The Tarim Mummies: Ancient China and the Mystery of the Earliest Peoples from the West 0500283729, 9780500283721

Citation preview

:.^^':^

ALLORY

.

VICTOR

H.

MAI

pent China and the Mystery

of fie Earliest Peoples from the West

IMkv

Thames & Hudson

The Tarim Mummies

J.

p.

MALLORY

•

VICTOR

H.

MAIR

The Tarim

Mummies Ancient China and the Mystery of the EarUest Peoples from the West

With 190

illustrations, 13 in color

^^^y Thames & Hudson

© 2000 Thames &

Hudson

All Rights Reserved.

No

Ltd,

London

part of this publication

may be reproduced

or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording or any other information storage

and retrieval system, without prior permission from the publisher. First published in

by Thames

&

hardcover

Hudson

Inc.,

in the

in

writing

United States of .\merica

500 Fifth Avenue,

New

York,

thamesandhudsonusa.com First

paperback edition 2008

Library of Congress Catalog Card

Number

99-66166

ISBN 978-0-500-28372-1 Printed and

bound

in

in

New

China by Hing Yip Printing Co. Ltd.

2000 York 101 10

CONTENTS

Introduction

Rediscovery

7

Chapter One

Beyond the Centres: Tarim Between East and West Chapter

34

Two

East Central Asia: Players at the Centre of the

Board

64

Chapter Three

The

Linguistic Landscape 102

Chapter four

Hoe

132

The Mummies Themselves

176

The Testimony

of the

Chapter

Chapter

Five

Six

Tartans in the Tarim

208

Chapter Seven Skulls,

Genes and Knights with Long Swords Chapter Eight

The Usual The

Suspects:

Indo-Iranians 252

Chapter Nine

Tocharian Trekkers

270

230

Chapter Ten

Who Were the Mummies?

298

Chapter Eleven

Legacy

319

Appendix One:

On

the Tocharian Problems

m

Appendix Two: Radiocarbon Dates From Selected

Sites 335

Bibliography 337 Sources of Illustrations 344

Acknowledgments Index

347

346

INTRODUCTION

The mummies

stalked into the world's pressrooms in 1994. London's

on Smniiiy could trumpet how the discovery of mummies in western China had stood 'history on its head'. Preserved in a state AL?/7

come out of ancient Egypt and decked in clothes gaudy even by 'New Age' standards, the mummies of Xinjiang, the westernmost and largest of the provinces of modern China, had clearly caught the public imagination. But it was also obvious from the popular that surpassed anything to

that were

accounts that journalists were going to have a hard go of

mummies were housed

in a

museum

it.

in Uriimchi. Where?

After

all,

these

They were more

ancient than the current population of the region, the Uyghur.

Who? And

what angle should one play? Although the mummies were certainly spectacular enough to fill a page or two of a colour supplement, they came from an area so far removed from the average reader's ken that it was difficult to find

common

short, the

ground, much less talk of revolutionary discoveries. In mummies looked very exciting; it just wasn't all that clear why they

shoulcl he so exciting.

What

lingered from the journalists' accounts

was

that the remains of

blond Europeans, possibly from northern Europe, had been found

tall,

in the

desert wastes of western China. Their earliest remains dated to about 2000

BC and they were found years ago.

to have survived in this region

And somehow

down

the Chinese historians had got

it all

to

about 1,700

wrong: China

had not developed its civilization in splendid isolation from the West but had been heavily influenced by prehistoric Europeans, so much so that it may have been Europeans who introduced both the wheel and bronze metallurgy, two of the primary technologies of civilization, to the Chinese. All of this was politically very sensitive and there were hints that the Chinese

As was

authorities were trying to keep these important discoveries under wraps. to

what ethnic or

linguistic label

might be assigned to the mummies,

it

suggested that they might have been the ancestors of the Tocharians.

Tocharians? The journalists tried their best: the Tocharians seemed to have

Rediscovery

What

did this mean? That prehistoric Irishmen had become lost in western China? But they also appeared to be tall blonds and looked more like Germans or Scandinavians. In short, remains of people closely related to those who read Western newspapers and magazines had somehow trekked their way across Eurasia to settle in the arid wastes between Qashqiir and Lopnur. Where? Before we move ahead we had better travel back a decade and recapture the excitement

been closely related to the Celts of western Europe.

of the recent rediscovery of the

New Friends Victor has visits to

in

made

mummies by

Victor

Main

Urumchi China

regular trips to

since 1981. Usually they have involved

archaeological sites in Xinjiang and Gansu, the far western portions

of the country, since his research speciality for the past two decades and

more has focused on early manuscripts found in that region. He had been to the Urumchi Museum many times in the past, but the summer of 1988 was different. As he walked through the old, familiar exhibition halls, he was totally unprepared for what he would encounter in a new gallery that had been opened at the end of the archaeological section. Parting the hanging curtains of the doorway, he entered another .world.

The room was

full

of

mummies!

Lifelike

mummies! These were not

wizened and eviscerated pharaohs wrapped

in

the

yards of dusty gauze that

one normally pictures when mummies are mentioned. Instead, they were everyday people dressed in their everyday clothes. Each one of the half-dozen bodies in the room, whether man,

woman

merely gone to sleep for a while and might

or child, looked as though sit

up

at

it

had

any moment and begin

whomever happened to be standing next to its glass case. Although he was supposed to be guiding an instructional tour

to talk to

for the

Smithsonian Institution, for the next three hours Victor totally forgot about all else in

the Uriimchi

of the

mummies

the world except the

Museum. For

in that dark, sequestered

the remainder of that day, he

community of ancient

souls in the

room

became

a

room of member

that, years later, he

would

return to visit again and again.

Although he

remembers as

best

swiftly

became

from that

'Charchan Man')

familiar with

initial

whom

all

of them, the

mummy Victor

encounter was the one (commonly

he came to

call

known

fondly 'Ur-David'. 'Ur' means

and 'David' refers to his second-eldest brother; the resemblance between the two gentlemen is startling. 'Ur-David' was sleeping peacefully in the far-right corner of the room (pis. I, II). He was lying on his back, his head propped up on a white pillow, his knees raised slightly, and his expressive hands - held together by a friendship bracelet twisted from 'primal' or 'earliest'

- placed

abdomen. 'Ur-David' was wearing trimmed with fine red pipmg and trousers of the same material. His knee-high socks of matted wool fibres (not quite felt) were as brightly coloured as a rainbow - with horizontal red and blue yarn

gently

upon

his

a brilliant reddish-purple woollen shirt

Rediscovery

Xinjiang The as

common IS

It

to be

political

found

in

designation of the far western territory of China

books

earlier

introduced the pinyin system of transcription, Sinkiang. The Territones', as they only

became

in

the 8th century ad, and the territory

Autonomous Region the province). The

often employ

miles, Isles,

one

1.

sixth of the entire

or,

in

1884. The main

who began

more properly known

more

47

as the

different ethnic

settling

Uyghur

groups

in

country of China.

which

we

600,000

sq,

neutral 'East Central Asia'

work. Xinjiang comprises a

Belgium, the Netherlands, France,

to spare bit

in this

or,

'Eastern Turkestan' or 'Chinese Turkestan' have also been

applied to this region, as has the politically will

is

of Xinjiang (altogether there are

names

Xinjiang

name means 'New

a province of the Chinese state

ethnic minority of the region are the Uyghurs, a Turkish people

there

is

than the Chinese spelling reform of 1956 which

Its

territory of over

area

Germany and

is

on the order of the

Italy

British

put together with room

from an American point of view, well over twice the

size of Texas

but a

smaller than Alaska.

Xinjiang, the

'New

stripes of flame red

most deliciously leather,

Territories', is

one of the most land-locked regions of the world.

and golden yellow, but occasionally alternating with the

faint blue imaginable.

Over the socks were boots of white

probably from a deer, that reached to his thighs. There was also a

mystifying leather thong wrapped around the middle finger of his

left

hand;

was secured by passing one end through a slot in the other end. Perhaps was a crop, signifying that he was a horserider, or merely a means of keeping the deceased from falling on his side.

it

it

Rediscovery

But

it

was the face that arrested Victor most. He

circleci

least a do/.en times, taking in

each tiny detail of this

whom

He wanted

he

felt

such an

affinity.

the glass case at

human

being with

to absorb everything

about

this

it was the face that kept drawing his attention. meaning of the spirals painted in ochre on his temples? Were they solar symbols? The horns of a sheep? Finally, he could do nothing but stare fixedly into those ancient eyes. They were blissfully closed, yet they spoke eloquently and powerfully. Then as his vision shifted to the top of the head he became conscious that the hair of 'Ur-David' was blond. His mind began to race wilcily, fillecl with a thousand insistent and unanswered

person from the past, yet

What was

the

questions:

What

is

this tall,

blond

man doing

here in the middle of Central

Asia where almost every person one sees today black hair? arrive?

How

did he get here?

What language

Where

did he

did he speak?

Victor looked at the label on the case. inside

his

remains dated to

It

said that the individual lying

Modern Standard Mandarin (MSM)) 1000 BC. He paced around the room, avidly

was from Charchan (Qiemo

and that

is much shorter and has come from? When did he

in

reading every single label - 1200 BC, 1800 BC, more individuals from 1000 BC.

The shock of

seeing these living fossils of Xinjiang

first led

him

to

question their authenticity. To be sure, at the turn of the century European explorers such as Sven Hedin of Sweden, Albert von Le Sir Aurel Stein

on behalf of Britain had

all

Coq

of

Germany and

recounted their cliscoveries of

desiccated bodies in their search for the artistic and literary relics of the

Buddhist shrines of Central Asia. In their magnificently illustrated accounts,

2,

A mummy of an

expeditions earlier

elderly this

man removed from

century

his coffin,

discovered by one of

ttie

Swedish

Rediscovery

they had even published drawings and photographs of these remains. In a

and time, these excited little interest and never prepared coming eye to eye with the prehistoric people of Xinjiang. With

different world

Victor for

many

questions

racing through his mind, he dashed off to recover his

still

Smithsonian charges.

Then

the Iceman

Came

On

the morning of 26 September 1991, Victor was sitting in the newspaper nook of the National Humanities Center in Research Triangle Park, North Carolina. When he picked up the New York Times, he was stunned to see the headline announcing the ciiscovery of the frozen bocly of a Copper Age

hunter-herder high Italy.

Except where

of the

left

in the it

Otztaler Alps on the border between Austria and

had been damaged by

a

pneumatic

chisel in the area

buttock and thigh during recovery, the body of this 'Iceman' was

almost perfectly preserved.

As Victor read more reports about this remarkable event, he learned that Otzi (as the Iceman had come to be called) was 5,300 years old, that he was equipped with carefully

sewn

a

bow, arrows, axe, and other implements, that he wore

and leather garments, and that he bore presumably

fur

therapeutic tattoos at strategic spots on his body. For Victor, however, what

was most

striking

was that

Otzi's icy grave

Austrian village, Pfaffenhofen,

in

which

was amazingly near to the was born and grew up.

his father

While he was devouring the news about Otzi, almost instantaneously

mind an ineluctable connection between the Copper Age Iceman of the Austro-Italian Alps and the Bronze Age denizens of the Tiiklimakan Desert whom he had first seen in 1988. Somehow or other, he sensed a distinct relationship between these ancient human beings. He became possessed by the clesire to determine whether this were indeed true and, if so, in what way the Iceman and the desert people were linked. Within half an hour of learning of the discovery of Otzi, he resolved to organize a research project on the mummies of the Tarim Basin (throughout this book we will frequently use 'Tarim' as shorthand for both the Tarim Basin proper and the neighbouring regions to its north and east). He was aware that DNA tests would be carried out on Otzi. Consequently, he thought that it would be both simple and precise to do the same for the Taklimakanians. Once the DNA of the Iceman and the desert people were compared, it should be easy to decide if they were of the same stock. That very afternoon, he began to write letters in preparation for a scientific expedition to the Uyghur region of China. there formed in his

The Quest Begins There were three main aspects to the flurry of activity into which Victor immersed himself during the coming months: locating reliable scientific

11

Rediscovery

expertise, raising the necessary funds,

Chinese government to engage tasks as equally challenging. in all three areas of

The

first

He had

Much

He

viewed

met with

to his surprise, he

all

swift success

known

to solicit the guidance

of the outstanding

and help of

a qualified geneticist.

work on population

genetics of the

distinguished Stanford scholar. Professor Luigi Luca Cavalli-Sforza.

unprepared for the approachability of ideas.

Not only

happy

three

endeavour.

was

step

long

and obtaining the permission of the

in cooperative research.

this world-class scientist

He was

and man-of-

did Cavalli-Sforza enthusiastically declare that he would be

to participate personally in the research project, he also suggested

several possible funding agencies.

he had already

come

With one telephone

close to killing

call to Cavalli-Sforza,

two of the three birds required

to set

the enterprise in motion.

The

Alfred

research

P.

Sloan Foundation

on the mitochondrial

made

a substantia! grant to

undertake

DNA of the Tarim Basin mummies during the

1993 season and they generously doubled the size of their grant following two years. Funds were also obtained from Victor's

own

in the

University

of Pennsylvania, the Henry Luce Foundation, and numerous individual

who

to do their part to Tarim Basin. The signals he was receiving from various sources in China were sufficiently encouraging to assume that there was at least a chance that he would be able to proceed. In particular, Wang Binghua, Director of the Institute of Archaeology in Uriimchi, exerted himself on behalf of the project. And although there were bureaucratic hiccups enough along the

citizens

are interested in world history

clarify the mysteries

surrounding the

and wish

mummies

of the

way, permission to go ahead was obtained from the Chinese authorities at the eleventh hour.

By

this

time Cavalli-Sforza had had to abandon the idea

of travelling to Central Asia himself because of a heart condition, so he

provided a superb substitute

in

the person of Paolo Francalacci of the

University of Sassari in Sardinia. After meeting up in Beijing, Paolo and Victor flew to Uriimchi on 22 June. Lengthy negotiations were required to

main concerns of the Chinese. It was agreed that their geneticists and archaeologists would be directly involved in all phases of the work, that they would be listed as joint authors of any significant publications emanating from the project, and that technology transfer would occur in the form of training in the new, advanced analytical techniques the American allay the

side

would be

samples was

some of

using.

It

cannot be stressed too heavily that taking tissue

a highly sensitive matter.

their scholars to

of study, to raise

work, and to seek

The Chinese

go abroad for brief

visits

side

was

also eager for

or for extended periods

money from American foundations to support their assistance in building a special museum for the preser-

vation of the scores of corpses that had been excavated and will continue to be excavated in the

Uyghur

region. Victor readily acceded to the mission

of helping the Chinese build what might affectionately be called the

'Mummy Museum'.

Rediscovery

Bronze Age People

On

15 June,

Wang

in

Poplar Gully

Binghiia, Paolo and Victor travelled by car 490

(304 miles) eastwards across the searing desert to Qumiil.

Known

as

km

Khamil

in Mongolian, the name of this important Silk Roaci town is transcribed in Modern Standard Mandarin as Hami. It is famous for its succulent melons

amounts of cotton are also town. Nearby was one of the most

suffused with fragrance and sweetness. Large

grown

in irrigated fields

important

Approximately Qaradovva a

little

surrounding the

sites for the project. 1

km

(0.6 miles) to the

(MSM Wupu)

is

northwest of the oasis village of

a terrace of pebble-strewn desert that lies across

which separates

valley called 'Poplar Gully'

from the town. From

it

mound of reddish known as Qizilchoqa,

the terrace one can see at the town's edge a distinctive

earth which gives

its

name

to the area. In Uyghur,

which means 'Red Hillock'. been fenced Institute of

in for

A

tract of

protection and

Archaeology which

is

is

it is

about 5,000

m

sq.

(53,820 sq.

responsible for

its

was

his

('Celestial

its

Mountains', Tian Shan

hunch, based on the

has

ongoing excavation.

The Qizilchoqa cemetery was discovered by Wang Binghua following the stream in Poplar Gully from

Tangri Tagh

ft)

under the jurisdiction of the Xinjiang

many

in

1978 by

glacial source high in the in

MSM)

to the north.

It

years he had spent as a practising

archaeologist in the region, that ancient peoples would have located their settlements along the stream because

it

of water. As he followed the stream bed, as to

provided a relatively reliable source

Wang queried

the local inhabitants

whether they had come across any old broken pots, wooden

and so

forth.

An

old

man named

Imit

artifacts

came forward with some ancient 3. Sites

mentioned

in

the introduction.

