Wrapped in Beauty: The Koelz Collection of Kashmiri Shawls 9781949098792

Table of contents :
Contents
List of Tables and Illustrations in Part I
Preface
Acknowledgments
Part I: Kashmir Shawls in Context
Chapter 1. Introduction
About the Koelz Collection
The Organization of this Work
Chapter 2. Walter N. Koelz and His Collections, by Carla M. Sinopoli
A Biographical Sketch
The UMMA Koelz Collection
Chapter 3. "Turbans" (Patkas)
Chapter 4. Kashmir and Persian Shawls in Their Historical Context
The Historical Background of the Kashmir Shawl
Early Development: Seventeenth and Eighteenth Centuries
Nineteenth-Century Production
The Koelz Collection in its Broader Contexts
Parts of a Shawl
Size and Shape
Weight
Shawl Materials
Understanding Weave
Producers and Consumers
The Production Process
The Loom
The Talim
Working Conditions
Expanding Demand and its Impact on Production
Chapter 5. Dissecting the Kashmir Shawl, Part 1: Piecework and Structural Proportions
Piecework as a Time-Saving Device
The Needlework of Piecing
Piecework Structures
Rectilinear Piecing in Long Shawls
Piecing in Square Shawls: Moon to Quatrefoil Format
A Multi-Moon Design Layout
A Quatrefoil Square Shawl
Other Variants on the Moon Shawl
The Elaboration of Fringe Gates
Changing Structures: Increased Figure Area
The Influence of the Jacquard Loom on Shawl Design
Chapter 6. Dissecting the Kashmir Shawl, Part 2: Decorative Motifs
The Boteh Motif
Symbolism of the Boteh: A Ubiquitous Natural Shape
Radial Flower Motif
Shared Motifs
Beading
Red-Flower-Green-Leaf Bel
Wishbone-Rose Border
Four-Fragrant-Flower Repeat
References Cited
Glossary
Part II: Catalogue
Turbans/Patkas
Long Shawls
Long Shawls Woven Whole
Yardage Long Shawls
Field Gallery Long Shawls
Block Pieced Long Shawls
Reworked Long Shawls
Strata-Field Long Shawls
Square Shawls
Moon Shawls Woven Whole
Moon Shawls Woven Whole except for Attached Hashias
Moon Shawls Pieced across the Middle
Square Shawls with Broad Field Gallery
Piecework Square Shawls
Strata-Field Square Shawls
Pieced Conglomerate Square Shawls
Square Machine-Woven False Kashmir Shawl

Citation preview

~

.

'

"')~

'I. e.

....'I). -" _.\

"\

~

1).

~

:'IJ.

\ ...

'"

'\

\

, ' .

'\ ~ I

~'\

.1).-'J."

~

SAUDIA ARABIA

. /' / -, -- ---, T K U WAC

'l)

_

~. \.

""''1).

Kashan

-Tehran."

1).

..,

.:}:A),,'"

1).

"" \-Sh"az ,\ .

.

'l!.

,~, , _,

'"

~ ""~ . ~1).>.'l\,-.,.~

.-

~

~>..

~').~~,. ...

j;:\ -;,.,N

\

.,

\.

/ /

i

l

i l

./ .,- . ...r.

L'

PERSIAN GULF

i

~

,

~

)

.. \. ~ \ "........"'\

{

/

.-.,

')

\

..-

j

-- 0'

i

/./'

E

E

-Bokhara

PAKISTAN

- Herat ",

i'... . .

•

i

'-')

......

.( .

'1.1)

,... c.:

-\ .

-

'\

~ ).

' .\

)'\~ I).,

","

,~

'\.

Meshed -

"

" .

'\.-;;:..

'\"'~

......, .-. \..

-Kerman 'J..", _-\-""\\

1).'"

').",-' "•

~ ~ .. ~~ ., ~>. i~~~ -Yezd ." -~~ " '.. ~ '1\.~_~ ~ -'ll.

.......

~

_~;~~~.

/

IRAN

.?:. ~

i'

,/

~ .

"1-'\.7 '\. , ." '\.... ).." .' { '}.. , - Isfahan

~'\....

.., . ~

~

':"\..\ \.: '\

'.....

.,., \'\ 1).~\ \'\

Basra- ('\

IRA Q

\..~'.

l

-Tabriz

,.,1). 'l!. ..,......1i.~ - ' 1\ X-'"\~ ~ ..... "'\. ~ .,-> '

:'~

j

-hdad\ ". I Bag .........

(/

'"

.

'-. ,.

\.;"

'\

\ "''1.,,-

~

~'-', ......(,..., ~" lo,

~

~,

.,."l\~

".,.;

INDIA

Delhi-

_Agra

'9"

_Yarkland

Anthropological Papers Museum of Anthropology, University of Michigan Number 93

Wrapped in Beauty The Koelz Collection of Kashmiri Shawls

Grace Beardsley in collaboration with Carla M. Sinopoli

Ann Arbor, Michigan

2005

©2005 by the Regents of the University of Michigan The Museum ofAnthropology All rights reserved Printed in the United States ofAmerica ISBN 978-0-915703-60-9 (paper) ISBN 978-1-949098-79-2 (ebook) Cover design by Katherine Clahassey

The University of Michigan Museum ofAnthropology currently publishes three monograph series: Anthropological Papers, Memoirs, and Technical Reports, as well as an electronic series in CD-ROM form. For a complete catalog, write to Museum ofAnthropology Publications, 4009 Museums Building, Ann Arbor, MI 48109-1079. Library of Congress Cataloging-in-Publication Data Beardsley, Grace, 1913-2003. Wrapped in beauty : the Koelz collection ofKashmiri shawls/ Grace Beardsley in collaboration with Carla M. Sinopoli. p. cm. -- (Anthropological papers/ Museum ofAnthropology, University of Michigan ; no. 93) Includes bibliographical references. ISBN-13: 978-0-915703-60-9 ((13 digit) : alk. paper) ISBN-10: 0-915703-60-2 ((10 digit)) I. Shawls--Iran--Catalogs. 2. Shawls--Jammu and Kashmir (India)--Catalogs. 3. Koelz, Walter, 1895--Ethnological collections--Catalogs. 4. University of Michigan. Museum ofAnthropology--Catalogs. I. Sinopoli, Carla M. II. Title. III. Anthropological papers (University of Michigan. Museum ofAnthropology); no. 93. GN2.M5 no.93 GT1420 391.4'4--dc22

2005018763 The paper used in this publication meets the requirements of the ANSI Standard Z39.48l 984 (Permanence of Paper)

Contents List of tables and illustrations in Part I, vii Preface, viii Acknowledgments, x PART

I:

KASHMIR SHAWLS IN CONTEXT

1. INTRODUCTION, 2 About the Koe1z Collection, 2 The Organization ofthis Work, 4 CHAPTER 2. WALTER N. KOELZ AND HIS COLLECTIONS, BY CARLA M. SINOPOLI, 6 A Biographical Sketch, 6 The UMMA Koelz Collection, 8 CHAPTER 3. "TURBANS" (P ATKAS), 15 CHAPTER 4. KASHMIR AND PERSIAN SHAWLS IN THEIR HISTORICAL CONTEXT, 20 The Historical Background ofthe Kashmir Shawl, 21 Early Development: Seventeenth and Eighteenth Centuries, 21 Nineteenth-Century Production, 23 The Koelz Collection in Its Broader Contexts, 24 Parts of a Shawl, 25 Size and Shape, 25 Weight, 29 Shawl Materials, 30 Understanding Weave, 35 Producers and Consumers, 40 The Production Process, 40 The Loom, 40 The Talim, 42 Working Conditions, 45 Expanding Demand and Its Impact on Production, 47 CHAPTER 5. DISSECTING THE KASHMIR SHAWL, PART 1: PIECEWORK AND STRUCTURAL PROPORTIONS, 50 Piecework As a Time-Saving Device, 50 The Needlework of Piecing, 51 Piecework Structures, 52 Rectilinear Piecing in Long Shawls, 53 Piecing in Square Shawls: Moon to Quatrefoil Format, 53 A Multi-Moon Design Layout, 55 A Quatrefoil Square Shawl, 61 Other Variants on the Moon Shawl, 64 The Elaboration of Fringe Gates, 66 Changing Structures: Increased Figure Area, 71 The Influence of the Jacquard Loom on Shawl Design, 72 CHAPTER

iii

6. DISSECTING THE KASHMIR SHAWL, PART 2: DECORATIVE MOTIFS, 76 The Boteh Motif, 76 Symbolism of the Boteh: A Ubiquitous Natural Shape, 83 Radial Flower Motif, 84 Shared Motifs, 84 Beading, 88 Red-Flower-Green-LeafBel,95 Wishbone-Rose Border, 95 Four-Fragrant-Flower Repeat, 96

CHAPTER

References Cited, 97 Glossary, 100

P ART II:

CATALOGUE

TURBANSiPATKAS, 104 UMMA 17387,104 UMMA17313,106

107 Long Shawls Woven Whole, 107

LONG SHAWLS,

UMMA 17312, 108

Yardage Long Shawls, 11 0 UMMA 17314,110 UMMA 17337, 112 UMMA 17339,114 UMMA 17348, 116

Field Gallery Long Shawls, 119 UMMA 17328, 120 UMMA 17327,122 UMMA 17333, 124

Block Pieced Long Shawls, 126 UMMA 17320, 126

Reworked Long Shawls, 128 UMMA 17310, 128 UMMA 17305, 130 UMMA 17334, 132

Strata-Field Long Shawls, 135 UMMA 17321, 136 UMMA 17309, 138 UMMA 17307, 140 UMMA 17335, 142

144 Moon Shawls Woven Whole, 144

SQUARE SHAWLS,

UMMA 17317, 146 UMMA 17323, 148 UMMA 17308, 150

iv

Moon Shawls Woven Whole except for Attached Hashias, 152 UMMA 17330, 152 UMMA 17342, 154 UMMA 17344, 156 UMMA 17304, 158 UMMA 17316, 160 UMMA 17324, 164 UMMA 17311,166 UMMA 17332, 168

Moon Shawls Pieced across the Middle, 170 UMMA 17306,172 UMMA 17326,174 UMMA 17319, 176

Square Shawls with Broad Field Gallery, 179 UMMA 17325,180 UMMA 17340, 182 UMMA 17343, 184

