Muted Memories: Heritage-Making, Bagamoyo, and the East African Caravan Trade 9781789201734

Citation preview

Muted Memories

Muted Memories Heritage-Making, Bagamoyo, and the East African Caravan Trade

 Jan Lindström

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2019 by Berghahn Books www.berghahnbooks.com © 2019 Jan Lindström

All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without wri en permission of the publisher.

Library of Congress Cataloging-in-Publication Data Names: Lindström, Jan, 1950- author. Title: Muted memories: Heritage-Making, Bagamoyo, and the East African Caravan Trade / Jan Lindström. Description: New York: Berghahn Books, 2019. | Includes bibliographical references and index. Identifiers: LCCN 2018056510 (print) | LCCN 2018061754 (ebook) | ISBN 9781789201734 (ebook) | ISBN 9781789201727 (hardback: alk. paper) Subjects: LCSH: Bagamoyo (Tanzania)—History—19th century. | Bagamoyo (Tanzania)—Commerce—History. | Ivory industry—Tanzania—History. | Slave trade—Tanzania—Bagamoyo. Classification: LCC DT449.B34 (ebook) | LCC DT449.B34 L56 2019 (print) | DDC 967.823—dc23 LC record available at h ps://lccn.loc.gov/2018056510

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

ISBN 978-1-78920-172-7 hardback ISBN 978-1-78920-173-4 ebook

 Contents List of Illustrations

vii

List of Maps and Figures

viii

Preface Introduction

ix 1

Part I. Heritage-Making, Branding, and Globalization

15

Chapter 1. Bagamoyo: A History of Practices, Principles, and Partnership in Heritage-Making

17

Chapter 2. Heritage-Making: The 2002 International Conference

41

Chapter 3. Fractures in the Image of Bagamoyo: Despair or Joy?

61

Chapter 4. World Heritage and Globalization: The Bagamoyo Case

87

Part II. Commerce, Competition, and Consumerism: Bagamoyo and the Caravan Trade

111

Chapter 5. Entrepreneurs and Explorers from the Heart of Africa

113

Chapter 6. Pawned, Preyed Upon, Purchased, or Punished: Slaves and Slavery in Nineteenth-Century East Africa

143

Chapter 7. Conflicts and Clashes in the Competition over the Caravan Trade on the Central Routes

177

Chapter 8. Bagamoyo and the Caravan Trade: The Entrance to the Heart of Africa

206

Chapter 9. Old Bagamoyo

234

vi

•

Contents

Chapter 10. Fluid Identities: Politics of Identity in Multicultural Bagamoyo

258

Chapter 11. Conspicuous Competitive Consumption and Communication by Means of Cloth

292

Chapter 12. Intruders and Terminators: The End of the Story

317

Epilogue

347

Glossary

349

References

355

Index

379

 Illustrations 1.1. The monument in place of the hanging tree in Bagamoyo. Photo by the author. 20 2.1. Part of the German Boma at the time of the 2002 International Conference. Photo by the author.

44

3.1. Doors to the cells at the Old Fort said to have housed slaves. Photo by the author.

71

3.2. The Old Fort. In the forefront, the defense walls built by the Germans. Photo by the author.

72

7.1. Coins minted by the Deutsch-Ostafrikanische Gesellscha . Photo by the author.

189

8.1. One of the mosques in ruins at Kaole. Photo by the author.

208

8.2. Sacrifices placed at an ancient tomb in Kaole. Photo by the author.

214

9.1. Ancient Islamic tomb at Kaole. Photo by the author.

236

11.1. Pocket watch from 1890, manufactured in Switzerland and owned by one of Bagamoyo’s traders. Photo by the author.

298

11.2. The popular rose e motif with foliage and waves. Photo by the author.

306

12.1. Participants at the 2002 International Conference viewing the German Boma. Photo by the author.

329

 Maps and Figures Maps 3.1. Some of Bagamoyo’s historical buildings identified for the inscription on Tanzania’s Tentative List. Created by the author.

69

5.1. East African central caravan routes and port towns. Created by the author.

114

9.1. Kusi: The monsoon from the southwest. Created by the author.

240

9.2. Kaskazi: The monsoon from the northeast. Created by the author.

241

Figures 11.1. Layout of a nyumba ya udongo or wa le-and-daub house and its yard. Created by the author, a er Tanzania: Bagamoyo township Survey, 1969.

300

11.2. Layout of a two-story stone house and its yard. Created by the author, a er Tanzania: Bagamoyo township survey, 1969.

301

 Preface The idea to write this book germinated when the government of Tanzania in January 2006 announced on its web page that Bagamoyo and the East African caravan trade route had been put on the country’s tentative list as a potential world heritage site. Bagamoyo was the most important trade port on the East African mainland coast during the second half of the nineteenth century. It was both the terminus for the trade in ivory and slaves from the interior of the continent and the entrepôt for the intercontinental trade in co on cloth, weapons, and ammunition. This announcement caught my interest because I had held a position between 1996 and 2002 as a sociocultural analyst at the Swedish Embassy in Dar es Salaam. I was involved in an informal group of concerned Tanzanians from the University in Dar es Salaam, UCLAS (University College of Architectural and Land Use Studies), the Department of Antiquities, and the Ministry of Natural Resources and Tourism, who worked to find financial recourses to arrange an international conference in Bagamoyo. The aims of the planned conference were both to identify the “outstanding universal values” of the town and the caravan trade, and to evaluate the potentials of Bagamoyo to be inscribed in UNESCO’s World Heritage List. This process culminated in September 2002 with the arrangement of the conference. The University of Gothenburg worked through a separate process, to establish heritage studies as one of the university’s strategic research fields. This inspired researchers at the university and Chalmers University of Technology interested in heritage-making and the Swahili world along the East African coast to formulate an application for financial support to undertake a fact-finding mission and to make contacts in Tanzania. The application was approved and a visit was carried out in 2011 to Tanzania, including meetings and discussions with researchers and experts on heritage conservation in Zanzibar, Dar es Salaam, and Bagamoyo.

x

•

Preface

The research team included, besides the author, Katri Lisitzin, Mikela Lundahl, and Inger-Lise Syversen. They formulated the research proposal “Frictions, Fractures and Cultural Resilience in Swahili Coastal Towns” that eventually received funding from Sida/SAREC. This grant made it possible for me to once again visit Bagamoyo in 2013 and reestablish contacts with some of the members in the informal work group that planned and arranged the 2002 International Conference. I am grateful for this support and the support received from the School of Global Studies, Gothenburg University, in the form of paid time to write this book. I am also grateful to colleagues and friends in Sweden and Tanzania who assisted me in this process, and in particular to Vilhelmina Poppius who helped me with the text. All remaining mistakes, misunderstandings, and misconceptions are, of course, mine.

 Introduction Many visitors to Bagamoyo fail to notice the Old Stone Town to their right, at the shore of the Indian Ocean, when they head for the Catholic mission station with its museum a few kilometers to the north of the center of the town. If they return for a tour through the town, which was placed in 2006 on Tanzania’s Tentative List as a potential World Heritage site, they may experience the closed, o en abandoned houses, and the tranquility of the streets. Guides will volunteer, for a negotiable sum, to take the tourists to the buildings correctly or incorrectly associated with the history of the slave trade. Horrors of the past are here mixed with the beauty of the palms along the coral sand beach and the genius loci or the spirit of the town, described by many as its slow rhythm and fresh air. Hidden behind the massive stone walls and solid carved-wood doors and under the dust on the narrow streets and lanes is a both fascinating and frightening history. Heritage-making, a concept purposely selected in this research to capture the various processes in the identification, formulation, and presentation of a cultural heritage, implies that different options may be available to identify a heritage, what it represents, and how it shall be presented. Memories are created and o en used to express aspirations and claims. Heritage-making is an ongoing process of forming an identity. Inherent in this process is that a heritage, particularly a world heritage, ideally should carry only one interpretation, image, or story. This should be captured in its “outstanding universal value,” the criteria that a cultural heritage must meet to become a world heritage. This logic implies that alternative stories or interpretations are silenced. Additionally, interpretations or claims of rival groups at the local level, in the academic world, and among experts on heritage conservation are o en marginalized through processes of exclusion when a particular interpretation is selected and promoted in the branding of the heritage.

2

•

Muted Memories

Heritage-making may be seen as a force that at the same time promotes nationalism and globalization since the same forces are behind these two processes, which are occasionally seen as contradicting each other. Both result in standardization and homogenization. Nationalism does this through an expected or imagined sociocultural similarity among the population within a given geopolitical territory. Globalization, on the other hand, does this in the standardized creation of the World Heritage List. In recent years, the world has had to witness the violent and purposeful destruction of world and cultural heritage as a means of resisting and rejecting both processes. This is partly a reaction against the heritagemaking processes that are seen the results of a Western-dominated globalizing world, and in a empts to erase what are seen as challenging and misinformed concepts, interpretations, traditions, and visions. One of the paradoxes in world heritage-making is that it is both an effect of and a force behind globalization and nationalism. The race among UNESCO’s (United Nations Educational Scientific and Cultural Organization) member states to have their cultural heritage inscribed on the World Heritage List (WHL) bolsters national pride and global homogenization or standardization, through the packaging of the heritage, which, by itself, is a globalizing force. UNESCO’s efforts to create a balance among the continents regarding the number of inscriptions on the WHL, which nurtures heritage-making in underrepresented continents, is both a force that promotes, and, at the same time, exemplifies the effects of globalization. The efforts to have Bagamoyo inscribed on the World Heritage List may be explored through the processes of heritage-making and globalization. Both processes contribute to the disembedding, or li ing out, of the heritage from its local context, when it is presented to a global audience. The world heritage expert system in the form of trained experts, rules, and power structures regulates the world heritage phenomenon. This system imposes a hegemonic matrix over a locally embedded cultural heritage. This is not to say, however, that the local understanding of a world heritage is dictated by the values ascribed to it at the global level, as shown, for example, by the destruction of Sufi tombs and mosques in Timbuktu, Mali, to which UNESCO ascribes “outstanding universal values.” These were acts on the ground to disembed the place from a globalizing world. It therefore seems more common that the interpretations at national and global levels correspond, since a potential world heritage, in order to be recognizable and recognized on the global scene, has to be identified, described, and presented according to UNESCO’s “best practice” format. The “best practice” concept implies the application of a normative power in the hands of UNESCO in heritage-making. The demand placed on applicants to the WHL to follow the “best practice” concept creates ho-

Introduction •

3

mogenization and standardization. Furthermore, the accepted formulas, procedures, and technical language that surround the promotion and acceptance of a potential world heritage may further contribute to its disembedding from the local context. UNESCO’s normative power is exercised when the qualities of both old and new sites are evaluated by using the defined norms. The disembedding process reveals another paradox in both heritage-making and globalization. To be qualified for the World Heritage List, an item of cultural heritage must be of “outstanding universal value,” which means that it must be both outstanding in comparison with all other heritage sites or a phenomenon on the global level in its own category, while being of universal interest. It must, in other words, be presented both as something unique but recognizable as a thing of interest for global heritage experts and tourists. The demand for both outstanding and universal places the identification, formulation, and presentation of the heritage squarely in the hands of experts despite the good intentions of UNESCO to have the local population’s acceptance and perspectives included in this work. The reliance on international experts and organizations easily implies that the presentation of the heritage is further disembedded from the local context. The identification and presentation of a potential world heritage site’s outstanding universal value is therefore an act performed on a tightrope between what is internationally accepted and acclaimed as world heritage, and what is locally understood as a lived in or lived with cultural heritage. The nomination dossier of a heritage site proposed for inscription onto the WHL must be stringent, coherent, and convincing. Although a world heritage site is inscribed on the WHL on the basis of fulfilling a number of criteria, the coherence expected in the interpretation and presentation of its uniqueness not only invites, but also demands, a univocal text or monocular image of what the heritage site represents. This fosters a tunnel view, as alternative interpretations, stories, or images tend to be muted or made invisible. This is what happened, as will be shown below, when the government of Tanzania in 2006 decided to present Bagamoyo as a representative site of the East African slave routes on the country’s Tentative List, a step before the heritage is formally nominated to the World Heritage List. When the heritage site is meant to a ract global tourists, it risks becoming further disembedded from its local context. Packaged and presented in order to increase the flow of money from tourists, the heritage site is conveniently framed in a way which makes it recognizable and a ractive for a specific category of global citizens. This is rarely as obvious as it was in the opening speech delivered by the Tanzanian Minister for Tourism

4

•

Muted Memories

at the 2002 International Conference arranged in Bagamoyo. She claimed that “it is necessary to put Bagamoyo on the World Heritage List so that people in America and the West Indies can come to Bagamoyo and trace their backgrounds” (author’s field notes, September 2002). With this, the history of the place was li ed out of its local context and translated into the global metastory of slavery and the slave trade. This transformation was taken one step further in 2006 at the launch of the site on the country’s Tentative List, when it was explained that “not even 150 years ago, millions of Africans had to bear a cruel fate. They were captured by slave hunters, chained together and forced to walk sometimes hundreds of kilometers to be sold for example to planters who used them as cheap labor in their fields” (UNESCO 2006). The presentation places itself squarely within the global metastory of the slave trade and slavery, which is modeled on the Atlantic slave trade and the US form of plantation slavery. The disembedding of the heritage was followed by the reembedding of a different story from a different part of the world and from a capitalist free market system into the pre-capitalist East African context. Different influences from the global heritage scene contributed to the branding of Bagamoyo, despite existing knowledge, as a representative of the slave routes. One was UNESCO’s launching in the mid-1990s of a number of route themes as the Slave Route Project (herea er SRP) and the Trade Route Project. The SRP had, at the time of the 2002 International Conference, contributed to both the growing research activities on the role of the slave trade and slavery in West Africa, the Caribbean, and North and South America, and the declaration of slavery as a crime against humanity at the 2001 conference in Durban. There is today a more nuanced understanding of both the local effects of the slave trade on the societies in West Africa, of the plantation systems in the western hemisphere, and the dark and destructive history of colonialism. The devastating effects of the slave trade on African societies have been known for centuries. Rosalind Shaw has more recently shown in her publications (1996, 2002) how the incredible cruelty during the centuries of slave hunting and trade in western Africa contributed to an increase in witchcra accusations. The incomprehensible violence, blood-sucking character of the business of the slavers, extraordinary richness concentrated in a few hands, and the breaking up of social ties, obligations, and solidarity were projected onto witches who lived, and still live, in a separate world called the “Place of Witches” from where they interfere in the lives of those in their communities. Here the witches live in affluence like members of the post-colonial kleptocratic class. It seems plausible that the increase during the slave trade in witchcra beliefs and accusations today find new fuel and targets through the inequalities and lack of trans-

Introduction •

5

parency that characterize many post-colonial societies and that the slave trade still affects the lives of people today. The SRP has also stimulated an increased interest among tourists to visit places associated with the slave trade. This is particularly prominent on the west coast of Africa with its slave castles and underground chambers, which have become sites of conscience. The tourist industry that grew up around these sites has been called “dark tourism” (Lennon and Foley 2000) or “thantourism” (Seaton 1996), while the places of memory themselves have been termed “places of dissonance” (Turnbridge and Ashworth 1996). These sites have become moorings for the descendants of those dislocated through the slave trade. Rosabelle Boswell, writing about descendants of former slaves on the Seychelles and Mauritius in the Indian Ocean, states that “slaves and others like them lost more than their identities, they lost the rich and diverse possibilities that could have made them into self-affirming human beings” (Boswell 2008, 20–21). The enforced displacement and open cruelty deprived many slaves of their families, ancestral graves, the characteristics of the familiar se ing and landscape into which so much of the memory is anchored in an oral culture. The tangible heritage from the slave trade may assist to enrich the intangible memory of the trade. Iron chains and slave castles may, besides representing a dark memory, communicate new knowledge and feeling among the descendants of slaves. They are concrete evidence of an open racism and systematic discrimination where the visitors may easily connect their own experience of racism and discrimination with the misdeeds of the past. This compression of time provides explanations and may give rise to strong emotions, as captured by Lisa McNair during a visit to a West African slave castle: “as the tour guides were explaining and walking through it, you could feel the spirit of our ancestors upon you. I don’t think you ever outgrow the experience of that feeling. Grown men in the group shed tears” (McNair 1998). The branding of Bagamoyo as a slave port, stimulated by the SRP and UNESCO’s wish to also include the Indian Ocean region in the project was, however, not based on solid information and evidence. Furthermore, the branding meant, as will be shown, that only one aspect of the caravan trade was li ed out and merged with a globally recognized metastory. The cultural heritage of Bagamoyo, the caravan trade along the central routes from Lake Tanganyika, and beyond, to the Indian Ocean, was transformed into a slave route in an a empt to a ract the growing “roots tourism”: people from the Americas go in search of their origins in places like Bagamoyo. When the image of the slave route was selected, the alternative stories of the East African caravan trade were muted.

6

•

Muted Memories

The East African caravan trade during the nineteenth century developed into an institution. A rather uniform caravan culture emerged where the caravans o en appeared as a whole society. This included sometimes three to four thousand men, women, and children, on the move, together with goods and other cultural elements such as languages and religions. The conventional view of nineteenth-century East African history was created by the many “explorers” who set out from, for example, Bagamoyo to “discover” the interior of the continent. What is less well known is that they followed well-trodden paths, established by caravan porters and traders from the interior of the continent. These porters and traders, at the end of the eighteenth century, began to extend their trading networks to reach the port towns on the Mrima coast, the area between the Tanzanian/ Kenyan border and Dar es Salaam. They brought their trade items, primarily ivory, but also slaves, in exchange for co on cloth, weapons, and ammunition, which they obtained from traders in the coastal towns. This initiative was later followed by traders with caravans organized from the coast that set out to reach the heart of Africa. Bagamoyo became the interface connecting the trading networks on the continent with the Indian Ocean networks and further afield to markets in China, Japan, Europe, and in the Americas. The demand for co on cloth and the tastes of the peoples in the interior of the continent stimulated both the industrial development of the textile industries in North America and India and influenced the textiles they produced. Simultaneously, the desire in Asia, the Americas, and Europe for ivory and copal, and the increasing demand for slaves in societies in and around the Indian Ocean and in the Americas triggered an increased caravan trade. This consisted of trading ivory from the interior, copal from the coast, slaves from the entire region, and cloth, weapons, and ammunition mainly from India, Europe and the Americas. The demand for trade items carried by the caravan porters to a trade town like Bagamoyo generated hectic processes of globalization. During the second half of the nineteenth century the town annually received tens of thousands of caravan porters, predominantly from the Nyamwezi area east of Lake Tanganyika and south of Lake Victoria. The lucrative caravan trade also a racted traders and financiers from India and the Arabic peninsula. European colonialists, first Germans and then those from the British Isles, trained to think in terms of tribes, soon ran into difficulties to comprehend what they met and saw when they tried to take political control over the cosmopolitan town. This study, therefore, is an illustration on how a more comprehensive understanding of an early phase of globalization may correct the previous view created by early colonialists, “explorers,” missionaries, and anthropologists.

Introduction •

7

The way we construct and use our heritage may help us to tackle and come to grips with some of the challenges in our o en conflict-ridden world. It is hoped that this study may contribute to challenge present-day growth of nationalist movements and arguments. Contrary to the arguments by neo-nationalist agitators, people may perfectly well, as will be shown here, enjoy living in a cosmopolitan society that provides multiple means to maneuver between and among different identities in the hierarchical construction of their society. Furthermore, people may equally enjoy living in a society without strictly defined territorial borders and in a sociopolitical context that stimulates and facilitates travels and trade over land and sea allowing multiple places of residence on different continents. This volume consists of two parts. The first is called “Heritage-Making, Branding, and Globalization,” and captures fi y years of heritage-making in Bagamoyo. Special focus is placed on the efforts to have Bagamoyo nominated for UNESCO’s World Heritage List as a representative of the slave trade of East Africa. It discusses world heritage-making as both an effect and a cause of globalization and the frictions and fractures that may develop during the efforts to identify and present the heritage. Frictions and fractures are likely to appear when the available information is stretched beyond its capacity to protect and promote a specific interpretation of the value of a potential world heritage site. It is argued here that this happened when the slave route theme was selected instead of the trade route theme in the launching of Bagamoyo as a potential World Heritage site. The second part, called “Commerce, Competition, and Consumerism: Bagamoyo and the Caravan Trade,” explores the memories from the caravan trade that were muted when the slave route theme was selected. It explains how the town could have been used to represent the interface between the East African caravan trade with iron, cloth, copal, and slaves, and the Indian Ocean trading networks in cloth, metal products, weapons, and ammunition. The focus in the first chapters in the second part is on the East African caravan trade as an institution; a society consisting of men, women, and children and specialized professionals oscillating between the interior of the continent and the Indian Ocean. In the last chapters, the focus shi s to the intensive process of politics of identity formation and communication that were triggered in Bagamoyo by the intense interaction of people from different backgrounds brought together by the new wealth from the caravan trade. The vibrant processes of hierarchization and creolization were fueled by three main characteristics of the East African trade towns: commerce, competition, and consumerism. This trade was o en embedded in practices and institutions that handicapped the European and American trad-

8

•

Muted Memories

ers who expected to trade according to market principles. The focus in the last chapters is on the cultural processes of creolization, inclusion and exclusion, struggle for power and prestige, control and colonization. The wealth generated through the caravan trade created a unique culture and society, initially beyond the control of the colonizers. The information in the first part of this work is to a large extent based on participant observation. The process of identifying Bagamoyo’s “outstanding universal value” at the 2002 International Conference on Bagamoyo as a potential World Heritage site consisted of a number of meetings, workshops, and conferences that were a ended by the author. The information on the process—from the arrangement of the conference to the nomination of the town on Tanzania’s tentative list in January 2006—is based on interviews and wri en documents. The second part of this work is a journey into the past. It is an a empt to present what was muted when the “slave route” theme was selected, at the expense of the caravan route theme. It is mainly a compilation of the rich, but sca ered, information available in wri en documents, and a reinterpretation of the established wisdom about East African caravan and slave trade. Experiences and insights from ten years of work and residence in Tanzania have guided me through the literature used for this text. The conventional perspective has been to view trade through the eyes of the “explorers,” missionaries, and traders who traveled from the coast and westward. In the present work, the view is turned 180 degrees to follow the Nyamwezi traders and porters in their caravans from their homeland in central Arica and down to the coast. It is generally understood today that they initiated the caravan trade that, a er some decades, was dominated by traders from the coastal societies. This work is in the field of historical anthropology. The work of, for example, Johannes Fabian (2000) on the reasons behind the caravan leaders’ frequent violence against their caravan porters and guides and Tanya Murray Li’s (2007) research on centuries of colonial and post-colonial projects and programs to change the life of “the natives,” the productivity in their cultivation systems, and the landscape on Sulawesi, Indonesia, have guided me to develop the approach adopted in this work. The text navigates in the border area between the search for causalities as to find answers to the questions of why it was the Nyamwezi people in the interior of the continent who initiated the long-distance caravan trade or why Bagamoyo developed into the most important port and trade town on the East African coast and, on the other hand, the local and global implications of the booming caravan trade. It also navigates between the fields of social anthropology, development studies, and critical heritage studies. It is inspired by the work of Jan Turtinen (2006), Lynn Meskell (2012, 2013,

Introduction •

9

2014, 2015), Christoph Brumann and David Berliner (2016) and Brumann (2016) by reporting what has been said and done (and sometimes what was not said or done)1 at meetings, workshops, and conferences aimed at the identification, branding, presenting, and, in the case of UNESCO, inscription or referral of a potential world heritage site. There are today a number of unpublished doctoral theses, as well as published books and articles, on the history of Bagamoyo and the life around the caravan trade. The rich information available in these studies, which expanded considerably in number from the mid-1990s, is used here to paint a picture of what took place when the caravan trade in ivory and cloth, but also slaves, between the interior of the continent and the coast grew and bloomed, during the second half of the nineteenth century and up to World War I. This was when the German colonizers began to use their railway between Dar es Salaam at the Indian Ocean and Kigoma at Lake Tanganyika with an extension to Mwanza at Lake Victoria, which implied the end of the caravan trade along the central routes in East Africa and a rapid decline of Bagamoyo town.

Part I: Heritage-Making, Branding, and Globalization The first chapter, “Bagamoyo: A History of Practices, Principles, and Partnership in Heritage-Making,” investigates the heritage-making efforts in Bagamoyo from the mid-1950s to the millennium shi . The first efforts, initiated during the British colonial regime, focused on the protection of specific objects such as doors and door frames carved before the year 1940. At the end of the 1970s and early 1980s the focus had shi ed to the protection of the Old Stone Town in Bagamoyo as a lived-in conservation area. At the beginning of the new millennium that focus had once again shi ed, now to Bagamoyo as a slave port together with associated sites along the central East African caravan routes and how heritage-making could be integrated into efforts to expand the tourist sector and generate resources for poverty reduction. Chapter 2, “Heritage-Making: The 2002 International Conference” explores the search for Bagamoyo’s outstanding universal value (OUV) during the international conference arranged in 2002 in the town. The conference was arranged with international and national experts to evaluate Bagamoyo’s potential to be inscribed on UNESCO’s World Heritage List. Four themes were tested during the conference: Bagamoyo and the slave and ivory trade, Bagamoyo as the entrepôt for Islam and Christianity to the heart of the continent, Bagamoyo as the entrepôt for Kiswahili to the heart of the continent, and Bagamoyo as an archaeological site.

10

•

Muted Memories

The conference agreed to recommend that the government of Tanzania nominate Bagamoyo under the heading “The Slave Port of Bagamoyo and Associated Sites and Routes of Memory” with the two entrepôt themes as supportive arguments. The third chapter, “Fractures in the Image of Bagamoyo: Despair or Joy?,” focuses on how the heritage was presented and what the nomination represents as a potential World Heritage site. The name Bagamoyo, a combination of the Kiswahili terms for “heart” and “lay down,” invites at least two different interpretations. One is that the slaves from the interior of the continent “laid down their hearts” at Bagamoyo before they were shipped to unknown destinations. Another is that the caravan porters laid down their cargo and let their hearts rest a er the long journey to the coast. When the first theme of suffering, humiliation, and despair was chosen in the presentation of Bagamoyo’s outstanding universal value in the nomination text, there was no room to present the caravan trade as a cultural phenomenon characterized by entrepreneurship, agency, and joy. UNESCO’s work with the World Heritage List is both an effect of and a force causing globalization, as shown in chapter 4, “World Heritage and Globalization: The Bagamoyo Case.” One of the effects of heritage-making is standardization. The heritage shall be described, presented, and conserved according to well-established procedures, principles, and practices. Processes of heritage-making may cause disembedding and homogenization, as well as the globalization of the unique. Globalization is at the same time causing heritage-making through both the race among nation-states to have their heritage inscribed on the List and through UNESCO’s work to have more nominations from “underrepresented continents” for inscription. The invention of new categories for inscription is meant to facilitate nominations from states and continents that hitherto are not so well represented on the List. It is argued here that the metastory on slave trade in UNESCO’s Slave Route Project, launched in the 1990s, which is dominated by the experiences from the Atlantic slave trade and US form of plantation slavery, influenced the theme and tone chosen by Tanzania for the presentation of Bagamoyo.

Part II: Commerce, Competition, and Consumerism: Bagamoyo and the Caravan Trade The first section in Part II turns the perspective 180 degrees. The conventional approach has been to approach the African continent, popularly presented as the “Dark Continent” from a vantage point of the coast. There is a long tradition in travelogues and other texts in describing how

Introduction •

11

the authors walk westward from the East African coast. In contrast, chapter 5, “Entrepreneurs and Explorers from the Heart of Africa,” explores the role that the Nyamwezi people, living south of Lake Victoria and east of Lake Tanganyika, played in the establishment of the caravan trade. When they extended their trading network with salt and iron tools in the interior of the continent to also reach the coast, they added ivory and also slaves as the major trade items. This created a direct interface between the continental and the Indian Ocean trading networks, with Bagamoyo as the pivot. The mercantile middlemen in the Swahili coastal towns were not late to profit from the expanding trade. At the end of the nineteenth century, between forty and sixty thousand, mainly Nyamwezi, men and women walked every year during the caravan season from the interior of the continent to the coast and back again. The initial control they had over the organization of the caravans, and the fact that they both constituted the majority of, and the most sought-a er of the porters, gave them a significant influence over the development of the caravan culture that emerged during the nineteenth century along the central routes through Tanganyika/Tanzania. All Nyamwezi porters who reached Bagamoyo were not free men and women. Some were slaves. The caravan trade in ivory was intimately interlinked with slavery and slave trade, as slaves were exchanged for ivory and ivory for slaves. Chapter 6, “Pawned, Preyed Upon, Purchased, or Punished: Slaves and Slavery in Nineteenth-Century East Africa,” explores the roles of slaves and slavery in the societies involved in the caravan trade and their appearance in the trade itself. Two areas have been identified in East Africa with high proportions of slaves; the Nyamwezi homeland area and the societies along the Indian Ocean coast, both heavily involved in the caravan trade. Slaves did not only appear as porters, but also as commodities to facilitate the progress of the caravan when bartered for ivory and food along the routes. Slave were sold as merchandise at the slave markets in Zanzibar in return for imported goods like cloth, weapons, and ammunition. The caravan trade developed into a sociocultural and economic institution much more complex than the hunting of slaves in the interior and the marching of them to the plantations on the coast. And contrary to what is stated in the nomination of Bagamoyo for the tentative list, the town was never a major slave port. The lucrative caravan trade not only a racted traders and financiers from the Indian Ocean trading networks, but also Europeans in the scramble for Africa. The former patricians in the ports on the Mrima coast, the middlemen between the two trading networks, came increasingly under pressure from two colonizing powers; the Sultanate on Zanzibar and the German colonizers. Chapter 7, “Conflicts and Clashes in the Competition

12

•

Muted Memories

over the Caravan Trade on the Central Routes,” captures the escalating struggle over the trade between the former middlemen-merchants and the intruding Arabic and European colonizers. Bagamoyo experienced an invasion during the late 1880s by rebels from Pangani and Saadani, two port towns north of Bagamoyo, bombardment from German naval ships, invasion by German soldiers, a siege by African rebels, and a blockade by German and British naval ships. The escalating completion over the trade resulted in riots and rebellions and eventually resulted in the formal German colonization of the area. Bagamoyo developed not only into the major terminus for the caravan trade but also became the major entrepôt not only for “explorers” and colonizers, but also for the spread of Islam and Christianity and the language Kiswahili to the interior of the continent. Bagamoyo, in other words, was not only the recipient but also the provider not only of material goods traded with up-country societies but also cultural elements that accompanied the trade. The result was the development of a cultural route through which people, goods, ideas, religions, and more flew in both directions. Chapter 8, “Bagamoyo and the Caravan Trade: The Entrance to the Heart of Africa,” expands three themes discussed during the 2002 International Conference during the search for the town’s OUVs. The caravans along the central routes became the avenues for the spread of Islam. Instrumental was the establishment of the Quadiriyya Brotherhood in Bagamoyo from where this form of Sufi Islam spread along the central caravan routes. These routes and the caravan trade were also instrumental to the spread of Kiswahili, which further accelerated when the German colonialists decided to use Kiswahili as their administrative language in their rule over the colony from their capital in Bagamoyo. Bagamoyo also became the place for the first Catholic mission station established on mainland East Africa where missionaries were educated and sent out to form up-country congregations. The chapter also touches upon the spread of lethal diseases along the caravan routes. Bagamoyo is a comparatively recent trade town compared to some other towns in what has been called the Swahili corridor along the coast. It did not exist during the “classical” Swahili period before the Portuguese entered the Indian Ocean and destroyed much of the intercontinental trade. It grew from the early nineteenth century when it became the interface between the continental and the Indian Ocean networks. Important reasons for the establishment of Bagamoyo as the main port on mainland East Africa was the proximity to Zanzibar and the wider world to the east, which was covered by the monsoon or trade winds belt, and the a raction of Nyamwezi caravans from the west that brought ivory and slaves to the coast. The movement of people along the coast and the arrival of new

Introduction •

13

groups from the east and the west created an intricate political landscape, which is captured in chapter 9, “Old Bagamoyo,” and constant negotiations of the established power balance. Caravan traders and financiers from societies around the Indian Ocean were a racted to Bagamoyo to organize their own caravans to the interior, when the town grew to become the most important port on mainland Africa, particularly for the ivory and copal trade. Suddenly the former mercantile class or patricians in the port towns, the long-since established middlemen between the two trading networks, saw themselves marginalized when new actors appeared on the scene, as presented in chapter 10, “Fluid Identities: Politics of Identity in Multicultural Bagamoyo.” The cosmopolitan character of the town created vibrant processes of fission and fusion, inclusion and exclusion, identity-making and boundary protection. This became particularly evident in the creation and re-creation of two broad categories of people in the coastal societies: the “civilized”, influenced by Islam and living in a town, and the porters, the “barbarian”, non-Muslims from up-country societies. The European and American “explorers,” traders, missionaries, and colonizers who entered the scene during the second half of the nineteenth century, trained to think in terms of tribes and stable monolithic cultures with well-demarcated boundaries, were ill-prepared to comprehend and describe what they met in a place like Bagamoyo. The dominant cultural codes in the coastal societies stipulated that wealth could be used to claim and maintain respect and influence. One of the most dominant means was through public displays of a trader’s or a family’s economic wealth. Precious material goods in this conspicuous consumption were the stone houses, wealth converted into food, music and dances during competitive feasting, Chinese wares, and cloth. Cloth became one of the items in highest demand along the caravan routes and became an institution in itself through its economic, religious, social, and political importance. The booming caravan trade created both a group of newly rich who tried to convert wealth into power and prestige, and increasingly indebted traders/middlemen, which gave their own dynamics to political life. Other means were to show generosity in patron-client relations or to finance feasts, which o en took the form of competitive dance events, another crucial cultural feature in the Swahili world. These forms of conversion of material wealth into religious, social, and political prestige are covered in chapter 11, “Conspicuous Competitive Consumption and Communication by Means of Cloth.” The caravan trade was embedded in specific institutions and traditions, which gave it its character. For the Nyamwezi porters, it was important to establish pseudo-kinship ties with people along the routes to and with

14

•

Muted Memories

traders in Bagamoyo. For the traders, it was equally important to have well-functioning relations with the traders and porters who brought their goods to the coast. These relations were more of a social character with ritualized exchanges of gi s and counter-gi s rather than relations based on market principles. To the frustration of the European and American traders, they could never compete over the wealth brought to the coast with the traders in Bagamoyo. Different a empts were made by newly arrived colonialists at the end of the nineteenth century to wrench control over trade out of their hands, as explored finally in chapter 12, “Intruders and Terminators: The End of the Story.” For a few years during the end of the nineteenth century, Bagamoyo was the capital in the German colony Deutsche Ost-Afrika. It was only when the German colonizers decided to move their colonial capital from Bagamoyo to Dar es Salaam that they managed to get control over trade. This occurred just before the outbreak of World War I when the Germans built the railway from Dar es Salaam to the western parts of their colony, which passed through the Nyamwezi heartland and facilitated a new trade route. This meant the end of a century-old cultural institution that had developed on the African continent during the first half and boomed during the second half of the nineteenth century.

Note  1. See, for example, the report in C. Brumann (2016) on what was not said or done at UNESCO’s 2012 Commi ee Session held in Saint Petersburg at the same time as the world witnessed the destruction of Sufi tombs and mosques in Timbuktu by the Ansar Dine Islamic group. For the destruction, see C. Joy (2016).

PART

I  Heritage-Making, Branding, and Globalization Part I explores established practices and priorities in heritage-making and branding memory. These processes are shown to be intertwined with other processes on the global agenda as practices and policies in development cooperation. This part also describes the physical qualities of a potential world heritage site that may change while paradigms for preservation and development partnership are refined and negotiated.

1  Bagamoyo A History of Practices, Principles, and Partnership in Heritage-Making This chapter provides a lens through which we may see both how the concept of and practices in heritage-making have developed, and how development-planning in Tanzania has changed from an emphasis on state-imposed self-reliance to a focus on the local a raction of global tourists. “Heritage,” writes Hafstein (2012: 502) “is about change. Don’t be fooled by the talk of preservation: all heritage is change.” Bagamoyo provides an excellent example of how heritage-making, especially in resourcepoor developing countries, is linked to national development priorities and dominant policies and priorities among bi- and multi-lateral development authorities. A constant theme since the late 1960s has been how Bagamoyo’s cultural heritage can be used to achieve increased tourism and economic development. This chapter will explore these intertwined processes throughout fi y years of heritage-making in Bagamoyo.

Bagamoyo as a Window into the History of Heritage-Making Bagamoyo not only constitutes a site of cultural or potential world heritage; the town also represents a history of heritage-making. The halfdecade of conservation efforts in the town illustrates the changes in ideals, ideas, and practices in the history of heritage protection and preservation.

18

•

Muted Memories

This chapter will capture global trends not only in heritage-making but also in development cooperation from the horizon of a small East African coastal town. This will be intertwined with an a empt to explore the different a empts to identify the object(s) worth preserving, the subjects, here understood as those supposed to have an interest in or responsibility for the conservation or preservation of the object, and, when identified, the memory or associative values that the preservation of the object will bring into the future. Different a empts have been made to conserve and preserve Bagamoyo’s cultural heritage (Ndagala 2000). International support has been mobilized, and national and international experts have worked on this since the 1950s. The government of the United Republic of Tanzania eventually nominated Bagamoyo in 2006 for its Tentative List—the country’s inventory of the properties it intends to consider for formal nomination to the World Heritage List. Progress has, however, been slow. Ten years a er the nomination Bagamoyo is still on the Tentative List waiting for its full nomination. As stated in the introduction, the name Bagamoyo (a combination of the Kiswahili words for “heart” and “lay down”) invites different interpretations. The name is o en associated with the thousands of slaves who were shipped from mainland Africa and who are said to have laid down their hearts and le them behind in Bagamoyo and the African soil. There is no doubt that slaves were shipped out of Africa from the port of Bagamoyo, although there were many more slaves shipped from Kilwa on the southern coast of Tanzania and probably also from Saadani a few kilometers north of Bagamoyo. Perhaps, on the other hand, the name Bagamoyo came from the thousands of porters, mainly independent free men from the Nyamwezi / Sukuma area south of Lake Victoria, who, a er the long and troublesome journey, laid down their elephant tusks and other goods and let their hearts rest.1 An eye-witness who visited this vibrant port town during the mid-nineteenth century painted the following picture: With the 20,000 inhabitants, among them 1,500 Indians, with the stone houses, the narrow streets, the noisy market place with hundreds of [sic] selling boots Bagamoyo looks really like an Oriental city. In the Caravan Serai caravan porters, coming from the interior, from Unyema and Uganda, lay down their cargo. During the caravan season Bagamoyo is sometimes overcrowded with more than 10,000 porters, traders and heavy cargo carrying slaves. (F. F. Mueller 1959, as reported in Ndunguru, Kadelya, and Henschel 2003: 4)

Is Bagamoyo a suitable “place of memory” for these brave, industrious, hard-working porters from inland Tanzania who came to the coast to

Bagamoyo

•

19

trade with the wider world? Or is it a suitable place to remember both porters and slaves? Can it be a place both of pride and of shame at the same time?2 These questions are well motivated since a dominant inner force in heritage-making is that only one story or one version of the story can be presented. A peculiar feature of heritage-making within the framework of the UNESCO’s World Heritage standards is the monopolization of what is to be remembered, the story, through the nomination procedures. It may be in the logic of the “outstanding universal value” concept that it promotes both standardization of the story and hegemony of one version of the memory. Alternative stories and interpretations disturb the narrative and have to be neglected and silenced in the presentation of the memory. Not long ago a tree stood close to the shore in Bagamoyo. It was the tree from which German colonialists hanged Africans. It was burnt down, but replaced by a concrete monument with a plaque telling the visitors: “Here is the place where the German colonialists used to hang to death revolutionary Africans who were opposing their oppressive rule.” Bagamoyo’s complex character and qualities as a “place of memory,” spanning a complex and rich history of the caravan and slave trade, colonization, rebellion, and the struggle for independence,3 provide a suitable field for research on the production of memory and meaning through heritage-making. The burnt-down tree may be seen as an act of resistance or, alternatively, as an act of trying to hide or forget the past. Memories may hurt.4 The identification and formulation of the memory is therefore a suitable field for an analysis of power, conflict, and contestation. The crucial issues concern who identifies what ought to be remembered, how the we associated with our heritage is defined and constituted and how the boundaries are drawn since “authority over heritage and political power within the community are . . . to some extent mutually translatable” (Hafstein 2012: 513). Heritage-making in Bagamoyo, culminating when Tanzania put “The Slave Trade Route” on its Tentative List in 2006, is actually a process that spans over fi y years. Its roots can be traced to colonial times when the British identified a number of carved wooden doors and door frames as a cultural heritage well worth protection and preservation. At that time, the focus was on specific material objects. A er independence in 1961, with strong emphasis on development, the heritage was soon associated with potential growth of the economy through tourism. Heritage got a utilitarian value or, expressed another way, “the World Heritage concept makes it possible to exchange a cultural value for a commercial value.”5 The following pages will try to follow this transformation of the heritage from specific material objects, which deserved to be preserved per se, to the use of the heritage for the develop-

20

•

Muted Memories

Illustration 1.1. The monument in place of the hanging tree in Bagamoyo. Photo by the author.

Bagamoyo

•

21

ment of society. Focus will be on four specific efforts of heritage-making in the town and the information available about them: the efforts by the British colonial government to protect the so-called Swahili doors in Bagamoyo’s Stone Town; a Tanzanian/Swedish survey of Bagamoyo carried out in the late 1960s; the Tanzanian/UNESCO work to identify buildings for preservation, including the 1983 workshop in Bagamoyo; and finally, the 2002 International Conference on Bagamoyo as a World Heritage site that resulted in the 2006 nomination for the Tentative List.

Heritage-Making in Bagamoyo Bagamoyo District is one of the poorest parts of the country, but the town of Bagamoyo is rich in historical significance, in heritage-making efforts, and in development and conservation plans, as captured in the following statement: “Detailed plans were formulated already at the beginning of the 70s. UNESCO assisted to develop [sic] a new plan during the late 70s and early 80s. NORAD [Norwegian development cooperation authority] gave additional support during the late 80s. Mr. Kamamba drew a new plan in the early 90s” (Ndagala 2000: 3).

Colonial Efforts Heritage-making has a long tradition in Tanzania. In 1937, the British colonial power instituted its Preservation Ordinance and, in 1957, established its Antiquities Division (Russell 1980: 17). The policy of conservation at that time had as its fundamental principle “to make good the structures as they stand, new work being added only where it is structurally necessary for the safety of the building, or where the greater part of the original materials are to be found in the vicinity and the original aspect of the parts to be reconstructed is to all intents and purposes certain” (as cited in Russell 1980: 16). The policy had a clear focus on material objects; in the Bagamoyo case the focus was on wooden doors and frames carved before 1940. Only limited a ention was paid to the symbolic importance of the doors, their integration in the layout of the house, their different functions in the houses in a trade town, the meaning of the symbols and motifs that decorated them, or what the different carving styles represented. Most of the more than three hundred doors found in Bagamoyo6 belonged to houses in Stone Town, but carved doors were also found in the neighboring villages (Areskough and Persson 1999: 29). These doors are a prominent feature in what has been called Swahili culture. The skill

22

•

Muted Memories

in carving them is intimately connected with Islamic mercantile culture. They signal the wealth and power of the house owner and they separated the private interior of the house from the public exterior. They are typically used in pairs where the le door represents the female side and the right the male (Areskough and Persson 1999: 29). Questions have been raised as to whether these doors are an East African cultural invention or whether the concept has been imported and then copied along the coast. Similar types of doors may be seen in other ports around the Indian Ocean but “East Africa is the only place where they still exist in large numbers and where a continuing tradition of doorcarving is practiced” (Aldrick 1988: 27). The most common names used for the identification and description of various styles of carving and door construction, like Oman, Gujarat, or Zanzibar (Indo-Persian) styles indicate strong influence from other parts of the Indian Ocean trading networks. Whatever their origins, these doors have long since been an integral cultural characteristic of the Swahili world.7 The doors and their decoration signaled messages to those who passed the front of the stone houses, as will be further explored in chapter 11. Li le of this caught the colonizers’ interest since they were mainly preoccupied by the protection of the physical doors and frames. They focused on the doors and frames per se and the cra smanship behind their production. The doors were seen as an expression of the sophistication of the coastal culture and as an evidence of the influenced from Asia/Arabia that stimulated the development of Swahili culture. As such they were well worth protection and preservation through colonial interventions and legislations.

