Textiles in Indian Ocean Societies (Indian Ocean)

RoutledgeCurzon Indian Ocean Series Editors: David Parkin and Ruth Barnes University of Oxford There is a need to understand the Indian Ocean area as a cultural complex which should be

Textiles in Indian Ocean Societies

This book concentrates on textiles as a major commodity, and primary indicator of status, wealth and identity in Indian Ocean regions Lavishly illustrated, it represents invaluable, and entirely new research

Textiles in Indian Ocean Societies considers the importance of trade, and the

transformation of the meaning of objects as they move between different cultures It also addresses issues of gender, ethnic and religious identity, and economic status The book covers a broad geographic range from East Africa to South-East Asia, and references a number of disciplines such as anthropology, art history and history

This volume is timely, as both the social sciences and historical studies have developed a new interest in material culture Edited by a foremost expert in the subject, it will add considerably to our understanding of historical and current societies in the Indian Ocean region

Ruth Barnes works at the Ashmolean Museum, Oxford University Her many

previous publications include Dress and Gender: making and meaning in cultural

contexts (co-editor); The Ikat Textiles of Lamalera: a study of an eastern Indonesian weaving tradition; Indian Block-Printed Textiles in Egypt; and Weaving Patterns of Life

RoutledgeCurzon Indian Ocean Series

Editors: David Parkin and Ruth Barnes

University of Oxford

There is a need to understand the Indian Ocean area as a cultural complex which should

be analysed beyond the geographical divisions of Africa, the Middle East, the Indian subcontinent, and South-East Asia, as its coastal populations have intermingled constantly The movement of people, goods and technology make it imperative that spatial concepts and the role of material culture be central in the study of the region by archaeologists, historians, ethnographers and anthropologists

Islamic Prayer Across the Indian Ocean

Edited by David Parkin and Stephen C Headley

Ships and the Development of Maritime Technology in the Indian Ocean

Edited by David Parkin and Ruth Barnes

Sufis and Scholars of the Sea

Anne K Bang

Textiles in Indian Ocean Societies

Edited by Ruth Barnes

Frontispiece Block-printed and painted cotton textile from India’s

Coromandel Coast is kept as a family heirloom in Eastern Indonesia, along with shell and ivory bracelets; the ivory is from East African or

Indian elephant tusks Lamalera, Lembata Photograph: Ruth Barnes

(1982)

Textiles in Indian Ocean Societies

Edited by Ruth Barnes

LONDON AND NEW YORK

First published 2005 by RoutledgeCurzon 2 Park Square, Milton Park, Abingdon, Oxon OX14

4RN Simultaneously published in the USA and Canada by RoutledgeCurzon 270 Madison Ave, New

York, NY 10016

RoutledgeCurzon is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2005

“ To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of

thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.”

© 2005 Ruth Barnes for selection and editorial matter; individual contributors

their chapters All rights reserved No part of this book may be reprinted or reproduced or utilised in any form or

by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission

in writing from the publishers

British Library Cataloguing in Publication Data A catalogue record for this book is available

from the British Library

Library of Congress Cataloging in Publication Data A catalog record for this book has been

requested ISBN 0-203-64425-5 Master e-book ISBN

ISBN 0-203-68804-X (Adobe e-reader Format) ISBN 0-415-29766-4 (Print Edition)

Contents

Introduction

Far-flung fabrics – Indian textiles in ancient maritime trade

The kofia tradition of Zanzibar: the implicit and explicit discourses of

men’s head-dress in an Indian Ocean society

Illustrations

1.1a

Key find spots and production centres of cotton in the Roman

11

1.1b&c

Resist-dyed cottons from Berenike: b (above) design with

possible lotus-bud motif; c (below) fragment with

incomplete triangular motif

14

2.1 Indian block-printed cotton fragment, Gujarat 17 2.2 Indian block-printed cotton textile, Gujarat 18 3.1 ‘Portuguese’ carpet, Khorasan, North-Eastern Iran 37 3.2a Asymmetric (Persian) knot open to the left 39 3.2b Asymmetric jufti (false) knot open to the left 39 3.3a Jufti knots bound regularly by only two weft passes 40 3.3b Asymmetric knots bound regularly by three weft passes 41

5.4a Drawing concentric circles using Popsicle stick tool 67 5.4b Attaching two pieces of the crown by sewing over-top the 68

5.7b Group of women embroidering kofias at a workshop 70

The thigh-supported spindle used for spinning cotton in the

88

6.7

Detail of an akotifahana, a cloth of reeled silk with

Tandroy women dancing at a mortuary ceremony with their

7.4 Siwa image and relief panels of garbhagriha (detail) 109

7.6 Detail panel #16 (scene 36) Candi Siwa, Loro Jonggrang 112 7.7 Vestibule, Candi Siwa, Loro Jonggrang complex 115 7.8 Kawang panels Candi Siwa vestibule, Loro Jonggrang 116

7.10 ‘Celestial roundel’ panel Candi Siwa, Loro Jonggrang 119

7.11 Floral panel (detail) Candi Siwa, Loro Jonggrang 120 7.12 Floral panel Candi Siwa, Loro Jonggrang 121

8.1 Weft ikat kain limar sarung (detail) from Jambi 131

8.3 Gold thread embroidered cushion end from Jambi 134

8.4 Sembagi cloth (detail) collected in Jambi 136

8.5 Contemporary siang malam cloth from Jambi 137 8.6

Silk selendang decorated with pelangi technique and with pauh

139

8.7 Jambi batik (detail) with durian pecah motif 140

8.8 Jambi batik (detail) with batanghari motif 142

9.1 Double ikat silk patolu, made in Gujarat and traded to Indonesia 147

9.2

Kewa Payong Amuntoda wearing an eighteenth-century head

Contributors

Ruth Barnes is research cataloguer of textiles in the Department of Eastern Art,

Ashmolean Museum, Oxford

Steven Cohen has written his Ph.D at SOAS on the representation of Indian carpets in

early Mughal miniature painting, and publishes on Indian textiles and carpets

Sarah Fee is an anthropologist who has spent several years of field research in

Madagascar and has recently completed her Ph.D in Paris She is currently associated with the National Museum of Natural History, Smithsonian Institution, Washington, D.C

Zulfikar Hirji is a Junior Research Fellow at Wolfson College, Oxford and Head of

Ismaili Living Traditions, Institute of Ismaili Studies, London He has written his D.Phil thesis based on research in Zanzibar and Oman

Fiona Kerlogue wrote her Ph.D at Hull University on the batik textiles of Jambi,

Sumatra She is Deputy Keeper of Anthropology at the Horniman Museum, London

David Parkin is Professor of Social Anthropology at the University of Oxford and a

Fellow of All Souls College

Himanshu Prabha Ray is Professor of History at Jawaharlal Nehru University, Delhi Mary-Louise Totton is an Assistant Professor in Asian Art History at Western Michigan

University who has written her Ph.D at the University of Michigan on the Central Javanese temple complex at Prambanan

Felicity Wild is an archaeologist who has worked at the site of Berenike, Egypt

John Peter Wild is Professor Emeritus at the School of Art History and Archaeology,

University of Manchester

of disciplines; on this occasion we had anthropologists, archaeologists, historians, and art historians present, both as speakers and as discussants

Textiles have been a major commodity in Indian Ocean societies from early historical times onwards to the present, both as trade items and as local products This was realised

by the first Europeans when they arrived in the region around AD1500, in search of spices and aromatics and with the desire to dominate this lucrative trade They discovered that textiles were the predominant item of exchange, taking the role of an international currency Without a stake in the trade in textiles, one did not have access to the markets

of Asia They entered a region that was extremely cloth-conscious Textiles were a major distributor of artistic design They also were a means of defining a person’s status and gender, a role they continue to play Then and now the great demand for cloth can only

be explained by understanding the importance of textiles in local societies The workshop convened attempted to make a contribution to this particular issue

In addition to papers given at the time, two articles were written especially for this volume (Steven Cohen’s and my own) Not all presentations were available for publication, but we gratefully acknowledge the contributions made by Mattiebelle Gittinger, John Guy, and Nandita Khadria Their participation in the workshop was most valuable The conveners also want to thank the British Academy for a travel grant which covered travel costs for Himanshu Ray and Nadita Khadria The Asian Studies Centre at