Piecework Square Shawls, 186 UMMA 17369,187 UMMA 17318,190 UMMA 17345, 192

Strata-Field Square Shawls, 195 UMMA 17329,196 UMMA 17315,198 UMMA 17341, 200

Pieced Conglomerate Square Shawls, 202 UMMA 17322, 202 UMMA 17331, 204 UMMA 17336, 206 UMMA 17338, 208

Square Machine-Woven False Kashmir Shawl, 210 UMMA47325,21O PART

III:

LIST OF COLOR PLATES BY SHAWL TYPE,

COLOR PLATES

213

UMMA 17387. Silk-embroidered cotton turban, 215 UMMA 17313. Fine creamy white wool turban or patka, 216 Detail of palla (UMMA 17313), 216 Detail ofboteh motif (UMMA 17313), 217 UMMA 17312. Woven wool shawl, 218 UMMA 17314. Yellow wool shawl, 218 UMMA 17337. Striped pashmina shawl, 219 UMMA 17339. Red and white striped wool shawl, 219 UMMA 17348. Medallioned ivory-white shawl, 220 UMMA 17328. Parchment-white wool shawl, style of 1820s, 221 UMMA 17327. Striped wool shawl, style of ca. 1825,221 UMMA 17333. Floral striped long shawl, style of ca. 1825,222 Detail of palla and color tab fringe gate (UMMA 17333), 222

v

UMMA 17320. Red wool shawl, 223 Detail of mihrab fringe gate (UMMA 17320), 223 UMMA 17310. Black wool shawl, 1820-1830, 224 UMMA 17305. Blue wool shawl, second quarter of the nineteenth century, 224 UMMA 17334. Striped wool shawl, possibly mid-nineteenth century, 225 UMMA 17321. Red wool shawl, early nineteenth century, 225 UMMA 17309. Wool shawl-coverlet lined in cotton sateen, 1820-1830, 226 UMMA 17307. Red wool shawl, mid-nineteenth century, 227 Fringe gate of pieced color blocks (UMMA 17307), 227 UMMA 17335. Red wool long shawl, mid-nineteenth century, 228 Two detail views ofUMMA 17335,229 UMMA 17317. Floral-striped moon shawl, perhaps 1800, 230 Detail of corner quarter-moon, end border, and fringe (UMMA 17317), 230 UMMA 17323. White wool moon shawl, style of ca. 1815,231 UMMA 17308. Blue and white wool moon shawl, 231 UMMA 17330. White moon shawl, early nineteenth century, 231 UMMA 17342. Ivory wool moon shawl, early nineteenth century, 232 Detail of end border and field of shawl, UMMA 17342, 232 UMMA 17344. White wool moon shawl, ca. 1825,233 Detail of end border and fringe (UMMA 17344),233 UMMA 17304. White wool moon shawl, possibly 1825,234 UMMA 17316. Brick-red moon shawl, early nineteenth century, 234 UMMA 17324. Red moon shawl, ca. 1815, 235 UMMA 17311. Red wool, multi-moon shawl, early nineteenth centUly, 235 UMMA 17332. Ikat medallion, wool shawl, ca. 1830s, 236 UMMA 17306. Red wool moon shawl, ca. 1825,236 UMMA 17326. Red wool moon shawl, style of ca. 1830, 237 UMMA 17319. Striped wool moon shawl, style of 1825-1830,237 UMMA 17325. Amber yellow and black wool shawl, style of second quarter of the nineteenth century, 238 UMMA 17340. Crimson and white wool shawl, mid-nineteenth century, 238 UMMA 17343. Embroidered wool shawl, ca. 1830,239 UMMA 17369. Crimson wool shawl, ca. 1870,239 UMMA 17318. Piecework shawl, third quarter of the nineteenth century, 240 UMMA 17345. Red wool shawl, third quarter of the nineteenth century, 241 Detail ofmihrab fringe gate (UMMA 17345),241 UMMA 17329. Red strata-field shawl, after 1820s, 242 UMMA 17315. Red strata-field shawl, post-1830, 242 UMMA 17341. Red strata-field shawl, post-1830, 243 UMMA 17322. White wool square shawl, mid-nineteenth century, 243 UMMA 17331. Yellow wool shawl, field motifs of early nineteenth century style, 244 UMMA 17336. Red wool shawl, post-1825, 244 UMMA 17338. Wine-red wool shawl, mid-nineteenth century or later, 245 UMMA 47325. Brocade, floral striped, black wool, square shawl, 246

vi

TABLES AND ILLUSTRATIONS IN PART I TABLES

2.1. 4.1. 4.2.

Koelz and Beardsley shawl classifications, 12 Cashmere goat hair compared with other animal textile fibers, 32 Quantities of shawls shipped from Indian cities in nineteenth century, 48

FIGURES

3.1. 3.2. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. 4.7. 4.8. 5.1. 5.2. 5.3. 5.4. 5.5. 5.6. 5.7. 5.8. 5.9. 5.10. 5.11. 5.12. 5.l3. 5.14. 5.15. 5.16. 5.17. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9. 6.10. 6.11. 6.12. 6.13. 6.14. 6.15.

Soumak technique as used in weaving, 17 Radial flower, late eighteenth-century motif, 19 Parts ofa shawl, 26 Shawl sizes in the Koelz collection, 28 Range of Koelz shawl weights, 30 Incidence of woven colors in principal areas, 35 Representative thread counts of Koelz shawls, 37 Two/two twill double interlocked tapestry weave, 39 Analysis of a turquoise floret with yellow center and black outline, 43 Part of a Kashmiri talim, 44 Partial field portion of Persian rectilinear block pieced long shawl, 54 Multi-moon layout of square Kashmir shawl, 56 Piecing diagram ofa Kashmir multi-moon shawl, 57 Design piecing system for one comer of a piecework shawl, 58 Layer weaving of shawl borders, 60 Piecing plan of a Persian multi-moon derivative shawl, 60 Piecing of different comers of a single shawl, 61 Proposed weaving plan for piecing parts of Persian shawl, 62 Partial piecing detail of a quatrefoil shawl's field comer, 63 Small, irregular pieces to be cut out and incorporated into piecework shawls, 65 Design development in nineteenth-century Kashmir square shawls, 67 Knotting of fringe gates, Balti long shawl, northern Kashmir, 68 Mihrab fringe gates, 69 Diagram of a single repeat in the fringe gate of a Persian shawl, 70 Scheme of fringe gate on a late nineteenth-century quatrefoil shawl, 70 Micro-threading draft illustrating point repeat, 73 Jacquard motif, 74 Celery botehs of the 1860s-1870s, 78 Mughal and Persian plant forms on seventeenth-century textiles, 79 Conventionalized seventeenth-century plant forms, 80 Transitional versions of the boteh composition, 81 Development of the boteh: designs evolving toward vase and dish, 82 Essentials ofyun-tsai-t'ou design from carpet and shawl borders, 84 Variations of the hooked vine motif, 85 Burgeoning hooked vine runners, 86 Some variants of bugle beading found in guards of palla tanjirs, 87 Thickened bugle beads in a rhomboid shape, 88 Some hexagonal beading forms, 89 Beading on early seventeenth-century hashia guard, 90 Yellow "birds" oftanjir guard flying with outstretched necks, 91 Residual beading of the later nineteenth century, 91 Wishbone-rose motif, 94

PLATES

2.1. 4.1. 4.2.

Walter N. Koelz, 8 Himalayan goat, 31 Mr. John Mohammed Sheikh demonstrating shawl weaving, 41

vii

Preface

S

tudy of the Walter N. Koelz Collection in the Asian Division of the University of Michigan Museum ofAnthropology (UMMA) has benefited from the efforts of two talented scholars over recent years. The work of the first, Carolyn Copeland, resulted in an exhibit and catalogue of the Koelz Tibetan and Ladakhi Buddhist monastic thangka paintings (Copeland 1980). Research by Grace Beardsley on the textiles in the collection has yielded the present publication. As the current curator of the Koelz collection, I can only express my profound gratitude and admiration for the efforts of these two dedicated volunteers in studying and sharing their knowledge concerning Walter Koelz and his remarkable collection of South and Central Asian artifacts. The present volume has a long history. In Chapter 2, we summarize the life and collections of Walter Koelz and review how the collection's extraordinary objects made the journey from Asia to Ann Arbor, Michigan. Chapters 3-6 focus on the artifacts themselves, specifically on the twill tapestry wool shawls produced over several hundred years and spanning a remarkable period in Indian and world history. These chapters comprise only a portion of Beardsley's study of the Koelz textiles; the remainder of her efforts must await future publication. Sadly, Grace Beardsley passed away before she could see the efforts of her decadeplus work on the Koelz textiles come to publication. Her work on the textiles began in the early 1980s when Karl Hutterer was curator of the Museum's Asian Collections. A self-taught textile scholar, Grace closely examined each of the 1DO-plus textiles in the Koelz collection, recording detailed information on construction, thread counts, color, and design. She also conducted extensive background research on the objects, and sought to understand them in their broader cultural and historical framework. Thus, her approach to the textiles was that of both a textile specialist and an anthropologist, not surprising given her prior education in decorative arts (University of Chicago) and anthropology (as a student of Alfred J. Kroeber at the University of California, Berkeley). Grace's route to our museum was circuitous. While at Berkeley, she met and married Richard Beardsley, a cultural anthropologist specializing in Japan, and they moved to Ann Arbor in 1948 when Richard took up a position in the University of Michigan, Department of Anthropology. The Beardsleys raised two daughters in Ann Arbor, making frequent trips to Japan. It was not until after Richard Beardsley's death in 1978 that Grace began her study of textiles (at age 65!) researching hand-woven belts ofTarahuamara native communities in northern Mexico and as well as the Koelz textiles. When I arrived as curator of the Museum of Anthropology in 1993, Grace had nearly completed her study of the textiles and had prepared a lengthy manuscript on the Kashmiri and Persian shawls, west Indian phulkari embroideries, Uzbeki embroidered household textiles (suzanis), fitted garments, and miscellaneous other textiles (saris, horse caparisons, rugs, etc.) in the Koelz Collection. A perfectionist, Grace was not

viii

satisfied with her ambitious and wide-ranging study, and when she left Ann Arbor in 1998 to live near her daughters in Oregon, decided to recast the volume into two separate books. As curator of the collection I was eager to see her work published, but neither Grace nor I made significant progress in meeting this goal for several years. We renewed contact in spring 2003, when I contacted Grace about the volume and requested her approval to extract the portion of her manuscript dealing with the Kashmiri and Persian shawls for a publication that could appear in concert with a planned exhibit of the shawls at the University's Museum of Art. Then ninety years old, Grace responded enthusiastically to this suggestion and began to work through the volume, writing detailed editorial suggestions and revisions in the margins, which she mailed to me in sections. As noted, Grace passed away before this volume could be completed, and it has fallen to me to complete the final editing and reorganization of the work. While I have necessarily reorganized the work somewhat, r have tried to be as true as possible to her intent and have maintained her graceful prose. An avocational weaver myself and a scholar of South Asian craft production, it has been a pleasure to see this volume through to completion, and to learn from Grace about the wonder of these beautiful textiles. I hope that the readers of this volume will share her love and enthusiasm for the objects described, and I am confident that they, like me, will learn a great deal from her dedicated and impressive scholarship.

