingly western in look.
First the portrait of a young girl carries in graceful pose on her
left shoulder a narrow necked jug of transparent ware and in her
11. Stein : Ruins, fig. 146.
12. ibid, fig. 147, p. 482.
13. ibid, fij. 148, p. 483.
right hand a patera. The Greek features, in the words of Stein,

‘mingling with Levantine or Circassian type of beauty are grace
fully blended with Iranian or Near Eastern dress, the white tur
ban, trimmed with a red bank and held by a large black knot on
the rights resting on the rich black hair’. Long ringlets descend
in front of the ears while a fringe of hair comes down on the for
ehead decorated with three bead strings of coral. The ears bear
graceful pendants in pink. A close-fitting vest with sleeves in a
deep red brown cover breast and shoulders, and from the head
dress hangs a veil of a delicate pale green. She is balanced on the
opposite side by a male head of a type distinctly Roman, strongly
built face, clean shaven, close-cropped black hair covering the
head. The dress consists of a dark red coat with a pale green
cloak thrown over the right shoulder.
Reference might as well be made to the Vessantara Jataka scene
earlier supposed by Aurel Stein as a princely procession14 scene.
The frieze above the dado at the south-east over a segment more than
eighteen feet long illustrates this Jataka, on the upper part of the
wall. There is an appearance of continuity of the several inci
dents in the painting without any dividing line such as a pilaster—
an architectural feature. The incidents seem to be separated
from each other by trees. Starting from the left, Stein records a
princely figure on horseback riding out of a palace gate, clearly
indicated in the picture with the wooden frame work and decora
tive carving on the gate. The horseman’s costume is very much
like that of the ‘Indian prince’ in the dado. A crimson cloak
descends across the left shoulder to below the waist while a green
garment resembling the Indian loin cloth (dhoti) covers the lower
parts of the body. A rich jewelled armlet and a broad necklace
painted in red mark the high rank of the rider. The horse,
remarkably well-drawn in white, has its head-stall and bridle de
corated with red tufts passing across its breast and apparently
reaching to the saddle is a broad belt of three straps or strings.
The whole suggested saddlery, according to Stein, corresponds
with the Roman sculpture of later times.
14. Stein : Ruins—Chapter XLIV, pp. 486 IT. This painting on the upper
part of the wall in the same shrine (M.V.) was not removed by Aurel Stein
and was subsequently destroyed by the clumsy opeations of a Japanese
'‘archaeologist’.
A chariot in front of the horseman with four white horses

abreast bearing harness of the same type as recorded earlier,
is being driven by a beautiful woman. Her hair descends in
black tresses below the neck, with two love-locks in front of the
ears. Her dress consists of a mauve bodice open in front and
held together by two head strings across the breast and of a
green mantle laid in heavy folds over the left shoulder. The fair
lady gives an impression of Iranian influence on some classical
model. Behind her are two standing children.
In the next scene is shown a richly caparisoned white ele
phant, truthfully painted and in a realistic form, as appears
from the expression of the animal’s eyes and face. The elaborate
adornment of the forehead and trunk consists of a diadem of
leaves, bands, bosses and rings, with a painted saddle-cloth and
carpet-like covering spread over it. A small neatly written
inscription of two lines in Kharosthi, above the right hand of
the elephant mentions the name of the painter .Tita—Titus and
his remuneration of 3000 Bhammakas for it. The elephant is
being led by a person with the same characteristic dress as in
the dado, and the rich jewellery shown on neck, ears, arm and
wrist, representing an Indian prince. His left hand supports the
elephant’s trunk, while the right one carries a peculiarly shaped
jug. This procession is met by four plainly dress figures with
their bushy hair and beards and their long staffs, symbolising
typical Indian anchorites. Comparing the frieze with the dado,
Stein finds traces of the Greco-Buddhist style in the former and
the contemporary art of the Roman Orient as transmitted
through Persia being reflected in the latter.15 On the basis of
the influence exercised by classical art in this remote corner of
Central Asia, the frescoed walls of the temples at Miran provide
excellent testimony. The inscriptional evidence confirms the
classical influence as it also proposes the probable date of these
frescoes which might be placed some time about the third
century A.D.
There is, however, considerable internal evidence in these
paintings to suggest their Indian conception and execution. The
men are Indian, some with full moustache and beard, dressed
15. ibid, p. 489.

in Indian garments with bare feet. The hands are also those of
Indians. In the destroyed painting of the Vessantara Jataka,
faithfully recorded earlier by Aurel Stein, the elephant shows
the accuracy of form and truth of action, as could be faithfully
rendered by the Indian artist alone. The girls, although sug
gestive of the Persian type of beauty, might well have been
Indian, as proposed by Andrew,16 perhaps influenced by contact
with Persian fashion. It is further proposed that the use of
Kharo?thi and the legend about an Indian colony in Khotan in
ASoka’s time strengthen the probability that Indian artists,
familiar with Buddhist lore, may have found employment for
their skill along the Silk Route between Khotan and China on
which Miran stood. The partial shaving of the heads of the
garland carrying boys is almost certainly Indian. There is noth
ing in the Miran paintings, which could be definitely indicative
of borrowings from Chinese art, despite trade relations with
China.
The technique of the Miran paintings, which are all in tempera
is in conformity with the well-developed methods. The design
is first drawn on paper and then transferred to the whitened
wall surface either by pouncing through the pricked drawing or
by other familiar means. The transferred outlines are then
lightly traced over with a pale colour for fixation. The colour
scheme follows next with the use of brush adding shading tints
to suggest chiaroscuro. The contours are strengthened with soft
brush lines of red or dark grey, blending to some extent with
the colours and also providing softness and roundness to the
edges. Emphasis is provided with the final touches of black or
red, while grey is used for high lights and the eyes. The colours
here are few and those locally available from mineral sources,
lamp black and indigo and occasionally of other vegetable
origins.
It is proposed that the wall decorations of Miran, on the
strength of their style and their Buddhist content, symbolic
compositions and motifs of the Kushcano-Iranian type form the
most extensive group of Gandhara school’s normal area of
diffusion. This school is supposed to have been transplanted
16. Wall Paintings, p. XXI.
into pictorial terms at Milan or in other words, the art of Miran

may he considered as an outpost of Gandhara art, suggesting
contact between the peoples of Central Asia and those of the
Kushana empire. In view of their style and iconography they are
equally rich in western elements brilliantly recast. The paintings
at Miran might appear mediocre to many, but their importance
for revealing western impact in the eastern ethos on a Buddhist
canvas cannot be lost sight of. As an earlier example of
Buddhist pictorial art, the Miran paintings do not appear to be
nascent but products of an evolved and fairly matured technique.
The Khotan Complex11:
The school of Khotan—as it is called, provides evidence of
absorption of Indian, Sassanian, Chinese, Sogdian and perhaps
also Chorasmian influences which are all assimilated and recasted
here as if in an original form. The imprint of Greco-Buddhist art,
however, predominates between the fifth and the eighth centu
ries. The complex provides examples of the stupa, of certain
Greco-Buddhist elements in the plan of monasteries and in
wall paintings, as for instance, in the' standing juxtaposed
figures of Buddha in Rawak, compared to the decorative
arrangements found near Hadda. These elements are also
traceable in decorative motifs and elements. Gupta influences
are equally noticed in various objects discovered in the
Khotan area—fragments of stylized drapery in stucco, oval
faces with simplified treatment of the hair and idealized
expression, closely clinging drapery, and in particular,
flying figures bearing garlands. In several works of art, of a
later date, from Dandan Oilik, Hindu influences are traceable.
These seem to have come from Kashmir in India, nicely blended
17. For a reference to Khotan and its art, see A. Stein : Ancient Khotan,
2 Vols Oxford 1907; ib id : Sand-Buried Ruins o f Khotan, London, 1907;
A. Grundwedel : Alt Buddhisteche Kulofaatten in Chinese Turkistan, Berlin
1912; P. Pelliot: ‘Les influences iraniennes en Asia Centrale et en Extreme
Orient’ Reveue d'histoire et de litterature religiouess III, 1912 pp. 97-119;
Stein : Ruins o f Desert Cathay, 2 Vols. London, 1912; Stein : Serindia, 5 Vols.
Oxford, 1921; R. Grousset : L’ art de 1’Asia Centrals et le influences irrainenns,
RAA. I. 1924. pp. 13-16; A. Stein ; On Ancient Central Asian Treks, London,
1932; F. Andrews : ‘Central Asia Wall Painting’, Indian Art & Letters. VII,
1934, pp. 1-21; Louis Hambis & M. Hallade : Sculpture et Peintures de T
Asia Central, Paris, 1937; also Encyclopedia o f World Art, Vol. VIII.
in the customary Greco-Buddhist art. Iranian details appear in

several examples, like the panels in Dandan-Oiliq No. X con
cerning the introduction of silk culture into Khotan. With this
blending of foreign influences emerged a new style evident from
the wall paintings of the shrines and monuments of the Demoko
group east of Khotan, and dated about the middle of the sixth
century. It is characterised by strictly frontal presentation,
highly developed stylization, a flat, almost two-dimensional desi
gns, and a tendency to geometric simplification which also occurs
in some votive paintings on wooden tablets from Dandan-Oiliq.
Laws of proportion and schematic designs are based on the
segment of a circle and the eclipse. Further, the Dandan panels
are supposed to represent a peculiar technique of drawing and
colour scheme by vague, sporadic reminiscence of Indian art
of the Gupta period.
The Khotanese school of painting centred round the mona
steries of Rawak, Dandan-Oiliq, Niya and Endere and Bala-
waste, as also Farhad-Beg Yailaki and Kuduk-Kol, all lying to
the north and north-east of Domko on the southern route about
four hundred and fifty miles south-west of Miran. This school
and its painters were much admired in contemporary China. In
late Sui times, about A.D. 620 the work of the Khotanese artists
Weich-in the elder, and the younger, were especially in demand
in the Chinese capital at Chan-gan. In Tibet, too, the painters
of this school were held in great esteem. This impact on Tibet
was great in the eighth century in the time of Ral-Pa-Chan
when there were Khotanese painters and monk translators there.
The prevailing influence exercised by the Khotan art on Tibet at
different times, however, varied in subject and type.18
18. Talbot Rice : Op. cit, p. 208. The Khotanese influence is traced in
Chinese Buddhist sculptures as well. According to Bachhofer, it is somewhat
surprising to find traces of the Knotanese type in the Chinese Buddha statues,
with the narrow face, the slit eyes, the large ufnrta, the same mouth and nose,
and the leaf-shaped mandorla. The plates of the garment, however, are
asymmetrical as they are in the art of north-western India or Gandhara. The
Khotanese type must have reached southern China in the fourth century.
It is very likely that a wave of influence from Khotan had reached northern
China at the same tine (Bachhofer : A Short History o f Chinese Art, London,
p. 65).
Among the specimens of Khotan school of painting may be

noticed the figure of Hariti from a shrine at Farhad-Beg-Yailaki
the surviving bust of a Tantric Buddha in meditation, probably
from Balawaste in the Domko region, as also the worshipper
from the same region, the Iranian Bodhisattva from Dandan-
Oiliq, the silk princess shown on a wooden tablet, the Bodhi
sattva of the Tantric type from Balawaste and the lovely Nagini
rising from the waters of an artificial fountain in Temple D. 11
at Dandan-Oilik. These might be noticed on a few, representing
paintings of the school from different centres. The goddess
Hariti19 in her regenerate aspect as the protectress and nourisher
of children and goddess of fecundity is attended by five children.
The two principal figures are shown in front view. In the
mingled sadness and sweetness of her expression, with heavy
lidded, half-closed, dreamy eyes, could be traced a smouldering
survival of the old fives capable of renewed activities as also an
introspective mood and nostalgic past. The robe worn by
Hariti has Iranian characteristics, and there is something of the
Persian houri in her appearance, characterised by the rather
insistent love-locks, the bloom on her cheeks (now discoloured)
and the voluptuous folds of her plump neck. Large rings are
noticed in the ears or the lobes which seem to be pierced with
gaping holes. The halo behind the head is turquoise green
surrounded with red and buff. In the badly damaged condition
of the painting, one could, however, notice Hariti, sitting cross-
legged with her right forearm bent to suppport one of the boys
sitting astride her wrist; another embracing her left breast, two
astride her shoulders, the one on the left shoulder wearing a
terra-cotta clout smock. The other three are nude. The fifth one,
badly defaced, on the left of the picture, wears a green smock
and seems to be dancing.
The bust of the Buddha from Balawaste,20 with the symbolic
19. Hariti with Five Children. Wall painting from Shrine XII, Farhad
Beg-Yailaki, Mid-Sixth century, 20.C.XII-004, Collection of the National
Museum, New Delhi. For a fuller description—see F. Andrews : Wall
Paintings, p. 17.
20. Buddha in Meditation, Wall Painting, probably from Balawaste,
Mid-sixth century. Haray’s collection of the National Museum, New Delhi;
Bussagli : Op. cir, pp. 55, 58.
motifs of the Tantric type adorning it, symbolises the influence

of Mahayana Buddhism with local influence accentuating it.
Among the Tantric symbols in this Buddha figure are the sun
and the moon, the two flaming jewels on lotus flowers, the two
books drawn on the upper arms, also surrounded by flames and
standing on lotus flowers, together with the Vajra (thunderbolt)
on the forearms. Other symbols include a chain ornament, a
central motif alluding to life and immortality, a galloping horse
and a crown alleged to be of the Sassanian type symbolising
royal powers. The base has radiating lines running down from
its junction with body to foot. Round the junction is apparently
wrapped a snake with a part of its body protecting like a cord
on each side and each part terminating in a snake’s head. The
whole of the device is perhaps a rendering of the churning of the
ocean.
Another painted figure from Balawaste,21 deserving notice is
supposed to be that of a worshipper or of Indra. The figure
is either kneeling or sitting with his legs crossed. His body
leans forward from the hips with the head tilted back. The eyes
are downcast and hands folded and uplifted to neck-level. The
thumbs, strongly abducted, are upright while the fingers point
horizontally. The head is covered with a close-fitting cap with
a head band in dark pink colour studded with white dots or
pearls. This beautiful picture from Balawaste seems distinctly
Indian in structure, colouring, costume and style similar to
some of the Ajanta figures datable to the early seventh century.
The eyes are heavy-lidded, large and dreamy. The eye-brows are
not so arched as in the Bezaklik paintings. In the ear, the ear
ring is a quatrefoil, the lowest foil being a green jewel. Round
the neck is a plain double band and on breast a massive carcanet.
On the upper arms are gold armlets, and there are two bangles
on the wrist. On the back of the right hand is drawn an eye in
21. Stein and Andrews : Catalogue—Op. cit, p. 13; Bussagli : Op. cit,
pp. 55, 57; P. Banerji in his paper on ‘Hindu deities in Central Asia’ (Viveka-
nand Volume, p. 285) considers this figure to be that of Indra. The presence of
the eye on the hand ,‘s supposed to be the conclusive proof of this identification.
Banerji also refers to the figure of Indra occurring on some other paintings
from Central Asia.(See also his paper entitled ‘Indra from Balawaste’ published
in Indo-Asian Culture, XVII, no. 4, pp. 14 If.)
black which makes the figure’s identification with Indra certain.
Besides various ornaments, the figure is supposed to have a
muktita, yajhopavita and is endowed with a nimbus.
The Hindu influence on the Buddhist paintings in the context
of Tantric symbols is evident from the painting of the Buddhist
God of the Tantric type from Balawaste.22 The fragment
shows on lower part a trimurti divinity with a small seated
Buddha above to left and the toes of a large figure standing on
a lotus to right. The trimurti figure sits full-face with head slightly
turned to left, the second and third heads, about two-thirds the
size of the central project on either side from behind the ears.
The central face has a third eye in the forehead and a long thin
moustache. The eyes are heavy-lidded and dreamy. On the head
is a skull set against the black top-knob with a pearlshaded
taenia. Large plain ear-rings adorn the normal sized ears. Other
ornaments include a heavy necklet, bangles and armlets. There
are four arms, two upraised holding the sun to the left and the
moon to the right. The right lower hand holds a pomegranate
against the breast while the left resting on the left thigh, grasps
an indistinguishable object. A long yellow stole is thrown round
the back of the shoulders while long black hair hangs behind.
There is no nimbus and the field of vesica is grey-green, border
ed with a red inner band and an outer band of red-brown, with
both hands contoured with thin white lines. An example of
this figure, with slight variations, is the one painted on
an extremely interesting wooden panel, found by Stein
at the ancient site of Dandan-Oiliq in 190023. In that panel the
figure looks the other way, and holds a white object instead
of a pomegranate in the right hand. This could be a drum
(tfamaru) or some other fruit. The positions of the two subsidi
ary heads are also reversed—the smiling (female ?) head to the
left and the demon to the right. The positions of the sun and
moon emblems (probably wrongly identified as Cakra and
22. Andrews : Wall Paintings—no Bal 0200 p. 22; Bussagli : Op. cit,
p. 60, 63. Of the four arms, the front p*ir are posed very similarly to those
of the Teacher in Miran Painting, III. 002, pi. 1. An example of this painting,
with slight variations, is that painted on an extremely wooden panel from the
ruined dwelling. D . VII at Dandan-Oiliq (Stein : Ancient Khctan PI. XL).
23. Stein : Ancient Khotan, PI. XL; Talbot Rice : Op. cit, PI. 1971.
Sahkha) are also reversed. In this panel the loins are covered
by a tiger skin. It has been suggested that this figure, undoub
tedly Siva, and one of the numberless importations into
Mahayana Buddhism from Brahmanic Iconography, is adopted
as one of the forms of the favourite Bodhisattva Avalokitesvara.
While the Hindu influences appear to have come through
Kashmir, the Khotanese school has example of purely Iranian
details and in some cases the Chinese features are also traced.
One such example of direct connection between local art and
the Sogdian style is corroborated in the Iranian Bodhisattva24 of
Dandan-Oiliq. It is characterised by the elongation of the body
and other features. Painted on a wooden tablet, this fine votive
image is four-armed, black-bearded, wearing a pale green close-
fitting tunic. He sits on a cushion and has nimbus, halo-crown,
dagger and other attributes. On the back of the panel is depicted
a three-headed goddess of the Tantrictype, probably of Saivite
origin but definitely connected with Buddhism.
Another wooden tablet depicts the legend of the famous
Silk Princess25, who secretly introduced the silk worm culture
in Khotan. This figure at the top is four-armed and has two
of the attributes of the Iranian Bodhisattva, the cup and the
knife with a short triangular blade, and both are wearing a crown.
Similar images are found, with minor variations, on a number
of panels from Dandan-Oiliq, thus testifying to a genuinely
Khotanese iconographic type. In the figure of the Silk Princess
her oval face, nose, costume and head dress, Chinese influence
could be traced. It is, therefore, proposed that various tenden
cies, Indian, Iranian and Chinese, co-existed in these particular
works of Dandan-Oiliq, dating sometime between the sixth and
eighth century as the site was finally abandoned in 791. The
best known of the Khotan paintings is supposed to be tho
24. The ‘Iranian Bodhisattva’ wooden votive Tablet from Dandan Oiluq
(Khotan) Sanctuary D. VII, probably seventh century, now in the British
Museum. For a description, see Bussagli : Op. cit, p. 61. It is proposed by
Natalia Diakonova that the figure is that of a goddess connected with silk—on
the strength of an iconographic analogy with other images, all of them from
Dandan Oiluq, which is not convincing.
25. Stein : Ancient Khotan. D. X. PI. LXIH. The Silk Princess wooden
votive Tablet from Dandan-Oiliq (Khotan) probably of the seventh century
is now in the British Museum, London; See Bussagli : Op. cit, p. 56.
lovely Nagini26 rising from the water of an artificial fountain in

Temple D. II at Dandan-Oiliq, signifying fusion of Indian and
Chinese elements. Unfortunately the painting is lost and the
photograph alone displays gesture of maidenly modesty. Fixed
symbolism and iconography connected with Buddhist thought
conditioned by the display of external factors and forces,
enriched Khotanese pictorial art, noted for its vitality and
assimilative tendency without sacrificing its originality. Some
of the greatest Khotanese painters worked in China in the
late Sui and early Tang period and made a deep impression
there. Wei-chh Po-Chihna—the elder, and the younger one
Wei-Chih (Vijaya), introduced in China a sense of colours
and a way of handling it with exceptional power of expression
and vigour of style.
The Northern Schools27
The art centres on the Northern Route Kizil, Kucha and
Turfan with the neighbouring centres. Kumtura (Kucharegion)
and Bezaklik, Toyuq & Qoco in the Turfan region also provide
illustrations of pictorial art which were equally connected or
influenced by the artistic developments of the neighbouring
centres—eastern and western. Political factors like the exodus
of the Sassanians as emigre’s at Kucha or the political domin
ation of China over Kucha (which in 658 became the seat of
the Chinese Government of the Tarim Basin) were responsible
for foreign influence on local art as also on the traditions con
nected with the art centres. The Kucha school of painting, re
presenting the western aspect of the Northern School, seems
26. Bussagli : Op. cit, p. 66.
27. For a general review of the Central Asian Paintings from centres on
the Northern Route, see Bussagli : Op cit, pp. 69 ff; those in the State
Museum, Berlin, are catalogued under Central Asian Art—in the work ‘Along
the Ancient Silk Routes’, New York (Op. cit) with a fine introduction by
Herbert Hartel, and a description of some of the important pieces, exhibited
at the Metropolitan Museum of Art. The finds from these art centres, rich in
Buddhist remains, have been studied by several eminent scholars—Russian,
German, British and French. Reference to these investigations is given in
brief by Andrews in his Wall Paintings (Op. cit, pp. xxiv ff). For a" fuller
bibliography, see note 1 of this Chapter based on the Entry in the Encyclopedia
of World Art, Vol. I, pp. 837-38. Individual references would be provided
separately. See also Talbot Rice : Op. cit, pp. 180 ff.
to have begun in the 4th century A.D. and continued in a more

liberal way till the end of the eighth century A.D. representing
broadly its two distinct phases, the first of Indo-Iranian type
flourishing around c. 500 A.D. and the second one, strongly
Iranian reaching its climax about A.D. 600, 650. The Chinese
influence representing the third phase is noticeable in the paint
ings at Kucha between the seventh and eighth centuries. On the
basis of these stylistic differences, two clearly distinguished
spheres of artistic influence in the area of the northern Silk
Route can be clearly marked out. The western one, fairly older
(about A.D. 500-700) centred around Kucha, and an eastern
school of the later times (c. A.D. 650-950) in the region of
the Turfan Oasis. The styles of the former school are especially
illustrated by the numerous wall paintings from the cave monas
teries of Kizol, near Kucha. They reveal Indian and Iranian
influences. The former is seen in the way of composition of
the subjects from the Buddhist texts and in the serene and
tranquil way of the Buddha portrayed in the frescoes. The less
obvious Iranian influence is traced in the pictorial details,
especially of the dress. Crowns and other ornaments are worn
by the bodhisattvas and other deities and many decorative
ribbons are of Sassanian origin; and numerous decorative
features are traced back to patterns from the Sassanian school
of silk weaving. On the other hand in the eastern sphere, the
Chinese influence is pre-dominant, quite clearly revealed in the
facial features as also in the dress of the subjects.
The use of motifs is no doubt helpful in placing any culture
in a historical perspective as also in tracing its origin but it
could be of little help in defining the style of any work of art.
further, in Central Asian art the absence of the personal data
consequent to the anonymity of the artist makes it difficult to
ferret out individual art style stamped in the area noted for
the evolution of hybrid culture. The artist very often himself a
Buddhist monk, had to conform himself to the ideals of the
Buddhist faith in stimulating the religious feelings of the votary
and motivate him to a higher degree of moral conduct as might
help him in coming closer to the goal of his spiritual needs and
aspirations. He no doubt fades into the background with his
personality submerged in his creations. As such, it is only possible
to analyse the pictorial art of different centres with reference