Post-Independent Heritage-Making in Tanzania The Preservation Ordinance from 1957, inherited from the colonialists at Tanganyika’s independence in 1961, was amended in 1964 and became the Antiquities Act. The new act extended the conservation concept from the Preservation Ordinance’s focus on specific items like “monuments” and “relics” created before 1863 and “protected objects” like carved wooden doors and frames created before 1940 to now also include “conservation areas” (Russell 1980: 17; Mturi 1983b: 115). The previous focus on objects as cultural heritage is clearly pronounced in UNESCO’s definition of Cultural Property: Cultural property as is in accordance with the various UNESCO definitions means: 1.1.1. Archaeological, historic, ethnological or scientific sites. 1.1.2. Structures of other features which are of historic, cultural, artistic, architectural or scientific significance whether religious or secular.

Bagamoyo

•

23

1.1.3. Buildings or groups or buildings which are of historic or architectural value including groups of traditional structures, historical quarters in urban or rural areas. 1.1.4. Objects which are of archaeological, cultural, historical or scientific significance including those existing in or being recovered from immovable property or those concealed in earth which may be found in archaeological sites or elsewhere. (Russell 1980: 47 [my emphasis])

The 1964 amended Antiquities Act protected “all the monumental remains and movable objects found in the 9th–18th century port-towns” (Mturi 1983b: 113). The focus in the Act is on the “classical Swahili” settlements such as Kilwa, Songo Mnara, Kaole, Tongoni and Kunduchi. Bagamoyo was excluded from the list of port towns since the town was first established at the end of the eighteenth century and early nineteenth century. The amended Antiquities Act was therefore ill-suited to protect the heritage in Bagamoyo.

The First A empt to Exchange Bagamoyo’s Cultural Heritage for Development The houses in Bagamoyo and their inhabitants have repeatedly been surveyed, measured, and demarcated. One of the first post-independence efforts was the joint Tanzanian and Swedish project “Housing, Building and Planning in Developing Countries”—a collaboration between researchers at the Department of Sociology, The University College, Dar es Salaam, Tanzania, and the Department of Architecture, University of Lund, Sweden. The findings are presented in the report Tanzania: Bagamoyo Township Survey 1969.8 The general picture of Bagamoyo painted in this text is of a poor, tranquil town without electricity or a sewage system and with very bleak potential for development and modernization. Population statistics showed that the town had fewer than six thousand inhabitants. The 1969 survey mentions the importance of agriculture and fishing as sources of income for the inhabitants, since half of the families in the survey had their own shamba, cultivated field, for subsistence farming. This should be compared with the formerly thriving agricultural sector during the second half of the nineteenth century that managed to feed the tens of thousands of caravan porters and traders for three to four months before they returned to the hinterland and even exported food crops to areas around the Indian Ocean. It may, however, be safe to assume that the agricultural system during the caravan trade depended on slave labor and that much of the system collapsed with the abolition of slavery and as a consequence of a culture that saw cultivation as a task for slaves.

24

•

Muted Memories

During the caravan trade at the end of the nineteenth century, Bagamoyo was an important trader with the societies along the caravan trails in salted, dried and smoked fish, which was produced in Bagamoyo and neighboring villages. At the time of the 1969 survey, fishing had gone back to its most basic form and only a small surplus was produced for trade outside the vicinities of the town. In this sense, fishing paralleled the development of agriculture in Bagamoyo. It is also stated in the report that at that time there was no tourism and no hotels in Bagamoyo. However, “it could be possible to a ract tourists to Bagamoyo, because it is so close to Dar es Salaam, there are fine beaches, and interesting buildings. In that case the need for parking, public latrines and washplaces, serving of food should be considered and maybe later a small hotel” (Tanzania: Bagamoyo Township Survey 1969: 36). Here the seeds are sown for a view that would become more pronounced in the coming decades: the town’s cultural heritage and fine beaches may be used to a ract global tourists , but at this specific point in time, they a ract weekend visitors from Dar es Salaam. The interesting buildings mentioned in the quotation are never further identified, but it can be read between the lines that the buildings of the town contribute to the cosmopolitan character of Bagamoyo’s architecture, which is aptly presented in the following way: In the centre of town and along the shore there are one-and two storey stone buildings in an Arab influenced style with beautifully carved wooden doors. In the south and southeast parts of the town, the Germans built a place for governor customs buildings, a post office and a few other buildings. Further south there is an area with English “low-cost”-houses and just south of the town there are three English bungalows for the higher officials in the town. In the other parts of the town, there are houses made of mud and poles, partly in a planned area with the streets forming triangles, partly in an unplanned area. (Tanzania: Bagamoyo Township Survey 1969: 6)9

The report specifies some suggestions regarding the development of tourism in Bagamoyo. The proximity to Dar es Salaam, the special atmosphere of Bagamoyo, the historic milieu and the nice beaches can be capitalized to develop the town into a future tourist and holiday resort. It is suggested that the local tourist industry should be run by the municipality and that resources should be secured for the “restoration of historical places of interest in the town” (Tanzania: Bagamoyo Township Survey 1969: 245). We can sense here the embryonic identification of a heritage that is more than a number of carved doors and frames, but nothing is said about why these are “historical places of interest” or “interesting buildings” that ought to be repaired and maintained. The report gives no clues to why

Bagamoyo

•

25

the history of Bagamoyo and its buildings are of interest at the end of the 1960s or what ought to be remembered of the town’s past and why. The 1969 survey is also silent on the need for support to meet the future, which is quite natural since it is a survey. It is found that most of the stone houses are in good condition and only need minor repairs such as the replacement of leaking roofs and plastering. A major hurdle identified in the study for their preservation is the lack of incentives for the owners to repair and maintain their houses. Many of the occupants are tenants, not very interested in paying increased rent for the maintenance of the buildings.10 Another problem was the bleak economic prospects in the town at that time, which made the owners hesitate to carry out repairs since few tenants could pay increased rent. It was therefore suggested that the authorities should buy some of the buildings and convert them into public use (Tanzania: Bagamoyo Township Survey 1969: 228–29).11 It should be noted that at the time of the 1969 survey Bagamoyo had neither identified historical buildings for conservation, nor was Bagamoyo identified as a historic town. It was only in April 1972 that the Ministry of Education, to which the Department of Antiquities belonged, appointed an ad hoc commi ee with representatives from the ruling party, Chama cha Mapinduzi (CCM), government institutions and the University of Dar es Salaam to come up with recommendations regarding the conservation of parts of the town (Kamamba 1983: 187–88). The government’s appointment of this commi ee should be seen both in the light of Bagamoyo’s exclusion from the towns protected under the Antiquities Act of 1964 and in light of the need to make amendments to the same act (Kamamba 1983: 187). The commi ee worked a few months before it presented a number of proposals in August 1972, with the aim to develop the town and the area surrounding the town. It identified the buildings and structures of historic significance, as “the Fort, the Caravan road, the Block house, the Mwanamakuka cemetery, the old stone town (i.e. India street, Customs street, School and fruit market street) and the Mission” (Kamamba 1983: 188). Bagamoyo Stone Town, or Dunda, is the area between India Street, Mangesani road and Uhuru Road (see Map 3.1).

Bagamoyo as a Potential National and Regional Center for Conservation and Heritage-Making The shi in perspective from the colonial focus on specific objects such as doors and frames, via early post-independence emphasis on interesting and historic buildings, to the 1970s efforts to preserve the historic town, was

26

•

Muted Memories

further accentuated through the government’s request in 1979 to UNESCO to identify and estimate the need for the conservation of Bagamoyo.12 The marriage between heritage and economic growth through tourism, germinating from the 1969 survey, was strengthened by the UNESCO report. The resources of Bagamoyo include, according to the UNESCO consultant T. N. Watson: “the beautiful sandy bay with its historical past, its easy way of life, traditional dancing and singing” (Watson 1982: 36). The easy way of life is not further explored or explained, but it probably has to do with the famous tranquility of the town and the o en commentedupon slow rhythm of life of its inhabitants. The economy of Bagamoyo, as presented in the 1969 survey, was still dormant. The first years of the 1980s have been called the “hamna time”; hamna in Kiswahili means “there is nothing” on the shelves in the shops. The national economy was in ruins a er the oil shock, the Uganda war, and a decade of strong central government interference in local life and production systems, such as the villagization (Ujamaa) policy during the 1970s. Against this background the UNESCO consultants saw tourism as the most promising sector in Bagamoyo. Watson (1982) writes: Bagamoyo is well located geographically, and it possesses an unusually fine architectural character; but although it is rich in history it has an insecure economic basis at present. The town is dependent on income from the peasant agriculture in the surrounding area, small traders, tidal fishing, and as a Government administrative centre. Despite its tourist a ractions, there is no appropriate overnight accommodation. (33–34)

However, the report envisages a bright future for the town with tourism as the economic backbone of the area. It is assumed that the town will a ract both national and international cultural visitors. In particular, it is mentioned that the growing population in the capital Dar es Salaam will need a recreational resort and Bagamoyo’s proximity to the international Airport of Dar es Salaam meant that “the town could provide [international tourists] a resting-point in beautiful surroundings before se ing off on the coastal circuits and visits to the game parks” (Watson 1982: 37). Tourists are expected to want to part with their money for the charms of the Indian Ocean, rather than the presence of historical and cultural values. The UNESCO report implied, however, a paradigm shi since the focus in future would be not only on weekend visitors from Dar es Salaam but also on visitors from abroad. One of the results from the report was that the proposed conservation areas were eventually surveyed and mapped and that, in 1983, the Ministry of Education declared the protection of these areas under the amended Antiquities Act of 1964 (Kamamba 2002: 2). The UNESCO consultants also suggested that the old German Boma13

Bagamoyo

•

27

be converted into a hotel in line with international standards and that “its architectural and historical quality could be preserved while providing accommodation for 172 guests” (Kamamba 2002: 37). The suggestions at that time on how to use the heritage followed utilitarian lines of thought: to a ract and host tourists and government officers. The people who actually lived in the stone houses were invisible.

Bagamoyo as an East African Center for Stone Town Conservation Practices and Strategies Heritage-making in Bagamoyo during the 1980s saw one new development and the pronunciation of a theme that had already emerged in the late 1960s. The new development was the identification of new partners who could be mobilized for the sake of conservation. The new paradigm also contained a further emphasis on the potentials of the tourist sector to raise resources for the development of Bagamoyo. The input from the UNESCO consultants also resulted in a recommendation to arrange a seminar in Bagamoyo and develop the town into a center for Stone Town conservation skills and cra s. The 1983 Bagamoyo Seminar, with delegates from Somalia, Kenya, Tanzania, and Mozambique, drew participants primarily from government or national institutions “directly responsible for the protection, preservation and presentation of cultural heritage” (UNESCO 1983: 3). It is therefore no surprise that the predominantly technical approach toward conservation adopted at the seminar stimulated the identification of mainly technical and financial problems in the work to conserve and protect East African stone houses. One of the identified problems was the difficulty of finding people with skills in stone house construction and maintenance; another was a lack of money (UNESCO 1983: 4). This process of identification of both problems and their solutions, according to Tanya Murray Li (2007), may be called “rendering technical.” The concept is borrowed from James Ferguson (1994) and stands both for the process whereby available solutions influence the definition of the problem and for the redefinition of political and socioeconomic issues into technical problems. Since predominantly technical solutions are o en closest at hand for conservation experts, the same experts identify predominantly technical problems. One of the problems identified during the 1983 Bagamoyo Seminar was, as mentioned, the lack of technical competence in the art of stone house conservation. To solve this problem one of the UNESCO consultants suggested the establishment of a Center for the Conservation and Restoration of Cultural Property at Bagamoyo. Bagamoyo was to become

28

•

Muted Memories

the leading center for training in conservation and restoration along the East African coast with the “Old Arab Fort” as its main building (Russell 1980). It is further specified that the proposed center should be directed by an international expert who should work at the center for five years while an expert from Tanzania should receive the necessary training to take over as the director of the center (Russell 1980: 19).14 The narrow focus on different technologies for conservation, which dominated during the 1983 Seminar, harmonized poorly with the Tanzanian focus on planned socioeconomic development in Bagamoyo. This may have inspired the seminar coordinator Amani A. Mturi to suggest the arrangement of a follow-up seminar, which “should draw its participants not only from the institutions responsible for the conservation of the cultural heritage but also from institutions and professions involved in rural development including architects, engineers, regional planners, sociologists etc.” (Mturi 1983a: 9). What is noticeable here is the inclusion or invitation of regional planners and sociologists in the conservation of Bagamoyo’s heritage. This opening of perspective depended partly on global development of a new concept of conservation during the la er part of the 1970s. One aspect of this new concept was the extension of the object: “conservation of historic towns should be geared towards the totality of the built or se led environment rather than individual features and structures of such an environment” (Mturi 1983c: 140). This paradigm shi resulted in new amendments to the 1964 Antiquities Act, not only to focus on specific objects, buildings or monuments, but also to embrace the conservation areas or the built and se led environment (Kamamba 1983: 186–87).15 The inclusion of the living space of local people called for a paralleled extension of the people at local, national, and global levels supposed to have an interest in and responsibility for taking care of the object, as seen in Mturi’s suggestion to arrange a follow-up seminar. We see in the quotation above that the group of new subjects also included an expert in development planning. This invitation could be extended almost infinitum, as explained by Bernard Feilden, an ICROM heritage expert: [T]he conservation of cultural property is multidisciplinary and inter-professional in practice. Many people may be involved in a single project: administrators; archivists; anthropologists; antiquarians; archaeologists; architects; art historians; biologists, building surveyors; chemists; conservators; cra smen; curators; ecologists; entomologists; ethnologists; engineers of many specializations; economic historians; geographers; geologists; hydrologists; landscape architects; legislators; mineralogists; museologists; petrologists; property managers; surveyors; seismologists; sociologists; town planners; valuers; and one could add volcanologists. (Feilden 1979: 3)

Bagamoyo

•

29

Feilden hastens to add; “The list is alphabetic and incomplete” (Feilden 1979: 3). Yet, the people who actually live in and make use of the built and se led environment has still to await its invitation to the process. The extension of the object and the invitation of new co-subjects were not without its problems as the former subjects, the heritage conservationists, could find their own professional interests threatened. The invitation of the new experts and authorities, on the other hand, was crucial since, at least in a country like Tanzania, with its long tradition of top-down planning (see Havnevik 1994), it is necessary to have policy makers and planners of socioeconomic programs on board in order for the new conservation strategy to succeed. Or, as explained by a leading Tanzanian expert on heritage-making: “a itudes and perceptions of planners both physical, regional and economic as well as policy makers will be crucial in determining the acceptance of the conservation policies and programmes” (Mturi 1983c: 152). A further specification of the aim with this new cooperation is revealed in a third contribution by Amini Mturi (1983b: 121–22) to the 1983 seminar when he pointed out that planning and management of socioeconomic development programs may undermine or even destroy Tanzania’s cultural heritage. We may here sense an awareness of a potential conflict between the development of Bagamoyo and its conservation.16 Hence, a cadre of planners, politicians, and policy makers commi ed to conservation had to be created through training and sensitization, since “conservation must be accepted as a positive factor in development by both policy makers and the planners of the social and economic development programmes” (Mturi 1983c: 152). Although the need for collaboration between conservationists, development planners and politicians is justified due to the need for successful planning of the conservation area, another motive for such collaboration surfaced during the workshop. As noted above, the lack of money was identified as one of the problems in the conservation of the stone houses and towns. The creation of a conservation-conscious cadre of policy makers and planners could open up new doors, which could lead to the “harnessing of funds from other sectors such as housing, education, culture tourism etc. for the financing of conservation programmes” (Mturi 1983c: 160–61). Sensitized to the need for and value of conservation, government officials in other sectors would, hopefully, see the need to use parts of their budgets to finance conservation activities in Bagamoyo.17 The merging of the cultural heritage with the development of Bagamoyo through an increased tourism sector was in line with a new development paradigm, which gained ground in Tanzania in the mid-1980s. The socialist experiment, self-reliance policy, and a empts to transform

30

•

Muted Memories

the rural sector through state-imposed villagization gave way to structural adjustments imposed by the World Bank/IMF, liberalization of the economy, and an increased dependence on foreign aid and, in the early 1990s also a liberalization of the political system with the introduction of a multi-party system. The local population, however, remained invisible and mute in this new concept or philosophy of conservation. Although other professionals were invited to cooperate, conservation was still a task for professionals, planners, and politicians. There was, however, a genuine conviction that the cultural heritage could be used for the a raction of global tourists in order to promote economic growth, although the main results of the 1983 seminar were restricted to the demarcation and protection of the conservation area in Bagamoyo and the training of a number of conservationists from the East African coast. Very li le came out in terms of the protection of the cultural heritage or the concrete conservation activities in Bagamoyo’s Stone Town, and very li le was resolved in terms of what this heritage actually represented or what memory should be inscribed on Bagamoyo’s Stone Town.

Heritage-Making and Poverty Reduction The end of the last millennium saw the development of new approaches and goals in multilateral and bilateral development cooperation. The focus was shi ed from development, o en understood as processes of modernization and economic growth, to poverty reduction. Development cooperation should from now on be demand-driven rather than supply-driven and be organized and implemented through partnership agreements and arrangements, as captured in the new buzzword partnership, which meant that the recipient country should own its own development process, as it was explained in development cooperation policy documents and agreements.

A Poverty Profile of Bagamoyo Another change during the second half of the 1990s in international development cooperation was, as mentioned, a more pronounced focus on poverty reduction. People from all walks of life agreed that poverty at the time of the millennium shi was widespread in Bagamoyo and in the surrounding villages. It was also pointed out that poverty was a major reason, together with uncertain forms of tenure, for the lack of maintenance or investments in the buildings in Stone Town. The characteristic

Bagamoyo

•

31

components of the perceptions of poverty include, according to Kombe (2002: 4): • lack of employment opportunities, especially for young people; • poor housing conditions and absence of basic services for most of the population; • “misery and helplessness,” especially among young people; • most households lacking the means to meet the daily living costs; and • most families finding it hard to meet the costs associated with education of their children. The lack of job opportunities was particularly severe among young women and men who were seen waiting for hours in the markets, at bus stations, or on the streets, or working as street vendors among the visitors. The specific atmosphere in Bagamoyo, described with words like tranquility, slow rhythm of life, or easy way of life (see the 1969 Survey and the 1982 UNESCO report referred to above), has been interpreted as the town’s genius loci (see Syversen 2007) or the “spirit of the place” (ICOMOS 2008b). In reality, it should rather be seen as an effect of poverty. Many migrate to Dar es Salaam in search for a be er life, while many others who remain in Bagamoyo turn to alcohol, drugs, and prostitution. The focus on poverty reduction threw its shadow over the discussions during the International Conference arranged in Bagamoyo in 2002 to investigate and evaluate the town’s potential for induction to UNESCO’s World Heritage List (see the next chapter). No wonder that the suggestions during the conference were centered, as they had been since the late 1960s, on how to exchange Bagamoyo’s unique cultural values for economic growth and social development. The agrarian character of Bagamoyo’s economy reveals itself in many ways.18 More than three quarters of the population in Bagamoyo town had their own farms for subsistence cultivation at the time of the millennium shi , but very li le was produced for market. It is sometimes suggested, as mentioned above, that cultivation beyond subsistence is despised in Bagamoyo since it is associated with the slave labor of the past. Another reason for the weak agricultural sector, given in a study carried out by Sam Maghimbi (1999), is that suitable farm land is too far away, and that most farmers still rely on the hand hoe (Kombe 2002: 2). This has forced many town dwellers to migrate to surrounding villages during the farming season, only to return a er the harvest is reaped (Kombe 2002: 4). The construction of the Dar-Bagamoyo road, which was opened just before the year 2000, worsened the situation for Bagamoyo’s poor people.

32

•

Muted Memories

Land grabbing became a new phenomenon in the district. Kombe (2002: 12) reports regarding the land along the road that: most of the land was being bought by the affluent for speculative purposes and therefore not put into productive use. . . . Because most of the land close to the road has been bought by the affluent, i.e. by mainly people from Dar es Salaam, farmers have to increasingly travel long distances to their farms, leading to further decline of food and cash crop production.

Employment opportunities at the time of the 2002 Conference were rare in the township since there is no major industry in the entire district. The total work force in Bagamoyo town was estimated at about fi een thousand people, but only 1,700 were formally employed, of which 1,270 were employed by the District Council. Of the rest, approximately 300 were employed in the construction of new hotels, and another 50 by NGOs, religious organizations, and national utility companies in power and telecommunication (Maghimbi 1999: 116, in Kombe 2002: 14).

The Two Faces of Bagamoyo: Dunda/Stone Town and Magomeni Bagamoyo township consists of two wards called Dunda and Magomeni; the administrative parts are under a division that is itself under a district. They correspond rather well with the two parts usually called Old Town or Dunda and the New Town or Magomeni. Magomeni ward constitutes the main part of Bagamoyo town, and in 1991 it covered an area of 14.75 square kilometers. Dunda ward contains approximately 2 square kilometers. The urban character of Dunda ward is obvious to any visitor: the area typically consists of two-story stone structures along narrow streets. The rural character of Magomeni is equally obvious: houses of the Swahili type made of wa le-and-daub walls and corrugated iron-sheet roofs intermingled with small gardens and unplanned winding footpaths. Dunda forms a triangle in Bagamoyo with India Street running along the Indian Ocean as the base with Uhuru Road and Mangesani Road as the other two legs. A survey carried out in preparation for the 2002 International Conference, “Bagamoyo’s Old Town Today: A Survey of Buildings and Their Inhabitants in Dunda 2002” (Knippel 2002) reported that a large number of houses in Dunda were temporarily or permanently closed. Only 54 buildings, compared to approximately 200 a century before, were inhabited around the year 2000 to house a total of 372 people. Almost half of these buildings (27) were occupied by the owners or their families, while roughly the same number of buildings (23) was inhabited by tenants. This situation affects the maintenance of the buildings, as this is generally the house owner’s responsibility. The uncertain ownership, which

Bagamoyo

•

33

surrounds many buildings in Stone Town, results in the absence or inadequate maintenance of many houses. However, the main reason given for the lack of repair is not that other people own the house, but the lack of money. The same survey showed that half of the families had an income in cash per month of less than 20,000 TSh per month, or US$ 0.6 per day, which was far below the US$ 1.00 per day accepted as the crude poverty line at that time. The vulnerability of the poor was also shown in their sources of income, which consisted of informal pe y trade and cultivation/fishing. The survey also showed that a quarter of the population had an income of between 20,000 and 40,000 TSh per month and another quarter earned more than 40,000 TSh per month. It has been estimated that 60 to 80 percent of the urban population in Tanzania at that time lived in what has been called informal se lements. Bagamoyo is no exception. Of the houses in Magomeni ward, 85 percent are built on unplanned land (Nilsson 2001, referring to UCLAS [University College of Lands and Architectural Studies] 1990–2000). The first informal se lements in Magomeni were established between 1965 and 1970 along the Mtoni Road. One reason for the rapid growth of Magomeni during the last decades was the declaration of the other major part of Bagamoyo, the Dunda area, as a Conservation Area, with restrictions and prohibitions on house construction. At the turn of the twentieth century, during the planning of the 2002 International Conference on Bagamoyo as a World Heritage only 140 new houses had been built on surveyed plots. It may be assumed that the informal se lements in the ward contributed to the rapid increase in population density since the national rule of 3 meters between houses was not followed in unplanned se lements. The density was therefore 51 houses per hectare in Magomeni, which was almost twice the number of houses (26) per hectare in the Old Town (Nilsson 2001: 14–18).

Integrated Development and Conservation: The SUDP The recognition of the widespread and deep poverty in Bagamoyo town and district and the changes in international and national development policies, priorities, and practices initiated the next chapter, beginning at the end of the 1990s, of Bagamoyo as a lens through which development planning and cooperation, as well as heritage-making can be viewed. At the time of the conference, national and local development planning and international development cooperation had developed into a new approach that built on “partnership,” integrated district development sup-

34

•

Muted Memories

port, and joint donor support, as well as an integrated approach for heritage protection and preservation. The emphasis on integrated approaches followed the development in global conservation policies and practices. The “European Charter of the Architectural Heritage” of 1975 (also called the Amsterdam Declaration) introduced integrated conservation. As the concept indicates, the cultural heritage was now seen as an integral part of urban development planning since “considering that the future of the architectural heritage depends largely upon its integration into the concept of people’s lives and upon the weight given to it in regional and town planning and development schemes” (ICOMOS 1975: 1). This approach was further developed in other texts. One is The Habitat Agenda adopted in 1996 at the Habitat II Conference in Istanbul. It specifies that member states have to address the following issues in order to improve the quality of life within human se lements: unsustainable population changes, including changes in structure and distribution, giving priority consideration to the tendency towards excessive population concentration; homelessness; increasing poverty; unemployment; social exclusion; family instability; inadequate resources; lack of basic infrastructure and services; lack of adequate planning, growing insecurity and violence; environmental degradation; as increased vulnerability to diseases. (The Habitat Agenda 1996: 2)

Another text is the Action Plan on Cultural Policies for Development, adopted in 1998 at a conference in Stockholm. The first objective identified and adopted through the integrative approach applied at the conference is: “To make cultural policy one of the key components of development strategy” (UNESCO 1998: 3). Member parties are urged to make cultural policies an integral part of “development policies, in particular as regards their interaction with social and economic policies” (UNESCO 1998: 3). Urban development planners developed their own versions of integrated development approaches in the form of the Strategic Urban Development Plan (SUDP). The SUDP had been developed during the 1990s as a tool for environmental planning and management by the UNEP/ UNCHS (United Nations Environmental Programme and United Nations Centre for Human Se lement). In the effort to reduce the poverty in Bagamoyo, planners, researchers, and politicians, once again cast their eyes on the town’s cultural heritage and the options to exchange the heritage for economic growth. The Bagamoyo District Council, the local authority for Bagamoyo town, decided in 2000 to broaden its approach from the conservation of specific structures and milieus, a result of the 1979/1983 UNESCO intervention, to the prepa-

Bagamoyo

•

35

ration and implementation of an integrated development and conservation plan: the Bagamoyo SUDP. The modus operandi for the SUDP process was to establish a partnership between the key stakeholders in Bagamoyo town,19 as actors in public, private, and civil society.20 The SUDP planning process built on participatory procedures to identify suitable arenas and activities for demand-driven intervention, which made it different from more conventional development planning, where blue-print plans were assumed to organize the change in a linear process—project planning, implementation, management. There is a strong belief in causality in development planning. The SUDP process, by contrast, employed a more issue-based and ad hoc solution approach (Karlsson and Tenga 2002: 1).21 We saw above how, during the early 1980s, conservationists tried to invite other government officers to join in conservation efforts. At the end of the 1990s the roles were reversed; the conservation efforts were integrated into broader programs and projects for development. This became particularly prominent when the integrated approach of conservation aimed at urban sustainable development while at the same time tried to reduce poverty. This is seen in the overall objective of Bagamoyo’s SUDP, which is “to maintain Bagamoyo as a Cultural Heritage Centre and as a pivot for pro-poor growth and development that benefits the citizens” (Karlsson and Tenga 2002: 4). The Bagamoyo SUDP was developed and processed at the time of the declaration of the Millennium Goals. The first among these goals is to combat poverty and eradicate famine. This list of goals was the result of a decade of rethinking the principles and practices in development cooperation, which in international cooperation should move from top-down planning to more democratic, demand-driven, and participatory procedures. Another novelty in development cooperation was the emphasis on partnership agreements and arrangements. Development cooperation should be demand-driven and meet the recipient’s actual needs rather than be supply-driven, i.e. fulfills the needs, wishes, and capacity of the donor. This was cached in two of the most popular buzzwords around the year 2000: the recipient should be in “the driver’s seat” and “hold the steering wheel.” In practice, this resulted in the establishment of sector-specific basket funds where the donors put their money at the disposal of the recipient country or into direct budget support where the money was channeled to the national coffers. This coincided with a decentralization reform in Tanzania, which made it possible for local governments, the District Councils, to enter into development cooperation agreements with international partners.22 These developments resulted in cooperation between the Swedish International

36

•

Muted Memories

Development Cooperation Authority (Sida)/Swedish Embassy as a main donor and the district authorities in Bagamoyo. A few years earlier Sweden had assisted the authorities on Zanzibar to have Zanzibar Stone Town inscribed on the World Heritage List. It therefore seemed appropriate to enter into cooperation with a heritage site on mainland Tanzania for the same purpose. The partnership with Bagamoyo has been expressed in the following way: “Sida is an active supporter of including Bagamoyo and the Trade Route in the World Heritage List [of UNESCO]. Support is also aiming at involving the local people in the burgeoning tourist industry and at the development of participatory approaches to planning and restoration projects in the town” (Sida 2004: 28). A number of consultative meetings were organized in Bagamoyo. The first, arranged in March 2000, a racted seventy participants of whom approximately ten were representatives from government, private, and civil sectors. The rest were Bagamoyo citizens. It was agreed during the meeting that Bagamoyo should engage itself in the SUDP process (Sida 2000) and a Steering Commi ee of eleven members was elected to further work on five critical issues. The first was to conserve and develop Bagamoyo Stone Town. Others, linked to poverty reduction and socioeconomic development were to utilize Bagamoyo’s coastal resources to a ract an increased number of tourists, which in turn was expected to strengthen the economic base of the town and district. A strengthened economy would in turn help to improve utilities and services to the town residents as improved sanitation and environmental quality. These efforts would together stimulate sustainable development of Bagamoyo. Through this, conservation activities were organized within a framework, which also encompassed economic growth and poverty reduction, improved social services, and urban development. To further broaden the participation, a second meeting was arranged in Bagamoyo in February 2001 to establish five “Issue-Specific Working Groups” to work with the five critical issues identified at the March 2000 meeting. Each group had approximately fi een members from all key sectors, totaling about sixty participants in the five groups. A third and final consultative meeting in Bagamoyo was originally planned to take place in April 2001. For reasons explained below, this meeting was postponed until the end of the year.23 Participatory processes may have had their limits and flaws. In October 2000, the steering commi ee changed its name to the SUDPF (Strategic Urban Development Plan Framework) Advisory Team, only to get a new name in October 2001—now called the SUDPF Implementation Team.24 This was not an innocent name change. The transformation from a steering commi ee during the planning process, via an advisory role to

Bagamoyo

•

37

becoming the implementing body meant contradictory roles among the members. This opened up the situation to potential conflicts. The members of the Implementation Team came from a broad spectrum of Bagamoyo denizens. The District Commissioner plus two members of the Bagamoyo District Council were members of the team. Others represented the Department of Antiquities, people from UCLAS, representatives from the Bagamoyo Investors’ Association, Bagamoyo District Business Association, the Bagamoyo Live Arts and Cra s Centre, and the Bagamoyo Young Artists (Karlsson and Tenga 2002: 8). However, the team did not include the Member of Parliament for Bagamoyo, or some powerful top officials in the local government.25 Local government authorities did not take long to respond to their marginalization through a empts to regain control over the process. In late 2000, the District Executive Director, representing the local authority, informed the members of the steering commi ee that the commi ee had ceased to exist since its purpose, as he understood it, was fulfilled. The role of the commi ee, he maintained, was to secure external support, which had been achieved in October 2000, when an agreement on the development of the SUDP was signed with the Swedish embassy and Sida. But more turbulence was on the way. A meeting was arranged in April 2001 to finalize the planning process through the adoption of the SUDP by the responsible authorities. The guest of honor at the meeting, the new District Commissioner, who had been appointed in late 2000, opened her speech by saying that she had received two different versions regarding the planning process. They were so contradictory, she said, that she had to adjourn the meeting and call the two sides to come to her office to explain their issues. She could not let this meeting adopt a document that was legitimately questioned by one section of the community. It was completely silent in the large assembly hall.26 It later surfaced that the District Commissioner had met a coordinator for a network of NGOs in Bagamoyo the day before and, later the same day, a temporarily assigned consultant and had understood that there were a number of issues that had to be clarified before she could recommend the adoption of the SUDP plan. The chairperson for the local NGO network was very much in favor of the participatory process through which civic society had been able to take an active part in the consultative meetings. Instead the conflict boiled down to the UCLAS seconded consultant’s attempt to put himself in charge of the implementation of the SUDP.27 Swedish support to Bagamoyo, guided by the integrated approach for development and conservation, was used, for example, to finance emergency repair of some of the historical buildings including the Arab Fort, Nassir Virji house and Caravanserai, and for the 2002 Bagamoyo Interna-

38

•

Muted Memories

tional World Heritage Conference (see next chapter). Another activity was to undertake an inventory trip in 2003 from Bagamoyo to Ujiji to document the East African Slave and Ivory Trade Route. Swedish support was also used for infrastructural improvements in the town. The prolonged process to develop and adopt Bagamoyo’s new development plan, which included the protection and conservation of some of the historical buildings in the old town, turned into a challenge of the concept of integrated development and conservation and of the potential to exchange cultural heritage for socioeconomic development. An agreement was therefore negotiated between the district council and Swedish International Development Cooperation Authority (Sida) to finance some emergency measures to show that the SUDP concept and process could bring development to Bagamoyo. The SUDP eventually imploded, but when looking back we see that it in fact resulted in a new bus station and a new market in Bagamoyo, the continued conservation of some of the historical buildings, and the arrangement of the 2002 International Conference in Bagamoyo as a World Heritage site. This conference will be the focus of the next chapter.

Notes  1. See, for example, W. T. Brown 1970; W. T. Brown 1971a; Kirknaes and Wembah-Rashid 1993; Kajeri and Henschel n.d.  2. A third interpretation will be explored in chapter 8.  3. During the 1960s and 1970s, Bagamoyo housed training camps and schools for freedom fighters from Mozambique and South Africa.  4. There is a message wri en by a visitor in the Bagamoyo Catholic Mission’s Guest Book: “Proudly being a Tanzanian. May God bless us not to remember the evils of the past” (23 February 2013, author’s field notes).  5. Personal communication with Jan Turtinen, 27 January 2014.  6. Russell reports that more than one thousand carved doors have been identified on mainland Tanzania: 729 in Dar es Salaam, 307 in Bagamoyo, 49 in Pangani, 42 in Mikindani, 30 in Kilwa Kivinye, 21 in Tanga, and 6 in Kilwa Kisiwani (Russell 1980: 12).  7. Athman reports that the oldest style of carving along the coast is called Bajuni and dates back to the eleventh century (Athman 1996: 13). Its name derives from a chip carving technique used to carve out circles and straight lines on the wood.  8. The findings are to a large extent based on previous surveys made by Rune Karlsson, one of the team members.  9. The doors mentioned in the text are probably those gaze ed by the British colonialists during the 1950s. The mixtures of architectural styles is also noticed by the World Monuments Fund (n.d.) when reporting that the Swahili

Bagamoyo

10.

11. 12.

13.

14.

15.

16.

•

39

architecture was supplemented at the end of the nineteenth century by the German architectural style and “this confluence makes Bagamoyo’s heritage style absolutely unique.” This is a consequence of the history of the town. Many of the traders le the town when Bagamoyo collapsed just before World War I as a consequence of the Germans’ decision to establish Dar es Salaam as the colonial capital and the construction of the Dar-Kigoma/Mwanza railways. Many traders who moved out still owned the houses, which were filled by tenants. This was the time when the state should bring both development and meaning to its citizens. Parts of the findings and recommendations are found in T. N. Watson’s report “Areas of Concern: Bagamoyo, Tanzania” (1982). Other parts are found in John Russell’s report (1980) “Proposed Centre for the Conservation and Restoration of Cultural Property at Bagamoyo.” The building is a landmark in Bagamoyo, but also an “itching memory” since it represents the memory of the German colonization of the area. Its landmark status is evident from the fact that a photo of the German Boma was on the front page of the pamphlet produced for the 2002 International Conference in Bagamoyo. The “itching memory” became obvious when the author, during a seminar on Bagamoyo as a potential World Heritage candidate, discussed the needs for potential German assistance in the conservation of the Boma with a German development officer. The discussion touched upon the ongoing rehabilitation of the old German Hospital in Dar es Salaam, which was carried out with German assistance. The message was that it was appropriate to assist in the conservation of the hospital since this building represented humanitarian concerns. It was more difficult, not to say impossible, to finance the conservation of the Boma since this building represented German colonial history (author’s field notes, March 2000). But the identified needs for external support was even greater: “Four more international experts should be provided for twelve man-months each year, to provide specialist tuition in various aspects of conserving and restoring cultural property. Whenever possible, local teaching staff should be employed in support of these experts.” (Russell 1980: 19). The fact that Bagamoyo was excluded from the Swahili towns protected by the 1964 Amended Antiquities Act and that the heritage consisted of a lived-in se lement became instrumental in the new amendments to the Act, as explained by Kamamba (1983: 186–87), “when policies were being evolved to cater for the protection of urban complexes like Bagamoyo with conservation interpreted to encompass both the protection and preservation of the heritage and its planned and controlled changes, the provisions of the Act were found to be inadequate. Consequently one of the main purposes of amending the 1964 Act was to introduce the concept of the protection and preservation of the totality of the built or se led environment.” It was revived when it was announced in March 2013 that a new port, the largest in Africa outside South Africa, should be constructed at Bagamoyo. For further details see the epilogue in this book.

40

•

Muted Memories

17. It should be noted that Mturi (1983c: 156) also saw many negative effects of tourism in historical towns since tourism is a user industry. On the other hand, tourism may, with appropriate fiscal policies and measures, contribute to increased revenue for the financing of conservation works (156). 18. As captured by Nilsson (2001: 17) regarding Magomeni, one of the two main parts of Bagamoyo, “When you approach Magomeni from the Stone Town you are immediately struck by the feeling of being in the countryside.” 19. It was recommended already at the preparatory workshop in Bagamoyo in the year 2000 that the SUDP “. . . should be developed through the stakeholders at all levels so that the final proposal incorporates their views, desires and aspirations” (Mafuru 2002, 4). 20. A consortium of District officials and planners and researchers approached Sida and the Swedish embassy for financial support for this process. 21. In their appraisal of the SUDP proposal Karlsson and Tenga (2002) conclude: “The normal stages of project implementation with its planning, monitoring and management phases are not clear in the case of the Bagamoyo project. It is in fact a very complicated situation with the involvement of different academic cultures, too many different stakeholders, unclear project ownership as well as a very academic ‘process approach’ to a very practical problem—the development and conservation of the Coastal and Rural Town of Bagamoyo” (2002: 2, my emphasis), a comment which reveals a different approach and the absence of an understanding of the integrated approach for development and conservation. 22. This meant bypassing the Ministry of Finance, which, so far, had held a monopolistic position to enter into such agreements. 23. A fourth Stakeholder Meeting was arranged in February 2002 in Dar es Salaam for the formal launch of the SUDP. The Framework was by then adopted by the District and Regional authorities, but not yet by the ministries concerned, i.e. the Ministry of Lands, the President’s Office and its Ministry of Regional Administration and Local Government, and the Ministry of Natural Resources and Tourism. 24. This renaming probably occurred for tactical reasons as some of the members of the team aspired to control the implementation phase. This, however, complicated the process since, as pointed out by K. Karlsson and S. Tenga (2002: 9) the term is misleading since the team has not been assigned any implementing functions. 25. The unconventional SUDP approach to build on participatory processes and practices sharply contrasted with the more conventional way in Tanzania to implement development projects through government bodies as the local authority or that a new implementation body is established within that authority. By contrast, the SUDPF process led to the establishment of an SUDP Implementing Team outside the framework of the local authority or District Council. Although the Council was the formal or legal owner of the SUDP, it was not recognized as a main actor. 26. Author’s field notes, April 2001. 27. Author’s field notes, April 2001.

2  Heritage-Making The 2002 International Conference Conferences are convenient workshops for heritage-making. This chapter will focus on the International Conference arranged in Bagamoyo in September 2002 with the aim of evaluating the town’s options to be inscribed on the World Heritage List. At the end of the last millennium, a legitimate question arose among a group of academics and heritage practitioners in Dar es Salaam concerning the possibility of arranging an international conference to explore the question of what to nominate for the UNESCO List. An object proposed for the List must carry or represent outstanding universal value in order to be of global interest. This means that the nomination dossier presented to UNESCO by Tanzania would have to prove that the proposed heritage is of an outstanding universal value (OUV).

The 2002 International Conference The operation of the List and UNESCO’s rules, procedures, and guidelines for nomination and inscription, provided much of the format and the framework for the conference. The three days that the participants spent in this small coastal town were structured around a series of plenary presentations, panel sessions, and workshops.1 The conference gathered international, regional, national, and local experts on history, archaeology, and anthropology, national and international experts on heritage conser-

42

•

Muted Memories

vation and presentation and museology as well as practitioners from local and national levels.2 The themes discussed and negotiated were selected to cover the various issues to be included in the nomination dossier needed for inscription on the list. Thirty-four papers were presented during the conference. They were divided into the following themes; history and culture (7 papers); museum and conservation (10 papers); rights, planning, and institutional set-up (4 papers); World Heritage sites (9 papers); socioeconomic development (3 papers); and the environmental se ing (3 papers).3 Most of the papers dealt with rather technical issues regarding conservation and protection of the heritage. Only one explicitly addressed the theme that came to dominate the conference: Bagamoyo as a slave port. Bertram Mapunda’s (2002) paper,4 presented as the third contribution during the first plenary presentation, had the title “East African Entrepots in a Period of Transition: Bagamoyo and the Slave and the Ivory Trade.” Another paper, “A Comparative Study between Bagamoyo and Mombasa” (Achola 2002), implicitly touched on the slave trade in the two coastal towns. Not a single paper had an explicit focus on the ivory trade, although this cultural phenomenon was o en mentioned during panel discussions. The absence of such a presentation should be contrasted to the six papers on socioeconomic and environmental conditions in Bagamoyo town and district.5 This imbalance must be understood from the approach that consciously or unconsciously guided the discussions before and during the conference: the nomination of Bagamoyo to the World Heritage List could be a means to reduce poverty.