Carla M Sinopoli December, 2004

ix

Acknowledgments

I

n a project taking more than 20 years from inception to completion, many debts have been accumulated. I am certain that if Grace were here, she would first and foremost acknowledge the support of her daughters Margaret and Kelcey. In the Museum of Anthropology, we thank Karl Hutterer, who as Curator of the Asian Division first encouraged this project. We also thank the then and now Museum Director, Dick Ford, for his support at the beginning and end of these efforts. Collection managers David Kennedy and Karen O'Brien provided invaluable help in making the shawls accessible for study and photography. Thanks to Karen O'Brien also for word-processing Grace Beardsley's original 400-plus page manuscript and Walter Koelz's manuscript which is included on the CD. The University of Michigan Center for South Asian Studies provided financial support for this work. Kay Clahassey drafted and redrafted all of the illustrations in this volume and has been with the project since its inception; her dedication, patience, and good humor are appreciated as much as her extraordinary artistic skills. Thanks also to Sally Mitani and Jill Rheinheimer who edited, designed, and nurtured this manuscript and rode herd on me to see it to completion. The University of Michigan Office of the Vice President for Research and College of Literature, Sciences, and the Arts provided funds for the photographic documentation of the collection. The photographs were taken by Dwight Cendrowski (Dwight Cendrowski Photographers LLC) in the hallway of the School of Architecture and Design; we thank Sherri Smith for her assistance in this effort. The publication of this manuscript coincides with an exhibit of the Koelz textiles at the University of Michigan Museum of Art. We thank exhibit curator Maribeth Graybill, and the museum's exhibit staff for providing such a wonderful showcase for the collection. Finally, we must acknowledge Walter Koelz who acquired this extraordinary collection in the first place; we dedicate this work to his memory. Carla M Sinopoli June, 2005

x

PART

I

Kashmir Shawls in Context

-1-

Introduction

About the Koelz Collection

T

he Koelz Collection of Asian textiles, Asian Division of the University of Michigan Museum of Anthropology, derives from Walter w. Koelz's travels in Southern and Central Asia in the early 1900s. The collection was made in what was then northwestern India, although the textiles themselves derive from a much larger area including Iran, Afghanistan, Turkistan, Uzbekistan, Tajikistan, Pakistan, Kashmir, India, and Xijang (Tibet). These areas are Central Asian, not only geographically but even more so historically: they are central to the Asian continent in having served for millennia as crossroads for trade, conquest, and cultural exchange. The Koelz Collection includes flat and fitted garments as well as a variety of textile household furnishings suitable for nomads, villagers, and urban dwellers. The textiles include wool pile weaving produced by nomads inhabiting areas just east of the Caspian Sea; silk brocades and embroidered cottons of the Middle East; fine woolen shawls fi·om Kashmir in the foothills of the Karakoram Range; Tibetan horse caparisons; and even a tribal sash from the Assam Hills of eastern India. Particular strengths of the collection are the tapestrywoven wool shawls of Kashmir, cotton shawls embroidered in the phulkari technique from India and Pakistan, and embroidered household pieces from Uzbekistan. Their creators likely included both household artisans who produced for family consumption and full-time specialists producing for international markets. I Koelz's discernment and respect for high-quality textile skills is impressive. Collecting these textiles was no easy task. In a letter written from Karachi in 1934, commenting on a shipment of items to the University of Michigan, Koelz observed: "the acquisition of good textiles is a slow process, more laborious even than the body-exhausting trek in the mountains after Tibetan paintings."2 This study of the Koelz textiles explores their origins and their uses as well as the objectives of their collector. The textiles also tell us about the societies that produced these 2

Introduction

3

goods and the period in which they were produced. Many of the pieces reflect the collision between traditional handwork and the industrialization of the nineteenth century. The Koelz Collection also casts light on changing textile technologies and the impact of changing expressions of taste. This study also touches on the search for suitable fibers and dyes, the development of ingenious devices for weaving and coded instructions for illiterate weavers, the patterned distribution of labor, and evidence of trade. Aspects of these are embodied, bit and fragment, in the cloth offonner days. Nor should the aesthetic dimensions of the collection be overlooked. Many ethnic cloths are of great beauty and are today recognized as such, rather than being regarded as mere "native" crafts, relegated to exhibition among ostrich eggs and other curiosities. We must also realize that in centuries past, a far greater value was placed on textiles. Cloth once ranked with gold, gems, weaponry, magnificent palaces, and large retinues as evidence of wealth and power. Cloth was, in fact, a mainstay in making palaces and retainers resplendent. The possession and distribution ofluxury items, including fine cloths, was both widely recognized and pervasively employed as evidence of power. A leader might distribute wealth to reward and control underlings or might present handsome and costly gifts to rivals in elaborate "Robes of Honor" ceremonies (Gordon 2001). Both gestures could strengthen and elevate the leader's position with his subordinates. Similarly, rulers and courts made opulent displays of wealth derived from plunder, tribute, local manufacture by court-maintained artisans, or gift exchange with other rulers or elites. Nor were handsome fabrics displayed only on special occasions. Taking an example from an earlier Central and Southwest Asian world to which some ofthe Koelz textiles are distantly related, the sixteenth- and seventeenth-century Safavid rulers of Persia used a dazzling array of fabrics in their daily domestic activities. Rulers, courtiers, and servants were clothed in silk, often gold- and silver-enriched. Their horses were caparisoned in equally handsome materials. Likewise, rich fabrics with woven or embroidered designs were used to fashion garden pavilions and gorgeous tents. These embroidered canopies were carried onto the hunting field to offer shelter for rest or regalement. On the field of battle, the contest in display was often almost as important as the contest at anns, and larger and even more elaborate textile constructions were erected. Within the Safavid palaces, floors, divans, and cushions were covered with rich weavings. Walls were sometimes entirely draped with hangings, and doorways were hung with curtains (Ackennan in Pope and Ackennan 1965:5 :2069). One wonders how all this textile wealth in support of power was produced for courts and how it was stockpiled. In the great Islamic courts of Persia, India, and Central Asia, the tremendous volume of sumptuous textiles required for courtly activities gave rise to the institution of palace factories known as tiraz (or karkhanas in the Mughal context). Some idea of the magnitude of manufacturers in such courts may be gained from a description of the fourteenth-century court ofthe Delhi Sultanate cited by Serjeant (1972: 122). In addition to the annor makers, jewelers, embroiderers and other master craftsmen, the state factories of Sultan Mohammad ibn Tughlak of Delhi (r. 1325-1351) employed four thousand weavers of silk. The tiraz storerooms also contained vast quantities of imported cloth from China, Iraq, and Alexandria. In the seventeenth-century Persian Safavid court, the need for splendid textiles was met by tribute as well as by the workers in the court's tiraz. Artisans were drawn from many lands,

4

Wrapped in Beauty

for skilled artisans were themselves valued assets. Among the manufactured products of the period, textiles became perhaps the most important. In 1667, Isfahahn had some thirty-two weaving workshops attached to the court, each employing some 150 skilled weavers-a total of 4,800 laborers working solely for the court (Lubell 1976:2:120). The use of fine textiles was of course not solely the realm of emperors. Textiles were also highly valued in the more modest households of local kings, elites, merchants, and others with the means to acquire them. The Maharajah Sawai Man Singh II Museum in Jaipur (Singh 1979: ix-xi) houses an outstanding family collection offabrics and costumes that is remarkable in revealing a dynastic family's long history of collecting cloth pieces and clothing; this is a rare well-documented and well-maintained textile treasure that has been conserved despite India's humid climate. Some of the earliest dated textiles preserved in India come from this collection, which includes fitted costumes for men, women, and children; shawls, odhanis (veils), and other draped wear; unused fabrics and furnishing materials, quilts, curtains, upholstery, carpets, and floor spreads; and palanquin covers and horse caparisons. The rajah's court workshops, karkhanas, which produced the textiles, were organized on the Mughal pattern: spinning, dyeing, weaving, carpet-making, printing, embroidery, stitching, and jewelry-making figure among the thirty-six workshops. Even among non-elites, smaller textile troves and more modest wardrobes were accumulated and passed on between the generations. The use of socially appropriate textiles placed the owner or wearer more securely in her or his social network. For example, the phulkari embroidered shawls in the Koelz Collection were prepared well in advance by women in villages of the Punjab for their daughters' marriages and distribution to wedding guests. Afterward, the shawls were carefully laid away to be worn at other festive events. In the same way, the Koelz suzanis from Uzbekistan, embroidered for home furnishings, were worked by family women in anticipation of marriage and the establishment of new households. While the textiles described above were made by women for members oftheir extended households, the Kashmiri and Persian wool shawls in the Koelz collection are the work of male professionals. Worn primarily by upper-class men, they were significant status symbols. The Koelz silk brocades are also the work of professional weavers, and were tailored to produce handsome clothing expressive of high social position. The Koelz textiles are claimants to no great age. The oldest fabric piece in the Koelz collection is probably the early seventeenth-century Shah Abbas silk fragment. Most of the shawls were purchased in antique shops of northwestern India in the early 1930s, and date to the nineteenth century. It is only for the wool shawls that a closer dating has been attempted, facilitated by the numbers of more specifically dated published shawls. Even so, the suggested dates for the Koelz shawls are approximate at best.