to common features, as also distinguishing them on grounds of
physiognomy, dress, colour scheme, narration and background
and media of light and shade. The characters of inscription, if
any, could suggest the probable date of the pictorial art, as for
instance the earliest paintings in the Painter’s Cave at Kizil
could be placed in c. A.D. 500. A few caves are painted in the
same style as the cave of the Painters, representing the oldest
group. One such being the cave of the Statues in Kizil which
provide the best specimens of the first Indo-Iranian styles; a
cowherd listening to a sermon of the Buddha28 and of Vajrapani.
The paintings show signs of Indian mannerism and style. The
vigorous cowherd leans on a knotted stick, watching over his
animals as he listens devoutly to the words of the Buddha. So
deep is his concentration that he is unmindful of the poor frog
being crushed beneath his stick. The two cattles, one dark in
colour and the other white, lie on the ground behind the central
figure. That of the Buddha shows him seated on the throne with
his hand raised in the gesture of teaching.
Vajrapapi29 in the other cave of the Statues, is shown here
gracefully seated on a wicker stool, holding in his left hand
the vajra, the thunderbolt, resembling the sceptre and with his
right hand fans the Buddha of whose figure only parts have
been preserved. He is elaborately draped in chains and strings
of beads or metal disks. He wears a brown skirt with ruffled
green borders and has long green scarves suspended from his
head dress with their ends almost touching the ground. On his
head is a diadem decorated with beads and discs with a white
band hanging from either side. Plumes of feathers at the sides of
the head dress are also noticed, with a large ornamented disk
in the Centre. Similar diadems appear in the Sassanian art.30
28. The Cowherd Nanda. Kizil Cave of the Statues c. 500. MLK. Ill,
8838; Le Coq & Waldschmidt : Die buddhistiche Spatantike in Muttelasian
1928-33.VI. pi. 3 c and p. 66; Bussagli : Op. cit p. 72; Hertel : Central Asian
Art no. 9, pp. 66-67.
29. ‘Seated Vajrapani’. Kizil Cave of the Statues c.500. MLK 8839.
Hertel ; Central Asian Art, p. 68-69; Le Coq & Waldschmidt : Op. cit, VI,
pi. 36, p. 66.
30. Ghirshman refers for example to a stone relief of the third century
A.D. from Sar Meshed depicting king Bahram II as a lion slayer (Ghirshman,
Only the bare knee of the Buddha is visible at the right edge of
the fragment, who. was seated on a carpeted throne. Below it
are two sharp-beaked falcon-like birds, one perched on the
ground and the other diving from above. Vajrapani’s almond-
shaped eyes, his narrow moustache and his cross-shaped navel
suggest Indian influence.
Both the paintings provide a very specific choice of colour,
with the use of finely distinguished shades of the same basic
hue in a variegated scale from whitish yellow through brownish
and reddish nuances to deep brown, sometimes merging into
black. Contrast is provided by a very bright green colour.
Contour to surfaces, objects and persons is rendered by fine
strokes of the brush. In the rendering of clothes falling into
folds, three-dimensional modelling of the fabric is perceptible,
as might be apparent from the cowherd’s loin cloth. The semi
circular concentric lines in the area of his right thigh shows the
material clinging to the contours of the body. The ground
colour, one shade darker determines the colouring of the lines.
Another illustration of the first phase is the picture of a
young ascetic*31 in his shells of foliage decorating either a vault
or a pendentive in the cave of the Navigator at Kizil. The
elongated, dreaming eyes of this figure, of about the same date
as the cowherd, show stronger Indian influence. The linear
rendering of the hair and the concentric, almost elliptical curves
are most conspicuous while the contrasting colours, the hand
ling of the beard and the ornamentation of the scarf over the
shoulders provide an interesting study. The image of this young
ascetic when compared to that of the mystic one MahakaSyapa32
from the so-called Large Cave at Kizil, at least a century later,
1962 fig. 215, 216). According to the late professor a royal crown decorated
with eagle features was originally the symbol of the Avestian god of victory,
Verethragna, who had the same function as Indra, the Vedic god of war, who
was included in the Buddhist pantheon (quoted. Hertel : Op. cit, p. 68).
31. ‘Young Ascetic’. Wall Painting from the Cave of the Navigator,
circa 500, IB 8389. State Museum, Berlin—Bussagli : Op .cit, p. 74; Talbot
Rice : Op. cit, p. 192, Illus. 181, who traces Indian influence, combined with
Graeco-Roman in this head of a young ascetic.
32. ‘Head of Mahakasyapa’—Kizil Cave above the Largest Cave, 7th
century, MIK.III. 8373a. Hertel : Op. cit no. 20 pp. 82-83; Bussagli : Op. cit,
p. 75.
shows a high degree of stylization with schematized eyebrows,

the folds of the mouth and chin and the shape of the eyes. The
facial features of Mahakasyapa are less powerful in expression.
The head is inclined before a flowered background, the hair is
cut short and shaved back from the temples in two wings. The
eyes, eyebrows and nose are strongly accented. Mahaka$yapa’s
face is framed by a light blue beard of the same colour as his
hair. The ear lobes appear stretched with the weight of the ear
rings once worn. The neck, chin and mouth are heavily lined.
A fragment of the green patch work robe is preserved on the
shoulder. This wall painting seems to form part of the Maha-
parinirvapa scene.33
The group of swimmers from the Cave of the Navigator,34
Kizil, about 500 A.D., now in the State Museum, Berlin is
another example of contemporary painting from that place,
supposed to be different in style from the figure of the monk,
and representing the personalities of different artists. The scene
shows three men swimming among water lilies, the middle one
seems hardly able to keep his head above water while the one
to the right exhibits his powerful strokes symbolising great
determination, which is evident from his expression and
gestures. A comparison of his face with those of the cowherd
and of Vajrapapi reveals stylistic similarity between the paintings
of this cave and those of the cave of the Statues. The elongated
eyes of the swimmers, the stylized swirls of the water as also
line work show Indian influence.
The second phase of Kucha art distinct from the previous one
in conception, treatment of space as also devoid of Indian
stylistic influence, is revealed from the Goddess and Celestial
33. According to Hertel, this wall painting depicting a MahaparinirvSna

scene may be compared with a similar better-preserved painting. It is based on
the Mahaparinirvana Sutra, a Sanskrit text found in Central Asia, and its
parallels in Pali, Tibetan and Chinese. Depictionsof Mahakasyapa are very
often found in connection with the Parinirvapa of the Buddha, while the
expression of great grief is noticeable in his face, it is difficult to suggest if
Mahakasyapa was shown kneeling at the head of the Buddha or at his feet
(ibid p. 83).
34. Hertel : Op. cit, p. 35 with other references from Grunwedel, Le Coq
and Bussagli, p. 73, No. MIK. Ill 8398—State Museum, Berlin.
Musician35 wall painting from the cave of the Painted Floor at
Kizil. Of course, there are Indian elements noticed in both
iconography and composition. This work is considered as the
most beautiful and best-known of all the Kucha paintings. In
a figurative style it is marked by decorative devices. The typical
rendering of the space is against the background punctuated by
stellar motifs, a rain of falling flowers. The Goddess and
musician, the subject of the picture, under a tree in blossom,
encircled by a wide double curve with decorative motif of a
geometric—abstract order enhance the beauty of the painting
as also suggest unity in composition. The style of this painting,
supposed to be a little over a century and a half later, is
different from the previous one on grounds of colours—the way
of their use and space and volume dominating the scene as also
suggestive of the movement of Buddhist thought in the direction
of mysticism. In the Avadana of Rupavati36 from the cave of
the Frieze of Musicians, Kizil, the figures are much less elonga
ted while the perspective and the figurative elements do not
depict any change.
The Kizil art is a composite one with traces of various in
fluences.37 That of Gupta art on the paintings is obvious, in
the rendering of anatonical details. The prevailing colour
35. ‘Goddess and Celestial Musician’. Wall Painting from the Cave of
the Painted floor. Kizil. 600, 650, IB 84206. State Museum, Berlin; Bussagli ;
Op. cit, p. 81 and p. 78 for description and comments.
36. ‘Avadana of Rupavati’ The sacrifice of the Bodhisattva (?) Wall
Painting from the Cave of the Frieze of Musicians, Kizil, IB 8390, State
Museum, Berlin; Bussagli, Op. cit, p. 84 and for comments p. 83.
37. The Kizil paintings are considered most important amongst the art
centres on the northern route, not because of any marked superiority in quality
or individuality of style, but because the surviving examples furnish a complete
time sequence for dating purposes. The earliest Kizil paintings are assigned to
the period A.D. 450-650, and the second phase ending in about 750. As
regards foreign influences that of Sogdia and Bactria are reflected in the cut
of the clothes, especially of the donors. The tunics often have the characteris
tically Sogdian single reverse. Hellenistic influence is traced in the pictures of
the Buddha* Women are, however, closer in'style to Indian than to Iranian
or Greek conceptions, while the men are nearer Sassanian models. Even
during the last phase while the artists were using highlights of the Chinese
type, many of the men still retained a markedly Persian appearance (Talbot
Rice : Op. cit, pp. 190-91).
scheme in the earlier style, closely related to that of the frescoes

of Bamian, is made up of greens and browns, dulled, soft and
without strong contrast. That in the second one consists of
malachite green and a shade of blue and shadows rendered
exclusively in orange. The frescoes at Kizil as elsewhere served
a dual purpose—religious instruction and providing decorative
scenes and an overall picture of Buddha’s life and activities.
The Greco-Buddhist influence is evident in many monographic
details while the undulating elements in the figures along with
refinement and lightness characterise Gupta or post-Gupta
influence. The Iranian influence is as well traceable in the
earlier phase of Kucha art in the form of decorative motifs and
certain details of the dress, emphasised rather greatly in the
second style. The motif of garland bearers in flight echoes late
Hellenistic tradition. This composite nature of art symbolised
with traces of influence from different quarters has indigenous
base, as is evident from the costume and the racial type of the
figure.
Another centre not far from Kizil on the Muzart River is
Kumtura38 where several of the caves are decorated with fres
coes in which the styles of Kizil are copied. These also contain
frescoes reminding of the second style of Kizil or of a later
period influenced by Chinese art. The temples containing these
frescoes are hollowed out of the rock except for a small temple
built in the open, in which was found the lower portion of a
statue of Buddha clad in stylized drapery like the models of
the Gupta period. The structural plans of the temples are
similar to those of the Kizil sanctuaries—predominantly rectan
gular with barrel vaults or square with a domed roof. There
are a good many examples of the first style in a domed cave
with the colour scheme consisting of green, brown and sienna-
reddish brown or brownish yellow with the figures like those
in the cave of the Peacocks and of the Painter at Kizil. The
second style of Kizil is copied at Kumtura in the cave of
Nagaraja with the same stylistic traits and colours. Specimens
of the third style in the caves of Nirvana of the Apsaras and of
38. See Encyclopedia o f World Art, Vol. I, p. 832 for a general des
cription of the site and its monuments. See also Bussagli : Op. cit, pp. 88, 89;
Talbot Rice : Op. cit, pp. 188, 190.
the Kinnaras are apparently influenced by the Chinese art of

the Tang period. The floral ornament is typical of the art of
Chinese derivation in Central Asia, as also of the figures floating
in clouds. The figures have round faces and slaunting eyes. A
typical Chinese impress is in the rendering of mountains and
trees according greater importance to landscape. The colours
in the third style are bright and tempered and the composition
is subdued, different from that of the Kizil paintings. Kumtura,
thus, represents the farthest point of penetration of Chinese
influence adding to the confluence brought about by those from
India and Iran for the service of the Buddha and his religion.
Among the important paintings from this place may be men
tioned the Buddha and the Praying Monk39 (c. 650) and
Divinities of the Tushita Heaven40 from the Cave of the Apsaras
(8th century), and the worshipping Bodhisattva41 (8-9th cent.).
In the last one the Mongol features, the head-dress, the drapery,
folds of dress and scarves—all traditionally Chinese, are pro
minent. In this third style, the composition and colouring are
also being increasingly dominated by Chinese tastes and
tendencies. Besides frescoes, there are sculptures of wood and
clay—the latter moulded around a wooden framework, some
times they formed part of the original decoration of the niche
and ledges in the caves. Mostly these were pious donation to
the monasteries.
Monastic establishments, systematically arranged around a
central court with separate stupas, were traced eleswhere as
well. In the ruins of the buildings, destroyed in fire at Duldur-
Akhur42 not far from Kumtura, were recovered fragments of
statues moulded of Stucco with straw or tow, as well as wooden
39. ‘Buddha and Praying Monk’ from the Cave with a Low Entrance
Kumtura. c. 650.1 B. 9024, State Museum, Berlin. Bussagli : Op. cit, p. 88.
40. ‘Divinities of the Tusita Heaven’. Wall Painting from the Cave of the
Apsaras, Kumtura. Eighth century I B 9021, State Museum, Berlin.
41. ‘Worshipping Bodhisattva’. Wall Painting from Kumtura, Eighth-
Ninth centuries, IB 8377, State Museum, Berlin Bussagli: Op. cit, p. 91;
Hertel : Op. cit, p. 126-127. This fragment is one of the finest examples of
richly coloured mural paintings.
42. For a general description of the site and its monuments, see
Encyclopedia of World Art. Op. cit, p. 833.
sculpture and some frescoes. The archaeological finds are

reminiscent of classical works, while the Chinese influence is
apparent in the wooden sculptures. Fragments of some frescoes
show a certain degree of similarity with the first style of Kizil,
while others reveal Chinese influences. The same phenomenon
of Indian, classical and Chinese influences alternating is also
traced at some other places like the ruins of Hinar and Tajzik
Karauh-Kosur.
Kara-Shahr*3
Kara Shahr also offers a complex of many temples and cave
temples in which fragments of statuary have been found. These
include the remains of a Parinirvana there. It is in the round
and has Indian influences. Important frescoes have also been
found in the cave temples with certain structural variations
from the usual rectangular plan of the second Kizil style. In
the same sector about 25 Kilometres to the south, south-west
from Karashahr extensive collection of Buddhist remains were
traced at Shorchuk and at Ming-oi—‘the thousand caves to the
north of it’. Both the places are of religious and archaeological
importance. The buildings are hewn out of rock and also
set up in the open air. Excavations at Shorchuk44 have brought
out many fired-clay moulds of not only the head but also the
torso and other parts of the body. The reliefs were generally
affixed to the wall, given finishing touches with a chisel, and
finally gilded, while the elegance of the bodies and the anatomi
cal details visible beneath the drapery betray the influence of
Gupta art. The stamp of Indo-Greco-Buddhist art, however,
predominates, especially in the smaller statues which appear
similar to those found at Hadda in Afghanistan and at Taxila.
A seated Buddha in clay from the Kirin Cave in Shorchuk now
43. For a general description of the site and its monuments, see
Encyclopedia o f World Art—Op. cit, pp. 833-34.
44. The geographical position of Shorchuq made it a place o f meeting
and mutual influence from Kucha and the "Chinese one from the east
which were absorbed here with the local artists portraying their specific
character distinguishing their contribution from others. Narration o f
Buddha’s life activities is replaced by t h e ‘three jewels’. The Chinese ele
ments in the paintings of Shorchuq and of the Karasahr generally differ
greatly from those of Kucha. (Bussagli : Op. cit, p. 94).
in the State Museum Berlin45, is a tine specimen of this art.

Seated on a high finely painted pedestal, supported by a rectan
gular block adorned with a pair of floral medallions, the Buddha
is clad in three garments, a transparent under one (antaravasa),
a yellowish brown robe (uttarasafiga) and a reddish upper
covering (sanghatl). The legs, crossed in the lotus posture, are
fully covered; the left hand holds one end of the upper cover
ing, while the right one—broken—was presumably in the pro
tection posture {abhayamudra). The hair is styled in waves,
with the hair line marked in red. The urna takes the form of a
stylised jewel on the forehead. The painted decoration is
unmistakably of Chinese type.
At Ming-oi,48 the number of grottoes, changing into
sanctuaries, was considerably large—may be ‘a thousand caves’
as the word seems to indicate. These shrines were adorned with
sculptures consisting both of large traditional figures of the
Buddhas and Bodhisattvas, as well as of numerous smaller ones,
depicting with precision and clarity people of every, age and
walk of life, with a sensitive grasp of pathos and humour.
While the geographical position of Shorchuk made it a meeting
place of artistic currents from Kucha and China also reflected
at Ming-oi, the local artists, assimilating foreign features,
equally displayed their ingenuity. Their paintings have a specific
character which distinguishes them from many others in Central
Asian Art. The subject matter, unlike the narrative tone at
Kucha, is taken from the lives of the monks rather than of the
Buddha, with deeper interest concentrated in the last of the
three jewels—Sahgha or the ecclesiastical community of
Buddhism—the ‘guardian of the Good Law’ or the Dhamma.
The paintings in general are of exceptionally high quality, with
45. Hertel : Op. cit no. MIK III. 7841 from Shorchuk, Kirin Cave
dated 7th-8th century, for description and cross references—Rowland :
The Art o f Central Asia (New York, 1974, pp. 176AF.),
46. The word Ming-oi meaning ‘A Thousand Caves’ was used to describe
the region in the Shorchuq district. It has come to be associated with the sculp
tures which adorned these shrines. These consist both of large, traditional
figures of Buddhas and Bodhisattvas, as well as of numerous smaller ones
depicting with vivacity and precision people of every age and walk of life
(Talbot Rice : Op. cit, p. 182). See also Andrews : Wall Paintings : Op. cit,
p. 28.
rich and dark colours, predominantly yellow, brown, rose and

green unlike the bright light ones in Kizil art. An extraordinary
variety of poses and physical types in the figures is another
notable feature of the style of Shorchuk47—which is supposed
to be a transitional one between the second and the third styles
of Kucha.
The Turfan Group48 :

TheTurfan region because of its geographical position was
the bone of contention between the nomad empires and China
as early as the time of the early Huns, and was exposed to a
series of ethnic fluctuations and political changes affecting art
development in many ways. The Chinese influence here was
strongly felt. Changes in religious thought left their impact on
remnants of art. Buddhism, Manichaeism and Nestorianism
and subsequent restoration of Buddhism signified changes in
iconography and style. The hold of Buddhism was no doubt
for a longer period of time. Its influence in the Tarim Basin is
supposed to have lingered on even when the forces of Islam
had swept over the neighbouring areas. This was due to the
interest and power of the Uighurs. Buddhism seems to have
survived till the beginning of the 15th century. TheTurfan
oasis in its complex includes religious and civil monuments,
including a great many watch towers, citadels and royal palaces
47. Encyclopedia o f World Art, Vol. I, Op. cit, p. 834. See also Andrews :
Op. cit, pp. 28-39 for reference to fragment of a painting from Ming-oi XIII,
now in the British Museum. This and a few others provide a series of scenes
in which an elderly teacher lectures to men of various ages dressed as monks
and holding a writing tablet and pen or stylus to take notes. Others are seated
each in a separate cave cell, apparently studying or writing. Aurel Stein in his
Serindia provides fuller information on Mingoi and its finds.
48. For detailed references to Turfan and its art, see Encyclopedia o f
World Art, Vol. XIV under Turfan, and also those noticed, mainly dupli
cations under Central Asian Art—ibid, Vol. I, pp. 837-38. Bussagli has a
special chapter on the Turfan Group in his work entitled Paintings o f Central
Asia, Op. cit, pp. 95-112. The history of exploration in the Turfan region is
recorded by Aurel Stein in his Serindia, Vol. Ill, pp. 1149 flf as also his Inner
most Asia, Reprint, New Delhi, 1981, pp. 568-89. In the course of his investi
gations, Stein examined cave and structural shrines at Khara-Khoja, Toyuk,
Bezaklik and other sites. See Andrews: Catalogues o f Wall Paintings, Delhi,
1933 for paintings from Turfan in the National Museum.
connected with the Uighurs. These are traced at Idikut-Shahri,

‘the city of Idikut’, and Khocho, while Mortuk, Bezaklik and
Senghira-aghiz provide best examples of religious art. The
degree of Chinese influence varies in these centres, as for ins
tance Murtuq retained many Iranian and Indian characteristics
and was directly influenced by Kucha and Shorchuk owing to
the monk’s adherence to Mahayanism. At Bezaklik on the
other hand, where Tantric influence dominated, the Chinese
influence was wide and deep. It constituted an artistic and
iconographic phenomena of its own, more complex than at
Kumtura. By oscillating between Iranian and Chinese forms,
Turfan was in fact a link between the tastes and tendencies of
pre-Islamic Iran and aesthetic currents from eastern Asia. The
different phases of this complex art are differently rendered from
one locality to another.
According to Hackin49, three phases can be traced in the
Bezaklik complex : a first period of Buddhist domination cover
ing the seventh and eighth centuries in which the influence of
Kucha was felt. The influence of Tang painting is reflected in
a most prevalent manner in this period. The second period
covering part of the ninth century is marked by Manichaean
domination of the Uighurs with a new influx of Iranian influen
ces. The last period of Buddhist supremacy from the tenth to
the eleventh century or perhaps even later, is characterised by
the setting in of the process of decadence, and for some time
direct representation of some of the Tantric divinities is traced
in certain temples. On the other hand, Murtuk of the same
epoch and close to Bezeklik presents a very different style. Here
Chinese influence is not predominant, but Indian influence is
clearly traced in the figure of the Buddha seated in the European
fashion and accompanied by a Bodhisattva of purely Indian
type holding a vase. The subject matter is taken from legends
connected with Mahayana Buddhism. In the flames surround
ing Buddha and in the coiffure of one of the donors the influence
of the art of Kizil and that of Iranio-Buddhist one is noticed.
49. J. Hackin : Buddhist Art in Central Asia : Indian, Iranian and Chinese
influence from Bamiyan to Turfan—quoted by L. Hambi in his article ‘Asia
Central’ in the Encyclopedia o f World Art, Vol. I, p. 835.
This introductory background of the peculiar features of the

Central Asian art at Turfan might be helpful in this context
noticing a few important specimens of paintings, now in the
Berlin Museum, as also in the National Museum, New Delhi.
The best illustration of western influence in Turfan area is
provided by a wall painting from Khochu dated between the
seventh and eighth centuries and now in the State Museum,
Berlin.50 It depicts the head of a Buddha beneath a canopy.
The shape of the eyes, the bold strokes of the eye-brows and
the dress resemble the sculptures of Shorchuk. Western figura
tive tendencies are revealed in other ways as well. This delicate
and pensive head of the Buddha from Khocho—called Idiguts-
chai 'the city ofldigut’ by the Uighurs in honour of a Turkish
hero of that name, is supposed to be an early example of
Uighur Buddhist painting. It is classified as one in the second
Indo-Iranian style. This beautiful image is inclined slightly to
wards the left and is surrounded by an oval nimbus and
surmounted by an unusual designed canopy.
Another wall painting51 depicts the ‘Great Departure’ of
Prince Siddhartha, who is partly visible on the left riding his
favourite horse Kanthaka ‘away from home to homelessness’.
In the mural, the forehead and bristling hair of one demigod
can be seen below the horse’s head and the fragment of the
figure of another one is noticed at the bottom. The partly
visible face has the same shape of one of the eye, and an
equally bold stroke of the eyebrow, as noticed in the earlier
figurehead of the Buddha. It is supposed to be a later execution
with realism when the somewhat austere colour scheme of the
Uighur art had developed. It is dated in the ninth century.
Central Asia also provides paintings portraying princes
and their families meant for hanging at the entrance or in pas
sages in the cave temples. One such fine painting on cloth from
Khocho, now in the Berlin Museum, is that of a Uighurian
50. Buddha beneath a canopy. Wall Painting from Khocho—Seventh-