St Antony’s College provided much appreciated hospitality, and we thank Dr Steven Tsang for the support he gave us Gina Burrows from the Institute of Social and Cultural Anthropology was responsible for much of the organisation of rooms, accommodation, and travel details, and she had to solve many last-minute problems She also took on the final preparation of the manuscript (assisted by Nadine Beckmann), including the collation of bibliographical entries, for which I am deeply grateful

Ruth Barnes

Ashmolean Museum

Map of the Indian Ocean region Note: Regions are numbered in relation to which chapter features them

Introduction

Ruth Barnes

Weaving is one of the oldest technologies, in many places predating pottery and certainly preceding metallurgy The processing and manipulation of fibres for weaving purposes was developed in Asia and the Near East at some time between 7000 BC and 6000 BC, with archaeological evidence for the use of both horizontal and vertical looms dated prior

to 6000 BC.1 While the function of textiles may initially have been protection against the elements, it soon acquired a social dimension As we can see in the elaborate forms of burial dress from Ancient Egypt, Central Asia, and North-Eastern China, textiles were used as a primary indicator of status, wealth, and ethnic or gender identity in human societies Writers of Mediterranean antiquity already mentioned that there was considerable demand for the exotic silks of China and the fine cotton muslin of India Textiles are fragile, though, and only survive under certain conditions; the dry climates

of, for example, Egypt and Central Asia, have preserved numerous ancient fabrics For the cultures of the Indian Ocean littoral there is little primary evidence that predates the Christian era, although small fragments of cotton fibres have been found at the Harappan site of Mohenjo Daro.2

Once historical documents can be referred to it becomes clear that textiles were a major commodity transmitted between Indian Ocean societies Both indigenous and traded fabrics had a significant cultural role, from East Africa to Indonesia, and from Arabia to Sri Lanka While this has been recognised in the past and is often mentioned in passing by historians of the Indian Ocean, so far no single volume has actually followed

up on this particular topic, or considered the question of why textiles are given such importance This collection of essays attempts to redress this issue and therefore considers the role of textiles in various societies with direct contact to the Indian Ocean Before exploring some of the issues set out in this publication, though, the non-specialist

in Indian Ocean studies may find it helpful to be referred to a small number of general works

Scholars have emphasised in the past that this particular maritime environment–like the Mediterranean – is a sea that connects rather than separates different cultures The scholarship on the subject is vast, of course, and it has involved historians of classical antiquity, India, the medieval Islamic world, and of Europe’s involvement with Asia after

1500, with some excursions necessary to draw on Chinese sources, as well For background to the history of Indian Ocean studies, Chaudhuri (1985) provides an accessible introduction He attempts to analyse the history of Indian Ocean societies in

the spirit of Fernand Braudel’s longue durée, as the latter applied it to the Mediterranean

with an emphasis on geographical and cultural spheres, rather than a historical understanding primarily determined by political and economic alliances.3 It is tempting to see the Indian Ocean in this light, and to draw out the often astonishingly close relations that have existed over vast geographical distances But the emphasis on unity can also

distract from the diversity explicit in local political and economic histories, as well as ethnographic accounts A balance has to be found between the two D.S Richards’s

edited volume Islam and the Trade of Asia was published more than thirty years ago

(1970), but still is a good introduction to the issues that concern scholars working in different geographic and historical areas of the Indian Ocean S.D Goitein’s publications (1963, 1967, 1971, 1978, 1983, 1988) on the eleventh- and twelfth-century Genizah papers from a synagogue in Old Cairo are very detailed and as a whole cannot be suggested as an introduction, but they do provide wonderfully humane insights into the life of communities connected with the western Indian Ocean Several symposia held in the 1990s have contributed substantial publications to the study of Indian Ocean archaeology and history (Boussac and Salles 1995; Ray and Salles 1996; Ray 1999) Abu-Lughod (1989) attempts an ambitious account of the historical and economic links between the different geographic and cultural spheres of Asia, the Indian Ocean, and the Mediterranean prior to the rise of European dominance; for its extensive collection of sources alone, her book remains a key introduction

Making textiles the focus of this volume means that it deals primarily with material culture Our contributors come from a variety of disciplines: archaeology, anthropology, history, and art history A few words on this interdisciplinary mixture may be useful For many decades the study of objects was largely discredited in the social sciences, and in art history the focus was heavily weighed towards aesthetics and stylistic analysis, often with only minor attention given to social context This meant that social historians and anthropologists on the one hand, and art historians on the other, had few interests in common In the 1970s and early 1980s, however, a shift in attitude towards material culture occurred I became aware of this change with the publication of Michael

Baxandall’s Painting and Experience in Fifteenth Century Italy (1972), and the reading

of Francis Haskell’s Patrons and Painters, already published in 1963, but not available in

an accessible paperback edition until 1971 Both studies had a formative influence on the art historical thinking of the time They helped to move that discipline away from the vagueness of style analysis, which was still prominent then in Britain, towards an approach that once again made greater use of social history This was by no means new

to the subject, but represented a return to the interests of many of the founding scholars in art history, such as Erwin Panofsky, Johannes Wilde, Wilhelm Fraenger, and Aby Warburg The field of art history, in its main stream of course an object-focused discipline, was now taking a new interest in the social role and significance of the material it studied

In the social sciences, in particular in social anthropology, this approximately coincided with a rediscovery of the world of objects, long since out of fashion and relegated to the historical corner of the discipline and a period that had been preoccupied

with evolution and migration theories Appadurai’s edited volume The Social Life of

Things (1986) had perhaps the most striking impact, no doubt because it was published at

a time when archaeologists and anthropologists were beginning to think again about the relationship between the making and using of artefacts, and the conceptual framework that this activity implies At some time during the more than twenty years that passed

between the publication of Andrew and Marilyn Strathern’s Self-Decoration in Mount

Hagen (1971) and Alfred Gell’s Wrapping in Images (1993) it became intellectually

interesting again for anthropologists to consider visual and material culture

Textiles in Indian ocean societies 2

It is relevant for this publication that the shift also coincided with a new approach to textile studies and the investigation of textile history and production Long dominated by either the study of technology, or the treatment of textiles as a minor part of art and economic history, the subject acquired a new ‘social life’ when scholars entered the field who had an interest in both art history and anthropology In African studies, this was first

apparent in Roy Sieber’s exhibition catalogue African Textiles and Decorative Arts (1972) and Robert Thompson’s African Art in Motion (1974), which was primarily a

study of the dress of West African masquerades For the Indian Ocean region, Bühler and

Fischer’s monumental study of The Patola of Gujarat (1979) was of foremost

importance Bühler’s major interest had long been in the history, geographical distribution, and technology of resist dyeing, and he had pursued this investigation in a series of meticulous but to the non-specialist often heavy-going publications, the

culmination of which was his three-volume study Ikat Batik Plangi (Bühler 1972) In The

Patola of Gujarat, however, he and Fischer moved beyond technology and also

investigated the social significance of a particular type of textile, the complexly patterned

double-ikat silk patola made in North-West India The patola were (and are) luxury

cloths for the Indian markets, but they also have played an important international role The publication therefore is a detailed investigation and account of local production and

design, but it combines that with a look at the social role of patola textiles, not only in

India, but once they were transmitted into a different cultural context As Bühler had

noticed when studying Indonesian ikat designs, patola were important as prestige textiles

traded to South-East Asia in particular, and their designs had a major impact on many of the indigenous textiles (Bühler 1959) This study opened up the way for several in-depth investigations by others who took a close look at textiles and their functions in the maritime region

No one did more towards establishing the field than Mattiebelle Gittinger Her

publications Splendid Symbols: textiles and tradition in Indonesia (1979), Master Dyers

to the World (1982), and Textiles and the Tai Experience in South-East Asia (Gittinger

and Lefferts 1992) are evidence for the emergence of a scholarly discipline They present three distinctly different aspects of Asian textiles in a scholarly manner: they introduce two South-East Asian traditions, as well as the cross-cultural significance of Indian textiles Her work inspired a new research generation The development of scholarship is perhaps most evident in the three symposia on Indonesian textiles, held at six-yearly intervals in Washington (1979), Cologne (1985), and Basel (1991) The proceedings record how over twelve years a new field evolved for the South-East Asian region, remarkable for its interdisciplinary nature, with anthropologists, historians, and art historians representing their subjects and finding it fruitful to expand their views through the medium of textiles.4 The progression of the field showed that ‘the most compelling entry for any critical discussion of [dress and textiles] is through particular, fine-grained ethnographic…studies’, to quote Nancy Lindisfarne-Tapper and Bruce Ingham from their

introduction to Languages of Dress in the Middle East (1997) In the last decade, textiles

and dress have been the focus of such detailed studies, many of them in edited volumes that look at specific topics, such as gender, status, personal and social power, and ethnic identity.5