Background to the Conference The preparation phase before the 2002 International Conference turned out to be a potent brew; both bi er and sweet and with stormy and calm moments. It is hard to identify a specific event or issue that triggered the process. Rather, it grew out of sporadic encounters, random meetings, and informal enquiries about why Bagamoyo was not on the World Heritage List, what had happened to previous efforts during the last fi y years to protect and conserve this historic milieu (Kamamba 2002). In short—what was the problem? Government staff responsible for the national antiquities and the country’s tourism sector; researchers on archaeology, history, anthropology, planned development, and architecture; and a representative from the donor community met irregularly and informally in Dar es Salaam during the years around the turn of the twenty-first century. The purpose of the meetings was to form a joint opinion on the appropriateness of arranging

Heritage-Making •

43

an international conference in Bagamoyo to explore the potential to nominate the town for the list.6 At the forefront of these informal meetings was the recognition of the urgency to protect and preserve Bagamoyo, captured by one of the participants in a preparatory workshop before the International Conference was arranged: The architectural heritage of Old Bagamoyo has decayed and deteriorated. In general, the larger part of the heritage both privately and publicly owned is in a state of neglect, disrepair and underutilization. This process has been accelerated during the 1990s resulting in the abandonment and collapse of a number of buildings including the landmark building of the German Boma, which was the district headquarters, and residence of the District Commissioner including a State Lodge. If immediate and urgent measures are not taken to halt this process, the whole rationale for the conservation and development of Old Bagamoyo town will be gone, unless our aim is to protect and preserve the historic town (Magofu) [as a ruined town] like the nearby Kaole. (Mturi 2000: 13)

The informal meetings resulted in the organization of a preparatory “Workshop on Conservation and Development of Bagamoyo Township,” held in Bagamoyo in March 2000. The workshop was opened by the Minister for Natural Resources and Tourism and closed by the Minister for Foreign Affairs and International Cooperation, who a few years later became the fourth President of the United Republic of Tanzania. This was seen as evidence of the national interest in the issue. The participants came from various ministries and authorities in Dar es Salaam (approx. 20–25 participants), from local authorities, civil society, and the business community in Bagamoyo (approx. 20–25 participants), and from bilateral and multilateral development partners.7 Six researchers were invited to present their papers.8 The participants at the workshop adopted a recommendation that Tanzania should find ways to promote the inscription of Bagamoyo on the World Heritage List as a representative for “The East African Caravan and Slave Trade.” Bagamoyo’s pivotal role, not only in linking the continental and the Indian Ocean trade networks, but also its role during early colonial times was pointed out by one of the invited researchers. Bagamoyo became the first capital in German East Africa and here the first German Custom House was built in 1895 as well as the mainland’s first post office. Bagamoyo, thus, represents “the time of increased trade and communication with the outside world” (Ndagala 2000). Bagamoyo’s role and uniqueness regarding the East African caravan trade during an intensive early phase of globalization was also pointed out during the discussions. Emphasis was on the caravan and slave trade, as summarized in a report from the conference: “Bagamoyo still has an

44

•

Muted Memories

Illustration 2.1. Part of the German Boma at the time of the 2002 International Conference. Photo by the author.

Heritage-Making •

45

important role to play as the pivot in international and global contacts. Its place on the World Heritage List as a representative for the East African Caravan and Slave Trade will keep the memory alive and assist to assure that the dark parts of our common history are never repeated” (Sida 2000: 6). In a parallel process, the Ministry of Natural Resources and Tourism developed a plan to relocate the education of staff for the hospitality sector from Kurasini in Dar es Salaam to Bagamoyo. This effort was based on the recognition that most of the jobs in Tanzania’s expanding tourist sector were being filled by Kenyans. Tanzanians could not compete due to lower competence, which motivated the rearrangement of the existing training and education. This could be facilitated by a move to new premises in Bagamoyo, or rather by the move to an old building, the so-called Nassir Virji house, one of the old stone house buildings in Bagamoyo’s Stone Town. Such a move would serve two purposes: to turn one of the classical stone buildings in Bagamoyo’s Stone Town into a living building, and to increase the quality in the training of staff for the hospitality sector. It seemed natural for the members in the ad hoc group that world heritage and tourism were good fellows, particularly so since Bagamoyo was one of the socioeconomically poorest parts of Tanzania. It must therefore have been equally natural that an international conference should be arranged to identify and present what should be nominated to the UNESCO List and that such a nomination could be used for the development of the town. The nestor of the ad hoc group, a prominent member from the University of Dar es Salaam, was assumed to take the role as the group’s unofficial leader during the preparation process. Uncertainties grew when, a er several meetings, he did not show an interest in fulfilling this role. Out of respect for his position and authority, no other candidate stepped forward to lead the group and the process. The issue was partially solved when other members in the informal group approached the lead donor with the suggestion to engage, with donor money, staff from UCLAS (University College of Lands and Architectural Studies) to steer the process. However, this created resentment among some academics at the university. The UCLAS’s staff is sometimes looked down on for their alleged weaker academic tradition and more practical orientation. At least one university member in the original informal group was never seen again in the meetings. Heritage-making may, in academic circles too, cause friction and fractures. The process gained in both speed and strength with the UCLAS consultants as the main organizers. The group identified the themes to be covered in the upcoming International Conference and experts to be in-

46

•

Muted Memories

vited. An application for financial support was dra ed by the UCLAS team leader, which resulted in a formal request for financial support that was submi ed to the Swedish Embassy in Dar es Salaam. The application was approved and financial support was channeled through the Ministry of National Resources and Tourism first for the arrangement of the preparatory workshop mentioned above, and also for the 2002 International Conference arranged in Bagamoyo.

In Search of Outstanding Values World Heritage is not produced on a clean slate. UNESCO has orchestrated the work with the identification, presentation, and conservation of the Heritage since 1972 when the “Convention Concerning the Protection of the World Cultural and Natural Heritage” was adopted, and since 1978 when the World Heritage List was established through the inscription of the first sites. Since then, the List has been both an effect of globalization, as an increasing number of state parties nominate their heritage, o en in an effort to be part of the “global ecumene of modernity” (Hannerz 1991; 1992; 1996) and a cause of globalization as heritages are created on “underrepresented” continents in order to have a more balanced List, as will be further explored in the next chapter.9 It is not surprising that the identification and presentation of universal values nourishes its own kind of conflicts and confusions: what, actually, did Bagamoyo represent? A friction, or rather a fracture, opened itself up during the official opening of the 2002 International Conference. The question at stake concerned what was to be seen in Bagamoyo and from what perspective. The contended issue concerned the extent of external or foreign impact on the development of Swahili culture along the East African coast versus the impact of African cultures on African soil. One of the participants from the University of Dar es Salaam argued forcefully and convincingly against the tendency o en encountered in texts on Swahili towns and culture, to deny Africans their role in this process. He warned against the tendencies to hail Arabs, Persians, and later Europeans for their contributions to various kinds of development in African communities and cultures, at the expense of endemic African efforts and developments. The discussions during the conference on what to nominate gave rise to four suggested themes: Bagamoyo and the Slave and Ivory Trade Bagamoyo and the Development and Spread of Kiswahili

Heritage-Making •

47

Bagamoyo and the Spread of Islam and Christianity Bagamoyo as an Archaeological Site As can be seen from the formulations, two of the four themes came close to the view of Bagamoyo as an entrepôt of predominantly foreign elements on African soil, while the first theme may be interpreted more as a terminus or export port for African goods and people. The three first themes capture a rather one-directional flow of people, goods, and other cultural elements either from the coast to the interior of the continent or from the interior to the coast. These three themes will be discussed below as they were presented during the conference.10 It should be noted that this chapter focuses on what was said, discussed, and concluded during the conference. Another chapter in this book titled “Bagamoyo: The Entrepôt to the Heart of Africa,” will further develop the themes on Bagamoyo as a point of entry for Islam and Christianity and the role of the caravan trade in the spread not only of these religions but also in the spread of Kiswahili by also using more recent research findings. Other chapters in this book will focus on the caravan trade along the central routes as a cultural route, i.e. the flow of people, goods, and other cultural elements in both directions and the hybridization or creolization processes which took place in a trade port like Bagamoyo.

In Search of Answers to What and Why With respect to the competence and knowledge of the participants, it must still be recognized today that the understanding of the “what” and the “why” questions have matured during the nearly two decades since the conference was held. More elaborate and powerful answers to both questions will be put forward in the second part of this book. Although the question of the “what” and the “why” by themselves may cause both frictions and fractures, two other questions may potentially be even more conflict-generating: who shall be empowered to identify them and how shall this be done? It is no surprise that various actors, guided by their knowledge and interests and based on their positions in the power structure, proposed different suggestions as to what shall be seen as “of outstanding value” in Bagamoyo and why this heritage ought to be on the World Heritage List. This can be seen in the four themes suggested for further elaboration. Archaeologists, anthropologists, linguists, historians, experts on the history of religion all contributed to the suggestions on Bagamoyo’s outstanding values. The task of answering the question “how” the nomination should be done, however, was spared for the nomination commi ee elected at the Conference. The result of the work by

48

•

Muted Memories

this commi ee, assisted by national and international consultants, will be further explored in the next chapter.

Bagamoyo and the Development of the Caravan Trade The controversy reported above regarding the impact of external influences versus endemic developments came as no surprise since heritage-making implies the identification and presentation of what is meant to be an authentic case. The procedure to identify, interpret, and present the outstanding value implies, however, that only one leading story, one “truth,” and one image can be put forward and that the selection of one story results in the silencing of others. A multivocal story with variations on a common theme cannot easily be fi ed into UNESCO’s matrix. During the conference two competing suggestions on the outstanding value of Bagamoyo came forward, each one linked to a different interpretation of how the long-distance caravan trade developed during the late eighteenth and early nineteenth centuries. The truth in one version held that the trade along the central caravan routes was the result of initiatives taken by Arabic/Swahili traders and European/American travelers from the coast. This perspective is aptly captured in a standard work on African history, The Cambridge History of Africa, in the following words: “East Africa in 1870 is best defined as the economic hinterland of the commercial entrepôt of Zanzibar. This area included much of what now are Tanzania, Kenya and Uganda, as well as parts of Rwanda, Burundi, Zaïre, Zambia, Malawi, Mozambique and Somalia” (Wright 1976: 539). A similar view was expressed by James S. Kirkman (1964: 22) when he explained that the “historical monuments of East Africa belonged not to the Africans but to the Arabs and arabized Persians, mixed in blood with the African but in culture u erly apart from the Africans who surround them.” The competing view maintained that long-distance trade was an outstanding cultural phenomenon created by industrious entrepreneurs in the interior of Africa who set out to find new markets for their ivory and slaves, even in the coastal towns along the Indian Ocean. This perspective will influence the presentation in the second part of this volume on the East African caravan trade along the central routes. Either we look at the caravan trade from the port of Bagamoyo, looking westwards, and try to get an understanding of how it linked to the interior of Africa or we view the caravan trade from the interior of the continent looking eastward. In this perspective, which is presented in chapter 5, Bagamoyo becomes the interface between the African caravan trade and the Indian Ocean trading networks. At the conference, however, the entry point perspective came to dominate, with an emphasis on Bagamoyo

Heritage-Making •

49

as the gateway for external actors and influences into the interior of the continent. This perspective, however, was not accepted without resistance. Why is it so difficult to accept, argued the professor from the University of Dar es Salaam mentioned above, that upcountry entrepreneurs paved the way for the caravan routes, which, at a later stage, became dominated by Arab and Swahili coastal traders and European and American explorers? Similar views were expressed twenty years before the 2002 Conference when James de Vere Allen (1981c: 230) stated in his comments on Mtoro Bakari’s book Desturi za Waswahili that “it [the book] should modify . . . the views of those (o en modern East African) historians and sociologists who like to see Swahili culture and civilization as something fundamentally alien to the continent, non-African, perhaps even in some sense reprehensible,” or deny people on the continent the capacity to develop the caravan trade to meet their own interests. Or, to paraphrase Adam Smith, why is it so difficult to extend men’s “propensity to truck, barter and exchange” to the peoples in African up-country societies? (Smith [1776] 1974). The late 1960s saw the development of a school in the field of historical studies at the University College in Dar es Salaam that focused on research on the African impact on Africa’s development. An example is research done by Andrew Roberts on the development of the Nyamwezi trading networks, which will be further presented in chapter 5. Historical studies on the East African coast continued, however, to be relatively unaffected by the ambitions and perspectives of the new school, since much of the explanatory power of what took place here was ascribed to immigrants from the Asian continent. Outdated Eurocentric views o en claim that the interior of the continent was isolated from the rest of the world and even “discovered” by European explorers, many of whom came through Bagamoyo. Today there is a growing body of knowledge about the depth and strength of the contacts between societies in the interlacustrine area (the Great Lakes Region) and eastern Congo and the societies along the East African coast. These findings show extensive trading networks predating Portuguese interference with the Indian Ocean trade around the year 1500. It may be assumed that the intracontinental trade came to a temporal end with the intrusion of the Portuguese. The birth date of Bagamoyo is not known. A plausible guess is that the embryonic start was a fishing village, which traded in salt and dried or salted fish with neighboring up-country societies (W. T. Brown 1971a, 74). A decisive development occurred when these, probably locally restricted networks, were merged with the trading networks in the interior of the continent. The most well-known of these networks had been developed

50

•

Muted Memories

by the Nyamwezi people who lived east of Lake Tanganyika and south of Lake Victoria. The increasing global demand for ivory, an item in great demand at that time in India, China, Europe, and the Americas, probably triggered the Nyamwezi traders and porters to bring an increasing amount of elephant tusks to the coast. These first visits from up-country traders and porters to places like Bagamoyo probably occurred in the late eighteenth century. It may be hypothesized that two different but connected processes contributed to the development of the long-distance caravan trade in ivory and slaves, for which Bagamoyo has been proposed as a place of memory. The first development was, as mentioned above, the organization of caravans from the interior of the continent that appeared in trading towns along the coast with their merchandise, such as ivory, livestock, iron, and slaves. Traders in Bagamoyo responded quickly and established themselves as the main agents in this rapidly expanding trade between the two different trading networks. The second process was the increasing pressure from the Sultan of Zanzibar to control trade on mainland Africa. This eventually led to an armed conflict in 1844 at Kaole, the ancient but now deserted Swahili trading town approximately 6 km south of present day Bagamoyo. It seems probable that the traders in Kaole moved to establish the new town of Bagamoyo when the Sultan tightened his grip on the area by establishing a garrison with soldiers in Kaole as a means to strengthen his control over trade. The move to the new port at Bagamoyo also meant that the old merchant class in the coastal towns could engage themselves in the development of the caravan trade. Soon, however, the sultanate on Zanzibar was tightening its grip over commerce in this new port too. This interpretation of the origin of Bagamoyo combines the two perspectives of external influences versus endemic developments. The original impetus may have been coastal people trading with salt and dried salted fish with up-country neighbors. The next development was the traders and porters from the interior who utilized already existing trading networks both in the interior and in the coastal hinterlands to develop the long-distance caravan routes. A further development came when the coastal traders, including the Zanzibari Sultans, gradually took over much of the caravan trade along the central routes in what is present-day Tanzania.

The Cosmopolitan Character of Bagamoyo In the search for “the outstanding value” of Bagamoyo during the conference, many comments and presentations dwelled on the importance of the external influences for the development of the port town. A recurrent

Heritage-Making •

51

theme was the town’s importance as an interface between the Indian Oceanic trade and the caravan trade. An argument for its outstanding value was the town’s cosmopolitan history and character during the la er half of the nineteenth century. The cosmopolitan character of Old Bagamoyo was explained with reference both to its involvement in global trade, but also to its a raction of people from different continents. Karsten Legère, a participant in the 2002 International Conference, referred to the German District Commissioner for Bagamoyo at the turn of the nineteenth century who gave the following population composition for the year 1900: 14,500 “Bantus,” 800 Arabs (Shihiri), 400 Indians, and 250 Baluchi.11 Although the elite class of Bagamoyo at this point was dominated by Arabs,12 not least in political, religious, and social life, it was the Indians who dominated in the economic sphere and organized the import of trade goods from India, Europe, and the Americas, which gave the town much of its cosmopolitan character. The large proportion of Indians in Bagamoyo contributed to the vibrant economic life in the town. The Indians were both financiers of the many caravans that started in Bagamoyo and set off to the interior of the continent, as well as shop owners who provided the caravan leaders and porters with the goods they brought into the interior of the continent. Milcah Achola, participating in the conference, reported that the nineteenth century caravan trade and the encouragement of the Omani rulers resulted in the immigration and se lement of Indian financiers. Living in the towns, hardly ever traveling inland themselves, they provided the necessary credit to Arab and Swahili merchants who wished to undertake caravan trade into the interior. They have been identified as the greatest financial beneficiaries of the nineteenth century caravan trade (Achola 2002: 5). In reality, the category Indians was a highly heterogeneous group of Bagamoyo’s population. The largest Indian Muslim community in Bagamoyo was the Khoja Ismailis who began to se le permanently during the 1840s (W. T. Brown 1970: 77–78). Other Indian Muslims belonged to different canons within the Sunni faith. Another religious group from the Indian subcontinent that contributed to the cosmopolitan character of Bagamoyo was the Hindus. Other religious minorities in Bagamoyo were Catholic Goans and Sunni Memons. Sultan Seyyid bin Said of Zanzibar (Sultan between 1832–1856) actively encouraged Indian merchants and financiers to se le in East Africa in order to get access to their commercial skills and financial resources. Bagamoyo Conference participant Bertram Mapunda (2002: 3) stated that “Asian traders became the economic backbone of the entire Zanzibar system of financing virtually all the caravans, which le the coast for the interior on credits they alone could afford to give, and controlling the export-import trade.”

52

•

Muted Memories

The most prominent Asian financier and businessman in Bagamoyo during the second half of the nineteenth century was Sewa Haji Paroo, who belonged to the Khoja Ismaili canon. Mapunda (2002: 7) reported during the conference that he was described as the person who “held the key to the interior.” Jesper Kirkenaes and John Wembah-Rashid capture what many explorers, missionaries, and caravan leaders of that time experienced when they organized their up-country expeditions: “From Zanzibar—where travelers to the interior started before going to Bagamoyo—one was always referred to Sewa Haji. And from him one obtained the expensive but right portage and safe passage and guidance to the interior” (Kirknaes and Wembah-Rashid 1993: 10). The counterargument from those at the 2002 conference who emphasized the role of the African influence in this development pointed out that Bagamoyo would probably not have reached its prominent position as the most important caravan port without the knowledge and skills accumulated by African caravan leaders. The skills among these early entrepreneurs and explorers, both men and women, became instrumental for many coastal Arabs and Swahili traders, but also for the first European “explores” and colonizers. The caravan routes paved through the bush by the first Nyamwezi caravans became the highways for the ensuing caravans from the coast. These caravans of traders, “explorers,” colonizers, and missionaries followed the well-trodden paths established by up-country traders and porters engaged in the intra- and inter-continental trading networks. It was explained during the Bagamoyo conference that these early African caravan leaders and guides also became instrumental for the first coastal traders in their a empts to learn more about the rich up-country sources of ivory and slaves.13 A Zanzibari-born Nyamwezi named Lief bin Said reached Lake Tanganyika in 1831 (Koponen 1988: 76), only to be followed in 1841–42 by Muhammed bin Sahli and Suleiman bin Nassor (Grey 1957). One of the most spectacular travelers and traders from the coast was Said bin Habin el-Afis who set off from the coast around the year 1844 in an effort to learn more about the interior of the continent. During his sixteen-year expedition he visited Luanda on the Atlantic coast three times (Chami 2002a: 3). He achieved this in approximately thirty years or one generation before Henry Morton Stanley crossed the continent on his journey from Bagamoyo to the mouth of the Congo River and on to the Atlantic coast. The most well-known guides, still today, are probably Chuma and Susi, who together with other Africans during nine months in 1873–74 carried the remains of Dr. Livingstone from central Africa to Bagamoyo. Felix Chami, professor of archaeology, suggested at the 2002 International

Heritage-Making •

53

Conference that the text above the door on Bagamoyo’s Anglican church, which currently reads “Through this door Dr. Livingstone passed,”14 should be changed to read “here is where Chuma and Susi (and others) passed carrying the body of Dr. Livingstone to Zanzibar and England” (Chami 2002a: 5). The purpose of his interpretation is to give credit to those who took care of the remains of the dead European and made it possible to repatriate him to his country of origin. The suggested new text is not just playing with words, but illustrates the struggle over the memory and the heritage. It also reveals that a heritage site o en carries only one story or truth at the expense of alternative or challenging interpretations.

The Second Theme: Bagamoyo and the Spread of Kiswahili One of the answers tested during the conference on the question “what is Bagamoyo’s outstanding heritage?” was its role in the spread of the Kiswahili or the Swahili language. The name is derived from the Arabic term for coast or margin—suahel (sometimes also sawahili or suwaail ). It is generally held that this language developed in the polylingual and multiethnic coastal communities and their surroundings through the influence of other languages spoken in the western Indian Ocean region but particularly the Arabic language. The people associated with this language are o en called the Waswahili or just Swahili. As was explained during the opening ceremony of the 2002 Bagamoyo Conference, there are uncertainties and disagreements regarding who the first Swahili were and who they are today.15 A fault line opened, as mentioned, over the question of the impact or importance of foreign cultures in the development of the Swahili language and society. In some circles it has long been claimed that the Swahili culture and language was the result of immigrants from Persia and Arabia who se led along the East African coast. When this view was presented during the opening ceremony of the conference it was immediately questioned by conference participants who argued that it was high time to recognize the African sources for the development of the caravan trade and the Swahili culture and language.16 An argument for the need to abandon the view of the Swahili culture and language as a passive recipient of mainly Asian influences was that many Arabs who came to the East African coast rapidly became Swahilispeakers and “Swahilized” as this was the language of the caravan trade. The assumed one-way flow of influence from Persia and Arabia into the Swahili culture and language was repudiated by one of the conference participants who showed that, for example, the rebel leader Bushiri, who led the 1888–89 rebellion in many coastal towns against the intruding

54

•

Muted Memories

Sultanate and German colonizers (see chapter 7), as well as the warlord and slave trader Hamid bin Muhammed el Murjibi, also known as Tippu Tip, who controlled a large part of central Africa, both preferred Swahili as their language. Karsten Legère, in his paper presented at the Bagamoyo Conference (2002: 3), suggests that Swahili was the most convenient language for them. Although there was a general consensus during the conference on Bagamoyo’s important role in the caravan trade along the central routes through which Kiswahili was spread to the interior of the continent, it was also agreed that the theme in itself could not form a solid enough foundation for the nomination of Bagamoyo to be inscribed on the list. It was therefore agreed that the spread of Kiswahili could be formulated into an argument that further strengthened the caravan trade and slave port theme rather than forming an argument by itself.

Themes Three and Four: Bagamoyo and the Spread of Islam and Christianity Islam has existed on the East African coast for a long time. During the conference it was reported that the oldest mosques at Kilwa Kisiwani, in southern Tanzania, at Kaole 6 km south of Bagamoyo and at Kunduchi just north of Dar es Salaam are from the fourteenth century. Other mosques on the East African coast are certainly older, but the point is that the building of these mosques is testimony to the cra smanship along the coast of building coral stone structures, a cra associated with Swahili culture. It was debated during the conference to what extent these mosques provide evidence of a strong influence from Oman or even Persia and to what extent they are evidence of an East African Swahili culture.17 The same frictions and fractures we have seen in other issues regarding the Swahili towns are found here. External influences may be detected, some conference participants maintained, although the style is unique to the East African coast. It is therefore more correct, from their perspective, to say that these mosques are evidence of an indigenous development along the coast rather than to see these buildings as evidence of strong influence from either immigrants or invaders. The importance of Islam in the creation of Swahili culture is unquestionable. The same is true regarding Bagamoyo’s role in the spread of Islam. One of the participants at the conference, Milcah Achola (2002: 7), mentioned that at the end of the nineteenth century “Bagamoyo had come to be associated with thriving trade and business, and a comfortable lifestyle and an impressive level of Islamic learning.” The theme was not discussed further during the conference, but will be more thoroughly explored in another chapter in this book.

Heritage-Making •

55

From the middle of the nineteenth century Bagamoyo also a racted Christian missionaries who came with the specific purpose of spreading the gospel in Africa and to convert those who were receptive to Christianity. The Bagamoyo Catholic Church became the first mission station on mainland East Africa. It was established in 1868 with the double purpose of converting people to Christianity and to working against slavery. The pastoral method developed here was to focus on young children, not yet influenced by Islam. Boys were primarily welcomed and brought to the seminar building at the mission station. Here some of them were trained as evangelists. The aim was to send them to up-country mission stations where they would convert the pagans. In this sense, Bagamoyo became the gateway for Christianity into the interior of the continent.18 Another strategy in the fight against slavery was to buy slaves, free them, and provide them with a freedom le er.19 The mission station organized “freedom villages,” where liberated men and women, hopefully organizing themselves in conjugal unions, could se le and live by cultivating the land. Some were stimulated to move and se le in up-country mission stations as a means to increase the number of converts there. The importance of Bagamoyo in the spread of Islam and Christianity was discussed during the conference. It was found that this was not a reason by itself to justify the inscription of Bagamoyo on the World Heritage List. It was, however, agreed that the town’s role in the spread of these two world religions was a powerful supportive argument when Bagamoyo and associated sites and routes should be nominated. Crucial issues in heritage-making revolve around the questions of what shall be remembered and why. To qualify to the List, UNESCO demands that a proposed heritage site be of “outstanding universal value.” The final declaration from the conference included a recommendation that Tanzania should nominate Bagamoyo for the UNESCO World Heritage List under the heading “The Slave Port of Bagamoyo and Associated Sites and Routes of Memory.”

Why Bagamoyo and Why World Heritage? Two different “why” questions surfaced during the process: why world heritage and why Bagamoyo? Different answers were put forward regarding world heritage during the preparation and arrangement of the conference. Conservationists and architects o en emphasized the need to preserve the unique material and immaterial values of Bagamoyo. The conservation issues are controversial, as many other issues in heritagemaking, since global and national interests may contradict and challenge more local interests.

56

•

Muted Memories

It may be noticed that the question “if Bagamoyo should be on the List” never surfaced during the preparation phase. In the 1980s, consultants from UNESCO had assisted in the identification and demarcation of the Stone Town and of some isolated structures identified with the heritage of the town. The legacy from the fi y years of heritage-making efforts in the town gave li le room for any questions about whether it was appropriate to nominate Bagamoyo or not.20 The contradictions between privately owned and maintained historical buildings versus governmental plans, restrictions and regulations and the perceptions of conservation versus development resulted in both innovative and interesting strategies. For some, the collapse, rather than the conservation of their stone houses was a desirable development. A participant at the preparatory workshop in the year 2000 could report that the present owners, although they had the economic resources to maintain their buildings, still let them collapse. To many of the owners “the significance of the buildings is that they occupy a commercially valuable plot and the buildings themselves are a desirable object in terms of the coral stones. Therefore the collapse of the buildings is a desirable object so that they release the site and building materials” (Mturi 2000: 14).

Poverty, Tourism, and Development Community development planners saw the options to reduce poverty by increasing the number of tourists expected to visit the town once it was declared a World Heritage site. That poverty was both widespread and severe in Bagamoyo was well known before the conference, evidenced through the SUDP (see the previous chapter). The socioeconomic surveys carried out as a part of the preparations for the conference showed that poverty affected the life of the majority of the population as well as the provision of socioeconomic services to the inhabitants. Given the grim socioeconomic situation in Bagamoyo it is easy to understand that part of the answer to “why World Heritage?” was to combat poverty through an expansion of the tourism sector. To marry outstanding cultural heritage to the tourist industry, however, is a tricky operation since the heritage must both be presented as unique while at the same time it should be seen as universal enough to a ract the a ention of tourists. This paradox is described by Sophia Labadi and Colin Long in the following words: “The constant (re)shaping of local heritage is in many respects part of and simultaneously occurring with the globalizing process itself. Indeed, heritage destinations worldwide may be adapting themselves to the homogenizing trends of global tourism,

Heritage-Making •

57

but, at the same time, they have to commodify their local distinctiveness in order to compete with other destinations” (2010: 8).

Memories of Conscience The Minister for Natural Resources and Tourism answered both the question “why World Heritage” and “why Bagamoyo” when she said in her opening speech at the 2002 International Conference that “it is necessary to put Bagamoyo on the World Heritage List so that people in America and the West Indies can come to Bagamoyo and trace their backgrounds.”21 With this she associated the history of Bagamoyo with the growing tourism of those tracing their family roots that had developed around the Atlantic Ocean. A prominent feature of this tourist industry was the visits by tourists from the United States and the Caribbean to slave castles on the West African coast. The conference was arranged at a point of time when the interest for humanity’s dark histories and memories was growing. UNESCO had launched its Slave Route Project in 1994 as a means to break the silence on slavery and the slave trade (for further information on this project see the next chapter). At the turn of the twenty-first century many European countries had begun investigating if and how they were involved in the Holocaust. Others were scrutinizing their involvement in the slave trade and misdeeds during the colonization of, for example, Congo. Robben Island in South Africa had been declared a World Heritage site to celebrate human rights and show that peaceful coexistence is possible. There was a general interest in and demand for documentation and communication of “dark memories.” It was a general conviction during the 2002 Conference that the slave trade on the Indian Ocean should also have its place of memory. With a growing global understanding of the need to document, talk about, and remember the misdeeds carried out in the past in order to prevent them from happening again, no further answer was needed to the “why” question. In one stroke, two questions had been answered: why Bagamoyo? Because of its assumed role in the slave trade. And which universal value? The memory of the slave trade, as stated at the opening of the 2002 International Conference arranged in Bagamoyo “so people from the Americas and West Indies can come and trace their background.”

After the Conference—Quo Vadis? The conference covered much ground in the process to compile the nomination dossier for Bagamoyo as a world heritage site. The only actor who

58

•

Muted Memories

can submit such a document, according to UNESCO’s rules, is a member state, or a number of states in the case of a serial nomination. This brings us to the last two questions: how was the nomination dossier produced and by whom? These are certainly pertinent questions since UNESCO’s requirement may place monopolistic power in the hands of governmental organizations, which, in the worst-case scenario, may limit both participation and transparency regarding the process and the product submi ed to UNESCO. Some people from academia and civil society who had taken part in the preparation of the conference and participated in the Conference itself felt that this had happened with the nomination of Bagamoyo. They were not invited to the meetings where the formulation of Bagamoyo’s nomination took place.22 The preparation and organization of the 2002 International Conference were highly participatory exercises, as already mentioned. The local civil society as well as representatives from the business communities in Dar es Salaam and Bagamoyo and members from both the main university campus and from UCLAS, Dar es Salaam, took part. In fact, one of the explicit aims of the conference was “[t]o . . . ensure the participation of ‘grass-roots’ in Bagamoyo in the process towards declaring Bagamoyo a World Heritage Site” (Lindström 2002). Although frictions and fractures occurred, still “the magic of the list” (Askew 2010) eventually managed to calm feelings and led to smooth cooperation during the conference, and the participants managed to agree on both what to nominate for the list and why. The way forward, however, was covered with hurdles and thorns. Perhaps it lies in the rules of the game that the task force appointed at the 2002 International Conference for the further dra ing of the nomination dossier was dominated by government staff, particularly from the Department of Antiquities at the Ministry for Natural Resources and Tourism. A suggestion that an external consultant from UCLAS should support the government side was also adopted during the conference. An international expert on heritage-making also assisted the task force in the formulation of the text that presented Bagamoyo as a potential site of World Heritage. Cynics would perhaps conclude that the government and state skillfully orchestrated the whole process. UNESCO not only promotes but demands popular or local participation in the making and management of a World Heritage site. However, it also stipulates, as mentioned, that only a state party, or a number of state parties in the case of a serial nomination, can nominate something as heritage and a site for inscription on the List. Since heritage-making involves the exposure of power and knowledge and triggers frictions and fractures, it is no surprise that the party endowed by UNESCO with extraordinary muscles, took the opportunity

Heritage-Making •

59

to use them. Some participants from the conference and the preparation phase found themselves marginalized a er the conference. We have seen that the theme suggested at the preparatory conference in the year 2000 on Bagamoyo as part of world heritage—Bagamoyo as a representative for the East African Caravan and Slave Trade—was narrowed down during the 2002 International Conference to a focus on an East African slave route. The perspective was, at the same time, widened to include Bagamoyo and associated sites of memory. Two factors were behind this development: UNESCO’s Slave Route Project that focused on the routes, and the recognition during the conference that what remained of Bagamoyo’s Stone Town and the historical buildings alone could not motivate an inscription on the List. One question remains: is Bagamoyo the prime slave port along the East African coast as suggested during the conference and in the presentation of the town on Tanzania’s tentative list? Or could other ports along the East African coast also aspire to this bizarre title? It has long been known that Kilwa, in present-day Tanzania, was the main port on mainland East Africa for the export of slaves to Zanzibar, the main slave market in the Western Indian Ocean world. Abdul Sheriff (1987: 226–35) estimates that approximately 23,000 slaves officially passed through the market annually at the height of the slave trade, of which approximately 75 percent came from Kilwa, or rather from the societies at Lake Nyasa (Malawi) and Lake Tanganyika (Koponen 1988: 85). Slaves were marched along the coast from Kilwa and northwards, sometimes all the way to Somalia a er the shipping of slaves was banned through the 1873 Zanzibar-British agreement and British ships started to chase Arab sailing ships on the Indian Ocean with slave cargo. Was the outstanding cultural value of the central caravan route an achievement by coastal traders primarily engaged in the slave trade or shall the heritage of the caravan trade be accredited to the up-country porters and traders who initiated the long distance caravan trade by bringing their goods to the coast? The next two chapters will look into the branding of Bagamoyo as an important slave port and its involvement in the slave trade, while the second part of this publication will tell an alternative story about Bagamoyo and the caravan trade.

Notes  1. Author’s field notes, September 2002.  2. Sources other than the texts presented at the conference will be used in this chapter to clarify and explain issues dealt with during the discussions at the conference.

60

•

Muted Memories

 3. Now, more than seventeen years a er the conference, different understandings may have developed concerning the issues discussed then. The aim in this text, however, is to present what was said and done before and during the conference, as a background to what was presented in the file when Bagamoyo was placed on Tanzania’s Tentative List.  4. Dr. Mapunda’s paper was presented by Dr. Msemwa.  5. Author’s field notes, September 2002.  6. Author’s field notes, 1999–2002.  7. From the Italian, German, and Swedish Embassies, UNESCO and UNDP.  8. Author’s field notes, March 2000.  9. In mind here is the creation of new categories as sacred landscapes which may suit Africa and Australia—continents “poor in culture but rich in nature.” 10. The fourth, Bagamoyo as an archaeological site, was discussed during the conference but not incorporated in the recommendations from the Conference to Tanzania. 11. With “Bantus” here understood as people from mainland Tanzania. The Baluchi were brought as mercenaries by the Sultan to protect his economic and political interests. 12. See chapter 10 regarding the difference made between Arabs born on the Arabic peninsula versus Arabs born on the East African coast. 13. See Donald Simpson’s (1975) book Dark Companions on the role African guides and porters played for the European and American “explorers,” missionaries, and traders. 14. This text is considered a fake by some, since the door was first fi ed into another church but moved to its present place in Bagamoyo at the hundred-year anniversary of Livingstone’s death in 1874. This door and sign is, however, an indication on the role Bagamoyo played in the spread of world religions. 15. See, for example, Arens 1976; Chami and Pwiti 2002; Constantin 1989; Fabian 2007b; Middleton 1992; Nurse and Spear 1985; Spear 2000; Trimingham 1964. 16. More recent research has shown that Kiswahili is basically a Bantu language and that the terms borrowed from the Arabic language are restricted to a few sectors like trade, arithmetic, and seamanship. In addition, they entered the language relatively late. 17. Author’s field notes, September 2002. 18. For further information on the caravan trade see chapter 5. 19. Another method was to welcome slaves freed by the British warships patrolling the coast in search for Arabic dhows carrying slaves. 20. Another reason, which surfaced in some of the discussions but was never put on paper, was that Zanzibar Stone Town was inscribed on the List in the year 2000. It seemed appropriate to also have a candidate from mainland Tanzania. 21. Author’s field notes, 9 September 2002. 22. Author’s field notes, March 2013.

3  Fractures in the Image of Bagamoyo Despair or Joy? Dynamics in Heritage-Making Fabian Kigadye was not quite correct when, in October 2003, as an employee at the Department of Antiquities, Tanzania, he told journalist Thorsten Bothe that it would take three years to complete Bagamoyo’s nomination dossier for the World Heritage List (WHL). On 20 February 2006, Bagamoyo was put on the Tentative List by the Tanzanian government, but more than fi een years a er Kidagye’s statement, it has not yet been successfully nominated for the WHL. The inscription of Bagamoyo on Tanzania’s Tentative List was officially announced on the government’s web page and it was said that Bagamoyo was designated as Tanzania’s seventh World Heritage site. Other pages on the Internet instantly picked up the news and labeled “Bagamoyo a World Heritage on the Nomination List” (World Monuments Fund), or “Bagamoyo UNESCO World Heritage Side?” (sic—should be Site, but notice the question mark) (Bagamoyo Friendship Society) or simply “Bagamoyo, Tanzania: UNESCO World Heritage Site on the Slave Route” (Africa Travel Magazine 2006). The placement on the Tentative List was probably seen as a formal procedure before Bagamoyo was accepted on the Heritage List. But what exactly was nominated to the Tentative List? The next pages will

62

•

Muted Memories

investigate not only what the government of Tanzania nominated, but also how and why. The International Conference arranged in Bagamoyo in 2002 evaluated the town’s prospects to be nominated to the WHL. The experts who participated in the conference not only assessed the site’s chances of entering the WHL, but also tentatively formulated what to nominate and why. By doing so, certain physical structures, aspects, narratives, and memories were selected and highlighted, o en at the expense of others, as will be shown below. Heritage-making is a paradoxical process, which tries to integrate and make use of the heritage in wider socioeconomic processes of development and change and, at the same time, preserve it. Three of the techniques used for this are: the employment of a selective eye, the disembedding of the heritage, and the standardization of the heritage’s image.

To See or Not to See: Two Sides of Bagamoyo Heritage-making is a process of selection; some aspects, stories, and memories are highlighted while others are not seen, made invisible, or hidden. The making of Bagamoyo as a terminal port for an East African slave route is an excellent illustration of what is selected to be seen and what is silenced or “not seen” (Wallace 1996). The very name Bagamoyo invites processes of selection. It is composed, as already mentioned, of the two Kiswahili words bwaga, which roughly means “to lay down,” and moyo, meaning, “heart.” Already in 1970, Walter Brown provided two contradictory interpretations; “the first holds that the town’s name originated with the lament of slaves who knew that, although they were to be shipped to distant lands, their hearts would forever remain in their beloved homeland” (W. T. Brown 1970: 69). There are reasons to assume that the interpretation was influenced by Western notions of nationalism as a territorialized “Vaterland,” an imagined community (Anderson 1983), which may have been at odds with the classificatory systems used by people in East Africa, who categorized themselves primarily along kinship and ethnic lines. The second interpretation, which Brown finds more probable, is that “Bagamoyo was named by the caravan porters who felt they had reached the end of their long arduous journey from the interior,” and could now “ease from worrying, cease from care and anxiety” (W. T. Brown 1970: 69).1 Both versions have been ripe for picking since the early 1970s. The first was selected for the nomination dossier of Bagamoyo to the Tentative List. Implicit in this choice was the silencing of the second interpretation.

Fractures in the Image of Bagamoyo

•

63

The two different interpretations2 of the name cannot have been unknown to the nomination commi ee. Brown’s articles from 1970, his Ph.D. thesis from 1971, as well as a booklet produced in the mid-1990s by the Catholic mission station in Bagamoyo wri en by Kajeri and Henschel give the two versions, as does the publication Bagamoyo: A Pictorial Essay of Tanzania by Kirknaes and Wembah-Rashid (1993). Crucial evidence for the second interpretation is the reproduction of the text of a song said to have been sung by the porters when they approached Bagamoyo. Only the first and the last of the six verses presented both in Brown’s article from 1970 and in Kajeri and Henschel’s booklet are reproduced here: Be happy, my Soul, let go all worries Soon the place of your yearning is reached: The town of Palms—Bagamoyo ... Be quiet, my heart, all worries are gone The drum beats and with rejoicing We are reaching Bagamoyo (Kajeri and Henschel n. d.: 4)3

The arrival of the caravans to Bagamoyo was also celebrated by the townspeople, as witnessed by German visitors, who in the jargon of the time and their trade,4 reported that: Women and children rushed to meet the caravan and received it with ear-piercing music. The blacks sang and danced and provoked the porters to join in their noise making . . . Those who were armed fired off their rifles. Once the caravan arrived in the town and the ivory and the goods were stored in houses and warehouses, the porters received their wages and stormed off. They rented themselves a hut, found themselves a woman, and played, drank and sang the whole day until the last hard-earned rupee was spent. (Sturtz and Wangemann 1890: 32; as reported by S. Fabian 2007b: 117)

The porters’ appreciation of the arrival to Bagamoyo is well documented. The sandy beach, the palms and the “good life” in terms of songs, dances and women appear in another song sung by the caravaneers: “Oh, what joy to see the dances, Where the lovely girls sway, In the evenings in Bagamoyo, Be still, my heart, All worries are gone, The call to rest thunders, and with jubilation we reach Bagamoyo” (S. Fabian 2007a: 452).5 A central feature of the Swahili world is the ngoma, which will be further explored in another chapter of this book. Poetry, sung as wimbo (sing. for song, nyimbo in plural) when performed, o en took the form of an ngoma, a multivocal term that stands for song, dance, and a musical instrument, the drum. Ngoma can be described as “any event in which music plays a part” (Campbell and Eastman 1984: 476). The dances were o en

64

•

Muted Memories

performed as single-gender events. Women danced in a circle and men in a line. One of the most popular female dances was the chakacha, o en performed by young girls, assisted by mature women. In the dance they tied a kanga, a piece of co on cloth, around their hips and danced in a counterclockwise circle. It may be assumed that this was the dance “where the lovely girls are swaying” since the characteristic feature of it is the rapid hip rotations performed by the dancers. Thousands upon thousands of porters each year laid down their cargo and celebrated a er they had arrived. The arrival not only meant the release of the heavy burden, but o en also a prolonged time, two to four months, of festivities, palm wine, exposure to an urban milieu and opportunities for status accumulation. During the nineteenth century, caravan journeys to the coast became popular among some people as a way to increase one’s economic, social, and cultural capital. This applied especially to the Nyamwezi people, but also to other groups in the interior of the continent. One example is an Uvinza (east of Lake Tanganyika) village chief, who became famous as a person who “had seen the world” since he had travelled to Bagamoyo (W. T. Brown 1971b, 631). The participants at the 2002 International Conference in Bagamoyo agreed, as mentioned, to recommend that Tanzania nominate Bagamoyo for the World Heritage List on the formulation “The Slave Port of Bagamoyo and Associated Sites and Routes of Memory” with focus on the central caravan routes.6 It was shown in the previous chapter that the “slave route” theme originated in UNESCO’s initiative in 1994 to launch the “Slave Route Project” (SRP). Up to the 2002 conference, this project had had a clear focus on the Atlantic slave trade with inscribed World Heritage sites on the West African Coast, as shown by SRP’s 2005 Evaluation Report. At the conference it therefore seemed appropriate, in order to remember slavery and the slave trade on the Indian Ocean, to add a place of memory for the East African slave routes (compare UNESCO’s 2005 valuation of the SRP). A er the 2002 International Conference, the further process of formulating the Bagamoyo case derailed in at least two important ways; the memory of Bagamoyo as identified as the main port in the caravan trade on mainland Africa, which meant that the caravan route where porters and traders walked in both directions during an early phase of globalization was reduced to a terminus on a one-way slave route that began in the interior of the continent and ended at the Indian Ocean ports. The second was that crucial parts of the central caravan routes stretching from central Congo via Bagamoyo to Zanzibar and further to Arabia, Persia, India, China, Europe and the Americas, were omi ed.