Eighth centuries. I.B. 8731. State Museum, Berlin. Bussagli : Op.c it, p. 97;
Hertel : Op. cit, No. 90, p. 154; Talbot Rice : Op. cit, p. 196.
51. The Great Departure. Khocho Ruins. 8th century. MIK .111, 4426—
Hertel : Op. cit, p. 155 No. 92; Bussagli : Op. cit, p. 26.
prince52 on both sides. Like all votive banners it consists of three

parts : the long, narrow, rectangular banner itself; the triangular
segment at the top, which here depicts a seated Buddha; and
rectangular strips of material at the bottom with a wooden stick
to weigh it down. The aristocratic prince with his long hair and
white beard and curved moustache puts on a fine robe, patterned
with a large floral design. This long-sleeved, round-necked gar
ment—the typical one of the country, reaches to the feet and
a belt fastens it in front with rectangular decorative pieces.
A slit in the side of the garment reveals a black knee-length
boot. The prince holds a stem in his hand with several blossoms.
The small squat body of the prince is in strange contrast with
the portraits of Uighurian prince and princess,53 painted on the
walls at Bezaklik, now in the Berlin Museum.
The mourning scene54 in Shrine EX from Bezeklik, dated in
the Eighth century, now in the National Museum, New Delhi,
is very realistic. It shows people mourning at the death of the
Buddha. These include dignitaries, princes and sovereigns of
different nationalities—Arabs, Iranians and Chinese, suggesting
popularity of Buddhism and its communion. Another shrine
No. XII from the same place also shows musicians and mour
ners5556*in a grotesque style.
The Turfan area was also a strong centre of Manichaenism
as also Nestorian Christianity. Most of the important finds by
the four German expeditions to Turfan included thousands of
manuscripts with and without illustrations. Most of the Turfan
miniatures, richly painted in gold and opaque colours are from
Manichaen books.58 The artists portraying these were probably
52. Hertel : Op. cit, p. 196-7, no. 136. Bussagli : Op. cit, p. 105.
53. Bussagli : Op. cit, pp. 106, 107.
54. Bussagli : Op. cit, p. 110; Andrews: Wall Paintings (Bez. XI. A C .
pi. XX, p. xxiv). It depicts a crowd of mourners gathered round the bier of
the Buddha, composed of representatives of the many kingdoms present on
that mournful occasion, and affords an interesting opportunity of identifying
the several types of communities usually at variance, but here united in
expressing a common grief.
55. ‘Musicians’. Wall Paintings from Bezaklik. Eighth century. Private
collection. Tokyo. Bussagli : Op. cit, p. 103.
56. Hertel : Op. cit, p. 174. The most important finds made by the four
German expeditions to Turfan were undoubtedly the manuscripts, both with
Sogdian. As Manism was based on Iranian dualism mingled with

elements from the Buddhist and Christian creeds, its philosophy
found its expression in art as well. The largest fragment of a
Manichaean miniature in the Berlin collection,*57 painted on
both sides, depicts on one side a Church ceremony. In the back
ground the principal character (his head being destroyed) ap
pears to be a high-ranking Manichaean priest in full vestments
with a prince or a king kneeling before him and three of his
attendants standing behind him. In the foreground are portrayed
Hindu deities : the one with the elephant’s head is Ganesa, the
boar’s head might represent Visnu as Varaha, the third and
fourth could be Brahma and Siva respectively. Facing them on
the left are two Iranian Manichaean gods. The miniature on the
other side depicts a religious celebration, the famous feast of
Bema, celebrated every year as the rnartyrdom day of Mani.58
It is suggested that even when the iconography became Mani-
chean and, by religious affinity, adopted Buddhist or other
Indian forms, which it promptly rearranged on new lines and
with new values, the encounter of the cultures of India and Iran
in the region of Turfan remained abundantly clear. A particular
stylistic idiom was born out of this encounter which moved
westwards and profoundly reacted on later Islamic art.59
and without illustrations, on various materials, such as palm leaf, birchbark

and paper. These provide significant information on religious, political and
philological aspects of eastern Central Asia. Mani’s teachings symbolised a
syncretism embracing many components in the form of a philosophy of nature.
It was based on Iranian dualism mingled with elements from the Buddhist and
Christian creeds.
57. ‘Leaf from a Manichaean Book’. Khocho Temple, 8th-9th century.
Manuscript Paintings, MIK.III.4079. This is the largest fragment of a Mani
chaean miniature in the Berlin collection and is painted on both sides. H ertel:
Op. cit, no. 114, pp. 176-177.
58. It was probably in the spring, for St. Augustine (who was himself at
one time a Manichaean) tells us that Mani died in March 276. A podium was
erected for the liturgical rites, with five steps leading upto it. It was draped with
sumptuous tapestries. The celebration fell into two parts : first, on the eve of
the faithful fasted in preparation and kept a vigil during the night in remem
brance of Mani’s death followed by the feast day itself (Hertel : Op. cit, p.
177).
59. It is suggested by Talbot Rice that Turfan with the Sassanian elements
in its sculpture and painting, together with Kucha, Kizil and Pendzikent surely
There are several other sites in the Turfan area including

Sengim and Toyuk as also Khocho and Kharakhoja which have
been centres of Central Asian Buddhist art. The wall paintings
at Sengim reveal a synthesis of different art influences. The
figure of naksatras, female lunar divinities, with their fluttering
scarves illustrate in Grousset’s words, 'the most felicitous synth
esis of Indian sinuosity, Hellenistic elegance and Chinese grace’.*60
These paintings are found in monasteries, and are well-preserved.
At Toyuk, also visited by Klementz and German and French
expeditions, Stein discovered61 fragments of Stucco reliefs there
of various types and sizes including a great many small heads
and ornaments in the form of necklaces. A painted ceiling in
the sixth century portrays a seated Bodhisattva surrounded by
rays with floral decorations like those in the Turfan sanctuaries.
Plurality and diversity of artistic life have been the keynotes
of Turfan’s contribution. Many artistic tendencies converged in
this area and had their impact on the Buddhist art. The type of
drapery, reduced to concentric folds often used as Pendzikent,
seems obviously connected with the art of Sogdiana. The style
of this region was to some extent determined by the synthetic
religious development, although one could trace differences in
Buddhist compositions of Bezaklik and Murtuq62 and those of
the Christian temple at Khocho. Forms of figurative expression
far removed from each other because of the different inflections
in their religious feelings were adopted. The Buddhist system
influenced both the Buddhist art of Tibet and the Islamic art of the Ghaz-
navids, Samanids and Seljukid Turks. Mongol painters appear, on the other
hand, to have been more deeply affected by the Vighurs, for the figural art
which the latter created did not die out when Khocho and Tun-huang declined.
(Op. cit, p. 203).
60. Encyclopaedia o f World Art, Vol. I, p. 836. For reference to paintings
from these sites, see Hertel : Op. cit, nos. 114-125, 136, 137, 138, 139;
(Khocho); 140, 144, 151 (Murtuk); 140, 144 (Tuyok); Andrews : PI. XI,
pp. 64 ff (Khara Khoja).
61. Innermost Asia, pp. 616 ff.
62. For Bezeklik and Murtuq finds see Stein : Innermost Asia, Op. cit,
pp. 633 ff. and also Andrews : Wall Paintings Op. cit Introduction pp. 32 ff.
and Plate VIII & IX. Those from Bezeklik are also noticed by Andrews ;
‘Introduction’, pp. 51 ff; Plates XII-XXXI. These are also mentioned by
Hertel : Op. cit nos. 81 -85, 108-110.
no doubt provided a wider canvas for the artists to display their

talents and the religious experience of the peoples was a great
asset in this direction. TheTurfan region on the eastern border of
ancient Kashgar was the last province of Central Asia to retain
any creative vitality at a time when the artistic activity in the
surrounding region had faded or was on the verge of extinction.
Tun-huang63
The great monastic agglomeration of Tun-huang further east
in Kansu, lying on the last stage of the caravan routes represents
a curious blending of Indian imagery and symbolism into the
Chinese style of painting. During the third and fourth century
A.D. numerous monasteries had come into being there and in
these a great many shrines and religious statues were set up.
Many of the shrines were located in caves and its walls and
ceilings were covered with paintings. Sometimes these also ap
peared on stupas. The painters here in Tun-huang adopted
Miran’s style and iconography as the basis for their own school
of art. In fact, Miran was the centre of artistic talents and these
influenced as well as helped working in other areas. Tun-huang
became a very important centre of Buddhist art. Buddhism reached
there by way of both the northern and southern Taklamakan
routes. The monastery founded in c. 333 on a site abounding
with grottos and caves, gradually expanded in number and space,
finally culminating in that of a thousand Buddhas and extended
all along with the entire area presenting a vast complex. Most
of the buildings have now crumbled with no decorations. The
63. A. Stein : The Thousand Buddhas'.Ancient Buddhist Paintings from
the cave Temples o f Tun-huang, London, 1921-22; A. Waley : A Catalogue
o f Paintings recovered from Tun-huang by Sir Attrel Stein, Oxford, 1931;
B. Gray : Buddhist Cave Paintings at Tun-huang, Faber & Faber, London,
1959; J. Hackin : Guide Catalogue de Musee Guimet—Les Collections
Bouddhique, Paris, 1923—Chapter II—Touen-Hovang, pp. 33 ff; Mission
P. Pelliot : Le grottes de Toiuien-houang, 6 Vols. of Plates Paris 1920-1924.
See also Bibliography under Asia Central—Encyclopedia o f World Art, Vol. I,
p. 837-8 and under Tun-huang (ibid) Vol. XIV pp. Bussagli : Op. cit pp. 115
ff; Stein : Serindia, Vol. III. Appendix E entitled ‘Essays on the Buddhist
Paintings from the Caves of the Thousand Buddhas’. Tun-Huang by Raphael
Petrocci and Laurence Binyon pp. 1392 ff. The fourth ‘Essay on the Art of
the Tun-Huang Paintings’ by Laurence Binyon, pp. 1428 ff.
surviving ones, however, contain the statues of Buddha and

Bodhisattvas which were set up there many centuries ago. These
are mostly made of stucco modelled with great skill,
sometimes executed in the Greco-Bactrian style, sometimes in
that of Gandhara. The paintings take the form both of murals
and of figural compositions executed on either silk or gauze. In
the earlier works the iconography conforms to the pattern of
the Chinese artists of the first half of the fifth century, but both
the paintings and sculptures range in date from Tang and Sung
to Yang times.
The Tun-huang paintings are illuminating documents for the
study of Mahayana Buddhism, but of yet greater interest for the
study of art. They provide an idea of Chinese Buddhist art in
concept and design. The paintings, despite a monotony of sub
ject matter exhibit a considerable variety of style. Purely Indian
art is represented by a small group of paintings which are
probably Nepalese; on the other hand, there are a number of
entirely Chinese ones as well. Between these two extremes are
pictures of an intermediate style—the productions of the local
schools of Turkestan, or in some cases a provincial Chinese
school. A few works are Tibetan. These frescoes at Tun-huang
belong to a period extending from the 5th to the 8th century
and may be divided into two groups : the first one of the Wei
period—between the 5th and 6th centuries—was executed in a
matured and perfect style, rather unsurpassed thereafter. That
of the Tang period, between the 7th and 8th centuries, of the
second style, is characterized by heaviness in the contours sug
gesting emergence of a provincial school of art, no doubt asso
ciated with skilled craftsmen. Besides frescoe paintings, those
on silk of the Tang period contain some of the masterpieces.
Some traces of western influence are noticeable in the rendering
of certain figures and the clothes, specially the scarves similarly
treated as in the paintings of Kucha or Turfan. While the treat
ment on the whole remains predominantly Chinese, most clearly
apparent in scenes illustrating the lives of monks and in the
floral motifs, the draperies, poses and facial expressions, as also
linear treatment and grouping suggest the impact of the Graeco-
Buddhist school and kinship with India.64 The colours promi-
64. Talbot Rice : Op. cit, p. 219.
nently used are malachite green, pink, pale blue, white and black
set with a red ground.
The Tun-huang pictures illustrate the mixed influences
prevailing there, of which an interesting example is that
large painting65 showing AvalokiteSvara seated on the lotus,
with an infinite number of eyes and hands symbolising the
infinite area of his compassion. Above are the sun and the
moon; and below are two demon kings wreathed in flame.
There is a border around the picture on which blossoms of
flowers are painted. Flowers are also shown dropping through
the air which is a favourite motive in Buddhist art. The
colours are glowing. In another painting of Avalokite-
svara,66 this time conducting a soul, a difference of mood,
style, genius is clearly manifest. In this suavity and flexible
movement replaces heavy symmetry in the composition as
also ‘solid hardiness’ in the drawing. Flowers seem floating down
in the air. The Chinese genius is traced in the instinct for living
movement, and love for sinuous line. Some of the large paintings
are seen to be repetitions on silk of the broad style which is seen
in the frescoes found at Tun-huang and other sites. The main
point to grasp, as pointed out, by Laurence Binyon is that the
tradition of Buddhist art which we find first formulated in
Gandhara after assimilating certain minor elements (chiefly
Iranian) in its passage across Eastern Turkestan, was transfor
med in China by the genius of that country’s art. Indian ima
gery and symbolism, Indian ideals of form were taken over by
the Chinese masters. The examples show that the Indian material
was fused in the Chinese style, and a really new phase of
Buddhist art was the result. The Buddhist pictorial art is con
strued by some as merely a continuation of the art of Gandhara,
enriched and shaped by the Chinese master painters according
to their formula evolved by the indigenous creative instinct.
Soviet Central Asia67
The western part of Central Asia had a cultural ethos of its
own and its art forms were handled in an original manner.
65. Stein. Serindia. PI. LXIV
66. ibid, pi. LXXI.
67. For the history of the ancient archaeological sites and the material
finds in these areas—see Aleksandr Belenistsky : The Ancient Civilization o f
About the beginning of the Christian era, Buddhism had conquer

ed part of eastern Iran and engendered a Greco-Buddhist art
which had originated about the end of 1st century B.C. in the
Gandhara region. It lasted in the same region until the 7th
century and to a limited degree in several localities until the 8th
century. It gradually spread to Serindia and China, and reached
Sogdiana through Bactria. Archaeological explorations and
excavations have now brought to light a number of sites in
Russian Turkestan, which were centres of urban culture and
have also revealed existence of Buddhist monasteries and san
ctuaries. The spread of Buddhism in Central Asia was an event
of outstanding importance in the Kusana period, and the
ancient sites testify to this fact. Among the important sites ex
cavated so far may be noticed Termez, Karatepe and Ajina-tepe
in the Amu Darya valley, Kafir-Kala and Kaji-Kala in Tadji
kistan, and Afrasayab and Pendjikentnear modern Samarkand-
Uzbekistan.
Termez (ancient Tarmita), an Uzbek site, situated on the
Vaksha river, was an important Buddhist centre in Central Asia
where a number of Buddhist monuments were discovered.
Fragments of stone statues of the Buddha executed in the
Gandhara style were found at the site. The excavations revealed
Buddhist temples with bronze lions. The recent discovery of
wall paintings, at Balatik-Tepe, near Airtam Termez, provides
interesting study. These are dated in the fifth century and
supposed to be contemporary with the earliest of the Tun-huang
paintings. Certain features, such as the ornamental motifs on
garments, the crowns with veils and bells as well as the
small kusti, completing the clothing of the figures, are supposed
to be of Sassanian inspiration. They confirm close relationship
between this peripheral region and the expanding Sassanian
Central Asia, London; A.L. Mongait : Archaeology in the U.S.S.R. (Pelicans).

Bussagli in his work ‘Painting o f Central Asia' notices Pjandzikent and the
influence of Sogdiana in a single chapter. The comprehensive bibliography
provided in Belenitsky’s book, includes a few in English language as well,
such as L.I. Albaum: Balalyk-Tepe, Tashkent, 1960; O.M. Dalton : The
Treasures o f the Oxus, London, 1926; A.V. Gudkova: Tok-Kala, Tashkent,
1964; G.A. Pugachenkova : Khalchayan, Tashkent, 1965.
world.08 The spirit and style of the figures, however, are suppo
sed to be original and independent which might have been
the result of East Iranian creations carried eastwards with ex
panding trade and the economic predominance of the western
state? of Sogdiana, Ferghana and Chorasmia. The Greek in
fluences often prevailed, and very frequently contained the rising
tide of Sassanian Persia, as pointed out by Talbot Rice.6869 The
earlier Indo-Hellenistic blend is noticed at its best in the frag
ments of a superb sculptured lime stone frieze of the first
century A.D. discovered at Airtam,70 a fortified Buddhist
settlement, located, some 18 Kilometres north-west of Termez
in Soviet Central Asia. It is supposed to have adorned a Buddhist
monastery. The frieze displays a row of youths and girls shown
half-length bearing garlands and musical instruments of local
origin. The style is clearly Kushan and the workmanship local.
Hellenistic influence is traced in the use of acanthus leaves to
separate the youths and girls. The Indian and Hellenistic ele
ments are reflected both in the modelling of the people and in
the choice of such decorative motifs as garlands, but their vital
rendering points to the hand of a local artist.
The ruins of Khalchayan71 in Southern Uzbegistan (near the
river Surkhan-Darya) to the east of Termez also provide intere
sting details of archaeological and artistic importance. The finds
include some fragments of clay sculpture which had decora
68. Bussagli : Op. cit, p. 35.

69. Op. cit, p. 131.
70. The first fragment of this frieze was discovered casually by a soldier
unconnected with archaeology. In October 1932, at the village of Airtam
(Ayrtam) on the right bank of the Amu-Darya, 8 miles above Termez, a
carved stone slab lying in the water was noticed by a soldier of the Soviet
frontier forces. Masson studied it and published a special monograph on the
subject. Later on, in 1936, seven other fragments of the frieze were discovered
at Airtam during excavation of the remains of a Buddhist shrine by the Termez
Archaeological Expedition led by M.E. Masson. The slab of marble stone is
some 20 inches high and has a total length of about 23 feet. On it are carved
in high relief high length figures of male and female musicians and bearers
of offerings, each figure being framed in acanthus leaves (Belenitsky : Op. cit,
pp. 98-99 pi. 49).
71. Belenitsky : Op. cit, pp. 99-102. For a detailed study of the site and
excavations, see G.A. Pugachen Kova : Khalchayan, Tashkent, 1965.
ted the walls of the iwan and the main hall. These included be
sides gods and goddesses—Athena, Apollo and Satyrs etc.—
various personages belonging to the native population of the
area. Along the top of the walls ran a frieze of garland carried
by boys like Italian putli with girls, musicians, dancers, satyrs
and other figures connected with the cult of Dionysus. In addi
tion to the sculpture, small fragments were also found at Khal-
chayan, with remains of human figures and many decorative
details. This site and its monuments are supposed to represent
an early development of the style which Schlumbergar calls the
dynastic style, as opposed to the temple art of Buddhism.
Toprak-Kala72 is another site of importance which was ex
cavated by Tolstov and his team of archaeologists. Situated
near Termez, the town was under occupation from the first
century A.D. until the sixth century as the capital ofChorasmia.
The town was noted for its large and busy market and contained
several hundred rooms, and was defended by three massive
towers. Certain sculptural finds including head of alabaster—a
stone readily available in Afghanistan—suggest stylistic affinity
to the Indo-Hellenic art of the Kusanas. A clay statue—life
size and painted—from Toprak-Kala, found in the ‘Hall of
Kings’ is said to be directly based on Hellenistic example. Stucco
and painted decorations, though much deteriorated, seem to
have been inspired by Hellenistic works.73 Both in style and
detail the decorations closely resemble those of Pjendzikent. The
finest murals appear in the ‘Hall of Kings’. Here the walls had
pinkish borders with white lilies painted above them on a blue
ground. In the 'Alabaster Hall’ the walls were decorated
with floral motives cut and engraved on alabaster. Next
room—the ‘Hall of Victories’—was decorated with sculptures
72. The site was discovered by S.P. Tolstov in 1938 and was excavated in
1945 and subsequent years. Summary in Belenitsky’s work : Op. cit, pp.
102-03 and also in A.L. Mongait : Archaeology in the U.S.S.R.—Op. cit,
pp. 239 IT. The excavations at Toprak-kala have shown the high level of in
dividual character of the artistic culture of Chorasmia.
73. See Talbot Rice,: Op. cit, pp. 118 fT. The sculptures are so correct
anatomically and so naturalistic in conception that they may well have been
inspired by Hellenistic works. They are of stucco and form an integral part
of the painted decorations.
296 Buddhisnn'n Central Asia
of seated kings feasting in thecompany of the goddess of victory.

The ‘Warriors Hall’ contained sculptures of warriors wearing
chainmail armour painted black and holding cane shields.
Other decorations included a painted frieze of gryphons, dancing
couples, vividly rendered musicians and a decorative piece of
composition of women gathering pears and grapes. Among the
fragments recovered from the filings were paintings of tigers,
horses and birds.
Pendzhikent74—the Hephthalite capital, some 65 Kilometres
south-west of Samarkand in the Zaravashan valley, founded in
the fourth and fifth centuries A.D. had a long history ofpolitical
vicissitudes of fortunes. It was at the height of its prosperity
in the seventh century, followed by its destruction at the hands
of the Arabs—its subsequent recovery and lingering on until
the ninth century, when it finally declined. The wall paintings
of Pendzikent adorned the larger buildings, temples and princely
dwellings. Their subject matter is often puzzling, with contra
dictory interpretations provided by scholars of Manichaean or
Buddhist interests. The mural decorations include both religious
and genre scenes. The former suggest a somewhat different
form of Zoroastrianism from the one practised in Persia, with
certain local forms of pagan sun and moon worship. The genre
scenes—concentrating on ordinary life—provide interesting in
formation. These are supposed to illustrate national epics. Some
of the paintings are said to relate some specific passages in
Firdausi’s Shdhnamd. Talbot Rice notices a wall painting from
Pendzikent 7576in which two persons are engaged in a game of
chess. He suggests that it has a symbolic meaning and is proba
bly based on an incident in Buddha’s early life. Another one,
something strangely evocative and appealing70 shows two
riders accompanied by attendants. The marked poetic
74. The ruins of Pendzikent form a complex archaeological pattern

with the ruler’s citadel, the town proper, a sub-urban settlement and the
necropolis as four clearly demarcated areas. Systematic excavations were
undertaken after the war in 1946 under the leadership of A.Y. Yakubousky.
75. Talbot Rice : Op. cit, illus. 88 p. 105.
76. ibid, illus. 90 p. 108.
quality in this painting is equally reflected in another wall paint

ing77—of a harpist.
Afghanistan78
This land-locked country occupying the central position
between the Middle East, the Indian sub-continent, Central
Asia and the Far East is described as a typical ‘round about’
civilization with trade routes and national interests converging
and diverging on it. Its importance has been not only in the
absorption and transformation of foreign forms but in their
dissemination throughout the ancient oriental world. The
relation of Afghanistan to Buddhism and Gandhara art centres
round the Kusana rulers, especially Kaniska, whose empire
extended from Khotan in Central Asia to Bihar in Eastern India
and comprised men of different nationalities and religions. It
is, however, proposed by Talbot Rice that when Buddhism
reached Gandhara, Hellenistic ideals were so firmly entrenched
that converts to the new faith must have turned to western
artists for the new votive statues they needed. The impact made
by the Buddhist sculptures produced by Hellenistic artists was
powerfully felt not only in western and eastern Turkestan but in
India as well. Buddhist art is supposed to have originated in
Gandhara from where it spread into Central Asia and eastward
into China. The Hellenistic influence retained its hold over this
art for a number of centuries. Economic prosperity and safe
communication consequent to the silk trade, however, led to
migration of western sculptors to the East. In the process of
rendering the Buddha figure, the Hellenistic element marked by
a touch of Parthian vitality underwent a measure of Roman
influence, with the cross-legged sitting Indian posture. Western
influence is also traced in the use of cointhian columns and
acanthus leaves for separating each scene. Broadly speaking it
can be said that the setting of Buddhist themes is largely Indian
77. ibid, illus. 92 p. 109.
78. For a comprehensive study of ‘Art in Afghanistan’ see Benjamin
Rowland’s book on the subject with the same caption, London, 1971, which
has also a comprehensive bibliography relating to Afghanistan in general,
as well as on individual centres and aspects of art. See also bibliography in
the Encyclopedia o f World Art, Vol. I under Afghanistan.
but the treatment is Greek. The Gandhara school that origina

ted in the second or the first century B.C. and lasted until the
end of the fifth or the sixth century A.D. exercised a profound
influence over Buddhist sculptors working in eastern Turkestan.79
Central Afghanistan during the Kusana period, was domina
ted by Buddhism and its specific culture. The typical example of
this culture is represented by Bamiyan,80 situated in the valley
between the Hindukush and the Kohi-Baba ranges. It occupied
in its heyday an important position on the trade route from
Bactria to Taxila. The two immense statues of Buddha represen
ted as Lokottara, the Lord of the World, cut in the rock at the
eastern and western approaches of the town dominate the
Buddhist complex in this region. The cliff between them cover
ing about a Kilometre in circuit is honeycombed with a conglo
meration of caves chapels, assembly halls and cells for the
Buddhist monks. Some of these grottoes are connected by
galleries within and along the front of the precipice for purpose
ofcircumambulation. The fifty-three metre Buddha, like the
smaller colossus, has provided access to its summit through a
system of stairways which was destroyed with the crumblingofa
great mass of the face of the cliff. At the time of Hsuan-tsang’s
visit the great Buddha shone golden in the sun—probably gilt
coloured. The Buddhist community at Bamiyam followed the
small vehicle or Hinayana, but the pilgrim Hu-i-chao who visited
that centre in the eighth century describes the monastery as
belonging to the Great Vehicle.
79. Op. cit, p. 149. For Buddhism in Gandhara see Talbot Rice’s inter
pretation. Op. cit. pp. 154-162.
80. Talbot Rice : Op. cit, p. 161. The most creative phase of Gandhara
art probably lasted from the first century B.C. to the second century A D .—
the time of Kani?ka and his association with Buddhism. Though Shapur I
destroyed the Kushan dynasty in c. A.D. 241, the school lingered on after the
original impetus had spent itself and probably owed its long survival to Roman
conservatism and influence. The Hellenistic tradition continued to hold its
own in the east for as long as Roman merchants retained control of the oriental
trade. That Romans or Rome’s eastern citizen, remained in touch with what
are now western India (Pakistan) and Central Afghanistan is proved not only
by the presence at Begram of Syrian glass of the fourth century and by the
discovery of hoards of Roman coins as far as Saripirl, on the northern slopes
of the Hindu Kush, but also by the predominance of Roman influence on the
late sculptures discovered at Hadda. (Talbot Rice p. 161)
Bamiyan is also noted for its painting. The surviving examples

reveal a synthesis of different styles and trends from Gandhara,
Syria, Sassanian, Iran and Gupta-India. Classical traditions are
assimilated by Buddhism. These wall-paintings are divided into
cycles of Indian, Sassanian and purely Central Asian styles.
The paintings still decorating the top of the niche and the soffit
of the vault above the smaller Buddha statue are entirely
Sassanian in style. The massive bulk and frozen lifeless dignity
characterising the reliefs of the Iranian kings are translated here
into painting. Sassanian influence is also traceable in the ipassive
figures of donors alternating with figures of Buddha, as also in
the essentially flat, heraldic patterning of forms noticeable in the
enormous decoration of the ceiling of the niche. The second one,
namely the Indian style of painting, bears resemblence to the
surviving examples of wall-paintings in India before the fifth and
sixth centuries A.D. The fragments of the decoration one
clothing the entire niche and vault of the Colossus Buddha are
supposed to belong to this group. The side walls of the niche
from top to bottom were painted with row upon row of figures
of seated Buddha, each in different and characteristic muclra.
Above may be seen medallions with flying divinities and finally*
on the vault of the niche, the whole pantheon of Bodhisattvas.
The third style of painting bears resemblance to paintings at
Kizil and other sites in Turkestan with the combination of line
drawing and areas of flat and brilliant colour providing closer
resemblances. The decorations surviving on the vault of the
bigger Buddha statue and in the adjoining caves belong to
a provincial Iranian manner; while the disintegrating frag
ments in the niche of this Buddha figure are supposed to be
local versions of the Indian style of the Gupta period. The blue
beautiful Bodhisattva of Group E is considered as a precursor
of the decorative heraldic mode of Kizil in Central Asia.
Practically all the colours used at Bainian were locally
manufactured ones with the exception of indigo which was
copiously used in the painting on the nicheof the Great Buddha.
The procedure adopted here was first the application of a thick
layer of mud to the rock wall with large pieces of chopped
wood in it to hold it firmly to the surface. This was followed
by a thin layer of finely powdered gypsum being spread
over it. On this slip the pigments wereapplied in tempera painting.