Introduction 3

Textiles and mobility

Why are textiles a particularly interesting subject of investigation for the Indian Ocean region? When discussing textiles in this maritime environment, it is their mobility that is particularly striking, as both Bühler and Gittinger demonstrated Cloth is relatively light and highly portable – and, initially at least, not at all fragile – unlike ceramics and glass Textiles have been a major trade item in the area, and the cloths of India have played a leading role in this From the time of antiquity into the middle ages, the lightness of Indian cotton and the quality of Indian dyes were unique This is taken up by the first two contributors to the volume Himanshu Ray discusses the historical evidence for textile trade and its economic significance in India and societies around the Indian Ocean; in her survey she makes use of significant new dating of actual textiles surviving She also examines the evidence for trade mechanisms, such as the role of the textile merchants as distinct from the producer For the Indian market, as well as the international trade in Indian cloth, it is quite certain that the weaver or textile printer had no influence beyond the production The distribution of cloth was turned over to the merchant In the evidence available to her regarding the international trade, Himanshu Ray has found that dealers in cloth are not mentioned separately Textiles were shipped as part of a group of staple commodities John Peter Wild and Felicity Wild present primary archaeological evidence that complements this historical discussion The fifth-century-AD cotton fragments discovered at Berenike, a harbour site on the Egyptian side of the Red Sea, are the earliest patterned textiles of definitely Indian origin so far recovered from an archaeological context, and they therefore are of foremost significance as evidence for the mobility of textile material This short, but important paper therefore is given the honour of initiating

the volume The first-century-AD Periplus Maris Erythraei (Casson 1989) already refers

to the trade in cotton fabrics from Gujarat, South-East India, and Bengal, but up to now

we have not seen any of the actual textiles surviving from the Near-Eastern pre-Islamic period.6

Indian cotton textiles probably remained a major export article for close to two thousand years There is a hiatus of several hundred years between the Berenike fragments and the next sequence of securely dated archaeological Indian textiles from Near-Eastern sources, but textual references of their trade to Baghdad during the ninth century suggest a continuity (Stillman 1986:737) The earliest substantial group of Indian textiles survived in Egypt, where they were traded to from parts of North-West India from the tenth century onwards.7 The Indian block-printed textiles were the original high-

status fabrics in East Africa as late as the nineteenth century, and the kanga cloths

discussed by David Parkin derived from them Women in Zanzibar wear sarong-like cotton wrappers which are printed with homilies or witty statements; they are worn to express the wearer’s emotional state and may comment on relationships with her husband and others in her household or immediate environment The sayings can be used to communicate intimate feelings between a wife and her husband or lover, but they may also be used outside the house to invite other women, possibly rivals, to participate in

competitive riddling, and can be used to provoke The kanga sayings are not generally a

statement on a woman’s social position, and they are not worn primarily to emphasise her

participation in the wider community They contrast in this respect from the kofia caps

discussed by Zulfikar Hirji in the second paper that offers material from Zanzibar He

Textiles in Indian ocean societies 4

presents a finely detailed description and analysis of the making of the caps, their designs and marketing, and their meaning in a local context These caps are made in Zanzibar and worn by Zanzibari males However, they also are signs of an important international

connection, worn by people going on the hajj to Mecca, and linking men with the origin

of Zanzibar’s ruling class in Oman The kofia is both a local product and a link with the

wider context of western Indian Ocean Islamic communities, especially those with close

family ties in southern Arabia Both the kofia cap and the kanga cloth are worn as a

personal message, but while the man’s cap is a statement about the wearer’s standing in the community and may be used to emphasise geographically farreaching connections, the woman’s cloth, with its specific sayings, is intended as a message about her inner self,

either temporary or long-term It is interesting to note that the kanga apparently had its

origin in imported Indian block-printed cotton cloth that was once a marker of high status, as well as an indicator of wide-ranging maritime contacts, but now has evolved into a local form of ‘text on textiles’

Exotic textiles and local practices

There is no doubt that patterned textiles have historically been a significant transmitter of design Their portability, however, can also bring about misunderstanding about their origin Here Steven Cohen’s discussion of the so-called Portuguese carpets provides revealing information These knotted carpets with seemingly exotic designs have been the subject of considerable discussion among scholars, both regarding their technical construction and their motifs, which combine certain conventional designs, typical for Iranian carpets of the seventeenth century, with figural representations that have their source in European imagery, and their origin of production One might think they were made to suit European taste, as they are dated to a time when the Portuguese presence in the Persian Gulf was still prominent But as Cohen shows, this is not likely to have been the case Instead their representations of maritime scenes, with ships and the occasional mermaid or merman, were probably made for local use but using European illustrations

as models, without always fully understanding the narrative meaning of the prototype, which would support the view that they were produced at some distance inland from the international setting of the Persian Gulf A further argument about these carpets has concerned their provenance, with the debate mostly favouring an Iranian source, but the possibility of an Indian, specifically Gujarati, production being proposed by one of the most eminent carpet scholars Steven Cohen addresses this issue and follows the history

of argument, and then establishes that the carpets’ likely place of origin was Khorasan in North-Eastern Iran This is argued primarily on technical grounds; a careful study of technology can indeed reveal much about the place of origin of an object, which is particularly true for textiles produced in a complex technique Although Cohen asserts that few people now believe the ‘Portuguese’ carpets to be of Indian origin, he sets out to explain why they indeed never could have been made in India: neither the technique of knotting nor the ply used for the warp match that of any carpet known to have come from

a Gujarati workshop

It is this close study of technology that must not be ignored when making historical connections But an understanding of technology alone does not always provide

Introduction 5

meaningful answers to anthropologists if it is not complemented by detailed ethnographic research This is argued by Sarah Fee in her discussion of Malagasy textiles The large island of Madagascar has a cultural history that has been affected by virtually all parts of the wider Indian Ocean The Malagasy language is Austronesian, with the closest linguistic connection found in southern Borneo; apparently the settling of the island occurred by Indonesians at some time in the first millennium AD Furthermore, loom technology and metal-working tools are closely connected to South-East Asia, as well, which indicates the movement of crafts people There are even linguistic connections between Malagasy tools and their Indonesian counterparts But these settlers also brought many influences from East Africa, so they seem to have moved to Madagascar from the continent’s coast Arabic culture had a major impact as well, as did contact with India Yet the question of overseas origins now is only of limited importance to the people of Madagascar Weaving can be a vital economic activity for women in the communities discussed by Fee, and it is connected to their sense of procreative power But beyond such general associations that may have resonance in the interpretation of cloth in South-East Asia and elsewhere in Indian Ocean communities, the meaning of textiles is often not fixed It can change from one occasion to the other; Sarah Fee makes a careful distinction and presents specific ethnographic situations

Appropriation and assimilation

All three contributions discussing textiles from Indonesia elaborate on the Indian connection, although this is always combined with an emphasis on the indigenous response The contact with India is only a facet in a locally more complex picture, regardless of whether this is assessed from a ninth-century Central Javanese perspective (Totton), from Jambi in south-eastern Sumatra in the nineteenth and early twentieth centuries (Kerlogue), or from the eastern Indonesian view of the present (Barnes) As Mary-Louise Totton demonstrates for the use and representation of textiles at the Central Javanese temple complex at Prambanan, textile patterns that have their origin in distant cultures, in India and China, have been assimilated and may be represented in stone relief sculpture that adorns the most sacred ceremonial temple space These designs indicate international connections and a cosmopolitan taste, but the integration into the Javanese temple’s innermost chamber points to a complete assimilation into a religious system that

is strongly indigenous in its interpretation of divinity, royalty, and ritual offerings, even when it makes use of South Asian cosmology and East Asian design

For the textiles of Jambi in southern Sumatra, Fiona Kerlogue establishes first and foremost a relationship to, and distinction from, the batik production of Java Many Jambi batik textiles, especially those with elaborate calligraphic inscriptions, were formerly given a North Javanese origin, as it was believed that virtually no Indonesian batik was produced outside of Java Kerlogue has shown here and in other publications that this view has to be reassessed It is still somewhat uncertain when batik was first developed in Jambi, and whether it was in fact originally an entirely Javanese introduction An independent relationship with India, and possibly the observation of South Indian resist dye techniques, may also have been a contributing factor in developing the skill in southern Sumatra A third factor addressed by Kerlogue is the strong local identification