Qaradowa Urumchi_^

"Subeshi Turpan

Gobi Desert

• ANCIENT SITE

t 1

50

-S

300 km

H 200

300 miles

13

REDISCOVERY

Wang was

objects that were exactly what

visited Qizilchoqa in 1993, Imit

During the were dug and, with a

first year,

in 1991,

fairly large

looking

was the guard

Victor initially

cemetery

29 graves were excavated. In 1986, 82 more graves

two additional graves were excavated

in

cooperation

team of Japanese archaeologists. Judging from the

and the number of obvious depressions

the tract

When

for.

at the

in the surface

size of

of the earth

(which indicate the existence of a grave beneath), there were probably at least

another hundred graves remaining to be dug at Qizilchoqa. But by June

when Victor returned to Qizilchoqa to make a film about the mummies Nova (WGBH-Boston) and Channel 4 (London), the entire cemetery had been hurriedly rummaged. Few graves remained untouched and litter from obviously hasty 'excavations' was strewn across the pebbly desert 1997,

for

surface. In

to

make

June 1998, Victor returned to Qizilchoqa once again,

a film for the Discovery

year's shallow scavenging still

this

time

Channel. The debris from the previous

had been cleared up and

it

appeared that

it

might

be possible someday to excavate the remaining graves properly

The Qizilchoqa

graves have been dated to approximately 1200

BC by

Chinese archaeologists because of the presence of bronze objects, the of painted pottery, and five radiocarbon dates. This puts the

style

site at a

key

point in the development of Chinese civilization, the height of the Shang

dynasty

It is

dated to essentially the same

moment as the

introduction of the

chariot and the rise of writing.

Aside from numerous well-preserved corpses, a wealth of artifacts was recovered from the graves. These were not luxury goods but simple items for

use in daily

(combs, mirrors, needles, bowls, pots, hooks, bridles,

life

spindles, bread, etc.).

Among

bells,

those that struck Victor most forcefully was

from an unexcavated grave. It was similar to another partial wheel that had been unearthed earlier from one of the other graves, which was kept in the Qumul Museum. These wheels, termed 'tripartite disc wheels', are of a peculiar construction which a part of a wheel that he spotted protruding

joins three thick planks (rounded at the edges) with tightly fitting dowels.

West Asia and throughout Europe a thousand years earlier. The spoked chariot wheel has not yet been found at such an early date in Xinjiang, but it has been found at a slightly earlier

The same

type of disc wheel

is

found

in

Nomhong, in Qinghai (Kokonor) province just to the east. The Qizilchoqa graves are relatively simple in their construction. About

site,

2

m

(6 ft

7

in)

deep, they are lined with large, unbaked bricks around the

The

were just big enough for the occupant(s) them on mats, lying on their sides or backs of the grave who were placed with their knees bent upwards. Above the buried individuals, about halfway down in the pit, was a ledge around all four sides. A layer of mats and reeds lower part of their sides.

pits

in

was placed here

to prevent the sandy soil

lined burial chamber. large,

rough-hewn

of sand.

14

The

from

falling

down

into the brick-

opening of the grave was covered by a row of

logs strewn with

mats and reeds over which was a

layer

4.

Mummified corpse on

the floor

of a bnck-lined tomb at Qizilctioqa

5.

Victor

Mair

witti a recently

exhumed corpse before

it

at Qizilchoqa

was returned

to

its

grave

REDISCOVERY

The most

impressive aspect of the Qizilchoqa graves

Due many of

is,

of course, the

unique combination of cHmatic

ancient corpses themselves.

to the

conditions

the bodies have been preserved intact

in the area,

through a process of natural desiccation. The corpses are marvellously coloured and patterned woollen fabrics,

and sometimes leather

coats.

They

felt

fully clothed in

and leather boots,

are clearly of Caucasoid/Europoid

extraction (long noses, deep-set eyes, blondish, light-brown or red hair,

and so

forth).

The men

are fully bearded

women

and the

have long, braided

hair.

In

and

company with about

his

a

dozen

crew spent the entire day

local archaeologists

at

and diggers, Victor

Qizilchoqa exhuming corpses from

had previously been excavated but, after preliminary examination and recovery of important artifacts, had been reinterred because of a lack of adequate storage facilities in Qumul or in Uriimchi. As one grave after another was opened, Victor was awed by the sight of bodies that had been graves that

lying there for over three millennia.

Paolo, wearing a face

surgical scalpels to

mask and rubber

own modern

the corpses with his

(and

gloves to avoid contamination of

much more

remove small samples of

tissue

potent)

DNA,

used

from unexposed areas

of the bodies (usually the inner thighs or underarms).

He

also took

fragments of a few bones (parts of ribs that were relatively easy to break

and teeth which preserve the tissue and skin.

The samples were placed

DNA

off)

perhaps even better than do the muscle

in collection jars, sealed

and

labelled.

While

Paolo was doing his work, Victor made a photographic and written record of the tissue collection. Altogether, they took double or triple samples from six corpses at

Qumul.

That evening the team feasted on an elaborate banquet held for them in the guest house where they were staying. Paolo was wearing his 'Ancient

DNA t-shirt and beaming broadly. The Trio From Charchan Although the 1993 expedition did not

Mandarin), which

in

lies

travel to the

town of Charchan (Qiemo

towards the eastern end of the southern branch of

the Silk Road, they were able to see half-a-dozen

important

site

of

Zaghunluq

mummies from

the very

that lay in the desert nearby Three of these

were the by-now world-famous 'Sunday supplement' family comprising 'UrDavid', the

woman who

is

thought to have been

his wife,

and the

with blue bonnet and blue stones over his eyes that was buried grave close by theirs. in the district /.

The temples of

'Ur-David'

were

designs

years.

16

at Korla

mummies from Zaghunluq were

which the 1993 team did

young woman of about 20 years old and

a

ochre.

museum

three

visit.

child

separate

While

kept

there,

they took tissue samples from two of the corpses that were best preserved,

ornamented with in

The other

little

in a

a

little

boy of approximately iVi

K

//.

'Charchan Man',

Victor Mair's 'Ur-David', reclines in tlie

UrCimchi

Museum.

III.

The

woman

from

Zaghunluq who

accompanied 'Ur-David'.

The 'Beauty of Kroran

IV.

with the wheat basket

and

(on the right)

winnowing the

left)

basl\ n \

^

b g d h

;>

z

\

/

h

-il

u

M

)»

;»

;

%

y

^ \A i

^

m

73

»

CD JO

n s

•^

J

\t

Q9

/?

j:^

;^.

> h

.Ol

J4

o
r

t

ih

T)

cna

major

theirs

over time and here are

of Estrangelo

(Es),

type employed

the

in

shown examples (Pa) and the

Palmyran

Manichean Sogdian

(MaSo) documents. The column at left

far

gives the transliteration of these

alphabets.

BC:

late as

to the lower

initially

proposed span, and perhaps more

burials in the flexed position that brick-lined pits

dated to

800-550 BC which could argue that the culture was closer

end of the

assigned to the Tort Erik (Sidaogou) culture.

63 The

initially

but a recent radiocarbon date from a textile places the

The

earliest graves

easily

were single

had been inserted into the base of

a pit

The

later

which was then covered with large logs smoothed on one

side.

from Qizilchoqa

graves comprised double burials in brick-lined pits; above was a second

comprised

chamber whose base was sealed by reed mats that protected the burial chamber from falling sand and the top of the chamber was covered with timbers. The two chambers were dug down to a depth of about 2 m (6 ft 7

layers of

buried remains and

covenngs

in).

This cemetery also yielded

we

mummies and

have more to say

extraordinarily well-preserved

and other organic remains, usually Euphrates poplar. The mummified remains here are of a Caucasoid type with light hair and large noses. The hands, arms and upper backs reveal

clothes (of which

will

later)

traces of tattoos, a trait again encountered slightly later

burials in the Altai Mountains.

tombs

142

Among

the

wooden

among the Pazyryk

objects found in the

there are also traces of wheeled vehicles such as three-piece disc wheels

The Testimony of the hoe

64. This disc

wheel

from Qizilchoqa reveals the tripartite

construction techniques

encountered across Eurasia

and hubs. There is no evidence to suggest that wheeled vehicles were independently developed in China and the general range of the earliest vehicles in Eurasia runs

from Southwest Asia northwards across the

Caucasus and then spans Eurasia from the Netherlands to the Yenisei Afanasevo culture

Andronovo

or,

cultures).

(the

more certainly, its successors, the Okunevo and The large disc wheels mark the initial horizon of

wheeled vehicles which persist alongside a

later

horizon of spoked-wheeled

2000-1600 BC), particularly the chariot. A number of other Age burials from Pazyryk, e.g. small wooden tables and trays, and a bronze mirror. Economic evidence comprises faunal remains from cattle, sheep, horse, donkey and camel, while flat cakes of millet and barley have been recovered. Timber remains of what is thought vehicles

(c".

features are reminiscent of the later Iron

some 90 cm (3 ft) long have also been recovered and among wood, bone and leather remains is harnessing equipment. As we will later see, Qizilchoqa was one of the cemeteries sampled for ancient DNA.

to be a plough

the

Cultures of the Gushi (Jushi)

Some 300 km

(186 miles) to the west of

Qumul

lie sites in

the vicinity of

Turpan oasis that have been assigned to the Ayding Lake (Aidinghu) culture. The lake itself occupies the lowest point in the Turpan region (at 156 m (512 ft) below sea level it is the lowest spot on earth after the Dead Sea). According to accounts of the historical period, this was later the territory of the Gushi, a people who 'lived in tents, followed the grasses and waters, and had considerable knowledge of agriculture. They owned cattle. the

143

The Testimony of the Hoe

They were proficient with bows and They were also noted for harassing travellers moving northwards along the Silk Road from Kroriin, and the territories of the Ciiishi and the kingdom of Kroriin were linked in the accoimt of Zhang Qian, presumably horses, camels, sheep and goats. arrows.'

because both were under the control of the Xiongnu.

60 (a

In the years around Chinese and was subsequently known as Jushi different transcription of the same name). IMl,

Gushi

fell

to the

There was variation among the different cemeteries of the region with some burials on wooden platforms or under timber beams and reed mats; other cemeteries revealed burials covered by small stone cairns. Burial was in the

extended position, head to the west, and might occur singly or as

a

double grave. Black-on-red painted handmade ware accompanied the burials

and there were also some wooden figures similar to those found Qawrighul culture. The Ayding culture itself is dated to c. 1400-700 part of the cemetery at Subeshi,

Ayding culture, must run

commonly regarded

later into the Iron

Age

the

but

to be related to the BC;. The portion of some 40 low stone

300

c.

in BC;

the Subeshi cemetery excavated in 1992 consisted of

on wooden beds. There were special entrances shafts and goods from later burials included iron.

cairns covering burials set into

some of

the burial

Evidence of diet was graphically preserved

in

such forms as a bowl

with millet and another containing the coccyx of a goat.

Among the

were seven very well-preserved

filled

burials

mummies

and the partial remains of some dozen more.

One

men wore

of the

leggings and

woollen underpants and was accompaniecl by a

bow

in a leather

case and a set of

arrows with wooden, bone, horn and iron tips,

apparently fashioned for different

intended prey

ment that

is

(ill.

66).

The archery equip-

reminiscent of that of the Scythians

we have

seen earlier. There was also the

mummy

superbly preserved

clad in a fur-lined cloak

of a

woman

and multi-coloured

woollen skirt which dated to about the 4th century BC. As the Subeshi cemetery provides us with our 'three witches',

have more to say about

it

in

we

will

the next chapters.

Settlements in Yengidala

About 200 km (124

V

in

miles) to the southwest

the vicinit\ of Lake Baghrash (Mongolian

Bostnur) are various

sites

of the Yengidala

(Xintala) culture which dates 65.

144

Double

burial

from a

pit at

Subeshi

c.

170O-1400 BC.

Here we have evidence of settlement mounds.

^ The Testimony of the hoe

one (492

at

Ycngidala 5 in

tt)

m

(16

and about 150

5 in) higli

ft

ni

diameter. Situated within the district of

Qarashahar (Yanqi), the Yengidala group affords us our earhest picture of sedentary communities in the vicinity of this important oasis centre, one of

the four garrison towns of the liistoncal period.

There was evidence for wheat, millet and stone sickles and querns; domestic animal hones were also recovered. The finds mclude painted potterv decorated with geometric and /oomorphic (bird or fish) motifs as well as incised

ware which

has been likened to that of the Andronovo culture of the steppe. Surface finds of bronze implements

such as

a

socketed axe have been retrieved.

Unfortunately burials of

this region

remain

unknown although some

disturbed

human

remains are reported from Chokkur (Quhui).

Kucha A

farther 250

km

(155 miles)

(Haladun) culture which

was the Qaradong

known almost

so far

is

entirely from surface finds except for a small

—

excavation at the settlement of Qarad()ng itself.

Here, near the major oasis town of

^

Kucha, archaeologists uncovered timber-postbuilt houses, traces of millet, sickles, grinding stones, a distinctive red-on-white

painted

pottery and about a hundred bone artifacts

(arrowheads, awls,

'%

198

The

Mummies Themselves

]]4.

A

I'/.i'-year-old

baby from Yanghe village in the

Gorge (near

Tuyuq Subeshi).

199

The Mummies Themselves

small bag were crushed plant remains, thought to be ephedra. These

Why

have been two of the medicines he was treating himself with. incisions in the chest region?

the autopsies of other

prompted the rather

We

may the

have no certain answer but, judging from

mummies, some form

of pulmonary disease

may have

drastic (and apparently unsuccessful) operation.

Related to the people and culture of Subeshi was a

little boy of 1 Vi years, Yanghe village in Pichan (Shanshan) County (ill. date from around the 3rd century BC.

whose body was found 114).

He

is

said to

in

Imammusakazim This

site

period

c.

which

lies

within the region of Khotan and which dates to the

AD 420-589 yielded two Buddhist

burials of

women. The

older of

two was about 30-35 years old, of Mongoloid extraction, and had brown hair with a black braid woven in. She wore a white patterned silk shroud. The younger individual was a girl of 10-12 years old who wore a red patterned silk shroud. Both women had silk bands to tie their mouths shut. the

Lopnur Region: Grave 7 Grave 7 was located 7.5

km

(4.7 miles)

Here Folke the best preserved of which was

southwest of Grave

5.

Bergman discovered three or four graves, numbered 7A. The coffin was half of a hollowed-out tree-trunk which was sealed by two long boards and mounted on four wooden legs. The mummified remains of 'an elderly, stately gentleman with a small white beard, a thin moustache and white hair' had apparently been tipped out by

plunderers (Bergman saw the hand of the ubiquitous Ordek

in this).

The

had been stopped by plugs of wool covered with silk, elaborate affair than the simple woollen stoppers found red a far more in the earlier cemeteries. Bergman was impressed with the non-Mongolian nostrils of the 'high nose'

features of the deceased

belonged to an Indian.

Silk

and suspected that the mummified remains was employed

in the clothing as well as the nose-

Bergman workshop about the 3rd century AD. Somewhat similar material, including woollen and silk nose-plugs, were found in the other graves and a series of low poles were found at the top of the mound in which the graves had been inserted. Among the graves excavated by Sven Hedin in the Lopnur region was Grave 35. Attracted by a standing tamarisk pole, Hedin's men dug down to plugs and there was a particularly elaborate collar which, conjectured, just might have been produced in a Persian

uncover a wooden coffin fashioned from half of a hollowed-out

log.

Upon

removal of the two boards that formed the lid, they found the remains of a young woman, 1.6 m (5 ft 3 in) in height. Her 'turban' and clothes were of silk and she was accompanied by a wooden drinking cup, a wooden foodtray, a sheep's skeleton and twigs of ephedra. Bergman conjectured that the

tomb might date

200

to the first centuries of our era.

The

At Yarliq-kol, Bergman numbered as Grave 10 a burial that he

found near the southern bank of

Qum-darya. Here

a coffin, a hol-

lowed-out poplar trunk some 2 (6

7

ft

was found

long,

in)

m

dis-

lodged from a yardang, the ridges

formed from wind action across

The

the region.

coffin's interior

had apparently been lined with contained the

thick

felt. It

ified

body of an old man Behind the

(4 ft 10 in)).

mumm(1.48

left

m

ear his

grey hair had been twisted into a

knot. His teeth were generally in

poor shape - the incisors were well

worn and

the lower molars were

missing- although he upper

in his

still

had

13

jaw.

Among the grave goods

were the

charred bones of a sheep's foreleg that

of

had been

felt

coverecl by a piece

to imitate the flesh,

fish vertebrae.