The Organization of This Work

The work is divided into two main sections. Part I consists of six chapters, providing the context for understanding the Koelz textiles. Chapter 2 presents a brief overview of the

Introduction

5

history of the collection, and of Walter Koelz, who procured it. Chapters 3 through 6 tum to the textiles themselves. Although there are only two turbans in the collection (and even these are arguably not turbans), these relatively early and finely woven textiles contain many of the elements important to understanding the more numerous shawls in the collection. These "turbans" are discussed in Chapter 3. The bulk of this study, however, is devoted to the shawls. Chapter 4 provides historical context for the shawls by considering their material constituents, technology of production, and the social contexts in which the shawls were produced and consumed from the seventeenth through the nineteenth centuries. Chapter 5 examines the shawls in even more detail, considering an important technological innovation of the nineteenth century-the development of elaborate pieced shawls. Chapter 6 turns to a discussion of major motifs important in pieced and fully woven shawls, and the impact of the European Jacquard loom on shawl design. Part II of this volume is a detailed descriptive catalogue of a portion ofthe Koelz textile collection: the elaborate shawls and "turbans" woven in Kashmir or Persia. Information on other textiles in the collection-saris, phulkari embroideries, fitted garments and household textiles (suzanis, rugs, book covers, and other pieces)-will be published separately in the future. Here, we focus on one important weaving tradition that emerged out of the Islamic cOUlis of Central and South Asia, and whose products came to be coveted and widely worn by European women throughout the nineteenth century. The transition from production for courtly elites to that of a broad international consumer base dramatically impacted both the forms and ornamentation of the textiles and their technology of production. For the most part, this monograph describes the collection in terms easily accessible to the general reader, keeping formal textile terminology to a minimum. But textile analysis is often technical enough to defY simple explanation and repetitively detailed enough to tire. Where feasible and expedient, data are condensed into tables and graphs. Nonetheless, certain sections of this work are replete with detailed descriptions of aspects of technology that have in the past been unfortunately overlooked or, perhaps, tactfully avoided. For the textile scholar, these details are essential. For the general reader, the pages are easily turned. Terms not usually found in standard collegiate dictionaries or that are used in a specialized way are defined in the text, and they have also been summarized in a glossary at the end of the book. English usage is employed for most terms, except where there is no ready equivalent. Where possible, the social background in which the fabrics were used is discussed. However, this is uneven and for some textiles in the Koelz collection, little information is available.

Notes 1. The Koelz collection of the UMMA also includes an outstanding assemblage ofwell-documented thangka paintings (Copeland 1980). 2. Letter to Professor Winter, 13 January 1934, copy on file, UMMAAsian Archaeology Division, University of Michigan.

-2-

Walter N. Koelz and His Collections Carla M Sinopoli

D

escribed by a colleague as the "last of the Victorian adventurers," Walter Nonnan Koelz was a man of wide-ranging interests whose adventurous travels at the top ofthe world generated the Museum of Anthropology's remarkable Koelz Collection (plate 2.1). Although fonnally trained as a naturalist, in his tours in Tibet, North India, and Central Asia Koelz was a catholic collector, focusing on artistic and craft goods as well as botanical and zoological specimens. Koelz kept detailed diaries! during these trips and maintained an active correspondence with numerous friends and sponsors. Many of his journals and letters were donated to the Bentley Historical Library at the University of Michigan, with which he had a seventy-four-year association. These records provide a valuable resource on Koelz and on the materials he collected. In this chapter, we present a brief biography of Walter Koelz, before turning to the collecting activities that resulted in the Museum ofAnthropology Koelz Collection and the woven textiles considered in this volume.

A Biographical Sketch Walter N. Koelz was born in 1895 in the small town of Waterloo, Michigan, where his father, a Swabian immigrant, was the village blacksmith. Upon graduating from high school, Koelz was awarded a scholarship from Olivet College in south-central Michigan. He began his study there in the natural sciences though he eventually chose to major in languages. After graduation, one of his fonner science teachers recommended him for the position of staff assistant at the University of Michigan's biological station at Douglas Lake, and Koelz joined the University as a graduate student assistant. This opportunity reawakened his earlier interests in biology and he decided to pursue a doctoral degree in ichthyology, completing his dissertation on Great Lakes whitefish in 1920.

6

Walter N. Koelz and His Collections

7

Koelz began his career as a world traveler in 1925, when he agreed to serve as naturalist to the MacMillan Expedition to the Arctic sponsored by the National Geographic Society. It was in 1930 that he undertook the first of what would become three decades of journeys to Asia, when he signed up to head the biological section of the Himalayan Research Institute of the Roerich Museum in Kulu, India. He spent nearly two years in the region before falling out with his employer and returning to the United States, moving briefly to New Mexico where he collected Native American pottery and traditional arts. In 1932, the Regents of the University of Michigan appointed Koelz as a Research Fellow, supported by the Charles L. Freer Fund. His main duty under this appointment was to travel to British India to collect biological specimens and material culture for the University's Museums of Zoology and Anthropology. 2 It was during this two-year trip that the majority ofthe Museum of Anthropology (UMMA) Koelz Collection was acquired. 3 Koelz returned to Michigan in 1934 and worked for nearly two years in a government parks project in his home town of Waterloo, Michigan. He returned to Asia in 1936 on behalf of the United States Department ofAgriculture, beginning a journey that would last some seventeen years, taking him to India, Nepal, and Assam, and from 1939 to 1946, to Iran. Koelz returned to his house of birth in Waterloo, Michigan, in 1953. He passed away on September 24,1989, at the age of ninety-four. With the exception of the 1932-1934 Freer trip, most of Koelz's efforts in Asia were devoted to collecting biological specimens-mostly plants and birds-for various institutions, including the American Museum of Natural History, Smithsonian Institution, United States Department of Agriculture, the University of Michigan Museum of Zoology and University Herbarium, and Michigan State University. His bird collections alone numbered more than 75,000 specimens. Throughout his journeys he was also a prolific diarist and an astute observer of local life and customs, recording information on ethnobotany, pastoral migrations, marriage patterns, and local politics. He was, as noted above, an avid collector of a range of craft goods and artistic products, collecting pottery, wooden objects, sacred images, sculptures, and textiles for himself, his friends, and the University of Michigan. With no formal training in Asian art, he had an eye for quality, and amassed an impressive and valuable collection. This brief summary, while perhaps capturing a bit ofKoelz's extraordinary energy and wide-ranging intellect, does little to capture the personality of the man, even to the extent that it can be gleaned from his writings. Stubborn and cantankerous, he was involved in frequent legal disagreements, including with his long-time collaborator and partner Rup Chand, who sued him for failing to maintain their financial agreements after they parted ways. Also in his correspondence are records oflegal disputes with the Roerich, and correspondence from aggrieved curators at the American Museum of Natural History responding to his complaints that they were not adequately caring for and documenting the valuable ornithological collections he had acquired on their behalf. Equally, he was a man who formed long and durable friendships, and his extensive correspondence and the many tributes published after his death, on file at the Bentley library, make clear that Koelz was a loyal and generous friend who left a profound impact on those he knew, as well as extensive information on the biology and culture of South Asia's isolated mountainous north.

8

Wrapped in Beauty

Plate 2. 1. Walter N . KoeLz ( from the Walter N. Koelz o llection of the Bentley Histori cal Library, Universi ty of Micbigan , used with penni ssion).

The UMMA Koelz Collection Koelz's appointment as a University of Michigan Freer Research Fellow provided him with a salary of $5,500 and a travel stipend of $1 ,500; his correspondence indicates that private donors further supplemented these funds. He left for India on November 2, 1932, and returned to the United States on Apri125, 1934. The diary from the first year of his journey was lost, but his 1933-1934 journal is on file in the Bentley library and includes detailed daily accounts of the travels he undertook with his assistant and companion Mr. Rup Chand, whom he had met in his first trip to the region. His writings provide rich information on his daily travels and collecting activities, focusing particularly on the time he spent in the mountains, including his visits to Buddhist monasteries in Ladakh and Spiti, where he acquired thangka paintings along with images, reliquaries, and other objects. There is much less discussion

Walter N. Koelz and His Collections

9

of his textile acquisition activities, which tended to occur in the winter months when travel was not possible in the high Himalayan mountains, and Koelz and Rup Chand spent time in lowland urban centers such as Delhi, Amritsar, and Lahore. While in the cities, Koelz scoured antique shops looking for textiles and other objects. In an April 27, 1933, letter written from the Punjab to friends and benefactors, Koelz described these efforts: And then there was constant intercourse with curio dealers within a radius of 500 miles. I saw several hundred old textiles, mainly shawls, and bot [bought] a half-dozen exquisite specimens. It is amazing that really few dealers have any idea of what constitutes merit in anything they sell and good and bad are priced alike. The more I see the more apparent it becomes that the paisley pattern represents the end of the art of shaw I weaving. The glorious work of the best period is something quite of art.4

While there is not a great deal of detail in Koelz's papers about his time in the cities, some hints can be gleaned of his textile collection work. Thus, his diary records that while in Amritsar on January 1, 1934, Koelz visited the shops of Radha Kishan BharanyS and of Gopal Das and observed, "Selected 20 phulkaries from latter to be held till we looked over the other ... [undecipherable] stock if offer wasn't accepted." Two days later he reported that "Gopal Das refused offer for phulkaris, so went to RKB and selected. Bot [Bought] 3 or four old Afghani or Kaziristan Phulkaris (the red and black) from G.D. Others haven't any or few. He had five fair or good ones." And on January 4, he noted "a man Ghulam Hussain had a nice black Persian shawl from Lahore. Asked 500 [rupees] and got 75. He also says a German has been buying for the Berlin Museum."6 By the end of January 1934, Koelz had amassed an impressive collection. Twelve large wooden boxes were shipped from Karachi on the President Polk on January 26; they arrived in New York a month later, on February 27, and in Ann Arbor in late March. Koelz wrote to Professor Winter at the University of Michigan: The contents of the boxes are for the most part objets d'art and natural history specimens collected this year. There are, however, some ethnological collections made by me in previous years and some other of my own property from previous collection, including a few shawls, two rugs, a few pieces of silver, and a few paintings. Some of the better paintings and rarer textiles I retain for the present. They are of both this year's and former collections. You will find a large plant collection, one 4000 numbers, and a very nice lot of bird skins. There are some mammal skins of the particular species desired by the Museum. Emphasis has been laid on the acquisition of both study and exhibit material in the field ofIndian and Tibetan art. There are about 40 very good Tibetan paintings bot [bought] this year and about 30 of mine. This I consider a most choice assortment. ... The textile collection consists principally of shawls, Old Persian and Kashmir. Roughly 40 were bot [bought] this year. There are a few pieces of old Persian brocade, old silk print, velvet print, silk and cotton saris, etc. The acquisition of good textiles is a slow process, more laborious even than the body-exhausting trek in the mountains after the Tibetan paintings. I am accumulating experience and should be able to do better work in the future. All the objects sent are not considered beautiful by me. Some have been bot [bought] as illustrative of type of design, execution, or otherwise. I shall be glad, however, if you agree with me that some ofthe things are really beautiful. 7