Textiles in Indian ocean societies 6

with the internationalism of Islam, and the influence the hajj has had historically on local

taste and the spread of non-indigenous techniques and materials The use of metal thread embroidery and supplementary weft, for which the gold or silver thread was originally imported from India, the appreciation of foreign patterns and their adoption into local textile designs, and the proud display of these outside influences on prestige cloth, all emphasise, in Kerlogue’s words, that ‘in Jambi, the textiles refer as much to the world beyond as to features within Jambi society’

The last chapter presents ethnographic research from Kedang in eastern Indonesia It records the recent innovation of two types of cloth in an area formerly affected by a prohibition on weaving, and hence entirely dependent on the import of textiles, especially those needed for ceremonial purposes and therefore of high local status The change from external to local production of prestigious cloth was largely brought about through the efforts of one individual, a woman who herself is a prolific weaver In doing so, she tried

to be sympathetic to Kedang’s past, and she developed a new man’s cloth which she considered appropriate for the taste of the community’s ancestors But she also initiated the local production of a woman’s ceremonial cloth that was formerly made outside the region, decorated in a complex technique that has long been appreciated locally, but was foreign to Kedang

Conclusion

All contributions in this volume emphasise that textiles play an important role in defining the person within the community Textiles are intimately associated with the human body, and their presentation helps establish the status, the cultural affiliation, and the spiritual or emotional state of the self Making the textile, weaving, sewing, adorning it in

a distinctive manner, often is gender-specific and may become closely linked with perceptions of gender It also may be seen as a creative activity in which a person’s sense

of innovation is called for This is a strong element in the making of a kofia in Zanzibar,

and it is certainly in evidence in Kedang, where the initiative of one person has transformed the weaving of cloth in her community and has had an impact on the performance of an annual ritual

But as was said earlier, textiles also have moved between Indian Ocean cultures for centuries, and a single type of cloth may be shared by many in the region Yemeni men

wear their futaq, a sarong-type garment usually made in Indonesia but sold in the local

souk Inexpensive, machine-woven Indian madras-type cloths are still exported to

Indonesia; they are common for daily wear, but may also become part of local exchange ceremonies.8 The kanga cloths of Zanzibar are inspired ultimately by Indian block- printed textiles, and the kofia can be bought in the market in Muscat, Oman The

transferability of textiles brings up a final, and possibly most interesting, aspect As they move between cultures, they can take on new meanings

We do not know the exact origin of the Indian resist-dyed textiles discussed by John Peter and Felicity Wild, but they probably came from North-Western India Block-printed textiles made in Gujarat for export survive from the tenth century onwards, at least Thousands of fragments are known from medieval Egypt, where they were used as garments and for furnishings The same material, when traded to eastern Indonesia as part

Introduction 7

of the spice trade, was eventually removed from the secular realm and became part of the sacred, as heirloom textiles and (in some cases) as gifts to be exchanged to establish or confirm a relationship between lineages, or between the ruler and the ruled Furthermore, the patterns of cloth may take on a new, locally-significant meaning This is touched upon by Totton in her discussion of designs adopted at Prambanan, and it may be an issue that is especially relevant to South-East Asia’s response to Indian Ocean contacts I have encountered it in my own field work in eastern Indonesia As is known from the earliest Portuguese sources writing about the area at the beginning of the sixteenth century,

patola silk textiles were a desired prestige item In the village of Lamalera on the island

of Lembata these cloths have become clan heirloom cloths of considerable metaphoric and ritual significance, and as such are considered inalienable Their designs, in turn, have influenced local patterns (Barnes 1989) The Indian-inspired motifs are given local names that clearly have no relation to the prototype, but are meaningful in the local

community In the most striking example a floral border frequently found on patola (Fig 9.1, this volume) is copied into the Lamalera repertoire The patola prototype represents a flowering bush; the Lamalera interpretation refers to the design as ata dikã, which

translates as ‘human being’ It is specifically associated with one lineage, which descends from the oldest son of the village founder The Lamalera interpretation is an elaboration

of a pattern found generally in the area, which shows stick figures represented in vertical rows, often interpreted to represent ancestor figures and used as a reference to genealogy and descent In other words, an outside design is assimilated and given a new meaning, quite distinct from the original representation, but significant in the new context

To some degree all contributors present evidence for this local reaction to foreign objects and images They give a picture of great diversity, because in each case the response to the world of the Indian Ocean is specific and differs from one location to the next If there is any unity in this diversity, it is in the comfortable integration of international contacts with local concepts How this relationship was initiated and continued to work, often remains unclear We have detailed (and often brutal) accounts for the time of the European entry on the stage of the Indian Ocean But relatively well-organised contacts preceded the European arrival by at least a millennium, and these no doubt required merchants’ accommodation, currency and trade agreements, as well as travel arrangements It is usually no longer possible to discern the nature of early historical interaction between small-scale communities and internationally oriented states We do know, though, that it must have taken considerable courage, navigational skill, and economic effort to travel over such vast expanses of maritime space, and required sound local contacts to gain access to the indigenous products pursued Textiles often provided the key and were the widely recognised currency

Textiles in Indian ocean societies 8

3 Braudel, Fernand (1966) La Méditerranée et le monde Méditerranéen à l’époque de Philippe

II Second revised edition Paris: Librairie Armand Colin [Trans The Mediterranean and the Mediterranean world in the age of Philip II, London 1972.]

4 The three volumes were edited by Gittinger et al (1980), Völger and von Welck (1991), and Nabholz-Kartaschoff et al (1993)

5 See Weiner and Schneider (1989), Barnes and Eicher (1992), and Tarlo (1996)

6 Pfister (1937:16) recorded some textile fragments from Palmyra which he believed to be Indian; but he emphasised that they were surface finds and could not confidently be connected with the third-century site

7 The most important collection of this material is in the Ashmolean Museum in Oxford (Barnes 1997a) The tenth-century date could be established by radiocarbon analysis

8 In 2001 I was present at a wedding ceremony and a funeral in Adonara, eastern Indonesia; on both occasions these Indian textiles were prominently on display and were circulated between clans

Introduction 9

1 Rome and India: early Indian cotton

textiles from Berenike, Red Sea coast of

Egypt

John Peter Wild and Felicity Wild

The port of Berenike on the Red Sea coast of Egypt (see Fig 1.1a) was founded by Ptolemy II Philadelphus in about 275 BC and named in honour of his mother From then until its final abandonment in about AD 520 it functioned as a trading port for commerce between the Graeco-Roman world, South Arabia, East Africa and India Excavations at the site since 1994, directed by Professor S.E Sidebotham of the University of Delaware and Dr W.Z Wendrich of Leiden University, have started to reveal more about the history of the site and the extent of its contacts with India The port appears to have been particularly active in the late first century BC and early first century AD After a possible decline in occupation in the second and third centuries AD, when evidence for activity is sketchy, the town experienced a renaissance of both occupation and commerce in the late fourth century which appears to have continued until its final evacuation, probably in the early sixth

Evidence for trade with India and, indeed, for the presence of Indians on the site, has come in the form of a graffito on a first-century-AD amphora in Tamil-Braāhmi,1 beads and pottery of South Indian origin2 and imported botanical remains including coconut, Job’s tear and large quantities of black peppercorns.3 Of particular interest, in view of the textiles, was the high proportion of teak wood, including re-used planks possibly from dismantled ships.4

The textiles from the site were not well preserved The proximity of the sea and the neighbouring wadi have led to the disintegration of the textiles from the lower levels of the site, while heavy dews alternating with daytime heat have attacked those on the surface Most of the textiles described below, from the last two seasons of excavation, were recovered from two midden deposits, one dated by the associated pottery to not later than AD 70, the other to the late fourth to fifth century AD Fragments were small and, in many cases, badly degraded by salt

Recognition that there were Indian cotton textiles at Berenike was not immediate Only after five seasons’ work and much debate can we now feel confident of the identification In the first two seasons (1994 and 1995) over 400 textile fragments were recovered from midden deposits on the