Most unusual was

that below his right hip his coat

was

and two

and under

a felt doll, apparently

with feminine features painted on

and hair

in the

Bergman

pubic region, which

interpreted as 'a symbol

of a feminine companion'.

The only burial

silk that

measured 60.5 cm

wide, 10

cm

standard

in

iately it

/

/

5,

The Yingpan

clue to the date of the

was that the coat contained (4 in)

the

(2 ft)

wider than the

Han and immed-

post-Han period, thus setting

at the

end of the

floruit of the

Kroranian kingdom.

mummy.

Yingpan Finally,

we should add here notice of the most recent mummy The Chinese press announced in January 1998 that

discovereci.

mummy

had been uncovered

that has yielded over 30

in the

tombs and

Kroran region is

at

to be a

new

Yingpan, a cemetery

situated to the west of Lopnur.

The

Mummies Themselves

The

/ 1

Mummies Themselves

6 Mask of the

man

buried at Ytngpan.

remains belonged to a male, aged 25-30 years, 1

1

in) tall

and had been

laid

out

in a

wooden

who

stood some

red woollen robe with yellow embroidered designs

wound around goods included

was attached bow and arrows, a

his waist; to the sash a necklace, a

a

1.8

m

(5 ft

He was wrapped

in a

and

was

casket.

a satin sash

perfume

glass

satchel.

cup and

a

Grave

wooden

comb. He has been dated

to the Han-Jin period, i.e. probably about the 2nd—3rd centuries AD. The most striking feature of the new mummy is to be found on his head. Resting on a satin pillow (we are obviously talking about someone of considerable status, Chinese archaeologists suspect a rich merchant - Sogdian?), the deceased wears a mask over his face. The

newspaper reports indicate that the mask

202

is

made

of

nici

which can mean

The

Mummies Themselves

ramie, and/or sisal. The wet 'hemp' was apphed in layers wooden mould and allowed to dry; the mould was then removed and the hard cloth mask painted white. The forehead of the mask was

hemp,

jute, flax,

over a

a band of gold foil. This is not the only discovery of a masked individual, because in 1996 Wang Bo recovered from Zaghunluq a mummified head with a tightly fitting leather mask. The custom of masks, of course, may well derive from China, but we should not forget that one of the characteristics attributed to the population of Kucha was that dancers

adorned with

wore masks.

Pazyryk and Xinjiang

The spectacular remains from East Central Asia are not the only mummies known in Asia. The excavations of the royal tombs at Pazyryk in the Altai Mountains, some 960 km (600 miles) to the north of Turpan, also uncovered a series of mummified remains, attributed variously to Scythians, Yuezhi, Wusun, Xiongnu and the Arimaspians of Aristeas, to name but a few candidates. They date from about 300-250 BC and were found preserved in large timber chambers sealed by stone cairns. Unlike the Xinjiang mummies, preparation of the bodies for survival was deliberate, while the means by which

it

was achieved was accidental.

We

can partly thank the

activities of

ancient grave robbers for the preservation of the organic remains in the

tombs because they breached the chambers in their lust to recover anything of value. This permitted the tombs to flood with water which, in the high Altai Mountains, froze: the stone cairns served to reflect sunlight and

1 1

7 Plundered

antiquity, ttie

::0.

o

in

tombs

remains of their occupants.

^;.,;(..!>-^:.;i.K.;C:»

;•

U)

"'.

1

m

203

The Mummies Themselves

preserve the ground under

them as a primitive box of frozen ground. The excavation of the mummies, which required heating water and slowly thawing out the remains,

ice

recovered

artificially

human

prepared

which evisceration and removal of brains had been practised and the body

remains

in

cavities stuffed w'ith straw before being

sewn up with horsehair or tendon sutures. Like the mummies of the Tarim Basin, hair had been artificially augmented: a false beard was prepared one of the mummies of Mongoloid

Pazyryk

at

for

extraction,

who

apparently

required

socially what his genetic background was unable to provide. Tattooing was also

extremely important and the skin of one of the

mummies

provided an artistic tableau

of mythical and real animals executed

in

the

'animal style' of the Eurasian steppe.

That there

is

a

possible ethnic relationship

between the Pazyryk mummies and those of the Tarim region is a fair inference although we cannot be too precise as to which ethnic groups (probably

may have been

Iranian)

mummification found

origins of the

presumption that

helci

what of the Pazyryk? The

involved. But at

long ago by Grafton Elliot Smith

mummification was inwnted

solely in

diffused from there to wherever else

Egypt and

we encounter

it is

hardly tenable on a worldwide scale (for example,

Peruvian

mummies

are the oldest in the world),

and

connections between ancient Egypt and the Altai

Mountains

are very difficult to contemplate.

As

to the

origins of artificial mummification, there are (at least)

two approaches. Archaeologists

working

in

the

incredibly

rich

Minusinsk Basin region have suggested that mummification evolved slowly and out of the need to preserve

bodies for longer periods as the construction of

/

18.

Tattoos

man from

on a

Pazyryk

we

find the features

tombs became more elaborate and required more time. This, they suggest, can be seen in the Tagar culture, a contemporary of the Pazyryk tombs. Here of the deceased preserved in clay masks. The preparation

of the burials required the removal

t)f

the soft tissue from the skeleton, the

building of an interior mannikin with grass, the sculpting of a facial

204

mask

The

I

from

Mummies Themselves

'J

clay,

the fastening of the bones together to

form

a 'body' to

which the

head was then attached, and then the painting and dressing of the body.

While elaborate, artificial

this

is

a far cry

mummification:

from the

result achieved in natural or

7

79.

The preparation

of a mannikln from the Minusinsk Basin involved de fleshing the

it

provides an effigy, not a fully preserved body. skeleton, reconstructing

Those who regard mummification as a one-of-a-kind invention might ponder how these effigies are remarkably similar to the Chinchorro mummies (effigies) of southern Peru and northern Chile that date to c. 6000-2000

the figure with plant

remains and then dressing

it.

BC:.

205

The

120.

Mummies Themselves

The preparation

of a mannikin

in Peru,

which long antedates those found

in Asia.

The second approach

is

to return to Hgypt, not because the techniques of

mummification diffused from there but rather because

it

might well provide

one piece of the puzzle. Like the Tarim Basin, the arid sands of Kgypt created suitable conditions for natural

we do

tombs of

the dynastic period

buried

simple pits where their flesh has been

in

moisture-depriving sands. This

206

mummification and prior

may be

find well-preserved in direct

to the elaborate

human remains contact with the

satisfactory as long as you are content

The

to bury your

dead

Mummies Themselves

shallow pits inserted into the desert sands. But as

in

complex and hierarchical, there was a social more elaborate tombs, large brick chambers (ultimately pyramids) which separated the corpse from the dehydrating effects of sand and dry air. In some cases, we have mummies that were naturally desiccated and wrapped up without evisceration but, in general, society

became

increasingly

drive to prepare increasingly

if

you wanted to

retain the integrity of the deceased, artificial techniques

were often required and

it

was these which

resulted in our traditional concept

of mummification.

While we have no direct evidence at least conjecture that in East

The

for such a transition in Asia,

one can

Central Asia a similar process evolved.

naturally preserved remains that one

would

regularly encounter in the

Tarim Basin had existed for nearly 2,000 years before we find evidence for deliberate mummification in either Han China or among the Iron Age steppe peoples.

We

have seen that pastoralists were

in

continuous contact with the

populations of the Tarim Basin and could not have failed to observe the remarkable preservation of the local dead. Indeed, a distinction between

some

pastoral groups

resident in the

who wandered north of the Tangri Tagh and those itself may be unwarranted. In this way, popu-

Tarim Basin

lations already imbuecl with either an ancestor cult, such as obtained in

ancient China, or with a preoccupation with the integrity of the

form or clay

masks

to achieve

The The

first

began

in the

3rd millennium BC)

mummification of

Silence of the

their

own

human

steppelands where

facial features (for instance the inhabitants of the

may have

stimulated others

dead.

Mummies on their dissecting examined the integrity of her

forensic scientists have laid out the 'Beauty of Kroran'

table.

They have weighed her

skin, retrieved nits

internal organs,

from her body and tested her with

scientific techniques.

But she

still lies

a battery of other

there, silent as ever

and might

well have a tag about her big toe, inscribed 'Jane Doe'. She cannot

just as tell

us

where her ancestors originated, what language she spoke, what her settlement was like, nor who her descendants are. She is no longer articulate,

mean that she has ceased to inform us. She and the other mummies all come to us clothed and, while we may not find a designer label, we may still recover quite a few clues to their identity from their garments. but that does not

207

CHAPTER

The

Tarim and Turpan basins

textiles in the ancient world.

to

offer

SIX

one of the

hirgest collections of

Their study has been primarily confined

Western scholars such as Vivi Sylwan who made detailed

examinations of material uncovered

in the earlier

The more

excavations of Sven Hedin,

Folke Bergman and Aurcl Stein. enormous accumulations of new material in much better dated contexts and some useful descriptive work by Chinese specialists, but more detailed analysis has been limited to the work of Irene Good of the University of recent discoveries have seen

Pennsylvania and Hiizabeth Barber of Occidental College, California,

who

were given partial access to some of the material (which provided a basis for

book on the Xinjiang mummies and their textiles). Chinese archaeologists have been somewhat loath to surrender their material to foreigners for analysis. From their viewpoint, the archaeological treasures of China have been pillaged quite enough by Westerners and no one wants to defer to 'foreign experts' to interpret one's own heritage, especially when it Barber's fine

concerns is

textiles, a field in

understandable,

it is

which China has traditionally excelled. While

this

frustrating to find the scientific reports which could

place East Central Asia in the larger picture of the development of textiles in

Eurasia so slow

in

coming. Textile production

cultural arsenal of a people: the materials

is

nor a mere adjunct to the

employed and the technology

production can be used to trace the course of cultural influences, possibly even migrations, while the decorative patterns employed in textiles or the cut of the material has long been known to be one of the

involved in

more

its

sensitive expressions of a culture's self-identity

A

Brief History of

it

is

uncertain

Twine

when humans began

to clothe themselves although the

circumstantial evidence of the freezing temperatures of the Ice

with the discovery of

208

flint

Age coupled

scrapers for cleaning hides would suggest that

Tartans

among

clothing (skin wraps at least) existed

the Neanderthals,

in

the Tarim

even

i.e.

modern humans in Europe. By about what are interpreted as bone and antler

before the appearance ot anatomically

20,000 years ago

we

find in France

needles that indicate the sewing of animal skins or textiles manufactured

from plant and

woody

tree fibres, e.g. bast, a

from the linden

fibre derived

and willow. Even earlier we have some evidence for nets about 26,000 years ago (from the Czech Republic) and a twisted cord of three-plied strands of spun vegetable

(bast?) fibre that dates to

about 15,000 BC.

A

massive

amount

of circumstantial evidence such as pendants, bead necklaces, etc., indicates the widespread use of string or cord both during and after the last glaciation.

continue

Flint or other stone scrapers

and remind us that wrapping oneself

abundance through the Neolithic

in

in the skins

of animals has always

been an option. Textile finds across Eurasia from the earlier part of the Neolithic are almost uniformly of plant fibres, particularly flax and hemp.

was probably the most widespread plant employed for clothing and, while we have no evidence for its domestication until c. 5000 BC, we do have evidence for linen textiles that precede such a date. This suggests either that people had domesticated the plant earlier Flax, which provides linen,

than present evidence allows or that they used wild flax

They

also exploited

hemp

production of rope and

but, being coarser,

Hemp

sails.

in linen

manufacture.

was more often used

in the

appears with reference to textiles by

about the 5th millennium BC on Neolithic in

it

sites

of the Yangshao culture

northern China but does not appear certainly as a material of texti)e

production also be

West

in the

employed

in the

until

about the

production of

millennium BC. Nettles could

1st

textile fibres

although solid evidence

no earlier than the 1st millennium BC. what is most remarkable is that sheep and goat were already domesticated some 10,000 years ago, possibly earlier. In chronological terms, it means that across Eurasia for a period of several thousand years shepherds, for their use generally dates to

Now

themselves dressed

Not ruped

really

of plant fibres, stood about tending their flocks

in clothes

of sheep and goats.

Were they missing something obvious?

Although our image of a sheep may be that of

(or for those

who occasionally

a

woolly quad-

have to evict sheep from archaeological

excavations, an incredibly stupid woolly quadruped), early Neolithic sheep

were not at

all

The coat on woven

woolly but covered with short, thick, coarse hairs or kemps.

a sheep thus

looked about as inviting for the production of

textiles as that of a deer.

The replacement

of the coarse short

kemps

by a woolly coat was a process that apparently began about 5000 BC and

woolly sheep did not become widespread earliest

appearance of woolly sheep

and neighbouring to the shift

territories

is

in

Eurasia until

3500 BC. The

generally thought to have been in Iran

from which

it

diffused east and west. In addition

from kemps to wool, there was also a

would eventually

c.

loss of

pigmentation which

result in purely white sheep (and wool). This provides a

useful chronological

marker

in the history

of dress: before

c.

can normally expect and do encounter skins and plant-based

3500 BC we fibres,

and

209

Tartans

in

the Tarim

after this period vvc find increasingly greater quantities of

The

woollen

textiles.

archaeologist works to a three-age system: Stone, Bronze and Iron; for

the history of textiles

it is

the

same although

the periods

do not

correlate

with those of the archaeologist: Skin (?-5000 BC), Linen (6000-3500 BC)

and Wool (3500 BC; onwards). The production of woollen fabric may he achieved by a variety of methods. Technologically, the simplest way is by felting. This only requires that one take the animal fibres, such as wool, and lay them out in the shape desired.

Then you wet

the fibres with

warm

the ability of the scaly fibres to tangle with

water or whey, which enhances

one another. Then apply pressure:

repeated beatings, crushings, kneadings, whatever fibres together intt) a ct)herent shape.

If

takes to fuse the woollen

it

decoration

small pieces of coloured material and subject

it

is

required, you can add

to the

same rough treatment

What you get is a textile that is water resistant and well insulated. Felt had many uses such as rugs, wall-coverings and tents, and hence it is very much associated with the pastoral peoples of the Eurasian until

it

also has adhered.

steppelands. hi terms of clothing, facture of protective headgear

When one considers inevitability

was

about the whole thing. Our

Beycesultan

in

typically

the processes by which

2600 BC, the date assigned to at

felt

employed

a felt

felt is

earliest

made, there

would have only been a some of their wool

it

in sheets to

rugs, etc. Felt

evidence dates to about

Ml)

I?/-;-'.-!^

into a

was an almost

inevitable

^7T?y

acci-

form that could occasionally be

consequence of keeping woolly sheep.

122. Plan of

rv'-'-^^

their

be fashioned into coverings such as hats, caps, small

Plan of plain weave.

K--':'^

a striking

some of matter of time before they would

gathered

l3

is

Anatolia. Given the fact that sheep would have to be

dentally trample

727

manu-

rug discovered on the floor of a shrine

occasionally penned up and that they would naturally shed coats,

in the

and stockings.

(^•..'•T^

twill.

Tartans

The main weaver to

alternative to felting

is

weaving on

a

loom. This requires the

of combinations, would run

in a variety

Among the enormous

the Tarim

forming the warp of the

set the threads into a fixed grid, thus

loom. Over and under these threads,

in

weaving patterns, ii'cft. two that we must pay particular attention to. The simplest form of weaving, plcun weave, or tabby, requires one only to pass the thread of the weft over and under each strand of the warp. Such a pattern is as old as the earliest evidence for the weaving of (plant-made) textiles around 7000 BC. A more complicated pattern can he obtained in the

the threads of the

varieties of

there are

production of

This involves,

twill.

itleally (there are

variations

on the theme),

running the weft over two and under two of the strands of the warp but alternating the pattern in each in a

row so that

it is

diagonal rib or alignment of the weave.

different

off-set

The

one

fabric

is

place. This results

not simply visually

from plain weave but because the weft hops over two

rows of the warp,

it

yields a denser fabric. Twill

The

associated with wool and not with linen.

is

(or

more)

almost invariably found

earliest

evidence for twill

is

from Anatolia and dates to the 4th millennium BC. This is followed by evidence from the Caucasus of the early 3rd millennium BC and then, after a considerable chronological gap,

Hallstatt culture in Austria

(c.

we

recover evidence for twill in the

1100-450 BC) and about the same time

in

Ferghana, the land of the 'blood-sweating horses', one of the western

approaches to the Tarim Basin.

Silk

and Cotton

The fabric synonymous with China, silk, derives from the silk worm (Bombyx mori), which is killed by heat within its cocoon so that the silk threads may be extracted before the moth breaks through the strands as it emerges from the cocoon. The earliest remains of such cocoons in an archaeological context come from the Yangshao, the north Chinese Neolithic culture,

5000 BC, but the

c.