10

Wrapped in Beauty

Included in the twelve boxes were 46 shawls, along with more than 600 other ethnographic and historic artifacts including iron arrow points, silver lings and ornaments, brass and copper teapots, wood seals, bronze and stone images, icon cases, additional textiles, and more than 50 Tibetan thangka paintings. Mixed in with these were numerous "hunting trophies" (zoological specimens) and dried plants. Koelz's packing list distinguished objects intended for the museum from those that were part of his personal collections. In an unpublished manuscript prepared upon his return, Koelz provided some further information on his collecting activities: In the winters of 1932-33 and 1933-34, collections of textiles in the various old culture centers of India were examined, a number of pieces of the best work were acquired. In the summer of 1933, a survey was continued of the art treasures of the great monasteries of West Tibet, including the Gelukpa district of Spiti and Upper Kunawar. Some 500 objects, chiefly old paintings, old brass, copper, and wood work, and modern silver work and jewelery were acquired in this area. The textile collection comprises some 40 shawls of Persian and Indian manufacture, representing the several types of weaving, printings, and combining of parts; one Persian silk print, several Chinese silk embroideries; three old Persian brocades of silk; one old Benares silk brocade; one old Rajput gold print on velvet; several miscellaneous pieces of old Indian silks; and some 25 large silk embroideries on cotton from India and adjoining countries. Outstanding among these articles are several large shawls. One of exquisite texture and superb color and design is clearly to be reckoned among the best products of shawl-makers art. It dates undoubtedly at least from the 17th century and is likely of Persian origin. Though virtually all of the shawls of the collection are of superior quality and considerable age, eight others are rare and choice specimens. 8

The shawl of "exquisite texture and superb color" is likely UMMA 17348, which Koelz described elsewhere as "unquestionably the masterpiece of the collection. Every feature may conservatively be described in superlatives. The coloring is beautiful, and the drawing superb."9This much repaired and fragile textile, likely considerably reduced from its original size, was backed by three layers of cloth (worn white silk, tan gauzelike cotton, and printed cotton), evidence of repeated preservation efforts that attest to the high value this shawl held for its former owners. Unlike the other shawls in the collection, its main field is of silk warps with silk and cotton wefts, rather than of wool. Beardsley also considers this an "accomplished piece of weaving" (see Catalogue), observing that it differs from all other pieces in the collection in design, color, fiber content, and weave. Both Beardsley and Koelz believe that it was manufactured in Persia. It is not possible to securely identifY the other eight shawls that Koelz considered "rare and choice." However, it is likely that UMMA 17320, a pieced long shawl that Beardsley dates to around 1870, and described by Koelz as "of outstanding quality," belongs to this group, as does UMMA 17307, a mid-nineteenth-century densely patterned red long shawl called by Koelz "one of the most imaginative designs in the collection ... (with) skillful conception" and "delicate and charming variations throughout" in color and pattern. 10 Other shawls that appear to have been especially esteemed by Koelz are UMMA 17305, and embroidered shawl Kl, which remained in his private collection. I I

Walter N. Koelz and His Collections

11

"A Study of a Collection of Textiles Made in the Indian Empire" Upon his return to Ann Arbor in the spring of 1934, Koelz, as noted above, worked for a government parks project in Waterloo, Michigan, acquiring land for the Waterloo Recreation Area. During this time he also prepared a lengthy manuscript on the shawls, writing to his friend and benefactor Ms. Margaret Watson Parker, in the summer of 1934: "I have been busy with a report on the shawls and can see the end. I have had many of the designs sketched so there will be something worth looking at at least. It is interesting in that the pieces fall into several distinct classes, which I interpreted to mean that there were several schools of weaving. It has been assumed that they were all made in Kashmir."'2 Writing her again in August 1934, he observed: I have written lengthily on the shawls and shall present the ms. to Dr. Aga-Oglu [Curator of the Oriental Division of the Museum of Anthropology] shortly. I have spent several days in the Detroit Library and find almost nothing about that class of textile. I find a reference to the excellence of shawl manufactures in India before 1600 and the fact that the industry was then in the hands of Mus sui mans. I infer the art came from the west with the conquerors. It seems too bad that so little scientific thot [thought] has been given to textiles in general.'3

Along with the manuscript, the accompanying illustrations eventually were handed over to what is now UMMA's Asian Division. They remained unpublished and on file until August 1955, when Koelz came to collect them so that he could prepare the manuscript for publication. Sadly, the manuscript was never published and while the text was eventually deposited in the Bentley library, the figures were not, and have not been located. We include Koelz's transcribed 260-page manuscript-Kashmir Shawls: A Study of a Collection of Textiles Made in the Indian Empire-without illustrations, on the CD-ROM enclosed with this volume. It is an interesting work, written several decades before any other scholarly publication on Kashmiri textiles (John Irwin's 1973 The Kashmir Shawl being one of the first such works). Koelz nonetheless discovered the important references that have since become canonical in any study of Kashmir textiles. He cites the observations of the chronicler Abu Fazl on the Mughal emperor Akbar's (r. 1556-1605) fondness for woolen shawls from Kashmir and the consequent expanding popularity of these textiles among the royal courtiers, and cites Francois Bernier, a French traveler in the court of Akbar's grandson Aurangzeb (r. 1658-1707), who described shawls as being the "staple commodity" of Kashmir. From the colonial period, he quotes extensively from the writings of William Moorcroft, who traveled in Kashmir in the early 1820s (see Chapter 4), and the 1866 writings of J. Forbes Watson (see Chapter 4). Nonetheless, Koelz's work is quite different from most other studies of Kashmir shawls, including the work of Beardsley presented in this volume. Contemporary scholars have available much better information with which to date shawls and chart changes in motifs and design structures than was available to Koelz. As a trained biologist, Koelz brought a Darwinian perspective to his discussion, suggesting that since shawls were manufactured in mUltiple locales and over a long period, their development "cannot have escaped the general

Square shawls with broad field gallery Piecework square shawls Strata-field square shawls Pieced conglomerate square shaw Is

Moon shawls pieced across middle

17331

17317 17323 17308

Moon shawls woven whole

17306 17326 17319 17325 17343

17321

Strata-field long shawls

Moon shawls woven whole with attached hashias

17310

17337 17348

17313

Floral-Design

Block pieced long shawls Reworked long shawls

Field gallery long shawsl

Yardage long shawls

Turbans Long shawls woven whole

Overlaid-Design

17341

17318 17345

17320

Block-Design

Table 2.1. Comparison ofKoelz's classification with Beardsley's organization of the shawls.

17311 17324

Red-Medallion

17338

17316

17334 17307 17335

17339

Red-Overlaid

17336

17329 17315

17340

17330 17344 17304

17309

17305

17328 17327 17333

17314

Butha-Design

......

::::

~

\:)

b:l (\)

S·

\:)..

(\)

~ ~ '"(5

N

Walter N. Koelz and His Collections

13

law of local differentiation." And in the absence of evidence of production locales, Koelz argued that regional variants should be identifiable through "the inductive path of scientific procedure," and considered weaving technique, color, and design to group the textiles into discrete classes. 14 Koelz defined six classes, distinguished by "more or less definite characters,"15 that he believed derived from distinctive production locales, mostly within Kashmir though he did not exclude Persian production for some shawls. Table 2.1 compares these classes with Beardsley's organization. Whereas Beardsley classified by shape, design structure, and manufacturing technique, Koelz did not consider shawl shape or production technology. Thus, square and long shawls were grouped together in three of his six categories, and pieced and fully woven shawls were sometimes also combined. His main classification criteria were based on design elements and design structure. Thus, his "Overlaid-Design Group," all described as "moon shawls" by Beardsley, was characterized by "wandering elements that are superimposed and intertwined," creating intricate patterns; Koelz considered this group among the best in the collection of weave and design. And his "Floral-Design Group" was "designed in classic simplicity," 16 built of small units of graceful vegetative elements, using simple color combinations. His termed his other groups: Block-Design, Red-Medallion, Red-Overlaid, and Butha-Design. Koelz's detailed discussion of the shawls in the UMMA collection is presented in full in the CD-ROM accompanying this volume, although it is somewhat difficult to follow in the absence of his illustrations. It is interesting to speculate how his approach might have shaped the study of these textiles had it been published closer to its completion in 1934. Now however, it is most interesting for presenting an alternate approach to the collection, generated by this biologist turned art historian, working in a time before serious scholarships on shawls had begun. Rather than explore this further, the remainder of this study consists of Grace Beardsley's masterly account of the shawls, which benefited from a somewhat longer legacy of textile scholarship, and her artistic eye.

Notes 1. Persian Diary, 1939-1941, by Walter Koelz. Anthropological Papers, no. 71, Museum of Anthropology, Univeristy of Michigan. Ann Arbor. This book is still in print. 2. University of Michigan Regents Proceedings, 1929-1932, p. 241, Bentley Historical Library, University of Michigan. 3. Unfortunately, the University of Michigan Museum of Anthropology and the University of Michigan Museum of Art have the same acronym. In this book, UMMA refers to the Museum of Anthropology and all collection numbers that begin with "UMMA" denote Museum of Anthropology collections. 4. Letter to Margaret Watson Parker, 27 April 1933, Koelz Papers, Box 2, Bentley library. 5. Described in its stationery, also in the Koelz file, as "Shawl and Indian Curiosity Merchants. Exporters of Shawls of Every Description." 6. Koeli Papers, Box 3. 7. Letter to Professor Winter, 13 January 1934, copy on file, UMMAAsian Archaeology Division, University of Michigan. 8. "Preliminary Report ofthe Oriental Expedition," Koelz Papers.

14

Wrapped in Beauty

9. Koelz n.d.: 107, manuscript on file, Bentley library, and reproduced in this volume in the accompanying CD-ROM. The page numbers cited from Koelz's volume refer to his original typescript; the reproduction manuscript has been renumbered. 10. Koelz n.d.:170. 11. Koelz n.d.:147. After his death, Koelz's private collection was sold at auctions by Sotheby's, Christie's and the Frank Boos Gallery, with proceeds benefiting the Nature Conservancy. 12.Koelz Papers, Box 2. 13. Koelz Papers, Box 2. 14.Koelz n.d.:20. 15.Koelz n.d.:20. 16.Koelz n.d.:85.