Fractures in the Image of Bagamoyo

•

65

Misconstrued Memories A careful reading of the presentation of the nomination of Bagamoyo as a potential site of World Heritage reveals an uncertainty about what the place actually represented. This became apparent during the 2002 Conference when different scenarios were tested before the eventual formulation “The Slave Port of Bagamoyo and Associated Sites and Routes of Memory” was agreed upon. As seen, the focus during the conference was on Bagamoyo while other sites and phenomena like the caravan trade routes and rest places and trade stations along the tracks were added to the port town. However, despite the recommendations from the 2002 International Conference to focus on Bagamoyo and the caravan routes, the explanatory text in the 2006 application describing the motivation was narrowed down to the slave route with the following words: “Until, not even 150 years ago, millions of Africans had to bear a cruel fate. They were captured by slave hunters, chained together and forced to walk sometimes hundreds of kilometers to be sold for example to planters who used them as cheap labor in their fields” (UNESCO 2006). The complexity and outstanding value of the long-distance caravan trade as a cultural route along the central caravan routes was hereby silenced when the slave route theme was selected for the official announcement of the nomination. Therefore, this nomination deserves a careful consideration. A first interpretation is that a general, not to say global, view is reproduced here according to which slave hunters, o en Arabs, captured Africans who were marched away in chains to work on plantations. This image was introduced by David Livingstone during the middle of the nineteenth century (compare Page 1974: 69–70) through his many texts reporting on his meetings with caravans with slaves in chains and encounters with slave traders. Slaves were called the “black ivory” and a common picture was of chained people who were marched to the coast carrying ivory tusks. The picture was recycled and perpetuated during the second half of the nineteenth century by many other travelers, o en to both propagate European interventions in the East African caravan trade, which at that time was dominated by Arabs, and, eventually, to motivate the German colonization of the area. The fight against the slave trade and slavery also turned into European mission activities to spread both Christianity and European civilization to the interior of the continent, captured in Livingstone’s triple C—to spread Christianity, Civilization, and Commerce to central Africa as a substitute for the slave trade. This is an image that sticks. Edward Alpers warned us forty years ago that “the old stereotype idea that most slaves were seized by marauding

66

•

Muted Memories

bands of Arab and Swahili traders is just another one of the myths which have grown up around the East African slave trade” (Alpers [1967] 1974: 24). A perhaps extreme but recent view of the Arab as both a slave owner and a Muslim can be borrowed from the USAID study “Bagamoyo Governance Baseline” (Torell, Mmochi, Spiering 2006: 4): During the Arab dominance, Muslim Arabs from Oman established several towns along the Tanzanian coastline and on nearby islands such as Pemba and Zanzibar. Coastal towns like Bagamoyo developed into independent economic units owing varying degrees of allegiance to the sultan. There were two social classes in the town, the ruling Arab class and the African laboring class—of which the majority were slaves.

The Arabs only constituted a small minority in the coastal towns and stood for only a part of the slave trade along the central routes, as will be shown below. In Bagamoyo they were o en outnumbered by the Indians during some years. Even the number of slaves in Bagamoyo was restricted. The sources disagree, ranging between 15 and 30 percent of the population during the peak of the caravan trade. The free inhabitants— middlemen-merchants o en called Shirazi and the free ordinary people o en called the Swahili—constituted the majority of the town.7 It is shown in other parts of this book, how frequently this myth—that the coastal societies and cultures were created by immigrants—appeared in texts on Bagamoyo and other coastal towns. It is o en taken for granted among both scholars and laymen, that influences from the Orient as well as Arabs in flesh and blood ignited the development of the Swahili culture, urban se ings, and the slave trade. One example is Peter Garlake (1966) who came to the conclusion regarding Swahili architecture and culture that “the religion was always Islam and the fundamental base of the culture came from abroad. . . . The culture was provincial—initiatives were always from abroad.” Forty years later, Graham Mercer states in his book about Bagamoyo that the Swahili culture is “a surprisingly advanced culture, for the Arabs and Persians (the la er known, even today, as Shirazi) who had se led here between 800 and 1200 AD, had traded genes with the locals, as well as commodities” (Mercer 2007: 17). These stereotypes about the Arabs in East Africa have provided them with a Janus face among Europeans and Americans; they are, on the one hand, given the noble task of bringing civilization to the area, but on the other hand, seen as those who brought violence, suffering, and slavery to the helpless Africans. Both stereotypes deserve to be questioned and challenged. The seemingly contradictory views on the influences from the Arabs can, however, be forged together with another stereotype, namely the helpless, defenseless, unsophisticated African who needed the protection of the European,

Fractures in the Image of Bagamoyo

•

67

in the form of colonization to be rescued from the cruel and cunning Arab. There is a clear fracture in the picture in the eyes of European travelers, missionaries, colonizers, and traders. The Arabs on the coast were assumed to have brought civilization during the centuries they had been there. They were a benevolent force as long as they stayed within the tenmile wide coastal strip, which was believed to be the “mainland dominions” of the Sultan of Zanzibar. Arabs in the interior of the continent were a malice since they interfered with the European interests in this fertile field for colonization, trade, and Christianity. This stereotypic portrait of the Arab and the Arabic, together with the stereotypes regarding the uncivilized African, o en included a fourth popular stereotype, namely that the Arabs organized the major caravans that set out from the coast. This view is present in the following portrait of the caravan porters who were full of joy when they returned home to Bagamoyo: “A er facing and overcoming all the hardship of the journey to and from the interior, they were united with the families and friends” (Kirknaes and Wembah-Rashid 1993: 4). Most of the caravan porters did not return “home” to the coast. In reality, nineteenth century large-scale caravans originated in the area west of Lake Tanganyika and south of Lake Victoria in the interior of the continent. The caravan trade was only gradually taken over by Arab and Swahili traders from the coast who relied on credits and loans from the Indian bankers/traders in the coastal towns and many, perhaps most, of the porters organized in a town like Bagamoyo for an up-country expedition were Nyamwezi, who had come to the coast with their ivory, hides, livestock, and slaves.

The Branding of Bagamoyo as a Slave Trade Port The “despair” interpretation of the name was chosen when presenting Bagamoyo as a potential World Heritage site in an a empt to brand the town as a major slave port on the East African coast. Historian and Bagamoyo expert Steven Fabian is very skeptical about the launching of the town as a sibling to the infamous West African slave port Gorée.8 He shows in an article (2013) that Bagamoyo was incorrectly made into a representative for the East African slave trade. Referring to the nomination of Bagamoyo for the Tentative List, he writes: “Bagamoyo, the Tanzanian deputation [those working with the nomination dossier] argued, has been the major slave port of the region. The presenters, however, could not produce enough hard evidence to support their claims” (S. Fabian 2013: 96). It is today well known that Kilwa, a port in the southern part of Tanzania, was the major slave port which supplied slaves to the plantations on the

68

•

Muted Memories

French islands in the southern Indian Ocean region and to the slave markets on Zanzibar. Fabian substantiates his claim that Bagamoyo as a major slave port is a misconception, probably created both willingly and unwillingly, through a number of different coinciding processes. As evidence of the minute role Bagamoyo played as a slave port he mentions the absence of a slave market in or around Bagamoyo, the absence of accounts on a slave market in the many travelers’ texts from Bagamoyo and the fact that the British Consul on Zanzibar, who constantly worked for the abolition of the slave trade and supervised the trade in Zanzibar and Kilwa, never saw any reasons to intervene in the trade in Bagamoyo (S. Fabian 2013: 96).9 On his own question about how this misunderstanding of Bagamoyo developed, Fabian concludes: “The answer appears to lie within Bagamoyo” (S. Fabian 2013: 101). The single most important factor behind the branding of Bagamoyo as a slave port is probably the work of the Catholic mission station in Bagamoyo established in the 1860s. Its small museum shows the horrors of the slave trade, the struggle for its abolition and the missionaries’ work with rescued slaves in the “freedom villages” on its premises. The message of slavery in Bagamoyo is also broadcast through a series of booklets sold at the museum.10 At least two of them are the result of collaboration between the mission station and the Department of Antiquities, Bagamoyo. It may be hypothesized that the branding of Bagamoyo as a slave port is part of the globalizing effects exerted by the World Heritage List in combination with UNESCO’s Slave Routes Project. A similar development is documented in the beach resort Shimoni on the Kenyan coast. Stephanie Wynne-Jones and Martin Walsh (2010) show how the caves (shimo in Kiswahili means “hole”) in the rocks suddenly became filled with stories about slaves being kept in the caves. But this storytelling only started to pop up around the year 2000. The people interviewed by Wynne-Jones and Walsh pointed out that the slave caves and stories became important in the late 1990s and early 2000s when this beach resort on the Indian Ocean tried to develop its tourist sector. Bagamoyo went through a similar process during the same period of time. Slaves have certainly been bought and sold and they have lived and worked in and around Bagamoyo and it has been estimated, as mentioned, that between 15 and 30 percent of the population were slaves at the end of the nineteenth century (S. Fabian 2013: 100).11 Visitors, who during the late 1980s came to the only bar, restaurant, and guesthouse at the shore beneath the Old Fort, encountered street vendors who o en offered the typical iron slave necklaces for sale. Bagamoyo will forever be associated with slavery. The point, however, is that it was only a er the turn of the

Fractures in the Image of Bagamoyo

•

69

twenty-first century that the town was made into the most important slave port in the area when it was ascribed an “outstanding universal value” as the slave port on the East African coast. In its presentation of the nomination of Bagamoyo the government of Tanzania explains that “important slave trade evidence include slave and

Map 3.1. Some of Bagamoyo’s historical buildings identified for the inscription on Tanzania’s Tentative List. Created by the author.

70

•

Muted Memories

slave descendants, buildings such as Caravan Serai, Von Wissman block (Dunda Tower), Old market, Customs house and the Old fort” (UNESCO 2006). In his critical examination of Bagamoyo as the most important slave port Fabian points out what may be more than a coincidence. Three of the buildings mentioned in the government’s presentation; the Old Fort, the Customs House and the Caravanserai, also figure on a map in the Catholic Mission’s museum as being connected to the slave trade. In an endnote, Fabian muses: “It is not clear if the museum was informed by local folk wisdom to identify them as such, or vice versa” (S. Fabian 2013: 111n33). More important, however, is that these three buildings lack substantial connections with the assumed slave trade and therefore constitute no evidence for Bagamoyo being the major slave port or an important slave market. Many visitors to the Old Fort have seen the high wall surrounding the yard and the massive doors facing the Indian Ocean, resembling the famous doors of “no return” at some West African slave castles, and they have o en listened to stories from self-trained guides about an underground tunnel used to take the slaves from the Fort to the ships at the beach. Underground tunnels and caves are o en associated with the slave trade. One explanation behind the tunnel story in Bagamoyo is that the loading of slaves, it is explained, had to be kept a secret to avoid the eyes of the British abolitionists (Wynne-Jones 2011: 332). In the Bagamoyo case this seems unrealistic since the British, a er the 1873 abolition agreement between the Sultan of Zanzibar and Great Britain never showed an interest in controlling the trade in Bagamoyo. The visitors have also looked into the Fort’s small rooms and wondered how many slaves were kept there? What is not obvious or told to the visitors is that the wall and the doors were erected by the German colonizers for defense purposes when they used this building as their first headquarters. The walls were not erected by slave traders to keep the slaves locked up. The Old Fort is the oldest existing building in Bagamoyo. It was built by Abdullah Seleman Marhib during the 1860s as a well-protected residential house. During the 1880s it was used by the Zanzibari Sultan’s representative in Bagamoyo—his liwali. From the time of the German colonization in the mid 1880s and, for most of the time until the end of World War I, it was a military base for the German forces. During this time the massive walls around the court were built. A er the British takeover and a er independence in 1961, the Old Fort functioned as a police station and/or prison. Hundreds of people have certainly laid down their hearts in despair in the small cells of the building, but as prisoners or people held in custody during colonial and postcolonial times rather than as slaves during the precolonial era.

Fractures in the Image of Bagamoyo

•

71

Illustration 3.1. Doors to the cells at the Old Fort said to have housed slaves. Photo by the author.

Another building identified in the presentation of Bagamoyo as a potential World Heritage site connected to the East African caravan trade is the Customs House,12 spectacularly placed close to the beach and the lateen-rigged sailing boats. Li le imagination is needed to see how slaves, according to the tourist guides, were kept right here on the shore before they were marched a few meters to the waiting boats. A picture of the Customs House decorates the front page of Kirknaes and Wembah-Rashid’s booklet Bagamoyo: A Pictorial Essay from Tanzania (1993). It is explained in the booklet that the Customs House was a “place for embarkation of slaves,” (1993: 2) which is not correct since the house was designed and constructed by the German colonizers during the mid-1880s (S. Fabian 2013: 102). In fact, the original Customs House built by the Zanzibari Sultan, which certainly was connected with the slave trade, was situated north of the current structure. It was dismantled by the Germans in 1885 and moved to the port town Saadani approximately 35 km north of Bagamoyo (W. T. Brown 1970). It was replaced with the two-storied Customs House that exists today. The new German Customs House was not erected on the place where the Sultan’s Customs House had been built to avoid potential connection between the regime of the Sultan and the German colonialists. Hence, there are no connections between the postcolonial and precolonial structure, now in Saadani, where slaves may have been kept, and the Customs House currently seen in Bagamoyo.

72

•

Muted Memories

Illustration 3.2. The Old Fort. In the forefront, the defense walls built by the Germans. Photo by the author.

Fractures in the Image of Bagamoyo

•

73

The third building in Bagamoyo, associated with the slave route in the presentation of the nomination of Bagamoyo, is the Caravanserai. At the time of the Bagamoyo Conference the daily paper The Guardian had an article about the conference with a picture of the Caravanserai and the text: “Remains of a building that used to accommodate slaves prior to their shipment to Zanzibar” (“Making Bagamoyo a World Heritage Site” 2002, September 5). The findings of an archaeological excavation of the Serai carried out around the year 2000 and presented at the 2002 Bagamoyo Conference13 revealed that this building had been a guesthouse (one of the meanings of the name Serai is guesthouse) “[f]or the comfort of travelers, traders and caravan leaders” (Wynne-Jones 2011: 332). The German colonizer Friedrich Ferdinand Mueller noticed, already during the 1880s, that caravan porters from the interior of the continent here laid down their cargo (see quotation in chapter 1, “Bagamoyo as a Window into the History of Heritage-Making”). The slave branding, however, continues to stick to the building. A sign outside the building, which now houses a museum and information center explains: “The history of the Caravan Serai is intimately related to the history of the slave trade. Slaves were used to transport the goods, and they were sold at the serai and at other sites in Bagamoyo. At the height caravans could hold up to 50,000 slaves per year, all arriving at the caravan serai.”14 No, the Caravanserai that today’s visitors to Bagamoyo encounter was neither a slave market nor a place where slaves were kept. German documents in the National Archives in Dar es Salaam show that the Caravanserai, as revealed by the archaeological investigation, was a guesthouse rather than a slave market (S. Fabian 2013: 102; Chami et al. 2004). The map in the Archives shows that ten long buildings that no longer exist surrounded the buildings erected by the Germans in 1892. These buildings could provide housing for up to five thousand porters. Each porter who used the facilities had to pay half a rupee for the accommodation (S. Fabian 2013: 105). Porters enjoying their time in this urban business center lived here, not slaves waiting in despair to be sold and/or shipped away.15 The major slave port on East Africa’s coast was Kilwa, not Bagamoyo. Initially it was the gold trade with Zimbabwe in the interior of the continent that brought prosperity to Kilwa. The establishment of the French colonies on the islands of the southern part of the Indian Ocean stimulated the slave trade from mainland Africa, o en via Zanzibar. The minted coins from Kilwa and its domed mosques bear witness to the flourishing past of this town (Unomah and Webster 1976: 271). A er the anti-slavetrade treaty between the Sultan of Zanzibar and the British government, Kilwa’s traders developed a new strategy for the exportation of slaves. For

74

•

Muted Memories

a couple of decades they marched the slaves in their possession overland from Kilwa to the ports at Saadani and Pangani north of Bagamoyo. The Arabic theme in the popular view of the East African slave trade suggests that slaves, when they le the coast for the slave market in Zanzibar, were destined for Arabia or the Persian Gulf. Studies have shown, however, that most of the slaves remained on the plantations along the coast and on its off-shore islands. The main proportion of the slaves brought from Africa to Asia came from the Red Sea area. Slaves from present-day Ethiopia and Sudan were exported eastward from ports like Suakin and Massawa (Austen 1988). The main slave market along the coast was in Zanzibar Stone Town. The clove and coconut plantations on the islands Unguja (also known as Zanzibar Island) and Pemba were the main absorbers of the slaves brought to the islands during the second half of the nineteenth century from the interior of the continent. Estimates from the beginning of the nineteenth century and onward show that two-thirds of the populations on Unguja were slaves (Nimtz 1980: 40) as were two-thirds of Stone Town’s inhabitants (Pearce 1920, quoted in Hino 1980: 109). Tanzania’s Director for Antiquities was not entirely correct when he told reporters from the magazine East African Business Week, as late as 13 January 2014 that “Bagamoyo was regarded as the main historical port whereby slaves had to be shipped abroad in about 160 years ago.” No, instead Bagamoyo was the major port for the long-distance caravan trade in ivory, cloths, weapons, ammunition and slaves and the short-distance trade in copal, and luxurious goods from China, India, Europe and America.16 The selection of the slave route brand in the presentation of Bagamoyo as a potential World Heritage site was unfortunate for at least two reasons. The first was that the case was not built on solid scientific knowledge. The second, as mentioned above, was that other stories were silenced.

Porters and Traders on the Central Caravan Routes The origin of this long-distance trade was already suggested by Benne ([1968] 1974, 210) when he wrote: “The initial stimulus for this movement came from Africans of the interior, especially the Nyamwezi, . . . who opened up trade routes that Arabs later followed.” It is well known today that the caravan trade along the central routes from the Congo area to Bagamoyo developed in the late eighteenth century, primarily through the initiative of the Nyamwezi people living south of Lake Victoria. The main item they brought to the coast was ivory tusks. However, they also traded

Fractures in the Image of Bagamoyo

•

75

in slaves, who at that time were brought with the caravans. Slaves were used as a means of exchange to obtain ivory, ca le, food, and the right to pass a certain area. Thus, they were primarily used in the caravan trade for other purposes than as a cheap means of labor brought to the plantations around the Indian Ocean. Tens of thousands of the Nyamwezi, both men and women, worked as porters and traders as they walked along the central caravan routes with their merchandise (compare Deutsch 2007; Roberts 1970; Rockel 2006a). A characteristic of the caravan trade was that the porters from the interior of the continent could not carry enough food with them for their own sustenance. They had to bring trade items and rely on the availability of food along the routes. This also applied to caravans that set out from the coast. One of the most important items of merchandise they brought with them to barter along the routes was, as mentioned, slaves. The intertwined connection between the ivory and the slave trade is captured by Henri Médard and Shane Doyle in the following way: Nor can the international slave trade in East Africa . . . be separated from that of ivory. Even if one product seems to dominate, without the other the trade would have collapsed due to the combination of lack of profit, transport and currency. Ivory and slaves were exchanged for a variety of imports from the wider world and facilitated the emerging of new long-distance trade with long-standing networks of African trade. (Médard and Doyle 2007: 8)

The dominant role in the caravan trade held by the Nyamwezi traders and porters influenced the development of a specific caravan culture (Rockel 2006a: passim). Their cultural elements not only followed in the tracks of their own caravans but were also imprinted on the caravans organized by Arab and Swahili traders from the coast (Rockel 2006a). The Nyamwezi caravan culture was a strong standardizing force on the central caravan routes during this early time of globalization. It goes without saying that standardization was a prerequisite for the organization of thousands of porters and also for techniques of measurement and modes of payment in the trade.17 The Nyamwezi caravan trade became the initial foundation of Bagamoyo’s merchants. The German explorer Oscar Baumann (1890: 30–31) reported that “one can identify the tastes of the Wanyamwezi in the countless Indian shops of Bagamoyo which are laden with many glass beads, wire, and other objects treasured by the Central Africans” (quoted in S. Fabian 2007b: 183). Fabian adds that “it [thus] makes li le sense to view these porters as outsiders to coastal urban communities like Bagamoyo; the economies of these towns revolve around the trade, needs, and desires of these upcountry traders” (S. Fabian 2007b: 183).

76

•

Muted Memories

By extension, the Nyamwezi porter/traders were also the foundation for a part of the Zanzibari Sultan’s economic empire. Through his control over what was transported to the mainland ports he usurped the tax collected on items under his control, like ivory and copal. Bagamoyo’s dominant role in this trade is evident from the following comparison: in the year 1872–73, Bagamoyo accounted for 61 percent of the value of the ivory shipped through Zanzibar from the entire coast, from Mogadishu via Lamu, Mombasa, Tanga, Pangani, Kilwa to Mozambique, and no less than 96 percent of all the copal (W. T. Brown 1970: 633). The porters/traders were reluctant to accept money in exchange for their ivory. Instead they wanted cloth, guns, and ammunition. Cloth was the prime item for the caravans that set out from the coast during the nineteenth century. For centuries cloth from India and England had been imported to the East African communities. During the first decades of the nineteenth century, however, American cloth began to dominate on the market. In particular, unbleached calicos, called merikani (based on “America”) from the textile mills in Boston and Salem, grew in popularity, not only to the benefit of the textile industries in these towns, but also for US trade with East Africa. Through the supply of merikani, US traders on Zanzibar could dominate the growing trade in copal, a resin used in US industries to produce a varnish for furniture and vehicles. During the 1850s “the number of American vessels and their total tonnage consistently surpassed all other Atlantic nations” (Prestholdt 2004: 772.). “Other Atlantic” here meant British, German, French, Portuguese, and Danish among others. In 1888, for example, a total of 15 million yards, primarily merikani, were imported a year to Zanzibar, mainly for up-country consumption (Prestholdt 2004: 777).

Tuning into a Metacultural Story In her opening speech at the 2002 Bagamoyo Conference, the Minister for Natural Resources and Tourism said, as already mentioned in a previous chapter, that “it is necessary to put Bagamoyo on the list so that people in America and the West Indies can come to Bagamoyo and trace their background.”18 This is what Roland Robertson (1992) may have called the “universalization of the particular.” In the choice of Bagamoyo as a memory of the slave route, Bagamoyo as a memory of the complex caravan trade route or a cultural route seems to have been lost. The presentation and representation of Bagamoyo as a site of World Heritage was formulated in a way that disembedded the heritage from what Bagamoyo

Fractures in the Image of Bagamoyo

•

77

and the caravan trade along the central routes was during the nineteenth century. The Minister’s opening words may be seen as a fine tuning of an interest which was growing at that time through UNESCO’s Slave Route Project (SRP), launched in 1994. It was the result of more than a decade of efforts to both break the silence that surrounded this part of our common history, but also establish a new understanding of Africa’s history or, as captured in the Preface to UNESCO’s eight volumes on the General History of Africa: “a phenomenon which did great disservice to the objective study of the African past was the appearance, with the slave trade and colonization, of racial stereotypes which bred contempt and lack of understanding and became so deep-rooted that they distorted even the basic concepts of historiography” (M’Bow 1981: xviii). The SRP was initiated in 1991 and formally launched a couple of years later. From its beginning it had two equally important aims; to give recognition to and stimulate research on the slave trade, and to stimulate the development of tourism (Wynne-Jones 2011: 318). The purpose of the first aim of the Slave Route Project was to: break the silence on the slave trade because to ignore or obscure such a major historical fact constitutes an obstacle to peace and mutual understanding. By virtue of the universal silence that shrouded it, the extreme violence that accompanied it, the troubling light that it sheds on the discourse to justify it and also, the cultural interaction it generated, the slave trade raises some of the most burning questions of our societies. (UNESCO 2007: 3)

In a similar vein it is assumed that the remembrance of slavery and the slave trade may positively affect today’s sociocultural processes and “contribute to the establishment of a culture of tolerance and peaceful coexistence” (UNESCO 2007) in a world where the efforts to establish racial and ethnic purity, o en leading to isolationism and xenophobia, gain in strength. A particularly disturbing feature in the defense of slavery in general and the enslavement of Africans in particular which emerged in the AngloSaxon world during the Atlantic slave trade and the establishment of plantation slavery on the American continents was the open racial arguments used for the treatment of people from Africa (UNESCO 2007; UNESCO 2014b). This view was not unique to the capitalistic form of slavery that emerged here, but it came to have a determining influence on the world view that tried to justify slavery as a means to bring civilization to the Africans. Similar racist a itudes toward the population on the African continent developed among the intellectuals on Zanzibar who argued, “as a mark of having become more civilized, the coast population’s skin became clearer

78

•

Muted Memories

and brighter than that of ‘barbarian’ (washenzi), ‘that is, the bush-people of the African interior’” (Glassman 2004: 743n). It has been reported that slave women when freed were rubbed with ground sandalwood powder to make their skins so er and lighter (Eastman 1988: 15).19 It was even maintained that “slavery was the very institution that had done the most to civilize African ‘natives’ imposing the control necessary to channel their anachronistic energy into productive labour” (for his references see Glassman 2004: 746n111). In an ethos so central for UNESCO, the memory of the slave trade routes predominantly treating Africans as merchandise is used as a means to stimulate a discussion and create be er knowledge about what happened during this dark chapter of our history in order to promote global peace and understanding. With regard to this, the project has been successful in its a empts and completed a major achievement when slavery and slave trade were declared a crime against humanity during the World Confederation against Racism, Racial Discrimination, Xenophobia and related Intolerance, held in Durban, South Africa, in 2001. There has long been a conspicuous silence in the Western hemisphere on the slave trade in general and the involvement of specific countries and their trade companies in particular. But the trade and slavery is also an itchy memory in many African societies. Slavery was well-developed and integrated in many local societies, which will be further explored in chapter 6. This puts into question the convenient and popular image of foreign Arabs hunting for slaves in up-country societies that they then marched to the coast. Another reason for the silence regarding slavery is that the descendants of both slaves and slavers may live as neighbors who decline to talk about each other’s background. Another reason for the sensitivity about the slave trade and even its existence is a specific notion of manumission of slaves involved in the East African caravan trade. Only the owner could do this, o en a er the slave had bought his or her freedom with money earned from the caravan trade (Deutsch 2006: passim). This means that descendants of slaves, freed by the British, and sometimes but not very o en also the Germans, along the East African coast at the end of the nineteenth century, may still be seen today as slaves in the local context. Another reason for the sensitivity of the issue is that slaves and slavery still exist and are parts of the socioeconomic structure in many societies.20 The centuries-old conflict between the primacies of rights of property versus the primacy of human rights has not always and not everywhere been entirely solved.21 This is captured in the concept “rights in persons,” which will be further elaborated in chapter 6. It may be assumed that the insecurity that exists in many societies regarding human rights makes the

Fractures in the Image of Bagamoyo

•

79

memories of the evils of the past inconvenient or they represent an undesired way of life. The message wri en by a visitor in the Bagamoyo Catholic Mission’s Guest Book is evidence of this: “Proudly being a Tanzanian. May God bless us not to remember the evils of the past.”22 It may therefore be hypothesized that it was convenient23 for the delegates who composed Bagamoyo’s file for the Tentative List to align themselves with the already existing global image of slavery and slave trade, particularly as the 2005 evaluation of UNESCO’s Slave Route Project 1994–2004 had identified the need to include the slave routes on the Indian Ocean. Furthermore, the evaluation also states that “Tanzania has taken initiatives to set up a Slave Route memorial from the coastal town of Bagamoyo through the inland areas and all the way to Lake Tanganyika” (UNESCO 2005: 14). The proposed slave route theme seems to have been communicated to the evaluation team one year before Bagamoyo was introduced on Tanzania’s Tentative List. What could be more appropriate, the year a er the evaluation report had been published, than to nominate Bagamoyo as a memory of the central East African slave routes? The shi in focus from the comprehensive caravan trade route theme to a memory of the central slave routes was further nourished by the postconference consultancy support provided by the Institute for Resource Assessment (IRA), at the University of Dar es Salaam to the Ministry for Natural Resources and Tourism. The support was meant to facilitate the writing of the nomination file “for the inscription of Bagamoyo and the Slave Trade Caravan Route as a UNESCO World Heritage Site” (IRA 2007: section 3.1.13). The shi of perspective from the caravan trade in, among other things, ivory, iron, and slaves, to a more pronounced focus on slave hunting and trade, soon resulted in a global interest for Tanzania as a destination for “roots tourists.” It coincided in time with the decision by the African Diaspora Heritage Trail (ADHT), to arrange its fi h international conference in Tanzania in October 2009. The theme of the conference was to facilitate “An African Homecoming: Exploring Origins of the Diaspora and Transforming Cultural Heritage Assets into Tourism Destinations” (ADHT 2006). The decision to arrange the 2009 conference in Tanzania came a er an invitation from the country’s new Minister for Natural Resources and Tourism at the 2008 ADHT Conference in Bermuda (eTN Global Travel Industry News 2009). The main activity during the conference in Tanzania was to launch “Tanzania’s new heritage trail, ‘The Ivory and Slave Route.’” But the presentation is considerably narrowed from ivory and slaves to only emphasize slaves and Arabs. It is explained that the conference:

80

•

Muted Memories

will offer a rare glimpse into the Arab Slave Trade in Eastern Africa, a major part of worldwide enslavement of Africans that many of us in the West are not familiar with. From the Slave Markets of Bagamoyo (translated: Point of Despair) to the slave chambers of Mwangapwane Beach, our delegates will be able to [learn about the] barbarity of slavery and celebrate the struggle for liberation that is also part of Tanzania’s rich tradition. (eTN Global Travel Industry News 2009)

The processes of disembedding not only work through the a empt to meet the anticipated needs of Westerners to learn more about their origin, as already identified by the Minister for Natural Resources and Tourism at the 2002 International Conference in Bagamoyo, but also through the projection of slavery and slave trade onto Arabs, regarded as a partly external or foreign element. The focus on the role of the Arabs, with its roots in for example the writings of Livingstone, can also be found in the motivation of the nomination text from 2006, which explains that “[t]he idea is not only to protect the still visible remains of the dark past like the Arab Forts and other historic buildings or parts of the route that are existing,” (UNESCO 2006). The focus on Arab buildings contributed to disembedding the trade routes by neglecting the role slavery and the slave trade had played at the time of the caravan trade in African societies; this focus also lead to a reduction of the cultural route theme to focus on concrete buildings, as will be further explored in the next chapter. As part of the disembedding process o en inherent in heritage-making, the caravan trade was li ed out of its local and historical context only to be replaced by the global understanding of the Atlantic form of slavery: slaves put in iron chains by alien slave hunters and marched through the African landscape, or cha el slaves toiling under the sun on plantations. It seems like the Atlantic slave trade and the American form of slavery have been imprinted on the perception of slavery on other continents—what may be called “a common global culture on slave trade and slavery.” The choice of the slave route theme has been interpreted as an a empt to put focus on the relatively unknown and invisible central slave routes in East Africa, which may explain the weak arguments in the nomination text. The task of the nomination commi ee has been described as a “struggle . . . with the dominance of certain models of slavery, based on the West African experience or derived from popularly held notions of what slavery ought to look like” (Wynne-Jones 2011: 336). With Barbara Kirshenbla Gimble ’s (2004) words one may say that the East African caravan trade was absorbed into a “metacultural production” of slavery and slave trade. It is therefore important to re-embed the memory of the East African caravan trade to “provide a local understanding of slavery” (Wynne-Jones 2011: 337), which will be done in the second part of this work.

Fractures in the Image of Bagamoyo

•

81

The choice of the slave route theme not only meant the re-embedding of a global metastory, but also the creation of a unidirectional route: the marching of slaves from the interior of the continent to the coast. The role of Bagamoyo and the caravan trade as an entrepôt and vehicle for the spread of Islam, Christianity, and the Swahili language into the interior of the continent was le out in the branding of Bagamoyo. This was done in contrast to the recommendations from the 2002 Conference, which suggested a much more multifaceted presentation of the East African caravan trade than just a focus on the slave theme.

Neither Head nor Tail An international expert on world heritage-making maintained at the 2002 Bagamoyo conference that “if you choose to focus on the slave route then you should also seek some kind of cooperation with other countries to the west, and also Zanzibar.”24 Two of the main sites in the central caravan routes were the eastern part of Congo and Zanzibar. The Manyema area in eastern Congo probably provided most of the slaves traded on the central routes and, as the elephant herds decreased in the Nyamwezi area, an increased number of elephant tusks. Zanzibar, with its slave market, was the terminus for the slaves traded on the central routes before they were either put to work on the island’s plantations or were shipped away to foreign destinations and societies on the islands in the Indian Ocean or on the Asian continent and even to the Americas. Zanzibar was also the interface through which ivory from the continent was sold to India, China, Europe, and the Americas and cloth was shipped to the ports on mainland Africa. The commi ee working with the nomination text, including national experts and international consultants, identified the stretch on mainland Tanzania as the geographical area covered by the central caravan routes. The text repeatedly mentions the Bagamoyo–Ujiji route, from the eastern part of Tanzania to the western part as the object for nomination. For example, it is explained that the nomination concerns the “Ujiji–Bagamoyo route as a whole” and that “Bagamoyo served as the terminal which starts from Ujiji” and that “Ujiji was the last major trading center of the central Caravan Trade Route located on the shores of Lake Tanganyika” (UNESCO 2006).25 One of the reasons for nominating only the Ujiji–Bagamoyo part of this long-distance trade was, as explained during the Conference, that Tanzania’s Department of Antiquities and National Museums Service were already the custodian of the Livingstone memorial in Ujiji and of buildings in both Unyanyembe (Nyamweziland) and Bagamoyo. The first two had,

82

•

Muted Memories

as noticed by Wynne-Jones (2011: 320f), been inherited from the colonial predecessors and had been made into heritage to commemorate the history of “explorers” and missionaries and the struggle for the abolition of the slave trade. They can therefore only represent the memory of the slave trade itself in a restricted way. The same can be said about the museum at the Catholic mission station in Bagamoyo which is “based on an abolitionist narrative derived from outside, rather than on the experience of slavery and forced migration within Tanzania. It seems that they cannot fully achieve the type of sensitive, social objectives called for by the Slave Routes project” (Wynne-Jones 2011: 327). The selection of this particular stretch of the central caravan routes is reiterated when the Tanzanian Director of Antiquities as late as January 2014 explained that the slave trade caravan route in Tanzania is the Ujiji– Bagamoyo stretch (“Bagamoyo Aims for Heritage Status” 2014). It may be assumed that it was both convenient and strategic to adopt a more narrow national perspective at the expense of a more regional or global concern. Even the consultancy support mentioned above provided by the Institute for Resource Assessment (IRA), including a fact-finding journey along the central caravan route, only covered the stretch from Bagamoyo, via five up-country trade posts, Mamboya, Mpwapwa, Kilimantinde, Kwihara and Ulyankule, before the journey ended in Ujiji (IRA 2007). The central caravan routes were hereby reduced once again to the Bagamoyo– Ujiji part. The central caravan routes did not, as mentioned above, end in Ujiji but continued on the western side of Lake Tanganyika. One of the areas most severely affected by slave hunting and trade was the Manyema area in today’s Democratic Republic of Congo (B. Brown 1971; McCurdy 2006; Page 1974). While other societies along the routes, such as the Nyamwezi, seem to have been spared the devastating effects of slave hunting since they were important as producers of food and providers of porters, the Manyema seem to mainly have provided slaves. Sheryl McCurdy (2006: 445) writes that: “When Zanzibari traders and waungwana [their henchmen]26 a acked villages, the waungwana killed the adult men and captured the adult women, adolescents, and children. They forced some enslaved women and girls into harems and domestic roles; they selected other women, girls, and boys for the caravan trade.” The Arabs who se led in the trans-Tanganyika area had a devastating effect on local societies. It has been claimed that the caravan trade here had an unusual impact both with regards to “the smashing of the old communities and the creation of Swahili-dominated ones”, and the readiness of the Manyema who survived to be Swahilized and strive for a waungwana identity (Wright 1976: 553).

Fractures in the Image of Bagamoyo

•

83

The devastating effects of the slave raids on the Manyema people became so extreme, Page explains (1974: 73), because the Manyema “lacked not only a central political organization but also any well-developed centralizing force within their society.” This facilitated people from the coast to gain political control over the area. Tippu Tip, the infamous slave hunter and trader noticed, regarding the Manyema, that “there are no hereditary chiefs; into the country come folk from other areas and offer wealth (goods) to those of the land. These, in turn, make one of the strangers chief” (Tippu Tip [1974] 1958/1959: sec. 87). Tippu Tip was not late to utilize this and establish himself as a major slave trader in central Africa. His emporium was built on slave hunting carried out by his waungwana, mainly in the area west of Lake Tanganyika, and trade along the caravan route and at the slave market on Zanzibar.27 Slave hunting among the Manyema resulted in the collapse of their social organization (Page 1974: 74). It seems like this negative impact resulted in two different strategies for those who survived. Some young men, o en slaves themselves, became ruthless slave hunters, waungwana, providing their owners with new slaves, while many enslaved women established a new identity in the places they were brought to, like Zanzibar. The spirit possession cults found among the African diaspora in the western Indian Ocean region is associated with the forced migration that took place during the slave trade. McCurdy writes with reference to a spirit possession cult Manyema women performed in these trading centers that: Manyema ethnicity is a cloth woven by women and men with the thread of their lives in response to the devastation of their communities. As slaves, dispossessed homeless, and others from Manyema drew on common practices, rituals, and clothing to build relationships and construct a new identity, they became a dynamic and vibrant community that others noticed, sometimes despised, and o en emulated. (McCurdy 2006: 469)

A considerable number of the Manyema arrived in the coastal society as porters and slaves, but also people from the Nyamwezi area came to the East African coast through the caravan trade and se led there. It is, for example, estimated that three main ethnic groups lived in the Dar es Salaam area at the time of Independence. They were the Manyema who were descendants from freed slaves, Nyamwezi who were descendants from the caravan porters, and Zaramo, the original inhabitants of the land (Leslie 1963, as quoted in Hino 1980: 110). Actually, the Wanyema, slaves or descendants of slaves, originally belonged to some twenty different groups that lived in the geographical area called Manyema. When they arrived in new se ings like Bagamoyo, Dar es Salaam, and Zanzibar they were classified by others on the basis of their place of origin and merged,

84

•

Muted Memories

probably also by their own efforts, and formed a new identity (see Jerman 1997). Nothing could be er illustrate the magnitude of the caravan trade than the appearance of Manyema women from the Congo, establishing a new identity and carving out a space for themselves in Zanzibar society. Thus, the nineteenth century Central Slave and Ivory Trade route as a global heritage did not and cannot stop at Ujiji in Tanzania but has to be extended deep into what today is the Democratic Republic of Congo. Nor can it stop at Bagamoyo, but has to be extended also to Zanzibar and further into the Indian Ocean region. A representative of the Department of Antiquities once explained that UNESCO was not pleased with Tanzania’s nomination of Bagamoyo and the slave trade route to its Tentative List, since it did not include the whole route.28 The caravan trade along the central routes is an excellent example of a heritage that may serve as a national Tanzanian memory since it originated among the Nyamwezi in what is today central Tanzania and as a global heritage remembering an early phase of globalization and intercultural connections, but also the cruelty which characterized the slave trade period in our common history. The caravan sites, rest stops, and trading stations along the routes could be a national heritage for the memory of the early entrepreneurs who, during the nineteenth century, engaged themselves in a growing global trade. But it did not stop here. During the la er part of the nineteenth century the caravan trade expanded as new actors from the East African coast entered the scene as the demand for slaves increased around the Indian Ocean and the demand for ivory increased in the industrialized world. The caravan trade developed into a transnational and transcontinental cultural phenomenon of global concern. In similar vein, the caravan route was much more than Arabs hunting and marching Africans to the slave markets and plantations. The slave and ivory caravan trade as a cultural phenomenon ought to be re-embedded in its East African context and understood, not as a disembedded metacultural phenomenon replicating the stereotypes of the Atlantic forms of slavery and slave trade, but as a manifestation of early world trade systems and globalization.

A Final Surprise in Pandora’s Box? The last surprise in the process of heritage-making in Bagamoyo is that this heritage, more than ten years a er the inscription on Tanzania’s Tentative List, has not yet been formally nominated to the World Heritage List for reasons we are only slowly beginning to understand.

Fractures in the Image of Bagamoyo

•

85

Notes  1. Elderly people interviewed by Walter Brown in the late 1960s recalled the former spirit of Bagamoyo and its comfortable life; “good food, magnificent clothes and love adventures” (W. T. Brown 1970: 69).  2. There is a third interpretation (W. T. Brown 1970: 69) documented during the 1860s, which is “the entrance to the heart (of the continent),” which will be further explored in chapter 8.  3. For the reliability of the recording and interpretation of this song see also S. Fabian 2007b, 185 and note 91.  4. J. Sturtz and J. Wangemann accompanied, according to Fabian (2007b), the two DOAG officials Gravenreuth and Deinhard on their hunting expeditions.  5. T. N. Watson (1982: 37) gives a different version of the song: “Oh, what delight to see the ngomas [dances] where the lovely girls are swaying in dance at night in Bagamoyo.”  6. Author’s field notes, September 2002.  7. See chapter 10 for a discussion on the porous boundaries between the groups and the politics of identity in the coastal towns.  8. There are a number of sites of conscience on the List that commemorate the dark side of human history. The slave site at the Island of Goreé in Senegal was inscribed on the List in 1979 and the same year the concentration camp at Auschwitz-Birkenau in Poland, while the Hiroshima Peace memorial, Japan, and Robben Island Prison in South Africa were inscribed during the 1990s.  9. This can be understood by the fact that it was Kilwa, not Bagamoyo, providing Zanzibar with the bulk of slaves used on the plantations on the islands or sold further (Sheriff 1987). 10. They include: Bagamoyo: The Beauty at the Beach (Kajeri and Henschel, n.d.); The Two Worlds: Bagamoyo and Slavery in the 19th Century (Henschel, Ndunguru, and Kadelya 2000); and Descendant of Former Slaves and Slave Owners tell about Slavery in Bagamoyo (Henschel, Ndunguru, and Kadelya 2001). 11. A different figure is given in a German source from the 1890s indicating that approximately 30–40 percent of the population in the town were slaves—compared to 60–70 percent on Zanzibar or 75 percent in Malindi and Mambrui (Nimtz 1980: 40). 12. Sometimes known as The German Customs House. 13. The investigation was carried out under the supervision of Professor Felix Chami. See the publication Historical Archaeology of Bagamoyo: Excavations at the Caravan-serai (Chami et al. 2004). 14. Author’s field notes, March 2013. 15. The branding of the Caravanserai as a place where slaves were kept probably has its origin in a map at the Catholic Mission Station. The map, which predated the construction of the current structure, depicts the place as a “slave market.” 16. For further details on this trade see chapters 10 and 11. 17. See also chapter 5.

86

•

Muted Memories

18. Author’s field notes, 9 September 2002. 19. Eastman refers to Strobel 1983, 122. 20. A er the first year of a six-year long service in Tanzania, living in a posh area of Dar es Salaam, the author learned that a man in a neighbouring house, Jumamoshi Mtumwa, was a domestic slave. Jumamoshi’s status or place of residence had not changed at the end of the author’s service. 21. For an interesting debate on these rights in mid-nineteenth century Britain, see Hochschild (2005) concerning whether the right to property was superior to human rights or vice versa. 22. Entry dated 23 February 2013; author’s field notes, March 2013. 23. It may be assumed that slavery as a memory both irritates and hurts. Bagamoyo’s twin city Zanzibar Stone Town, on the World Heritage List since 2000, was the major slave market during the second half of the nineteenth century. However, this World Heritage has not capitalized on its potential as a Slave Route Memory, as noted in UNESCO’s evaluation of its Slave Route Project. 24. Author’s field notes, Bagamoyo Conference, 9–12 September 2002. 25. In a telephone interview with the former Swedish Director for Antiquities (26 February 2014), who is well experienced in the Bagamoyo process, it was mentioned that there could be a le er of understanding between Tanzania and the Democratic Republic of Congo, which opened up an avenue for Tanzania to move ahead since the Democratic Republic of Congo would follow when the documentation of the slave routes in Congo was available. 26. The waungwana identity will be further explored in chapter 10. 27. His house is today a tourist a raction in Zanzibar Stone Town. 28. Author’s field notes, March 2013.