2250 BC.

When

silk

is first

of the Shang dynasty

{c.

well developed that one

The Chinese

earliest actual

evidence for

silk

dates to

mentioned on an oracle bone from the

1200 BC)

it

c.

3300-

later part

seems to have become so abundant and

must presume

a long tradition of prior evolution.

aristocracy kept the technique of

its

manufacture secret under

Age contexts Mountains by about 300-250 BC. We have already seen how Han imperial policy employed silk as a bribe to hold off the depredations of the Xiongnu in the first centuries BC. The technique of boiling the cocoons to obtain the unbroken silk fibre from domesticated moths did not reach the West until the 6th century AD and the earlier discoveries of silk in the West, especially those prior to the penalty of death, yet silk

dating to

c.

encountered

in the

600-500 BC and from Pazyryk

opening of the

Silk

wild silkworms. in India,

is

A

West

in Iron

in the Altai

Road, can perhaps be explained by the exploitation of possible exception to

all this is

the development of silk

which may predate any putative contacts between

it

and China.

211

Tartans

in

the Tarim

Finally, brief

earliest in India

mention should be made of cotton which was domesticated

sometime

in the

3rd millennium

according to more recent evidence).

Harappan

civilization

It

is

found

possibly the 4th

BC; (or

Indus Valley or

in the

which maintained trading links

all

over Central Asia

and also with Mesopotamia. Nevertheless, the appearance of cotton outside India seems to be relatively late and it does not generally appear on archaeological sites or 1st

in literary

references beyond

Armed with some rudimentary knowledge to the evidence of the 123.

Baby from

Qawnghul.

its

homeland

until the

millennium BC.

for textiles

also derives

is

Tarim Basin.

not limited to the

We

of textiles,

we can now

turn

might emphasize that the evidence

mummies

but that

much

of our evidence

from burials where only the skeletons have survived.

Second Millennium BC: Qiiwrighul The

earliest attested textile

Xinjiang were discovered

2nd millennium at

BC;

remains

in

at sites of the

such as the cemetery

Qawrighul. As might be expected, the

textiles

from

primitive.

this

There

is

period are the most

no evidence

for seams,

piping, sleeves or trouser legs; dyeing is

limited to very small areas of cloth.

Generally,

all

woollen

textiles are of felt

or employ plain weave.

The 'Beauty

of Kroran' from the Towiin

hood contwo pieces of

River cemetery wore a woollen sisting of an underlay of

dark-brown woollen cloth covered by weather-resistant felt overlay.

a

Her mid-

was made of leather with the warmth and her cloak, which reached to her knees, was plain woven (with extra weft looped in) sheep's wool. Her boots were ankle-high moccasins. The other woman from the same period {c. 1800 BC;) and the Konchi River region wore a similar felt hood and her cloak was a golden-brown woollen length skirt

fur turned to the inside for

wrap consisting of two weave that had (the

strips of plain

been stitched together

loom employed seems

to have been

incapable of producing a piece of cloth

wide enough to cover the woman). The child from the same graveyard was

212

Tartans

in

the Tarim

in a brown and beige woollen The proud mother or grandmother who wove this blanket could

wrapped blanket.

almost be accused of showing off as

it is

a

tour-de-force sampler of different weaves

which changed every few inches.

The

textiles

from Qawrighul are more

We may

than a fashion statement.

recall

that the range of the wild sheep does

not extend as far east as China and the earliest

evidence for sheep

2nd millennium

to the

of woollen textiles

in

China dates

in

BC:.

The presence

the

Tarim Basin

indicates that by the early 2nd millennium

BC domestic sheep from the west (along with domestic wheat, possibly barley)

had been introduced

Whether

it

to the

was introduced by the

ancestors of the Qawrighul

not

is

Tarim Basin. direct

mummies

or

impossible to determine, although

the association of a western physical

humans with

type of

a (more) westerly

domestic livestock suggests

just

such a

conclusion.

While we cannot be certain of the date of a

number

of other

mummies, some

from the same general region may also be discussed here.

The

by Bergman at a

site

mummy

discovered

near Lopnur, which

he numbered Cirave 36, also haci a double-

two feathers and with ear-flaps that could be tied under the chin. The old woman in this grave was enveloped in a large dark brown woollen cloak trimmed with a yellow and red border and fastened layered

felt

on the

left

cap, pegged with side

together with skirt of red

wooden (and one bone)

and undyed woollen

pins.

fringes.

Around her waist was

a string

Her shoes were of rawhide with an

T24.

Hat from Grave 36

near Lopnur.

inner sole of lambskin.

more difficult to discuss unless we presume Bergman at Ordek's necropolis date to approximately the same period on the basis of similar ritual and finds (they may actually be much later). Even here we will not get very far. The mummified young man from Grave 5A, the best preserved of the burials, wore a white felt cap with ear-flaps and five feathers mounted on pegs; it was

Male

dress of this period

is

that the burials excavated by

21.5

Tartans

in

the Tarim

fastened around his chin with a cord. This at least indicates that the 'Robin

Hood' style of headgear was worn by both males and females; whether there was a more subtle 'gender statement' being made in these hats we cannot tell without a greater sample. Certainly unsubtle was the fact that around his waist he wore a narrow loin-cloth (only 5

anchored to

cm

was made out of wool

(2 in)

wide) which had been

was also the large mantle (2.1 m b\ 1.55 m (6 ft 10 in by 5 ft in)) which was wrapped around him. This unfortunately is the only piece of male outer clothing from the area. As such a mantle may have served equally as a shroud, it is difficult to his penis.

It

as

1

be certain

how males

Bergman

actually dressed during this earlier period.

did observe that the young

man was shod

with brand new moccasins of

ox hide.

The

Millennium BC: Zaghuniuq

1st

The recently excavated mummies from Zaghuniuq (near Charchan) indicate new techniques that had appeared in East Central Asia by the 1st millennium BC. The quality of the weaving had improved and now included not only plain weave but twill. The quality of the wool was also superior: it was a fine wool with

To

fit

little

evidence of kemps.

out 'Ur-Da\id\ the gentleman from Zaghuniuq, we can start with

his array of caps

and hats

brown hat was made by in

(he

had ten altogether).

very stylish beret-like

known as nalbinding ('needle binding' made from a single piece of yarn), which knitting. This is one of the earliest known examples the technique

the Scandinavian languages,

gives the

appearance of

of nalbinding in the world. Another

analyzed by Chinese

brown

felt

and

a

mid-seam had

beret

from Qizilchoqa has been

having been knitted;

textile specialists as

be the earliest example of that technique 125 Beret-type hat

A

in the

this

too would

world. Another hat of white

a rolled piece of felt in the front to

form two 'horns'

cartoon viking) and there were two flaps with braided strings

recovered from

(like a

Zaghuniuq.

for fastening the hat to the chin. Still

another hat was made by sewing

together two pieces of thick brown

felt

stitches of white thread to

high (32.7

cm

peak with is

(1

a

ft)),

with neat

form

a very

rakishly tilted

turned-up brim that

also edged with the white

decorative stitches. shirt

His

was crudely form-

fitted:

it

consisted of

two rectangular

bolts of

burgundy coloured wool that were draped over

each shoulder to which were attached tubes to serve as sleeves; the fabric here

214

was plain

Tartans

weave with red piping

m

Also discovered

his

in

the Tarim

alont^ the seams.

tomb was

a bulky,

double-faced twill-weave sweater

made

of

heavy cream-coloured threads; the sleeves

would only have come down

man's

to the

elbows and the low-cut, open neck was tied together

Below

with thread.

his waist the

pair of woollen trousers

man wore and

it

a

might

be noted that a pair of unaccompanied trousers were also recovered from else-

where on the piping were

site.

in

Seams

evidence.

as well as red

He

also

wore

a

multicoloured belt (red, brown, blue, green and yellow) fashioned from

woollen yarn. The

concept of a five-coloured belt

wore and

practice in India whereby brahmins

still

[pjuca-vupj) thread over their shoulder. This

is

reminiscent of the

wear belt,

a five-coloureci

however, did not

126 Homed' hat made of white felt with chin strap: from

Zaghunluq

hold his trousers up but merely fastened the two sides of his upper it was his They were double layered

garment. Deerskin boots rose to his knees but stockings that really caught the eye.

with robin-egg-blue red

felt

underneath for insulation and bright-

and yellow woollen

wound around the outsicle to known in the prehistoric world,

strips

create the loudest socks

although they would have been entirely covered by his bot)ts.

One more garment

is

to be associated with his

- an enormous outer coat made of extremely thick brown thread which was set about 80 cm

dress

(2 ft

7

in)

from the surface of the grave and

served to protect the lower

tomb

(ill.

felt

128).

On

chamber of the it was a white

top of

blanker on which

turn were

in

placed a leather sacldle and a simple,

round-based black

jar.

The

saddle

and trousers go together while Westerners

may look

ancient Greece or the

home

well:

Rome

to for

of their philosophies,

laws and literature they have aciopted the costume of horseriding barbarians

who,

for practical

(and obvious) reasons, developed trousers. It

might be noted that the ancient Greeks mocked the concept of trousers,

suggesting that close-fitting pants repressed the sexual abilities of their

127. Tall

peaked hat

of brown

felt

with

white stitching, from

northern neighbours!

Zaghunluq.

215

Tartans

in

the Tarim

i

128 The coat of Vr-David'.

The woman accompanying 'Ur-David' wore

a finely

dark red that reached to her calves. This was perfect that

it

a

woven

twill

almost had a sheen and would no doubt fetch a large sum

produced today. Under her knee-high boots of white deerskin her

was wrapped

in red

wool

flannel

possible fashion statement that

and her

now

right in yellow

legs

One

of the other

were covered with

spirals.

The excavator

weave with red

The

women

in

left

if

foot

and sky blue,

a

escapes us (of course, textiles were

'women's work' so we must presume she had quite worn).

robe of

marvellous garment, so

the

tomb wore

a yellow- (or white?-)

a bit of say in a

what was

dark-red dress and her

dyed woollen cloth with red

also recovered a blue shawl of loosely spaced plain

stripes.

infant buried next to this tomb, 'Baby Blue', with the fluffy blue

bonnet and blue stones over

its

eyes,

wore

a

red-brown plain-weaxe shroud

or baby blanket which was then wrapped in a white

felt

was found on the man's

dress.

Moreover, the same kinds of thick strings of

twisted blue and red strands that

bound

the hands of 'Ur-David' to his chest

were used to bind up the baby's wraps. This seems to be clothing was

made

The same brown wool,

blanket.

beautiful, rich shade, probably achieved by applying red dye to

by the same

woman who,

a family

whose

not unexpectedly, produced

colour-coordinated textiles for members of the same family In the

more

recently excavated 'sacrificial' burials at

the tripartite mature

woman,

woman),

woman wore

the mature

Zaghunluq (with

the 'Scream Baby' and the mutilated younger a beautiful violet-purple

upper garment

and a coarsely woven greyish lower garment. The baby was dressed

amazing brown shroud decorated with

216

red strips

in

an

on which were tie-dyed

a

Tartans

scries of yellow circles.

baby's

mouth shut

A

Irene Ck)od analyzed burials.

The extremely

the site and

Good

yellow woollen headh>\nd u.is

(the family of 'Ur-David' used a

employed

burgundy

in

the tarim

to tie the

strap).

one additional piece of fabric from the Zaghunluc] fine thread suggests the possibility of

cashmere on

points out that goat skulls are associated with one of the

Zaghunluq burials. To this we might add a large quantity of spindlewhorls, some of which were ornamented with interlocking spirals.

Tartans at Qizilchoqa Although dating to about the same period as Zaghunluq, the cemetery Qizilchoqa to the northeast near

Hami

far-reaching historical connections have been suggested. of the Qizilchoqa cemetery

is

at

yielded different weaves for which

problematic: the

initial

The

precise date

dates placed

it

at

about

1200 BC, contemporary with the later period of the Yanbulaq culture, but a

new radiocarbon date of

c".

8(){)-530 Bc; suggests that

Tort Erik (Sidaogou) culture. The abundant evidence a variety of clothes, including

it

woollen robes with coloured belt bands and

fur coats (the fur turned inside) with integrated gloves,

wooden buttons

(ill.

130).

belongs to the later

for dress here revealed

But our main story

lies

which fastened with

with the woollen

textiles.

729 Robe from Qizilchoqa with a

belt.

217

Tartans

in

the Tarim

Irene Ciood

made 6

examination of

a detailed

fragment (15

a textile

hy 4

in

em

The main weave

cm

by 10 the

frt)m

in))

site.

here was

normal diagonal

twill (pi.

X) bur the decoration

in-

voked the production of plaid, the same type of technique

decorative

one might expect on Scottish tartan

(pi.

a

XI). This

mvolved the weaving of wide and narrow colour-strips on both

warp and

the weft

and here the

employed were threads of

urs

blue,

te

and brown, each thread made

of

some

to 40 fibres.

,•>()

The white

d brown thread are natural while e blue

thread

dyed. This small

is

has been invested with

trip of cloth

iea\y historical implications.

As we have seen, the twills

known

region between Turkey

Caucasus where

and

the\'

130. Fur coat (with the fur

turned inwards) and

integral gloves

arly at the site of Hallstatt.

occasionally themselves)

from ,^^i,^^.^_

Qizilchoqa.

^^

^,^^.

Here miners

in

left

Bc; in

the

millennium

are

abundance from the millennium

the

^n^.\

the\ are dated

to the late 4th-3rd 1)C\

earliest

derive from

found late

m

2nd

Europe, particul-

residues of their clothing (and,

the protective environment of Austrian salt

Hallstatt culture occupied a territorv which classical authors

would associate with Celts only

a

few centuries

later,

it is

generally

presumed

that the miners here (and the warriors and others buried in the neighbouring

cemetery) were also Celts or Proto-Ck'lts. The easternmost finds of

dating from the centuries around 1000 B(

(or

fragment from Qizilchoqa and many others

like

(some very Scottish looking); true into the

1st

millennium

Al).

twills are

The Qi/ilchoqa

somewhat it

from the same cemetery

unknown twill

twill,

later), are the

is

in

China

until well

\irtuall\ identical to

the textile fragments recovered from Hallstatt with respect to both style and

technique (hence one of the arguments employed by the tabloid press for placing kilted Celts

in

the

Tarim

Basin).

We

are not talking simply of the

diffusion of a particular weaving and colour pattern. As Hiizabeth Barber writes: 'the regular

combination of plaids and

twills in the

the similar pla\ of wides and narrows in the plaids

218

move

same

cloth and

us into a border

Tartans

^

/one where

sum

ir's

total as aeeidentaP.

There

is

also a

similarity in the weight of eloth.

and

Qi7.ilchoc]a materials, for

;

;

colours while the Qizilchocja

used

six colours. In addition,

there are even differences

Tarim

plaicls

plaids. Irene

Good

among

the

has noted that

the weaving traditions of

Zaghunluq

$

i

*

-Vt^ \

•

S5«.\ \

>.

•

. >.

^

J.,.:-

*

'

»•,

•*• .-

* i

.

a different '

'

'^

5 s :*.

.

warp rather two as found at

over three stems of the

than the more typical

.-^^

** -

".

;

V

> N > s s

>\

';

k

,

*

*•

*.

*

^

•*'N*W-

-,•''* .•:->.

«

••

•-

••..

V

-

i

5

;

^;

s •-

^.

.

S

•*

•

s s

between the Huropean and East Central

•.

«•

VI

:-

>

s ^

\

•;

>«.fc»r"^

?

"•-'::

Qizilchoqa. in weighing the similarities

«»

^ N

^

'•

.,

"

^

.

i*.

•.

•v^,.
r/yy y/fy//// /yyyyy yyyyy /A

%«»^.»»^>>>>;»!«

nryyyryyyyy^

Yyf^^a^^

163.

A

pot

(A)

footed bowl

and

(B)

from Keremchi, and a typical pot (C)

and censer

(D)

from

the Afanasevo culture.

307

Who Were

the

Mummies?

Nevertheless,

if

the similarities with the Afanasevo and subsequent cultures

can he translated into dates, the evidence provides additional support for the expansion of a steppe-derived culture into East Central Asia by the

2nd millennium and

Bt;.

Keremchi burials are found also

as far south as Uriuiichi

and the Yanbulaq

in the

southern Altai

culture.