The same process was employed in Turkestan and at Tun-
huang. The dating of ihe paintings at Bamian is provided by the
evidence in the style81 of the two giant Buddhas. The smaller
one with the hair in Apollonian ringlets and the drapery in
rather sharply defined classic folds corresponds to the end of the
period of greatest Roman influence in Gandhara proper, suppo
sed by Rowland to be from the early second to the end of the third
centuries A.D. The lesser colossus is placed not earlier than
the year 200 and the larger one about the fifth and sixth centur
ies A.D. The small fragments of painting, recovered from Group
G at Bamian, to the east of the lesser colossus and actually slightly
in front of the main cliff, are supposed to be oldest. The paint
ings recovered froip these ruined vaulted grottoes formed part
of an illusionistic scheme of decoration with the combination
of sculpture and painting.82 These may be dated in the seventh
century A.D.
81. See Chapter VII of the Art o f Afghanistan by Benjamin Rowland
with the bibliography provided at pp. 65-66 which includes Godard, A and Y ;
and Hackin, J : Les Antiquities bouddhiques de Bamiyan—Memoirs de la
Delegation Archaeologique Francaise en Afghanistan, II, 1928; Hackin and
Carl, J. : Nouveilles recherches arch eoloques a Bamiyan—ibid.III. 1933;
Rouland, B. The Wall Paintings at Bamiyan, Central Asia and Ceylon (Boston.
1938).
82. Rowland : Art in Afghanistan, p. 35 IT. A stylistic analysis of the Wall
Paintings of Bamiyan is made by Akira Miyaji in the Report of Japan-
Afghanistan Joint Archaeological Survey for 1974— published in 1976, pp.
17 IT. According to the Japanese scholar, there are four syles of Bamian Paint
ings. The first one confined to the paintings in the inche of the West Grand
Buddha received the tradition of Gupta art and digested aspects of far
Hellenistic-Roman art. It displays the most classical and naturalistic aspects
of all the paintings. The sillful drawings of the Bodhisattvas, sometimes show
ing the tribhahga form and in three dimensional representation, are painted
in calm colours with red brown outlines on a dark background. The second
style shown by the paintings in Caves I, 140, 176, 223 has skilful outlining,
detailed depiction, sensualistic shading, bright polychrome and a tendency
towards design and deformation. Bright colours such as blue, green and Ver
million are used. Lotus becomes a quite schematic pattern. Folds of shawls
are schematically depicted. The painting of this style seems to reflect trends
seen in Central Asia, Kizil, Khotan etc. The third style seen in the paintings
of the niche of the East Grand Buddha. Here the artists abstain from three-
dimensional depiction and concentrate mainly on the frontal unity of the
One of the series of caves immediately to the west of the

smaller colossus, in the vestibule of Group D, had the room
originally painted with a design made up of medallions filled
with human heads, boar’s heads, birds and floral motifs, all
supposed to be devices of a purely Sassanian origin. Every
detail represented on this ceiling could be traced to prototypes
in Sassanian stucco and textile decoration. This design was also
used in weaving and its surviving specimens from Turkestan
suggest its popularity. These are assigned to the seventh century
A.D.
Another Buddhist centre of pictorial art in Afghanistan was
the ravine of Kakrak,*83 several kilometres to the cast to the
cliff of the Great Buddhas. The painted decoration of these
grottoes has been removed in its entirety and shared by the
Musie Guimet and the Museum at Darul-Aman. The paintings
at Kakrak are the earliest known examples of the mystic diagram
of mandalas of esoteric Buddhism, in which the cosmic Buddha
Vairocana or his regents, the Buddhas of the Four Quarters,
are surrounded by the galaxy of their emanations. Each of the
domes at Kakrak was painted with a Buddha image in the centre
of the vault, surrounded by a ring of medallions enclosing
smaller Buddhas. An outstanding work is the so-called ‘Hunter-
king’ painting,84 adorning the drum of a dome. It is remark
able not only for the richness and variety of its colours, but
also for its lively drawing, intelligent stylization and a definite
picture. Dramatically opposed to the classical first style, it seems to be in

fluenced by Sassanian art. Bright flat colours such as blue, yellow, red-brown
and white are used. The fourth style has pseudo-naturalistic or ornamentalistic
features and is the latest of the group, with the use of rough flat colours—pale
blue, light yellow, reddish brown and white. The outlines are considerably
thick. Generally the work is gloomy. Lotus figures are depicted in pairs. It
is the most retrogressive of those found at Bamiyan.
83. According to Rowland, the style of the heads of Bodhisattvas, Yak-
shas and arhats from cave G is extremely fresh and cursive in execution,
vivid in colour, and the actual types appear like a stylized translation of the
familiar forms of Gandhara sculpture into painting. The head of an arhat is
similar to many portrayals of holy men in the sculpture of Tumshuk in Central
Asia. These fragments appear to be related to a single remnant of Gandhara
wall paintings recovered by Tucci in Swab. (Art in Afghanistan, p. 39)
84. Talbot Rice. Op. cit, p. 166. no. 151.
touch of refinement. The main figure, a haloed sovereign wearing

a crown ornamented with three crescents is seated cross-legged,
holding the bow in his hands and the arrows sticking on the
ground. The architectural motif of columns and pediments is
suggestive of Gandharan inspiration. The Buddha figures in two
different postures with haloes are seated on lotus fiowers. The
bright cotrasting colours are in harmony with each other. This
painting is dated in the sixth century A.D. when the Gupta
influence in pictorial art had been completely assimilated. The
employment of an Indian type of abstract shading in the painting
of the ‘hunter-king’ and other fragments from Kakrak suggest
the fusion of Iranian and Indian elements found at Bamian, a
hybrid style that is the direct antecedent of the Buddhist paint
ings in Central Asia. The varied styles and techniques of the
paintings at Bamian and Kakrak closely resemble those of the
wall paintings of Kizil and Murtuq. These point to the role of
Afghanistan in the diffusion of influences to Central Asia and
the Far East.
Fondukistan85
The Indian styles of painting and sculpture at Fondukistan
in the hidden valley of Ghorband brought to light by the French
mission in 1936, provide an interesting study. The chapels in
this monastic area, arranged as complete iconographical
sculptural ensembles, are packed with figures set off by gaily
painted backgrounds. These statues were simply moulded of
unbaked clay, mixed with straw and horse hair as a binding
medium and built up around a wooden skeleton or armature.
The material replaces the use of stone and lime plaster. The
statues were made in a technique similar to one used for
manufacturing large unbaked bricks in the Ghorband valley
even now. The Indian character of the images is reflected by
the soft and sensuous opulence of the anatomical form and the
85. Rowland. Op. cit. pp. 43 ff. Chapter 8 entitled Buddhist Art o f the
Pilgrimage Roads : Fondukistan with bibliography at p. 66. See Hackin,
J : Le Monastere bouddhique de Fondukistan in Diverse Recherches archaeo-
iogiques en Afghanistan—Memoirs—op. cit (MDAFA). VIII. 1959, published
in English in Journal of the Greater India Society. VII. 1940; The description
is mostly based on these works.
elegant rhythm of their swaying postures. The seated figure of

a Bodhisattva, now in Musi6 Guimet, in a position of volop-
tuous ease and soft modelling with the warmth and breathing
fullness of the bodily form, is one of the best examples of
Indian sculpture from Fondukistan. The exquisite flower-like
gestures of the hands and the radiant face suggestive of serenity
and spiritual calm appear to be in agreement with the godly
concept of the great icons of the Gupta period.
The best specimen of painting is that of Bodhisattva Maitreya
—with its thoroughly Indian character. The position of the
body suggests its borrowing from movement in the Indian
dances with the same elegance of gestures and the same fusion
of sensuality and abstraction in the contour of the body as may
be seen in the Bodhisattvas and Devatas in Indian caves. The
figure is more provocative in its exaggeration of the body. It is
proposed by Rowland88, that stylistically and spiritually the
style is strangely parallel to the Tantric forms of Pala art in
India, and perhaps both are the eventful outgrowth of qualities
inherent in the Gupta sculpture and paintiug of India. The
inner stronghold of this late Buddhist style may be placed on
an arc running from Kashmir through Central Afghanistan and
beyond this in a concentric zone in Central Asia, including
Kizil and Tumshuk where it becomes desiccated and patter-
nized.
Begram8687
The ruins of Begram, ancient KapiSa about seventyfive
kilometres to the north of Kabul near the banks of the Panjshir
River, were discovered by the French mission which carried
on excavations from 1937-1940. A vast treasure trove of objects
from the Mediterranean and the orient were discovered inside
two rooms. The extraordinary collection of wares included
glass vessels from Syria or Egypt, bronze vessels of Western
86. Op. cit. p. 48.
87. For a general study of the excavations at Begram, see Ghirshman. R.
Begram, recherehes archaeologiques et historiques sur les Koushan’s—Memoirs
(Delegation Archaeologique Afghanistan) Op. cit XII. 1946. Hackin, J.
Nouvelles recherehes archeologiques in Begram (1939-1940) ibid. XI. 2 Vols.
1964. See also Rowland : Op. cit, p. 65 for full bibliography.
manufacture, Roman steelyard weights and statuettes, caskets

and plaques of carved bone and ivory from India. Lacquer
bowls from China were discovered in another room. The
Kushan coins of Kaniska and Vasudeva were also found in
conjunction with the treasure. The magnificent collection of
ivory carvings include those of large figures of dancers or
courtesans carved almost in the complete round. These ivory
carvings belong to the Indian tradition. One of the most beauti
ful panels shows ladies at the toilet. The Begram collection
contains a number of examples of painted glassware and also
bronze statuettes. The Begram art appears to be purely secular
and the finds point to active relations between India and the
West with the adoption of Western and Indian styles from the
first to the third century A.D.
Haddass
The ruined monastic city of Hadda, located some eight kilo
metres to the south of Jalalabad, with its myriad towers and
monasteries, was at its apex in the fifth century A.D. when visit
ed by the Chinese pilgrim Fa-Hsien. In the time of Hsuan-tsang
in 632 there were many Sangharamas with few priests and the
stupas ruined and desolate. Here, like the Buddhist structures
around Peshawar and at Taxila, the stupas and viharas were
completely encased with stucco decorations with a school of
stucco sculpture flourishing in this ancient site of Nagarahara
between the middle of the third century A.D. ’and the fifth
century. The time of the Hephthalite invasion while preserving
the dryly spiritualized Gandhara art for the Buddha images, the
sculptures here portrayed a living and natural array of facial
types of such figures. The work at Hadda represents a genuine
religious art. The talents of the craftsmen of Hadda are traced
not only in these figures but in the numerous fragments of
figures and heads of attendant figures in large relief composi
tions.8889 While the wavy hair and the diademed crown are
suggestive of a Greco-Roman prototype, the sensuous fullness of
88. See Barthouk, J.: Les Fouilles de Hadda, Figures et figurines—Me

moirs. VI. 1930; Rowland : Op. cit, p. 65 for fuller bibliography.
89. Rowland. Op.cit, fig. 109, p. 32 for details.
the face with its sharply modelled brows and eyes point to the
Indian ideal of the Gupta period. The extension of the Hadda
medium of stucco sculpture throughout ancient Gandhara as
also in Chorasima or Khwarazem in Russian Turkestan is
equally interesting. It is not certain that the statuary of Toprak-
Kala was a local development of an earlier Bactrian tradition
or it was a northerly provincial extension of the figural sculp
tural art of Hadda and the Kabul valley. The technique of
stucco or lime plaster, as used at Hadda and throughout the
ancient region of Gandhara, is supposed to be invented in
Alexandria in the late Hellenistic period as an inexpensive subs
titute for marble.90
Gandhara Region91 :
The ancient Indian mahajanapada or state of Gandhara,
bounded on the west by Lamghan and Jelalabad, on the north
by the hills of Swat and Buner, and on the east by the river
Indus and on the south by the Kalabagh hill, is closely associa
ted with Buddhist art. It is also described as Graeco-Buddhist
art and is closely connected with the Kushan rulers, especially
Kaniska. Some describe the Gandhara school as the eastern
most appearance of the art of the Roman empire especially in
its late and provincial manifestations. The subject matter of
the Gandhara carving is almost entirely Buddhist* but purely
classical motifs are used for decorative purposes. Kaniska is
considered to be the founder of this school, or its great patron.
It is assumed that from the days of Kaniska until the end of
Buddhism in this area, the practice of importing foreign
artisans continued, although the contribution of native crafts
men was fairly great. In this school, the first representation of
the Buddha in human form is traced. His image is supposed to
be ‘a compound of iconographical and technical formula adopt
ed by foreign sculptors from the repertory of the classical
90. Rowland. Op. cit, p. 28.
91. A comprehensive bibliography on Gandhara and its art is provided
by H. Deydier in his work entitled : Contribution a l’etude del'art du Gandhara,
Paris, 1950. Other later contributions include J. Marshall : The Art o f Gan
dhara (Cambridge, 1951); B. Rowland : Rome and Gandhara—East and West,
IX, 1958, pp. 199 ff.
world’. So also are the earliest representatives of Bodhisattvas

portrayed in all the finery of contemporary nobility of India with
western influence traceable in the stiff swallow-tail folds of the
dhoti or skirt. As such, the Buddhist iconography here is trans
lated into readymade foreign patterns. While the analogy with
Roman art and technique are too apparent to dismiss the in
fluence of the Roman West in the development of classical art
under the Kushans, it is suggested by some scholars92 that
Gandhara art is the final development of a still undiscovered
Greco-Bactrian civilization, rather than the result of inter
course with the Roman world. The Kushan’s fine temple at
Surkh-Kotal with its long inscription in Bactrian Greek is
considered by Schlumberger93 to be a link in this evolution of
the art of Gandhara from the lost culture of Hellenistic Bactria.
The Gandhara art enjoyed a greater period of existence with
its monotony of expression, and repetition of type and techni
ques. Probably the invasion of the white Huns in the fifth
century put an end to creative artistic activity in Gandhara. The
final chapters of Gandhara art are said to have their setting not
in Gandhara but in Kashmir and such remote centres as
Fondukistan in Afghanistan where artistic activity lasted till the
seventh century A.D. The examples of Gandhara stone sculp
tures in Afghanistan are extremely limited, being fragments from
Paitava and Shotorak, near Begram, now in the Kabul Museum.
A seated image from the former place shows the Buddha in
dhyanamudrd with figures of Indra and Brahma in his halo.94
This image like others from Shotorak belongs to a late phase
of the Gandhara style more in keeping with Indian ideals and
form. The original classical drapery is reduced to a convention
92. This view of Roman Ghirshman and Daniel Schlumberger is quoted
by Rowland. Ghirshman points to the extraordinary finds of Hellenistic,
perhaps Bactrian art at the Parthian capital of Nisa near Akshabad in Soviet
Turkmenistan as an ultimate source for the influence of Greek art in Asia.
Schlumberger proposes that the Kushan fire temple at Surkh-Kotal with its
long inscription in Bactrian Greek is a link in this evolution of the art of Gan
dhara from the lost culture of Hellenistic Bactria. (Art in Afghanistan, p. 24);
Schlumberger, ‘Lc Temple de Surkh-Kotal en Bactrianc’, Journal Asiati-
que, J.A. CCXC. 1952 pp. 433-53).
93. Journal Asiatique. CCXC. 1952 pp. 433-53.
94. Rowland : Op. cit. p. 26.
of string-like ridges attached to the surface of the body. This

formalism of the late Gandhara Buddhas provided the models
for repetitions of the type in Central Asia and China as late as
the eighth century.
Tibetan Art95
A switch over from the south-western to the south-eastern
sector of Central Asia concentrating on the countries lying at
the periphery no doubt means change in the art technique as
also in the form, but the matter substantially remains the same.
It is in the context of Buddhism in Tibet that the art of this
region need be considered. Its source of inspiration was no
doubt India. The contribution of Indian Buddhist scholars in
the propagation of the ideals of Buddhism in the local ethos
was fully advanced by the local artists who assimilated the art
traditions of the Palas, as also those of Kashmir. The Khotanese
artists who trekked into Tibet are said to have brought with
them their own styles and iconographic traditions, imposing
these on the Tibetans. Further, with the inclusion of numerous
deities in the decoration of their temples, with their personal
colours,96 banners—painted and embroidered—were used for
embellishment and decorative purposes. Painted Replicas of
these banners were portrayed on the walls of the temples and
95. A comprehensive bibliography on Tibetan Art is provided in

Encyclopedia o f World Art, Vol. XIV, pp. 61 ff. as also in Coomaraswamy:
History o f Indian and Indonesian Art, New York, 1965, pp. 148n. Some
important ones include Hackin, J. : Illustrations tibetanies d'une ligende du
Divyavadiina, Musie Guimet, Paris, 1914; ibid : Guide— Catalogue Du M usii
Guimet, Paris, 1923; ‘Notes on Tibetan Paintings’—Rupan, 7,1921; Getty.A. :
The Gods o f Northern Buddhism, Oxford, 1914; Francke, A.H. : Antiquities
o f Indian Tibet, A.S.I. Calcutta, 1914; Roerich, G.N. : Tibetan Paintings,
Paris, 1928; Gordon, A.K. : Tibetan Religious Art, New York, 1952; Tucci,
G. : Indo-Tibetica, 4 Vols., Rome, 1932. A special number of the MARG
Vol. XVI, Scptember-November 4, contains rich material on Tibetan Painting,
bronzes, calligraphy and ritual objects—with numerous illustrations. As
regards Tibetan Paintings at Tun-huang, see Aurel Stein : Serindicr, Vol. II,
pp. 836 fT.
96. Each Buddha in the Tibetan hierarchy was invested with a personal
symbol and a colour appertaining to his rank, as for example, Vairochana
was awarded dragon white, Ratna Sambhava lion yellow, Aksobhya elephant
blue and so on. (Talbot Rice. Op. cit, p. 210).
shrines. These depicted religious ceremonies and divinities.

Some of the banners were painted on canvas, some on paper
and occasionally some on silk. The native style is prominently
displayed in these banners than on the wall paintings. Here the
subjects chosen are mystical in content and include ritual
dances as well. These banners with their pictorial art demand
special attention.
It is considered likely that painting in Tibet developed mostly
under Indian influence. The wall-paintings in Tasparang,
Western Tibet, reveal the influence of Ajanta. Chinese and
Persian influences, however, mingled with this basic Indian ore
in Tangut area in the north-west. Later, the influences were
mainly Indian and Chinese with the colour symbolism and fine
brush work of the former and the line drawing of the latter
along with the fanciful projections of feeling for landscape,
cloud effects and dramatised imagery. The paintings mostly
represent the divinities of the pantheon or saints, as the principal
figure with scenes from his life as well as minor divinities figur
ing in on a smaller scale. There are paintings which show the
mantfala or the domain of the saint, and also important monas
teries. Some of the prominent pictures show the wheel of life
in this transitory world (samsara). In these a big dragon pro
jects a disc formed of three concentric circles. The smallest of
these, in the centre, encloses three animals, the snake, the pig
and the chicken, symbolising anger, ignorance, and voluptuous
ness. The larger surrounding circle is divided into six segments
showing the life of the six categories of living beings.97 The
third circle contains twelve images representing more or less
vaguely the connection between twelve causes and effects.
An important class of painting98 known as Tsog-shin re
presents the total Pantheon of the Lamaite divinities. The
97. These include two kinds of beings living upon earth : men and beasts.
Those living beneath the earth are condemned to an eternal hunger and thirst
drink by reason of their narrow throats, which shook forth fire when they
desire to-drink. The inhabitants of hill are divided into twenty classes according
to their torments. And then there are those living in heaven, the asmas,
who always struggle against the gods. And above them are the various divi
nities themselves. All these beings are seen in movement in symbolic colours.
(Marg. Op. cit, p. 19).
98. ibid, p. 20.
Buddhas (48), the Bodhisattvas (12), the feminine divinities (9),

the protectors or tutelary gods (42), the defenders of the faith
and the eight terrible ones (27), and the minor divinities and
the guardians of the four cardinal points. The whole painting of
this class is concentric. The main divinities are nearest to the
central point of the pyramid occupied by Avalokitesvara.
Above the chief divinities are the three ranks of Bodhisattvas,
and on his right are the three Buddhas. Thus, the central group
includes the main divinities surrounded on the right and the
left by the goddesses. Above the principal group is the double
rank of the protectors with feminine divinities of the second
order on the right and the left. Under Avalokitesvara are
the Buddhas overflowing on the sides of the pyramids while
different Bodhisattvas appear below. Further down are others
connected with Tibetan Lamaism with the bottom rung occupied
by the defenders.
The subject matter of all Tibetan art is purely religious, pro
viding as elsewhere in Central Asia and here in India too, the*
outward form and expression of the inner ideals of Buddhism.
Incidents from the life of the Buddha, portraits of the great
Indian scholars and teachers of Buddhism in Tibet who founded
philosophic school and propagated Buddhism there, as also
portraits of their disciples, illustrations of the Jataka tales, the
Buddhist hierarchy of gods and goddesses and Buddhas and
Bodhisattvas, including the future ones—often in series—are all
included in the theme of Tibetan paintings. Frescoes and
illuminated manuscripts also engaged the attention of Tibetan
artists in the first centuries after the introduction of Buddhism
into Tibet. Despite Indian or Chinese influence on Tibetan
paintings, there is a blending of one with the other in the
Tibetan ethos. As pointed out earlier, the Chinese influence is
apparent in the treatment of landscape—with the clouds, rivers,
mountdins, flowers and vegetation, all finely depicted.
The Indian influence is traced to the time of the Palas of
Bengal. The monastery of Nalanda not only gave form to the
religion but also inspired the art of Tibet under king Dharma-
pala. Two artists from Northern India—Dhiman and Bitpala
are supposed to have helped in the task of executing paintings
and modelling statues at the same time. The former was
influenced by the Eastern school while the latter was indebted

to the art of Madhya Pradesh. Besides these two Indian master
minds in the realm of art, there were others who came to Tibet
from Central Asia, Swat, Kashmir, Nepal and from China."
Some inscriptions do mention the school of the artist as for
instance the style of Lilugs suggest that the painters came from
Li viz Khotan. Some paintings discovered by G. Tucci in the
small temple at Mangnang in Western Tibet are accurate imita
tion of the style of painting that probably existed in Kashmir
in the 12th century. So also the paintings of lha Khan of Alchi
in Ladakh can also be attributed to the Kashmir school of a
much later period.
The Chinese influence is mostly confined to the monastery
of Shalu—the residence of Bu-stan (1290-1361). Artists here
were called from China and Mongolia. This style assumes pro
minence in the 17th and 18th centuries. These paintings con
form to fixed norms, precisely defined in literature. They are
rich compositions in which Buddhas, Bodhisattvas and other
divinities are always depicted with inferior retinues of acolytes.
Deities also dominate the composition amid a choir of elect in
their own paradise of which the sceneries are also portrayed
according to set norms. The colours vary according to the
nature of the principal subject: white is used for god in his
calm aspect; red or dark blue is used for the terrifying one or
in combat. Aspects of Tibetan life and landscape are depicted
with lively colours and great animation and landscape is just an
echo of the Chinese one. In ancient paintings no part of the
surface is left free. It is completely covered with images, figures,
symbols etc. The illustrated manuscripts have images of divini
ties and the Buddhist masters painted on the first and last page,
and sometimes miniature paintings decorate the inner pages as
well.
Besides paintings, reference might as well be made to Tibetan
architecture and sculpture99100—especially the bronze. - Not much
can be said on the subject of Tibetan architecture from the
Indian point of view except that the Tibetan stupa or Chorten—
99. Encyclopedia o f World Art, Vol. XIV, Op. cit.