4  World Heritage and Globalization The Bagamoyo Case Heritage-making entails contradictions, conflicts, and competition since it is essentially about the mobilization and exercise of power over objects and subjects at global, national, and local levels. World heritagemaking in the modern or even post- or hyper-modern context concerns, as shown in a previous chapter, both formulating a globally recognized and accepted memory and, at the same time, the production of meaning at the local level. The heritage presented to the global audience, both experts and tourists must carry an aura of “outstanding universal value” and exert a performative power over the inhabitants in such a lived-in heritage as Bagamoyo’s Stone Town in order to meet global acceptance and expectations. The inhabitants of Bagamoyo are expected to appreciate and take care of their heritage and to share it with a global audience if their town is placed on the World Heritage List. This chapter focuses on global, national, and local efforts to have Bagamoyo inscribed on the World Heritage List as a representative for the nineteenth century East African slave routes. More correctly, it is the slave routes along the central caravan routes and “The Slave Port of Bagamoyo and Associated Sites and Routes of Memory,” as shown in the previous chapter, central nodes in this transcontinental trade, which were nominated to Tanzania’s Tentative List.

88

•

Muted Memories

Bagamoyo on the Global Scene Bagamoyo, today a small town was once the major port for the caravan trade that covered much of what is today Tanzania, Rwanda, Burundi, and eastern Congo. It was the interface in a global trading network that included the interior of the African continent, the societies around the Indian Ocean and South China Sea, Europe, and America. The International Conference held in Bagamoyo in 2002 evaluated, as reported in a previous chapter, its prospects to be nominated to UNESCO’s World Heritage List. A er the conference, Bagamoyo was once again placed in the global arena when, in 2006, Tanzania put this coastal town on its Tentative List to become a World Heritage site in the hope of being inscribed on the World Heritage List (WHL).

Changes during FiĞy Years of Heritage-Making in Bagamoyo Previous chapters in this volume, spanning the period from the 1950s to 2010, have focused on three developments or transformation in the efforts to conserve and present Bagamoyo as a cultural heritage site. The first concerns the redefinition of the object; from the efforts during the colonial period of the 1950s to protect and conserve individual objects such as carved frames and doors of the stone houses in the town, via the protection of specific houses and milieus in what is called Bagamoyo Stone Town to the transformation of the presentation of Bagamoyo, as a result of the 2002 conference, as a proposed Slave Trade/Slave Route site together with associated sites of memory along the central East African caravan routes.1 The second transformation is related to the question of who is concerned by Bagamoyo as a cultural heritage site. This process took the form of inviting and incorporating new stakeholders: from the 1950s, when the experts’ interests were focused on the conservation of doors, frames, and buildings, via the invitation during the 1970s and 1980s of experts in community development planning and the people who actually live in the old Stone Town of Bagamoyo, who were seen both as the custodians of and the beneficiaries of the heritage. The latest redefinition of the stakeholders or guardians of the heritage of Bagamoyo took place during and a er the 2002 International Conference when the global public, in the form of development experts and tourists in general and “roots tourists” from the Western hemisphere in particular, became prominent stakeholders. The experts and scientists, who saw the universal values of Bagamoyo (and associated sites) as a memory, and international development partners, who saw the cultural

World Heritage and Globalization

•

89

values of Bagamoyo as a vehicle for development and poverty reduction, also became active agents in the process of heritage-making. The last transformation concerned the redefinition of the “memory,” or what Bagamoyo represents: from the original focus on the cra smanship behind man-made objects, like doors and frames carved before 1940, via a focus on Bagamoyo as a specific place with historical buildings of Arabic and Indian origin existing side by side with German colonial architecture and its particular genius loci, to the launching of the historical site as a part of World Heritage as the terminus in the East African slave trade route. The “memory” was hereby transformed from representing a specific localized cra smanship, via a lived-in se lement representing a vibrant localized port town, to Bagamoyo as a disembedded heritage together with associated sites representing the memory of an inter- and intracontinental trade during an early phase of globalization. One of the reasons for the latest transformation of the question of “what to remember” was captured by Fabian Kigadye, the conservator at Tanzania’s Department of Antiquities, when he said that the first ideas to nominate Bagamoyo dated back to 1987, when the focus was on the entire township, but today:2 “many of the historic buildings are no more than ruins now while others have been vanished completely. In 2002, during a conference of international experts held in Bagamoyo, it became obvious that the township alone would not meet all requirements to be listed as a World Heritage. As a result, the whole idea was extended to the entire trade route” (Bothe 2003). As a global heritage, Bagamoyo had to be packed and presented in a way that could be understood on the global level. We see here a universal process of continuous change in the identification and presentation of a heritage, as described by Gil-Manuel Hernàndez I Martí in the following words “It could be described as a social construction, understood as a symbolic, subjective, processual and reflexive selection of cultural elements (from the past) which are recycled, adapted, refunctionalized, revitalized, reconstructed or reinvented in a context of modernity by means of mechanisms of mediation, conflict, dialogue and negotiation in which social agents participate” (Hernàndez I Martí 2006: 95). The processes behind the reinterpretation and reformulation of the memory a specific site or object represents are captured on the global level by Jean Musitelli: “Heritage finds itself being bestowed with meanings and goals that go beyond the simple imperative of conservation, making it, alternatively or simultaneously, a sign of identity, an engine of development, a tool of cooperation, or a vehicle of tourism” (Musitelli 2002: 330). The quotations above, which identify the social construction of heritage, its ability to change and conform to global processes, and its poten-

90

•

Muted Memories

tial to link with the tourist sector, will set the stage for the investigation below into the interconnections between heritage-making and globalization and the presentation of Bagamoyo as a representative for East African slave routes.

The World Heritage List The dynamic character of heritage-making is the result of the many paradoxes inherent in processes of globalization. One such paradox is the tricky balance between universality and particularity in packing and presenting an item as a world heritage. Another is that heritage-making in Bagamoyo is both an effect and the cause of globalization. Yet another paradox concerns the heritage’s capacity to both stir up nationalistic feeling and to link to wider global processes. Lastly, heritage-making may, on the one hand, give the already dead a second life, and on the other hand, freeze ongoing sociocultural processes into standardized forms of conservation. All this calls for a closer exploration of the connections between heritage-production and globalization. Globalization is not a new phenomenon. The Nyamwezi traders/porters (see the next chapter) who during the nineteenth century marched by the thousands from the interior of the continent to the Indian Ocean with their merchandise were agents in the globalization process. On their way back, they brought goods from other continents into the heart of Africa and the ever-increasing demand for cloth on the continent ignited the technological revolution in remote locations like Salem and Boston in the United States and Bombay (Mumbai) in India (Prestholdt 2004). It has been suggested that the World Heritage endeavor can be seen as a “somewhat romantic act of resistance” (Brumann and Berliner 2016: 3) since it o en focuses on the demarcation of very localized and welldemarcated sites in a time of globalization. It can also be argued, however, that heritage-making is not only an effect and a force behind globalization but also at the same time a cause that contributes to nationalistic ambitions and pride, and an effect of nationalistic projects (see also Meskell 2015). The interconnections between national and global heritage and the factors that contribute to the transformation of a national heritage into a world heritage are captured by Arizpe in the following way: “the value given to cultural heritage will depend on the meanings that are chosen among those constantly travelling along a web of cultural exchanges and recombinations. At present, as never before, trade, globalization, migration, and tourism, as well as telecommunications and telematics, are rapidly adding more and more exchanges to that web” (Arizpe 2000: 32).

World Heritage and Globalization

•

91

This raises questions concerning the relationship between and forces behind nationalism and, in particular, the “national cultural heritage” created, recreated, and utilized in the construction of a “national identity” on the one side, and on the other, the heritage identified as a global heritage, when UNESCO’s member state nominates an object of national heritage to be accepted as a part of the world heritage and inscribed on UNESCO’s WHL.

Universality versus Particularity During its forty years of existence, the WHL has gained iconic status. But the WHL and the sites on it have also been contested for being a primarily Eurocentric creation with li le credibility and legitimacy outside the North Atlantic cultural and political sphere. Such challenges motivate an investigation into the origin and current operation or utilization of the concept of “our common heritage,” and the role of the WHL. The proliferation of the message, reported above, that Bagamoyo is a site of World Heritage, may be explained by what Marc Askew calls “the magic list of global status” (M. Askew 2010: 19), or what Lynn Meskell (2012) calls the “rush” among government parties to have their heritage sites and monuments inscribed on the list. The a raction for a state to nominate an item of its heritage for inscription as World Heritage must to a large extent be understood from the seemingly conflicting desires to both be able to exhibit its glorious past but also to be both modern and part of the contemporary global world. It should also be acknowledged that the WHL functions as a means to exchange values from one sphere to another; from cultural or natural heritage to tourism and economic development. A national cultural heritage such as East African caravan trade along the central routes may be transformed into a global heritage through the ascription of an “outstanding universal value”—a criteria for a site to be inscribed on the WHL. This concept is explained in UNESCO’s Operational Guidelines (2012) with the words: “Outstanding Universal Value means cultural and/or natural significance which is as exceptional as to transcend national boundaries and to be of common importance for present and future generations of all humanity” (para. 49).3 These values are expected to “float” above or beyond their historical, political, and cultural contexts and, in an essentialist way, as the objects or sites inscribed on the list are assumed to carry this “outstanding universal value” as an inherent quality. The idea that a “heritage” is inherently valuable in its own right has been challenged by many. Laurajane Smith (2006), following Barbara Kirshenbla -Gimble (2004), is perhaps the most outspoken in her constructivist position by pointing out that it is the present-day cultural pro-

92

•

Muted Memories

cesses that make a specific heritage valuable (compare M. Askew 2010: 31). The previous chapter demonstrated how the memory of the East African caravan trade was replaced by a global metacultural understanding of the Atlantic slave trade. Another issue which is related to the ideas about “our common heritage” is the question if there are strong arguments for a “cultural heritage internationalism,”4 which claims that certain heritages, through processes of deterritorialization or disembedding (Giddens 1990: 21–28), belong to the entire humankind or are the principle heirs of the heritage always localized in space?5 The genealogy of the concept of a World Heritage that belongs to humankind as a collective goes back at least to the end of World War II and the Nuremberg Trial. The trial addressed, among other crimes, the purposeful destruction of European cultural heritage and declared that the destruction of this heritage was a violation against humanity. The idea of a global heritage was expressed in 1954 in a much-cited paragraph in the Hague Convention for the Protection of Cultural Property in the Event of Armed Conflicts: Being convinced that damage to cultural property belonging to any people whatsoever means damage to the cultural heritage of all mankind, since each people makes its contribution to the culture of the world; Considering that the preservation of the cultural heritage is of great importance for all peoples of the world and that it is important that this heritage should receive international protection. (UNESCO [1954] 1999: 1)

The fact that the vision of our common heritage grew out of the devastating wars during the twentieth century stimulated thoughts about a heritage that belongs to all of us, and highlighted the fact that the heritage most important to protect and conserve was the heritage seemingly most exposed to war activities, i.e. monuments and architecture. This background is certainly one of the reasons for what has been called the Eurocentric view on what is regarded as “outstanding” and the dominance of European monuments, so exposed to destruction on this war-ridden continent, on the WHL. In its work with the WHL, UNESCO has to balance the promotion and protection of universalistic concepts of a common shared cultural heritage against another main responsibility: to promote and protect the “right to culture” and defend cultural diversity (M. Askew 2010: 23; see also Eriksen 2001). The responsibility to promote cultural diversity is captured in the Preamble to the Constitution (UNESCO 1945) with the following words: “[T]o develop and increase the means of communication between . . . peoples and to employ these means for the purpose of mutual understanding and a truer and more perfect knowledge of each other’s life.”

World Heritage and Globalization

•

93

UNESCO, like other organizations in the UN family, is tasked with the duty to promote global understanding and peace, and a means to achieve this is through a be er understanding of and respect for alternative ways to organize human life. It is an understatement to say that it is tricky to balance universal values against cultural diversity, as noted by Thomas Hylland Eriksen (2001) and Askew (2010). Eriksen (2001) shows how the UNESCO document Our Creative Diversity (1996) oscillates between the propagation for a universalistic global ethic on the one hand and, on the other, the particularistic cultural variety and “the right to culture.” A culture in this perspective is seen as “something which belongs to a particular group of people, something associated with their heritage and their ‘roots’” (Eriksen 2001: 5). This opens up processes in which “the culture” may be used by one group to exclude other groups.6 The defense of cultural diversity o en provides fuel for politics of identity and processes of exclusion. It has been noticed that heritage-making for nationalistic purposes implies the exercise of power that o en leads to frictions, fractures, and resistance. The cultural heritage, writes Uffe Juul Jensen (2000: 43) “can be—and o en is—constructed to support the activities or dominance of powerful groups or nations at the expense of other groups or nations.” Nationalistic endeavors, including national heritage-making, express exclusiveness and create or demand processes of exclusion. Herein lays one of the paradoxes or contradictions in heritage-making and the vision of our common heritage. There is a growing tendency in today’s world to utilize UNESCO and the WHL to promote nationalistic interests and to gain global recognition as a part of the particular form of modernity (Meskell 2014: 218) that developed in the Western hemisphere. This politization has been seen as a direct result of the composition of the World Heritage Commi ee. Its twenty-one elected members represent their own nations and, consequently, may use the commi ee to push the interests of their own nation. It may be more tempting for a member of the Commi ee to promote national interests, not least in terms of a potential increase in global tourism, than to work for the identification of what is outstanding and universal on the global arena and to use the heritage to promote mutual understanding and peace (see Meskell 2013: 489). The work of UNESCO during the last decades has been characterized by constant challenges from member states that use the organization to promote nationalistic interests. As expressed by one of the experts interviewed by Christina Cameron and Mechtilda Rössler (2013: 166): “li le by li le we witnessed the arrival of diplomats, ambassadors, ministers, people who were sent by their countries, who transformed the course of the

94

•

Muted Memories

World Heritage Convention and who accentuated . . . the political aspect, perhaps not in the best sense of the word, of the Convention.” The work to promote the universalistic view of our common heritage has been increasingly challenged during the last decades by member states that work for the promotion of their own nationalistic interests. The politicization of the nomination and inscription procedures has resulted in a situation where members of the World Heritage Commi ee make pacts with each other in ways that lead to diplomatic decisions above the heads of the experts in UNESCO’s advisory boards (Cameron and Rössler 2013). This creates a magnificent challenge against what the WHL stands for and represents and how the criteria of “outstanding universal value” should be interpreted. As a consequence of the power balance in today’s world, it may also complicate the efforts to promote more inscriptions on the WHL from resource poor countries and continents.

Standardization as a Globalizing Force One of the most powerful forces behind both nationalism and globalization7 is the process of standardization (Anderson 1983; Eriksen 2007; Sassen 2006; and Scholte 2005).8 To standardize is to homogenize the particular, as in UNESCOs work with the sites inscribed on the World Heritage List. Efforts to standardize are, however, expressions of power, which may give rise to many forms of resistance. Two hundred years ago, Benjamin Constant expressed his resistance to Napoleon’s a empts, o en by military means, to create uniformity over large parts of Europe using French standards with the following words: The conquerors of our days, peoples or princes, want their empire to possess a unified surface over which the superb eye can wander without encountering any inequality, which hurts or limits its view. The same code of law, the same measures, the same rules, and if we were gradually to get there, the same language; that is what is proclaimed as the perfection of the social organization . . . the great slogan of the day is uniformity. (Constante 1804)

In his book Seeing Like a State (1998), the political scientist James Sco develops the brilliant but challenging metaphor of the scientifically managed German forest. The dawn of modern society saw the emergence of many scientific disciplines and professional experts. One of them is the ForstwissenschaĞler or forest scientist—an expert in forest management, whose task was to calculate and regulate production, yield, and regrowth. Sco writes: The fact is that forest science and geometry, backed by state power, had the capacity to transform the real, diverse and chaotic old-growth forest into a

World Heritage and Globalization

•

95

new, more uniform forest that closely resembled the administrative grid of its techniques. To this end the underbush was cleared, the number of species was reduced (o en to monoculture), and planting was done simultaneously and in straight rows and on large tracts. (1998: 15)

Power was exercised over nature in the name of science and productivity: “The German forest became the archetype for imposing on disorderly nature the neatly arranged constructs of science” (1998: 15). A similar grid of standardization and uniformity is imposed over objects nominated for UNESCO’s World Heritage List. Rules and regulations determine what may grow in a Prussian forest or what may be done in a conservation area, when habitat and habitus is converted into heritage (Hafstein 2012: 510). The process of heritage-making implies a high degree of standardization of the “unique,” as the object is identified and circumscribed by technical and juridical language. The object is identified and demarcated in its local context through the use of global language, measurements, and expert knowledge; a process of disembedding, which we will return to below. Furthermore, as a part of World Heritage, a place like Bagamoyo has to make itself accessible to the global public. It is expected to be able to manage an increasing flow of tourists who arrive to consume what is supposed to be a unique experience encountered already in the global marketing of the heritage (Salazar 2010: 132) and adds fuel to the disembedding or “li ing out” of the heritage from its local context.

World Heritage-Making as a Globalizing Force UNESCO balances, as already mentioned, between on the one hand protecting and promoting cultural diversity against what is seen as the devastating and destructive forces of globalization, mainly understood, in the view of UNESCO, as economic globalization, commodification, and homogenization, and on the other hand, protecting and promoting the idea of a common heritage, universal values, and global ethics. This has caused contradicting views among various researchers regarding UNESCO’s own role in the globalization processes.

UNESCO, Power, and Globalization In his article “Globalizing Heritage,” William S. Logan (2002) asks if UNESCO’s World Heritage practices and procedures have caused a “cultural globalization.” Focusing on the work of UNESCO, ICCROM (International Centre for the Study of the Preservation and Restoration of

96

•

Muted Memories

Cultural Property) and ICOMOS (International Council on Monuments and Sites) he notes that: these organizations continue to play a powerful role on the global scene, laying down international standards for professional practice—“world’s best practice”—in the cultural heritage field as well as influencing thinking in those fields in less direct ways. The activities of UNESCO and its associated bodies are said to be a empting to impose a common stamp on cultures across the world and their policies creating a logic of global cultural uniformity. They seek to impose standards of “good behavior” onto Member States and other states. (Logan 2002: 4)

This, he argues, is a result of the organization’s modus operandi, which includes (a) developing “best practice” codes, (b) defining the values of cultural heritage, and (c) establishing standardized management practices in World Heritage sites. All these components include strong prescriptive normative forces through the concepts of “best practices” and “good behavior.” It should be noted, however, that Logan is eager to also question the role of UNESCO as a globalizing force. He maintains that the power, which the “center” exercises over the “periphery,” is counterbalanced by UNESCO’s capability to constantly adopt new influences and initiatives from the member states in the “periphery.” An example is the adoption of the Nara Document on Authenticity from 1994, which challenges the predominantly European views on preservation and authenticity. Marc Askew (2010: 20) delivers a different argument in an a empt to explain the limited capacity of UNESCO being a globalizing force. Due to its constitution and the role given to the member states, UNESCO has been hijacked by the members since “the globalized and institutionalized heritage system has not overcome nation-state-based power structures and nationalist agendas, but has rather enhanced them, and this severely compromises the ideal of forging a countervailing meta-national zeitgeist evoked by the term ‘World heritage’” (Askew 2010: 20). The power of UNESCO may seem limited in the struggles between the organization and the sovereign member states if only the “first face” or dimension of power (Lukes 1974) is considered, i.e. that UNESCO’s capacity to make formal decisions that are against the interests of the member states is strongly restricted. The organization, as other UN organizations, adheres to the Westphalian tradition (Chemillier-Gendreau 2002) of the sovereignty of the territorialized state that, in combination with the organization’s strategy to work through diplomacy and consensus, makes UNESCO sensitive to the wishes and whims of member states. However, there are other sources of power available to the organization. One is to put a World Heritage site on the “danger list,” which means

World Heritage and Globalization

•

97

it could eventually be erased from the list. At the time of the nomination of Bagamoyo to Tanzania’s Tentative List, no less than 40 percent of the sites put on the List of World Heritage in Danger were found on the African continent (Boswell 2008: 23). They had problems meeting the UNESCO standards regarding management and protection. The high proportion certainly have multiple causes as a country’s inadequate competence or resources to protect the heritage, conflicting local views on how an area or the specific structures shall be used, or lack of a genuine ownership of the heritage when decisions are made far away from those who, for example, live in Bagamoyo Stone Town. Another form of power is exercised when UNESCO, its staff, and affiliated experts declare that a nomination file is incomplete or is of substandard quality and therefore recommend and/or demand amendments. This is the exercise of the second dimension of power, the “nondecision-making” power (Bachrach and Baratz 1962), which implies that an issue or question is obstructed from entering the agenda for decision-making. This has been an effective tool through which a member state that tries to nominate its heritage may be disciplined, its nomination dossier standardized and eventually, if successful, the heritage becomes a part of “our shared global heritage.” This source of power is built into the procedures of the organization. Its advisory body in the field of cultural heritage, ICOMOS, established in 1965, provides evaluations of properties proposed by member states to be inscribed on the WHL and inspects the status of properties already inscribed. An evaluation team may make four different recommendations: Inscription or “go ahead”; Referral, which is a request for minor editing of the proposal and additional information; Deferral, which is a request for major editing; and, finally, Noninscription.9 This dimension of power was experienced in Bagamoyo a er the government had put the town and the caravan routes on its Tentative List. Early evaluations showed that the dossier was incomplete. One part missing was maps showing the central caravan routes and the “associated sites.” Another was an information center in Bagamoyo.10 The further processing of Bagamoyo as a World Heritage site had to wait until the gaps were filled, illustrating UNESCOs “nondecision” power. Ten years a er the nomination, these gaps have not yet been filled and Bagamoyo as a potential site of World Heritage has not been put on UNESCO’s agenda for the final decision. Another power strategy o en used by UNESCO is what Steven Lukes calls the third face of power, i.e. to instill UNESCO’s ideology, aims, interpretations, and procedures in member states and help them adopt UNESCO’s interests as their own.11 This is achieved through, for exam-

98

•

Muted Memories

ple, “capacity building” and “technical assistance” extended to member parties. Once established, the interest shared between UNESCO and a member state is maintained through self-discipline generated by the rules and regulations, which accompany an inscription together with the constant monitoring, carried out by UNESCO and its advisory bodies. This is a most suitable and elegant form of power exercise for an organization, which searches for, and works through diplomatic and consensus solutions (Meskell 2013: 490). This kind of power has been exercised in Bagamoyo since the end of the 1970s. The report “Areas of Concern: Bagamoyo, Tanzania” produced by a UNESCO consultant (Watson 1982) both identified the historical buildings worth protecting and preserving and suggested external support to capacity-building in conservation techniques. The 1983 seminar/ workshop and a new consultancy report “Proposed Center for the Conservation and Restoration of Cultural Property at Bagamoyo” (Russell 1980) detailed the need for capacity development and the dissemination of “best practices” to East Africa. During the 2002 International Conference, both international and national experts orchestrated the discussions to make sure that the decisions and recommendations were in line with UNESCOs policies, priorities, and practices. The shi of presentation from “The Caravan and Slave Trade” to “The East African Slave Route” may be seen as an aligning with UNESCO’s interests by using the form of its “Slave Routes Project,” which, from its inception, should also include the Indian Ocean. To be on the WHL implies compliance with the interests and demands of UNESCO, which may be called the power of the WHL.12

The Globalizing Power of the WHL With regards to similar experiences and in contrast to Logan and Askew, researchers have identified and pointed out the globalizing role of organizations such as UNESCO. Ulf Hannerz (1991, 1992, 1996), for example, identifies the role of international organizations in the establishment of what he calls a “metaculture of modernity” or the “global ecumene of modernity,” which may facilitate the understanding of one of the paradoxes in heritage-making: the relationship between nationalism and globalization. It is o en assumed that nationalistic and globalizing forces oppose each other since nationalism is sometimes understood as a reaction to globalization. It has been shown, however, that they share the same underlying forces, for instance, standardization.13 Herein lies one of the contradictions in heritage-making: the national heritage may be used to acquire global recognition when it is:

World Heritage and Globalization

•

99

constructed with reference to an ideal of human flourishing that has been recognized by various, o en opposite, traditions and communities throughout Western history. That ideal also plays an important role in great parts of the world today, among groups and communities outside the Western world who are struggling to ensure self-determination and to become respected members of the world community. (Jensen 2000: 43)

To be respected in the “global ecumene” implies compliance with established and internationally accepted rules, roles, and standards, one of the strongest forces behind globalization. One of the ways to achieve this and to gain recognition as a member in the modern world is to have sites on the World Heritage List,14 as has been dramatically experienced during recent years through the purposeful destruction of such heritage to demonstrate a rejection of belonging to that world. It has been shown that the purposeful destruction of Sufi tombs and mosques in Timbuktu, Mali (Joy 2016), was more about the future than the past (Brumann and Berliner 2016: 16); i.e. to disembed Timbuktu from a globalizing world and dictate a different future for the town. Both acts may seem paradoxical: the national heritage, surrounded by an aura of antiquity is used as evidence that a particular state belongs to the modern civilized world or, alternatively, this heritage becomes associated with the modern Western world and should therefore be destroyed. Besides UNESCOs globalizing power through the WHL, Hernàndez I Marti (2006: 99–100) adds the following globalizing forces and tools in UNESCO’s work: (a) the concept applied by UNESCO of a cultural heritage is a modern Western invention exported to other parts of the world in what is called a “westernizing project”; (b) the assets inscribed on the WHL have been deterritorialized and merged with other assets to make up a “common global culture”; (c) the establishment of institutions, an organizational setup and the deployment of experts have helped “spread a homogeneous conception of what should be understood as heritage, and of how this heritage has to be studied and preserved”; and, finally (d): there is a series of standardized patrimonialized formulas and categories which have spread all over the world: cultural asset, museum, eco-museums, theme park, . . . a set of formulas and instrumental categories which homogenize the definition, classification and management of cultural heritage throughout the world, as happens with the devices for management, conservation, protection, definition, evaluation, commercial exploitation and categorization of cultural heritage. (Hernàndez I Martí 2006, 100)

Many of the commentators referred to above point out that UNESCO’s interest in list making and standardization of procedures and presentations are prominent parts of wider processes of globalization. This is captured in Hernàndez I Martí’s third issue above, i.e. the organizational

100

•

Muted Memories

set up and dissemination of a standardized concept of heritage and procedures for its preservation. This effort to standardize became apparent during the 2002 International Conference in Bagamoyo and took the form of a long list of plans and documents identified as essential parts of the nomination dossier. These include (the list is not exhaustive): a Land Use Plan, a Conservation Area Plan, a Tourism Plan, a Tourist Facilities Plan, an Infrastructure and Service Plan, a Visitors Facilities Plan, a Restoration and Maintenance Plan, a Monitoring Plan and any additional plans needed for the specific World Heritage Site (see, for example, Kyessi 2002). The plans for Bagamoyo as a World Heritage site are meant to cater to various needs. They can be categorized in plans for the protection and maintenance of the heritage (three plans), plans for the utilization of the heritage (four plans), and a plan for the surveillance of the local and national performance. It is a mixture of plans reflecting UNESCOs desire for surveillance through follow-ups of the town’s performance as well as Bagamoyo’s wish to a ract an increasing number of tourists. Through this procedure, Bagamoyo’s cultural heritage is cast in a standardized form for how a World Heritage should be understood and presented. The requested and required plans and documents not only explain how the heritage should be preserved, but also how it should be utilized and made accessible to a global public, which in turn is supposed to stimulate economic growth, and at the same time monitor compliance of UNESCO’s rules, regulations, and procedures. There is no doubt that many of these plans are urgently needed, but UNESCO’s insistence on their existence and on their form and content is a strong standardizing and globalizing factor in today’s world. A bureaucratic grid is rolled out over the place and the life of the Wabagamoyo, the people of Bagamoyo; it is regulated and surveyed, like the trees in the scientific German forest. Through its role in World Heritage-making and its work with the WHL, there seems to be no way UNESCO can avoid being a powerful globalizing force. The globalizing power of UNESCO is explained and summarized by Jan Turtinen (2006) when he points out that the organization not only has the capacity to mobilize experts and decision-makers around the task to conserve and preserve the heritage, but has also managed to steer the representational processes and hereby impose its own interpretation in form of the “meaning” that shall be assigned to the specific site.15 In the case of Bagamoyo this is seen in the shi of perspective from a memory of the East African caravan and slave trade to a slave trade/slave route memory.

World Heritage and Globalization

•

101

World Heritage as an Effect of Globalization A few years a er the establishment of the World Heritage List in 1978, it was discovered that the sites inscribed were located in a restricted number of countries and continents. Furthermore, heritage from “early” or “ancient civilization” in the Mediterranean and Near East area and pre-Hispanic times in Latin America dominated other historical époques. Areas that had produced large-scale physical structures dominated the list, in particular European architectural monuments, o en with a Christian background. Henry Cleere from ICOMOS noted in his speech at the 2002 International Conference in Bagamoyo that “what these all have in common is the monumental nature of their cultures, manifested in historic towns, in architectural masterpieces, and in massive archaeological remains” (2002: 3). In short, the dominance of monuments in combination with the inflation of the European heritage raised arguments and accusations concerning the list’s failure to represent all continents. In 1993, almost half of the sites on the list were located in Europe or North America compared to less than 10 percent from the African continent (Ellio and Schmutz 2012: 269f.). This underrepresentation of the African continent has been interpreted as a continuation of the devaluation Africans and the continent faced during the slave trade and colonization. This imbalance is not very surprising since UNESCO grew out of the devastating wars on the European continent and the damages they caused on the cultural heritage there. The situation, however, was embarrassing for an organization, which works for all people on earth and nurtures the idea of a creative cultural diversity. Therefore, in 1994, UNESCO adopted its “Global Strategy for a Representative, Balanced, and Credible World Heritage List” in order to broaden the representation. One way to achieve this was to identify new categories, such as outstanding industrial heritage, cultural landscapes, underwater heritage, coastal, marine and small island sites, and deserts, caravan, pilgrim, and trade routes.

Heritage-Making on the African Continent The new category called “cultural landscape” was introduced in the Operational Guidelines since this “opened the World Heritage Convention not only to new types of sites, but also to regions of the world which were not able to present nominations under the previous criteria” (Saouma-Forero, Hö , and Rössler 2000: 5). In other words, previous conventions and guidelines had excluded certain parts of the world from contributing to

102

•

Muted Memories

“our common heritage,” since the dominant view on what constitutes heritage with “outstanding universal values” had favored heritage on the European continent. The same kind of reasons can be found behind the introduction of another new category called “intangible heritage,” which may be seen as a remedy to this. So far, the countries on the African continent had been much more successful in nominating candidates in the category of “natural heritage,” so the new category of “cultural landscapes” was supposed to provide a window for new inscriptions. However, twenty years a er the launching of the Global Strategy, no less than forty-three sites, or 47 percent of the ninety-one African sites on the WHL, were natural sites and three mixed—both nature and culture—as compared to the global scene where no less than 77 percent were cultural sites while less than 20 percent were natural sites (UNESCO 2013). Hence, the original design of categories and the Operational Guidelines of the WHL contributed to perpetuate the view of Africa as more “nature” than “culture.”16 The introduction of the new category of “cultural landscape” was supposed to be a remedy to the imbalance since this category is seen as a result of the combined work of man and nature. Dawson Munjeri17 explains: “Africa’s scenario is one where there is a symbiotic relationship between the ecosystems and the ethnosystems. This is so because more than anywhere else, the cultures of Africa have evolved out of nature and still draw their authority from it” (Munjeri 2000: 37). This is a gross generalization and unjustified characterization of both the African continent and other continents. It is clear, however, that the adoption of this new category was meant to assist African countries, as well as Australia, to be able to nominate new candidates for the list. As a byproduct, however, it perpetuates the notion that Africa is more nature than culture since this new category was needed to assist African countries in taking part in global heritage-making or, in Hannerz’ words, become members of “the global ecumene.” No less than five regional meetings were arranged on the African continent in order to increase the number of sites on the World Heritage List between the time of the launching of the Global Strategy in 1994 and the Bagamoyo International Conference in 2002. Progress, however, was slow and the ICOMOS representative at the 2002 Bagamoyo Conference had to confess, despite efforts to correct the imbalance in representation between different continents that “in June 2002 there were only 27 cultural sites and monuments in sub-Saharan Africa out of a total of 586, or around 5 percent. This figure compares with 318 in Europe, 99 in Asia, 67 in Central and South America, and 56 in the Arab states. Only the Oceania/Pacific region is less well represented than Africa” (Cleere 2002: 1).

World Heritage and Globalization

•

103

UNESCO has also developed a number of route themes to facilitate inscriptions on the WHL. The previous chapter showed that the slave route theme was chosen for the branding of Bagamoyo. This was done at the expense of another route theme that was discussed during the 2002 International Conference. Other categories of routes are collectively known under the label “Caravan, Pilgrimage, Cultural and Trade Routes,” which open up both for the identification and promotion of intangible heritage as a trade route’s importance for the regional and global interaction and the movement of people, and exchange of goods, ideas and cultural elements as well as to provide an opportunity for a number of countries crossed by such routes to cooperate in order to formulate a nomination together.18 The category “cultural routes,” or “heritage routes” as they first were called, was discussed in 1984 at a World Heritage Commi ee meeting in Madrid on “Routes as a Part of Cultural Heritage.” The theme was proposed as a new concept among the different cultural heritage categories for inscription on the WHL. It was further elaborated in 2003 in a new meeting in Madrid in connection with a revision of the Operation Guidelines. A cultural route is, according to ICOMO’s definition: a land, water, mixed or other type of route, which is physically determined and characterized by having its own specific and historic dynamics and functionality; showing interactive movements of people as well as multidimenssional [sic], continuous and reciprocal exchanges of goods, ideas, knowledge and values within or between countries and regions over significant periods or time; and thereby generating a cross-fertilization of the cultures in space and time, which is reflected both in tangible and intangible heritage. (ICOMOS-CIIC 2003: 3)

Different types of cultural routes have been identified. Some have been established with a specific aim, such as the Inca and Roman Empire routes. Others, such as the African trade routes or the Asian Silk Route, are the result of a “long evolutionary process in which the collective interventions of different human factors coincide and are channeled towards a common purpose” (ICOMOS 2005: 2). The identification of new themes opened up for countries not characterized by having monumental heritage to become members in the club of countries on the WHL if a number of states joined forces in heritagemaking. Fabian Kigadye’s statement at the 2002 International Conference in Bagamoyo that “it became obvious that the township alone would not meet all the requirements to be listed as a World heritage. As a result, the whole idea was extended to the entire trade route” (Bothe 2003) should be seen in light of this. The route theme was presented by national and international experts on heritage-making at the 2002 International Conference. The Director for

104

•

Muted Memories

Antiquities explained that “the values of Bagamoyo is determined mainly by the past of Bagamoyo. This past is related to a number of events that happened not necessarily in Bagamoyo alone” (Kamamba 2002: 2). He continued by explaining that its past is connected with the expanding ivory and slave trade between the interior of the continent and the coast and “the interaction and encounter between different societies and cultures including different belief systems, African religious systems, Islam, Christianity and Hinduism” (Kamamba 2002: 2). To conceptualize a route is, however, tricky. Slave routes were most often routes which transported the prime goods, slaves, in a single direction. Cultural routes, by contrast, are characterized by a two-way flow of goods, peoples, and ideas. The caravan route theme was captured by Fred Lerise, a participant in the 2002 International Conference in the following way: The decision to expand the scope of Bagamoyo town to the entire slave trade route was the result of the World Heritage Conference [in Bagamoyo] in 2002. At present we are at a preliminary step in this process. It is called the central slave caravan route. It should not be seen as a tunnel but as a channel of several activities, moving from mainland to coast and from Bagamoyo to the interior. Religions—Christianity and Islam—were brought to the interior through this route. And so was Ki-swahili, the most commonly spoken language in sub-Saharan Africa today. If the route played a role in unifying several countries, it has a value not only to Tanzania. (Sida 2004: 29)

The concept of the caravan trade as a channel, which is one of the core characteristics of pilgrimage, trade, and caravan routes, was explored during the conference. Caravan routes have been the channels for the flow of people, goods, ideas, religion, and languages in both directions or, as expressed by Stephen J. Rockel (2014: 172), “over the course of their travels they spread new crops, aesthetics, fashions, and ideas.” This characteristic was behind the recommendation from the 2002 conference that the role of Bagamoyo as an entrepôt for Islam and Christianity and the role of the caravan trade for the spread of Kiswahili were crucial characteristics of the caravan trade, which, among other things, brought ivory and slaves from the interior to the Indian Ocean trading networks. An ample illustration of the flow of people, goods, ideas and more in both directions along the caravan routes is the establishment of a Manyema community on Zanzibar that “others noticed, sometimes despised, and o en emulated” (see the previous chapter). The activities that took place via the caravan trade along the central routes in Tanzania were so much more than the marching of captives from the interior of the continent to the ports at the Indian Ocean. The central caravan routes’ capacity to generate “a cross-fertilization of the cultures in time and space”, as expressed in the ICOMOS definition

World Heritage and Globalization

•

105

above was not, as mentioned in the previous chapter, further explored in the 2006 presentation of Bagamoyo as a potential World Heritage site. There is a hint in the presentation that this heritage could be a “Transnational Nomination with neighboring countries” (UNESCO 2006a), but this idea was not further explored in the text. It seems like the caravan and trade route theme was forgo en when the nomination commi ee opted for the slave route theme. It may seem ironic that one of UNESCO’s initiatives to promote global understanding and peace through the role of the caravan routes “in unifying several countries,” as Lerise expressed it, was muted by another initiative to achieve a more balanced list through the promotion of the Slave Route Project, but this may also be interpreted as one of the effects of heritage-making, i.e. the silencing of alternative narratives. The same holds true for the passage in the presentation text that opts for the inclusion of the intangible heritage of the caravan trade “since one of the aims of the nomination of the heritage is to document the memories about the era and preserve the culture and the traditions of the communities living along the route” (UNESCO 2006). This may have been inspired by other activities in the Slave Route Project that have documented the memory and impact of the slave trade among populations in western and southern Africa and on the islands in the Indian Ocean. This intangible memory is the result of the enforced movement of people along the caravan routes in eastern and southern Africa. People subject to forced migration through the slave trade rarely or never managed to bring memories with them in the form of material objects. In many case, the forced and o en abrupt separation from the place of origin and the cruelty in the treatment of slaves had the purpose of erasing their memories and making them accept their new conditions and identities. It is therefore not surprising that descendants of former slaves today search for their origins or roots. The slave castles in West Africa have a racted an increasing number of tourists from the Americas. They have become very concrete evidence of the cruelty and open racism that characterized the Atlantic slave trade. The castles with their chambers, chains, and “doors of no return” represent memories of conscience and are also used in UNESCO’s Slave Route Project to illustrate the open racism that developed with the slave trade and the a empts to justify both slavery and the treatment of Africans. In eastern and southern Africa the memory of the slave trade and slavery o en take a different form. Here the intangible heritage comes to the forefront in the absence of concrete constructions that can be associated with the slave trade. Oral traditions have been in focus in efforts initiated by UNESCO to document the impact of trade and slavery on the societies

106

•

Muted Memories

in these parts of Africa and on the islands in the Indian Ocean. This intangible memory is the result of the forced movement of people along the caravan routes in eastern and southern Africa and their rebellion against deportation and slavery. It is a memory that has been carefully preserved and inherited over generations. However, Tanzania’s text is, as mentioned in the previous chapter, focused on concrete, mainly “Arabic” buildings, but it also contains an a empt to develop the route perspective through the identification of “Mango Tree Avenue” or “Mango alley” (UNESCO 2006a). The interest in mangoes, Stephanie Wynne-Jones (2011: 336) explains, “stems from the commonly held belief that mango trees formerly lined the route on each sides; the rumour is based on the existence of avenues at both Ujiji and Tabora. It is commonly maintained that a single avenue used to stretch from Ujiji to the coast.” She interprets this as an a empt by the commi ee members who worked with the text to conceptualize “the central route as a physical pathway” (2011: 336).

Paradoxes in Heritage-Making The contradictory principles of universality versus particularity are met in UNESCO’s relations with its member states. Many commentators point to the toothlessness of the organization. One of the reasons for this weakness derives from the structuring of the international relations between the organization and its member states. These relations are described by Monique Chemillier-Genderau in the following way: “international law is built on a central norm, around which its whole normative body is articulated: the norm of the equality of sovereign states” (2002: 384). This norm creates a burdensome contradiction—it is the independent state that is responsible for suggesting what is to be considered part of our universal common heritage. Hence, UNESCO’s Constitution places considerable power in the hands of the member states since they, rather than local or global communities, are given the monopoly to nominate candidates for the list. There is li le space for genuine bo om-up processes in which the local actors in the form of local authorities or the organizations in the civil society are given the responsibility to nominate potential world heritage sites and objects.19 Researchers have seen a distinction between globalization from above and globalization from below (Jokilehto 2007: 23–24). Globalization from above is triggered by market forces and the efforts by politicians and bureaucrats to stimulate the flow of capital, people, and goods. Another force from above is UNESCO’s efforts to create heritage on underrepre-

World Heritage and Globalization

•

107

sented continents and brand them in standardized forms that are recognized and accepted in the global market. Globalization from below finds its energy in the a empts of national and local actors to qualify for globally defined standard principles and procedures in hope of being recognized as members of the modern world. The a empt by the government of Tanzania to adopt the theme of a slave route may be seen as such an effort. It is also met in the competition between countries that may result in the rush for inscription. A paradox in heritage-making concerns the aura of eternal life that surrounds the concept of heritage versus the life-draining processes that may occur during the processes of disembedding, identification, formulation, and presentation of the heritage through internationally accepted procedures. This is evidenced in the Bagamoyo case. When an object is identified as heritage it is also viewed as something produced in the past that shall be protected and preserved for future generations or, alternatively, to erase that specific memory in an a empt to get control over the future. The idea to have Bagamoyo and the East African slave routes and trade on the World Heritage List so descendants from the slave trade between Africa and the American continents can learn about their background is in line with the first a empt while the purposeful destruction of World heritage sites in Afghanistan, Iraq, Syria, and Mali were attempts to edit the future. That heritage is subject to changes becomes particularly obvious when it is a lived-in conservation site as Bagamoyo Stone Town. The decision during the 1980s to demarcate and protect this area affected what the people who lived here and owned the houses could do with their lived-in habitat and the buildings. As a protected habitat, they cannot be demolished to make way for new buildings, which resulted in a strategy among some owners to let them deteriorate until they collapse, as reported in a previous chapter. The physical deterioration of the buildings is not only the result of a conscious strategy among some of the owners but also the result of the weak economic base of those who live in the protected area, in combination with the incapability of the Department of Antiquities to assist in the protection of the buildings. Twenty years a er the demarcation of the protected area, it is best characterized as an urban milieu in decay and deterioration. The falseness of the aura of eternity, which surrounds the idea of a cultural heritage, may partly be understood through Anthony Giddens (1990) characterization of modern society as a “post-traditional society.” This should not be understood as a claim that modern society is without

108

•

Muted Memories

traditions. Rather, he argues that traditions in modern society have to be nursed and taken care of in order to survive. They continue to live only if we constantly recreate and rearrange our relations with them, but they do not have in themselves an internal quality of representing an eternal value. Traditions are like zombies—the living dead. Cultural heritage has been characterized in a similar way (see Hernàndez I Martí 2006: 11). It has been said about cultural heritage-making that it is “a mode of cultural making that gives the endangered or outmoded a second life as an exhibition of itself” (Kirshenbla -Gimble 2004: 56). The branding of Bagamoyo as a representative for the East African slave trade/slave route implied the creation of a zombie. The town was deprived of its soul and blood when the complex caravan trade was reduced to the marching of slaves from the interior of the continent to the coast. The next part of this work will therefore try to unearth and explore the blood and soul of the caravan trade along the central routes and Bagamoyo’s role in this cultural phenomenon.