Splitting the Differences

We

must reconcile ourselves

to the fact that the earliest

Bronze Age culture,

possibly the earliest farmers, in East Central Asia were representatives neither

of the Afanasevo culture nor of the

Afanasevo boasts

earlier dates

BMAC

per

se.

Of

the

two

cultures, the

and more mobile antecedents. So

far

it

reflects

the earliest evidence for the exploitation of domestic animals in the

steppelands far east of the Urals. To connect the earliest Bronze Age settlement of East Central Asia with the Afanasevo culture would seem to require one of the following models.

One might propose

that there

was an

earlier

common

source for both

the Afanasevo and Qawrighul cultures that spread across Kazakhstan

eastwards to the Yenisei and southeastwards into the Tarim Basin. early such a

phenomenon might be

Afanasevo dates would carry rassingly early given

its

is

constrained by the fact that the earliest

this culture

back to

supposed antecedents

in

c.

3500

B(;

(already embar-

the west) and so

have to envisage a spread of stockkeepers eastwards during the the 4th millennium BC. This agricultural expansion

is

is

1

64. Hypothesis of a

source for

both the Afanasevo

and Qawrighul The

common

would

cultures

source

ultimately

lie

west of the Urals but a staging area for both cultures in Kazakhstan IS

purely conjectural.

308

we would

first

half of

not impossible but evidence for such an

not supported by archaeological data and

of cultures east of the Urals that date to the 4th millennium

common

How

we know

BC; that are

Who Were

clearly ;zon-agricultural. Unless the archaeological picture of alters radically (a possibility since there

common source for both

is still

so

the

Mummies?

Kazakhstan

much unknown) an

the Afanasevo and Qawrighul cultures

is

earlier

probably

not the preferred option.

and stockbreeding

Alternatively, the spread of agriculture

which

is

indicated by the Afanasevo culture

movement south

into East Central Asia.

lations such as the

may have

If

this

to the east

also seen a subsequent

were the case, then popu-

Qawrighul culture would either have had

to have

abandoned the deposition of pottery in graves (as is well known in the Afanasevo and later Andronovo cultures) or possibly to have neglected the manufacture of pottery altogether and replaced it by organic containers (basketry

highly developed at Qawrighul). This latter situation

is

unheard of and

Age

in Ireland

is

is

not

encountered, for example, at the transition to the Iron

and western

Britain.

But neither settlement of the steppelands

of the Minusinsk Basin nor the uplands of the Altai should have prepared

such newcomers for the environmental regime of the Tarim Basin:

how

could

mixed stockkeepers and farmers from the north have successfully developed the irrigation agriculture required to exploit the oases of East Central Asia? It is

this

need to develop an irrigation-based economy that turns our

attention westwards back to West Central Asia, specifically Bactria and

Margiana, where we have seen that developed by

c.

2000

BC".

of a direct import of the

a very successful oasis

But we have also seen that so far there

economy was is

no question

BMAC into the Tarim Basin and the parallels tend

to be general rather than specific.

The

case for external influences improves

markedly between the Tarim Basin and the west, specifically between the

Tarim Basin and Ferghana, in the later 2nd millennium BC, a time after the initial settlement of the Tarim by Bronze Age farmers. For this reason we have an incongruity between a burial rite and material culture that may be derived from the steppelands and an irrigation

from the agricultural oases of Central Asia.

economy

that should derive

How can these two very different

components be reconciled? Our most obvious recourse is the structure of relationships between the steppe populations and those of the oasis communities. We have already seen in West Central Asia that steppe tribes could settle and adopt irrigation agriculture. But in West Central Asia, where we witness such phenomena, this symbiosis always occurs where we have prior agricultural settlement. This is not a model that we can easily transfer to East Central Asia since we have no substantial evidence for agricultural settlements prior to the establishment of the Bronze Age, nor can

we demonstrate

that the physical

type associated with such an intrusion of farmers predates those from the

northern steppelands. To prepare a 'northern steppe culture' for life in the Tarim Basin, we might expect that it first came into contact with one of the existing West Central Asian oasis-based cultures before it arrived in East Central Asia.

Some

linguistic evidence

may support

just

such a hypothesis.

i09

Who

Were the Mummies?

The

We

Linguistic Stratigraphy of Tocharian

have seen in the preceding sections that one of the

correct ordering of our evidence, be influences.

The problem

it

now

before us

critical issues

is

the

that of populations or archaeological is

arranging

in correct

chronological

order the different prehistoric languages spoken in the Tarim Basin. Various

accompany such an

degrees of difficulty the primary goal

exercise but from our perspective

to arrange correctly the relationship

is

between the

Tocharian and Iranian languages as these are the languages most have served as the vernaculars of

its

As we have already seen, every language contains something of cultural history in

vocabulary.

its

likely to

prehistoric populations.

Some

its

own

of the vocabulary will be inherited

from Proto-Indo-European, some will be new words created from older inherited elements. These do not hold the same interest for us here that loanwords do, since borrowed vocabulary, particularly diacritic cultural

may point to the time and place of foreign contacts. As we have remarked before, Chinese appears to have had a very minimal impact on the Tocharian language. To be sure, we have the odd loanword,

vocabulary,

e.g.

Tocharian klu

names

for units of

toil'), all

'rice'

from Old

measurement,

Sinitic e.g.

'"ghw (modern Chinese dao) and

tow

sank,

{- 10 sank) and cak (= 10

transparently from Chinese. But the impact here

culturally predictable

and comparatively

is

minimal,

recent. Otherwise, there

is

no

evidence that the Tocharians gained their vocabulary for the native flora

and fauna of the Tarim Basin from

earlier

Chinese inhabitants, nor do they

appear to have gained their agricultural vocabulary from

this source.

This

lends additional support to the argument that the Chinese were both

latecomers into the Tarim Basin and did not have a significant impact until relatively late in the

Now many

easternmost regions.

of the other loanwords found in Tocharian are connected

with the religious and social

life

associated with

Buddhism and hence

the

sources here are Buddhist Hybrid Sanskrit, Prakrits and other Indie languages, occasionally perhaps filtered through Iranian intermediaries.

These may be

enough associated with the

easily

relatively late spread of

Buddhism among already established populations, both Tocharian and Iranian-speaking.

number of loanwords in Tocharian would appear to derive from Saka or some other East Iranian language and these fill out a larger semantic range. In some instances they appear to have been loanwords associated with exchange, e.g. Tocharian B pito 'price' may derive from Saka plha- 'price', and Tocharian A pare, Tocharian B peri 'debt' would be derived By

far the greatest

from Saka

plra- 'that

which

is

to be paid'.

There are other words which

might also be assigned to the general realm of commerce, e.g. units of weight and measure. Iranian military terms also had an impact on Tocharian and

when one considers

depictions

in

Buddhist temples of Tocharian knights

kitted out like Sassanians with long swords,

310

one might

recall that the

Who

Tocharian B word for sword, kerttc, language, filling

our similar political contexts,

A

kJkmiirtik

Tocharian word for related in

'iron'

some way

'ruler'.

e.g.

It

is

kamirdo

'head', 'chief (god) gives

also important to note that the

(Tocharian enctiivo or ancii) would appear to be

to Iranian (Ossetic). In

some

instances the loanwords

are connected with exotic animals, e.g. Tocharian

Saka mitya-

some East Iranian

derived from

Avestan karata- 'dagger'. Bactrian also supplied loanwords

e.g.

Tocharian

is

Were the Mummies?

'tiger',

B mewiyo

'tiger'

from

or Tocharian B eksinek- 'pigeon' which seems related

to Saka assanaka- 'pigeon'. By difficult linguistically to

and

make any

large, however,

it

would be exceedingly

case for the temporal priority of Saka in

Tarim Basin with respect to Tocharian. Of the approximately 25 or 30 loanwords, most could be easily explained as later borrowings passed between communities along the Silk Road connected with the rise of urban or Buddhist institutions. Most importantly, there is no evidence that the vocabulary of agriculture in the Tarim Basin specifically derives from Saka. the

Asses, Canals and Bricks

Although Saka and Sogdian are the attested in the

earliest of the Iranian

languages directly

Tarim Basin, there are some Iranian loanwords

in

Tocharian

that may derive from a still earlier period. Two of these have only come to light in the form of a Tocharian B text that documents the

of an estate. Within the text there canal'

and the

arte^

boundary. Douglas are early loans

is

that can serve as an estate

has recently suggested that both of these words

from an East Iranian language

(or proto-language) and, while

how

these loans could have taken

there are a

number of ways

place, only

one

is

to explain

not problematic. As irrigation agriculture

of agriculture that one can practise

how

transfer

reference to the orotsa newiya 'great

some form of watercourse

Adams

recently

in the

Tarim Basin,

is

it is

the only type

difficult to see

the Tocharians could have eked out a living there before they

knew

of

and Tocharians both arrived same time, the Iranians passing onto the Tocharians the technology of agriculture before the latter had starved to death. The simplest explanation would be to place the Iranians in position first and then have the Tocharians, largely pastoralists, wander in and adopt agriculture; this is precisely the type of pattern we have earlier encountered in West Central Asia where steppe pastoralists moved into the oases and adopted the irrigation.

It is

also unlikely that the Iranians

at precisely the

agricultural techniques of the earlier inhabitants. But, as

seen,

we do not

really have typically

West Central Asian

we have

already

BMAC

farmers

occupying the Tarim Basin, nor does our review of the physical evidence suggest that populations from West Central Asia arrived

Turpan basins

earlier

than those from the north.

None

in the

Tarim and

of our usual models

will do.

Adams

suggests that the most convenient explanation would involve

the adoption of irrigation techniques and East Iranian terminology by the

311

Who Were

the

Mummies?

Tocharians en route to their historical those archaeologists

BMAC

in

who

This model would

seats.

have noted that while there

fit

nicely for

no evidence

is

for the

East Central Asia, certain similarities between the Yanbulaq

culture and that of the

BMAC do suggest some form of

mediating culture,

Andronovo type. We may moved from the steppe, through the Altai and Tangri Tagh, and south into the Jungghar, Tarim and Turpan basins, settling in the oases of the latter two to engage in irrigation presumably

a

more mobile pastoral

culture of the

suggest then that Tocharian populations

agriculture. It

might be objected that no one entering the Tarim and Turpan basins

who had likely to

make

to

way through

the surrounding

mountain passes

is

have carried the techniques and vocabulary of irrigation agriculture

with them. This

word

their

is

not a serious objection as the underlying East Iranian

we find borrowed into Tocharian may also mountainous regions where it is known in Sarikoli, one of the

for 'irrigation canal' that

be found

in

Pamir languages related northwest of India

words connected with

and cognates are also known

to Saka,

in the

Nuristani and Dardic languages. That

irrigation can be

approaches to East Central Asia

Of

problem before us but

if

perhaps present

common

found on the western mountainous

at least suggests the possibility that

Tocharian speakers passing through terms and techniques.

in the far

this region

course, a western entry

East Iranians were

in the Altai as well,

it

in

could have adopted such

would not

occupation

might provide

really solve the

in

Ferghana and

a plausible route for

northern steppe peoples southwards into East Central Asia.

Another

instructive piece of evidence

Tocharian B kercapo which gardahha-.

If

this

is

a

is

is

the Tocharian

word

for the ass,

universally agreed to be related to

borrowing,

it is

a very early

Old

Indie

one when the underlying

Proto-Indic and Proto-Tocharian was something like "'gordebhos.

form

in

Now

such a form could just as well be Proto-Indo-European and so

we

are

not dealing with some word borrowed from the period of Buddhist

word ''gordebhos should have been 2nd millennium BC or earlier. We are uncertain as to whether the original word referred to the domestic ass or the wild onager, and was then transferred to the domestic ass. The ass was originally domesticated in North Africa and then expanded into Mesopotamia by the 4th millennium BC, spreading from there both northwards and eastwards; its first appearance in northwest India is around 2000

expansion

in

circulating

among

BC.

On

East Central Asia; a

peoples

in the

the other hand, the so-called 'half-ass' or onager ranged across the

and south into Central Asia and northern India. Whatever the meaning of the word, the linguistic evidence suggests some form of contact between Tocharians and Indo-lranians long before the opening of the Silk Road and the spread of Buddhism. We have already seen that one of the cultural features that tends to link the prehistory of East Central Asia with that of West Central Asia is to be seen in the use of mud bricks in domestic and funerary architecture. Of

entire steppe

original

312

Who

interest then

the fact that Tocharian iscem 'clay'

is

would appear

Old

to Avestan istyam 'brick', zamoistva- 'clay-tile'.

and Khowar

Were the Mummies?

to be related

Indie istakd 'brick',

northeast hido-Aryan language) ustu 'sun-dried brick', 'large

(a

clod of earth'. This would again point to either a loan from a very early stage of hido-lranian or the mutual inheritance by both language stocks of a

common So

eastern hido-European proto-form

far the linguistic evidence

is

'''isti-

'clay', 'brick'.

hardly overwhelming, but

it

does hint at

contacts between Tocharians and Indo-Iranians prior to the spread of

Buddhism

East (Central Asia. These contacts need not have been

in

particularly intense

we were

and we would be giving

to portray the Proto-Tocharians as

a very mistaken impression

if

wandering nomads who only

learned their agriculture from settled Proto-Indo-Iranians. To be sure, they entered the Tarim Basin with their livestock whose Indo-European

were retained,

Tocharian B

e.g.

ke^^,

okso,

a^^

names

and stnvo are cognate with the

English words 'cow', 'ox', 'ewe' and 'sow' respectively.

They

also retained at

B and words for the basic agricultural pursuits of ploughing, sowing, threshing and grinding. The ancestors of the Tocharians were already mixed farmers (agropastoralists) and were not ignorant of cereals before they entered the Tarim Basin, nor did they require contact with Indo-Iranians to learn about least

one of the inherited Indo-European words

tano from a

farming.

late

One

Proto-Indo-European

for 'grain' (Tocharian

''dhoh-j^neh^j-),

curious word, Tocharian B kanti 'some form of bread'

possibly related to Indo-lranian words for 'wheat'

(e.g.

Khotanese Saka ganama) and Hittite kant- 'wheat'. Although

common word

been a

in the eastern

is

Avestan gantuma-,

Indo-European world,

this it is

may have generally

taken to be a loanword from some Near Eastern language that spread

through Asia.

Finally,

one other feature should be noted: there

is

no evidence

of the Indo-lranian languages adopting Tocharian words; our loanwords

only seem to go

So

far then

in

one direction.

we have found recurring evidence

(?)Proto-Europoid

one follows the typological school)

(if

= steppe = = Tocharian; and

that north

west = oasis = Indo-Afghan/Pamir-Ferghanoid (or Hemphill's Bactrian-

Alwighul-Kroran group) = Indo-lranian. Before we pat ourselves on the back

it is

The

about time we get a grip on things and have an 'assumption

alert'.

as Proto-Tocharians

hangs on our identification of the northern intruders and not some form of Indo-Iranians. But we have

already seen that the

BMAC

logic of

all this

is

early Indo-lranian identity; the

not the only culture to which

Andronovo

culture and

we

some of

assign an

its

western

neighbours on the European steppe are also widely regarded as Indo-lranian. Indeed,

we have

given strong reasons to believe that the Indo-Iranians were

originally steppe peoples themselves

of West Central Asia.

And we

who came

dominate the oasis culture

have also seen that the Qiiwrighul physical

type might he related to that of the liberty to

to

Andronovo

culture.

One

is

at perfect

propose the equation: north = steppe (Andronovo) = Proto-

Europoid = Indo-lranian. The Occam's razor we employed

earlier to

313

Who

Were the Mummies?

separate Tocharians and Indo-Iranians has suddenly

instrument indeed.

It is

time

we played

become

a very dull

the Afanasevo card.

The Afanasevo Card: The Short Trek

We

have worked our way into a logical corner but at least we have some

notion of what type of device

We

predicament.

it

is

going to take to extract us from our

have the Tocharians

in

the northern and, accepting the

Kroranian evidence, southeastern part of the Tarim Basin by the of the

We

millennium AD.

1st

should have been there at least by the of Buddhism) and

first

half

have set out arguments to indicate that they 1st

we have seen why

millennium BC (before the spread

it is

easier to derive

them from the

north than from any other direction (although we must admit that Iranians

might also have come from

this

same

direction). Furthermore,

we know

from our examination of the separation of the various Indo-European stocks that there are

no particular reasons

to associate Tocharian genetically

We

closely with Indo-Iranian than with any other stock.

reviewed the linguistic evidence and seen that opinion tends to be

between assigning Tocharian

its

among

closest relations

and arguing that

more

have already split

various European

separated from the other

stocks (Germanic, Greek,

etc.)