100. See—Encyclopedia—Op. cit and Marg—Op. cit, pp. 14 ff.
a bulbous dome set on one or more square bases was dedicated

to great events from the Buddha’s life, such as his Nativity and
Nirvana. It is like the Nepalese type surmounted by a square
harmika and a mast upholding a teer of ‘telescoped umbrella’
with a flame finial at the top. A large monument at Gyantse
has an unusual plan and elevation. It is erected in fine stepped
terraces on a polygonal plan with multiple recensions or step-
backs. The actual dome built on this prasada or pyramid was
circumambulated at each successive level. In plan it resembles the
great Mahayana monument of Borabudur in Java. The sky
scraper structures like the Potala at Lhasa, built of stone and
sun-dried bricks with the white-washed walls thicker at the
bottom, provide the contours of the surrounding mountain
peaks. This palace-monastery at Lhasa is a noble pile of succes
sive stages.
The sculptural art is confined to bronzes. There is a legend
that the first images were brought into Tibet by the Chinese and
the Nepalese wives of king Srong-san-gaupo in the seventh
century. And they went on being imported from India and
Nepal for a long time afterwards. The talent of the Tibetan
craftsmen introduced a great dynamism into the imagery. These
craftsmen took the dimensions, the number of heads, arms and
legs and other symbols from the canon but displayed their free
dom and technical virtuosity in their creation. Metal sculpture
of the various divinities, big and small, multiplied and the
Tibetan bronze, brass and copper image makers flourished in
their art. The images are made by cire-perdu or lost wax
method. The figure is modelled in wax and covered with sand,
fixed by silicate and plaster or some other substance. Then it is
baked. The wax melts leaving the mould free. Metal is then
poured into the cavities. After the metal cools down the mould
is destroyed. The figure is then finished by hand with the chisel
and tools. A cavity in the image is always left to insert the
magic formula printed on rolls of paper. After the insertion the
hole is sealed and the image becomes sacred.
The Buddhas—past, present and future *are shown in contem
plation, with their hands shown in different mudras. The
Bodhisattvas wear rich garments and jewellery, unlike the un
adorned and uncovered Buddhas except the robe covering the
left shoulder. There are tutelary divinities called yei-dam taken

mostly from the Tantra. The Dhyani Buddhas are shown with
their consorts in the image called yab-yum101—the honourable
father and mother. The Dharmapalas or the guardians of the
law are shown in a furious mood with numerous heads, arms
and leg. The Kalachakra or the lord of time has four faces and
twenty-four arms with his consort having only eight arms. The
former tramples on demons. Jambala or the god of wealth Is a
fat Kubera-like figure holding a bowl of jewels in one hand, a
mongoose vomitting jewels in the other. Vajrapani with many
forms and Avalokitesvara symbolising wisdom are also included
among the bronze icons. Reference might as well be made here
to a beautiful bronze image of the Buddha in the Bhumtiparfa-
mudrd.101102 The elegance and grace of this image with its delicate
and sensitive handling of the whole body in repose is an example
of the highest workmanship in Tibetan bronze making. The
Pala workmanship may have contributed to its elegance.
Another bronze103—representing a yab-yum image—is equally
interesting with the melodramatic expression equally graceful.
The male figure inclining backwards with his tender eyes and
open mouth is inter-locked with his Sakti in a close embrace.
The lines of the arms and the torso are drawn with a delicate
pliant touch. The collections at the Music Guimet are fairly
rich in Tibetan bronzes and paintings.104
101. Zimmer : The Art o f India Asia, p. 195. This posture is a common
one in Tibetan Buddhist images, and was derived from the earlier archetype
of Siva and the Goddess. The example of Vajrasabha, the president of the
group of DhySni-Buddhas—the five others being Vairochana, Ak$obhya,
Ratna-Sambhava, Amitabha and Amoghasiddhi, is shown with his Sakti
(Zimmer p!. 610). The male and female principles in eternal embrace, represent,
as in Sivaitup iconography, the coincidence or union of opposites. The divine
couple are both the goal and the way; fulfilment and the means or process of
attaining it; enlightenment or way to enlightenment, (ibid, p. 196).
102. Marg. Op. cit, p. 44.
103. ibid. no. 2, p. 45.
104. For a fuller and detailed reference, see J. Hackin : Op. cit,
part IV, Le Tibet, pp. 63ff. Chapter II, pp. 70 ff entitled’ Les peinturcs et les
bronzes’ notices the biographical scenes with general characteristics, scenes
from the life of the Buddha, the Dhyani-Buddhas and Bodhisattvas, female
divinities, Mahasiddhas and saints, grand Lamas and ritualistic objects. It
might be out of context to make a detailed study of Tibetan paintings and
Central Asia with its eclectic outlook took over and recasted
with considerable success the art forms of the great sedentary
civilizations, those of the semi-classical of the East, of Iran, of
Indo-Greek and Indo-Scythian art, of China and of Gupta
India. The borrowing results are traceable in the cultural evolu
tion of different zones, no doubt varying in their extent and
intensity and extending from minute details to essential concepts
of art style and iconography. The Iranian elements, however,
predominate in the western zone, while south of the desert the
Indian influence or traits have an upper hand along with the
intruding Persian influence. The eastern zone has certainly close
Chinese impact. Despite these diversities, the Central Asian
peoples were close to each other in their spiritual outlook and
artistic expression, with extraordinary analogies of style and
iconography. The inexhaustible subject matter was provided by
the Buddhist legends—the Jatakas—for the artists to display
their talents. When elements from the Indian complex mytho
logy, with the rich possibilities offered by Hindu and Tantric
importations, were added, the scope of the artist was immensely
extended. The vast material collected by the explorers and
excavators from different nationalities no doubt reveals a be
wildering diversity of style and treatment. This was due to the
complex social and political conditions prevailing in the zones
of artistic activities. The change of political powers during the
long course of Central Asian history, passing on from one race
to another differing in ideals and traditions in art and varying
in psychological conception, was responsible for this lack of
uniformity except on a religious plane.
In the light of the above observations the Buddhist artistic
contributions in different areas have been recorded in sequence
of time and space. The earliest ones were, of course, Miran and
the sites lying on the old silk route. The paintings here suggest
familiarity with the Gandhara art. Besides classical elements
like the motive of the festooned garland carried on the shoulders
of amorini, as also the wringed angel busts, there is continued
bronzes from this Museum from the iconographic or artistic point of view.
A general observation of Tibetan Art in the context of Central Asian Art alone
is provided here.
narration as provided in the Vessantara Jataka scene. Indian

conception and execution could be noticed in the type of persons
putting on Indian dress, but the Persian type of beauty suggests
Iranian contacts. The reference to Titus and the fees paid to
him—as recorded in a Kharosti inscription—confirms the role
of the Indo-Greek artists in this area. The Gandhara school is
supposed to have been transplanted into pictorial lines at Miran
where Roman-Hellenistic influences are quite apparent. In the
Khotan complex the imprint of Greco-Buddhist art predomi
nates between the fifth and eighth centuries. Indian, Sassanian,
Chinese, Sogdian and even perhaps Chorasmian influences are
no doubt assimilated here. Out of the blending of foreign in
fluences emerged a new style characterised by strictly frontal
presentation, highly developed stylization, flat almost two-
dimensional designs and a tendency towards geometric simpli
fication. The elongated matching eyes are drawn in Indian
fashion and the beautiful figure of a worshipper kneeling in
prayer from Balawaste in structure, costume, colouring and
style reminds one of the Ajanta figures. The direct connection
between the local art and the Sogdian influence is evinced from
the painting of the Iranian Bodhisattva of Dandan-Oiliq. The
figure of the Silk Princess as also other ones suggest Chinese
influence co-existing with Indian and Iranian art tendencies and
style. The Khotan School had its impact on Tibet and references
to Khotanese painters are traced in Tibetan records.
The Buddhist art centres on the Northern Routes—Kizil,
Kucha and Turfan—were equally connected with or influenced
by the artistic developments of the neighbouring ones. The
Kucha centre beginning about the fourth century A.D. continued
till the end of the eighth century with its two distinct phases: the
first of the Indo-Iranian type (c. 500 A.D.) and the second
strongly Iranian, reaching its climax about 600-650 A.D. The
Chinese influence is noticeable here in the paintings between
the seventh and eighth centuries. The vigorous cowherd leaning
on a knotted stick is a good illustration of the first type. Indian
influence is more prominent in the group of swimmers scene at
Kizil. The paintings at Kumtura form a third phase marked by
Chinese influence with more attention to line and the colours.
Iranian types and motifs are virtually eliminated or absorbed
in the Chinese element. The Mongolian features, head d«-ess,

drapery and folds of dress and scarves are traditionally Chinese.
The local painters at Shorchuk, a meeting place of figurative
currents from Kucha and China, concentrate on the lives of the
monks and the other two of the Three Jewels—the Dharma and
the Sangha.
The Turfan group of Central Asian paintings, covering a
period of three hundred years, exhibit mannerism in physiog
nomy, pose and costume of the figures. The different phases of
the complex art in the Turfan region vary from one locality to
another with varying degrees of Chinese influence and dominance
of Tantric features, associated with Mahayanism. At Khocho,
the Turkish name for the ancient capital Kao-Chang, Manich-
aean and Nestorian influences dominate, and so ‘also different
phases of interaction of Chinese and Iranian cultures are traced.
Marked differences could be seen in the figures of Buddha at
Khocho and at Bezaklik. The mourning scene at the latter
place in shrine IX shows people of different nationalities distin
guished by their faces and dress demonstrating the universal
aspect of the Buddha religion.
Reference has also been made to paintings from Tun-huang—
the meeting place of the two routes as also noted for the caves
of the Thousand Buddhas. They provide an idea- of Chinese
Buddhist art in concept and design. Purely Indian art is re
presented by a small group of paintings. A few works are
Tibetan. These frescoes at Tun-huang belong to a period bet
ween the fifth and the eighth centuries A.D. Those on silk used
as banners contain some masterpieces. While the treatment on
the whole remains predominantly Chinese, traces of impact of
the Graeco-Buddhist school and kinship with Indian art could
be traced.
The western part of Central Asia—now comprising part of
Soviet Central Asian Republics, had its own cultural ethos as
also its art forms. A number of sites in this region were ancient
Buddhist centres. The recent discovery of wall paintings at
Balalik-Tepe near AirtamTermez are supposed to be of Sassanian
inspiration with the Greek influences often prevailing, and
occasionally containing the rising tide of Sassanian Persia. The
earliest Indo-Hellenistic blend is noticed at its best in the
fragments of a superb sculptured limestone frieze of the first

century A.D. The ruins of Khalchayan, Toprak-kala and Pen-
dzhikentthe Hephthalite capital, have provided stucco figures,
and wall paintings adorning the larger buildings. Their subject
matter is rather puzzling.
Both Afghanistan and Tibet at the periphery of Central Asia
are notable for Buddhist art. The former was connected with
Gandhara art and provides examples of rock-cut statues, scul
ptures and paintings as well as stucco figures. The pictorial art
here is remarkable for richness and variety of colours and a
definite touch of refinement. The fusion of Iranian and Indian
elements produced a hybrid style, the direct antecedent of the
Buddhist paintings in Central Asia, especially at Kizil and Murtuq.
The paintings in Tibet are supposed to have developed mostly
under Indian influence, with the Chinese and Persian influence
mingling with it at the western and eastern parts of Tibet res
pectively. The subject of all Tibetan art* is purely religious pro
viding as elsewhere, the outward form and expression of the
inner ideals of Buddhism. Here artists from Northern India as
well as from Khotan participated in the realm of pictorial art.
The art of Central Asia—eclectic in character—assimilated
the forms of great sedentary peoples around them. At an early
stage it came to be known, according to the French archaeologist.
Schlumberger, as the non-Mediterranean descendant of Graeco-
Roman art. It subsequently gave birth to a series of indepen
dent movements which influenced and were influenced by the
major civilization of Asia. Despite the multiplicity of ways of
rendering space and perspective as also providing colour scheme,
the theme—the all-pervasive Buddhism, was the religious force
unifying a large part of Asia in the spirit of humanism and
universalism. Art only catered to the projection of this concept.
CHAPTER VII
THE SUMMING-UP
Central Asia is supposed to have been the cradle of human

civilization. Its geographical position, however, accords it a
receiving as well as a transmitting centre. In metaphorical
language it has not been so much a basin as a pool in a tidal
river, flowing alternately to and from the sea. In such a pool
could be found creatures of different provenance. Currents—
cultural and political—passed through it from east to west and
in the reverse direction leaving some remnants over there. Being
in touch with Bactria and the regions conquered by Alexander
and through them with the western world in its art and thought,
the Tarim sampled the stream fragments that had drifted from
Asia Minor and Byzantium through the Iranians. Chinese
civilization and administration were imported from the east
while the south provided other currents connected with Buddhism.
From the dawn of history down to the middle ages warlike no
mads continually passed through this region picking up and
transporting the ideas and institutions of others. This factor
necessarily involved simultaneous use of a number of tongues
for popular as well as for learned purposes. This is evident from
the recovery of great polyglot libraries at Tun-huang, and manu
scripts in several languages at other places including Indian
Buddhist, Manichaean, Syriac, Sogdian, Uighur and Chinese
ones. Written on palm leaves, birchbark, plates of wood or
bamboo, leather and paper from the first century A.D. onwards,
these manuscripts convey the story and doctrine of the religious
cults with which they were associated.
A fuller account of the religions of which these codifications
are a part, no doubt, demands a detailed study of the available
source material. This material is ample, confined to art and
literature as also to epigraphic records which shed light
on the material culture as well. Explorations and excavations

have revealed numerous sites connected with Buddhism. The
religious monuments of Central Asia comprise stupas, caves and
covered buildings which were used as temples or viharas. The
caves were decorated for Buddhist worship with frescoes as well
as figures executed in stucco. Many stQpashave also been found.
The ideas and designs of Indian Buddhism dominate in these
monuments except in regions nearer to China. Information
relating to Central Asian relations with China is provided rather
in detail by the Chinese annals. These relations were often
interrupted and the data available from the Chinese accounts
have occasional missing links, particularly in relation to the
history of Buddhism in Central Asia and its transmission to
China. This study of Buddhism in Central Asia, based on these
varied sources of information is, therefore, projected within the
framework of geographical information, particularly, the land
routes passing through the important political and trade centres,
kingdoms and their rulers who patronised Buddhism, the role of
the Buddhist missionaries and savants and their contributions—
religious and secular.
Detailed information about the routes passing through Central
Asia is provided by the Chinese accounts including those of the
pilgrims who passed through these on way to* India. These are
also known as the silk-routes during the first few centuries of
the Christian era. Two routes passed through the Tarim basin
from the frontiers of China upto Balkh. These were used by
the Buddhist monks and savants as well for the dissemination
of Buddhist thought and culture in the States of Eastern Tur
kestan and finally in China. The ancient route, from the capital
of China, Chang-an (present Sian) in the province of Shansi,
after crossing the Gobi desert reached the oasis of Tun-huang
from where, approaching the Taklamakan desert, it bifurcated >
in two directions. The northern one passed through Hami,
Turfan, Karasahr, Kucha, Aksu, Tumshuk and Kashgar to
Samarkand. The southern route traversed via Miran, Cherchen,
Keriya, Khotan and Yarkand to Herat and Kabul. It w as'in
the first century A.D. that Buddhism was taken to the states
lying on the routes. Peoples from Kashmir and North-west India
proceeded to Khotan and Kashgar and' set up small colonies
The Summing-up 319
with kings claiming descent from Indian regal families. The

routes starting from north-west India in those days passed by
Hadda and Nagarahara (Jalalabad) and reached Bamiyan before
crossing the Hindukush. Bamiyan was thus a halting place for
the Indian monks and grew into animportantcentre ofBuddhism.
Further to the north, Bactriana (modern Balka)known asNava-
Sarigharama, was a great centre of Buddhist learning. It was also
the meeting place of two different roads leading to Central Asia
and China.
The importance of Kashgar both from the .point of view of
commercial activity as well as of expansion of Buddhism was
equally great. Its location provided relief and hospitality to
the travellers and the pilgrims alike and numerous monasteries
had come up here. From this place again there were two routes
extending up to the borders of China. The southern one running
along the fringes of the Tariin basin passed through Yarkand,
Khotan, Niya and also a number of other sites like Dandan-
Oiliq, Endere and Miran. These places were notable for
Buddhist monasteries and as centres of learning. According to
the accounts of the Chinese pilgrims, Fa-hienandHsuan-Chuang
they were associated with the Mahayana school of Buddhism.
The centres on the northern route Kashgar, Kucha and Turfan
were, however, HInayanist. The inhabitants of Kashgar were*
sincere Buddhists and there were more than a thousand monks
of the Sarvastivadin school, but their knowledge was not in
proportion to their zeal. This school flourished at Osh and
Kucha and also at Balkh and at Bamiyan. The Great vehicle
was predominant at Yarkand and Khotan as also in Kapisa.
The people and language of the countries or centres on the
northern route were different from those of the south, but both
had common affinities in the form of Buddhist religion and
culture.
Kucha or Kuchi, like its counterpart Khotan on the southern
route, was the most important centre of Buddhism with Indian
names of rulers like Suvarnapuspa, Haradeva, Suvarriadeva
etc. The Buddhist monks of this place were well conversant
with Sanskrit, as is confirmed by the finds of manuscripts in
Sanskrit and bilingual documents in ICuchean Sanskrit. In the
words of the Chinese pilgrim Hsuan-Chuang, it was not only a
centre of Buddhist studies but it also provided famous scholars

and savants who translated Buddhist texts and also gave dis
courses. The noted Buddhist scholar Kumarajiva was one such
scholar who shone as a luminary radiating his lustre throughout
Central Asia and China. He was connected with the Kuchean
regal family. The grottos in the hills in the neighbourhood of
Kucha provided a safe and solitary retreat for the monks.
Agnidesa in the region of Karasahr—the next place on the
northern route was linked with Kucha in culture and language
as also in ethnical affinity and common religion—Buddhism.
The finds of literary texts and art objects in this region testify
to its importance as a Buddhist centre. Turfan, the next stage
on the northern route further east was equally connected with
Buddhism with strong Chinese cultural impacts. Extensive
literary remains include words in Sanskrit, Chinese and various
Iranian and Turkish idioms in two dialects.
The Northern as well as the Southern routes met on the
Chinese frontier at a place called Yu-men-Khan or the ‘Jade
Gate’, close to the famous Tun-huang which was noted for its
thousand grottos. These were carved out between the fifth and
the eighth centuries A.D. for the Buddhist monks proceeding to
China as also for holding discussions between groups of Buddhist
•scholars from various countries. Some Indian families had
settled down at Tun-huang in the third century A.D. promoting
activities both secular and religious. The finds of a large number
of manuscripts suggest its importance as an active centre of
Buddhist learning. Twenty-thousand manuscripts were dis
covered from the walled up chamber at the Tun-huang. These
are in Brahml, Kharosthi, Tibetan, Turki, Uighur and Tokharian,
and point to the cosmopolitan character of the monks and others
frequenting or living in this area of Central Asia. A study of
the geographical factors as also the nature of the peoples of
different regions—nomadic as well as pastoral, receptive to ex
ternal influences but equally independent, has, no doubt, been
necessary in this context. The process of the introduction of
Buddhism in Central Asia and through it into China was gradual
and steady. It was the work of missionaries, political exiles,
savants and scholars rather than of zealots. The Chinese
sources point to royal invitations extended to Indian Bud
The Summing-up 321
dhist scholars to visit their country and propagate the

message of the Tathagata. These were mostly from the Buddhist
religious establishments in Central Asia itself. Tradition connects
Buddhism in Khotan with Kunala, the destitute Mauryan
prince from Taxila. A number of Indian colonies had come up
in Central Asia as recorded by the Chinese pilgrims. While the
earlier picture of Buddhism and expansion of Indian culture in
Central Asia might not be distinct and vivid, that of the later
period is more pronounced and specific as might be evident
from the finds of Buddhist manuscripts, monuments, sculptures
and paintings. Patronage to the Buddhist scholars was extended
by local rulers whose historical account no doubt demands
some study.
The study of political history of Central Asia in relation to
Buddhism and its expansion in this vast area as also in China
involves reference to those rulers who patronised Buddhism
in their kingdoms. While the date of introduction of Buddhism
through Kashmir or Afghanistan is uncertain, it is clear from
the accounts of the Chinese pilgrims that both the schools of
Buddhism were flourishing in different centres. That suggests
not one but at least two or may be even more currents of
Buddhist religion which swept over this vast expanse. The
earliest account of Buddhist activity could be traced to the
first century A.D. with the despatch of Kasyapa Matanga and
Dharmaratna proceeding to China in A.D. 65. The arrival of
the first Buddhist missionaries is no doubt mixed up with the
legend of the Emperor Ming of the Han dynasty seeing a golden
man in a dream—apparently the Buddha—who asked him to
invite the two Buddhist monks. Emissaries were for inviting
these two savants and they did go to China with sacred texts
and relics. The first Buddhist monastery was built for them in
the capital. These monks spent the rest of their lives translating
Buddhist texts into Chinese and doing evangelical work. They
are mentioned as natives of the country of the Yue-che i.e.
Tokharistan, since the word Che (from Yue-Che) is prefixed
to their names. This country provided other monks as well for
academic and missionary purposes. Lokaksema of Tukhara
origin, and a man of exceptional learning went to Loyang in
A.D. 147 and translated there some of the important Buddhist
texts into Chinese. He stayed there till 188 and was followed by
one of his young disciples, Che-Kien, who left towards the end
of the second century and was in China for nearly three decades.
Dharmarak$a—another Buddhist monk, called Fa-hu in China,
belonged to a Tukhara family. He settled down in Tun-huang
towards the middle of the 3rd century A.D. A master of thirty-
six languages he was widely travelled in Central Asia before
leaving for China in A.D. 284 and worked there till A.D. 313
translating nearly 90 Buddhist texts. He was followed by another
Tukhara monk named She-lun who came to China in 373, and
eleven years later by Dharmanandi from the same kingdom.
Both the monks translated Buddhist texts into Chinese. Intelle
ctual scholarship and religious zeal accounting for the migra
tion of Buddhist scholars to China on invitation could be
possible only in an age of awakening and maturity as also in
the area of Buddhist learning. Tokharistan—the land of the
Tukharas—Tu$arhs must be crowded with Buddhist savants. It
was during the Kusana period that Buddhism was taken to
Central Asia where the people from Kashmir and North-Western
India (now Pakistan) had set up small colonies with kings
claiming descent from Indian regal families. Bamiyan in
Afghanistan had risen to be a great centre of Buddhism in the
early centuries of the Christian era, and it continued to retain
that position for a number of centuries. Bactriana (modern
Balkh) —Chinese Fo-ho and Indian Bahlika, further north, was
another Buddhist centre. Here the religion of the Tathagata
was introduced in the first century B.C. or even a little earlier.
Its political history slides from the Greeks to the Sakas, follo
wed in turn by the political ascendancy of the Yue-ches and the
Huns.
The establishment of Buddhism in Tokharestan under the
Kusanas stimulated its process of expansion. The Parthians to
the south-west took active interest in this region, and equally
participated in the emigration programme of their Buddhist
savants to China. According to the Chinese Annals a number of
such scholars distinguished by the prefix An (Ngan) went to
propagate and translate the gospel of the Buddha. A Parthian
prince, known to the Chinese as Ngan-She-Kao or Lokottama,
visited the western frontier country of China with a load of
The Summing-up 323
Buddhist texts. He was a prince of the ruling family who waived

his claim for the throne in favour of his uncle and took up the
robes of the Buddhist order at a young age. He settled down at
the White Monastery in A.D. 144—which was earlier built for
the two pioneer Indian monks Kasyapa Matahga and Dharma-
trata. He founded a school for translating Buddhist texts and
personally contributed more than a hundred items in this
programme. Other Parthian scholars who were associated with
the school of translators of Buddhist texts were Ngan-Hivan,
originally a merchant, and Ngan-She-Kao and the Indo-Scy-
thian monk Lokaksema. Ngan-She-Kao madethe first organised
effort to translate the Buddhist canon into Chinese. This acade
mic exercise of translating Buddhist texts was later undertaken
by some Parthian monks who, though not very celebrated, went
to China during the third and fourth centuries A.D. Besides
doing the translation work, they also contributed towards the
propagation of Buddhism. Some Sogdian monks also collabo
rated with Ngan-She-Kao, including the illustrious Seng-hui
(Sahghamati ?). The names of the Sogdian monks are disting
uished by the prefix Kang, a short form of the Chinese Kang-
Kiu, the name given to Sogdiana. He was the first to introduce
Buddhism in Southern China in the third century A.D.
The contribution of these Buddhist scholars from Tokha-
restan, Parthia and Sogdiana is suggestive of the prevailing
prosperous condition of Buddhism in their homelands, enabling
them to move out for evangelical work in distant China after
transgressing natural barriers in the way. The credit for Bud
dhist activity and expansion should go to Kaniska, the patron of
the fourth Buddhist Council and" himself an ardent Buddhist.
Buddhism seems to have made considerable progress not only in
India but also in Central Asia during this period, as seems evi
dent from Buddhist monuments at Bamiyan of Surkh-Kotal in
Afghanistan and Adjina-Tepe in Tadjistan (now in Soviet
Central Asia). The knowledge of Indian Kharosthi script is
revealed from the documents found at Niya, not very far from
Khotan, and such Buddhist scriptures as the Gandhari
Dharmapada. Further, it is proposed that the interference of
the Yueh-chih in the political affairs of Kashgar accounted for
the introduction of Buddhism in that kingdom. This assump
tion seems to be in agreement with the tradition recorded by