Notes  1. The nomination identifies Bagamoyo as just one of the links in the chain.  2. A similar picture was presented by Tanzania’s Commissioner for Culture in his speech at the 2000 preparatory workshop in Bagamoyo when he compared the situation at the turn of the twenty-first century with the situation of the stone houses during the early 1970s when he worked in Bagamoyo: “Today, exactly thirty years later, more than half of these buildings are mere ruins or are almost in a state of disrepair” (Ndagala 2000: 6).  3. However, it has been noticed that the term is used thirteen times in the English version of the Convention without being defined (Cameron and Rössler: 2013). Others have described the OUV as “some sites are so exceptional that they can be equally valued by all the people of the world” (see Labadi 2013: 11) or that such world heritage sites “belong to all the people of the world irrespective of the territory on which it is located” (UNESCO 2012).  4. The term is borrowed from J. H. Merryman (2005).  5. The supernational level assumed to exist in “our common heritage” is paralleled in formulations about outer space, Antarctica, and the high seas (Brumann and Berliner 2016: 8).  6. This is nothing new. The area, people, culture, and language in focus in this study reveal an extraordinary example of how identities, although primarily not ethnic, have been created and manipulated to achieve inclusion and exclusion. Other parts of this work will illustrate the fluidity in the coastal identities, how material resources were used to achieve, express, and maintain status and power, and how various groups and actors constantly struggle to transgress, close, or break sociocultural boundaries.

World Heritage and Globalization

•

109

 7. Globalization is understood here as the accelerated flow of people, goods, capital, ideas, information, knowledge, conflicts, environmental pollution, etc. Globalization is seen as a more complex and multifaceted process than just an issue about globalized financial movements, instruments, and markets.  8. The acceptance of common frameworks and standards made the imagined community (the nation) possible, and contributed to the accelerated flow of people, goods, money, ideas, etc.  9. During recent decades, this source of power has been seriously challenged by the elected members of the World Heritage Commi ee, the organ which, according to Article 8 of the Convention is the major decision-making institution regarding inscription of proposed heritage sites. A growing number of members have begun to follow other interests than those recommended by the advisory boards. Nominations recommended to be further developed have been inscribed and heritage sites recommended for Noninscription have been given a second life through decisions on referral or deferral (Meskell 2012). 10. Author’s field notes, March, 2013. 11. This may also be seen as a normative power. 12. See also Turtinen 2006. 13. In a thought-provocative argument Saskia Sassen suggests that “nationalism, o en seen as an obstacle to globalization, is a product of the same forces that are shaping the la er” (as formulated in Eriksen 2007: 25). 14. See Breidenback and Nyíri 2007 regarding the current desire of China and Russia to nominate and have parts of their heritage inscribed on the WHL. 15. Cf. Lukes’s third dimension of power. 16. Duncan Light (2015: 144) identified a difference between the continents in the heritage tourist industry. In Europe, it focuses on historic buildings, sites, and monuments. In the Americas, it focuses on natural landscapes, while in other parts of the world “heritage tourism also includes indigenous culture and traditions.” (2015: 144). 17. Deputy Secretary-General, Zimbabwe National Commission for UNESCO and Deputy Permanent Secretary Delegate of Zimbabwe to UNESCO. 18. Called “serial nomination.” 19. One of the most absurd consequences of this is that even indigenous people, who have reasons to minimize their relations with the state they o en do not even recognize, are made dependent on the same state for the nomination of, say, sacred landscapes where they live.

PART

II  Commerce, Competition, and Consumerism Bagamoyo and the Caravan Trade There is a long tradition in the presentation of East Africa’s history to associate the negative developments in or “underdevelopment” of the area with external forces. This is pointed out by Jennifer Kopf (2005: 8) when she refers to Edward Alpers’ conclusion in his book Ivory and Slaves in East Central Africa that “examining Portuguese, Umani Arab, Brazilian, Indian, British and American relations with east Africa,” Alpers finds that “in each situation there emerged a definite structural subordination of African economies to a capitalist economy, with the main powers of decision-making residing outside Africa” (Alpers 1975: 265). This long tradition to diminish the African impact is captured in UNESCO’s publication General History of Africa with the following words: “For a long time, all kinds of myths and prejudices concealed the true history of Africa from the world at large. African societies were looked upon as societies that could have no history . . . and [it was] argued that the lack of wri en sources and documents made it impossible to engage in any scientific study of such societies” (M’Bow 1981: xvii). The same flaws are met in many presentations of the history of East Africa. “One of the many errors of much of the historical literature on the coast of East Africa,”

112

•

Muted Memories

writes Hatim Amiji, “has been to view the coast in the precolonial period as though it was a tabula rasa onto which have been impressed more or less intact, certain Asiatic ‘influences’” (Amiji 1983, 66). The second part of this book will begin with an exploration of the indigenous influences and impacts on nineteenth-century caravan trade along the central routes from eastern Congo to the Mrima coast on the Indian Ocean. The exposé will follow the caravan porters from the interior of the continent to Bagamoyo. Their initiative, to bring ivory to the coast, not only stimulated the trade in the town, but also a global trade in ivory and co on cloths. The impact of the growing caravan trade on Bagamoyo resulted in hectic processes of creolization and cultural hybridity as well as struggle for inclusion in and exclusion from life in the town. The booming caravan trade generated the wealth that fueled a characteristic feature in the coastal societies; status and power-seeking and communication through consumerism and conspicuous consumption. The struggle over status positions and recognition by means of the wealth generated through the expanding caravan trade, turned nineteenth-century Bagamoyo into a powder keg. The stiff competition between domestic actors and intruding regional and global powers over the control of the caravan trade on the central routes eventually led to the colonization of the area. The construction by the German colonizers of a railway between Dar es Salaam and the western parts of their colonial territory meant the end of the caravan trade and Bagamoyo as the most important port on the Mrima coast. This story, hidden behind the o en-closed doors and windows of the stone houses of present-day Bagamoyo and behind the a empts to brand the site as a World Heritage site for the East African slave trade will be further explored in the subsequent chapters.

5  Entrepreneurs and Explorers from the Heart of Africa A theme put forward in the first part of this work is that alternative stories are hidden or silenced when a particular perspective on a potential World Heritage site is chosen. In another chapter it will be claimed that we can look at the central caravan route either from Bagamoyo as an entrepôt, looking westward; or we can look at Bagamoyo as the terminus of the long-distance trade, looking eastward. In a previous chapter it was also suggested that we can either chose to see Bagamoyo as the main slave port on the coast or we can chose to see the town as the most important for the trade in ivory, gum copal, slaves, cloth and other items, and as an interface in a cultural route that connected the interior of the continent with the Indian Ocean region. The text below will focus on Bagamoyo as a terminus for the ivory trade, which stimulated the development of the town as a vehicle for the spread of people, goods, religions, languages, and ideas into the interior of Africa by the porters and traders operating on the central caravan routes of East Africa. The focus will be on the Nyamwezi1 porters and traders from the Unyamwezi area south of Lake Victoria and east of Lake Tanganyika (see Map 5.1). Most researchers today agree on the Nyamwezi’s role in initiating this long-distance trade, and their role as porters in the caravans both from the interior of the continent and from the coast, although traders from the coast began to dominate the caravan trade during the second half of the nineteenth century.

114

•

Muted Memories

Map 5.1. East African central caravan routes and port towns. Created by the author.

Ivory has been exported for several millennia from East Africa. The Egyptians imported ivory from Nubia four thousand years ago and East Africa exported ivory to the Roman Empire two thousand years ago (Chami 1999). A millennium later, traders from Asia searched for ivory along the coast. The ivory from the African elephant is easier to carve that the bri le ivory from the Asian elephant. Their tusks are usually much larger, and ivory was in high demand, for example, in India where it was used to produce bridal decorations. The demand, which increased at the end of the eighteenth and early nineteenth centuries during the industrialization in West Europe and North America, was behind the development of the central caravan routes and the role of the Nyamwezi people in the caravan trade. It was reported at the mid of the nineteenth century that in caravans from the Nyamwezi area: there is no desertation, no discontent, and, except in certain spots, li le delay. The porters trudge from sunrise to 10 or 11 A.M., and sometimes, though rarely, they travel twice a day, resting only during the hours of heat. They work with a will,2 carrying uncomplainingly huge tusks, some so heavy that they must be lashed to a pole between two men—a contrivance technically called mzega-zega. Their shoulders are o en raw with the weight, their feet are sore, and they walk half or wholly naked to save their cloth for display at home. They ignore tent or covering, and sleep on the ground; their only supplies are their country’s produce, a few worn-down hoes; intended at times to purchase a li le grain. (Burton 1859: 408–9)

The long-distance caravan trade from the interior of Africa developed through the centripetal and centrifugal forces of early globalization; centrifugal in the sense that the caravan trade distributed goods like ivory,

Entrepreneurs and Explorers from the Heart of Africa •

115

slaves, and co on cloths over the globe and centripetal since it brought together producers and consumers on different continents into networks of trade. The following pages have two aims: to present the involvement of porters and traders from the interior of the continent in the creation of Bagamoyo as the most important caravan port along the East African coast and to get a fuller understanding of the caravan trade.

The Origin of the Long-Distance Trade along the Central Caravan Routes A specific physical condition of this part of Africa contributed to the development of this form of caravan trade. It is, as Walter T. Brown (1971a: 15) writes “conspicuous for the total absence of long, navigable—and thus potentially commercial—water-ways.” Three rivers—Pangani, Kiangani/ Ruvu, and Rufiji—are navigable between 100 and 200 km into the continent while the main areas for ivory and slaves were up to 1,000 km away from Bagamoyo. The absence of rivers suitable for transport between the interior of the continent and the East African coast le the human body as the prime means of transport. Hence, the East African caravan trade was built upon the feet, legs, shoulders, and heads of the porters (Horton and Middleton 2000: 105) or, as Thomas Beidelman, quoting John Roscoe (1921: 4f), captures it “human burden bearers who was the sole means of transport” (1982: 612). In contrast to many other caravan routes, where horses, mules, camels, llamas, yaks, reindeers, or donkeys were used as pack animals, this trade depended on the body of the human beings. The caravan trade in this part of Africa is many centuries old, as witnessed by the geographer Muhammad al-Idrisi during the twel h century: “As there are no pack animals among these people they are obliged to carry on their heads and backs the objects for the cities where they buy and sell” (Brady 1950: 28, quoted in Rockel 2006a: 3). This statement, however, only managed to catch a part of the truth, the transport of trade goods to the trade towns. These porters and the African caravan leaders along the central caravan routes from, for example, Bagamoyo to Lake Tanganyika were, as we shall see, a sine qua non for all the travelers, traders, and missionaries from Europe and America who set out from the coastal towns in their a empts to penetrate the continent and later on also for the European colonizers. These intruders depended on the skills, knowledge, and social and cultural competence of those who worked with the caravans. The image of the East African caravan trade as a slave trade is very sticky. Many accounts present a picture of slaves tied together and marched from the interior to the coast, or they place the merchandise traded within this long-distance trade, mainly elephant tusks, on the shoulders and heads of

116 • Muted Memories slaves. The image resembles the misconception of Bagamoyo as the main slave port on the East African coast. The image of slaves captured in the interior of the continent and forced to carry elephant tusks to the coast where the slave porters were sold to work on the plantations is mainly incorrect, since most of the porters were free men and also women, who either took part as small-scale traders cum porters or sold their labor for a wage. Many porters were, however, slaves who were recruited in the coastal societies where the owner was compensated for the service of his or her slave. These slaves could carry trade goods, but also the tents and other items of the owner’s/trader’s household, which made this kind of journey more comfortable. Sometimes they even carried the “explorers,” traders and missionaries. Already Burton (1859: 408) identified three kinds of “original” caravans in East Africa: “The most novel and characteristic are those composed only of Wanyamwezi; secondly are the caravans directed and escorted by Waswahili freemen or fundi (slave-fattori) commissioned by their patrons; and, lastly, those commanded by Arabs.” From the middle of the century a fourth category of caravans may be added; those arranged by European or American travelers, missionaries, and colonizers with the aim of traveling fast and without the purpose of transporting often bulky trade goods between the coast and the interior. Burton’s categorization corresponds rather well with three different techniques to carry the loads, each one associated with a specific ethnic group (Rockel 2006a: 103–4). The Manyema and other people from the Congo forest, often slaves who first had been brought to the coast and there recruited as porters in caravans organized by coastal traders, carried their loads on their backs. Loads carried with the help of ropes suspended from the forehead of the carrier cannot be too heavy, but at the same time they cause a minimal obstacle when travelling along forest paths. Porters originating from the coast, often slaves used as porters in the caravans run by coastal traders, preferred to carry their loads on their heads. This is a technique suitable for bales and rolls with cotton cloth, the most important trade item transported from the coast to up-country consumers. Cloth imported from India, Europe, or the Americas was repacked in a trade town like Bagamoyo into rolls or bales for further transport to up-country societies. This technique, too, is less efficient when compared with the technique preferred by the Nyamwezi, since the porter can only carry a lighter burden of approximately 23 kg. The Nyamwezi, by contrast, carried their loads on their shoulders. A typical load of ivory weighed 70 lbs or approximately 26 kg. Elephant tusks have a shape that naturally suits the transportation on shoulders; cloths were packed in rolls, if carried by Nyamwezi porters suitable for this

Entrepreneurs and Explorers from the Heart of Africa •

117

technique. In addition, the porters also carried food, water, and personal possessions and weapons such as flintlocks. Stephen Rockel (2006a) estimates that the total weight a porter carried could amount to 35 kg or more.3 As will be shown below, the organizational techniques developed by the Nyamwezi caravan tradesmen were copied by other caravan leaders. There were, however, some noticeable differences. The porters in the Nyamwezi caravans should be regarded as free men and women, while both the Swahili and Arabs o en used slaves. Europeans and Americans used both free persons and slaves in their caravans, although this is not always explicitly explained.4 There is today a growing literature on the origin, organization, and life conditions of the caravans and their porters (see for example Flint 1976: passim; Roberts 1970; Rockel 2000a; Rockel 2006b). These sources describe the emergence of a specific culture connected with porterage as well as how demands and tastes in the interior of Africa, through the mediation of the caravan trade, affected the industrialization of distant parts of, for example, the United States and India (Prestholdt 2004). This is a cultural heritage, which deserves a be er recognition. Bagamoyo was, in the past, conventionally seen through the eyes of those who came over the Indian Ocean, particularly from the east, from Zanzibar. In the present text, the perspective will be partially shi ed and reversed since the focus will be on how the caravan trade and Bagamoyo may have been experienced by the thousands and thousands of porters from the interior of the continent when they approached this final destination a er months of walking along the central caravan routes. A shi of perspective from the coast to the interior will also put the focus on the consumers in the heart of Africa who o en have been marginalized in our understanding of the Indian Ocean trading networks (Riello and Roy 2009: 20). Recent research has, however, challenged the former view that made the Africans into an “under-considered population” or, as expressed by Pedro Machado (2009: 57) “far from being marginalized in this commercial nexus, African consumers were able to negotiate the terms of trade and their engagement in relations of exchange of which they formed an integral part.”

“Those Who Work with a Will”: Explorers from the Heart of the Continent The following quotation will provide the stage for this investigation in the caravan trade: “If Zanzibar was the hub of the ‘commercial empire’ (Sheriff 1987), the Nyamwezi caravan was its basic institution” (Rockel 2000b, 769). The development of the caravan trade was part of a global-

118

•

Muted Memories

izing process that spread trade goods over the globe. It is estimated that during the peak decades of the caravan trade one-third of the male population in Unyamwezi, the land of the Nyamwezi people, annually undertook the caravan march to the coast (Roberts 1970, 66; see also Rockel 2006a, passim). The long-distance caravan trade, as we meet it during the la er part of the nineteenth century, was a response to trading activities initiated around 17805 by people from the interior of Africa. Rockel (2000a: 1) in colorful language writes: Beginning in the late eighteenth century a new way of life developed, indeed a new culture, as people from what is now western Tanzania, especially those called by others the Nyamwezi or “people of the moon,” made pioneering journeys. Their caravans at first searched out opportunities in the regional trading networks of the high plateau between Lake Tanganyika, Victoria and Malawi. Many caravans searched out opportunities in almost every part of East and Central Africa, at first starting from the Nyamwezi homeland on the central plateau, travelling east across the nyika or wilderness, to climb the coastal ranges and reach the Indian Ocean.

In short, this interpretation tells that their ancient trade in salt and iron between up-country societies was extended with new items and destinations: ivory, livestock, hides and also slaves were brought to the small trading ports along the Indian Ocean. Ivory from East Africa had for a long time been in high demand in India. It was mainly used to manufacture the bangles that women, rich and poor, Hindu or Muslim, wore for the marriage ceremony (Koponen 1988: 57). The first porters/traders who made their way to the coast probably arrived in ports like Kaole, Saadani, or Mbwamaji (or probably more correctly Mbu Maji or Mosquito Water, situated 14 km south of Dar es Salaam). Mbwamaji was, a er Pangani to the north and Kilwa to the south, a most important port when the caravan trade gained in volume and a large proportion of the ivory trade passed through this port during the end of the eighteenth and early nineteenth centuries. It seems like this port became less important when Bagamoyo was established and had by the 1860s “sunk into obscurity” (Rockel 2006b: 1). The first Nyamwezi caravans that reached the trade towns on the coast were probably small compared to the caravans at the end of the nineteenth century. It has been estimated that the pioneering caravans consisted of thirty to forty people (Rockel 2014: 180), probably groups of relatives or neighbors who undertook the journey to the coast. Some caravans grew in scale with the booming trade. It is estimated that some of them contained up to 2,500 or 3,000 people during the decades before colonization (Rockel 2014: 180).

Entrepreneurs and Explorers from the Heart of Africa •

119

The origin of this long-distance trade was explained by Norman Benne ([1968] 1974: 210) when he wrote that “the initial stimulus for this movement [caravan trade] came from Africans of the interior, especially the Nyamwezi, . . . who opened up trade routes that Arabs later followed.” This history is also part of Nyamwezi oral traditions. Andrew Roberts, writing on Nyamwezi trade, reproduces an oral story from an interview he held in the Nyamwezi area in 1967 for his study on the history of Nyamwezi trade. When the first traders, trading in iron, particularly iron hoes and salt produced in western Tanzania and ivory, pushed eastward toward the coast, they passed the Gogo area, today the central part of Tanzania where: they met people tilling the soil with wooden tools. So they sold their iron hoes for domestic animals . . . When they were in Uzaramo [the area around Bagamoyo] they were told that further to the east there was a big lake [Indian Ocean]. They asked to see it and went on as far as Bagamoyo . . . Here they went to Bushiri and found beads and cloth. This Bushiri was a lightskinned man . . . They came back again and began looking for ivory. They carried this to Bagamoyo where the long-bearded men lived. These were the first Arabs. When the Arabs saw this, they wanted to go to the countries where the ivory was obtained. (Roberts 1970: 48)

Roberts is concerned about the truth value of this account and warns that all details in the story should not be taken wholly at face value (1970: 48). He points to the anachronism of Bushiri, who lived during the la er part of the nineteenth century while the first Nyamwezi reached the coast a hundred years earlier. Furthermore, Bushiri could not have been on the scene at the same time as the Gogo of what today is the central part of Tanzania used wooden tools and “the first Arabs” came to the coast. This oral history seems somewhat condensed. With these warnings in mind, the legend may still communicate other messages such as that the Nyamwezi caravans could bring trade items such as iron hoes, which were in high demand among the people along the caravan routes. Porters from up-country such as the Nyamwezi, Sukuma, and Kimbu could not carry enough food-stuffs for their approximately three-month march to the coast. They had to rely on trade items in demand along the routes. The salt and iron they traded within their already-established up-country trading networks and later on slaves who walked on their own, were such items. The long tradition of iron and salt-making in the area between Lake Tanganyika and Lake Victoria supplied the caravans with the means of exchange, in combination with tobacco and smallstock, which made it possible to procure food along the routes. The legend gives us a hint about the demand for iron hoes among the Gogo.

120

•

Muted Memories

The legend also reveals a division of labor between the groups. Ugogo or Gogoland, the dry central part of Tanzania, was rich in elephants, but the Gogo never became involved as traders or porters in the caravan trade. They hunted elephants but, in contrast to the Nyamwezi, they never or rarely brought their ivory to the coast. Instead, they bartered their tusks for cloth, beads, weapons, and ammunition with the caravans coming from the coast, and they bartered for slaves, salt, iron hoes, and tobacco with the caravans passing their area from the west. The increased demand for food that followed from the caravan boom stimulated the intensification of their food production, which may be implied in the message that they shi ed from wooden to iron tools. Although their role in the caravan trade differed from the role of the Nyamwezi, they could still benefit from the trade. The caravan trade brought a certain degree of prosperity to those parts of Gogoland that were involved in the provision of ivory and food and that could extract tolls and gi s from the caravans (Sissons 1984). Their role as provider of food stemmed from their geographical position between two uninhabited semi-desert areas along the central caravan routes. A caravan traveling from Tabora in Nyamweziland to the coast usually spent one-third of its time passing the land occupied by the agropastoral Gogo people. Before entering their area in central Tanzania, the caravan had to pass a dry area called Magunda Mkali or “fierce field” area. Food and water for the passage of Marenga Mkali had to be brought from Nyamweziland and carried by the porters until they reached Gogoland. Here the caravans passed through a number of chiefdoms where a road tax called hongo had to be paid before food and water could be obtained from the local population. Sissons (1984: 76–98) estimates that a caravan passed through eight to twelve chiefdoms on its way through Gogoland where hongo was demanded. The number varied depending on whether a caravan used the southern, central, or northern tracks of the central caravan routes. East of Gogoland the caravans entered the notorious Marenga Mkali or “bi er/brackish water” area known for its lack of open water sources except a er the rains. The Nyamwezi porters had to obtain both food and water from the Gogo before they continued their march toward the coast and camp sites like Mpwapwa. Iron hoes, tobacco, as well as ca le and slaves were used to purchase not only the necessary provisions but also ivory in this area rich in elephants.

Why the Nyamwezi? For a variety of reasons, a cluster of three or more interrelated groups with the Nyamwezi at the forefront developed the caravan trade along the

Entrepreneurs and Explorers from the Heart of Africa •

121

central routes in Tanzania (see Map 5.1). The other groups, such as the Sukuma, Sumbwa, Konongo, and Kimbu, are all closely related to the Nyamwezi and belong to what R. G. Abrahams (1967) calls Greater Unyamwezi. The most commonly mentioned reasons for “why the Nyamwezi” will be presented below. There is no need here to weigh them against each other, but suffice to say that they, in an interlinked manner, triggered the historical epoch we know as the caravan trade along the central routes. The findings from Andrew Roberts’s (1970) thorough study, presented in the chapter “Nyamwezi Trade” in the edited collection Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa Before 1900, documents the ancient trade in salt and iron in the inner part of Africa. Nyamwezi were strategically placed on the routes and could benefit from the demands for these goods. Salt was extracted at the brine springs in Uvinza to the west of Unyamwezi, the land of the Nyamwezi. Estimates by European travelers who visited the area during the last decades of the nineteenth century say that approximately twenty thousand people came to work at the springs each year, each one producing approximately 25 kg of high quality salt (Roberts 1970: 54). Salt was also extracted from Lake Eyasi and Lake Kitangiri to the east of the Nyamwezi heartland. Iron was at that time produced by ironworking specialists in the Ha and Zinza areas (Roberts 1970: 44) (see Map 5.1), located to the northwest of Unyamwezi and was in high demand in the societies to the east of Unyamwezi, as explained in the legend referred to above. When the caravan trade gained in momentum the Nyamwezi carried new and old iron tools to the lands of the Gogo, Sagara, and Zaramu and exchanged them for ivory and food (Unomah and Webster 1976: 287). The importance of iron in the caravan trade is captured in a song of the ironworkers in which they portray themselves as the genuine producers of the wealth and the source of the trade with the coast: “Only the charcoal of the mubanga tree brings you the fine cloths and sights of the coast” (Roberts 1970: 53, who refers to Stern 1910: 155, 157). Hence, the Nyamwezi were strategically placed as middlemen on the east–west axis that later became the central caravan routes in a manner resembling the position of the merchants and middlemen in the ports towns on the coast who became an interface between the networks on the continent and those over the Indian Ocean region. To the southwest, their trading networks extended into copper-rich Katanga in what today is the Democratic Republic of Congo and northwest to Lake Edward where new elephant hunting grounds were sought. The origin of the extension of the trading networks from the interior to the coast may also be sought in the relay-type of trade that had existed for centuries between the Great Lakes region and the coast. Trade items

122

•

Muted Memories

traveled from hand to hand in these barter systems. Richard Francis Burton (1859) wrote, with reference to this system, that traders from the coast penetrated westwards, probably to the Singida region in central Tanzania, where “the caravans made no stay; they built neither house nor store, but lived in tents and under hides, and a er hurriedly completing their barter they returned coastward” (Burton 1859: 300). These encounters could have been a source of information to the Nyamwezi traders on the demand for ivory and slaves at the coast. Another reason for “why the Nyamwezi” was the availability in their area of the prime goods in the developing long-distance trade: elephants and their tusks. Already in 1811, people on Zanzibar knew that ivory was in abundance in Unyamwezi (Roberts 1970: 49). This information triggered trading activities on the coast, and in 1820, an Arab expedition set off westward in search of ivory, an initiative also explained in the book of the infamous slave trader Tippu Tip [1974] 1958/1959, 63). When elephants became scarce in Unyamwezi, the Nyamwezi hunters/traders extended their hunting grounds westward into Manyema-land in Congo and northwards as far as Lake Edward (Roberts 1970: 68.). Unyamwezi was, at the time of the development of the caravan trade, covered in a woodland landscape. The southern part of the area was covered with the miombo woodland, while the woodlands in the northern part had been cleared for cultivation and to become grazing lands (Roberts 1970: 41). The drier area to the east was also sparsely covered with trees and bushes. These wooded habitats were more favorable for elephants than open cultivated lands. And tusks were, a er all, the prime trade objects. A different kind of reason for “why the Nyamwezi” is related to the economic organization of Nyamwezi society. The population was primarily dependent on agriculture for subsistence, while keeping livestock was of a more limited importance compared to the Sukuma people to the north and the Gogo people to the east. The importance of agriculture and the fact that many, perhaps most, porters also were cultivators, determined the timing of the caravan season. This was particularly true for the small-scale traders and porters, who were part-time farmers and parttime porters/traders. The harvest season determined the organization of caravans from upcountry societies to the coast. Just before the harvest was taken, usually in April, a drummer went around in the villages to notify presumptive traders/porters that a caravan was to be formed within a couple of weeks (Rockel 2006a.) The division of the year and agricultural calendar in a rainy and a dry season when the harvest was taken made cultivation difficult or impossible during the dry season. This was the time when the

Entrepreneurs and Explorers from the Heart of Africa •

123

porters could be absent. The favorite time to start the march down to the coast was the season when li le work could be carried out in the fields. The determining effect of the cultivation calendar had a peculiar effect on the caravan trade since the caravan porters and traders, on the one hand, had almost unlimited time to bargain for the best price of their ivory while at the coast, and on the other hand, were expected to return home in time for the onset of the cultivation season. The Indian traders who dealt with the ivory trade were well informed about the Nyamwezi agricultural calendar and “knowing that a Nyamwezi caravan could eventually become anxious to get home in time for the sowing season” acted accordingly (Burton 1860: 1, 39). The Indian trader who could wait would wait until the price went down when the time came to start the up-country march. The demand for the labor of the porters in the agricultural sector was also sensed in another way. Non-Nyamwezi caravan leaders o en found it difficult to obtain porters to continue beyond Tabora and the land of the Nyamwezi to Lake Tanganyika and further westward into Congo. A caravan leader who wanted to continue westward was o en faced with new negotiations or had to find new porters. Yet a further explanation for “why the Nyamwezi,” a question Rockel (2006a, 6) asks, is the pronounced employment of slave labor, especially female slaves, in their cultivation system (Deutsch 2006: passim; Rockel 2006a). It is suggested that the temporary absence of male labor in Nyamwezi agricultural regimes was compensated through slave labor. The gender division of labor put a large responsibility on women when the men traveled to the coast. These practices increased as many porters invested their earnings from trade into the purchase of livestock and slaves, a strategy that must have been encouraged by the cheapness of slaves. Slaves were in fact cheaper than a donkey, a cow, or a bag of salt—and used as a currency in the same way hoes and salt were (Roberts 1970: 60).6 Slavery and the use of slaves in the East African communities is an old phenomenon (see the next chapter). Roberts reports that “in 1881 Becker passed through the Nyamwezi chiefdom of Ugunda and wrote of ‘un petit bourg, fort de 500 habitants, dont 40 hommes libres et le reste esclaves!’”, which Roberts for good reasons believes is an exaggeration (Roberts 1970: 59–60). Other sources, however, testify to the high proportion of slaves in Nyamwezi society. Jan-Georg Deutsch (2007: 76; referring to Reichard 1890: 277) reports that “an official report in 1890 [reveals] that about twothirds of Tabora district’s [in central Nyamweziland] total population of around 350,000, i.e. approximately 233,000 people, was locally regarded as slaves” was accurate. The increase in the caravan trade and the expanding trade links brought about changes in local food production. The introduction of new crops

124

•

Muted Memories

increased the yields. One of these crops was rice, probably brought by Nyamwezi traders/porters when they returned home from visits to the coast and by coastal people when they pushed westward in search for ivory and slaves. An advantage with rice was that this crop could be grown on land previously unsuitable for cultivation, as land flooded during the rainy season (Roberts 1970: 62). When irrigated wet rice was introduced this kind of land, mbuga, could be turned into productive paddy fields. Maize and cassava were other crops that entered the area during the eighteenth century. It is probable that these crops, from the American continents, came through trade links with societies on the Atlantic coast. Together with rice they meant a diversification of food production and an increase in the total output, which probably released many Nyamwezi women and men from some agricultural duties and made it possible for them to take part in caravans to the coast. “No doubt,” Roberts writes “cultivators with slaves—who might be sold if food ran short—were readier to take such risks than those without” (Roberts 1970: 72). It may be hard to accept, but slaves and livestock seem to play the same role in this kind of production and exchange system. With a surplus from cultivation or trade, both slaves and livestock could be obtained. Both could be taken along the caravan route and exchanged for ivory, and both could be exchanged for food in times of famine. Both were simultaneously a storage of wealth and a medium of exchange.7 Yet another reason for “why the Nyamwezi” was their political organization. Unyamwezi consisted of a number of relatively weak chiefdoms. They were weak in the sense that the freedom of ordinary people to travel was not heavily restricted by those in power as o en was the case in more centralized political organizations. They were weak also in the sense that the political elite could not easily tie people to the land in order to maximize production and to usurp a surplus. It has been suggested that the fact that the chiefs had no special weapons and few resources meant that dissatisfied villagers could move elsewhere, particularly since land was abundant (Roberts 1970: 42). Thus, people seem to have been free to travel and, by extension, to take part in the caravan trade. A further dimension of the political organization, which may have favored the development of the caravan trade from the Unyamwezi area, was the relative peace and stability in the area. Compared to the northern route from Pangani and Tanga through Arusha and Maasailand, where large-scale conflicts and famine had severe repercussions on trade from 1840 onward, the central routes could operate under more peaceful conditions, with the exception of the middle of the nineteenth century when the Ngoni from South Africa invaded and disrupted the order in parts of the area.

Entrepreneurs and Explorers from the Heart of Africa •

125

A contributing reason of a somewhat different character was the arrival of the Sagara people to the Nyamwezi area, who had moved westward from their mountain homeland in eastern Tanzania. It has been assumed (Rockel 2006b: 9) that slaving expeditions from the coast to the Uluguru and Usagar mountains inhabited by the Sagara resulted in their migration further west to the Ukimbu area. When they arrived to the southern part of the greater Unyamwezi area during the eighteenth century they came with items from, and probably also knowledge about, the trade and demand for ivory on the coast, which thus linked the coast with the interior (Rockel 2006b: 9). It is assumed that this stimulated the Nyamwezi to extend their previous routes to also reach the coast and to add ivory to their merchandise. As explained by Aylward Shorter in his monograph about the Kimbu, one of the groups related with the Nyamwezi: “The coming of Nyitumba from Usagara [the land of the Sagara people] was an event of the first importance, for they opened up a route linking the interior to the hinterland of the coast, and ultimately drew the peoples of the interior to the coast in search for the new objects of value they had brought” (Shorter 1972: 225). It may be hypothesized that the demand for ivory from up-country sources increased when the first source of ivory exported from East Africa, the coastal elephant herds, were almost extinguished (Håkansson 2004; Koponen 1988; Sheriff 1987), and the Nyamwezi were not late to respond to this demand.

Caravan Organization and Culture The caravans8 that set out eastward from the interior were of different shapes and sizes. At the beginning many caravans were rather small and formed by a number of individual traders and porters who came together to carry their own trade goods or small traders who employed a handful of porters. These small traders and porters could also came together to form larger caravans comprised of free and relatively independent traders. Other caravans contained thousands of porters who either were their own traders or hired porters employed by a trader or a group of traders doing big business. One such “bigman” was the Sukuma (a group in the Nyamwezi cluster to the north of the Wanyamwezi) caravan leader named Telekesa. He arrived in Bagamoyo in September 1889 with his caravan containing 3,500 people, 400 tusks, 2,500 ca le, the same amount of sheep and goats, and 50 to 60 donkeys. He came to the coast every second year with the porters and goods he assembled in Sukumaland (S. Fabian 2007b: 163; referring to Schweinitz 1893). The porters in Telekesa’s caravan, how-

126

•

Muted Memories

ever, were their own independent traders who bargained each on their own strength with coastal traders.

Caravan Culture on a Cultural Route It goes without saying that such an amount of people on the move with their goods demanded an elaborate organizational setup, which explains that: “Among the porters a shallow kind of specialization developed: there were cooks, tent boys, carriers, personal servants; selected men were trained as askaris [soldiers]. The caravan personnel formed a community of its own with its own interests and ‘culture,’ expressed, for instance, in songs and stories” (Roberts 1970: 66).9 The porters developed their own culture from sociocultural features borrowed from the Nyamwezi society. This is convincingly captured in Rockel’s study on the long-distance trade aptly called Carriers of Culture: Labor on the Road in Nineteenth-Century East Africa (2006a). The porters’ indigenous model for caravan organization was copied by non-Nyamwezi caravan leaders. In this sense the Nyamwezi had a standardizing influence both on caravans lead by Swahilis/Arabs and by European traders, missionaries, and travelers. The caravan trade along the central routes became a cultural route in the sense of UNESCO’s routes projects. A crucial feature of Nyamwezi culture incorporated in the caravan tradition was the establishment of joking relationships—utani. This institution was the grease in the machinery for the porters, or as Lloyd Swantz (1956: 22) put it, utani-ship provided “safe conduct, passport and hospitality when travelling” when they passed societies whose members were either friends or foes. Utani-ship meant the establishment of quasikinship ties, which were used to get access to support and protection, but also responsibilities. It was common that a Nyamwezi porter/trader established such a joking relationship with a town dweller, preferably a trader or a person working for a tradesman, in, for example, Bagamoyo. This person benefi ed from the arrangement since he was the first to be notified when his Nyamwezi friend approached the coast with his goods, which prevented the up-country traders and porters from dealing with the underlings in a port town like Bagamoyo (see also S. Fabian 2007b: 180–81). A hint about the common understanding of the background and role of this kind of brotherhood is explained by the Bagamoyo-born author Mtoro bin Mwinyi Bakari10 who writes (J. W. T. Allen 1981: 167): “In the past the inland people and the people of the coast feared and robbed each other. If an inland person came to the coast with goods, he was robbed, and if the coast people went inland, they were robbed of their goods and

Entrepreneurs and Explorers from the Heart of Africa •

127

murdered. So they wanted blood brotherhood to remove the fears from their hearts.” The role of utani-ship should not be underestimated in a time and place where other overarching or integrative sociopolitical institutions were lacking. This quasi-kinship system gave a structure to a sociopolitical landscape that was otherwise fragmented into a number of small polities. The establishment and maintenance of predominantly fictive social relationships made it possible to travel to distant places in the absence of overarching political, juridical, or military structures. As will be shown in the next chapter, the kinship system provided the idiom through which utani-ship and slavery were perceived and articulated. Rich traders, who could be of Nyamwezi origin and belong to the local elite, as well as Swahili, Arab, and European traders borrowed the structure and organization from the Nyamwezi caravans and adjusted it from the original format of small-scale traders who o en operated individually in the caravan or in household clusters. This pa ern was adopted and adapted to fit the large-scale caravans with well-defined owners and leaders and that occasionally contained more than two thousand porters. These caravans are best characterized as a society on the road. Other examples of the Nyamwezi cultural influence on the caravan culture are “its own timetables deriving from Nyamwezi social norms” and customs that “defined working standards and regulated disputes related to working conditions, food, wages, and payment as well as the weight of the load the porters had to carry, the length of marches, and the compensation for extra duty” (Wisnicki 2008: 115).

Caravan Organization The organization of large scale caravans was o en described in the travelogues produced by a wide range of travelers, traders, missionaries, and colonizers. The largest caravans could form a column of participants that covered 3 km or more. They contained staff with specified and hierarchically organized functions who accompanied the porters. The most important members in the trade caravans were the traders. They usually walked or occasionally rode donkeys at the rear of the column so they could oversee the progress and detect deserters. The following account gives the details of the organization of the caravans: The kirangozi is immediately followed by the aristocracy of the caravan, the ivory-carriers, who in their burden take pride which salves the sore of its dead galling and undiminished weight. The tusks are poised upon the shoulder; large ca le-bells dangle from the point which is carried in the front, whilst the porter’s travelling-kit is fastened to the bamboo be-

128 • Muted Memories hind. Next in order come the cloth and the bead carriers; then those laden with wire; and in the rear a rabble route of slaves corded or chained in file; women and children in separate parties. (Burton 1859: 414)

The leaders of the caravan were the wanyampara. The term is a derivation of the Nyamwezi term banamhala, meaning “old men” (Cory 1954: 63); both a status title and a power position in a gerontocratic society. This original meaning, writes Rockel (2000a: 183), was probably lost when the term was adopted and misspelled in the context of caravans with costal origin. At the front of the caravan was the guide or kirangozi (pl virangozi), (sometimes spelled viongozi ) who had made the journey several times and knew the route. The caravan guide was crucial in caravans from the coast, since he had knowledge about the local demands along the caravan route. Such knowledge was important for the caravan traders to select the items that could be used to obtain food and the right to pass a village or chiefdom, and could also be exchanged for ivory and slaves (Prestholdt 2004: 764). Usually he was elected among the porters, another feature that may have been borrowed from the Nyamwezi/Sukuma society. Hans Cory (1954) writes about a special category of community leader in this society who was elected to stay in office for usually three to five years, after which a new leader was elected. This elected leader was called nsumba ntale, a title Cory translates as “The Great Commoner.” His duties were mainly to organize communal work gangs for cultivation and to supervise the welfare of the society (Cory 1954). It seems plausible that this institution was behind the elected kirangozi who was usually also entrusted to negotiate the road toll, hongo, with the chiefs of the polities the caravan passed through. The negotiation skills were essential for the successful closure of the negotiations and handing over of the hongo, which in turn opened up for the purchase of food and water. When all is ready, writes Burton (1859: 414): the kirangozi guide, shouldering his load, and holding his red tattered flag furled but ready for floating when spectators turn out to gaze. . . . The dignitary is conspicuous in some wild and wonderful attire; for headdress the spoils of tippet money, a pair of antelope horns, or a wild cat’s skin dangling from forehead to shoulders. . . . His other insignia of office is the kipunga or fly-flapper of zebra’s tail.

The many dangers along the route called for another specialist: the medicine-man or mganga in Kiswahili. He could offer services like divination and ritual protection of both men and ivory. Tusks marked by him with lines, spots, and figures would be safe and eventually arrive at

Entrepreneurs and Explorers from the Heart of Africa •

129

the trading posts on the coast (Rockel 2000a: 184). Considering the many dangers the porters could encounter during the long march, all kinds of protection and guarantees for luck were reasonable. The last specialists to be mentioned here are the askari or armed guards. Authors writing about European caravans o en mention Baluchi mercenaries. They also seemed to be preferred by Arab caravan leaders, while Swahili caravans were o en protected by armed slaves. The Nyamwezi caravaneers took care of themselves, which meant that they both had to carry the trade goods and the weapons like bows and arrows or rifles for their protection. The different ways to solve the problems with protection and security indicate important differences in the three types of caravans identified already by Burton (see above). The self-employed Nyamwezi played multifunctional roles in their caravans. Arab and Swahili traders, whose caravans were composed of hired porters and/or slaves, needed either paid soldiers as the Baluchi mercenaries or slave-soldiers. Some of these slave-soldiers, called waungwana, had been kidnapped or purchased as young boys and trained to extreme loyalty toward their owners but also extreme ruthlessness toward other people. They became among the most brutal slave hunters, particularly in the Manyema area in eastern Congo, which will be further explored in the next chapter. The caravan organization developed by the Nyamwezi was copied by caravans that originated from the coast. One simple reason was that Nyamwezi porters, who had arrived in Bagamoyo beginning from June to August, were the most sought-a er porters in the caravans that departed in November/December from the coast. Throughout the era of the caravan trade, they either operated as traders/porters who carried back the merchandise they had obtained in exchange for the ivory and slaves they had brought to the coast, or as porters who exchanged their labor for trade goods on the market.

Female Caravanners We can see in the quotation above on the marching order that even women and children took part. There are many accounts on this and even photos of Nyamwezi porters in Bagamoyo, some of whom were women. An early testimony from the end of the nineteenth century from the missionary Johann Ludwig Krapf, who met a caravan from the Kimbu area in the greater Nyamwezi complex, tells us: “These people who came with their wives and children from the interior and lived in small huts by the seashore, told me that they spent three months on the journey, and had brought slaves and ivory” (Rockel 2000b: 756f.). Another contemporary observer, Hans

130

•

Muted Memories

Hermann von Schweinitz (1893: 481), adds: “The Mnyamwezi with wife and children annually move off for the coast, o en the whole family, o en the entire village, leave their home area behind not to return before half a year or three quarter of a year” (as quoted in Brandström 1986: 2–3). Another source also details some of the duties performed by woman caravaneers: women are of immense help to men, and consequently to the leader of an expedition. The porter, loaded with his boxes or bags of sixty pounds, his rifle and ammunition and mat, has quite enough to carry through eight hours of marching . . . arrival at a camp she prepares his evening meal, gets the camp ready, and, if necessary, washes his clothes for him, and helps in a hundred ways her tired husband. (Stairs 1891: 958)

The Nyamwezi type of caravan trade was o en carried out as a family enterprise or by kinship groups. The two sources of information above describe the caravan women as wives, and it seems reasonable to assume that they moved with their husbands and children between their upcountry homes and the coast. One reason for categorizing female caravan workers as wives is that independent women who a ached themselves to the caravans sooner or later found themselves in matrimonial relations. No information has been found on the existence of independent female porters who traveled and traded. Women usually moved ahead of the main column.11 They carried out many different roles such as establishing the first contact between the caravan and the societies they had to pass on their way to the coast or to their up-country destinations. They collected information on these societies and the members’ taste regarding trade goods and they also, above all, get hold of food for the porters. Their role in the procurement of food should be seen in relation to the scarcity of food in some places along the central routes, particularly in the almost uninhabited Magunda Mkali and Marenga Mkali areas in central Tanzania (see Map 5.1). Caravans from the Nyamwezi area had to purchase food for eight days before they entered Magunda Mkali on their way to the coast. When they reached Kilimantinde on the eastern side of the area they had to get hold of food for another five days to pass through Marenga Mkali. Women were indispensable in the search for food in the local communities.