Indo-European stocks

such an early date that there are no grounds to

presume

it

was

a large part of

a close

at

it

neighbour of any other Indo-European group during

early development. Either way, the Tocharians are clearly

its

not part of the greater Indo-Iranian superstock and so, for linguistic reasons,

we

will prefer

any model that keeps some distance between them and the

other stocks throughout a considerable part of their early evolution.

come

On

the

when they had already emerged as an independent stock, presumably sometime after 2500-2000 BC. It is most likely that it was during the early part of the 2nd millennium BC that Proto-Tocharians came into contact with already settled other hand, they did

into contact with the Indo-Iranians

Indo-Iranians and borrowed a few terms and techniques relating to irrigation agriculture, brickwork

From demands

and possibly the domestic

an archaeological perspective, one is

way

ass.

to

accommodate

all

these

to regard the Proto-Tocharians as an offshoot of the Afanasevo

culture of the Altai-Yenisei region. As

we have already

seen,

it

displays

genetic connections with the cultures of the European steppelands (and

hence might represent the eastern extreme of an Indo-European linguistic

continuum) but

it

was also

isolated

from the other steppe cultures with the

European steppe and and then replacing it. If an offshoot of the Afanasevo culture moved southwards into East Central Asia in the centuries around 2000 BC, it could hardly have avoided contact with Indo-Iranians of some sort just to its west and it may thus have adopted the rudiments and some of the vocabulary of irrigation agriculture from them. It would then have moved into the northern and southeastern Tarim Basins where it established later

Andronovo

the Yenisei River

314

culture filling the gap between the

Who Were

itself,

the

Mummies?

with marked cultural change and adaptations, and continued until

the historical expansion of the Uyghurs.

model the Proto-Tocharians moved south from the Altai region in about 2000 Bc: to settle in the northern Tarmi and the southeast (Kroriinian). Subsequent movements of populations from the north may have carried Iranian speakers into the Tarim Basin as well but they never achieved linguistic ascendancy in the north and east. The Saka, however, entered the In this

Tarim from the west during the 1st millennium BC and established themselves in the northwest of the Tarim and in the south at Khotan. How does such a model work geographically?

165. Northern

approaches

to East Central Asia lead

at the map of the approaches to the Tarim Basin then the would be from the northwest towards the southeast, i.e. from territories that may have once been part of the Indo-Iranian chain of languages of West Central Asia to the Altai southeast into the land of our If

we look

natural route

earliest

mummies. To

this day, the best passes for entering the

Jungghar

one

into the eastern

end of the

its

northwest side where one can skirt the

forbidding centre of the basin and move along the southwestern mountain slopes to the south. This

>^^'

would bring northern immigrants through

a funnel

from the

Afanasevo culture would carry

one through lands

occupied

Basin from the north are along

Tarim Basin.

A movement

(later)

Andronovo

by the

culture

before arriving

in

the

Tanm region

vaSevDl

(r=^

^m

a? ^/

I

y\ua"^

O c \n the Kroranian document). According to him this con-

Agnean

firms that the source of the

preted as a 'Kuchean

woman"

(if

related to Sanskrit istlkJ) or a

member first

of

a

Brahmin family

isciike

is

has been regarded as equivalent to

ktisitirie

'Kuchean" thus securing,

Kuchean language

itself,

in the

the equation

between Sanskrit tokharika, i.e. a 'real" Tocharian (of Bactria), and a Kuchean. We also have a Tocharian B text that glosses 'Tocharian' with 'Kuchean', thus

that they both regarded them-

Tocharian (KA) and as Chinese historical

selves as 'Tocharian'.

To (1)

these equations

word was

we may add

that

'Tocharian' was widely used before

tradition derives

Buddhism from the

Yuezhi (Kushans), this further secures the

= Kushan = Tocharian

text

this

equation

us only

tells

is

what the

Turks (Uyghurs) called the language of it

(or themselves).

It

possible, for

is

example, that the Indie original of the Maitreyasamiti was

first

translated into

"toxri (some Iranian language)

and

only then translated into Tocharian

A

and Turkic (Winter, pers. com). 3. As for the St Petersburg document, kucatine reveals an a where we would expect

;,

i.e.

k"sinne 'Kuchean', there-

fore, there are

the

kucanne

grounds to doubt that = Kuchean (Winter

suggests that the referent here

may

be 'Kushan' although this requires one

and after our Kuchean-Agnean (KA) documents for peoples and places in East Central Asia (and not just Bactria), and (2) if we follow the Chinese equation and

equation Yuezhi

usage of

as 'the four Tuyaristan' (e.g. Persian

'Tocharian woman', and yet the Toch-

cahar tuyaristan, Sogdian

arian equivalent iscake

Yuezhi

Bactria), there

is

for

Tocharian

(of

persistent evidence for

(KA) language.

We

should note here the issue of the

political entity of the

to explain

why

the L'yghurs were inter-

ested in a people

Tarim Basin known (adj.)

ctfi'r

who had

existed a

millennium before). More importantly, the Sanskrit

word

clearly refers to a

is

either a

333

On

the Tocharian Problems

masculine or neuter noun,

i.e.

It

frus-

any attempt to translate the word

trates

'woman". While we cannot be absolutely certain what 'the four Tuyaristan' refers

both Sanskrit "white",

as

brilliant

4.

family

to, there

is

some evidence

that

it

may

be equated with the 'Four Garrisons' of

Tang dynasty and

the

as these

sviti-

and Avestan

'brilliant",

ones",

Iranian

the

cf.

who were

sfnti-

=

k"ci

i.e.

'the

royal

expected to possess

solar brilliance, the Avestan x'-'argnah-.

The equation

phonologically possible

is

but there are serious problems: 'kwiti-

not attested

is

in the

1 )

(

a root

Tocharian

mendicant' and when modiit refers to Sanskrit and not Tocharian A. According to D. Q. Adams, it is more 'religious

fying a language

Agneans called themselves what all our other sources call them, something like "akhi which could derive from a Proto-Tocharian \jke 'end', likely that the

included Qashqiir and Khotan, both of

languages although there are several

'limit", i.e.

which would have spoken an Iranian rather than a KA language, it is

other words for 'white' or

of Qarashahar were the 'borderers',

difficult

how

see

to

reference

this

the genitive of Kuci

word

that the

supports the equation of Tocharian -

plural,

KA

bright ones';

Tarim Basin. The argument of many of the objections concerns either the rejection of in the

phonological equations as spurious, or imprecision of any ethnic term

the

regarding Kushan, Yuezhi,

etc.,

because

these words were regularly re-applied

\n other

'gold"

example,

For

Kucheans were Buddhists and Buddhism the Tocharians/ Kushans, therefore the Kucheans were Tocharians/ Kushans, or Kushans in East Central Asia adopted the name 'Yuezhi' because this equation already existed in Chinese tradition (and not because the Kushans

came from

felt

themselves ethnically or linguistically

'Yuezhi', whatever that meant).

A

doc-

ument from Dunhuang, dating to AD 966, indicates, for example, that the state between Khocho and Kucha, i.e. the state centred on the town of Yanqi (Agni), was known also as Yuezhi. One

Kuci cannot refer to 'the

i.e.

(3)

while 'the brilliant ones"

makes sense as an ethnonym, it does not name; and (4) although the Chinese referred to the royal house of Kucha as 'Bo', the recurrent element of the Kuchean royal names was Sanskrit Suvarna 'gold", i.e. Tocharian B Ysasse.

and may have made any number of interpositions.

Kuciii indicating

singular rather than

is

words, the

and not

tie-in

should be with

'white".

What did the Agneans call themselves? Some propose that the native name for Agni (Chinese Yanqi), the region of Qarashahar where Tocharian

A

been recovered,

texts have

that this ones'.

A

word

Arsi and

is

also indicates 'shining

A

Tocharian

text refers to arsi-

kiintw-d, literally 'in the Arsi-tongue'.

Some

from an earlier on the same root that provides Sanskrit drjuna- and Latin jrgentum 'silver'. Two Old Turkish inscriptions from Mongolia (Kiil-Tegin and Bilga Qayan) from stelae that date to the period around ad 732-735 both linguists derive arsi

''argeyes

which

built

is

refer to the

to this territory even after the departure

which, the stelae suggest, were located

and thereby

Greater Yuezhi

the

associate the Yuezhi with territory in

which we

Or this

later recover

could merely

KA

manuscripts.

reflect the

tendency

of the Chinese to label ethnic minorities

whose cultures had been heavily influenced by Buddhism as Yuezhi. The claims and counter-claims concerning the equation Tocharian

= KA continue

with no clear sign of major agreement.

Two

additional problems, the

designation

of

the

Agneans, require comment as

What selves?

state

We have seen

=

MSM

''kuw-dzi.

well.

did the Kucheans call them-

was Kuci

k^ci)

self-

and

Kucheans

The

that the

name

of the

Tocharian B Qiuci = early medieval

(or '"Kusi, in

adjective

is

kiisinrie, e.g.

kusinn oroccepi lante yaitkorsa 'by the order of the great king of Kucha'. As the

Chinese name for the royal house of the

Kucheans was Bo sought to derive plural

334

'white", linguists have

fe"c';

from

kuitcyes which

is

a

nominative

cognate with

'marchers", the type of

we

ethnonym which

find elsewhere in Europe, e.g. Ukraine

('beyond the borders") or Old English Mercia ('the Marches').

toquz

tirsin 'the

nine Arsin'

where we would expect the Arsi,

just

in

the vicinity of Qarashahar.

Although

equation

this

logically possible,

almost

all

is

phono-

the points

proposed can be challenged because (1) the underlying form is not attested in either Tocharian language (although extensions of the same root underlie

Tocharian

B arkwi);

A

arki 'white' and Tocharian

Tocharian word

possible that the

is

it

(2)

is

singular rather than

plural as the interpretation seems to

demand and,

if

so,

it

fails for

the

same

reasons given in our discussion of Kucha; (3) all

external references to the

Agnean

kingdom show

a hard velar (kig) (i.e. Khotanese Argi, Buddhist Hybrid Sanskrit Agnidesa) and never the s demanded by the proposed equation with J«;; and (4) the assumption that the jr5(-tongue refers to Tocharian A is rejected by most Tocharian specialists

who

see

it

as a translation of Buddhist

Hybrid Sanskrit Jrxa-

Further Reading H. W., 1970 'Tokharika", /oi(r>M/ of the Royal Asnitic Society, 121-122. Bailey, H. w!, 1936 'Ttaugara', BSOS 8, Bailey,

883-921.

Haloun, G.

can then argue that the Yuezhi held on of

the people occupying the area

as a place

exocentrically by the Chinese and Turks,

logical

is

'brilliant'; (2)

in

the

meaning of

19^"^

'Zur Ue-tsi-Frage',

ZDMG 91, 243-318. Henning, W.

B., 1938 'Argi and the "Tocharians"", BSOS 9, 3, 545-571. Henning, W. B., 1949 'The name of the

"Tokharian" language'.

Asm Major

1,

158-163.

Henning, W. B., 1978 'The first hidoEuropeans in history', In Society and History, Essays in Honor of Karl August Wittfogel, ed. G. L. Ulmen, 215-230. The Hague, Mouton. Konow, S., 1933 'War "Tocharisch" die Sprache der Tocharer?', Asia Major 9, 455-466. Miiller, ¥. W. K., 1918 'Toxri und Kuisan (Kiisan)",

SBAW,

566-586.

K W. K. and E. Sieg., 1916 'Maitrisimit und "Tocharisch"",

Mtiller,

.S'BAVV,

395-417. Narain, A. K., 198^ 'On the "first" IndoEuropeans: The Tokhanan-Yuczhi and their Chinese homeland". Papers on Inner Asia, No. 2, Bloomington, Indian, 1-28. P., 1934 'Tokharien et koutcheen", Journal Asiatujue 224, 23-106. Sieg, E., 1918 'F'lii einheimischer Name fur Toxn", SBAW, 560-565. Sieg, E., 1937 'Und dcnnoch

Pelliot,

"Tocharisch"", Sieg, E.

and W.

SBAW,

130-139.

Siegling., 1908

'Tocharisch, die Sprache der

Indoskythcn",

Thomas, W.,

SBAW,

915-932.

1981 'Zu skt. tokharika

seiner Entsprcchung

ini

und

Tocharischen",

/CZ95, 126-133. W, 1984 'Zur tocharischen Entsprcchung von skt. tokharika'

Winter,

.

KZ

97, 131-1.53.

Wylie, T.

v.,

1962 The Geography of

Tibet according to the 'Dzam-GlindRgyas-Bshad. Serie Orientate Roma 25, Rome, Istituto Italiano per il medio ed cstremc) oriente.

Xu Wcnkan 19% 'The Tocharians and Buddhism", Studies in Central and East Asian Religions 9: 1-1^.

TWO

APPENDIX

All d.itcs .irc i.,ilibr,uc-2.?(); aiiJ h.lscJ

to 'H",. proh.tliilit> oni|ilo\

LolMiou

l.jh.

No.

int;

Mjtcrnil

on d.ua

sets

of

C^.ilihr.itioii

Stiiivcr,

a laboratory error factor ot

Rcimcr

l'ioi;r.im Ri.\ lt

.il.

4.1.2, Stiiuir.

.\1.

.iiul

I',

j.

KcimLr,

I44S, RjdiiH.irhon 40: l041-l()Si. D.itis

\'->'->'\,

.irc i.,ilihi\UcJ

1.

Date HP

Cal liC

l.()i\ili(»i

i.ah.

No.

Miitcrhil

D.ite

HP

Cat

BC

ALWIGHUL (ALAGOU) M13

ZK-2045

wood

1890±70

41-AD321

M14

ZK-2046

wood

1830±70

AD 27-384

111

M15B

ZK-2047

wood

1870±90

46-AD 383

111

M16

ZK-2048

wood

1800±70

AD 68-409

111

M19

ZK-2049

wood

2050±95

360-AD 131

Ml

WB77-14

wood

2410±80

792-262

111

M3

WB78-24

wood

2360+90

782-202

111

M4

WB77-21

wood

2260±65

406-169

M21

WB77-27

wood

2140±80

391 -AD 47

M28

WB77-26

wood

2490±130

900-234

M30

WB77-25

wood

2260±65

406-169

M32

WB77-24

wood

2350±80

762-204

CHONG BAGH 1

CHARWIGHUL (CHAWUHUGOUKOU) M4 1

1

1

1

1

1

1

1

ZK-1328

wood

2600±80

903-414

M6 M20

ZK-1329

wood

2610±75

903-522

ZK-1330

wood

2580±75

894-412

M25

ZK-1331

wood

2750±80

1125-796

M29

ZK-1332

wood

2670±75

995-672

M31

ZK-1333

wood

2690±90

1015-600

M32

ZK-1334

wood

2550±90

894-403

M59

ZK-1335

wood

2530±75

827-404

II

1

1

1

(QUNBAKE)

Ml

ZK-2113

wood

2500±70

804-401

M2

ZK-2114

wood

2720+100

1185-673

M3

ZK-2115

reed

2600±90

919-411

M3

ZK-2116

wood

2620±75

913-541

M4

ZK-2117

wood

2420±80

794-264

M9

ZK-2143

wood

2190±80

400-2

M10

ZK-2144

wood

2480±95

824-386

1M27

ZK-2145

wood

255O±80

833-406

M34A

ZK-2146

wood

2380±75

765-233

M4

ZK-2288

wood

2570±80

894-409

M7

ZK-2289

wood

2600±75

900-519

1

1

1

M60

ZK-1336

wood

2450+80

799-386

11

1M06B

ZK-2031

wood

2150±75

391-AD 17

11

M10

ZK-2290

wood

2440±75

796-386

IM09

ZK-2033

wood

2740±90

1185-788

11

M12

ZK-2291

wood

2530±75

827-404

IM025

ZK-2036

wood

2460±75

799-392

M18

ZK-2292

wood

2230±90

410-45

IM030

ZK-2037

wood

2720±90

1108-765

1

11

LOPNUR (LUOBUBO)

1

M035

ZK-2038

wood

2930±80

1

M043

ZK-2039

wood

2550±70

829-409

MA2

WB80-24

wood

2010±75

199-AD 131

1

M045

ZK-2040

wood

2640±70

919-562

MB1

WB80-25

wood

1870±80

41-AD 376

M2

ZK-2110

wood

2510±80

824-400

ZK-2111

wood

2380±70

764-259

NURASAY COPPER MINE (NULASAI)

ZK-2112

wood

2330±75

759-202

Mine

BK77001

charcoal

2580±170

1185-260

M13

ZK-2043

wood

1870±75

38-AD 339

Mine

WB82-51

wood

2340+70

759-206

II1M12

ZK-2044

wood

2090±70

357-AD 64

11

11

11

111

M6 Ml

5

1

389-904

.3.?5

Radiocarbon Dates from Selected

Location

Lab.