-3-

"Turbans" (Patkas)

L

ittle used in the West, turbans are a fonn of headdress with a long history in South and West Asia. These wrapped headdresses perfonn both protective functions of defense against climate and physical violence, and ceremonial functions of identifYing rank, office, religious affiliation, and social or personal identity. According to Chandramani Singh, the turban was the most important article of elite male costume in medieval and early modem north India: No one was allowed to enter in the royal household or the king's court without wearing a turban . . . . Removal of his turban was the most insulting punishment for a nobleman. Prisoners were not supposed to wear turban .... Men were not supposed to perform any religious ceremony without a turban on their head. [Singh 1979: xxii]

A description of the turban appears in the writing of the nineteenth-century traveler and adventurer Annin Villnbery. In 1863, Vambery wore a turban for a year while traveling through Turkistan disguised as a dervish. He observed: The turban, it is well known, represents the pall that every pious Mussulman must wear on his head as a continual memento of death. The Koran only enjoins a pall having a length of seven ells. I But zealots often exceed this measure, and carry about on their heads four to six such palls, thus making altogether from 28 to 42 ells offine muslin. [V{lmbery 1979(1864):211]

Anthropologist Bernard Cohn commented further on the symbolic importance of the turban, noting that, "in the eighteenth and nineteenth centuries, an Indian would place this turban at the feet of his conqueror as a sign of complete surrender" (1989:314). Further, "nineteenth century guide books written for Englishmen traveling in India warned their readers never to touch a Hindu's or a Muslim's turban as this was considered a grave insult" (Cohn 1989:314-15).

15

16

Wrapped in Beauty

By the mid-nineteenth century, the turban was in wide use throughout India, though its size and mode of tying differed by community and by caste (Watson 1982[1866]:13). The rigidity of the system was not relaxed until the end of the nineteenth century. Commonly of cotton, turbans may be made also of silk or wool. They are either rectangular or square, with great variety in styling. A typical rectangular size might be 5.90 to 9.80 m (15-25 yards) long and 25 to 30 cm (9-12 in) wide, although some turbans are both longer and wider. Square turbans average about 3.0 m per side (range 1.37-3.64 m) (Watson 1982[1866]:20). The ornamentation of turban fabrics is usually confined to the portions that are exposed to view when worn. The Koelz collection contains two turbans, which both appear atypical in size and ornament. Indeed, as I discuss below, it may be that neither was in fact a turban. The first (UMMA 17387) is a nineteenth-century Hazara silk-embroidered cotton turban. The Hazara tribe, originating in the mountainous Hazarat region of central Afghanistan, is known for the weaving of flat textiles such as kilim rugs and saddlebags. We have little information from Koelz concerning his acquisition of the turban. It appears to have been collected in the northern Punjab, probably from Afghani or Hazara migrants who had settled in the reglOn. The Koelz turban is indigo dyed and embroidered across much of its surface. The embroidery is mostly in henna-red silk, with additions in light green, yellow, white, and fuschia silk. An interesting aspect of the embroidery is the use of a soumak-like stitch, set in columns, as area filler for some ofthe large squares and side triangles. Soumak is a wellknown weaving technique, used in non-pile carpets and other kinds of weaving, and widely employed in rug-making areas throughout the Middle East. Essentially, soumak consists of weft-wrapped groups of warp threads (Fig. 3.1). The use of soumak as a phulkari embroidery adjunct instead of a weaving technique is interesting and unusual. Although catalogued as a turban and quite possibly used as such, at 196 cm, this piece is far shorter than the average 590-980 cm length that Watson described. Moreover, its ample embroidery throughout is atypical of turban ornament. Rup Chand (pers. comm.), Walter Koelz's co-collector and traveling companion, suggests that this piece may have been a tent band or was at one time used as a tent band. The second turban in the Koelz Collection is a late eighteenth-century Persian piece that Koelz acquired in Bombay (UMMA 17313). This piece is more than 6 m long by 70 cm wide and is woven of very fine goat wool. The central creamy-white plain field is sheer enough to permit easy reading of a typed page placed beneath, though the ornate silk end and side panels are opaque. These woven end panels or borders (pallas) and pieced side panels or borders (hashias) have elaborate multi color floral motifs, and were woven continuously for the full 626 cm length of the piece. The extreme length of this piece suggests a very long turban or, perhaps, a sash. Singh (1976: xi) describes eighteenth-century portraits featuring men in sashes (kamarband or patka) that were long enough to wrap around the waist two or three times, fasten in front, and fall to the knee. Like the Koelz piece, these patkas had matching floral panels at either end, a feature not necessary in a turban, since only one end would be visible. Often, an elaborately

Koetz s "Turbans"

17

Figure 3.1. Soumak technique as used in weaving. For embroidery, fabric warps or wefts are regularly counted in same way as loom warps are here (after Eilan 1973:27).

worked dagger was worn tucked into the patka. Singh also notes that in medieval society, soldiers were supposed to wear such waistbands and that kings and emperors presented patkas to their officers and nobles in recognition of their services. Similar pieces are known from Persia. An observation made in 1753 is pertinent: "even in the presence of kings, the Persians wore their sashes as turbans which were made of costly Kermani wool" (Hanway 1753:1 :33). Sashes are depicted in numerous miniature paintings ofIndian Mughal and Persian elites, which show elaborately clad courtiers wearing patkas tied in front with ends hanging to the knees. The ends show panel designs much like the pallas ofthe Koelz piece. In fashions, these two imperial courts had much in common. The Koelz turban or patka has important similarities to contemporary eighteenthcentury shawls from Kashmir and Persia, which are the main focus ofthis study. I therefore describe this patka's construction and design in some detail to serve as a datum point for the subsequent discussion of the Koelz shawls in Chapter 3. In size, this Koelz piece tallies with published measurements of late eighteenth-century Kashmir patkas. Earlier patkas were shorter, only about 200 to 300 cm long by 35 to 50 cm wide (Singh 1979: xxix). They gradually increased in both dimensions. Among late eighteenth-century Kashmir patkas illustrated by Ames (1986:216, 222-23) are three that closely relate in design and size to the Koelz piece. They range from 634 to 640 cm in length and 67 to 69 cm in width, and have narrow side borders similar to the three cm hashias of the Koelz patka. Ames (1986:50) gives 3.5 cm as a typical eighteenth-century hashia width and states that the patka went out of fashion about 1810.

18

Wrapped in Beauty

Warp counts for the hashias, or side borders, of this patka are higher than for the rest of the piece-58 compared to 42 threads per cm. The warps are also of silk, not wool. Silk warps for the garment edges provide greater strength and durability than wool and may help the garment to hang better; however, their use creates certain weaving problems. The wefts crossing the denser silk 58/cm warp area of the hashia require slightly firmer pressing into place with each shed (see Chapter 4 for discussion of loom operation) than those crossing the 42/cm warp area if the woven edge is to advance evenly. That is to say, the hashias would tend to build up faster than the palla. This was one of several factors that encouraged the separate weaving of has hi as with subsequent attachment to the shawl proper, a practice prevalent in nineteenth-century Persian and Kashmir shawl manufacture. Another consequence of dense hashia warps is the altering of yam diameter ratios and appearance. The same weft yarns and color range provide decoration in both palla and hashia of the Koelz patka, but the yams are more compressed by the hashia's higher thread counts and their colors are thereby delicately subdued. The motifs also have greater clarity in the hashia. The palla design of this patka is important for its bearing on later developments in Kashmir shawls and certain other textiles. A row of six rather columnar floral botehs (a conventional flower motif) comprises the principal motif of each palla (see UMMA 17313, detail, in Color Plates). Each boteh ascends from a red vase set on a low pedestal tray, and includes a generous selection of at least twelve different flower shapes. Flowers intrigue designers not only for their beauty, but also for the many ways in which they may be presented and still be perceived as flowers. The designer and weavers of this patka did not stint for floral variety, whatever the cost in time and effort. The botehs are neither the naturalistic plant forms of the early eighteenth century nor the compact, highly stylized, bent-tipped, floralfilled botehs of the later nineteenth century, but instead resemble a form slightly later than that noted by Irwin as 1740-1770 (Irwin 1973:12, Fig. 4). In the space between the botehs, a rose sprig rises from the lower tanjir (the palla's lower border). Above, a small pot of flowers magically floats between botehs. In the base of each boteh reside three radial flowers (Fig. 3.2). The central one rises directly from the larger red vase; two lateral ones from small yellow pots each sit atop a leafy support. Tanjirs above and below the end panels show alternate red and blue flowers, perhaps a full-face red rose and a blue carnation in profile. But the lower tanjir differs slightly in that a red rose in profile has been substituted for the full-face, yellow-centered red rose above. Four flowers repeat regularly along the patka side panels or hashias: yellow lily, red rose, blue carnation, and crimson Persian lilac. This fragrant quartet forms 11 cm repeats and is borne on a meandering leafy green vine with many tiny florets and buds. It is a relatively complex design and a delight in its finely woven clarity of expression, thanks to its high thread counts and unstinting workmanship. The same four flowers in the same coloring and sequence frequently appear in hashias of late eighteenth-century Kashmir patkas. The 34 knots preserved in the fringe at one patka end occur about 1 cm beyond the end of the weaving. This is a self fringe formed by extension of the patka warps, a common practice in the late eighteenth century. These were tied off with single overhand knots. Presumably there was once knotting across the entire woven width, but friction has taken

Koetz s "Turbans "

19

Figure 3.2. Radial flower, late eighteenth-century motif. Drawn from Persian patka, VMMA 17313.

its toll both here and at the opposite patka end, where the original fringe is not completely worn off, leaving a frayed edge. In trying to determine a date for this patka, a number of traits must be considered, including its size, high thread counts, light weight, narrow hashia ,relatively shallow pallas, number of colors, and knotted self fringe . Together with the design characteri tics depicted by this assemblage, it seems reasonable to attribute the Koelz patka to the late eighteenth century.

Note 1. A measure varying in size in different countries; in England, 45 inche . Therefore, even English ells would equal 8.75 yards.