Figure 1.1a Key find spots and

production centres of cotton in the Roman Empire and India

site, all late Roman.5 To our surprise, almost half of them were of cotton This is without parallel in the Roman Empire: at best there is a handful of cotton fragments from stratified Roman contexts in the Lower Nile Valley and in Roman Syria, Palaestina and Mesopotamia,6 with occasional outliers north of the Mediterranean.7 The proportion of cotton found at Berenike was equally high in subsequent seasons

The cottons at Berenike can be divided technically into two distinct groups: the one is woven exclusively from S- or anticlockwise-spun yarns, the other from Z- or clockwise-spun yarns Ancient spinners were highly conservative, and the tradition in Egypt and neighbouring Roman provinces was for the S-direction: Z-spun yarn was only used for special weft.8 It would be fair to assume accordingly that the S-spun cottons were produced in Egypt or at least in the Nile Valley Roman writers mention the growing of cotton in Egypt9 and there is papyrological evidence for its cultivation in the oases, probably under irrigation.10 S-spun cottons are not infrequently found in the Meroitic and X-Group cemeteries along the Upper Nile (first to sixth century AD)11 and there is archaeological evidence for cotton growing in Ethiopia.12

The Z-spun cottons are a separate problem: they are best described as ‘intrusive’ There are isolated examples of Z-spun cottons on the Upper Nile, but the largest group, still only 13 pieces, comes from Quseir, the Roman port of Myos Hormos further up the Red Sea coast and Berenike’s main trading rival.13 To look for an overseas source, therefore, for the Z-spun cottons seems plausible

The weight of ancient literary and documentary evidence indicates India to be the only practical source, given its long-established cotton industry Herodotus writing about 425

Rome and India: early Indian cotton textiles 11

BC comments on wild cotton and cotton clothing in India,14 while a century later reports from observers on Alexander’s expedition to North-West India refer to intensive cotton cultivation.15The picture is confirmed by the Greek envoy Megasthenes resident at Pataliputra on the Ganges16 and by contemporary Indian regulations recorded in the

Arthasaāstra.17 The Graeco-Roman shipper’s manual known as the Periplus of the

Erythraean Sea (dated to between AD 40 and AD 70)18 has a wealth of information about the cotton trade and production centres in India, on which a map can be based (see Fig 1.1a) The filled triangles on it indicate the production centres Some of the manual’s terminology is obscure; but it could be argued that North-West and South India and Sri Lanka were the principal suppliers to the Roman world of largely plain cotton while the east coast and especially the Ganges Valley offered the finest qualities A shipper’s contract on papyrus governing the ownership and conveyance of a consignment of Indian cotton sheets may even have been drawn up at Berenike.19

Cotton finds in India of Harappan date are well known;20 but we have yet to learn of any relevant material of later date with which to compare the finds from Berenike However, the circumstantial evidence provided by other classes of Indian material imported into Berenike has become very strong In particular, the Indian teak planks and their association with shipping led us to identify some pieces in our collection, all in Z-spun cotton, as characteristically sailcloth A high proportion of these were from the early Roman midden deposit

Among the numerous plain cotton fragments there were at least six strips of cloth, with a distinctive form Up to 30 cm long and often sewn end-to-end to others, both raw edges on the long axis are turned under to leave a band about 3.5 cm wide Their purpose was not apparent until 1998 when a piece of medium-weight tabby turned up, attached to which, at right angles to one another, were strips of cloth of similar quality The edges of the strips had been turned under, sewn down to the main cloth on one side with running stitches, on the other with overcast stitching: two sections were sewn end-to-end The strips concealed nothing, certainly not a seam in underlying fabric: they were reinforcement A square patch had been sewn to the main cloth in one place to repair a small hole

Ships’ sails as depicted in Mediterranean art frequently bear a grid pattern that has been variously interpreted.21 Most recently the suggestion has been advanced that the grid represents reinforcing bands,22 and that accords with our finds at Berenike A relief of early third-century date from Ostia, the port for Rome, shows not just reinforcing strips in

a grid pattern, but brailing rings attached to them through which the brailing ropes passed.23 There are two brailing rings from Berenike, one still with cotton string through its holes Although shipwrecks feature in Indian art,24 we have yet to discover a representation of reinforced sails The sail suggests a single-masted Mediterranean ship, such as that mentioned above from Ostia, rather than the two- or three-masted vessels of Indian art

There is only one published parallel to the Berenike sails, a linen sail from a (Roman?) grave at Thebes in Upper Egypt, complete with webbing reinforcements and brailing rings.25 Our torn strips may in fact be a substitute for woven webbing, of which we have

a few examples One has stout plied warp, and close to the selvedge on each side are traces of a largely missing pinstripe in dark brown plied cotton Incidentally, this may be

Textiles in Indian ocean societies 12

the first evidence that naturally brown cottons grew in Ancient India, as they did and do

in Peru.26

Four or five composite cotton textiles were uncovered, each nothing but a mass of patches reflecting repeated repairs The patching had been carried out with great care; the warp and weft of the patch were aligned with those of the ground weave and the sewing was carried out in Z-spun thread In one case a hole was darned rather than patched The patching and darning presumably began as running repairs to sails, which may ultimately have been re-used as tarpaulins protecting cargoes

The bulk of the cotton collection consists of featureless plain scraps, some of which

nevertheless are very fine and may be the traded goods listed in the Periplus One

recognisable object is a small bag It is sewn together with S-spun cotton thread, so, though the fabric is Indian, the bag was actually made at Berenike

Indian weavers practised a number of decorative techniques including creating blue checks in a variety of designs A number of these have been found at Berenike The weavers counted out carefully the sequences of blue and undyed yarns and compensated for differences in density between warp and weft to achieve a symmetrical pattern The blue dye is probably from indigo rather than woad.27 The fifth- and sixth-century wallpaintings of the Ajanta Caves show blue checks worn as clothing, as by a serving maid in Cave 1 (late fifth century),28 and covering a bolster on the couch of King Viruna

in Cave 2 (AD 500–550).29 One check from Berenike may even have been a sail

Two fragments showed colour and texture added to undyed cloth by rows of green dyed tufts, hooked under two warp-threads during the weaving This may be an

blue-example of the kaunakai, pile fabrics, produced in North-West India according to the

Periplus.30 Equally complex is what looked at first sight like drawn-thread work: with the aid of two pairs of countered twining threads the warp was grouped into threes, fours, fives and sixes to leave a narrow reserved line in the weft direction

The most exciting textiles are undoubtedly the tiny fragments of blue resist-dyed fabric, of which there are now seven or eight, mostly with incomplete motifs One simple rosette pattern, undyed on a faded blue background, is based on 12-petalled flowers arranged in regular rows On one side the outline of the motif is crisp; on the other less boldly defined The resistmedium, wax or perhaps a clay or lime paste,31 was painted on

by hand – it is not a block print like the well-known medieval Indian prints.32

Other fragments can be reconstructed as a large petalled flower, and as a lotus-bud (Fig 1.1b) All seem to have echoes in Indian art: in Egypt they may have served as hangings.33One piece with a triangular motif (Fig 1.1c) was patched with S-spun linen thread, i.e the repair was done at Berenike It is interesting that the resist-dyed cotton textiles only occur in late Roman levels at Berenike None of the few other examples of resist-dyed fabrics known in the Roman world can be dated earlier than the third century.34

At present Berenike boasts the largest corpus of recognisably Indian textiles from the first five centuries AD I emphasise ‘recognisably’ A

Rome and India: early Indian cotton textiles 13

Figure 1.1b & c Resist-dyed cottons

from Berenike: b (above) design with possible lotus-bud motif (BE96 0219);

c (below) fragment with incomplete triangular motif (BE98 1191)

resist-dyed cotton from Niya in the Han-period kingdom of Shanshan, south of the Taklamakan, may be Indian;35 but in the light of the finds from Berenike in the West it should now be possible to identify confidently as Indian some of the humbler textiles which Aurel Stein brought back from his Central Asian expeditions on the eastern fringes

of the Indian world

Berenike Project, Professor S.E Sidebotham and Dr W.Z Wendrich, gave every practical assistance