No.

Sites

Materiiil

Date BP

CalBC

QAWRIGHUL (GUMUGOU)

Location

i.ah.

No.

Material

Date BP

CalBC

TORT ERIK (SIDAOGOU)

M38

BK81042

wool

3390±100

1939-1441

T5(3)H50 WB77-29

charcoal

2260±80

480-112

M38

BK81043

sheepskin 351 0±1 70

2295-1431

T4(3)

WB77-30

charcoal

2400±65

775-379

M38

BK81044

wood

3430±70

1918-1596

T3(3)

WB77-31

charcoal

2270±80

504-119

M4

BK81045

wood

3560±80

2137-1688

T1(4)

WB77-32

charcoal

2360±65

760-233

Ml 2

WB81-28

wood

4140±80

2902-2471

T1(4)H44 WB77-33

charcoal

2320±65

537-203

IIM23

ZK-1 003(1)

wood

3550±60

2106-1694

T2(5)

WB77-34

charcoal

2800±70

1206-810

IIM23

ZK-1 003(2) wool

2120±105

397-AD116

H4

WB77-35

charcoal

2510±80

824-400

T6

WB78-27

charcoal

1990±60

M2

ZK-1 052

wood

1480±70

II

II

II

QIZILCHOQA (WUPU)

64-AD

1

30

Ml 9

WB79-12

wood

2990±65

1408-1004

(2)M4

WB79-13

wood

2760±80

1185-797

M26

WB79-14

wood

3010±85

1436-999

'Beauty'

ZK-1 001

sheepskin

3480±70

2009-1621

M101

WB79-15

wood

3030±85

1491-1004

F4

WB80-23

wood

1860±75

36- AD 376

TOWAN RIVER

(TIEBANHE)

YANBULAQ (YANBULAKE)

SAMPUL (SHANPULA) M02

WB84-11

reed

2780±90

1210-797

M64

ZK-2186

wood

2970±55

1386-1004

M6

WB84-12

reed

1720±95

AD 81-540

M70

ZK-2187

wood

3300±75

1743-1413

M01

WB84-14

reed

2050±65

345-AD 80

M45

ZK-2188

wood

3130±65

1521-1219

M02

WB84-15

wood

1940±75

105-AD241

M54

ZK-2189

wood

2580±55

829-541

M2

WB84-17

wood

2140±75

388-AD 22

M20

ZK-2194

wood

3610±55

2137-1777

M02

WB84-19

charcoal

2230±65

402-111

M14

ZK-2195

reed

2410±80

792-262

M7

ZK-2196

wood

3250±90

1739-1319

M64

ZK-2197

wood

2600±85

910-412

SHAMBABAY (XIANGBAOBAO) M13

BK 77002 wood

2370±70

762-233

Ml 3

WB77-19

wood

2400±70

783-263

M40

WB78-01

wood

4270±90

3096-2584

YEWIRGHUL (YU'ERGOU) = ALWIGHUL CULTURE

M17

WB78-12

wood

2539±65

824-409

M67

WB78-21

wood

2580±85

898-408

M21

WB78-13

wood

2670±65

969-765

M37

WB78-14 wood

2570±65

831-414

M42

WB78-20

wood

2370±85

782-206

M58

WB78-40

wood

2300±130

789-4

M3

WB81-61

wood

2160±70

391-1

M55

WB78-19

wood

2100±65

357-AD51

MB

WB82-05

wood

3060±75

1495-1053

M81

WB78-23

wood

2060±80

355-AD123

M30

WB78-02

wood

2010±80

201-AD133

M47

WB78-22

wood

1890±60

36-AD316

SUBESHI (SUBASHI)

TOMURLUK

(TIEMULIKE)

M2

BK82107

wood

2470±60

796-399

M4

BK82108

wood

2140±60

378-1

ZAGHUNLUQ (ZAHONGLUKE) No published dates but He Dexiu dates

336

1

AD 424-670

in

the range

c.

3200-2700

indicates there are five BP, i.e.,

c.

1

500-850

BC.

,1990 IIh- Cainhridi:.e History

Abbreviations

of harh' Inner .Asli.

llhS-

hhio

loiiin.il of

hiii()l)i\in

(Cambridge Tikh\nisk\, eds.,

General

amhridgc:

C

and Southeast Asian

and B. A. I it\nisk\ Vostochny I iirkestaii r

19,S.S

Drernosti

1

Hudson,

of Vniuiihi. N'ortdii;

onJon; M.ieinillan.

I

C. W2

Ir.mk, A.

I

Ccntr.il Asi.i.

U

\

I

Cardlity of AmstcrJ.im: //.
1. 46: .After Stein, Aurel,

1.

V.uijiusuo, 1992,

miidi nui/aiig," Xin/iaii'i

199?, 'Shanshan Subcixi iiun.|un

16.41: 1996), table

(

larendon Press, plate CXIll,

12, fig. 2.

1994),

|.,

1961), 6",

Press, ?49. 47: After

Wnpu

luhitaii Di(.|U

Stem, Aurel, 1921, Seriiidu,

Oxford, N.

Maura (

l.uiii

'\m|i.iiig Sh.insh.ui Subashi gu

?8:

Klio, Beiheft 41, taf.

von Ciabain

'I

Xiii|iaiig Weiiv\

fig.

and Hicbert

2". 6?: .After Xm|iaiig

?, tig.

mu/.mg,"

40,

.\ni.\

fig.

uer Zi/hii|u Weiihuating

Pringle. 40: .After

48.2. 42: Junge,

Stiidien

VX'eivN

Pringle. ?9:

After Daniels and Bright

"1,

York, (tovmi

to xiikyii

62: After Xin|iaiig

Beaiitv) (1992), 1()\

linvj,jrn\i "1.1, 14.

Drawing by Maura vondabain (1961),

Publishing, 214-215. 20: After Harlcv (

I

fig. ".

1995), 258,

Kingdom

der

Dr.iwiiigb\ Victor M.iir.

?7:

1921, Seriiidhi, Oxford,

Anil, 196", Ihc Art of i:hme,

After

16, fig. \5. 78:

Dchamc-

Francfort (1989), 194, fig

14. 79:

Xinjiang Shehui Kcxueyuan Kaogu 'Xinjiang Alagou

"laniiiisuo, 1981,

mu

shu.xiic niiiguo

Weiufii

I, fig.

jianhao,"

faiiic

Photo Victor

5. 80:

Mair. 81: After Dehaine-lrancfort, (19S8),201,fig. 21.82: After Zlimiin, i99S,

ulniral

"C

An

complexes of

Age ni the larim Basin,' The Bronze Age dnd h.iily IvDti /Aye

in

Peoples of F.iistern Central Ayjkh Srednego Yeiiiseyii.

Q., 1984, 'The position of Tochariaii

Lm^iu's iiido-ctiroftceniies. 1

V V

135, fig. 8.7.

Drawing b\ .Maura Prmgle. Drawing by Maura Priitgle.

Pringle. table p. 288: .After

90. 168: After

fig.

:

1997, 'The epicentre of the Indo-

'Kin tocharischcr

19S6!, 66,

Temple 1986), 86, fig. T. 169: After Temple 1986), 72, fig. 46. 170: After Lu Liancheng (1993), 827, fig. 4. 171: After Ciening, V. K, (i. Zdanovich .\nd

m

site:

.Xinjiang archaeology: Idris .Abdursul, Sabit

Israfel Yusuf,

He Dexui, .Abdulka\um Hoja, Lii Wang Binghua, Wang Bo, Yu Zhang C^huan, Zhang Yuzhong and

Ping; for providing essential research materials: Flileen

Murphy, Xu Wenkan; for generous aid: James Buchanan, and for moral support and encouragement: Finiear Mallorv, Liching Chang and Thomas Krishna Mair.

,,

Niimlicrs to

att-r

it.ilicf

111

1

1

illiistr.itioiis; tluiM-

111

bold refer to coldiir pl.itcs.

Ahu

~S Ach.K-niciikU -/J Ad.inis, n. g. 22!, 2-?. 2''5, 2S4, 2SS-9, ill, li.ikr

14v

?9, I4>,

A\ kh.miiin 9s,

I4~.

15S,2,>6, 24S,252,

A/o\

260-1.263-^.294-5.

A/tecs }^

lil,ick-.\iiJ-\vliitL-

306-", i09. 312-13. 516-lS, !2i; ^"6, I4S.

luhj 14"

/s;.

B.ihvlon

;6>

;5,v,

Anglo-S.ixon 300

?

Bergm.in, h.lke 29,6?, lSO-2. 180. 185-".

19,

?2 -2 I

Anxi

{ii'c .ilsii I'.irilii.ins

51, s9

IS.ictr.i

Any.uig /70

46-", 51, 5\ SS-60. 69-"(), 94-8.

146. 236. 241-2, 24.S.

Aorsi

5S

104. 108. 125. 131.

294-6. >()6-9, ?14. ?16-1S; /vV, ;^9. I hi,

Ap.ir\t.le 46; /6

141.244.253.255,

95; ;9, ;2 14i.

Afnii.iscvo culture

41,46, 94,

Affih.inist.in

lOS,

9,S,

Apollo ?9, 44, 146 Al]su 6S.~2-3, 112; 2~, J2. ;5

;9,

160

5,

Ar.ib 44, 2S?, 321, ?2?

?>

I

li.ictri.i

li.ictri.i-M.ir.ui.in.i

Afne.i 246

Ar.ibie ^l.Sl, 111

Arch,ieoloj;ic,il

Ajinc.m 29. 2~4, 2S()-

Ar.ilSe.i

41,46,96, 10\ 131,262.264 Ar.tm.ue 102-5. UP-S,

Complex

1

VV^. V?4

v?V5

2'"4. 2S().

Aiini

2Sl.VVi—t

Ajinidcs.i

1

Aielinghu (sci' .il,isi.|i]e

-i;r

be.id

141, 152.

274. 276;

333 bicycle 323-t Bh.iruk.i

160,209

204;

255. 279, 300.

116, 122-3. 132. 165, 170-2, 193,200.

248-9, 252-+. 270-2.

Budim

163,239,287.

?16-|7; S4, 55

Btsan-po ???

Bl.i-ni.i

140, 154-5, 182,

bl.mket

213,215-16.229 Blood tvpe C) 245 blouse 220 Blumenb.ich, |. K 2?2 BMAC 131,262-3. 26S-", 269. 304-5. ?08-9, ?| 1-1 ?; j>l),

116, ?10

Bolton, |. n. 44-5, 146

)

;

Bugur "2; 27 Bukhara 100.254 Buku khan 101 Bulgarian 120 Burrow, Thomas 2"8-9 1", 299 Burushaski button 21" 1

By/ainine

C;ad()ta 68

P.

40, 42,

(^almadan.i 68

t^ambridge camel 41, 8", 9s,

1

159,

144 25, !", "5, 89,

I,

92, 126, 144. Is8, 196,

202,222; // liower, H.imilton 62 bowls 14 I'.r.ice. C:. L. 236 br.ichyceph.ilic 2?? lU.ihmi 15. 122, 2S4. 2"2; 49 1

?12-14

brick

120 14,

141-2, 14",

ri.

Bedouins 283

bridles

Behistun

Bnndisi 39 British I'lritish

51,

"?,"6, 85,

16. 123,

l?8-9, 143-+, 147,

i

269, 2"8

1

llni\ersit\

28

Bosh.m "4

bow

100

B\/ainium 99

Bore.is 44

IV

120,271; /9

?8

Bui; ?-

Caesar 272

l>l. 161

306;

92,

.S'9,

;44

Bishkent V.dlev 1?S Bi.ick Se.i 10.99. 106.

160,

105

???-(

Buddhist 10, "?,"s', ""-8,80, 82, 111-12,

Buddhist H\brid

212, 2?0, 245; 96, 9";

68,26

5.

Bishkein culture 265;

lUeton

Beilii

1

?21-2, 3??;2.S',

of Kror.in 140, 181, 18 5,20",

Beiim.u

1

?10, ?1?,314,

Bilg;i

bird

16, 9?,

4s"

"(s,

2^6,284,311-12.

100 (.J.n.m ??4 126, 1?9, 145

r>e.uit\

?04;

62, "0,

"9,9", 101, 104, 110.

Biig;i k.ijihan

be.ird

132, 14?-+, 147-8.

102, 1"0

1~5,222, 249, ,V J. \44

I5e/,;iklik

Bostiuir

68,92

18\ ?20;6O;

182,

1""',

S,V

192

135, 143. 146-^.

b.iskets

r9, ?00

i9

II.

2",

21?, 266, 269

Assyri.ius 259 132.

I'.uii

B.irkol (L.ike.i b.irlev

104

Bessemer process 321 Bevcesult.m 210

bonnet 216 boot 9, 16. 182, 189, 196,212.215-16,222,

r", 193.208.218-19, 228-9 B.irber,

^5. Ss.-S".

Assyri.t

s7i

69 B.m Ch.io 69,^9, 86, 95-6

r>.irher, Kli/.ibeth

5s

Asii

Avdinn L.ike culture 1?

1

/

1"4, ids

I'j.ilucliist.in

95 B.m'c.u 69 B.ui Vong S6 B.in Zh.U) 69

104

Avest.in

55

294; 52

B.in.u..iMi

if>

Buddhism

129.

299 B.iltic 120-1,285,288, 291-?; SO, s2 B.ilto-Sl.ivic 28S-6. 289,

h.imhoo

?3

>2,

88

222;

.i\e

Ainxim.tnder Andir 27S

2?

1

120-1, 129,

5s

259,266, 288; h.iluchi 256

10?, 120-1

1"1, 1"?,

Sanskrit 144

l?.\lts

AulusCelluis ?9 Av.ir 42,99,321 AirsLi 105, r, 25",

An.itolui 45. 129.

J

n

222, 299; ;oo, lOl Berlin Fthnologic.il

330-1,333;

?22. 328.

Buddh.i 2"0-|

14

1

B.in Bi.io

,itsehi

104

4

;

1

BMAC)

10,256,

U.uiev, H.irold

Athens 36

Darv.i 41, 5S-9. 100,'

li.ii

1

Ast.in.i

?cS

1

1,~4, S6. 88.94, 126, 144-5, 152. 154. 158-9, 162. IS", 196,

,lss

Altun 59 AKvighul 159, 16?. 2 r. 2?9. 243-4. 26H. ?l?,

?

B.iik.il

>l)

Ary.iiis

57,60. 164, oVi Alpine 2?2-i,2>6 Alt.u 29, 44-S. 99, 142, 146, iSK, 160, 20i-+. 2ll.2Vi-7, 241-2, 260. 294-5. ?06-9.

;

11)

Annhi

Am

9",

U.ictri.in

B.inhd.id

Aristotle 4~

Alnnorsi 5S Al.ins 106. 2?9, 260 Alb.inian 120-1; S2 Albert P. Slo.in Found.ition 12 Alc\.\ilder the dre.u 4"^, 59, 94-S. 100.

?I2.

15; 4/

I

Arim.ispMiis 42-1, 54. 203; 9, 12 Ariste.is M), 1 9-42, 44, 203

j/>o

!«('('

10,

(.•.((•

1

200-1.208.213-14,

Museum

262-3. 26S-9. 280-3, ?0I.304. 306-". 309. ?3?-l; /{, 24

broii/e ", 14, ss, ?6, ?9, 141, 14", 158-9,

162-4. 189. 196.269. >29; /"6

belt-pl.H|ues

2"4

1"0

brocide ~s |

?21; ;29

10,281

1

Museum

liritish

120

38

Ik-lsk

li.Kon, hr.incis

"9

"?,"8

/i,7>7./

BeiortissMii

belt 75, 21s. 21", 228,

l'..icliu

160, 162;

•iiiim.il style

1114

106

Se,i

>(14-S.

14

109 Libr.ir\

1"4

167,260.266.

cannibal 45 cannibalism 40

cannon

''ill

cap "5, 108, 18s"-6, 193, 210.213-14, ?()6;42, 11(1

C;ape Verde 47

card l(yH carpet 70,80, 137. (iarpini. (iiovanni da Plan del "I c.ishmere 2I~ C.ispian Sea 45-6. 48.