-4-

Kashmir and Persian Shawls in Their Historical Context

T

he flowered and opulently figured wool and goat hair shawls of Kashmir and Persia (modem Iran) are well known and highly esteemed. In the eighteenth and nineteenth centuries, Kashmir was the more renowned center of production; Kashmir shawls were widely regarded as superior to the Persian shawls. However, during the latter half of the nineteenth century, the Kashmiri shawl industry virtually collapsed, overtaken by the machine-driven shawl industry of Europe. Perhaps best known of the rival European shawls are those woven in the Scottish town of Paisley, a name frequently associated in the public mind with Kashmir shawls. The dominant, conventionalized teardrop-shape floral element, known as boteh in the Kashmir and Persian shawls, came to be known in Europe and America as the paisley motif. To properly understand the discussion of the forty-three woolen shawls that Koelz collected, one must know some ofthe context of their production and use. The enthusiastic European discovery of the Kashmiri textiles revolutionized the economics oftheirproduction over the course of the nineteenth century, and that process is detailed in this chapter. All forty-three ofthe Koelz woolen shawls are hand-woven in a technique known as 2/2 twill tapestry, in which a figure is worked in a 2/2 twill ground, a weave to be described later. Nearly all were purchased from antique shops in northwestern India, with the exception of two from Lahore in present-day Pakistan and two that are undesignated. Based on the provenience reported to Koelz, they are catalogued as ten from Kashmir, thirty-one from Persia, and two unknown. Efforts to distinguish the two main groups by criteria other than reported provenience-whether by size, coloring, design, thread count or other traits-proved futile. This is not to say that there are not differences. Indeed, Koelz remarks ofthe antique shops in Isfahan, for example, that among other items were "shawls of the quality we had learned in India to regard as Persian" (Koelz 1983 :49). Unfortunately, he did not elucidate. In this study, all of the woolen shawls are treated as a single group. However, much of the

20

Kashmir and Persian Shawls in Their Historical Context

21

general background presented revolves around Kaslunir rather than Persia because the Kaslunir shawl excelled in quality and because its production is better reported, thanks to the writings of colonial travelers and subsequent scholarship. Indeed, with the exception of the distinctive piece UMMA 17348, discussed in Chapter 2, there is little reason to question a Kasluniri origin for all of the shawls in the collection. In this chapter, I provide detail on the history and contexts of production and use of the Kashmir shawl. Where possible, I also include information on Persian shawls. I consider how European influence affected shawl design and production in the nineteenth century, leading ultimately to the decline of the shawl industry in Kaslunir. My goal in this chapter is to provide an overview to allow the reader to better understand the extraordinary labor and skill involved in the production of the elaborate and beautiful shawls in the Koelz collection, and the social significance they had to their European and Asian consumers.

The Historical Background of the Kashmir Shawl Early Development: Seventeenth and Eighteenth Centuries

The precise beginning of the shawl industries in either Kaslunir or Persia is clouded at best. Both regions have long histories of textile production, and the specific techniques employedtapestry and twill weave-can be traced back at least as early as the seventh century A.D. in Persia (Geijer 1979:84-85). Indeed, twill weave is visible on clay impressions recovered in archaeological sites in the Ferghana Valley (southeast of Tashkent) and dating as early as the eleventh to fourth centuries B.C. (Korobkova 1962, cited in Barber 1991 :211-12). Specific evidence for the production of shawls comes much later. During the Persian Safavid Dynasty (1502-1736), elaborate shawls for elite consumers were produced in workshops, or tiraz, in Kerman and Khurasan. In Kashmir, the earliest known pieces of the classic "Kaslunir shawl" date to the seventeenth century, and these are few and fragmentary. The eighteenth century is somewhat better represented, but the great body ofKaslunir shawls in museums and private collections today is of nineteenth-century manufacture. Although this cannot be confirmed, shawl weaving in Kaslunir is said by some to have been established in the fifteenth century by the ruler Zain-ul-Abidin (r. 1420-1470), who brought in Turkistani and Persian weavers for the purpose. As a youth, Zain-ul-Abidin spent seven years as a hostage at Timur's court in Samarqand, where he acquired a taste for the many luxuries produced there. Upon succeeding his father to the Kasluniri emirate, he sent emissaries to Central Asia, Persia, and Arabia to enlist artisans of every calling for his own court. Among them were weavers of shawls (Gans-Ruedin 1984:31). To this day, the weaving technique and tool terminology used in Kaslunir is entirely Persian, which lends credence to this account and to the possibility that the fifteenth century marks the advent of a Persian style of weaving in Kashmir. A sixteenth-century impetus to Kasluniri shawl weaving was provided by the Mughal emperor Akbar (r. 1556-1605). Akbar was an enthusiastic admirer of the shawls ofKaslunir and kept his wardrobe well stocked, even suggesting innovations in their wear and design.

22

Wrapped in Beauty

Akbar greatly encouraged the shawl industry by importing to his court Persian-trained weavers. It is less clear what his impact was within Kashmir itself. Although Akbar 's armies had conquered almost all of northern India by 1576, it was not until 1586 that he added Kashmir to his domain, and he could not have had direct impact on production until then. Thus it would seem that Akbar's influence on Kashmiri shawl making was that of patronage rather than introduction. During the late sixteenth century the reigns ofAkbar and of Shah Abbas the Great of Persia (r. 1587-1629) overlapped, and their courts rivaled each other in brilliance. Geographically they were about as far apart as Chicago and Salt Lake City, with mountains and deserts between. But the connecting route, part of the lower ancient Chinese Silk Route, was well traveled. A lively interchange of textile techniques and motifs passed along this artery, as did artisans themselves, for workers as well as goods circulated among the courts of Asia. Therefore, the question of the "beginning" of shawl weaving in Kashmir is awkward at best. Still, it is interesting. If already in the sixteenth century the shawls were renowned as gifts and sent to distant countries (Irwin 1973:9-10), as well as hoarded as a form of wealth (Wilson 1979[ 1866]: 166), then it seems likely that an earlier foundation was already in place. Undoubtedly, a weaving tradition of some sort was already present in both Kashmir and Persia. What we know ofthe history of the Silk Route suggests a greater age and wider knowledge of 212 twill tapestry weave than has been generally considered. In both Iran and Kashmir, tapestry shawls were originally the prerogatives of the elite, worn by kings and royal courtiers. Persia had long held a position of preeminence in the arts and was no doubt the fountainhead for the Kashmiri style of shawl weaving. Under the Sassanids (224-651 A.D.), Persian textiles set a high standard for excellence, especially the famous patterned silks whose designs were widely copied long after that dynasty's decline. In the sixteenth century, when Shah Abbas I again made Persia a great power through military exploits and effective administration, he also actively encouraged commerce and the arts. Silk manufacture was a royal monopoly. Weaving flourished. The best Persian carpets date from the sixteenth and seventeenth centuries. The pomp and pageantry of the court of the seventeenth-century ruler, Shah Abbas II (r. 1642-1660), dazzled European envoys and travelers. It is difficult to image that shawls of the period were other than of the highest quality. Subsequently, however, Persian shawl weaving apparently declined in the eighteenth century, a period of regional chaos, siege, plunder, assassination, and intrigue coupled with probing for advantage by rival European powers. It did not recover until the nineteenth century. Ames (1986: 102-4) summarizes some of the observations made by European travelers to the region on Persian shawl weaving. Such travelers, bent on learning about the shawl weaving industry, often focused on Kashmir and ignored Persia almost entirely. However, seventeenthand eighteenth-century European visitors did comment on how highly the Persians valued weavings of goat fleece, reporting fabrics of goat hair more delicate and expensive than those of silk. They also described the Persian nobleman's passion for wearing exorbitantly expensive sashes, or patkas, wrapped about the waist and with elaborately patterned ends falling in front to the knees. Some were of silk ornamented with flowers in gold; others were woven from the fine wools of Kerman, a weaving center in south-central Iran.

Kashmir and Persian Shawls in Their Historical Context

23

Little mention was made of shawls per se until G.A. Olivier, a French emissary in the 1790s, reported on the large shipment of Kashmir shawls that arrived annually in Baghdad by caravan. He added that Persian shawls woven in Kerman were exported, but they had "neither the beauty nor the fineness" (in Ames 1986: 103) ofthe Kashmiri shawls. Even if not ofthe perfection of Kashmir, Kermani shawls woven from the "fleece of baby goats grazed on the mountains surrounding Kerman" (Ames 1986: 103) were acceptable commodities, worn at home and shipped abroad, even to India. Other shawls were made in Yezd, Kashan, and Baghdad itself. Up until about 1800, the faults of Persian shawls, in European eyes, seem to have been their less imaginative and more limited variety in design, as well as their coarser texture when compared with shawls of Kashmir. From the second quarter of the nineteenth century, Persia competed with Kashmir in the shawl trade. Persian shawls continued to differ from those of Kashmir in quality and in design, particularly in the use of border floral treatments and greater architectural emphasis. In addition, a predominant color in Persian shawls was a deep red not characteristic of Kashmir (Irwin 1955: 17). Ames suggests that Kashmiri influences began to appear in Persian manufactures around 1820, as the former buti-filled Persian stripes gave way to pronounced allover patterns with motifs then prevalent in Kashmir (buti is a small boteh). Unlike in Kashmir, there was no transitional development of the boteh/paisley motif in Persia, suggesting that this was a direct import rather than a local development. In addition, natural calamities in Kashmir during the 1830s resulted in the migration of many Kashmiri weavers to the Punjab. It is possible that some emigrated even further and found employment in Kerman (Ames 1986: 104). As noted above, in my study ofthe Persian and Kashmir shawls in the Koelz Collection, I have been unable to differentiate between them by any objective traits. Of course, the sample is small and some shawls are constructed of disparate pieces salvaged from earlier shawls. Moreover, most date to the nineteenth century when patterns converged. Inasmuch as Persia in the eighteenth century was in turmoil and artistic production was limited, an interesting comparison might be made between Persian and Kashmiri textiles of the seventeenth century through examining illustrations in portraits, even where preserved fabrics are scarce. This, however, remains a project for future scholars. Nineteenth-Century Production In the nineteenth century, shawl design in Kashmir received a powerful external stimulus and change of course when European attention impinged upon local tastes. The virtual craze for Kashmir shawls that seized European ladies of fashion in the nineteenth century had its beginnings in the late eighteenth century. Shawls began to find their way into fashionable European wardrobes, some brought by travelers, some brought to English ladies by men of the East India Company. But the greatest fashion influence was exerted by the influx of shawls presented to French ladies by Napoleon's officers returning from the Egyptian campaign. In 1798, Bonaparte successfully attacked Egypt, at that time part of the Turkish empire under Selim III. Among the booty gathered from Turkishjanissaries were Kashmir shawls, which were adopted by delighted French court ladies and became high fashion. The