14 Herodotus, III, 106; VII, 61, 1

15 Theophrastus, Enquiry into Plants IV, 4, 8; Arrian, Indica, 16 Megasthenes 1926

23 Graefe 1979:121–123, Abb.133, Taf.124, 2

24 Swamy 1997; Schlingloff 1988:195–218; Behl 1998

25 Rougé 1987; Black and Samuel 1991

33 Wild and Wild 2000

34 Bender Jørgensen and Vogelsang-Eastwood 1991:3

35 Vogelsang-Eastwood 1990:63 pl 2; von Wilckens 1993

Rome and India: early Indian cotton textiles 15

2 Far-flung fabrics – Indian textiles in

ancient maritime trade

Himanshu Prabha Ray

It is appropriate to start with an explanation for the title of this paper In recent years there has been an increase in the corpus of dated Indian textiles either found in archaeological excavations or in museum collections, which have been subjected to radiocarbon investigation These come from archaeological sites within the Indian subcontinent, as also from recently excavated sites such as Berenike along the Indian Ocean rim These specimens are still few and far between and come from spatially and temporally different contexts, but they relate to the wider world of Indian Ocean trade networks

Two of the collections thus covered are those of the Ashmolean Museum, Oxford, and the Victoria and Albert Museum, London (Barnes 1997 a, b; Guy 1998) The two are by

no means similar While the Victoria and Albert collection is rich in specimens from the colonial period, especially the European-market chintz collection (Irwin and Brett 1970), the Ashmolean houses more than 1,200 textile fragments collected in Egypt and involved

in the western Indian Ocean trade The series of radiocarbon dates, ranging from the tenth

to the fifteenth centuries for the Ashmolean collection, available since 1996, has further enhanced its value (Barnes 1996) (Fig 2.1) Textiles with similar or at times identical designs as the Ashmolean specimens have also been reported from Indonesia; while many of these bear Dutch East India Company (VOC) stamps, it is significant that several large pieces acquired from Sulawesi can now be traced back to the fourteenth/fifteenth century The first cloth to be thus identified again is in the Ashmolean Museum (Fig 2.2) How then does one explain the finding of Indian textiles from Gujarat at two ends of the Indian Ocean network? Thus rethinking the nature of transoceanic trade in the region is imperative How did this trade muster regional and local communities of craftsmen and traders? Was it state-controlled and hence directed towards satisfying the requirements of the elite? To what extent can one talk of the autonomy of the producer or the craftsman?

These are some of the questions that this paper attempts to address, and which form part of the author’s larger study on the archaeology of the Indian Ocean It is suggested in the first section that the Indian Ocean

Figure 2.1 Indian block-printed cotton

fragment, Gujarat, traded to Egypt, radiocarbon dated to tenth century AD

Ashmolean Museum Oxford, Newberry Collection (EA 1990.250)

trading system was radically transformed as a result of increasing European interest in the region It is important to understand that in the ancient period, the Indian Ocean trading network functioned within different parameters, but was nonetheless a continuous system subject to the usual fluctuations in demand and supply Leading on from there, the discussion shifts inland to peninsular India and to the different partners that made these maritime connections possible, i.e the trader, the craftsman, the State and the religious institutions Contrary to received wisdom on states, control of craft production, this paper suggests multiplicity of control, hierarchy of craft production and diverse uses of the textiles themselves One final issue addressed here relates to the increasing practice of using the ethnographic present to explicate past practices The discussion here focuses on one specific example, the lace makers of Narsapur It is evident that this paper covers a

Far-flung fabrics – Indian textiles in ancient maritime trade 17

wide time span and perhaps a reasonable method of presenting an overview is by addressing the issues thematically

Figure 2.2 Indian block-printed cotton

textile, Gujarat, traded to Sulawesi, eastern Indonesia, radiocarbon dated

AD 1400 ± 40 Ashmolean Museum Oxford (EA 1995.61)

Indian Ocean trade in the early period

Conventional studies have tended to categorise maritime trade in the Indian Ocean in the pre-colonial period into national networks, such as Roman trade, Arab trade and European trade in the western Indian Ocean, as well as Indian trade across the Bay of Bengal The basis for these identities was inevitably the nature of the sources available

For example, the first century AD text, the Periplus Maris Erythraei provides the most

comprehensive account of trade and trading commodities between the Red Sea and the west coast of India The text itself nowhere refers to ‘Roman trade’, though it does mention a diverse range of communities involved, including the Greeks, Arabs, Indians, and the Ichthyophagoi, or coastal fishing communities Nevertheless, in popular parlance references to Roman trade supplying the markets of the Empire with luxury items such as Indian muslin continue Similarly, the spread of Islam to India and South-East Asia is often associated with the stimulus provided by the demand for spices.1

Textiles in Indian ocean societies 18

Implicit in this is a second perception that Indian Ocean trade in antiquity was largely

a trade in luxuries Empires required these luxuries and in turn promoted long-distance trade (Hall 1996) To quote Bentley, ‘the consolidation of the Roman and Han empires deeply influenced the dynamics of world history Most obviously, the two empires both organized vast territories into orderly and coherent polities, enabling their subjects to carry on trade and to seek prosperity under conditions of relative stability’ (1993:64)

A third perception that is often repeated in historical writing is that ancient maritime trade was a reflection of the better-documented European system The pattern of trade explicated on the basis of European records of the seventeenth to nineteenth centuries is often used to explain earlier developments For example, in the colonial period, cloth from India was used in intra-Asian European trade to pay for spices from the Indonesian archipelago A similar pattern is presumed for an earlier period when a quest for spices is linked to Arab participation in maritime trade (Guy 1998:8, 14) This generalisation does not take into account the disjunction in Indian Ocean maritime trade that occurred as a result of protectionism and monopolistic legislation in England in the seventeenth and eighteenth centuries This legislation started a new trend with the linking of the interests

of the State with those of the trading companies, such as the East India Company As a result, from the sixteenth century onwards, the European companies found it necessary to procure cargoes from one part of Asia with goods obtained in another, shipping Indian cloth to Indonesia to pay for spices being a prime example (Prakash 1997:239)

It is being argued in this paper that the ancient trading network in the Indian Ocean functioned within a different background from that of the later European system The interest groups and consumption centres were located within the Indian Ocean region and not outside it The commodities involved in the Indian Ocean trade were by no means limited to luxuries Instead these may be divided into various broad categories such as aromatics, medicines, dyes and spices; foodstuffs, wood and textiles; gems and ornaments; metals; and plant and animal products These categories find mention in a

range of textual sources from the Periplus Maris Erythraei (Casson 1989:39–44) to the

Genizah documents, Chinese and Arab accounts (Tampoe 1989:131–153)

The beginnings of this indigenous system may be traced to the exploitation of marine resources in the Mesolithic period when fishing and sailing communities settled in coastal areas By the third millennium BC, these early beginnings had evolved into trade networks between the Makran and Gujarat coasts of the subcontinent and the Persian Gulf Earlier studies had emphasised the Harappan–Mesopotamian connection based on cuneiform references to the import of luxuries from North-West India, but recent archaeological data, especially from sites in the Gulf, has considerably expanded this notion

One of the sites crucial to our understanding of Harappan maritime trade is that of Ras al-Junayz located in an area which is still the major landmark for boats crossing from Pakistan to the Gulf and Africa (Tosi 1993:365–378) The site represents one of many such coastal sites, with little or no hinterland but involved in a wide-ranging exchange system in subsistence goods along the northern rim of the Arabian Sea The stretch between the Makran coast and Africa is arid and inhospitable, with few resources apart from the rich marine life The availability of foodstuffs from sites further east along the Indian coast made the economic integration of the region in the wider Indian Ocean

Far-flung fabrics – Indian textiles in ancient maritime trade 19

network possible This pattern continued into the historical period, albeit with fluctuating fortunes, as also shifts and changes in routes

Around the beginning of the Common Era, this network had expanded to include large parts of the western Indian Ocean A range of communities participated in this, such as the Nabataeans, Sabaeans, Homerites and Arabs, in addition to Indians Muza at the

mouth of the Red Sea is described in the Periplus Maris Erythreai as a port of trade

without a harbour (section 24) but with a good road-stead for mooring, and teeming with Arabs – ship owners or charterers and sailors (section 21) Leuko Kome on the Red Sea coast was the harbour of the Nabataeans where craft, ‘none large’, loaded with freight from Arabia (section 19) Trade in cloth, wood and agricultural products sustained the Indian Ocean network as is indicated by the presence of guilds of weavers, potters, oil millers and so on, in the list of donors mentioned in the inscriptions from the Buddhist monuments of peninsular India (Ray 1986:112) This is further substantiated by the botanical evidence from the archaeological excavations at Berenike which included imports from South Asia in the fourth to sixth centuries AD, such as pepper, coconut, Job’s tear and possibly rice (Sidebotham and Wendrich 1996:447; Cappers 1999)