The Acquisition, Transportation, and Classification of Ivory The whole caravan trade in ivory depended, in the last instance, on elephant hunting, which, in turn, depended on specialized skills. The recruitment of elephant hunters was more open than recruitment for iron-work, which

Entrepreneurs and Explorers from the Heart of Africa •

131

was monopolized by specific cast-like groupings, but more restricted than becoming a porter (Roberts 1970: 68). The hunters were organized in exclusive societies, and membership was more regulated by experience and skill than kinship ties. The hunting of elephants demanded skill, experience, and also magical support. A hunting expedition was led by a fundi who knew the special medicines and songs needed to a ract the elephants and, one may assume, also to make a hunt successful (Roberts 1970: 68).12 The fundi was the leader of a group of five to six men called bahemba (apprentices). Under the leadership of the master the neophyte learned the skill to kill elephants. When the neophyte had killed several of them he may have been given the insignia and hunting charms, which qualified him to be a fundi himself (Roberts 1970). As an apprentice he could also be a part-time porter and trader and carry tusks to the coast and cloth or beads back home to trade with along the route or in the up-country communities. Elephant hunters among the Nyamwezi, wayege, established a guild known as the uyege. Their importance in the society may be measured by the role they played in the ceremonies that accompanied the installment of a new ntemi or king. Before he was fully recognized as the new king, he would sit on the carcass of an elephant and would have his hair shaved. The elephant was specifically killed by the wayege for this purpose: “By this practice the ntemi was in essence initiated into the guild of hunters. In other words, membership of the uyege was regarded as a necessary status which the ntemi must a ain before he could be accepted (Unomah and Webster 1976: 305). But the supply of elephants and, hence, tusks, was not infinite. The Nyamwezi hunters had to search for their prey ever further afield in what today is eastern Congo and southern Uganda. Complications arose, however, when the border was demarcated between German East Africa and the Congo Free State, later Belgian Congo. European interests began to have an influence on the caravan trade. Another consequence, as already mentioned, was that the Belgians tried to divert the tusks from the Congo area to the ports on the Atlantic coast, away from the Nyamwezi porters and Bagamoyo traders. Additional interference was experienced when the German colonial administration in 1894 tried to impose restrictions on elephant hunting (Pallaver n.d.: 13). The effects are seen in the amount of ivory brought through Tabora, the major up-country trading post situated in the Nyamwezi area. Pallaver (n.d.: 14), referring to German colonial records from Tabora, reports that 80,000 kg of ivory was registered in 1891. In the season 1900–1 it had decreased to 25,000 kg, only to decrease further to 13,000 kg in the season 1905–6.

132

•

Muted Memories

The ivory brought to Bagamoyo was classified according to quality and weight. Two different categories were used: gendi meaning hard ivory, and pembe for so ivory (W. T. Brown 1971b: 634). The so type, which is opaque and easy to carve, demanded a higher price since the hard type was difficult to cut. The tusks were also classified on the basis of their weight. The smallest of the so type, weighing less than 10 kg (25 lbs), were called pembe calasia. A medium size tusk around 15 kg (40 lbs) was called pembe bab Cutch, while the largest at more than 15 kg were called pembe bab Ulaya. In general, a tusk weighing 13 kg (a frasila) of the so type fetched three times the money of a hard type of similar weight (W. T. Brown 1971b: 634).

The Cloth and Bead Economy The bulk of the cargo carried from the coast by human bearers was co on cloth, most commonly either undyed sheets from the United States called merikani or dyed cloth, o en from Britain but dyed in India, called kaniki. Cloth played a multifunctional role in the caravan trade and was the prime trade good brought into the heart of the continent to be exchanged for ivory or slaves. Cloth was, during the early phase of the caravan trade, the only means of payment accepted by porters and used by them to obtain other goods locally once the caravan season was over. Furthermore, the porters demanded cloth as a daily allowance (posho), which they used to obtain food, water, and services as they moved across the landscape. Standard payment was a six-foot (one fathom) piece of cloth called upande or shuka per day per porter. Cloths were also used to express identities and status or to support claims to elevate the owner’s status. Last, but not least, cloth was used by the caravan leaders to pay the hongo tribute or tax to the leaders in the societies they passed through in order to be allowed to pass, draw water, buy food, collect firewood, and camp. The imported cloths were of two main kinds: monochrome textiles like the undyed co on sheeting called merikani or the dark-blue dyed kaniki and the multicolored cloths called “cloths with names.” Both types usually came as full-sized body wrappers. The monochrome types were directly imported from United States or from Britain via India as in the case with kaniki, while the multicolored types, during the first half of the nineteenth century usually came from western India or Oman. The two types of cloth not only differed in their colors and prices (the multicolored textiles were up to one hundred times more expensive than the monochrome cloths), but also how they were integrated in East African societies. The monochrome types became the most important means of exchange between

Entrepreneurs and Explorers from the Heart of Africa •

133

porters and traders in the caravan trade and between the caravan people and the people in the societies a caravan passed, while the multicolored types achieved much more of a status function and were used to express power and prestige but also as part of the ritual exchange caravan leaders had to undertake when they gave gi s or road tax, hongo, to the leaders of the societies a caravan passed. Reports from “explorers” show that “‘colored cloth’ accounted for up to 10 per cent of caravan total outlay costs and 25 per cent of porters’ loads” (Fee 2012, 78). The merikani gradually became the principal standard of value in the caravan trade. During the first decades the standard measure was the shuka or a two-yard length of cloth. It got its name from the piece of cloth men wrapped around the waist. One shuka could be divided into four mikono or hands and used as fractals in, for example, the purchase of food. It may be assumed that the growth in the caravan trade triggered the need for another measure than the shuka. The doti became the major measuring unit during the second half of the nineteenth century. A doti is a twelvefoot length of cloth that is the piece of cloth needed by women to cover their hair and bodies. This measure, which has its name from the most common dress among women, was usually cut into two shuka (Sissons 1984: 42–52) in a similar way that the kanga cloth was cut when this colorful textile became popular among women in East African societies. The largest unit in the cloth economy was the “piece” or jorah. It could vary in length, but a standard measure was thirty yards (Sissons 1984). The multifunctional role of cloth in the caravan economy meant that an enormous amount of bales or rolls with cloth had to be transported by each caravan that set out from the coast. The goods were usually packed in bales or rolls sown up in coconut-fiber sheets and carried on the head or shoulder of the porter. The standard weight of a bale was 60 lbs or a li le less than 30 kg. Such a roll contained a li le more than 100 m of cloth. Heavier cargo could be transported on a carrying pole (mandala or zega zega) slung between two or more porters. The heavy loads carried by the porters affected the speed of the caravan and the distance it could cover per day. A standard distance was between 12–15 km per day during the 4–5 hours they marched from sunrise to midday (non-trading caravans with lighter cargo could move twice as fast) (Beidelman 1982: 609–10). The use of cloth as a means of exchange and for the accumulation of wealth gave a specific character to the caravan culture in East Africa. Each caravan, organized by traders, travelers, missionaries, or colonizers, had to bring large quantities of cloth with their caravans to pay their porters, to obtain food, to purchase local trade items such as ivory and slaves and to pay tolls and taxes along the caravan routes. Mtoro, the Bagamoyo-born writer, refers to different types of cloth in his book. The different varieties

134

•

Muted Memories

are identified by their place of origin, quality, and price ( J. W. T. Allen 1981: 297). “Feeding hundreds of porters and guides,” Sarah Fee reports, “required much cloth, creating a major expense—and logistical nightmare—for caravan leaders” (2012, 4). It has been estimated that 80,000 doti per year were needed to procure the 1 million kg of food from the farmers in Gogoland (central Tanzania) to feed the caravans which passed through the area (Sissons 1984: 191).13 The amount of cloth needed for coastal caravans to reach Unyanyembe in Nyamweziland or Ujiji at Lake Tanganyika depended on both the speed of the caravan and the availability of food in the communities through which a caravan passed. In 1871, Henry Morton Stanley, for example, reached Unyanyembe a er 84 days. Richard Francis Burton and John Hanning Speke needed 134 days to travel the same distance, while caravans lead by Arab or Swahili traders usually needed between 72 to 120 days to reach Unyanyembe (Stanley 1970: 20–21) On their way to Unyanyembe, at Tabora, in central Tanzania, the caravans on the central routes passed areas struck by recurrent crop failures, which made the caravans pay more for the available food. Stanley on his way through Ugogo in central Tanzania noticed that “we crossed broad bleak plains, where food was scarce and cloth vanished fast” (Stanley 1970: 193). The role of cloth as a medium of exchange was explained already to Vasco da Gama when he visited Calicut in 1498. The Indian Ocean trading network, he was told, depended on goods from various parts of the world that could be exchanged for each other. “To buy pepper in Calicut he needed gold, which he could obtain from Kilwa in East Africa in return for textiles from Gujarat” (Pomeranz and Topik 1999: 228, quoted in Davidson 2012: 7). As mentioned above, cloth has been imported to East Africa at least since the tenth century from Gujarat and Kutch in northwestern India (Fee 2012: 3; Machado 2009: 53f.). “This intra-regional engagement,” writes Machado, “linking Indian weavers with African consumers, was mediated in critical ways by networks of Indian mercantile groups” (2009: 55). Textile producers in India early on specialized their production to meet the demands around the Indian Ocean. The Indian hegemony over the textile trade stemmed from three factors: their control over the supply of co on for the textiles, the technique to use a variety of vegetable dyes, and the availability of cheap, skilled labor (Davidson 2012: 1). The colorful block print co on textiles produced in Gujarat were in high demand in the societies around the Indian Ocean and, together with textiles from Europe, came to influence the dress style in East Africa. Indian traders could, through their control over the textile trade, also control much of the Indian Ocean trade in ivory, gold, slaves, and other

Entrepreneurs and Explorers from the Heart of Africa •

135

commodities since textiles were the only “currency” accepted by traders/ porters from inland Africa in exchange for their trade goods and labor. One of the strengths of the Gujarati textile industry was its capacity for production differentiation, which made it possible for Indian traders to meet changes in local taste (Machado 2009: 59–60). During the first decades of the nineteenth century the undyed merikani co on cloth became increasingly popular in East Africa (Ryan 2013: 85) and American textile mills and traders could temporarily compete with the Indian manufacturers. The effect of the exploding demand for this type of cloth acted, as mentioned, as a catalyst to the mechanization of the textile industries in Salem and Boston in Massachuse s (Fee 2012: 3; Prestholdt 2004). The competitiveness during the 1850s of the American textiles on the East African coast has been described in the following way: Americans dominate the trade of the entire coast, they trade to the ports or at times unload easily as contraband at any place they regard as convenient the goods [which are] suitable for exchange with Africans, such as cloths. They [Americans] introduce be er and cheaper co on goods than those from Diu, Daman . . . which until now were used for this commerce, and which they [American textiles] are forcing out of market . . . (As explained by the Portuguese Carlos José Caldeira, quoted in Machado 2009: 82)

Another item in high demand in up-country societies were beads in different shapes, as pearls, cylinders, and discs; in various colors as red, white, black, blue, or pink; and in different materials as glass, porcelain, and coral. Glass beads functioned as means of payment very similar to the role of cloth. They were used to pay compensations in cases of offenses, to pay tolls and tributes along the caravan tracks, to pay the porters and other caravan functionaries, and they were used to procure foodstuff as well as ivory and slaves (Pallaver 2009). The trade in beads, like the trade in cloth, demanded a deep knowledge about taste, fashion and demands in the societies along the caravan routes. Traders have made the mistake, writes Raymond Wendell Beachey (1967: 273) of “assuming that because blue beads sold well one year, it would do so the next, but found to his sorrow that fashion had changed.” The profit could be jeopardized if the wrong beads were brought to the interior. The bead trade on Zanzibar was, according to Burton, in the hands of Banyan (Indian Hindu) traders from India, who imported beads to East Africa by the ton (Burton 1859: 424). During the time of the caravan trade about four hundred varieties with their own particular name were available on the market.14 It has generally been assumed that the main reason for a caravan trader to be equipped with a variety of these beads was to be able to meet the whims and wishes of the consumers they met along the caravan trails.

136

•

Muted Memories

Beads and wires came to play the role of fractal coins of the cloth currency. One of the advantages of these means of payment was that they could be split up in smaller units, which was rarely done with a piece of cloth. Beads were, for example, used by the independent porters in the Nyamwezi caravans in exchanged for food, while cloth was used to procure ivory and slaves. Caravan leaders from the coast, by contrast, o en used cloth to buy food and water for their porters. The absence of an accepted multipurpose currency stimulated the development of an intricate system of multiple currencies in which the beads played a most crucial role in the transactions, although cloth provided the standard for evaluations. The functionality of the beads depended on their large variation in shape, color, and material. Other products were sold in quantities that were determined by the value of a particular string of beads. One of the early travelers, Captain Stair, found that food for sale in the markets in Tabora was arranged in quantities that corresponded to a particular string of beads (Pallaver 2009: 24). The consumers’ changes in preferences over time and space, regarding the shape, color, and material of the beads may be a ributed to the role beads played in bodily ornaments. Beads were used in multiple ways to construct individual and collective identities and to display the social body. “In the interior of 19th-century Tanzania,” writes Karin Pallaver (2009: 25), “glass beads were used to decorate beards and hair, to embroider cloths, masks and dolls, and to make jewellery, like necklaces, bracelets, and so on.” Items used for the beautification of the human body have since ancient times been made from gold, ivory, silver, stones, glass beads, or other materials judged as precious. The beads were a means of payment and functioned as storage of wealth. One of the main characteristics of the caravan trade was that the movement of peoples, goods, ideas, and wealth, through different economic systems between the interior of the continent and the coast, was the dependence on universally standardized and locally accepted means of payment. Pallaver (2009: 25f.) therefore proposes that the changes over time and space in the preferences for specific types of beads could rather be explained with the variations in the items offered on the markets to the caravans than “be a ributed to the rapid changes to the ‘fastidious tastes of African women’ as Stanley put it” (Pallaver 2009: 26). The four hundred different varieties of beads used in the East African caravan trade were not only means of transactions but also of translation between different economic systems. If co on cloth was the glue and grease in the interface between the continental trading networks and the networks on the Indian Ocean, established through long-distance trade, beads became the glue

Entrepreneurs and Explorers from the Heart of Africa •

137

and grease in the interface between the caravan trade and local economies that linked into the caravan trade. Recent research has shown that people in the interior of the continent not only passively received what the traders, including missionaries, travelers and colonizers, brought to them but “had precise desires that merchants had to satisfy for exchange to occur” (Fee 2012: 1). This globalization of the caravan trade was a result of two interacting processes; an increased demand and rise in the price of what was exported from Africa, in particular ivory, which followed with industrialization in Western Europe and North America, and falling prices on the imported textiles due to an industrialization of textile production.

The Development of Local and Global Markets and Demands The second half of the nineteenth century saw a dramatic increase in the demand for porters. The number of perambulating European and American travelers, traders, colonizers, and missionaries increased, as did the number of caravans organized by Swahili and Arab merchants. This resulted in increased competition for porters and in the bargaining power they had in the struggle to increase wages. Occasionally the demand surpassed the supply, which in turn gave rise to “recruitment agencies” that tried to secure contracts among the porters for their customers. But the caravans not only gave rise to a labor market where salaries and compensations were negotiated. It also gave rise to expanding trade between various local economies in the area. Alfred A. Tucker reports how people could accompany caravans set up by missionaries stationed in the interior of the continent destined for the coast only to obtain goods there, which could be bartered, preferably for ivory, food, or slaves, in up-country societies: A drum was sent out to all the villages round in order that by its “safari” beat the people might know that there was a caravan going to the coast. On hearing it they would throng out of the villages with eager enquiries: who is going to the coast. The drummer, on such occasions, would usually return with a crowd of followers—some anxious to carry loads to the coast—others wishful simply to follow in the [missionaries’] train, as a protection. It was no uncommon thing for a Missionary to have twenty paid porters and two hundred unpaid followers. The men who did no work paid their own way down-country by selling tobacco or spades (hoes) of their own manufacture . . . The main object of porters and followers alike was to get to the coast and to carry back a load for usually a good wage was paid. (Tucker [1911] 1970: 57)

138

•

Muted Memories

The ivory trade seems have had positive effects along the central caravan route (Roberts, 1970: 51). Ivory was not in great demand in the up-country communities, but could easily be traded with merchants on the coast or with caravan traders visiting societies in the interior. The trade greatly expanded the markets for most kinds of goods by both bringing new items to trade that came through the caravans, but also stimulating the production of local goods when the demand for them increased. But it also had negative effects for the people and areas the caravans passed through. The intensification of food production, which o en led to the cultivation of marginal land, sometimes in combination with the sociopolitical turmoil that followed in the footsteps of the expanding caravan trade, resulted in some areas in an environmental degeneration (Biginagwa 2012; Håkansson 2004; Håkansson, Widgren, and Börjesson 2008). Roberts notes that the trade in ivory also stimulated trade in slaves. One source indicates that slaves from Unyamwezi were sold on Zanzibar already in 1811, although Swahili and Arab slave hunters and traders did not regard the western parts of today’s Tanzania as prime hunting ground for slaves. “This was due partly to the valuable links which had been developed between local trade and caravan trade and partly to the opening up, in the 1870s and 1880s, of new areas of exploitation” (Roberts 1970: 59). Hence, it seems like the interconnections of the two economic circuits spared the Nyamwezi area from excessive slave hunting. The slave trade that existed in the area “chiefly served to meet the demands for slave labour within the interior” (Roberts 1970: 59). The Nyamwezi society was an importer rather than exporter of slaves (Deutsch 2006: 26). The massive economic expansion did not, however, lead to a thorough reformation of the economic organization and its institutions. Except for the market for labor and some items, the local economies were never transformed from subsistence into a market economy. “Markets themselves were not,” writes Roberts (1970: 63) “on the whole, characteristic of the region.” Hence, trade was not carried out on specific markets or market places, “but in ad hoc gatherings at a caravan halt, or by visiting a particular place of commercial interest, such as a center for salt or iron production” (Roberts 1970: 64). A second piece of evidence on the limited economic transformation toward a general market economy was the fact that the porters rarely, if ever, accepted money for their ivory or in exchange for their labor. General purpose money as the Austrian Maria Theresa (MT) Thaler or dollar (riale), a silver coin, was most widely accepted along the coast. It was first struck in 1751 and became widely used in the Arabic trading networks in the Indian Ocean and along the Red Sea.

Entrepreneurs and Explorers from the Heart of Africa •

139

Further south, on Zanzibar and along the Mrima coast the Indian Rupee15 (rupia), worth half a riale, was more in use as a result of the dominance Indians had over trade and the financial sector (Sissons 1984). Both the Thaler and the Rupee, with its fractural coins the Anna and the Pice or Pesa, had fixed values. Less frequently used and accepted was the Spanish Dollar or piaster with a floating value. None of these currencies were accepted in trade with people in the interior of the continent (Pallaver 2009: 21). Instead they asked, as mentioned, for cloth, particularly merikani, copper and iron wire, weapons, ammunition and other items. In this sense the economy never became thoroughly monetized. The Nyamwezi long-distance caravan trade not only influenced the market and shops in Bagamoyo, but also the industrialization of parts of America and India. It is conventionally assumed that East Africa imported ready-made industrial products and exported raw materials (Machado 2009: 55). Studies have, however, demonstrated the opposite (Prestholdt 2004), since the taste and demands in up-country societies affected the global trade in a considerable way.16 It is well-documented how traders bought goods, particularly cloth, from England and America and customized them in India a er the latest news from African up-country societies concerning the goods in fashion and demand there (Prestholdt 2004). The remaking of industrial products also took place on the African continent. The artisans in the coastal towns were important in this process. Local weavers would, for example, reap up imported textiles from India or China in order to get access to gold or silk threads that they wove into locally manufactured textiles in order to both meet the local demands and increase the value of their products (Prestholdt 2004: 766). Even the caravan porters took part in the a empts to meet the local taste, which may shi from one area to another. Prestholdt writes: Since fashions changed quickly, it was necessary to redesign pieces of cloth or restring beads on the road. Caravan leaders purchased the kind of cloth or beads they thought to be in demand in a particular locale, and, just outside the area where they intended to trade, they would stop and redesign cloth in a way that appealed to the local customers. (Prestholdt 2004: 765)

American traders on Zanzibar could report home that the demand for the unbleached cloth surpassed the supply. The merikani was the most important trade item to East Africa from the United States from the 1830s to the Civil War during the 1860s. “Throughout the 1850s,” writes Mackenzie Moon Ryan (2013: 89), “two to three dozen American vessels carrying between 7,000 and 10,000 tons of goods docked each year in Zanzibar. The vast majority of imports came in the form of domestic unbleached co on sheeting from the co on mills near Salem, Massachuse s.” Prest-

140

•

Muted Memories

holdt (2004) hypothesized that this trade stimulated the mechanization and development of the textile mills in Salem and Boston through the introduction of steam machines. This example shows how early globalization had both centripetal processes—binding together the demands and taste of people in East African up-country societies with industrial areas in America; and centrifugal effects in distributing products over the globe (see also Fee 2012: 3). The US dominance in textile import came to an abrupt end in the Civil War. One reason was that the Boston and Salem mills could no longer compete with the cheap Indian textile industry. The US cloth had been produced from co on grown by slave labor on the plantations. The production price increased a er the war and a er the abolition of slavery in the United States. As a consequence, India, once again, took over as the main producer of co on cloth and Bombay became the main exporter of unbleached cloth to East Africa. Prestholdt (2004: 777) reports that figures from 1890 show that nearly half of the annual exports from Bombay, or 13 million yards of cloth were exported to Zanzibar. The same year traders on Zanzibar also imported 9 million yards of cloth from England, bringing the total amount that year to 22 million yards. Zanzibar was the single largest export market for umbrellas, furniture, and glassware manufactured in Bombay. In this way the taste of East African consumers, mediated by porters in the caravans, contributed to the industrialization of Bombay, in a similar manner as they had done forty years earlier in the industrialization of Salem and Boston. The US Civil War, on the other hand, triggered the taste and pa erns of printed co on textiles on the other side of the globe. The decline in the supply of durable white merikani inspired manufacturers and traders from Europe to supply their own version of this popular trade item and British and British Indian manufacturers tried to improve their substandard products by adding a white substance to the cloth “to bolster the perceived weight and quality of imposter merikani” (Ryan 2013: 87). Another invention to improve the appearance of this kind of textile was to dye and print designs on the cloth which may have led to the use of woodblock prints in the area which in turn may have resulted in the earliest form of East African kanga textiles (Ryan 2013: 87) (see chapter 11 for further information on this development). Nyamwezi traders/porters dominated the caravan trade along the central routes until the middle of the nineteenth century when coastal traders from Bagamoyo, and by extension Zanzibar, gradually took over. This was partly the result of the increased dominance from the Sultanate on Zanzibar which imposed a new tax system which favored traders from the coast, at the disadvantage of traders from the interior of the continent

Entrepreneurs and Explorers from the Heart of Africa •

141

(Deutsch 2007: 22). Another reason was the decimated elephant herds, which moved the prime hunting ground further west on the continent and thereby beyond the direct control of the Nyamwezi. A third, and what Deutsch mentions as perhaps the most decisive reason, was that coastal traders had se led in strategic up-country se lements to gain control over the caravan trade (Deutsch 2007: 22). The final blow not only to the Nyamwezi but to the whole caravan trade came when the German colonial government in 1912 opened the railway between Dar es Salaam and Kigoma at Lake Tanganyika and Mwanza at Lake Victoria.

Notes  1. The name, which means “people from the west” was probably applied by coastal traders to most of the people on the central plateau of present-day Tanzania, and communicated to European and American travelers. The bulk of the Nyamwezi porters, however, originated from the area around Tabora in what is known as Unyamwezi, the land of the Nyamwezi. In this text the ethnonym Nyamwezi is primarily used for the Nyamwezi proper, but occasionally also for the other groups in the larger cluster.  2. It is here assumed that Burton contrasts the free Nyamwezi porters with the slaves used as porters in caravans from the coast.  3. No wonder porters put down their hearts and loads in joy when they reached Bagamoyo.  4. The acquisition of slaves for the caravans and plantations on the coast will be in focus in the next chapter.  5. Indian beads dating back to the fourteenth century found in the Great Lakes Region indicate that trading networks between the interior of the continent and the coast may be much older (Chami et al. 2004: 25).  6. Globally, salt had been an important trade item and means of exchange (see Kurlansky 2002). Iron hoes have long been important trade items and used in barter economies on the African continent.  7. People and livestock were both a source and measure of wealth, which made it of interest for those in power to try to restrict their movements.  8. The specific caravan culture which developed among European and American travelers, traders and colonizers will be further explored in chapter 12.  9. See his note 4 on page 66 for a number of references on caravan life and culture. 10. The German linguist Dr. Carl Velten asked Mtoro bin Mwinyi Bakari (herea er Mtoro) in the 1890s to record the traditions and customs of the Swahili people. Mtoro was the chief editor of the Desturi za Waswahili (Customs of the Swahili people). He may have come from the Zaramo people, who inhabited the area around Bagamoyo where the first German colonial capital was established. Mtoro was also a Mswahili or Muslim coastal person as well as

142

11. 12.

13.

14. 15.

16.

•

Muted Memories

a Mbagamoyo, an inhabitant of the town. He is an ample illustration on how a person in old Bagamoyo could have multiple identities (see chapter 10). Mtoro’s texts were compiled by Velten and published in 1903. The text was later edited and translated to English by J. W. T. Allen and published in 1981 by the University of California Press (Ryan 2013: 75). See also Burton 1860: 1, 348. The names of some of the most well-known hunters were recorded by Unomah during the 1960s. The most famous got the nickname Kapiga miti (he who shot down trees); others were known as Kapanda, Mlivbe, and Mkuvba (Unomah and Webster 1976: 282). In order to supply a hundred men with food during two years, Henry Morton Stanley estimated that “4,000 doti = 16,000 yards of American sheeting; 2,000 doti = 8,000 yards of Kaniki; 1,300 doti = 5,200 yards of mixed colour cloths” were needed (Stanley 1872: 22). See Burton (1859, 1: 225–27) for a description of some of the varieties. The German East African Company used the Rupee as their standardized measure of value on the first coins minted in the 1880s by the Company in its a empts to get control over the caravan trade. Machado (2009: 55) suggests that East-Central and South-East Africa together with Gujarat in India constitute “. . . regions with connected Indian Ocean histories.”

6  Pawned, Preyed Upon, Purchased, or Punished Slaves and Slavery in Nineteenth-Century East Africa Not all Nyamwezi men and women who arrived in Bagamoyo were free traders and porters. Some were slaves. Nineteenth-century Bagamoyo was a slave society. A slave in this society was someone who was owned by a member of the society, or even by another slave,1 and therefore lacked the right to take a place in the ranks of the “owners” of that particular society, culture, and territory. A main characteristic of many slaves was that they were “socially dead,” since they had neither ancestors nor elaborate kinship relations or links to the land and group that could be used to claim to belong to that group. It was also commonly held that slaves, at least slaves who had arrived recently at the coast, were not even capable of creating their own cultures and communities, and they were scorned for not knowing how to behave or speak in the society in which they lived. That is why they could easily be defined and tolerated as social and moral inferiors even in the strictest system of categorization, since they fell outside ordinary society and life. They belonged to members of that society but did not belong in that society.

144

•

Muted Memories

As an institution, slavery was socioeconomically embedded in the society but did not define it. Rather, the society defined its form of slavery. This institution was integrated with other social, economic, and political relationships, such as family and kinship relations of dominance and submission, patron- and client-ship, status of insiders and outsiders, and principles of being “civilized” or “barbarian,” human or dehumanized. Slavery played an integral part in the US capitalist plantation system, as well as in the African domestic or lineage-based production systems. The boom in the caravan trade in ivory coincided with the boom in the slave trade in East Africa. An increased demand for slaves on the newly established plantations along the coast, and on the islands in the Indian Ocean under the Sultanate, swallowed up an ever-increasing number of slaves. The expansion was facilitated, as will be shown, by the existence of older forms of relations of dependency and servility (Deutsch 2006; Kollman 2005; Médard and Doyle 2007; Meillassoux [1986] 1991; Wright 1993) and the existence of institutions like the slave markets on Zanzibar through which most of the slaves for sale passed. Slave trade along the East African coast is, however, much older than the caravan trade explored in this text. O -cited evidence is the so-called Zanj rebellion between the years 869 and 883 in present-day Iraq.2 The majority of the slaves brought to Arabia and Persia via Lamu, a Swahili trade port in northern Kenya, came from Madagascar (Vernet 2009: 46). It was first with the development of the French plantation economy on the Mascarene Island and the translocation of the Sultanate from Oman to Zanzibar that Kilwa in southern Tanzania grew to become the most important slave post on mainland Africa. It has been estimated that more than three-quarters of the slaves brought to Zanzibar during the 1860s came through Kilwa (Cooper 1975). Edward Alpers reports ([1967] 1974: 8) that a British captain estimated that in 1811 some 6,000–10,000 slaves were sold annually at the slave markets on Zanzibar. Half of them probably came from Kilwa. Zanzibar was a well-established slave market3 since the boom in the slave trade that occurred a century earlier with the establishment of plantation economies on the French islands Ile de France (Mauritius) and Bourbon (Réunion) in the Mascarenas archipelago. Frederick Cooper (1975: 87) concludes that “the slave trading business, not just the use of slaves, was a fact of Zanzibari life throughout the first half of the century. It remained so until at least the 1870s.” He suggests that 10,000 to 15,000 slaves were sold in Zanzibar’s slave market during the first half of the nineteenth century, while Abdul Sheriff estimates that approximately 25,000 slaves arrived to Zanzibar during the 1860s and 1870s (Sheriff 1987: 226ff.). The plantations demanded a much larger labor force than was found on the islands. The

Pawned, Preyed Upon, Purchased, or Punished

•

145

French slavers first turned to Madagascar and the Portuguese colony of Mozambique in their search for slaves. Soon, however, they turned to Zanzibar in search of a more reliable supply. The French eventually turned to Kilwa since most of the slaves sold on Zanzibar came from this port town.4 The explosion of the slave trade during the second half of the nineteenth century took place in parallel with the efforts, particularly by the British, to abolish slavery. Their strategy was to persuade the Sultan on Zanzibar to take measures to forbid the trade in slaves, which, by extension, would imply the termination of slavery as an institution. An anticipated side effect of abolition was the collapse of the Arab plantation economy, which would open up the area for European traders and se lers.5

The Atlantic Metanarrative Much of the global picture of the slave trade has been modeled on the Atlantic trade. In a similar vein, the American plantation system has o en informed our understanding of slavery as a production system. As noted by Alpers: “As in the historiography of the African slave trade, the Atlantic dominates” (Alpers 2007: 20) in the creation of our knowledge and view about slavery and the slave trade. However, the 400-year-long epoch of the Atlantic slave trade and gang system of slavery on US plantations was a rather unique form of slavery and trade in human history (Deutsch 2007; Glassman 1995: 80). The form of slavery, which developed under the capitalistic plantation economy in the Americas, differed in many respects from the forms of slavery developed in other economic and political systems. One reason for the differences was, as mentioned, that it is not slavery that determines the type of society and economy but the other way round: the type of society and economy determine the kind of slavery in that particular society and to what extent the slave was an alien, a permanent stranger or “outsider,” or a kind of property that could be gradually integrated in the owner’s family, kin group, and society. One major difference between the American/Atlantic type of slavery and many forms of slave systems found on the African continent concerned the distance between masters and slaves. In the “closed” American type of slavery, the slave was an alien and dehumanized stranger who had no chance to integrate in the owner’s family or kin group. In many, but not all, African slave systems, the slave could gradually be adopted into the master’s family and kin group. The specific characteristic of the US slave system has sometimes been used in a empts to define slavery as an institution. It has, for example, been suggested that slavery “is beyond the

146

•

Muted Memories

limits of family relations” (Brion Davis [1966] 1988: 31). Such a definition is not valid for many African slave systems. Another difference between the systems concerned the barriers established between master and slave, and the open racial violence that characterized the US slave system. Here master and slave were characterized by their skin colors. This difference was rarely a prominent feature in many slave systems on the African continent. “Race” was not a prominent element in the social construction of differences between segments of the precolonial societies. Furthermore, the sharp divide based on skin color in the Atlantic region was blurred in the Indian Ocean region as Circassian female slaves from the Volga area were in high demand in some societies around the Indian Ocean, as in the harem of the Sultan of Zanzibar. One of his daughters with a Circassian concubine, Princess Salme bint Said ibn Sultan al-Bu Said writes in her autobiography that the Sultan had only one horme (legitimate wife) while the rest of the women he held in his harem were sarari (surîe in sing; suria in Kiswahili) or concubines (Said-Reute [1980] 1994: 5). The racist ideology, which not only made slavery possible but also motivated it in the Christian (US and European) world, was built on a series of arguments. One was that violence was justified or even necessary to obtain heathen slaves and make them work. As a consequence of this, the US racist ideology also developed a defense for slavery in the United States with its liberal constitution from 1787, which included a declaration of human rights. It was claimed by the defenders of slavery that “in the midst of the horrors of slavery and the slave trade, the masters had, in part at least, performed the office of advancing and civilizing the negro” (Brion Davis 1966: 24, referring to George Bancro 1862: 3:408). This kind of justification is generally not seen in Africa south of the Sahara except for the East African coast, for example, on Zanzibar, where the learned class during the late nineteenth century developed the argument that slavery was good for mainland African (Glassman 2004). This line of thinking maintained that slaves originated from uncivilized societies, which became an argument for the civilizing effect that slavery was supposed to have on the slaves. Here members of the learned classes could argue that “slavery was the very institution that had done the most to civilize African ‘natives’, imposing the control necessary to channel their anarchic energy into productive labour” (Glassman 2004: 746). This argument was built on the strict categorization of the “civilized” versus the “uncivilized,” rather than on skin color. Slavery was recognized by the Koran, and “bringing barbarians into the light of religion and civilization has once been a key justification of enslavement in Islamic doctrine” (Glassman 2004: 746). Although slaves in some Muslim contexts were “outsiders,” “brute beasts,” and, as a proverbial expression explains, the

Pawned, Preyed Upon, Purchased, or Punished

•

147

“enemies of God and the Prophet” (Glassman 1995: 79), slavery, under these circumstances, could even benefit the infidels since the institution could be seen as a “method of religious conversion and indoctrination for those who were not Muslims” (Lovejoy 1981: 17). It was therefore a good act to take a slave from the world of the pagans since it would “save him from death and make him a Muslim” (Cooper 1981: 283). The civilized-barbarian dichotomy, so fundamental for the ruling classes on the islands in the Zanzibar archipelago and in the coastal towns, became an argument for slavery and for the integration of slaves into society. Researchers have frequently pointed out that Islam and Sharia law protected the rights of the slave in Muslim societies. Slaves had the right to food and shelter even at an advanced age, and if the food was insufficient or the slave was mistreated, she or he had the right to demand a transfer to another slave owner. The law also stipulated that a child who was the result of a relation between a slave woman and her free owner should not be sold, and that such a woman at the death of her owner should be free (Sheriff 2005: 114).

Studies on African Slavery The history of the Western world’s view on slavery and slave trade is complex. During the scramble for Africa the misdeeds of Arabs as slave hunters and traders were reported in European/American travelogues and other reports from their meetings with the interior of the continent. In particular, the mention and portrayal of slave children (Alpers 2009: 27)6 served to motivate European interventions to save the hapless Africans from the cruel and cunning Arabs who, at that time, dominated the caravan trade. The Brussels Anti-Slavery Conference Act of 1890 on the abolition of slave trade and slavery helped to justify the European colonization of the continent. There is extensive literature on slavery in Africa (see, among others, Alpers [1967] 1974, 1975, 2007; Campbell 2005, 2006a; Cooper 1975; Goody 1980; Kopytoff 1982; Lovejoy 1981, 1983; Médard and Doyle 2007; Meillassoux [1986] 1991; Miers 2006; Miers and Kopytoff 1979; Robertson and Klein 1983). It may be assumed that some of the peculiar characteristics of slavery in Africa motivated researchers to try to define what slavery is in general, and on the African continent in particular. A common theme in both the a empts to characterize “African slave systems” and their integration in respective societies is what has been called the “domestic form of slavery,” which included the potential for the slave to gradually integrate into the owner’s family and kin-group and eventually also into that society. Another feature, which applied to some forms of East African

148

•

Muted Memories

coastal slave positions, was that some of the slaves were entitled to take part in the caravan trade and generate their own wealth as traders, which meant that they had the right to hold property, including other slaves. In his classical work on the history of the East African coast, Reginald Coupland (1938, 1939) suggested that the expansion of the slave trade had its origin in the penetration of Asian traders along the coast. The slave trade, however, was facilitated by the existence of the institution of slavery in East African societies. The combination of an external demand for slaves exported from the East African coast through the trade links over the Indian Ocean, coupled with an internal institution of slavery, had contributed for millennia to the low population density in this part of the continent. A similar interpretation is offered by Alpers ([1967] 1974: 4) who argues that East African slave trade was recent, and had its origin in the first half of the eighteenth century. According to him, two factors contributed to this development. One was the involvement at this time of Arab slave traders and hunters; the other was the establishment of a plantation economy on the islands in the Indian Ocean and along the East African coast. This economy depended on slave labor, which in turn encouraged slave hunting and trade. It should be noted that both of these factors were external, but they depended on the ancient internal institution of slavery in the East African societies, which fueled the rapid development of the slave trade during the nineteenth century. The interrelationships between the existing forms of internal slavery and the external demands for slaves to, for example, work on the plantations on the coast, was captured already by Mtoro who wrote at the turn of the nineteenth century (J. W. T. Allen 1981: 169) that “prisoners of war, pawns, and persons taken in adultery whose families cannot redeem them become slaves. Arabs and other slave traders go inland and buy them and bring them to the coast and sell them to others.” Reports show that slavery was common in the early nineteenth century in Nyamweziland, evidence of the ancient history of slavery as an institution in the area, which is of interest for this research. Hence, it cannot be argued that the slave trade here developed as a response to a growing caravan trade that transported slaves from the interior of the continent to the coast and the societies around the Indian and Atlantic Oceans. Slavery was an endemic institution that existed in the area before the traffic between the interior of the continent and the East African coast began (Deutsch 2006: passim; Goody 1980: 28; Médard and Doyle 2007: 7).

Classification of Systems of Slavery As mentioned above, various a empts have been made to classify and characterize different forms of African systems of slavery. Crucial determining features are: the employment of slave labor, domestic or on

Pawned, Preyed Upon, Purchased, or Punished

•

149

plantations/farms; how closed the system was, options for integration or total exclusion; and the hierarchical order, human beings or dehumanized brutes (see Alpers 2006: 51). The relations between the various forms were, under certain conditions, interlinked. An example is the increase in slave trade during the booming caravan trade. People who became slaves, for example, through debt systems or pawn-ship (see below) in societies with a more domestic form of slavery, were sold, in increasing numbers, to caravan traders who brought them to the coast, as explained above by Mtoro. For the slave, this meant not only a shi in owners or place of habitation, but also a shi of system, from the domestic form to the plantation form of slavery (Kollman 2005; Wright 1993). Frederick Cooper (1975), in his now classical study “Plantation Slavery on the East Coast of Africa in the Nineteenth Century,” elaborates on the various types of slave systems found in the area. He maintains that a slave in the pre-plantation Muslim societies along the coast was a social inferior but not associated with a particular place in the economic order (Cooper 1975: 8). They carried out many different tasks in society and strengthened the prestige and political power of individual owners and competing descent groups. With the introduction of the plantation economy many slaves found themselves reduced to farm laborers. A controversy in our understanding of slavery concerns the supposedly “benign” character of some African slave systems. This was frequently touched upon in the travelogues of early travelers. Richard Francis Burton together with David Livingstone, both of whom frequently wrote about slavery and the slave trade, have been most influential in shaping somewhat contradicting views of African slavery. Burton, when comparing the situation of slaves or “property” and free caravan porters, was of the opinion that: “The ‘property’ is well fed and li le worked, whereas the porter, belonging to none but himself, is le without hesitation to starve upon the road-side. The relationship is rather that of patron and client than lord and bondsman; the slave is addressed Ndugu-yango, ‘my brother’ and he is seldom provoked by hard words or stripes” (Burton 1859: 351). As will be shown below, the acquisition of slaves was o en a means to increase the local labor force and also, through their incorporation in the owner’s kinship group, a means to reproduce the number of kinsfolk more rapidly than what natural reproduction would permit. The fact that a slave could refer to his or her owner as “father,” and the owner refer to the slaves as “children,” has been taken as evidence of a “benevolent” type of slavery, which has also been called “domestic slavery,” since it was assumed that the slaves mainly performed domestic services and were assimilated into the owner’s household. It has also been said that the system was “open,” since the slaves could be integrated into the owner’s kinship groups in contrast to the “closed” American plantation slave system.