Hsuan-tsang about the princely hostages from Sha-la or
Kashgar residing in a Buddhist convent as a resultof Kaniska’s
conquest of that region. It is further presumed that the Bud
dhist Church and its establishment, whatever be its period, was
the result of impact from Baktra and not Khotan, as Mahaya-
nism was the prevailing form of Buddhism in Khotan, while
Hinayanism was popular in Kashgar.
The information relating to Buddhism in Kashgar is no
doubt given by the Chinese pilgrims who visited this place on
their way to or from China. To the early Chinese geographers
the country was known as Shu-le, and later on it was called
Kie-Sha. Its general description, including that of the people,
is substantially the same as given by Hsuan-Chwang. Chih-
mong (A.D. 404) saw at Chih-sha (Kashgar) Buddha’s alms
bowl as also his spittoon made of a variegated colour stone.
The famous savant of Kucha, Kumarajiva also visited this
place about A.D. 400 and he records the miraculous quality of
this bowl changing its weight according to the merit of the
person carrying it. Fa-hien also describes his stay at Chich-
Cha (Kashgar) enjoying royal hospitality and participating in
the great quinquennial assembly of monks. It was a Hinayanist
centre pf Buddhism with a thousand monks residing here. The
place also boasted of a tooth of the Buddha for which the
people had set up a stupa. In Hsuan-Chuang’s time, Buddhism
was in a flourishing condition. He records several hundreds of
Sangharamas with some ten thousand followers, studying the
Little Vehicle and belonging to the Sarvastivadin school.
Several stupas were explored close to Kashgar. Two other
pilgrims, Dharmacandra of India and Wu-kong of China
passed through Kashgar, the former on his way back from
China and the latter on his way to Gandhara.
Indian scholars frequently visited Kashgar. Kumarajiva, son
of Kumarayana and Ilva of the Kucha royal family, on his
return journey from Kashmir, where he had gone for his
education, stayed at Kashgar for nearly a year and studied
Abhidharma of the Sarvastivada school with its six divisions
under able teachers. It is reported that Hi-Kien, a master of
Tripitaka requested the ruler to retain Kumarajiva permanent
The Summing-up 325
ly in Kashgar. Before leaving for home, Kumarajiva also stu

died the four Vedas, the five sciences, the Brahmanical Sastras
and also astronomy at Kashgar* During the period of his stay
(last decade of the 4th century), two other important persona
lities—Suryabhadra and Suryasoma, the two sons of the King
of So-Kiu (Karghalik-Yarkand) were also ordained here and
Kumarajiva taught them Satasastra and Madhyamika-âstra
before leaving for Kucha. He in his turn was assisted in his
studies by Buddhayasas of Kashmir who had earlier come here
and stayed here as an adviser to the ruler. He later on joined
Kumarajiva in China where he was taken by the' Chinese
General Lu-Kuang. Another Buddhist scholar at Kashgar was
Dharmacandra who was originally from Magadha and had
gone to China in 730 from Kucha at the invitation of the
Chinese ambassador. On his way back in 741 he stayed for
some time at Kashgar. Political unrest in the way finally landed
him in Khotan where he died two years later.
On the southern route from Kashgar to Khotan across the
Sita (Yarkand Darya) river was Che-Kiu-Kia or Tsen-ho
according to the Chinese sources and Cu-gu-pan of the Tibetan
texts, corresponding to Cokkuka of the Central Asian docu
ments. It has been identified with Karghahalik-Yarkand.
According to Hsuan-Tsang, the people were sincere Buddhists
and the country had some tens of monasteries with more than
a hundred monks who were all followers of Mahayanism. The
two princes of this place Suryabhadra and Suryasoma had gone
to Kashgar and received initiation from Kumarajiva and
studied Mahayana texts with him. The ruling family seems to
be of Indian origin.
Further south was the kingdom of Khotan which in the time
of Hsuan-tsang had a hundred monasteries with five thousand
monks who were all followers of Mahayanism. Buddhism was
supposed to be introduced here in the time of Vijayasambhava,
the grandson of Kutsana, its founder. The monk Vairocana,
supposed to be an incarnation of Maitreya, had come here from
Kashmir bringing with him the relics of the Buddha. The ear
liest monastery was built for this monk by the ruler, followed
by the settingup of several other such establishments by the
Khotanese rulers including the one built by the Chinese prin
cess Punesvara (Pu-nye-shar), queen of king Vijayadeva in

honour of Kalyanamitra. The ruler’s eldest son Dharmananda
who had joined the Buddhist order also established several
monasteries. The school of the Greater Vehicle (Mahayana) was
prospering in Fa-hien’s time as well with three thousand monks
in the famous Go§nnga monastery there. Its greatness as a
centre of Mahayana Buddhistic studies in the third century
A.D. is evident from the visit of the Chinese monk Chu-She-
hing who came here in A.D. 290 for the study of Buddhism
under eminent teachers and sent through his disciple 9000 bund
les of original Buddhist texts to China for translation. The
noted Buddhist scholar Buddhasena was a famous savant,
known as ‘the lion of learning’. There are references to visits of
Buddhist scholars from China and carrying with them texts
from Khotan. The quinquennial assembly no doubt was a
great attraction for such inquisitive visitors on a religious
mission.
Besides Khotan, there were several other Buddhist centres on
the southern route. Fragments with inscriptions in Brahmi,
manuscripts and other records, as also KharosthI documents of
the 3rd century A.D. were found in excavation. Both the
schools of Buddhism prospered in this part of Central Asia
around the fourth century A.D. Lou-lan, the Na-fo-po also
called Shan-Shan in the Han Annals, was another important
centre—a stronghold of Buddhism as also of Indian culture.
According to Fa-hien there were 4000 monks here, all followers
of Hinayanism. The original name of this place was Kroraina
or Kroranjinaof the Kharo§thi documents. These records from
the Central Asian sites on the southern route at Lou-lan, Niya
and Endere shed considerable light on the material culture of
the people in this area. The names of donors in these inscrip
tions appear to be of Indian origin. Frescoe paintings from
Miran with the Buddhist subject matter and linked with Gan-
dhara art bring out the importance of this place in the history
of Buddhism and its expansion in Central Asia.
The famous Buddhist centres on the Northern route—Aksu,
Kucha, Agnidesa or Karasahr and Kao-Chang-Turfan with a
common racial heritage and language, of course with minor
dialectical differences of a local nature, had a uniform cultural
The Summing-up 327
background. They were linked with each other in the sphere of

Hinayana Buddhist school. Among these States, Kucha was the
most important politically as well as for its eminent Buddhist
savants. The date of the introduction of Buddhism here is un
certain. It was no doubt in a flourishing state in the third
century A.D. According to the Annals of the Tsin dynasty
(265-316), at that time there were nearly one thousand Bud
dhist stupas and temples in Kucha and it also sent Buddhist
monks to China for translating sacred texts into Chinese. In
the fourth century A.D., the capital was virtually changed into
a Buddhist city with numerous monasteries controlled by the
Abbot Buddhasvamin—a great scholar, a follower of the
Agamas. Among his disciples was the famous Kumarajiva.
Nunneries for the Bhiksuriis included those from royal and
noble families, observing a strict disciplined life. The role of
Kumarajiva and his disciples in the dissemination and expansion
of Buddhism in Central Asia and in China is a well-known fact,
and his life history is a saga of service to the religion of the
Lord. Kumarajiva was well-versed in all branches of Buddhist
learning as also in languages—oriental and Chinese. He could
remove doubts and dispel erroneous interpretations of scholars
from all quarters. He was joined at Chang-ngan by his Sarva-
stivadin guru Vimalaksa who was staying all along in Kucha
after Kumarajiva’s departure. Another scholar Dharmaraitra
from Kucha also joined him in his new sojourn, followed by the
Kashmirian scholar BuddhayaSa under whom Kumarajiva had
studied in Kashgar, and Buddhabhadra from the same place.
The Kuchean savant was responsible for introducing Mahaya-
nism in the countries of the Tarim basin and also in China in
a more responsible and authoritative manner. He was one of
the greatest exponents of the school of Buddhism and also of
the Madhyamika philosophy. As an institution in the true
sense, he drew votaries to his shrine of learning both in Kucha
as also in the Chinese capital where he stayed till the end of his
life. Many other scholars joined Kumarajiva in China after
spending quite some time in Central Asia. These included
Sanghabhuti, Gautama Sanghadeva and Punyatrata—all from
Kashmir—who went to China by the end of the fourth century
A.D. Dharmayasa, pupil of Punyatrata, was also associated with
the translation of a number of important texts of the Buddhist

Sarvastivadin school. He stayed there till 453 and then returned
to Central Asia and was finally back home in Kashmir.
Among the Buddhist savants of the first quarter of the fifth
century A.D. were Buddhajlva also from Kashmir who collabo
rated with Fa-hien in translating a couple of manuscripts
brought from India. A great teacher of Vinaya and a follower
of the MahiSasaka school, he translated three important works
of this school. Gunavarman, a prince of the royal family of
Kashmir, another erudite scholar from Kashmir, was invited by
the Chinese emperor, and during his short one year’s stay at
the Jetavana monastery he translated eleven works into
Chinese. Another inmate of the same monastery was Dharma-
mitra from Kashmir.
The Buddhist savants from Central Asia who contributed to
its thought and literature in later times can be assessed on the
basis of their translation of Buddhist texts into Chinese as also
their frequent visits to Central Asia for the collection of textual
material. Dharmaksema, originally from Central India and a
follower of Mahayanism translated 25 texts into Chinese. He
wished to return to Khotan in A.D. 433 but was killed in the
way. This was in connection with the incomplete text of the
Mahdparinirvana-Sutra of which the other part was probably in
Khotan. His pupil Tsiu-Kiu-Kingsheng, a nobleman, later went
to this great Mahayanist centre and studied the texts of this
school with Buddhasena, a great scholar. This centre of
Buddhist learning with- its famous savants continued to attract
many Chinese monks, and equally played an important role in
the transmission of Buddhism to China in the Tang period (A.D.
618-907). One of its greatest scholars in this period was
Siksananda who went to China and stayed there till his death in
A.D. 710. Another scholar from this place was Devaprajna or
Devendrajnana, who stayed in the Chinese capital till A.D. 689.
Shih-Kiyen of the royal house of Kustana (Khotan) was another
emigrant to China. All these scholars translated a number of
Buddhist texts which are fully recorded. Many such scholars
from other parts of Central Asia, who went to China and trans
lated Buddhist texts include Mitrasanta, Ratnacinta, Thien-si-
tsai (-deva), Prajiia, and Danapala from Tukhara, Kashmir and
Kabul respectively. Tibet also contributed in this matter.
The Summing-up 329
Pa-ho-sz-pa—Bash pa—a sramana of Tu-po (Tibet) was the

confidential adviser of Kublai-Khan as also his altar-ego. He
translated one work and in A.D. 1269, devised an alphabetical
system of the Mongol language which was utilised for writing.
While the Chinese source material provides information about
the life and activities of Central Asian savants as also their
contributions, archaeological finds of stQpas, sculptures and
paintings equally testify to the llourishing state of Buddhism in
Central Asia with its two prominent schools—HInayana and
Mahayana and their offshoots. The available artistic material
warrants a detailed study of Central Asian Buddhist art in all
its phases and facets. The finds of Kharosthi records, numbering
nearly eight hundred in a language which might be termed Pra
krit with various Iranian idioms and vocabulary introduced into
it, provide sufficient information for a detailed study of material
culture as also of its impact on the life of Buddhist monks. This
aspect is taken as a separate area of study. The religion of the
Tathagata in its relation with Brahmanism as also other religions
has also been brought out. The Tarim basin and the lands of
the Oxus afforded the necessary base for the mingling of
different religions and cultures; and it is quite likely that besides
Hinduism, Zoroastrianism and Christianity as also Manichaeism
had their impacts on Buddhism. In fact, Central Asia appears
as an area representing exchange of religions, ideas and art
from India as also from Iran, as is evident from the finds of a
number of manuscripts in different Iranian idioms. So also is
its debt to India unquestionable. But equally important is the
assessment of Central Asian contribution to Buddhism parti
cularly in respect of Buddhist literature and art. The relation
ship with Tibet was, of course, on a mutual basis. The Tibetans
occupied the Tarim basin for nearly a century, and the Buddhist
scholars and artists from Khotan made their contribution in the
‘land of the snow’.
The Buddhist literature as also its art from Central Asia
represent several periods and strata. The older one is revealed
from the fragments of Sanskrit Agamas, found at Turfan, Tun-
huang, and in the Khotan district, fragments of dramas and
poems of ASvagho?a from Turfan; the Pratimoksa of the
Sarvastivadins from Kucha and numerous versions of the antho
logy called Dharmapada or Udana. These along with the

Prakrit version found in the neighbourhood of Khotan, as also
fragments in Tokharian and Sanskrit—all those representing
the Buddhist canon as it existed during the time of Kaniska
coinciding with the eflloresence period of Gandhara art—are
supposed to represent the older stratum. The latter one is
symbolised by the abundant discovery of the Mahayana sutras.
The new stratum is said to comprise Mahayanist sutras,
found in abundance, particularly the Prajha-Paramita, the
Saddharma-pundarlka and the Suvarna-prabhasa. The last one
was translated from Chinese into Uighur, and into ‘Iranian
Oriental’. The ‘Dharanis’ or magical formulae discovered in
large numbers belong to a later period. It is proposed by
Sylvain Levi that some Mahayana-Sutras were written or re-
edited in Central Asia, since they contain references to Central
Asian place names, which could be the result of local patriotic
instinct. This is evident from the praises showered on the
mountain Gosringa, near Khotan in the Suryagarbha-Sutra or
from the list of holy places in the Chandragarbha-Sutra. Mani
festations of Buddha through his ray of light in Central Asia
also outnumber those mentioned in Indian texts. Further, one
of the Turkish Sutras discovered at Turfan contains a discourse
of the Buddha to the merchants Trapusha and Bhallika who
are described as Turks, and Indra is called Kormusta, that is
Hormuzd. In another Sutra, Brahma is called Asura, identified
with the Iranian deity Zervan. All these facts point to the local
colouring of the Buddhist doctrine; and might as well imply
amendations made in the Buddhist texts here.
There are also traces of interaction of Chinese ideas—Con-
fucian and Taoist—on Buddhism, as also the admixture of
Buddhism in Manichaeism. The dated inscription of the temple
erected in Turfan in A.D. 469 is suggestive of the former in
fluence. This record in honour of Maitreya, regarded here as
the future Buddha and an active and benevolent deity manifest
ing himself in many forms, also speaks of heaven (tien) as
appointing princes, and of the universal law (tao) ‘and it con
tains several reference? to Chinese literature. The Chinese edict
of 739 confirms the second hypothesis by accusing the Mani-
chaeans of falsely taking the name of Buddhism and deceiving
The Summing-up 331
the people. This is not surprising as Mani mentions that

Zoroaster, Buddha and Christ had preceded him as apostles.
His followers in Buddhist countries naturally adopted familiar
words and symbols to propagate the message of the Prophet.
Manichaean deities are represented like Bodhisattvas sitting
cross-legged on a lotus, and Mani is called Ju-lai or Tathagata.
Besides reference to Amida’s Paradise with holy trees bearing
flowers which enclose beings styled Buddhas, the construction
and phraseology of Manichaean books resemble those of a
Buddhist Sutra.
The co-existence of different religions in the Tarim basin,
with their impact on each other, did not exclude Nestorian
Christianity, introduced into China by A-lo-pen in A.D. 635,
almost simultaneously with Zoroastrianism. While the Nestorian
monument at Si-ngan-fu, commonly called the Nestorian stone,
and the finds of the fragments of the New Testament at Turfan
mostly of the ninth century could be definite proofs of this new
creed and its propagation in Central Asia, the reference to many
Buddhist phrases, such as Seng and Ssu for Christian priests
and monasteries, and the deliberate omission of the crucifixion
suggest adaptation of the Christian doctrine to suit the native
sentiments. It is thus possible that in western China and Central
Asia Buddhism, Taoism, Manichaeism, Nestorianism and
Zoroastrianism made mutual impacts on each other. There is
of course’not much evidence of the modification of Buddhism
except in local colouring, as proposed earlier. Buddhism in
strength and in numbers was the most important of all the
religions and equally the earliest to reach Central Asia.
From the point of view of material culture, Central Asia
presents a picture of rich and assimilative cultural pattern.
Physical configuration, with vast areas ' covered by the deserts,
and mountainous and hilly tracts, accounting for difficult com
munication tracts, had no doubt, contributed to the develop
ment of local cultures particularly nomadic and pastoral.
Historical factors, however, helped in establishing some form of
cultural integration. This was symbolised by. Buddhism and a
common way of life, at least in those areas where it was domi
nant. The finds of inscriptions at Lou-lan, Endere and Niya no
doubt present a picture of material culture in its multifacet
forms. The plantation of Indian colonies and that of others

on the trade routes had set the process of cultural fusion into
operation. The Chinese pilgrims passing through Central Asia
between the fourth and the seventh centuries have recorded
this fact with reference to Buddhist thought and way of life.
There are references to rulers bearing Indian names in both the
sectors, such as the Vijaya rulers in the south at Khotan and
Puspa and Deva ending rulers at Kucha, and those with Arjuna
suffix at Karasahr (Agnidesa) on the Northern route. Other
Indian names appearing in records are Ananda and Buddha-
mitra, Dharmapala, Punyadeva and Vasudeva, as also
Epic names like Arjuna, Bhimasena, etc. The titles taken by
the rulers are also Indian, like maharaja rajatiraja and
avijitasitpha—‘unconquered lion’. Sometimes there are mixed
names like Vasu Mogiya, Vasukekeya etc. suggesting under
standing and assimilation in the cultural ethos. The impact of
Buddhism seems to have eliminated caste consciousness, but
there were several classes of people, ones including chiefs, feudal
lords (cozbo), affluent householders and administrative officers,
forming the upper crust of society, while the lower one included
slaves, workers and artisans. Inter-class or caste marriages as
also between the Indians and the natives—apparently as their
names suggest—were equally known. A priest Sariputra receives
an adopted child Dewga Ainto—naming her Siratoyae and
finally marries her to another priest Buddhavarma in a lawful
manner. Their daughter later marries the priest Jivalo Athane.
Three generations of cross relationship through marriage is signi
ficant. The family continues to be a well-knit joint unit consis
ting of the father, mother, brother and younger sisters, as also
grand-father. Indian terms are used, such as kula and parivara
for the family. The head of the unit exercised control over other
members and also looked after their welfare. Adopted children
and even slaves formed part of the family. Ladies of the upper
stratum were commonly educated and could communicate in
writing with friends and relations. Slaves constituted the pro
perty of the master who generally utilised their services in farm
ing. They could be immune from bondage after payment of
the money invested on them. Even Buddhist monks kept slaves
to look after their farms and property interests and there could
The Summing-up 333
be change of masters as well. Food, dress and ornamentation

as well as pastime and items of recreation appear to be more in
tune with the Indian pattern of social life. It appears that along
with Buddhism, Indian music was taken to Central Asia—an
inference based on the depiction of such musical instruments in
paintings from Khotan and Kucha regions. Actors, musicians
and dancers were actually taken by the Chinese General Lu-
Kuang to China in A.D. 382 when the kingdom was con
quered.
The economy of Central Asia was confined to agriculture,
cattle rearing and trade as well as some avocations. Irrigation
had its importance which was well-realized in the absence of
rainfall and that too when the land was barren or not so fertile
and productive. There are references to several types of land—
arable, mishi and barren—akri. The ploughing and sowing
operations are also recorded as also the amount of seed required
fo ra measured plot of land which equally determined its market
value. Slaves and women could own as well as alienate their
lands. The Khotan area was noted for cotton and hemp pro
duction. Animal husbandry, ‘connected with pasture lands
equally engaged Central Asian peoples in their economic activity.
Camels and horses were used for transport purposes and the
former also catered to exchange deals. Sheep provided wool,
while rams carried light goods. Cows yielded milk from which
ghee was produced. Ghrita—or ghee,—a purely Indian product—
was made on a big scale, as might be inferred from the reference
to a hundred ghee jars. Industries, especially related to carpet
and silk, leather, cotton and utensils are also mentioned
in records. Vineyards provided grapes for processing and the
distilleries had the support of the government. There was a
special department for the collection of taxes on old and new
wines. There were also state dairy farms, like the state distil
leries. An officer called Satavida looked after the State enter
prises.
There appears to be a mixed economy with the participation
of different interests, not excluding the Buddhist monks who
were well-off and not much different from other householders.
They could marry and also own property, including slaves. The
sale of yellow metal—gold is recorded in several records. There
are also references to different types of coins—sadera-stater,

trachma —drachm, muli and ghare. These suggest trade dealings
in cash as well. Weights and measures are also mentioned such
as khit and milima etc. Labour and transport facilities were
equally available to prevent the economic stagnation. Adminis-
tra .i. n too had close links with the rural economy as also with
industries and avocations, and the despatch of taxes was not
delayed. The material finds in excavations reveal a fairly good
standard of socio-economic life, closely associated with native
traditions and equally receptive to foreign impacts and influen
ces. It is no doubt difficult to sift or apportion the contribu
tions—Indian, Iranian and Chinese—to the composite culture
of the Central Asian peoples in different sectors. There was not
much change till the tenth century A.D. when Islam had spread
its influences and gradually changed the tenor and temper of
the converted population.
Buddhism and its expansion from India to Central Asia and
thence to the Far East, no doubt, provided the base for the
artistic activities in both the areas. Buddhist art centred round
the life activities of the Buddha and his previous births based
on the narrations in Buddhist texts. In this enterprise all the
Buddhists joined hands or acted independently in offering their
services. That accounts for the impacts or influences from
different quarters—classical Greek and Roman, Persian and
Sassanian—in the pictorial and sculptural art of Central Asia.
While the theme continues to be Buddhist, the actors in the
pictorial drama change with the painter who makes full use of
his imagination and background and evolves his own colour
scheme. In this context the earliest impact was with Gandhara
and its artists who are supposed to have been inspired by Greek
traditions as modified in Rome. These artists carried with them
their pictorial and sculptural art to Central Asia. The native
talent accepted these artistic influences with discrimination. A
facile technique is, thus, traced in Central Asian art. The
examples of painting in caves and free-standing shrines suggest
an advanced state and productive capacity with the expansion
of Buddhism and the setting up of a large number of stupas
and shrines, the demand for' artists and sculptors considerably
increased. The commercial enterprise on the trade routes with
The Summing-up 335
the patronage and contribution of traders and merchants pro

vided incentives for artists—experienced and young, who were
roving from one centre to another. They received handsome
payments for their services.
The paintings rightly grouped into two—on a geographical
basis—reflect, as proposed earlier, the impacts and influences
of diverse origins. Well-defined styles of Central Asian paint
ings, no doubt, suggest Indian and Chinese inspirations in their
respective areas. Indian qualities dominate in the south with
Persian influence intruding, while the Chinese features are per
ceptible, of course, with modification by Tibetan, Uighur and
others in the north. The common factor in both is the Buddha
and his legend—variously expressed. Hindu and Tantric im
portations as well stimulated the imagination of Chinese and
Tibetan artists who were more interested in the decorative part
than the spiritual element which inspired the earlier Indian
renderings. The manner of expression in art renderings differs
widely in style and treatment. This could be due to the chang
ing political situations through the centuries and the complex
social milieu. Diversities in design and treatment did not com
pletely rule out identical mannerism in some compositions at
places which might be at considerable distance from each other.
This might be due to some artist and his troupe of painters
moving from one centre to another.
The earliest Buddhist paintings are from Miran and the sites
lying on the old silk route. Indian conception and execution
are noticed in the type of persons in Indian dress; the Persian
type of beauty no doubt suggests Iranian contacts. The reference
to Titus and the fees paid to him, as recorded in a KharosthI
inscription confirms the role of the Indo-Greek artists in this
region. TheGandhara school is supposed to be transplanted in
to pictorial lines at Miran where Roman-Hellenistic influences
are evident. This influence from Gandhara predominates in the
Khotan complex between the fifth and eighth centuries with the
assimilation of other foreign characteristics. The Centres on the
Northern routes—Kizil, Kucha and Turfan—were equally
connected with or influenced by others. At Kucha—the Indo-
Iranian (c. 500 A.D.) and strongly Iranian (c. 600-650 A.D.)
phases are noticed. Indian influence is prominent in several
paintings like the one depicting the cowherd listening to the