58.69.84. 107, 131. 164,239.256.259, 262-3. 304. 3 17; 4 5.54 (Catacomb culture 239. 294

.M7

1

'

1

Index

41-2, ~?,76. 87,

c.ittlc

lU,

92, 126,

143,

l,?S,

154, 158-9, 184,249,

260, 262, 266, 33

Caucasian {see iilso Caucasoid) 232, 237, 246,249,251; 147 Caucasoid 16,25,36, 44-5, 55, 63, 70, 92-3, 95, 136-7, 141-2, 147, 156, 158-9, 161^, 173, 190,229-31, 236-7, 239-40, 243-4, 248-50, 256, 258, 279, 295, 299-300, 306, 326, 329; 143

Caucasus 84, 106, 14?. 211,218-19,229,256, 281; 37

331,333; /,S', ,S',V, I4\ t7S, 177 C^hinchorro 205 Chinese 30,44,71,86, 109-11, 118, 120-1,

164-5,236,252,

123,

C^hiou-I'eiiii, T/elnie\

329

Chokkur

145

ChongBauh

158, 163,

6"

335;

Choresnna

113,'

Cavalli-Sfor/a, L. L.

(acero 272

Chu

76. 126, 134, 159, 162

254

331

1,

31^

288, 289-92, 331; 50,

coat 16, 138, 196,201.

censers 294

centum 121,233,285;

no

110, 12H,

Dog Roiig s3 dolichocephalic 233 doll 187,201,222; /33 Don 37-8, 106,241,292 Dongxiang 249-50; 145 donkey

ceramics {see pottcrv)

Dravidian 259; /47 dress 48 Duldur Aqur 274 Dunhuang 51-2,58-9,

Cockburn, Aidan 32 coins 59,62,74,78,

260, 282, 306

283,322,334; U, 67, S3

s,

61,91

Four Fwgr VV?

Mahdv, Christine

Frisian

Hao

Fu

P.

1

{see also

Khocho) 101 garrison town {see also Four Garrisons) 155-6 Gaulish 299 Geloni 38

Gelonus 38 German 120,279

Germanic 41,99,

120,

Germans

8, 109, 155,

266,288-9,291; ;56 Getae 99 ghee rGimhutas, Mania 128 girdle 79,88 glass 202 goat 126, 138, 140, 144, 159-60, 209. 2 r, 224, 260. 262, 266 Gobi 84 gold 43-5, 53, 62, 64,

74,89,92,97, 126, 158, 160, 178-9,203; 79

Gomati 79 Good, Irene

27, 208,

217-19 goose 126, 181 Gothic 119-20 Goths 99 gourd 155, 164; 74

gown

155, 196

Graeco-Scythian 171 grain {see also cereal)

36-7, 56, 173 grape 57, 164, 169,333 Great Scribe's Records {see also Shi/i) 56 Great Wall 58,-1,84-5,

Greece 129,215,224; 10 Greek 30,36,43,45,55, 94-5,97, 102, 104, 107-8, 119-20, 146,

233,286,288-9,291,

120 14"

314, 333; 50, 52, /55

36, 80, 88-9,

r,

144,

Gaochang

89, 113 12, 16,

Fukang 68 fur

158,242,250,

147,

254,281,283,304-5, 328, 333; 57

181-2,212,217;

IM) Further Beilu 68 Further Jushi 68 Further Pulei 68

Greeks 34,37,48,53, 96, 106,259,266,288; 9,

;56

griffm 43-5,53; ;2, /5 grinding stones 159

Griinwedel, A. 63, 171, 173

Guchung 68

331

elephant 4^,53. 81,

/6

foodtray 200 Four (iarrisons 70, 80, 14\ r,4

25,246-7; 141 France 209, 232 French 109,284

EclcLi

112

76

flute

147, 155, 169,207-8,

Ebi-nor 68

Daiidan-Uifiq

159,203,209

flax

Francalacci,

330,

>9, 17:;,

262,327

41, 126, 145,201

122-3, 131-2, 134-7.

214,230,236-7,240,

Cvrus the Great 46, 108 Czech 120

1

fire altar

fish

Fleure, Herbert 259

102

133

139, 141, 144,

Finns 247

{see j/so

jungghar Basin and Yarish Basm) 146, Mr D/iingarian Gate 44-s

cup 200,202 Cyclops 43

dagger 126, 248 Dahe/hiiang

figurine

152; 62, 7/

D/ungana

cowrie 141 crane 126 cranial index 2? 1-2; 139 craniometries /36 Ctesiphon 14

Dacian 99. 299; 50 Dadicae 46

181,

8,

94, 100, 133-5, 142,

glove 217; /30.

185-7,213 16,44,70,74,80,

feather

333

165,212,222

152, 155,

175,248-9;

;4J

3 12; 24, 161

earring 189, 192 tast Central Asia 9, 29-30, 33-5, 63-4, 66,

128; 53

IX

eye colour 49, 70, 93, 125, 156,

Gansu 56,60,80,92,

124,284,286,288-9, 291-3,314, 331; 50, 52

142, 147, 164, 169,

Corded Ware

Dadiwan

Ksarhaddon 91 Eshek-bashi Ola 74 Kstrangelo 41 F.truscan 102, 124,299

211,255,304,309,

Dzjni-filin^-r^Yds-hshjJ

Cyrillic

Kriitou 328

dve 212.218,225 1

269, 2X4, 296, 303-4,

348

/9, 2?.

322Constantinople 14 Conway, R. S. 234 copper 74, 126, 136, 139,295,327-8; I7.i copper oxide 74 Coptic 110 3 9,

1

170, 171

Kratosthenes 47-8

Dushanbe 107 Dutch 120

1

Chaghan-tungge 68

302;

s4,

1

225

compass

126,313 Chadir 72

cereals

Chang'an

14, 140, 152,

F.pidauros 95

99, 111, 113, 174,254,

95-7, 193,254, 283;

comb

enihalming 29 Hndere 112 Kngland 300,319 Faighsh 120,236,279, 284, 300 ephedra 138, 152, 185-7,200,262.269, 305-6

62-3,68,81,86,92,

Columbus 48

iilso

142, 146-7,

134, 143, 147,

266, 328

Colosseum 48

5/

81,

1

215,217,220,248;

Celts 8,32, 155,218-19, 222,235,266; (56

Dnieper 106, 129,241. 263, 325 Dniester 106

119-20, 124,286-",

52

?4, \^9,

dog

Claudius 49 Cleopatra 48 cloak 75, 144.212-13, 222, 225, 229

Celtic languages 41,

1

Vacher 233 Delphi 34 Devoto, Giacomo /56 DhjiiuujpJilj 253 Dian 329; 20 Dionysius 108 Discovery Channel 14 DNA 11-12, 16,28.32, 143,230,245-8;

114

Cimmerians 28 170, 172-4;

KP

96. 1()\

('horesmi.ins 46,256; ;6, 4) ChristianitN ~0, 101,

no,

9, 126,

209,215-16 de Guignes, Joseph 62 de Lapouge, Georges

141

C^hristians

28 cave shrmc 93, 94

299,

310,333-4; ;47 Chinese Turkestan 9

cauldron 45, 162-3,331; i77 12,

283^,

262, 272,

deer

ri 52

Gaelic 233

Guhu

Galcha

Can

2V-,

Gallo-Roman Ciamkrelid/e,

2M-,

lomas

68; 26

0\us)

{see also

53,

58-9, 95 (iitizi

322

Guishuang 95 Gandaridae 46; 16 Gandhara 80,97,

ri-2

Gumo

68, 73; 26

Gunuigou

{see also

Qiiwrighul)

Gandhari 285

gun 322

Cianfu 56 Ciangcs 4^-8, 94, 96; ]9

gunptmder Giishi

136,336

il9, i22

S8.60, 99. 143-4

^

.

Index

Hephthahte "9-80,

GiitLMibi.Ti;, loh.iniK--.

^S-^;

.522

Guti

Herodotus M). 35. 3--9. 40-1.44-5.47,88,92,

W, 2S1-2

Hai ^^\

98, 102, 106, 108, 163,

colour ~, 10, 16, 49,93, 125-6, 141-2,

h.iir,

156, 159,

17.?,

260,328; 12. I\2I Himal.ivas 4''. 281,

r^.

200-1,204, 230-1,

Hindu Kush 36,4", 268;

145,304 211,218-19, 229, 235 Ha mam "4 Q.ir.iJoug)

hook

328. 329;

Mummy

Han,

(Jviiastv

188; 10

55-7.60-1.69.73. 75-6. 84. 87, 89-95, 126, 128-9, 134, 138, 143-4, 147, 152, 154, 157-60, 162, 167, 169.

i

30, 55, 57,

59-6 66,69-70, 72, 75-6,78,81,85,87, 9(K?,96, 99, 118, 1

,

177.211,223,257,

136-", 152, 163^,

260, 262, 264, 266,

ro, r~,

i6~, 169,

201,20-, 211,230, 243.253,283,300,

303,318,329,331,

My,

V

horse 25,37,41-2.45.

16,92, 140, 14'^, 217, 219, 237, 300; 5.S' 13.

Hami

10, 14

horn 153-5, 192,22s,

Qumul)

{sec ^iho

68; 26 H.wshii 5^, 6~, 68, 69, "2, -4, —-8, 88, 91, 95, 152

haotiui

262

138,

Harappan

culture 212,

259, 304; /47

hat (see dlso cap) 25, 32, 88, 162, 182, 196,210,

9. 41

65, 133

Huaiigniangniangtai 134"

—

Hubei Hui 249-50; Huisheng 82

Hunan P" Huns 30, 55,

>5, 4';

Hecataeus

Ching

Iberia

'

10

92, 155

99 124

Iberian 299 Icelandic 120

(62

Hemphill, Brian 244. 268.313 henna 248 Henning. W. B. 281-2 Henrv luce Foundation 12'

Indo-Scvthian 171 Indra 257 Indus 59,96,244 Indus civilization 33,

174,212,259

1

ls4, 21s"

lar

jaxartes 41,49, jiaugbin 75 jianshi 95

s'S,

j in

dynasty -?, 202; IX

in Vallev

55,91-2.94,

lllynan 299; 50

Imammusakazim 200 Imaus 47,281. r>? Imit

13. 14

Incas 33 India 31,45,47.53.70,

-5,^8,80,82-3,97-8, 103, 109, 114-15, 118. 122. 125, 131, 165,

1~, 180,211-12,215, 224, 255, 25", 259, 262, 26-, 2^8-9,

iilsi)

Kum.irajiva) 75, 81

liuquan 60 Jones, William

119-20,

123

of IndoEuropean Studies 27 10 lud.usm jumi 68; 26 jungghar Basin

315;

14,

lute

also

?()-,

?12,

20?

262-3,282,331, 111

Iranians 41,91,96-7, 10(1, 104, 106-9,

Kamberi, Dolkiin 2~

Kangiu 58-9 Kaniska 96-7

Kasia 51

299,302,310-11,

Kazakh 249- s'O; ;45 Kazakhstan 135, 139,

Ireland 50.309 8.

120,236.2-6,

288 iron 74,91, 141-2, 144,

147-8, 152, 158-9,

162^, 167, 196,311, 319,321,328 irrigation

135, 142, 169,

262-4, 266, 269, 304,

309,311-12.314, 317-18 River 41 Islam 41,62,^0,81,

Iset

110,2-3 Issedones 40-?, 84,92; 9, 12

Issedon Scvthica 41

147, 158,

263,

U Museum

246; 6

Kroraina (see also Kror;in) 65,81,278 Kroran 30,58,60,63, 65,68,79,81,84-7, 99, 116, 132, 1?7, 140,

184-5,201,222,237, 239, 244, 253, 268.

2-8,299,301,313,

314-15 81

Khan

81

Kucha 30, 52, 60, 63-4, 69,72-5,77, 122, 145,

146

282-3, 285, 288, 294,

174.263

19,

Kror;in)

Kashgar (see also Qashqar) 64 Kashmir 96, 304

114.

K6k-turaq 160;67, .VO Konchi River 137,212 Koiigque River (see Konchi River) Korea 99 KorLi 16,25,52,68; 5,

Kublai

124-5,204,253,256, 259-62, 266-8, 277-9,

?14-16, 318, 326. 331. r,?-l;-/3. 50. 52. ;55

248, 270; XII

Kror.iyina (see also

153; I4h

K.irasuk

knights with long swords 171-2, 175,

Kror;inian 29. 123, 2-8-9, 300, 302,

125, 131, 139, 1-4,

209, 220, 248, 259,

IM

?30;

318; 19.24. 2(y,SS, S3. 95. i33. I.iS. ;53. ;60

5,V

144, 16-; 26

jushi

158-9,325,328,

knife

144, 164-5, 181, (,iir>hil

Achaeology

1

162 Kirghiz 100

Kokonor 40

68; 26

lie

'I'arish

sl,-{l, 104,

Khruschev, Nikita 232 Kidantes 98 Kikkuli 257

knitting 214

165

236

(Llrumchi) 2-. 246;

100

1

Inner Mongolia 269 Institute of Archaeology

Irish

243, 255, 283; H2

84-6, 14S, 165,200, 208, 236, 294 hemp 1^8. 20?, 209, 262, 294, ?05-6, ?21;

;55, ;56

Il-baliq 281

HeDexRi

10, 63,

1

Iraq

139, 160, 162,239,

Hedin, Sven

jade Gate Pass 60, i04 jainism 9japan PI Japanese 14, 1(N, 11

liaolie

Indo-Iranians 32, P, 125, 129-30, 138,228,

Iceman isee jlso Ox/a) 11,228,24-

Hecvka ^^^ 155, 178, 193

/56. /57 Indo-Greek 94-5

111-13, 116-18, 121,

'l46

latioi

headdress iO headgear 229 'He.ivenlv Questions' ?30-l Hebrew 102-3

50. 52, 53. 55, i55.

Iran

58, 62, 98,

Hurrian 257, 259 Hvperboreans 39, 44,

/

219,228,232-3, 258-9,265-7,270-1, 281,287,290-1,292. 293,314.316-17.331;

Institute of /-/s'

;;;, 124. 12s. 126. 127

hawthorn 126 Hazaras 41 headband 22

120,

122-4, 126, 129, 131,

(Bei|iiig)

1

214, 220, 321; 66, X2,

?30

;

312-18; 52, 56, 14~,

Hunve 92 Hunza 117

harp 29

50 Indo-Kuropeans 33, 1

lade s8. -8-81, 8s', 138-9, 14-, 322, ?28,

158

106, 239; 35

hare 126 harness 321

Ivrkae ?S

JinshK 73 jiumoluoshi (see

Huanghe

l>S

kilt

252-3, 256-7, 259, 262, 265, 267-8, 288, 33

95,

299; 47

haiiov, VvacliesLu

Ivolga 88

17-18,

1

,S'3,

254, 272, 278, 284,

299

291,293-5,303,305,

Hou Can 86, 165 HoumaCity 190 Huang Wenbi 63

Hanmi

/56

265-8, 285, 287, 289,

horseriding

141

313; /J7 Indo-Arvans

67,

60 Khotanese 78, 81, 112, J

Italo-Celtic 286 Italy

U,

24, 30,

289, 291-2; 50, s2,

77

76.

249 KanfjvMi 236-tO, 242-4; / r, ; 5.S', HI)

97, 121,2-1,

Mi;

88-9. 141,23', 240,

Han

Arvaii")

266, 287, 288,

jingiue 68; 26

86; IX; ethnic j^roup

hitiifilii

300. 302.315, 334; 19,

Italics

233,252,257-61,263,

horse-bit J,S'.

Italians

Indic [see jIso 'Indo-

269, 273, 276, 287-8. 294. 324-5. 328, 330,

horse-gear 162 horsemen 328-9; /67

9, 22. 24. ?7.

239, 243, 253-5, 278,

160,281,333 275,321

121-2^ 124-5, 131,

17

i,

120,25HIA 245 hood 212,222

H.illst.itt

Hami

18s,

Hittite

jIso

{si'f

linduism 9". l"!

I

245, 248-9; cut "5, -9; J;;X1I

164. 16", 1-4, 200,

Issvk kul

237,243,268,301-2,

Hindu 114-1\ 180

Issedon Seni.a 41,8s

531,333; /6 Indians 96; 9

277-8, 284-5, 302, 310, 333-4; 52 Indo-Afghan 238-40,

333

181-2, 1S4, 1S6-9, 191, 193-4. 196,

H.ii.idun

?(10-1, ?04, 312, 321,

)'9

169, 171-2, 175,203,

239, 248, 272, 274,

160,236--,

293^, 308-9; 164

Ke'ermuqi (see also Keremchi) 146,307 Kellv, William 321 Keremchi 146, 30--8;

276-7,303,318, 333-4; 19,24,26,27, 2,S', 32, 58,67,68,