Maritime voyages in the early centuries of the Common Era were regarded as profitable ventures and Buddhist literature describes a variety of social groups who were

involved (Jatakas, Book I, no 4; Book X, nos 439, 442; Book XI, no 463; Book XVI,

no 528) In addition to merchants there are references to princes who travelled across the seas to make money (Book XXI, no 539) The prosperity and status of these mariners is evident from the donations made by them to the Buddhist monastic establishments and recorded in the inscriptions of the Early Historic period It was also at this time that maritime activity between South and South-East Asia intensified leading not only to increased commercial activity between the two, but also the spread of the religions of the subcontinent, especially Buddhism and Brahmanism

One of the, as yet, little known trading groups in the western Indian Ocean is that of the Christian communities The spread of Christianity beyond Palestine into Arabia, the Persian Gulf and further east, is a subject inadequately researched, though archaeological data for the presence of Christian churches and cemeteries is gradually emerging In the fourth to seventh centuries AD, several tribal groupings in the regions now in the countries of Jordan, Syria and Iraq are known to have adopted Christianity Christian presence is known from south-west Arabia and Christian monasteries located on the caravan routes in western Arabia functioned as caravanserais

The archaeological record from the Gulf further supports these references Recent excavations at al-Qousour on the island of Failaka have uncovered a Christian church assigned an eighth to tenth century date This data together with Nestorian Church records and Arabic literary sources indicates the presence of a stable Christian community in the region for a considerable period of time both before and after AD 630

In the sixth century, an Egyptian monk, Cosmas, surnamed Indicopleustes wrote a learned treatise for the express purpose of disproving the sphericity of the earth In the

midst of several irrelevant topics, the Topography of Cosmas contains interesting details

about the island of Sri Lanka as also accounts of his travels in the three gulfs: the Mediterranean Sea, the Red Sea and the Persian Gulf He mentions a church of Persian Christians on the island of Sri Lanka and states that due to its central position the island

Textiles in Indian ocean societies 20

was much frequented by ships from all parts of India and from the regions of Persia and Ethiopia.2

It is then evident that a Christian community was settled on the south-west coast of India in Kerala and Tamil Nadu from about AD 600 to the ninth century and that this community used the Persian or Pahlavi language (Gropp 1991:83–88) One of the earliest churches to be built in India was in Kerala and a copper-plate grant of the year AD 849 records several rights made over to the church, such as the grant of land and the services

of certain families In addition, several trading rights were granted to the church,

including customs duties and the right to collect four kasus (a kind of coin) per boat,

every time it came in and went out of the harbour (Abraham 1988:20–29) century Pahlavi inscriptions from the Buddhist Cave 90 at Kanheri on the west coast provide a record of further visits by Persians In contrast to the other inscriptions at the site, these do not mention donations but seem to have been inscribed with the objective of

Eleventh-documenting visits of hamdinikan or co-religionists listed by name (Gokhale 1991:142–

147)

In the three centuries preceding the emergence of Islam, the Nestorian Church had exerted a unifying influence on the region’s population, both amongst the tribes of northern Arabia and among the settled communities of the coast (Potts 1990:354) By building on the unity forged by Christianity in an earlier period, the spread of Islam was able to expand the frontiers of this trading network The earliest literary sources that describe this network date from the ninth century onwards Archaeological excavations at the site of Siraf located on the coast of Fars have provided evidence for the participation

of the site in trade with China as early as the latter half of the eighth century This was accompanied by an increase in architectural activity and the construction of a mosque at the site (Tampoe 1989:101)

The establishment of the Abbasids and the subsequent shift in main centres of the Caliphate from the Mediterranean region to the eastern provinces of Iran and Iraq led to

an increase in maritime activity with Basra, Siraf and Sohar being the major centres In the ninth century, trading settlements were established on the islands off the coast of east Africa While the attribution of these Swahili cities to Islamic foreign foundation is a matter of debate, it is possible that they received immigrants from the region of the Persian Gulf It was also during this period of Siraf’s ascendancy that the Persian sailing

instructions known as Rahnamaj were translated into Arabic (Meglio 1970:106)

Maritime trade continues to figure prominently in literature from the Indian subcontinent, especially Jaina non-canonical texts dated from the eighth to the eleventh century AD These contain frequent references to long-distance trade and to sea voyages

undertaken by merchants owning ships ( pota-vanik) to distant lands in quest of wealth.3

Merchants from different regions of the subcontinent are described as meeting at the

coastal centre of Sopara on the west coast (desiya-vanika-meliya; Jain 1990:36, footnote

10) Their deliberations provide indications of the commodities traded which included horses, elephants, pearls, and shell, ivory, silk, arecanuts and so on, as also a list of the places travelled to, namely, Uttarapatha, Dwarka, south India, South-East Asia and China (ibid.) Often merchants pooled their resources and organised themselves in a caravan to undertake the arduous journey for trade, but occasionally a rich merchant took the initiative, had a drum beaten and proclamation made asking for associates to accompany him (Jain 1990:39)

Far-flung fabrics – Indian textiles in ancient maritime trade 21

From the ninth to the mid-fourteenth centuries two of the merchant guilds that dominated economic transactions in south India were the Manigramam and the Ayyavole (Abraham 1988) Associated with these two merchant guilds were associations of craftsmen such as weavers, basket-makers, potters, leather-workers and so on (ibid.: 117)

As late as the seventeenth century, the Ayyavole seemed to be concentrated in the producing areas of Andhra Pradesh Though these two guilds originated indepen dently

cotton-of each other, from the mid-thirteenth century onward, the Ayyavole association became

so powerful that the Manigramam functioned in a subordinate capacity to it Not only did these merchant associations develop powerful economic networks, but they also employed private armies The range of their operations extended well beyond the boundaries of the Indian subcontinent into South-East Asia Several clusters of Tamil inscriptions have been found on the eastern fringes of the Indian Ocean from Burma to Sumatra In contrast, the major states of Java and Bali reacted some what differently to the advent of Indian merchant groups In the tenth century, local versions of these

merchant guilds, termed the banigrama, appeared in the north coast ports of both the

islands While some foreign merchants may have been included in them, these groups appear largely as indigenous organisations associated with the local economic networks

as tax-farmers (Christie 1999)

Another important source for a study of medieval trade is the Genizah collection The

Hebrew word genizah means burial and the term Genizah papers designates a repository

of discarded writings found in a lumber-room attached to a synagogue in Fustat or Old Cairo These documents present a random collection of business correspondence, family letters and legal deeds dated between the eleventh and thirteenth centuries AD Generally, the script used in these documents is Hebrew, while the language is Arabic While a majority of the documents deal with the commercial transactions of the Jewish community in the Mediterranean, there is much that is of interest for a study of trading activity in the Indian Ocean (Goitein 1967:1–19)

For one, the Genizah documents indicate inter-linkages between merchant groups from different regions and distinctions on the basis of the scope of operations and organisation of trading activity For example, the distinction between retailers and wholesalers is well developed Many of the retailers were trading artisans who marketed

their own products The term ‘tajir’ was used for big merchants who traded in a wide

variety of goods They were divided into three categories: one who stored large quantities

to sell when the price was high; one who transported goods from one country to another; and one who sent shipments abroad (Tampoe 1989:126) One such merchant was Nahray bin-Nissim; a wholesale merchant mentioned in 250 of the Genizah documents (Goitein 1967:149–150, 155)

The forty-two India letters of the Cairo Genizah provide significant information on the modes of transactions between partners albeit of different religious affiliations, such as Jews or Muslims.4 A letter dated 11 September 1149, written by Madmun bin Hasan bin Bundar and sent to Ibrahim bin Perahya bin Yiju, the owner of a brass factory in

Mangalore, sends greetings to Ishaq, the bania It also asks the latter to send a boat from

Mangalore, as also all the pepper, iron, cubeba and gingembar at their disposal (Goitein 1980:528–529) Similarly, letter no 9 (dated 1145) mentions business relations between

an Indian nakhoda, or ship owner, named Tinbu, and the author of the letter, also a

nakhoda named Mahruz bin Jacob The boat from a certain pattanasvami is mentioned as