sermon of the Buddha or in the group of swimmers at Kizil.
The paintings from Kumtura suggest a third phase characterised
by Chinese influence eliminating or absorbing the Iranian motifs
and types. The Mongolian features, head dress, drapery and
folds of dress and scarves are traditionally Chinese. The Turfan
group covering a period of three hundred years is noted for
mannerism is physiognomy, pose and costume of the figures.
Its different phases vary from one region to another with chang
ing degrees of Chinese influence and dominance of Tantric
features. At Khocho, Manichaean and Nestorian influences are
conspicuous with interaction of Chinese and Iranian cultures on
local art.
The paintings at Tun-huang, the meeting place of the two
routes is noted for the caves of the thousand Buddhas, provid
ing an idea of Chinese Buddhist art in concept and design. A
small group is supposed to represent Indian art, and a few are
Tibetan. These paintings are placed between the fifth and eighth
centuries A.D. At the other end the western part of Central
Asia now in Soviet Central Republics had its own art forms as
revealed from a number of sites representing old Buddhist
centres. The recent discovery of wall paintings at Balatik-Tepe
near Airtan Termez, are supposed to be Sassanian inspiration.
The earliest Indo-Hellenistic blend is noticed in the first century
A.D. superb sculptured lime-stone frieze.
The two countries at the periphery of Central Asia—namely
Afghanistan and Tibet—are also significant for their contribu
tion to Buddhist art. The former was connected with the
Gandhara art, and provides examples of rock-cut statues,
sculptures and paintings. Here the fusion of Iranian and
Indian elements produced a hybrid style, the direct antecedent
of the Buddhist paintings in Central Asia, especially at Kizil
and Murtuq. The paintings in Tibet are supposed to have
developed mostly under Indian influence with the participation
of artists from Northern India as also from Khotan. The Art of
Central Asia was Buddhist in character, assimilating the forms
of great sedentary peoples around them. At an early stage it
depended on the Greco-Roman art of Gandhara, but subse
quently gave birth to a series of independent movements which
The Summing-up 337
influenced and were influenced by the major civilizations of
Asia.
The Central Asian Buddhist art and religion exhibit certain
traits like deification, pantheism, creation of radiant or terrible
deities, extreme form of idealism or nihilism. Buddhism lately
borrowed many personalities from the Hindu pantheon, and
one finds Buddhas and Bodhisattvas such as Amitabha Avalo-
kita or Avalokitesvara, ManjuSri and Ksitfgarbha which might
not be having antecedents in India. It is suggested that they
were borrowed from some other mythology. Amitabha, a
benevolent deity with his paradise of light is compared to
Ahuramazda. All the features of his paradise are said to be
Persian, though the concept of Tusita heaven is Indian Bud
dhist and fairly old. Avalokita is also connected with Ami-
tabha’s paradise. There is no reason for his special association
with Central Asia, since he assumes distinctness and import
ance much earlier in India. Later works describe him as the
spiritual son or reflex of Amitabha. The Zoroastrian doctrine
of the Fravashi—the spiritual being conceived as a part of a
man’s personality but existing before he is born and inde
pendent of him—seems to be more appropriate in defining the
relationship between a Dhyani Buddha and his Bodhisattva.
Further, Bodhisattva ManjusrI according to Sylvain Levi is
supposed to be of Tokharian origin being worshipped at Wo-
tai-Shan in Shan-si. He is connected with China according to
the Indian tradition, while local ones associate him with
Nepal, Tibet and Khotan. There is no clear proof of his
Central Asian origin. The same might be true about the Bodhi
sattva Ksitigarbha as well. While he was known in India in
the fourth century A.D. his cult was not prominent here; it
flourished in China, finally making him a popular deity in the
Far East, second only to Kuan-Yin. He was gradually trans
formed into a god of the dead.
Central Asia seems to have played a very active role as a
transit centre, sending ideas, icons and merchandise from one
end to the other. It further provided its Buddhist savants and
scholars to propagate Buddhist doctrines and canons in the Far
East, especially in China, translating these in the language of
the country. From the beginning of the Christian era onwards
monks went eastwards from Central Asia to preach and trans

late the scriptures and it was through this country that Chinese
pilgrims came to India ‘in search of truth’. A long history of
Central Asian Buddhism is a saga of creative religious and
artistic activities displayed on a wide horizon, in which peoples
of all shades and ethnic origin seem to have taken a fairly
active part.
SELECT BIBLIOGRAPHY
BOOKS
Andrews, F.H. : Descriptive catalogue of Antiquities recovered
by Sir Aurel Stein during his explorations in Central Asia,
Kansu and Eastern Iran, London, 1936.
-------- Wall Paintings from Ancient Shrines in Central Asia,
London, 1948.
Anand, Mulk Raj : Tibetan Art. Marg Special Number, Vol.
XVI, No. 4, 1963.
Bachofer : History of Chinese Art.
Bagchi, P.C.: India and Central Asia, Calcutta, 1955.
Le Canon Bouddhique en Chine, Tome I, Paris 1927; tome
II, Cal. 1938.
-------- India and China, Calcutta, 1944.
Banerji, J .N .: Development of Hindu Iconography, Calcutta,
1956.
Banerji, P. : Hindu Trinity from Central Asia, Bulletin, New
Delhi, 1970.
Beal, Samuel: Buddhist Records of the Western World, Vols. I
& II, London, 1906.
Benjamin, R.: The Wall Paintings of India, Central Asia and
Ceylon—a comparative study, Boston, 1938.
Belenitsky, A. : Central Asia, London, 1969.
Bhattacharya, Chaya: Art of Central Asia, Delhi, 1969.
Boyer, A.M., Rapson, E.J. & Senart, E.: Kharosthi Inscriptions
discovered by Sir Aurel Stein in Chinese Turkestan, I-III,
Oxford, 1921-29.
Brough, J.: The Gandhari Dharmapada, London, 1962.
Burrow, T.: The Language of the Kharosthi Documents from
Chinese Turkestan, Cambridge, 1937.
-------- A Translation of the Kharosthi Documents from Chinese
Turkestan, London, 1940.
Bussagli, M.: Paintings of Central Asia, Geneva. 1963.
Chattopadhyaya, A : Atisa and Tibet, Calcutta, 1967.

Coomaraswamy, A. : History of Indian and Indonesian Art,
London 1927, 1965.
Dabbs, J.A. : History of the Discovery and Exploration of
Chinese Turkestan, The H§tgue, 1963.
Eitel: Handbook of Chinese Buddhism.
Eliot, Charles : Hinduism and Buddhism, Vols. I-III, London,
1921.
Eramerick, R.E. : Tibetan Texts concerning Khotan, London,
1967.
Foucher, A. : L’art greco-bouddhique du Gandhara, Vols. I.
III. Paris, 1905, 1918, 1933.
Thomas, L. A : Beginnings of Buddhist Art, London, 1917.
Gaulier, S. et others : Buddhism in Afghanistan & Central Asia,
2 Vols. Leiden, 1976.
Giles, H.A.: The Travels of Fa-hien, Delhi, 1972.
Grey, B. : Buddhist Cave Paintings at Tun-huang, London,
1959.
Grunwedel, A.,: Buddhist Art in India, London, 1901.
-------- Alt buddhistiche Kultstattin in Chinese Turkestan,
Berlin, 1917.
Hackin, J. : Buddhist Art in Central Asia etc. London, 1937.
Hambly, G .: Central Asia, New York, 1969.
Hambis, L .: L’Asie Histoire et Civilisation, Paris, 1977.
Hedin, S.: Central Asia and Tibet Towards the Holy City of
Lassa, 2 Vols. London, 1903.
-------- Scientific results of a journey in Central Asia, 1899-1902,
Stockholm, 1904-7.
-------- Through Asia, London, 1898.
Hopkirk, P. : Foreign Devils on the Silk Road, London, 1980.
Huloewe, A.F.P. : China in Central Asia—The Early Age,
Leiden, 1979.
Ingholt, H. : Gandhara art in Pakistan, New York, 1937.
Kumagai, N. : Central Asian Painting.
Legge, J . : Fa-hien’s Travels, Oxford, 1886, Tokyo.
Le Coq, A. Von. : Buried Treasures of Chinese Turkestan,
London, 1928.
Mac-Govern, W. M : The Early Empire of Central Asia, Univer
sity of North Carolina Press, 1939.
Select Bibliography 341
Majumdar, R.C. : The Age of Imperial Unity, Bombay, 1960.

-------- Classical Age, Bombay, 1964.
-------- The Age of Imperial Kanauj, Bombay, 1964.
Marshall, J. : Gandhara Art, Cambridge, 1968.
Pelliot, P .: Lesgrottes de Touen-houang, 6 Vols. Paris, 1920-24.
Petech, L. : Northern India according to Shui-Ching-Chu, Rome
1950.
Petrucci: Les peintures bouddhique de Touen-houang, Paris,
1916.
Puri, B.N. : India under the Kushanas, Bombay, 1950.
Remusat, Abel. : Histoirede la ville de Khotan, Paris, 1820.
Rice Talbot, T. : Ancient Arts of Central Asia, London, 1965.
Rapson, E.J. : See under Boyer.
Rockhill, W.W. : The Life of the Buddha, London, 1884.
Rowland, Benjamin : The Art and Architecture of India,
London, 1969.
-------- The Art of Central Asia, New York, 1974.
-------- Art in Afghanistan, London, 1971.
Rudolf Hoernle, A.F. : Ms. Remains of Buddhist Literature
found in Eastern Turkestan, Oxford.
Stein, A .: Sand Buried Ruins of Khotan, London, 1904.
-------- Ancient Khotan, 2 Vols. Oxford, 1907.
-------- Ruins of Desert Cathay, 2 Vols. London, 1912.
-------- Serindia, 5 Vols., Oxford, 1921.
-------- Innermost Asia, 5 Vols., Oxford, 1929.
-------- On Ancient Central Asian Tracks. London. 1920.
-------- Ancient Buddhist Paintings from the Cave Temples of
Tun-huang, London, 1921-22.
Walters, T. :On Yuan-Chwang’s Travels in India, London, 1905.
PAPERS
Aalto, P .: On the Role of Central Asia in the spread of Indian
Cultural Influence. (Vivekanand Commemoration Volume,
Madras 1970, pp. 249-62).
Agrawala, R.C. : Some Aspects of Indian Culture in the Kharo-
sthi Documents from Chinese Turkestan (Vivekanand
Volume, pp. 275-81).
-------- Position of Women and dependents in the KharosthI
documents from Chinese Turkestan (Indian Historical

Quarterly—IHQ—1952).
--------Position of slaves and Serfs in the Kharosthi documents—
IHQ 1953, pp. 97-110.
-------- Numismatic data from Kharosthi documents—Journal
of the Numismatic Society—JNSI—1953, pp. 103-06.
--------Numismatic data in Niya Kharo?thi documents—JNSI—
1954, pp. 219-30.
-------- A study of weights and measurements in the Kharosthi
documents—Journal of the Bihar Research Society—
JBS—pp. 365 ff.
-------- Form of Taxation in the Kharosthi Documents, IHQ.
1953, pp. 340-53.
-------- Some family letters in the Kharosthi documents. IHQ.
1954, pp. 50-67.
-------- A study of textiles and garments in the Kharosthi docu
ments. (Bharatiya Vidya) Bombay, 1953, p. 75-94
-------- Life of monks and serfs in the Kharosthi documents.
(Swaroop Bharati) 1954, pp. 173-81.
Bagchi, P.C.—Indian Culture in Central Asia, JBORS, 1946,
pp. 9-20.
Banerji, P. : Indra from Balawaste. Journal of Indo-Iranian
Culture, 1968, Vol. XVII.
-------- Hindu deities in Central Asia. (Vivekanand Volume) pp.
281-88.
Bhattacharya. C. : India—a major source of Central Asian Art
(ibid, pp. 289-298).
Bailey, H.W. : Indo-Iranian—Bulletin of the School of Oriental
and African Studies. BSOAS. 1949-50. pp. 121-39; 920-
938.
--------Gandhari, ibid. 1943-46. pp. 764-97.
-------- The Khotan Dhammapada—ibid. pp. 43-56; 488-512.
-------- The Rama story in Khotanese (Journal of the American
Oriental Society) JAOS, Vol. LIX, pp. 460-8; See also in
BSOAS.
Binyon, L. : Essay on the art of Tun-huang Paintings. Serindia,
Vol. Ill, pp. 1428-31.
Brough, J. : Legends of Khotan and Nepal, BSOAS, 1947-48, pp.

333-39.
Burrow, T. : Iranian words in the Kharosthi documents from
Chinese Turkestan, BSOAS, 1934, pp. 509-16; 779-90.
-------- Tokharian elements in the Kharosthi documents from
Chinese Turkestan, JRAS, 1935, pp. 667-79.
-------- Further Kharosthi documents from Niya, BSOAS, 1937,
pp. 111-123.
Bushell, S.W. : Early History of Tibet, JRAS, 1888.
Bussagli, M. : Indian Influence in Central Asia (5000 years of
the Art of India).
Das, S. C. : Contributions on the Religion, History etc. of
Tibet. (JASB, 1881, pp. 187 ff; 1882, pp. Iff and 87 flf.)
-------- Indian Pandits in Tibet—Journal of the Buddhist Society
of India, I 1893.
Diakonova, N .: A Document of Khotanese Buddhist Iconography,
Artibus Asie, Vol. XXIII, 1960, pp. 229-30.
Dikshit, K.N. : Buddhist Centres in Afghanistan (Vivekanand
Volume, pp. 229-238).
Filliotzat, J. : The expansion of Indian Medicine abroad (ibid,
pp. 239-248).
Giles, H.A. : Tun-huang, JRAS, 1914, pp. 703 ff; JRAS 1915,
pp. 41 ff.
Gupta, S.P. : Pre-historic Indian Cultures in Soviet Central Asia
(Vivekanand Volume, pp. 239-248).
-------- Some Aspects of Buddhist Art in Central Asia (A.N.
JhaVolume-I, pp. 43-60).
Hackin, J. : Buddhist Art in Central Asia Indian, Iranian and
Chinese influences from Bamiyan to Turfan (Studies in
Chinese Art and some Indian influences, 1937, pp. 14.)
-------- The work of the French Archaeological mission in
Afghanistan.—Indian Arts and Letters—IA & L, Vol.
XVV, pp. 41-9.
-------- The Buddhist Monastery of Fondukistan (Journal of
the Greater India Society—JGIS—1940, pp. 1-14; 85-81).
Hambis, L. : Asia-Central (Encyclopedia of World Art, See also
articles on Khotan, Tibetan Art, Tun-huang and other
centres by the same author in this work. Vols. I-XIV).
-------- Sculpture et peintures de Haute Asie—Inedits de la

collection Pelliot La Revue des Arts. Ill, 1956, pp. 3-8.
Hirth, Friedrich: Story of Chang Kien—China’s^ Pioneer in
Western Asia. Text and translations of Chapters 123 of
Ssi-ma-Tsicn Shiki, JAOS, XXXVII, pp. 89-152.
Kingsmill, T.W. : The intercourse of China with Eastern Turkes
tan, JR AS, 1882, pp. 74 ff.
-------- The Migration and Early History of the White Huns,
JRAS, 1878.
Konow, Sten: Documents relating to the ancient history of the
Indo-Scythians.
-------- Strater and Drachin in old Kharosthi Inscriptions (Acta
Orientalia—AO, 1928, pp. 255-56).
-------- Some Notes on Central Asian Kharosthi documents
(BSOAS, 1943-46, pp. 513-540).
Litvinsky, B.A. : India and Soviet Central Asia (Vivekanand
Volume, pp. 263-274).
-------- Buddhism in Central Asia (Diushanbe Conference
papers, 1968).
Le Coq, A. Von : The Fourth Turfan expedition (JRAS, 1928, pp.
227-30).
Pelliot, P .: Indian Influences in early Chinese Art in Tun-huang
(IAL, 1926).
Puri, B.N.: Geographical Information from RajaSekhara’s works
(Journal of Indian History—JIH 1962, pp. 412ff.)
Rowland, B. : Studies in Buddhist Art of Bamiyan—Art &
Thought, 1947, pp. 46-50.
Stein, Aurel.: Archaeological Discovery in the neighbourhood of
Niya (JRAS, 1901, pp. 569-72).
-------- Archaeological work about Khotan (JRAS, 1901, pp.
• 295-300).
-------- Ancient Manuscripts from Khotan (JRAS, 1906, pp.
695 ff).
Rudolf Hoernle, A.F.: Ms. Remains of Buddhist Literature (JRAS
1910, pp. 834-8; 1283-1300; 1911, 447-77).
Ross : The caves of the Thousand Buddhas (JRAS, 1913 pp.
434 If).
Siren, O. : Central Asian Influences in Chinese Paintings of the
Tang Period, Arts Asiatique III, 1956, pp. 3-4
Silva, A : The spices and Silk Road (Vivekanand Volume, pp.

299-304).
Thomas, F.W. : Buddhism in Khotan & its decline (Asutosh
Mookerji Volume III,pp. 30-52).
-------- Glimpses of Life under Tibetan Rule in Chinese Tur
kestan Man. Vof. XXXIII, pp. 101 ff.
-------- Some Notes on the Kharosthi documents from Chinese
Turkestan (AAO, XII, 1934, pp. 39-90).
-------- ibid. XIII, pp. 45-80.
-------- Some Notes on Central Asian Kharosthi Documents
(BSOAS, 1943, pp. 513-40).
-------- The terms employed in Central Asian documents
(BSOAS, 1935-37, pp. 789-94).
Williams, J. : The Iconography of Khotanese Paintings (East &
West, Vol. 23, 1973, pp. 109-154).
INDEX
Asia—Central: Geographical features Iranian and Chinese phases 275,
I, Trade routes and people's move Western and eastern spheres 275,
ment in—3, Silk trade—3, political Indian mannerism and style in the
division of—4, Cultural exchanges first phase 275, Vajrapapi 276,
5, China and the West 6, Peoples Colour scheme 275, a young ascetic
of—7, Scythians and Hunas 8, 277, the group of swimmers from
Sakas and other nomadic tribes Kizil 277, the second phase at Kizil
of—9, Yuechi and the Kusanas 10, and absence of Indian influence 278,
Hephthalites—Huns 11, Turks in— the goddess and the celestial musi
II, Parthians and Mongols in—12, cians 278, the Kizil art and its
Seljuk Turks in—15, Routes— composite character 279, Kumtura
Northern and Southern 17, Kucha frescoes and Chinese influence 280,
and its importance 21, Jade Gale temple structures— examples of
and Tun-huang 22, Exploration of different styles 280, Nagaraja Cave,
ancient sites and finds in—24, caves of the Nirvana of Apsaras
Texts and monuments 25, New 281, Mongolian features 281, Dil-
languages of—25, Tibetan language dur-Akhur 281, finds of fragments
and literature in—26, Chinese texts of stucco statue 282, Kara-sahr 282,
from—26, Frescoes and silk Paint Cave temples at Shor-chuk and
ings from—27, Introduction of Mingoi—extensive collections of
Buddhism and its impact in—28. Buddhist remains 283, Impact of
Abhidharmcipitaka 195, 195n. Ir.do-Greco-Buddhist Art 283,
Agamas 190, 193 Seated Buddha in the Kirin cave
Asiani—identification 36n. (Shorchuk) 283, Paintings at Mingoi
Asanga 201 283, The Turfan group 284, Beza-
Aftasahasrika 197 klik and strong Tantric influence
Asvaghosa—fragments of Sariputra- 285, Three phases in the Bezaklik
prakarana of—from Mingoi 24, complex 285, Wall paintings from
205, finds of works of—24, —and Kho-chu 286, The Great Departure
Nagarjuna 13. scene 286, Secular scenes 287,
Asfadasasahasrikci 200 Turfan complex, Manichaeism and
Art of Central Asia—Buddhism and Nestorian Christianity 287, Mani-
art activities 255, Buddhist pictorial chaean miniatures in the Berlin
art and its dating 256, grouping of Museum 287, Hindu deities in the
Central Asian paintings 257, Indian background 288, other sites in the
influence 258, Persian and Chinese Turfan area—Sengin, Toyuk,
influences 258, Buddhism—the Khocho and Kara Khoja 289,
common link 258, Miran and its synthesis of Indian sinousity,
frescoes 260, an outpost of Gan- Hellenistic elegance and Chinese
dhara art 260, Buddha with six grace 289, Stucco reliefs at Toyuk
monks 261, Gautama Bodhisattva 361, Tun-huang—the caves of the
seated on throne 262, Lay wor thousand Buddhas 362, Impact of
shippers 263, Miran Paintings and Miran’s style and iconography at—
Gandhara School 263, the Khotan 290, mural paintings and figural
Complex 268, Centres of Khotan compositions 291, date of paintings
School 269, Specimens of Khotan 291, mixed influence on—291,
school 269, Bust of the Buddha and AvalokiteSvara 292, Indian influ
figure of Indra from Balawaste 270, ence fused in Chinese style at—292,
Hindu influence with Tantric sym Buddhist art in Soviet Central Asia
bols 272, examples of Iranian details 292, Termez 293, Wall-paintings
273, the silk princess scene 273, the at Balatik tepe 293, Indo-Hellenic
Northern School, 274, Indo-Iranian, blend in the Airtan freize 294, ruins
of Khalchyan 295, Toprak-Kala 92,—and the Yuch-chi 93, Tokha-