24

Wrapped in Beauty

Empress Josephine, ever extravagant, is said to have owned as many as sixty (Levi-Strauss 1987:19). Although the Turks are said to have favored buti-striped shawls from Egypt, the women of the French court seem to have been more enchanted by the boteh palla among the Mameluke shawls from Egypt, to judge from elegant portraits of the time. Not only was the shawl high fashion, but its high cost made it a status symbol, and its exotic origin and design fed into nineteenth-century romantic enthusiasm for all things Oriental. Fashion coincided with economy and politics as Europeans expanded commerce and colonies in Asia. Aspiring wives of the century's newly rich industrialists followed the style of their aristocratic sisters and shawls became essential items of dress. Supply could not meet demand and European manufacturers hastened to share this lucrative market. Not only did they strive to copy styles of Kashmir, but also to modify designs for European tastes. Eventually, they also sent designs to Kashmir to be woven and then returned to Europe for sale. This situation lasted through the first three quarters of the nineteenth century, by which time the shawl craze in Europe had run its course. At the same time, as Persia became more stable, shawl weaving there revived, and by the second quarter of the nineteenth century, Kashmir weavers faced competition from Persian looms. Both countries, however, experienced similar problems caused by European demands: the demand both to produce far greater numbers of shawls and to modify their forms to meet the desires of their new clientele. A dramatic change occurred as old patterns gradually were abandoned and Kashmiri weavers undertook to produce designs drawn in Europe. These designs entailed European interpretations of Asian styles, and strongly reflected the increased mechanical potentials for pattern expression resulting from technological advancements in weaving technologyspecifically, the invention ofthe mechanized Jacquard loom early in the nineteenth century (Karwatka 1999: 17). The draw loom, formerly used in Europe for pattern weaving, could weave repeat patterns of limited size-according to Ames (1986: 114), of no more than 12 to 15 cm. The possibilities for pattern size were greatly expanded by the Jacquard loom, which, with accumulating improvements, was capable by mid-century of weaving a single pattern occupying virtually the whole width of the shawl. Hand-weavers of Kashmir found themselves competing with products made by the Jacquard machine. To cope with the altered market situation, hand weaving procedures changed and labor conditions deteriorated. The Koelz shawls reflect many of these changes.

The Koelz Collection in Its Broader Contexts It is not essential to undertake a full accounting of nineteenth-century shawl weaving developments in either Kashmir or Persia in order to discuss the Koelz shawls. Yet a wider background is both interesting and helpful. For the interested reader, a great deal of information is ably presented by Frank Ames in The Kashmir Shawl and Its Indo-French Influence (1986) and by Monique Levi-Strauss in The Cashmere Shawl (1988). Both authors describe the French shawl industry as well as Kashmiri production and show fine collections in colored plates. John Irwin's earlier book, The Kashmir Shawl (1973), is brief but well

Kashmir and Persian Shawls in Their Historical Context

25

rounded and focused on English involvement in Kashmir. To all three of these authors I am particularly indebted. As noted earlier, because relatively little appears to have been written about the Persian shawl industry, it receives less attention here. Many developments in Persian shawl making may be assumed to parallel those of Kashmir. In the following material, for either country, only background information deemed pertinent is introduced as various aspects of the Koelz shawls are in tum considered. Parts of a Shawl

Throughout this work, I will use local, predominantly Persian, terminology to describe the various elements of shawls. Figure 4.1 provides schematic illustrations of long and square shawls, with their main elements-field (matan), palla (end borders) and hashia (side borders)-Iabeled. Size and Shape

Kashmiri and Persian shawls were made in two forms: the dochalla or long shawl; and the rumal or square shawl (Ames 1986:49). Of these, the long shawl is generally regarded as the older form and in Central Asia was worn primarily by men, thrown around the shoulders for display or draped about the body for warmth. Many were of generous size, about comparable to a single-bed sheet, and therefore were versatile garments with many uses. Study of the Koelz shawls and review of published sources (e.g., Ames 1986; Irwin 1973; Levi-Strauss 1987) reveal that both long and square shawls increased in size throughout the nineteenth century. These changes may be attributed to changing consumer tastes stemming from the European market, as women's costume evolved from the slender flowing silhouette of the Empire period toward more substantial and voluminous full-skirted fashions. Perhaps the earliest shawl measurements we have are those reported by Franyois Bernier, the first European of record to visit Kashmir, in 1664 (Bernier 1968[1891]). He recorded shawl length as 180 cm (71 in) and shawl width as 120 cm (47 in). He also noted the delicacy and softness of Kashmir shawls and reported them as being worn by both men and women in winter (Levi-Strauss 1987:14). Using Bernier's figures, the total area of seventeenthcentury long shawls might be assumed to be approximately 21,600 sq cm. Comparable figures for the Koelz long shawls, taken as representative of the nineteenth century, range from 34,305 to 50,552 sq cm, with a median of37,788 sq cm. The size oflong rectangular shawls, then, nearly doubled in two hundred years, with an accelerating increase during the nineteenth century. Square shawls were worn by men as early as the eighteenth century. Certain styles, especially the moon shawl (chandar), were favored by court ladies of Persia. Upon becoming popular among nineteenth-century European women of fashion, their early dimension of about 130 cm per side increased to approximately 180 cm per side by mid-century (Ames 1986:56-58). Square shawl woven area nearly doubled in just over half a century, from 16,900 to 32,400 sq cm. During the first quarter of the nineteenth century, square shawl production greatly accelerated in response to European demand. The moon style peaked

26

Wrapped in Beauty

lIIilIlU\!l\1111l1l1111111\IIHIlIIIIUllllIUIIIIIIIIIIIlII111\l\lIIlIlIlII\1\\II1I11I\\\\\lIl111III11Il\\\I\\IIIll\ll\\h

palla

gallery

_~-i-I--

field-+-+-..

hashia tanjir or end border 1--C===============~==~Lfringe 1lillllJlIIIIIIIIJIIlillllllllllllllllllllllllllllllllllllilIJIUIIIIIIIIJlIIIIIIIIIIIIIIIJII'IIIIIIIIIIIIJIIIIIIII/JI1II1I1

gafe

1~11111l\Ifollllll~\II~mlIL~~li'IIVl'IIIIIIIII~nlll~IIIU1I1IIt1'II\I\IUalill~:ttlf~11I~IUlllmlllllnanll~UJl~IIi'f:II'!JIIIIIII[';']illl~Llltll'II,!Jlntlllllil\lln~nlIWlb1dllllilIIIIIII'1

~tllll

Figure 4.1. Parts of a shawl.

during this early period, and was followed by other square styles, predominately either field gallery or quatrefoil composition. For both long and square shawls, the doubling of woven areas over the nineteenth century seems remarkable given the increased demand for greater numbers of shawls. Dimensions and shapes of the Koe1z shawls are illustrated in Figure 4.2. Shawl length is usually the warp-wise direction ofthe field and includes fringe gates (i.e., fringes and the immediate fabric from which they descend). The Koelz shawl dimensions clearly indicate only two shapes-long rectangles and square-with little overlap between. It should be mentioned however that some ofthe shawls appear to be remakes destined for the antiques market, where their beauty and fine weaving would still have significant value. With a few exceptions, the remakes are consistent in size with the shawls in original condition. In comparing Koelz shawls with other long shawls, data drawn from various published sources (Ames 1986; Irwin 1973; Levi-Strauss 1987; Singh 1979; Watson 1982[1866]) indicate that nineteenth-century long shawls averaged around 250 to 350 cm in length. Koelz long shawls accord with these measurements, though they average on the shorter end of the range, perhaps reflecting the number of remodeled shawls in the collection.

Kashmir and Persian Shawls in Their Historical Context

27

The Koelz square shawls accord well with the 180 cm size suggested by Ames (1976:57) as typical. Shawl types within this group are fairly evenly distributed, except for the piecework quatrefoils, which are among the largest. The sizes of strata-field and conglomerate piecework shawls are shown in Figure 4.2. The fact that strata-field square shawls all fall within the mainstream width ofiong shawls argues for their being remakes from pallas offormer long shawls. The small sizes of conglomerate types no doubt reflect their having been assembled from suitable scraps. Long shawl widths didn't vary as much. This is a function of traditional loom widths. Kashmiri looms typically were constructed for three men sitting abreast. The weavers can't crowd together any closer than three body-widths, and they can't spread any farther apart than 40 or at most 50 cm and still effectively count and control the complex tapestry patterns. Individual weavers had to count warp yam pairs with a density of 17 per cm (43/in); a 40 cm width yielded 680 warp pairs or 1,360 threads. During actual counting, half the pairs would be raised to form the shed, so weaver count would be eight pairs per cm (20/in), or 340 warp pairs for 40 cm. This is close work and weavers commonly ruined their eyes by age forty. The average Koelz long shawl width is about 140 cm. Divided by three, this allows for a width of 47 cm to be worked per weaver. This figure is a little high because the separately woven hashia panels and occasional additional, separately woven, wide field borders or galleries have been included in total shawl widths. Slightly narrower widths (120-140 cm) are reported for a preponderance ofthe nineteenth-century Kashmir long shawls conserved in Jaipur's Maharajah Sawi Singh II Museum (Singh 1979:95-101). Six eighteenth-century shawls are in the same width range, which allows a space of 40 to 47 cm per weaver. This estimate is supported by Leitner's account of the number of yams prepared for warping the loom (Leitner 1882: Appendix, xxvi). When he wrote in the 1870s, the development of piecing shawls from multiple textile segments had led to the use of three loom sizes, designed for operation by one, two, or three men, the last being the most widespread. Using the weaver's term of nals (pairs of warps that occur in 2/2 twill, see section on weave), Leitner states that warping was done in standard nals of700 for the small loom, 1,300 for the middle-sized, and 1,900 for the large. Doubling these numbers to get total warps and dividing by 34 (the median warp count for Koelz shawls) leads to a width controlled by one man of 41 cm for the small loom, 38 cm for the two-man looms, and 37 cm for the large. Thus, for the close work of tapestry weaving, the width of the weaver's shoulders dictated loom widths, constraining cloth widths, which in tum became a modular unit for shawl construction. This relationship is borne out by examining the widths of the various Koelz tapestry pieces selvedged on both sides and incorporated into the shawls. There are more than forty such pieces. Rounded to the nearest 10 cm, the widths of these pieces range from 40 to 180 cm. Half are in the 120 to 140 cm range, which suggests a three-man loom, with each man weaving 40 to 47 cm. The only other selvedged widths that occur in any frequency fall at 40 cm (presumably woven on a one-man loom) and at 160 to 170 cm. Ifwoven by three men, the latter wider widths would require each to work in the lower 50 cm range, atypical but feasible. Thus, several approaches lead to the same general conclusion: individual weavers usually worked a 40 to 45 cm width, plus or minus a few centimeters.

28

Wrapped in Beauty

400 390 380 370 360 350

. .

::::::::::::::::::::::::: ....

340 330 320 310 300 290 280 270 260 250 240 230

~