Textiles in Indian ocean societies 22

having reached Aden (letter no 38), while his large ship foundered while being driven by winds to Berbera on the African coast (Goitein 1967:349–350) The references are primarily to interactions with the Malabar and Gujarat coasts and very rarely to regions further east Maritime voyages, when successful, could lead to enormous accumulation of wealth; a characteristic example being the story of Ishaq the Jew as related in the ‘Ajaib al-Hind’ of Buzurg bin Shahriyar Towards the end of the ninth century, Ishaq left Oman

for the East with a capital of no more than 200 dinars For thirty years he amassed a great

fortune abroad and returned to Sohar in AD 912 with a shipload of musk, silk, porcelain, jewels, precious stones and other Chinese merchandise The musk, silk and porcelain

alone were reported to have been worth three million dinars (Williamson 1974:94)

Thus it is evident that maritime trade in the Indian Ocean was a continuous process involving a range of communities and marked by growing complexity and expanding geographical frontiers The commodities involved in the trade were by no means limited

to luxury items, but included textiles, foodstuffs, dyes, medicines, woods and metals What is also significant is that these commodities were transported through coastal

voyages and sold at markets along the Indian Ocean littoral For example, the Periplus

refers to Opone (Hafun) as a port of trade, which provided a market for goods from Barygaza on the west coast of India These included agricultural products and cotton

cloth: the monache and the sagmatogene ‘Some ships sail principally to these ports of

trade but some follow the coast and take on whatever comes their way’ (section 14).5

The inland network: the trader and the State

It is then apparent that textiles formed an integral part of the Indian Ocean trading network in antiquity Dealers in cloth are not referred to separately; instead textiles were one of the commodities mentioned as having been historically traded across the Ocean The letters of merchants found in the Cairo Genizah are a useful source for an appreciation of the economic dependence on seafaring activity, not only in the Mediterranean, but also the western Indian Ocean

The reason for this interest in sea traffic is easily understood: the local market and the prices of goods for import and export were entirely dependent on it It was these ships that brought wares from abroad and carried the prospective buyers of goods produced locally or destined for re-export

(Goitein 1967:302)

A letter written in the eleventh century refers to the price of flax falling after the arrival

of a ship at Mazara on the south-western tip of Sicily, whereas at the end of the sailing season the price of lead rose as no more supplies were expected (ibid.) The arrival of a single ship in port was a major event that could create a stir even in a port the size of Alexandria Similarly, ‘an uncommonly quiet season in sea traffic put the whole local market out of gear’ (ibid.)

At the same time there are references in early Buddhist sources to local merchants

who booked a share of the incoming cargo by paying money in advance (Jatakas, Book I,

Far-flung fabrics – Indian textiles in ancient maritime trade 23

no 40); a practice which could often be manipulated to control prices Nor was this limited to the first half of the first millennium, as is evident from the Genizah papers When a convoy of foreign ships reached a harbour, some local merchants would try to monopolize all the imports and thus drive up the prices – a practice against which the Muslim law books were already fulminating, namely with regard to incoming caravans

(Goitein 1967:303) One of the major consumers of expensive textiles was the State, the other being religious

institutions The Arthasastra of Kautilya contains a long list of valuable goods considered

important to be included in the king’s treasury and this includes a range of various

textiles such as silk, where a distinction is made between patrorna, kauseya and

cina-patta (II.11.107–114) Patrorna has been identified as uncultivated silk collected from

various trees and together with kauseya it forms the Indian varieties of silk (Scharfe 1993:290) Kauseya is already mentioned in the fifth and fourth centuries BCE grammar

of Panini (IV.3.42) and also occurs in the Sanskrit Epics These references add to the

discussion in the Periplus Maris Erythraei, which mentions the inland city of Thina from

which silk floss, yarn and cloth are shipped via Bactria to Barygaza and via the river Ganga to Limyrike (section 64) It is, however, not indicated how much of the export comprised of Indian silk

It is not surprising then that the Arthasastra of Kautilya presents a picture of regulated trade and textile production in the early centuries AD The panyadhyaksa is

state-described as being in control of trade He fixed prices of various commodities (IV.2.36) and allowed a certain percentage of the profit to traders The text devotes a whole chapter

(II.23) to a discussion of textile production and refers to the sutradhyaksa as the

officer-in-charge Women were involved in the spinning of yarn from wool, bark-fibre, cotton, hemp and flax and there are references to separate establishments for the production of silk (II.23.8–9) The wages were determined on the basis of the quality of the yarn (II.23.2–3), and it even refers to slaves being employed in textile production

This information is, however, at variance with that provided by inscriptions, Buddhist and Jaina texts, as also from Tamil sources These latter sources describe a complex mechanism of craft production and trade largely handled by entrepreneurs with little control by the State A hierarchy of commercial transactions is mentioned in the Buddhist Jatakas, ranging from barter to those conducted by merchants and guilds For articles of royal consumption, it was the valuer in the court who decided the price, while commodities for the consumption of city dwellers was transported and traded by merchants in caravans (Ray 1986:108)

The Divyavadana refers to shops at Sopara on the west coast, which exclusively dealt

in cloth imported from Kasi or Varanasi (kasika-vastravari) and also shops which stocked phuttaka cloth which has been described as printed muslin (Upreti 1995:42; Chandra 1973:33) Puspapatta or ‘flowered cloth’, either printed or embroidered, was

also known There was a regular street of cloth merchants at Madurai in the south who stocked ‘piles of bales, each containing a hundred lengths, woven of cotton, hair or silk’

(Cilappatikaram XIV: 98) In the streets of Kaveripattinam in Tamilnadu, weavers (karukas) ‘brought their fine silks and all kinds of fabrics made of wool or cotton’ (ibid

Textiles in Indian ocean societies 24

V: 18) The town also possessed ‘a spacious forum for storing bales of merchandise, with markings showing the quantity, weight, and name of the owner’ (ibid V: 21)

The State no doubt derived revenue from taxing trade transactions at entry points and there are several instances of these taxes being transferred to religious establishments The charter of Visnusena dated to AD 592 from Lohata in the Gujarat, Kathiawar region, provides a detailed list of 72 trade regulations or customary laws to be followed by the

community of merchants (vaniggrama) established in the region The charter assures protection to them and endorses their continued functioning (Epigraphia Indica, 30:163–

181) Some of the regulations are of considerable interest For example, it is specified that merchants staying away for a year were not required to pay an entrance fee on their

return Other clauses specify duties that were to be paid A boat full of vessels

(bhanda-bhrta-vahitrasya) had to pay twelve silver coins, but if the vessels were for a religious

purpose then it was only one and a quarter silver coin In the case of a boat transporting paddy, it was half this amount Other commodities mentioned as being carried on boats included dried ginger sticks and bamboo (ibid.)

Around the beginning of the second millennium there is increasing evidence for fortified settlements in coastal areas with attempts by the State and religious institutions

to localise the sale and exchange of goods either within the fortified precincts or in the vicinity of the Brahmanical temple Indicative of this shift are inscriptions from Gujarat,

as also from south India The tenth-century Siyadoni inscription from Gujarat records donations made between 903 and 968 by merchants and artisans and the endowment of shops in the textile market, main market and so on (Jain 1990:139) Another example from south India is the Piranmalai inscription of the thirteenth century from the Sokkanatha temple ‘built like a fortress’ The importance of the temple lay in its strategic location astride routes crossing the southern part of the peninsula from the Malabar coast

to the east The record lists a range of commodities on which cess was levied for the benefit of the temple These included cotton, yarn, thick cloth, thin cloth, thread, etc (Abraham 1988:105)

The craftsman

A distinction has to be made between trade in textiles, and production of textiles and the two functions were not generally combined The pattern that emerges around the beginning of the Common Era is a hierarchy of craft production Spinning, for example, was done by women as a household chore The employment of women in ‘factories’

owned by the State is mentioned in the Arthasastra and was referred to earlier This is

further supported by references in early Buddhist literature, which indicate that spinning with the aid of a spindle was a household chore performed by women A three-phased process is referred to: the preparation of the cotton; the spinning of the thread; and ginning or smoothing, before the weaving of the cloth These three processes are well illustrated in a painting of the Mahajanaka jataka in Cave 1 at Ajanta (Schlingloff 1988:181–187) It has been suggested that since cotton spinning was a part-time domestic occupation it allowed for fluctuations in demand without social disjunction (Varadarajan 1999:362)

Far-flung fabrics – Indian textiles in ancient maritime trade 25