295, Stucco and painted decora rian Buddhist monks 94, Parthia
tions 296, Pendzhikent and its wall- and its Buddhist scholars 96, An-
paintings 296, Afghanistan and its Shi Kao in China 97, Parthian
Buddhist art 297, Hellenistic ele (An-Ngan) Buddhist scholars 98,
ment with a touch of Parthian Buddhism and Buddhist monks
vitality and Roman influence on from Sogdia (Samarkand) 99,—and
Buddhist art 298, Bamiyan and its the Kusanas 100, Kaniska and the
wall paintings 299, Sassanian in fourth Buddhist Council 100, Early
fluence 299, The Indian style of Buddhist missions to China 101,
painting 299, resemblance of paint translations of Buddhist scriptures
ings at Kizil 299, the two giant in China 102, Schools of—in the
Buddha statues 300, animal and Ku$ana period 123, Buddhist centres
floral motifs and designs 301, in Afghanistan 103, Buddhism in
Other Buddhist centres—Kakrak ScJthern States 104, Buddhism in
301, mystic diagrams of mandalas Kashgar 104, Kumarajiva and
of esoteric Buddhism 301, archi Dharmacandra— 104, Vijaya rulers
tectural motifs of columns 302, of Khotan and Buddhism 107,
Fondikistan 303, Indian style of Chinese princess of Punesvara and
paintings and sculptures at—303, of Khotan 109, Hsuan-tsang’s
Bcgram—treasure trove 303, Secu account of— 110, Buddhist texts
lar art at—303, Hadda 304, Stucco in— 112, Buddhist scholars and
Buddhist statues 304, Gandhara Chinese monks in—113, Aurel
region 305, Tibetan art 307, Paint Stein on Buddhism in— 113,
ings and Indian influence 307, ancient Buddhist sites in— 113,
Tsog-shin paintings and Lamaite Kharosthi documents from— 113,
divinities 308, Concentric ones— Buddhism and Buddhist scholars
307, subject matter 307, Chinese from Northern States 114, Bud
influence 311, Tibetan architecture dhism in Kucha 114, Kumarajiva
and sculpture 311, bronzes 311, and his life activities 116, —in
Tantric divinities 312, A review of China 118,—and Mahayana Bud
Central Asian art 314. dhism in the Tarim basin 117,
literary contributions of— 117. and
Bactria—33, 33n, Greek occupation his contemporaries 121, Tibetan
of—37, political importance of— Buddhist scholars 128, Buddhist
37, Trapusa and Bhallika of—90, schools and their centres in Central
Hsuan-tsang at—90. Asia 129, Brahmanism in—129,
Bagchi, P.C. 90, 179 Beliodora and Bhagavatism 130,
Bailey, H.W.—on Gandhari lS3n, Krsna legend in Armenia 130,
184; On Ramayana in Central Asia the Narayana cult 132, Vi§nu, Siva
90. and Surya worship in— 132, Impact
Bernahtam, A.N. 99. of Vai$navism on Buddhism 132,
Bezeklik—largest Buddhist site of Saiva and other Brahmanica( sculp
Turfan 83. tures in Afghanistan 133, Saivism
Bower Manuscript 186. in Sogdiana 133,—in Dandan-Uliq
Brahmanism in Central Asia 130ff. 133,—in Eastern Turkestan 133,
Brough, J. 184n. Brahmanical gods in Central Asian
Buddhabhadra 111, 122. art 134, Gancsa in Tun-huang 134,
Buddhadeva 121. Lokapalas 134, the Rama legend
Buddhayasa 107, 122. in Khotan 135, Manichaeism,
Buddhist Council—fourth 37, 100, Nestorian Christianity and Bud
171. dhism in Central Asia 135, admix
Buddhism and Buddhist savants of ture of Buddhism and Manichaeism
Central Asia 86IT. in— 139, Nestorian Christianity in
Buddhism and State patronism 87. the Tarim basin 139, Buddhism and
Buddhism in Afghanistan, Bactria and Christianity 140, Some new trends
Parthia 89, Hsuan-tsang's refer in Buddhism 141, Amitabha’s wor
ence to—90. ship 142, Zoroastrian impact on
Buddhism and Indo-Greek rulers— Buddhism 142, Mafijusri and his
91 IT—Demetrius and Menander Tokharian origin 145, Bodhisattva
Index 349
Ksitigarbha and his cult in— 146, — Fa-hien 17, 113—on Central Asian
impact on Manichacnism 147, monks and Indian languages 24,—
Tibetan Buddhism 147, Indian on routes to India 24,—on Buddha’s
scholars in Tibet 148, Padma- bowl 51, —on Khotan 55.
sambhava 148, Kamalasiia 149, Fa-ling 126.
Tantric Buddhism in Tibet 151,
Atisa Dipankara and his contribu Gandhara—art of. —influence on
tion 151, Tibetan translation of Central Asian art 258, Foreign
Buddhist texts 154, Buddhism in influence on Gandhara art 256.
Mongolia 186, the role of Pag-spa Gandhari language 183
187, Later history of Buddhism— Gautama Sanghadeva 123.
157, the grand Lama and Lamaism Ghrishman 40, 40n.
161, Lamaism— 159, the Kalacakra Gho$aka 94.
system—Vajrayana in Tibet 162, Gunavarman 125, 328.
the last phase in Tibet 162, Hevajra,
Buddhakapala and Yamantaka, Hoernle—Manuscript remains I3n,
Hayagriva 165, the praying wheel 179n, 207.
167, the old and the new order 168, Horse Culture 50.
Different sects—Ge-lug, Kar-gyu, Hsuan-tsang—on Central Asian
Sa-kya 169, A review of Buddhism routes 24,—on political kingdoms
in Central Asia 170IT. of Central Asia 45, —on Kashgar
51, —on Khotan 53, —on Kucha
Chang-Chien mission 46. 79, —on Chokuka 168.
Chingez Khan 15, and Sakya Pandit Huns 10, 11, 43.
157. Hunnish nomads 8.
China—Silk trade with—225, Bud
dhist scholars in China 96ff, — I-tsing —on translation of Sanskrit
Buddhist texts translated in—135ff. texts into Chinese 208.
Chinese—hold over Central Asia 45,
Political mission in—46, Decay of Jdtakamdld 207.
Chinese power in—47, resumption
of—76.
Che-ma-to-ma—Calden Cher-Chen 64 Kalpanamanditika 206, 206n, 207.
Chih-meng 51. Kamalasiia 150n.
Chokkuka 108. Kani?ka and Buddhist Council 37.
Coedes on Menander 93. Karasahr 17, ancient Agnidesa 21,
Culture—See under Material Culture. 74f.
Cultural diversities in Central Asia Kashghar —importance of — 20, 46,
225. Buddhism in —50, Hsuan-tsang
Cultural integration 227. in—51, Script of — 106, Hinayana
Cross cultural fertilisation 231. monks in — 106, Scholars of — 106.
Khotan-Kustana 20, 20n, 52, —rela
tions with China 54, 55, Indian
Dandan-Oiluk—Buddhist monastery monks in—61,65, Mahasanghikas
at—58, 58n, finds of manuscripts in—55, Texts translated in —62,
at— 112, Stein discovery—58n, Vijayasambhava of —53.
Saivism in— 133 Khotan —an active centre of Bud
Dhammapada 19, 92, 184 dhists 61, Ramayana legend in
Dharmacandra 108 — 135.
Dharmagupta 81 Koshelanko, G.A. 97.
Dharmayasa Kucha —its importance 18, 21 n,
Dharmaksema 1, 125 —rulers called Po 79,—Kumarajiva
Dharmanandi 190 of —81, 81 n, Savants of —80,
Dhdranis 202,—in Central Asian finds —Centre of Buddhism 114, 115
203.' Kuchean Texts 223.
Kumarajiva —life and activities 2In,
Economic Life—See under Material 8 In, 116ff, —and his contempo
Culture raries 131ff, —in China 122f, —and
Endere—KharosthI records from—19, his literary contributions 122.
226. Kumaralata 105n.
Kujanas —Contact with Rome 39, Samitindtaka 212, Sidd/tasdra Jlva-

role of — in Central Asia 40. kapustaka 211 212. Buddhist texts
in Sogdian and Uighur 213, Contri
Lamaism 163f. butions of local scholars 214, Geo
Langa-darmas 167, 167n. graphical and topographical texts
Languages and Literature—Central 215, Sanskrit texts in Afghanistan
Asian 174, Kuchean 174, Nor- 216, Tibetan Buddhist literature
tharyan-Saka 174, Khotanese 175, 216, Dulva, Prajndpdramitd, Tantras
Finds of manuscripts 176, Sanskrit 218, Tanjur 219, Kanjur 220,
and its use in learned society 177, Ralpchan and Tibetan Buddhist
Northern Tokhari-Kuchean langu literature 220, A Survey 222.
age 177, Nordarisch Saka 180, Levi, Sylvain 206, 332.
Iranian Aramaic alphabets 181,
Sanskrit texts in Brahmi and Kha- Mahasahghikas 103,105,—in Khotan
rosthi scripts 230, Turkish dialect 109, 111.
in Uighur alphabet 182, Tibetan 182, Mahdparinirvanasutra 194, —its
Chinese texts 182, Gandhari Prakrit search by Dharmak$ema in Khotan
183, Dhammapada in Gandhari 183, 193.
Sukhdvativyuha 185, other texts in Manichacism 79, 194.
Gandhari 185, Buddhist texts in Matriccfa 207, —Satapancdstaka 207
Chinese 185, Gandhari Prakrit in Mahayanism 129, 111.
Kharo$Jhi script 185, Brahmi and Mahayana texts 89f —See under
Kharo$|hi in Central Asia 185, literature.
Asvagho§a’s dramas and Kalpana- Material Culture 225, Cultural diver
mancjiiikd in Brahmi Script 186, sities 226, Horse culture 226, Indian
Sanskrit introduced by the Sarvasti- cultural integration 227, the role
vadins 189, Hinayana literature in of the Indian rulers of Khotan and
Kucha and Agnidesa 189, Maha- Kucha 228, Indian colony of
yana literature in Khotan, Kashgar Karasahr 229, Indian missionaries
and Kucha in Sanskrit 189, Manus 229, Buddhism and cultural integ
cript finds of Sanskrit —Tokharin ration 229, Cultural data from the
and Sanskrit —Chinese lexicons Kharo$thirecords 231, Indian names
189, Kantdra Sanskrit grammar 189, and the Ramayana legend in Khotan
Canonical literature—Agamas 190, 231, Family life 232, Position of
Vinayapitaka, Pratimoksa Sutra, women 233, Marriage 234, Slaves
193, B/iik$unipratimok <ta 194, Mana- 234, Food and Food habits 236,
par inirvuna Sutra 194, Abhidharma- Dress and Ornaments 238, Dress
pifaka, Sangitiparva 195, Suttapitaka of the upper strata in paintings 238,
Udanavarga 196, Mahayana texts items of ornamentation 240, pas
—Aftasdhasrikd Prajndpdramitd, time and recreations 241, music in
Saddharmapumjarika, Suvarna- Khotan and Kucha 242, Agricul
prabhdsa, Gandavyfiha, Tathbgata- tural Economy 243, Irrigation 244,
guhyaka Samacihirdja, Hasubhumi- Crops 245, pasture lands 245,
svara 197, Prajndpdramitd, Malta- Handicrafts 245, Silk industry 247,
prajndpdramitd Sutra 199, Affddasa- dealers in food and drinks 248,
sdhasrikd Prajndpdramitd 200, National Economy, Medium of
Vajracchedikd Prajndpdramitd 200, Exchange and barter 248, Weights
Ratnadhvajasutra 202, Dhdranis 202. and Measures 249, Labour & Trans
Non-Canonical texts 204. liuddha- port 250, Silk Route of Central Asia
carita, Saundarananda & Sdriputra- 251, Administration and Rural Eco
prakarana 205, Sutrdlankdra, Srad- nomy—Assessment 252, Cesses and
d'lotpddasdstra, Sutrdlankdra 205, Taxes 252, The Indian element in
Kalpandmanditikd, ,206, Catur- administration 253, A Review 253f.
Satavastttra 206, Satapancdstaka McGovern—on the role of Central
stotra 207, Catuhsataka-stotru 207, Asian hordes 5, 5n.
Jdtakamdld 208, Chandaviciti 208. Miran —explored and excavated by
Sanskrit Medical texts 209 —local Stein 67, ancient Buddhist struc
translations 210. Udanavarga, tures and frescoes at —68.
Uddnastotra, Uddndlankara, Kartna- Moksagupta, 82.
vibhanga, Yogasataka 211, Maitreya Mok$aia, 110.
Index 351
Nestorian Christianity and Buddhism Saddhannapundarika 197—translated

137. by Dharmaksema 197, —by Ku-
Non-Canonical texts —See under marajiva 198, —by Jinagupta 198,
Literature. by Dharmagupta 198, —into
Tibetan 198, Hoernle’s reference
to — 198, finds of fragments of
Padmasambhava 149, 151, , — 198.
Parthians & Sogdians —role in the Saivism 133fF.
expansion of Buddhism 12, — Sakas 31, 32, Movements of —34.
monks 13, Samadhiraja 242.
Political History—Early Peoples of Sangitiparavaya 240, 258,
Central Asia 30, Massagatae Con Sanghabhuti 123.
federacy 30, Sakas of Herodotus Sang-Yuan 128.
and Samaritans 31, Achacmenian Santarak$ita 149.
domination over Central Asia 32, Sanskrit —in Central Asia 188, intro
Alexander in —32, Yuch-chi, duced by Sarvastivadins 189, patro
Wu-Sun and Hiung-nu 34, the nised by other schools 190, Canoni
Kusanas in —36, Kani§ka and the cal literature in — 189f, —metrics
Ku$ana hold over—37, Buddhism 209.
under the Ku$artas in Central Asia Sdriputraprakarana 24, 178, 178n.
37, Sassanians in —40, Hephthalites Sarvastivadins 103, 104, 112.
in— 42, Turkish empire in —42, Sassanians—Conquests of Central
Mongolia and the Uighurs 44, , Asia by —40, 4 In.
Political States of —45, Kashgar Satapahcasatika Stotra 207.
45, Chinese hold 47, Tibetan Sautrdntikas 105.
domination over the Tarim basin Saundarananda 205.
46, Buddhism and Buddhist pil Scripts 186.
grims in Kashgar 50, Khotan 52, Sik$ananda 127.
early history 53, foundation of Sogdian monks—with the prefix Kang
—52, The Kutiala tradition 53, in China 101, Buddhist Texts 213;
Contacts with China 54, Fa-hien’s „ Sogdian colonies 173.
account 55, The Vijaya dynasty of Sraddhotpddasiitra 248.
—58, end of Chinese supremacy Stein, A.—on Jade Gate 22, finds of
over —55, explorations and ex Chinese Buddhist texts 22, Dis
cavations in —60, Translations of covery of Painted Panel at Dandan-
Buddhist texts in —60, Hsuan- oi-liq 112.
tsang and his account of —65, Sten Konow 10, 92n.
Remains of Stupas and Buddhist Strabo —on Saka Conquest 9.
shrines in —65, Miran and its finds Sutridamkara 207.
67, Paintings in Miran 68, Political Suryabhadra 325
kingdoms on the Northern Route Suryasoma 325
69, Aksu 69, Kuche 69, Chinese Summing-up 3171T. Central Asia—the
account 69, Chinese relations with Cradle of human civilization 317,
the Tarim basin States 70, Kucha, the geographical factor—movement
Aksu and Uch-Turfan 73, Later of peoples of—318, Centres of trade
accounts 73, Buddhism in Northern and travel 319, monastic establish
Slates 72, Buddhist scholars from ments 319, Centres on the Southern
Central Asia in China 81, Buddhism route—Yarkand, Khotan, Niya
under the Uighurs in Central Asia 320, —on the Northern route—
83, A Review 83. ' Kucha, Karasahr, Tun-huang, the
Prajhdpdramitd 198, also Malta—200, meeting place 320, Early history in
Vajracchediku 200, Astadasasdha- relation to Buddhism 321, Indian
srika 200, Prdtimoksa 193. missionaries to China 321, Bud
Puncsvara —Chinese Princess 109, dhism in Tokharestan 322, —in
21 In. Parthia and Sogdiana 323, —in
Punyatrata 124, 125. Kashgar 322, Buddhist scholars
and their contributions—Kumara-
Rama legend 135. jiva, Suryabhadra, Suryasoma 325,
Ratmeinta 325. Mahayana Buddhist Centres at
Rawak Stupa 112. Kashgar & Khotan 326, Centres on
the Northern routes 326, Other Bud Tocharistan and Buddhism 97.
dhist scholars—Buddhajiva, Guna-
varman, Dharmaksema 327, Prajna,
Dhanapala 328, Academic acti
211.
Tocharians 96; —Translations 206,
Tu-ho-lo—Endere 19.
vities 329, Buddhist literature 329, Turfan—Kao-Chang 77.
Impact of Chinese thought on Tun-huang 23, finds of manuscripts
Buddhism 330, Peaceful co-exis- at —23, 23n, 181n, 264, Paintings
tence of different religions 331, at —22.
Material culture— rich and assimi
lative 333, Agriculture and trade Udanavarga 209, 211.
economy 333, Mixed economy 332, Uighurs 44, 89, 139n, 182; Buddhism
Buddhism in relation to Buddhist under the—83
art 334, Art centres of Central Asia Uighur Texts 213.
335, Central Asia—a transit cultural Vijaya dynasty of Khotan 135
centre 336. Vijayasiipha 52.
Vimalaksa 81 n, 121.
Tibetans in Central Asia 13,14, 88. Vinayapitaka 190.
Tibetan Buddhism 147ff—See Lama-
ism. Yuch-chi 9, 9n; —equated with the
Tibetan Literature 259. Ku?apas 93.
Plate I. Buddha with Six Monks—from Miran (p. 261)
(Courtesy, National Museum, New Delhi)
Plate III. Two Girl Worshippers—from Miran (p. 263)
Plate IV. Hariti—from Farhad-Beg-Yailaki (p. 270)
Plate V. Bust of Buddha—from Balawaste (p. 271)
Plate VI. Worshipper or Indra—from Balawaste ( p. 271)
Plate VII. The Trimurti Divinity with a Small Seated Buddha—from Bala waste
(Courtesy, National Museum, New Delhi) (p. 272)
PUteVIH. Cowherd Listening t0 ,he Serm on-from Kizil fn 2761
(Courtesy. Berlin Museum. West Germany) P'
Plate IX. Head of MahakaSyapa—from Kizil (P -277
(Courtesy, Berlin Museum, West Germany)
Plate X. Group of Swimmers— from Kizil (P 278)
Plate XI. Goddess and Celestial Musicians—from Kizil (p. 279)
Plate XII. Buddha and Praying Monk—from Kumtira (p. 281)
Plate XIII. Worshipping Bodhisattva—from Kuratira (p. 281)
(Courtesy, Berlin Museum, W. Germany)
Plate XIV. Buddha under a Canopy—from Turfan (p. 286)
Plate XV. An Uighurian Prince—from Bezeklik (p. 287)
Plate XVI. Bodhisattva—from Hadda (p. 304)
(Courtesy, late V. P. Trivedi)
Buddhism in Central Asia is a saga of
peaceful pursuit by Buddhist scholars from
Kashmir and Kabul to propagate the
message of the Buddha. This vast region
between the Ticn-Shan and the Kunlun
ranges was the centre of activities of these
Buddhist savants. Here people of different
races and professions, speaking many
languages, were finally blended into a
cosmopolitan culture. This created an
intellectual climate of high order. In this
context, the famous silk trade route was
helpful in adding to the material prosperity
of the people in this region.
The present study, therefore, is not one of
Buddhism in isolation. It equally provides
an account of the political forces
confronting each other during the course of
history of this region for well over a
thousand years.
For centuries the drifting desert sand of
Central Asia enveloped this civilization and
.the religion connected with it The
late-nineteenth and early-twentieth century
explorers and archaeologists successfully
uncovered it at different centres along the
old Silk Route. This has been helpful for a
comprehensive study of Buddhism with its
literature and art. The finds of hundreds of
inscriptions have added to the cultural
dimensions of the study.
ISBN: 81-208-0372-8 R s. 295

Buddhist Tradition Series
Edited by Alex Wayman
(ISBN: 81-208-0287-x)
n Buddhism: A Survey with Bibliographical Notes— H ajim e N akam ura
Nagwjuniana: Studies in the Writings and Philosophy of Nagaijuna— O n . U n d tn tr
Chinese Monks in India— I-Ching, Latika L a h m , Tr.
Buddhism in Central Asa— B N P u n
Dharmakird's Theory of Helu-Centridty of Anumina—M angala ft Chinchore
The Legend of King Aioka: A Study and T ransladon of the AJokavadana—John
S. Strong
Buddhist Insight—Essays by Alex Wayman, George R. Elder, Ed.
Buddhism Transformed: Religious Change in Sri Lanka—Richard Combnch and
G ananath Obeyesekere
The Buddhist Tantras: Light on Indo-Tibetan Esotericism—Alex W aym an
The Lion's Roar of Queen Srimili—Alex W aym an and H idtho W aym an, Trs.
The Buddha Nature: A Study of the Tathagatagarbha and Alayavijnana—Brian
E. Broom
Evolution of Stupas in Burma: Pagan Period: 11th to 13th Centuries A.D. —
S u ja ta S cn i
Buddhist Parables— Eugene W atson Burlingame, Tr.
The Debate of King Milinda: An Abridgement of the Milinda Pah ha—B hikkhu
Pesala, Ed.
The Chinese Madhyama Agama and the Pali Majjhima Nikaya—B hiksu Thick
M in h C hau
Sudden and Gradual: Approaches to Enlightenment in Chinese Thought— Peter
N. Gregory, Ed.
Yoga of the Guhyasamajatantra: The Arcane Lore of FortyVerses—Alex W ayman
The Enlightenment of Vairocana: Study of the Vairocana-bhisarhbodhitantra
and Mahavairocana-Sutra—A lex Waymanand R .T a jim a
A History of Indian Buddhism: From Sakyamuni to Early Mahayana— H trakaw a
Akira, P a u l Groner, Tr. and Ed.
Introduction to the Buddhist TanUric Systems—E D. Lessing and A lex W aym an,
Trs.
Anagatavamsa Desana: The Sermon of the Chronicle-To-Be— U. M eddegama,Tr.
and Jo h n C Holt, Ed.
22 Chinnamasta: The Aweful Buddhist and HinduTantric Goddess—Elisabeth A n n e
23 On Voidness— Fernando Tola i n d Carmen Dragonetti

24 Nagaijuna's Refutation of Logic (Nyaya) Vaidalyaprakarana— Fernando Tola
and Carmen Dragonetti
25 The Buddhist Art of Nagarjunakonda— Elizabeth Rosen Stone
£ 3 $
Discipline: The Canonical Buddhism of the Vtnayapi{aka—/. C. H olt

Philosophy and its Development in the Nikayas and Abhidhamma—F. W atanabe
Untying the Knots in Buddhism: Selected Essays— Alex W ayman
MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED

Buddhism in Central Asia is a saga of
peaceful pursuit by Buddhist scholars from
Kashmir and Kabul to propagate the
message of the Buddha. This vast region
between the Ticn-Shan and the Kunlun
ranges was the centre of activities of these
Buddhist savants. Here people of different
races and professions, speaking many
languages, were finally blended into a
cosmopolitan culture. This created an
intellectual climate of high order. In this
context, the famous silk trade route was
helpful in adding to the material prosperity
of the people in this region.
The present study, therefore, is not one of
Buddhism in isolation. It equally provides
an account of the political forces
confronting each other during the course of
history of this region for well over a
thousand years.
For centuries the drifting desert sand of
Central Asia enveloped this civilization and
.the religion connected with it The
late-nineteenth and early-twentieth century
explorers and archaeologists successfully
uncovered it at different centres along the
old Silk Route. This has been helpful for a
comprehensive study of Buddhism with its
literature and art. The finds of hundreds of
inscriptions have added to the cultural
dimensions of the study.
ISBN: 81-208-0372-8 R s. 295

Buddhism in Central Asia by B.N. Puri

Uploaded by

Copyright:

Available Formats

Buddhism in Central Asia by B.N. Puri

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Buddhism in Central Asia by B.N. Puri

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What region does the book cover?

What region does the book cover?

BUDDHISM IN

Editorial Advisory Board

MOTILAL BANARSIDASS PUBLISHERS

© MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED

Also available at:

BY JAINENDRA PRAKASH JAIN AT SHRI JAINENDRA PRESS.

Central Asia is still an enormously important area of the world

The study of Buddhism in Centra! Asia cannot be carried

connected with the growth of the Kusana empire, maybe pre­

to other religions, especially Manichaeism and Nestorian

on the western model without alienating Indian characteristic

II. EARLY HISTORY OF CENTRAL ASIA 30

III. BUDDHISM AND BUDDHIST SAVANTS OF 86

Buddhism in Afghanistan, Bactria and Parthia 89

IV. LANGUAGE AND LITERATURE 174

V. MATERIAL CULTURE 225

VI. THE ART OF CENTRAL ASIA 255

VII. THESUMMING-UP 317

PLATES (after p. 352)

I. Buddha with Six Monks

The term Central Asia is supposed to connote the Tarim Basin,

aux—Tibet on the south, enclosed by the Himalayas, and Iran,

1. G. Hambly. Central Asia (London, 1969)—henceforth Hambly, p. 2.

Tarim Basin without making their occupation a permanent

Silk, one of China’s chief mercantile commodities became the

settlement. Many of its river valleys, however, with their fertile

2. McGovern : Early Empires o f Central Asia, Chapel Hill, 1939

manners and customs were transmitted from one region to an­

lished communication between India and China, and it was from

7. McGovern : Op. cit, pp. 6flf.

following, however, witnessed the turn of fortunes8 with the

8. While Turkestan was ruled by the Scythians, Mongolia in this

extended their tentacles in different directions, dominating

Turks are mentioned in the Chinese accounts as Tu-Kiu i.e.

Besides the Parthians, the Sogdians are also worth recording

blished the foundations of a civilization conducive to the efflore­

Shamanists, with the Shaman (boge) exercising a dominant role.

predecessors. They made occasional inroads into Persian territory,

19. Op. clt, p. 11.

tury, Italian merchants including the famous Marco Polo, had

Ancient Routes :20

tries in the periphery has been assessed rather summarily, it is

south of Kesch (Sahar-i-sabz) he reached the country of Tokharestan with

royal families. The routes starting from North-West India in those

Tokharians, the Tu-ho-lo of the Chinese, near Badakshan, and

The countries along the northern route from Kashgar to the

The grottos in the hills in the neighbourhood of Kuchar met

onwards, its importance28 increased and it also attracted forei­

and point to the cosmopolitan community of peoples that lived

Buddhist Finds—Literary Texts and Monuments3132

the script of the time of Kaniska. It is the oldest known Sanskrit

in Central Asia. These are written in an alphabet of Aramaic

36. Eliot : Op. cit, p. 191.

as well as Buddhism. There have not been any Tibetan trans­

39. Stein : Ancient Khotan. Op. cit, p. 483 fT.

connected with the growth of the Kusana empire, maybe pre

manners and customs were transmitted from one region to an

blished the foundations of a civilization conducive to the efflore

onwards, its importance28 increased and it also attracted forei

as well as Buddhism. There have not been any Tibetan trans

here, as for instance, some of the caves at Ming-oi have dome

of Guzan (Kani§ka of Kusapa family) and helped him in his con