150

•

Muted Memories

What was not fully understood by these early observers was that the idiom of kinship relations was used to express relations of dominance and subordination, and relations of seniority, and that the slaves were lumped together with other “legal minors,” i.e. women and children. The application of the kinship terminology, however, opened windows for the slaves to negotiate and renegotiate their position. From having been aliens who lacked kinship relations with members of the societies and households of the owners, many slaves, in particular women (Deutsch 2006: 5) in the “lineage mode of production” system, seem to have been gradually transformed into accepted members of their new societies. This argument, based on the “slavery-to-kinship continuum,” introduced by Suzanne Miers and Igor Kopytoff (1979: 22), captures the feature that “earning personal freedom was an in-built feature of the various forms of pre-colonial slavery” (Deutsch 2006: 7) in East Africa. The concept “slave-kinship continuum” can be interpreted in two ways. One captures the gradual transfer from being a slave into becoming a kinsman. There is enough ethnographic evidence to show that this transformation, which Jonathon Glassman (1991: 283) calls “the reduction of the slave’s marginality by his gradual absorption into the kinship and community structures of the non-slave population,” occurred in many African societies.7 The other interpretation of the concept is as a continuum with two poles with a number of variations in between. It captures different degrees of dependence and bondage between the master and slave positions, common in many precolonial African societies. This is just another example of the differences between the closed American slave plantation system, and some of the African systems, as will be further developed below in the presentation of various categories of slaves in precolonial Bagamoyo. Here we have to distinguish between slavery and slaves. Slavery was an integrated institution in many East African societies. Individual slaves, however, could escape the institution by transforming themselves from their slave status into free members of their societies. In the “lineage mode of production” system, presented below, this could happen when a slave, on the initiative of his or her owner, was adopted and integrated into the local kin groups (Isaacman and Isaacman 1977). The possibility for individual slaves to exit the slave status o en depended on the availability of new “raw” slaves, which guaranteed the continuation of slavery. In the more commercially oriented societies involved in the caravan and plantation economies along the caravan routes and along the coast, the emancipation of slaves o en occurred when the slave had been able to accumulate wealth and could buy his or her own freedom. It has even been argued (Deutsch 2006: passim) that the emancipation of slaves in German East Africa was more the result of the slaves’ own struggle for

Pawned, Preyed Upon, Purchased, or Punished

•

151

their freedom through, among other ways, paying their owners for their own market value than the result of the colonial government’s acts and decrees to abolish slavery. Both forms of emancipation, gradual incorporation into the owner’s family, and self-emancipation through the payment of a compensation, concerned individual slaves. Neither the transformation from slave to “kinsperson,” nor self-emancipation threatened the system of slavery as long as new slaves were available when the older slaves managed to quit their slave status (Goody 1980: 41). In some of the African systems of slavery, including the systems in Bagamoyo, there were possibilities for a slave to initiate his or her own transformation from being a slave to become a “free” servant of the former owner. Still, a slave was a slave even in the African systems. There has, through history, been a cunning trick to present the relations between master and slave in domestic and kinship terms that makes it possible to present the inequalities in their positions as given by nature, as the inequalities between husband and wife and a father and his children. These lines of thought made it possible during the colonization of East Africa to present some African form of slavery as more digestible than the US form of plantation slavery. The relatively humane treatment by some of the owners of their slaves must never blind us to the inhumanity of slavery as a system (Finley 1980: 122, quoted in Lovejoy 1981: 12–13). The owner had the right to sell, with some restrictions as will be explained below, mistreat, manhandle, and even kill a slave. It should be kept in mind that the terms “domestic slaves” or “domestic slavery” have been used to represent at least two different phenomenon. The Sena of Mozambique (see below) may represent one form of slavery where slaves could be incorporated rather rapidly into the owner’s family and society and increase the labor force of that household. Another is the frequent use of kinship terms to articulate, as mentioned above, authority and submission, independence and belonging or, as in the case of the forms of joking relationship mentioned in the previous chapter that developed between up-country traders/porters and Bagamoyo, a form of friendship and mutual trust.

Slavery in East African Socioeconomic Systems Kinship systems o en provided the idiom through which belonging was perceived and expressed.8 Many such systems had the capacity to absorb new members or to alienate members, sometimes through what Jan-Georg

152

•

Muted Memories

Deutsch (2006: 17) calls “contractual transfer of people between different descent groups.” The relatives in many African kinship systems formed a corporate group with shared interests, o en under the authority of a male head, who held rights over the members but also the responsibility to care for and protect them. Hence, to belong in such a group, and to belong to members of such a group, provided security for most of the members.

Rights-in-Persons and the Kinship-Slavery Continuum The term “rights-in-persons” was coined by Miers and Kopytoff in the introduction to their volume Slavery in Africa: Historical and Anthropological Perspectives (1979).9 The term captures one of the most central human proprieties studied by anthropologists: our capacity to establish, maintain, and terminate relations of belonging. In many preindustrial societies without social security systems, the idea and discovery of not belonging in various social groups, or to other people who have an interest in your survival and well-being, must have been a terrifying experience. Peter Rigby, who carried out anthropological fieldwork among the Gogo people in central Tanzania, writes that: homesteads were probably much larger before colonial penetration than they are today. This was not necessarily correlated only with the possibly greater cohesion of the kin groups then, but also because men with li le or no property who may today set themselves up as homestead heads would not have been able to do so in the past. They would not have been able to marry wives as easily as they can today, nor would they have felt as able to survive physical and supernatural a acks upon themselves and their herds. (Rigby 1969: 185)

But the opposite was also true. Poverty and famine were powerful forces, which led to the alienation of vulnerable members of the group, like widows, orphans, the poor, concubines, and others, “and isolated people without connection became very vulnerable to enslavement” (Médard and Doyle 2007: 15). Members in the kinship groups constituted assets in terms of a labor force, exchanged items in marriage and blood money transactions, and were pawns when support or loans were sought in times of distress. Hence, kinship ties were crucial evidence for claims to belong in such a group, in contrast to the status of belonging to another person, or, in the words of Henri Médard and Shane Doyle: “The absence of a kin group puts slaves at the margin of society. This is why kinship is the most distinctive feature in the definition of slavery in Africa” (2007: 27). As the slave trade expanded with the booming caravan trade, the number of people who met the fate of being kidnapped and enslaved or taken as

Pawned, Preyed Upon, Purchased, or Punished

•

153

the spoils in armed conflicts between various groups increased. People enslaved through brutal violence were o en subject to forced migration when they were marched away from their home areas. When taken from their own societies and kin groups, the enslaved became socially dead. Cut off from their relatives, separated from their ancestral land containing the graves of their ancestors, they were denied and deprived of most of the qualities that characterize a human being. Many, not all, were relegated to the margins of their new society (Médard and Doyle 2007: 27) or even denied, initially, the right to belong in that society. We saw in the previous chapter that the transportation of goods along the central caravan routes depended on human legs, feet, heads, and shoulders. Most other productive activities, agriculture in particular, depended on human arms and hands, as the prime energy source. The control over other peoples’ bodies became a means to achieve control over economic, political, social, and ritual life, and as a source for the accumulation of wealth, prestige, and power. Some Marxist-inspired studies of African societies during the early 1970s identified a “lineage mode of production” system (Resch 1992) based on the kinship lineage structure. Or, in the terms used by David Graeber (2011: 130–33), these were “human economies” in contrast to a “capitalistic economy” like the American plantation slave system. These societies were o en based on extensive forms of agriculture with an abundance of land but limited access to human power to till the land. Under these circumstances the acquisition of additional human beings beyond the kinship and marriage systems became crucial.10 Slaves often became the solution to this problem.11 The use of slaves in the local agricultural regimes explains why women were in higher demand than men in many African societies, since many African agricultural systems primarily depended on female labor, which contrasted with the higher demand for men in the US gang system (Goody 1980: 37–38). An illustrative example was the agricultural regime among the Gogo in central Tanzania, strategically positioned along the central caravan routes halfway between Lake Tanganyika and the Indian Ocean. It has been suggested that the relative political stability in this area, in contrast to the political turmoil along the northern routes through the Maasai country and the southern route close to the Southern Highlands and the Hehe area, favored the caravan trade along the central routes. One reason for this stability was the need for the caravans to obtain food in the area, and one reason for the production of surplus food was the employment of slaves to carry out the agricultural activities (Sissons 1984: 88). All caravans on the central routes depended on the purchase of food from the inhabitants in this area. Carol J. Sissons (1984: 88) suggests

154

•

Muted Memories

that the important service role of the Gogo may have spared them from the devastating effects of slave hunting and it may therefore be assumed that their position in the caravan trade paralleled the role of the Nyamwezi. The Gogo provided food while the Nyamwezi provided the bulk of the porters and the ivory for the caravans on the central routes. Slaves in the lineage mode of production system were, as mentioned, categorized through the idiom of kinship. When newly acquired they were classified as kinless, outside the existing kinship systems, and were in that sense deprived of the human qualities of establishing social relations and belonging to a group (Lovejoy 1981: 11). They were simply a ached to their owner. However, in the existing classificatory schemes some of them were soon integrated under the kinship idiom, and may well have been referred to as the “children” of their owners or “fathers.” This transformation is well documented among, for example, the Sena of Mozambique (Isaacman and Isaacman 1977). The system may be exceptional in the rapid integration of slaves into the owner’s kinship group, and may therefore serve to illustrate the conditions and processes that characterized one form of slavery in the area. The acquisition of slaves in Sena society o en had purely utilitarian purposes since “the size of the surplus that a lineage segment could accumulate varied in direct proportion to the number of its economically productive members” (Isaacman and Isaacman 1977: 107). Many slaves in the lineage mode of production system shared many of the living conditions with their owners, worked side by side with them in the fields, and ate from the same pots. Upon arrival to the owner’s house, a Sena slave was integrated into the owner’s lineage by being given the owner’s clan name. This was usually the first step of the transformation of the slave from a kinless, socially dead person to a socially recognized member of society. In many societies, the end product of this process was a former slave who had acquired her or his dignity as a kin member, not by revolting against the master’s slave society, but by working to be integrated into it. Another strategy was to create relations through marriage, which, according to Barbara and Allen Isaacman, was “the principal mechanism for incorporating recently acquired adult akaporo [slaves] into the lineage segment” (2007: 111). Marriage between free Sena and acquired slaves became an institutionalized form of manumission, which also implied a crucial step in the integration of slaves into Sena society. The result was an increased number of hands that could be employed in cultivation under the control of the lineage or household head. The reasons for this kind of desire for and treatment of acquired aliens, has to be sought in the production systems and in the political organization of Sena society. In the absence of institutions like the police, courts,

Pawned, Preyed Upon, Purchased, or Punished

•

155

and prisons, which could capture and return runaway slaves, the most effective means to ensure that the slave stayed was for the owner to gradually establish social ties with the slave(s). For a slave, who might have been pawned or sold into slavery by her or his family, social ties developed with the owner was a means of being integrated in the new family and society. Hence, both master and slave had interests in forging social ties. For the slave this meant recognition of a society based on slavery rather than rebellion against it, or revolution to transform the entire system. A different form of slavery developed in the hierarchical coastal societies. The fundamental divide in these societies run between those who considered themselves as civilized, clean, respectable and with honor and those they saw as barbarian. Slaves usually fell into the la er category. “Raw” slaves were, according to Mtoro ( J. W. T. Allen 1981: 171f.) not allowed to sit on stools and could not share food from the same plates or bowls with their masters, which contrasted with the situation for the slaves in the domestic mode of production systems. Evidence of their exemption from civilization and cleanliness were that they were forbidden to wear shoes or cover their heads and that the owner presided over the marriage of two slaves. Such a marriage was not blessed by a religious leader. The humiliation of slaves continued a er death. They were not buried among the free-born and did not have Islamic ceremonies.

The Acquisition and Making of Slaves The variations in slavery and economic systems in nineteenth-century East and Central African societies also influenced the dominant modes of slave-making and acquisition. In short, four main methods in slave-making may be identified: by force, such as in wars and kidnapping and enslaving; by natural disaster, which forced the poor to either pawn a family member or barter their own freedom for food and other forms of assistance; by judicial process, which could force a convicted culprit and sometimes also the whole family into slavery; and by being born to slave parents.12 The last category, slaves born into captivity, was one in which the slave was occasionally treated as one of the family since the service and loyalty of such a slave was crucial for the master’s comfort. The principal processes and logic of the first three forms will be further explored below.13

Brute Force The first method, to use brute force to enslave other human beings, was used all over the area, but was probably not the most common form

156

•

Muted Memories

(Wright 1993: 7). The infamous slave trader/hunter Tippu Tip explains in his autobiography how he acquired porters in the vicinity of what today is Dar es Salaam, for one of the caravans he set up on the coast. On the day of departure Tippu Tip discovered that the porters were sca ered throughout the many villages in the vicinity of Bagamoyo. He therefore visited the Indian trader Hila and: “[I] asked him for strips of metal, and he gave me as much as I wanted. These I took back to Mkamba and the cra smen who were with me made chains. I put the whole lot of porters in these chains” (Tippu Tip [1974] 1958/59: 19–21). The political turmoil that followed the booming caravan trade sometimes led to social disintegration, which made life insecure for the weak and defenseless. The “brutalization” (for the term, see Iliffe 1979: 50) of life along the caravan routes was a result of an increased demand for slaves, the spread of firearms, and colonization. The outright hunting for slaves became particularly common in the societies south and west of Lake Tanganyika, areas that probably suffered more heavily from the slave hunting and trading expeditions than other areas. The brutalization and militarization of these areas was caused by invading traders from the coast and Ngoni warriors from southern Africa. The situation became chaotic when local leaders adopted the Ngoni military strategy in their hunt for slaves to sell to the coastal traders (Wright 1993: 6). In these kinds of raids, women and children were usually captured while the men were killed (Deutsch 2007: 88). A main force behind the sociopolitical havoc that developed in many up-country societies with the caravan trade was the widespread appearance of firearms. It has been reported that traders from the coast supplied local chiefs with firearms to facilitate their capture of slaves from neighboring societies. Slave raids among neighbors probably produced more slaves during the peak of the caravan trade than raids carried out by Arab and Swahili traders from the coast (Biginagwa 2012). The political and economic turmoil which followed in the wake of the rapidly developing caravan trade, in combination with the logics of slave hunting and trade, led to widening circles of violence in many African societies. As noticed by Francis P. Nolan, “The introduction of firearms and the appreciating value of prisoners of war in the slave trade, seem to have changed the nature of war, people replacing ca le as the prime object of plunder” (quoted in Deutsch 2006: 26n84). There are many accounts from the end of the nineteenth century on how people simply disappeared on their way to or from the fields or wells or in the forests while collecting firewood. This seems to have been a common experience in societies along the caravan routes, which explains why some villages became heavily fortified while others were moved away from the routes (Deutsch 2006: 37).

Pawned, Preyed Upon, Purchased, or Punished

•

157

This brutalization not only affected upcountry societies involved in the caravan trade but also life in the coastal communities. The struggle for control over the lucrative caravan trade, and the development of a plantation economy in some places along the coast, occasionally resulted in violent disruptions. The so-called Bushiri rebellion in 1888/89 (see the next chapter) led to considerable disturbances in and around the Bagamoyo area when he and his fighters a acked the surrounding villages in the area and forced people into slavery (Glassman 1995: 251–52). In the turmoil many slaves ran away and were captured by Bushiri and his men. For the first time, a slave market was temporarily established by him in Bagamoyo.

Pawnship and Natural Disasters Another slave-making force was the o en weak production base of many extensive forms of East African cultivation systems. The production system together with vulnerable methods for food preservation and storage made many households and villages sensitive to climatic, environmental, and sociopolitical hazards. The subsistence economy o en balanced on a thin thread between self-sufficiency and disaster. In cases of crop failure and famine, the kinship networks o en became stretched to the extent that further help was no longer available from relatives to those in need. In these situations two solutions were available: to pawn a family member in exchange for food or other forms of assistance or to barter the individual freedom for food and protection by becoming another person’s slave, as explained by Mtoro (J. W. T. Allen 1981: 169) “in time of famine people would sell themselves to each other. Or a person may pledge a child or a brother-in-law, and when he has not the money to redeem him, he becomes a slave.”14 Furthermore, the cholera and smallpox epidemics, which followed in the wake of the caravan trade, resulted in such depopulation of many local communities that their food production must have been threatened. This was surely another factor behind the decision of many family heads to pawn kin members or many farmers’ decisions to sell themselves as slaves in order to survive. Pawns given away as security for a loan were usually young women (Deutsch 2006: 55). The institution built on the rationale that the pawned would work in the creditor’s household, or on the fields, until the loan was repaid. The debtor had the right to repay the loan and thereby reclaim the pawned relative. The creditor, on the other hand, had the right to enslave or even sell the pawn, if the debtor failed to honor the debt. Pawning, Deutsch suggests (2006: 56), “was o en a means to disguise the outright sale of a person by his or her family, since it was regarded as an

158

•

Muted Memories

exceedingly dishonorable act.” The sensitivity in this act is illustrated by Jean-Pierre Chrétien (2007: 228) in the statement that a child given away during famine should be seen as a “gi ” to its new family, a positive measure since giving is seen as an act which gives merit. A well-documented case is the fate of Swema (see Alpers 1983). When Swema was around ten years old her mother pawned her to a creditor to obtain a loan, but she was unable to repay the debt. The creditor then sold Swema to an Arab to recoup his loss. Debts could cause enslavement in many up-country societies where trade took the form of barter and loans the form of material objects like food. The pe y traders in these localized trading networks “had a role in the overall penetration of commercial values and the exploitation of indebtedness” (Wright 1993: 8). They became the interface mentioned above between the established barter economies and the intruding market economy when coastal traders came to comb up-country societies in search of slaves. When the interface was established between the local slave systems and the intruding traders who sought slaves for the plantations along the Indian Ocean and on its islands, the slaves in the local forms of slavery were the first to suffer. Pawns were not slaves, but were easily enslaved. In many East African societies a household head or a parent could, as already mentioned, offer a dependent member as surety for a loan. Pawns were not usually cut off from their families as slaves were, unless the debtor failed to repay the loan. When the debt was repaid the pawn was reclaimed and returned. If not repaid the pawn could, as in the case with Swema, be enslaved and sold. “Only with repeated transfers,” writes Marcia Wright, “and the passing of currency-like trade goods, and by coming into the possession of traders, did people finally become commodities” (1993: 7). In the coastal societies, with the stiffening competition over prestige and positions, it became common for Indian traders and financiers to award loans to men who lacked their own financial means for their weddings. As surety these men o en pawned themselves. Many were enslaved and sold when they failed to repay their loans (Deutsch 2006: 71). Many up-country porters also experienced that they could easily obtain loans from traders in the coastal towns which, when they failed to repay them, caused them to end up in slavery, o en on the plantations along the coast or on the Zanzibar islands. Another strategy to survive economic stress was to give up one’s own freedom and accept the status of being another person’s slave. In the lineage production system this meant to exchange the labor force of a family or kin member, actually the physical body, for food or protection. In con-

Pawned, Preyed Upon, Purchased, or Punished

•

159

trast to the pawn, who could remain in contact with the family of origin, a person who offered herself or himself as a slave became “socially dead” in relation to her or his family of origin, since in this case there was no debt that could be repaid in order to reclaim enslaved family members. It has, for example, been reported that “during the year 1884, a terrible famine occurred in East Africa, and occasioned a revival of the slave trade there, many of the poor natives having sold themselves for food” (from a Church Mission Society pamphlet from 1885, quoted in Woulfin 2011: 87–88) The increased demand for slaves may have changed the practice of pawnship. It has been reported, for example, that the Busaid dynasty on Zanzibar “refused to return Mijikenda daughters [a group on mainland Kenya] who had been pawned to them in exchange for grain, opting instead to sell them as slaves to Arabia” (Spear 1978: 218). The second half of the nineteenth century saw a rapid commodification and brutalization in the trade of human beings.

Punishment In the more monetized societies along the coast, personal debts from loans in the form of money or trade goods taken from financiers or traders were a more common reason for enslavement. In his study Debt: The First 5000 Years, David Graeber (2011) argues that the main function of money, when it developed as a means of transaction, was to measure and regulate debts. But the purpose to measure and regulate debts depended, in turn, on the existence of compulsory means to force the debtor to repay the loan. Hence, the development of debts, money, and the centralized execution of physical violence, o en in the form of police, courts, and prisons, went hand in hand in human history. This is also how the financial sector in Bagamoyo and other trade towns along the coast developed. This may be seen in the efforts from the sultan on Zanzibar to establish not only customs houses but also military garrisons, courts, and prisons, introduced legal systems, and appointed judges in places like Bagamoyo. These institutions secured the sultan’s own economic interests along the coast, but also the interests of the Indian and Arab traders he encouraged to se le in Bagamoyo. The establishment of the sultanate implied a new centralized political order manifested in the sultan’s right to put insolvent debtors into prison or enslave them. This function was a prerequisite for the development of a banking system, a service in the hands of Indian immigrants that lay behind the boom of the caravan trade.

160

•

Muted Memories

Mtoro, the Bagamoyo-born author, explains how debt and punishment functioned in the caravan trade. An ill-fated trader who had taken a loan from an Indian trader or financier but who had lost his merchandise through robbery or natural disasters still had to repay his loan. If he failed due to lack of resources, “he would be given a further advance to go on a second expedition. All the profit that he made would go to the debt; but he was not imprisoned or bankrupted” (J. W. T. Allen 1981: 207). But imprisonment waited if he failed a er the second advance. Sewa Haji Paroo was the leading businessman in Bagamoyo during the last decades of the nineteenth century. He had gained a particular reputation that any caravan leader who wanted to reach the interior of the continent or obtain porters had to pass through him. It may be assumed that a substantial part of these porters were his own slaves, whom he had obtained through loans he had given to caravan porters, to be used for their own trade in caravan tours to the interior of the continent. Indebtedness, writes Deutsch (2006: 24), “was probably chiefly responsible for the impoverishment of the Nyamwezi porters who were sold as slaves to work on the Pemba clove plantations in the 1870s.” Sewa Haji supplied European travelers and colonizers with guides and porters. “It is very likely,” Deutsch (2006: 71) writes, “that the la er were slaves.” Evidence is that those who hired porters from Sewa Haji only paid the porters their posho, or daily food allowance. Their owners received the pay for their work. The European/American travelers, traders, colonizers, and even missionaries, however, are for reasons we only can guess, silent on the status of their porters. When reports on their use of slaves as porters in their caravan reached the readership in Great Britain it provided fuel in the agitations from the abolitionists who accused the British government for being implicated in the slave trade in East Africa (Woulfin 2011: passim). In many less monetized economies in the interior of the continent, mischief other than the inability to repay loans could also lead to enslavement. In some up-country societies, in the absence of institutions like garrisons and prisons, justice was administered through other means. Common punishments for criminal acts like adultery, homicide, bodily harm, or “witchcra ” were expulsion, execution, or enslavement and being sold to foreigners or, alternatively, young women were o en given as compensation to the offended person or group for offenses commi ed by male relatives.15 Ostracism, in turn, o en resulted in either execution or enslavement by foreigners. It has been suggested (Unomah and Webster 1976: 311) that the expanding caravan trade and increasing demands for slaves lead some up-country rulers to pervert the traditional judicial systems to be able to punish more people in their courts into slavery.

Pawned, Preyed Upon, Purchased, or Punished

•

161

Slaves in the Caravan Trade In his thorough study of German and other archives, Jan-Georg Deutsch (2006) identifies two major areas in what became German East Africa with high proportions of slaves. One was the coastal societies, increasingly involved in the caravan trade and plantation economy. The second was the Nyamwezi area south of Lake Victoria and east of Lake Tanganyika. Statistics from early German colonial reports on the number of slaves held in various parts of the colony gave 233,000 slaves in the Nyamwezi area, compared to 1,500 in the Iringa District, an area of comparable size (Deutsch 2007: 76). The same source reported 125,000 to 200,000 slaves along the coast, of whom 125,000 to 175,000 were held in the two southernmost districts Kilwa and Lindi, while 2,000 were held in Bagamoyo. The number of slaves in Bagamoyo probably referred to the number of slaves residing in the town, which did not necessarily include the number of slaves who temporarily appeared in Bagamoyo either as porters or merchandise. Early accounts from the 1880s and 1890s show that the majority of the people in the Nyamwezi area were slaves (Deutsch 2007: 76). The slaves held here were, as has already been mentioned, primarily used in their cultivation systems, presumably to release Nyamwezi men, and women, to be able to take part in the caravans, but also to produce the agricultural surplus needed to supply the many caravans that either originated in the area or passed through it. It may even be hypothesized that the existence of slaves in the Nyamwezi society was a determining factor for the emergence both of the caravan trade in the area and for the development of the Nyamwezi as the dominating caravanning group during the early phase of the caravan trade. Slaves also appeared as a medium of exchange in the Nyamwezi caravan trade. Slaves, usually obtained from societies to the west of the Nyamwezi, were brought with the caravans to be exchanged for elephant tusks, livestock, or trade goods that were brought to Bagamoyo and other ports on the coast. This is captured in a standard work on African history in the following words: Nyamwezi traders, even porters, travelled to Ujiji and to Buganda to buy slaves. Some of them co-operated with the coastal traders to raid the fragmented Manyema communities in eastern Congo. It would appear however that the greater proportion of the slaves purchased by the Nyamwezi in the Congo or from the Ganda were not exported outside East Africa. Some were kept by the Nyamwezi themselves as domestic servants, others were given in exchange for ivory and foodstuff in Ugogo [the part in central Tanzania through which the central caravan routes passed]. (Unomah and Webster 1976: 299)

162

•

Muted Memories

A specific feature in the Nyamwezi slave system was, as already mentioned, the high proportion of female slaves. One reason for this was that many of the tasks in the local agricultural system were carried out by women. Men usually took part at the beginning and end of the cultivation cycle. Its rhythm was given by the seasonal changes between the dry and the rainy seasons. Soil preparations began with the onset of the rainy season, while the main harvest was reaped in the early part of the dry season. A er having prepared the soil, many Nyamwezi men le their homes to go with the caravans to the coast while women, enslaved or free, took over in the fields, particularly the task of weeding during the periods of plant germination, growth, and maturation. Another reason for the popularity of female slaves in Nyamwezi society was that they were considered more “docile.” Yet another reason could have been that women and girls “were more universally assimilable than were boys” (Wright 1993: 2). It was more difficult for women than for men to run away and abandon their owners since they faced a greater risk of being slave-napped again. It may also be assumed that the spread of new diseases along the caravan routes (Kjekshus 1977: 23–24) led to depopulation, which in turn increased the demand for slave labor in the Nyamwezi cultivation systems and in particular female slaves who could perform both productive and reproductive functions. Their capacity to give birth to new members and thereby contribute to the reproduction and increase of the owner’s lineage and its labor force was appreciated in the domestic or lineage production systems. The slave systems on the coast and in the Nyamwezi area may be seen as communicative vessels. The slave traders had to look for new outlets for their slaves when the British-Zanzibar anti-slavery treaty was signed in 1873. The southern part of the Nyamwezi area, with its intensive cultivation, became one of the major areas for the sale of slaves. A contributing factor to this was the introduction of new crops like maize and rice. In particular, paddy rice made it possible to turn the mbuga soils (black co on soil), temporarily flooded during the rainy season, into productive lands. This development increased the demand for more human hands in the cultivation systems, which in turn led to employment of an increasing number of slaves in Nyamwezi agriculture.

Trading Slaves The fact that slaves in the Swahili world could obtain credit, take part in the caravans, and operate as traders reveals a specific feature of the East

Pawned, Preyed Upon, Purchased, or Punished

•

163

African slave systems: the tight social relations of trust and dependence that could develop between master and slave (see Miers 2006: 5). These tight relations helped the owner tie the slave to himself and his family in order to avoid escape, and made it possible for the slave, in the logic of the “moral economy,” to ask for favors and improvements from the owner or master. Slaves in the coastal societies could join a caravan both with and without the owner’s consent. They not only functioned as porters but could also operate as traders. Already Richard Burton, in his publication The Lake Regions of Central Africa ([1860] 1869) noticed that a slave who had been entrusted by his master with cloth and beads for trade could undergo a social transformation. When a trading slave returned to a place like Bagamoyo, the slave had to hand over a profit to the owner. The owner, however, was o en responsible for loans taken by the slave and used for his own business during the caravan route. The slave was not only the owner’s property but also a “junior” not fully responsible for his or her own acts.16 Slaves could obtain loans for caravan trade from the Indian merchants in the coastal towns, even against the will of their owner. It seems to have been common that slaves simply disappeared with a caravan, only to reappear some months later and then either offer a share of the profit to the master or offer wealth that represented their own market value in order to be freed. This was possible since East African slave systems accepted that slaves worked in a large number of professions and also that they could hold property, including other slaves. A common profession was to work as a caravan porter/trader. The expansion of the caravan trade and with it the number of merchants, traders, loans, and debts probably led to new forms to negotiating, reclaiming debts and obtaining security for loans. The intrusion of European colonial powers also influenced the economic system, as noticed by Glassman: “Indian merchants, who, as British subjects, were forbidden to hold slaves, apparently welcomed such opportunities [to extend loans to trading slaves] to bind other men’s slaves to themselves through debt” (Glassman 1995: 75). Why did trading slaves return to the coast and their masters instead of running away when they were in the interior of the continent? One simple reason was that a runaway slave, without family and kin, was easily captured and enslaved by a new master. Another was their aspiration to be assimilated into the coastal culture and society. The owner’s own involvement in the caravan trade may have contributed to a confidence that the slaves would fulfill the long journey and then return to the coast both with the owner’s profit, and possibly also their own. The fact that many of them

164

•

Muted Memories

were allowed to act as porters/traders in the caravan trade, the backbone activity for the coastal towns, must have strengthened the slave’s feeling of belonging in that culture and society and, by extension, to the owner. To be enslaved meant to be deprived of one’s honor, which was publically displayed by the free members of the society when they scorned the slaves (Schoenbrun 2007: 47–49) for their lack of knowledge about local traditions, and how to behave and speak (Deutsch 2007: 94). The only way to regain honor was to be recognized as an honorable person in the owner’s society. Hence, to assimilate to that society, and accept its cultural rules and codes, was the most secure way to move from the margin of that society to become a member of it.

Trade in Slaves The main slave markets along the East African coast were first Kilwa and then Zanzibar. Kilwa, the first port for long-distance caravan trade (Deutsch 2006: 33), was for many years the most important slave port along the East African coast. Many of the slaves taken from the area at Lake Tanganyika and Lake Nyassa (Malawi) were shipped to Madagascar and the French islands of Reunion and Mauritius to work on the plantations there. The French islands were the main destination for slaves until the Napoleonic wars, when the British navy emerged as a global force and began patrolling the waters between the African coast, Zanzibar, and the French islands in search of slave vessels. Zanzibar became, despite British efforts to abolish the slave trade through agreements with the sultan, an outlet for slaves from Kilwa. Actually, Stone Town, the capital of Zanzibar with the sultan’s court, was built from the wealth created by slaves, as captured by Laura Fair: “Slaves, and the profit derived from both their sale and their labor, were at the very heart of Zanzibar’s transformation from a small and relatively unimportant Swahili town to the center of a vast trading empire stretching from central Africa to halfway around the world” (Fair 2001: 12). Furthermore, the stones used to build the stone houses, coral rag, “used universally for building [were] quarried beyond the town, and carried in by slaves—a tedious process—and lime made from the same material, and burnt in conical piles at the yards around Shangani Point” (Elton 1879: 51). The houses were built with slave labor and the town was built with the wealth created through the slave trade and the use of slaves on the plantations. The establishment of a plantation economy in some places along the coast, both to feed the caravan porters and for food export to the Ara-

Pawned, Preyed Upon, Purchased, or Punished

•

165

bic peninsula dramatically increased the demand for slaves. Visitors to Zanzibar reported that approximately 40,000 to 45,000 slaves were sold in Zanzibar in 1839 (Alpers [1967] 1974: 11). At the end of the nineteenth century up to two-thirds of Zanzibar Stone Town’s population could have been slaves, and up to a quarter of a million were slaves on the two islands Unguja and Pemba (Kollman 2005: 42–43). The trade in slaves continued to increase with the booming clove plantation economy on the islands until slave trade was banned in 1873. Slavery as an institution was, however, banned on Zanzibar and Pemba only in 1897, in Kenya in 1907 (Alpers [1967] 1974), and in German East Africa, Tanganyika mainland, in 1922 a er the British had taken over the territory (Deutsch 2006: 2). Zanzibar traded in slaves from the area between Mogadishu in the north to Sofala in the south, but primarily from Kilwa. One of the main destinations was, as mentioned, the Mascarene Islands (Mauritius, Reunion, and Rodrigues), which received European immigrants and African slaves in the seventeenth and eighteenth centuries (Boswell 2008: 24). The Kilwa-Zanzibar link was undermined first when the Indian traders in Kilwa were forbidden, in their capacity as British subjects, to take part in the slave trade. The picture painted by the abolitionists and missionaries of Arab slave hunters and traders also hints at the Arabic Peninsula and Persian Gulf as the main receiving areas of African slaves. As explained by Captain James Frederic Elton (1879: 4): On the East Coast of Africa, . . . a form of slave-trade went on, as from time immemorial, supplying the needs of the Mohammedan nations in the East— of Persia, Arabia, and Turkey, and notably of Zanzibar, . . . [And] In these countries slavery is an institution as deep-rooted as the religion of the Prophet, which makes domestic slavery essential to the peculiar status of women, and gives every encouragement to agricultural slavery among freeborn Mohammedan men.

There are, however, disagreements among researchers regarding Zanzibar’s role in also providing slaves to the Asian continent. Sheriff (2005), for example, argues that relatively few slaves were shipped from Zanzibar to the societies in the northern part of the Indian Ocean, compared to the number of slaves sold to the French colonies, and the number of slaves kept locally on Zanzibar (or Unguja) and Pemba when the clove plantation economy exploded from the early nineteenth century in what has been called a “clove mania,” and with it also the demand for slaves (Sheriff 1987: 51–53). Cooper (1975: 100) writes about the tremendous expansion in the 1840s, which leveled out a er 1850.

166

•

Muted Memories

Slaves and Slavery in Bagamoyo Bagamoyo’s vibrant economy, the result of the booming caravan trade, created a demand for slave labor in many different trades while at the same time opening up possibilities for some slaves to learn new skills and to seek to advance in the hierarchy of tasks, duties, and positions. Much of the sociocultural and political dynamics in the coastal societies were generated by internal identity-making processes among the slaves in combination with the way the free members of the society characterized and treated their slaves. A dominant characteristic of these societies was the strong tendency towards hierarchization and processes of inclusion and exclusion, which affected not only the free members of the port towns but also the life and agency of the slaves. An example of the categorization strategies and exclusion mechanisms employed by the free-born members of the society against the slaves was the naming system. The nonhuman character of the recently acquired or “raw” slaves was publically revealed denying them permission to have proper Muslim names. Slaves could not carry names from the Holy Quran or other Islamic texts. Their names were o en reduced to refer to the days they were born or the functions they performed for their masters, for instance, Jumamosi (Monday) or Fedha (a slave’s capacity to bring money or wealth) (Suzuki 2012: 230–34). Even the dress codes signaled the difference between free and slave. Slaves were denied the right to carry the emblems of civilization, which meant that they were denied garments that covered the entire body. Instead, a male slave would wear a loincloth tied around the waist and a slave woman would be dressed in a black cloth tied around her body but she would be unveiled. This ideal was, however, not always strictly followed since many slave owners provided a leeway for their slaves to articulate an increasing acculturation into the coastal culture and society. This seems to have been particularly prominent among female slaves, who were both decorated to perform on public occasions, and who could carry colorful textiles tightly tied around their bodies.

Categories of Slaves Different classification systems were applied to the slaves in the coastal societies. One was based on the slave’s relation to his or her master, another on the slave’s main occupation, for example domestic or trade duties in the town, or agricultural work on the surrounding fields. Still another was based on the slave’s religious affiliation, place of origin, and degree

Pawned, Preyed Upon, Purchased, or Punished

•

167

of assimilation into coastal culture and life. These intersectional criteria opened windows for identity-making, and possibly also avenues for emancipation. The generic Kiswahili term for slave, mtumwa, covers many different forms of dependency on the owner (Deutsch 2006: 65). The word is composed of m-, a prefix showing that the word refers to a human or animal, -tum-, the root of the verb to send, -w-, and an infix showing that it is the passive form of the verb (Eastman 1988: 16).17 As there are different reasons for a person to be sent, there are also different possible interpretations of the word. One common practice in preindustrial societies was to send a kinsperson as hostage or pawn to a partner or ruler to forge an alliance. This has been a common practice for traders and societies involved in the trading networks in the western Indian Ocean, i.e. the strategic placement of kin in the houses of allies, rulers, and actual or potential trade partners. Another reason to keep slaves who could be sent may be sought in the division in Swahili society of an open male sphere and a secluded female domain. Freeborn wives and daughters were ideally secluded in the stone houses and in their court yards. Slaves, as “nonpersons,” were used to carry out the tasks that potentially brought shame (aibu) on free women and men, or threatened their cleanness (usafi) and respect (heshima). No harm could be caused to slaves who were sent out to operate in the public sphere, since they had already been deprived of respect. Many different forms of slave statuses developed in and around the town with the expansion and diversification of Bagamoyo’s economy. A major difference, as in so many dimensions of coastal life, was made between slaves born on the coast and the “raw” slaves brought to the coast as trade objects or porters through the caravan trade. A slave born on the coast, a mzalia should, ideally, not be sold by her or his owner (Glassman 1995: 89). One reason was that the wazalia were born by slaves who resided in a free household in a port town like Bagamoyo. They had been raised in the owners’ houses and were expected to have developed feelings of loyalty, which could be exchanged for a degree of independence and the right to engage in profit-making activities, to own property, and also to invest in their own slaves (Fair 1994: 31–32). A slave from the interior, a “barbarian” in all aspects of the word, had a much more insecure life situation. A “raw” slave, in Kiswahili called mjinga, was “a recently imported cha el, a slave from the wild interior, a man incapable of talking Kiswahili” (Elton 1879: 59–60). A “raw” slave (for the term, see also J. W. T. Allen 1981: 170–71) who reached Bagamoyo through the caravan trade could either be retained in or around the town or be shipped to the slave markets on Zanzibar. If kept on the coast, the slave was most likely put to work on the owner’s, or

168

•

Muted Memories

another person’s fields, mashamba (plural, shamba in singular in Kiswahili). The demand for agricultural products increased with the booming caravan trade and influx of porters and traders to Bagamoyo. Female slaves constituted a high proportion, possibly the majority (see Miers 2008: 175), of the agricultural slaves since women, as mentioned, could perform most of the tasks in the local cultivation regimes. Many town dwellers controlled fields in the vicinity of Bagamoyo, which were cultivated by slaves. A common regime was that slaves worked for five days on their master’s shamba and two days, always including a Friday, the Muslim day of prayer, on a plot of land allocated to them for the production of their own subsistence. They did not receive a “salary” but were compensated in kind (Deutsch 2006: 70). They worked on the fields under the supervision of a more senior slave, who was entrusted this task by the owner. The “raw” slaves also lived in separate villages under the control of the more senior slaves. This arrangement provided them both with their own subsistence, and generated an agricultural surplus, which was taken by the owner. Slaves born on the coast, wazalia, could also be engaged in cultivation but sometimes under a different regime. They could be entrusted to organize their own cultivation on the condition that they provided a tribute to their owner. This tribute, called taja or ijara, should be handed over either each month or once a year (Glassman 1995: 86). Although still slaves, the wazalia could be seen as clients or, as Glassman expresses it (1995: 86) “bonded peasants.” Another category of slaves in Bagamoyo was called vibarua (sing. kibarua), who worked as day-laborers. They were rented out by their owners, or if they belonged to the category of wazalia slaves, could rent themselves out for a compensation, which, either fully or partly, went to the owner. Alternatively they found a paid job and handed over a fixed sum each month to their owner (Deutsch 2006: 71). It has been suggested that the British banned Indians from owning slaves, but allowed them to hire slaves from non-Indian slave owners (Mackenzie 1895: 15–16) may have increased the number of vibarua slaves. In contrast to the “raw” slaves, they rarely worked in the fields, but were involved in tasks either in the town or in the caravans. This facilitated their adoption of “civilized” town culture and their struggle to accumulate the wealth needed to buy their own freedom to get away from the slave status. The booming caravan trade created an increased demand for people who could supply different kinds of services and goods. Deutsch identifies the following tasks for slaves in the coastal towns: “pe y trade as hawkers or in unskilled menial jobs as day laborers, messengers, porters, stevedores, mangrove cu ers, builders, and—in the case of women—cleaners,

Pawned, Preyed Upon, Purchased, or Punished

•

169

water carriers and sex workers” (2006: 71). Other tasks included digging up and cleaning copal, which was in high demand in China, Japan, and the Americas, and working with the copra from the coconut plantations. Some also worked as a substitute for a kinsperson for their owners’ creditors if the owners had failed to honor their loans. The vibarua slaves, most o en men, who made up a significant part of the population in some coastal towns, were o en owned by less affluent owners or even other slaves. They provided an avenue for investment for people who did not own or control large fields where the slaves could be put to work. Instead, they were either rented out to other people who needed their service, or sent out to find their own paid jobs. The category of slaves, which occupied the highest position within the coastal slave community, was the artisan or trading slave (Glassman 1995: 87–88). These skilled slaves, Deutsch indicates (2006: 71–72), could have been day-laborers who had learned a particular skill and thereby had been able to climb the ladder. Some became caravan guides or leaders who received the other caravaneers’ respect, and who could exert their power even over free porters. This illustrates the flexibility and fluidity of the coastal systems of slavery and the means to establish hierarchies, power, and prestige. They were o en called fundi in Kiswahili, which indicates a profession as artisans. Typical occupations in the diversified economy boosted by the caravan trade were caravan specialists, weavers, silversmiths, and palm wine tappers. Other occupations, connected with the transport over the Indian Ocean, were captains, fishermen, sailors, boatmen, and boat builders. There were also those related to the “civilized” urban life in Bagamoyo, such as carvers, carpenters, metal-workers, bricklayers, lime burners, and stone masons (Deutsch 2006: 71–72). Female Slaves

Not all female slaves were engaged in cultivation. Some were domestic slaves who tended to household chores and, thus, relieved free women in the household from these duties. Many of these tasks meant leaving the private spaces in the houses to enter and work in the public areas of the town, which, ideally, was not possible for the free female household members. Another category of female slaves was the suria or concubine. They o en accompanied their masters on caravan tours into the interior of the continent, journeys that the trader’s freeborn wife could not undertake. On other occasions, they could be dressed up in expensive clothes and jewelry, as “decorated slaves,” and publically displayed. There are frequent accounts in travelogues wri en by mid-nineteenth century visitors to East Africa on the public display of female slaves, as a contrast to those

170

•

Muted Memories

toiling on the fields and plantations, to boost the owner’s prestige. James Frederic Elton (1879: 29), for example, writes that “slave women [could be seen] in bright cloths wrapped tightly below the armpits across the bosom and flowing to the knee,” which contrasted with the dress code of the freeborn women, who wore “the gilt masks of Muscat, which hide nearly all the face under a mass of filigree-work” (1879: 46). It is possible that a specific feature among the freeborn population in the coastal towns may have contributed to an increased demand for female slaves. In the early twentieth century, a Provincial Commissioner in the Kenyan colony reported that a large proportion of Swahili women were infertile (Eastman 1988: 13). As a child from a union between a free man and a slave woman who belonged to the father’s kin group, the infertility of the free-born wife/wives could have been compensated through the inclusion of slave concubines in the urban households, which would explain the popularity of keeping female slaves. For example, it has been reported that female slaves at the turn of the nineteenth century outnumbered the male slaves in Mombasa town.18 The high demand for female slaves may also explain that they were more expensive to acquire than male slaves.19 Female slaves were not only used to support the reproduction of the owners’ patrilineages, but also to perform various productive duties as cooks, nannies, and other domestic duties. They could also, in contrast to the owner’s wife/wives, leave the house to buy or sell items at the market, draw water from wells, and accompany the owner on his caravan journeys if he was a trader. Many traders traveled with a large number of concubines who, in various ways, cared for them. No wonder that the price for female slaves was higher in the coastal societies than for male slaves. The high proportion of female slaves, in combination with a strict division between a male and a female sphere, and the seclusion of the freeborn women led to a clear divide, according to Carol Eastman (1988), between a male waungwana world and a female wanawake (women) world. The urban, civilized, Muslim waungwana world was, strictly speaking, a man’s world. The secluded world in the houses of the patricians was filled with free-born women and children, domestic servants, o en slave women, and female concubines. Both servants and concubines probably came from up-country societies characterized by the household heads by ushenzi or “unculturedness.”20 Gender segregation, according to Eastman, resulted in spatial segregation, which in turn produced cultural and religious segregation in the society. The Kiswahili language may support the hypothesis that the high employment of female slaves, o en from up-country societies, resulted in dual culture in the Swahili world with, on the one hand, a free, civilized,

Pawned, Preyed Upon, Purchased, or Punished  • 

171

urban, male waungwana world, and, on the other, a subjugated “uncivilized,” rural-influenced female ushenzi world. Most researchers today agree on its Bantu origin and the considerable influence on it by the Arabic language. This influence, however, is relatively late, and restricted to domains such as trade, science, craftsmanship, and the maritime world. In contrast, the domains where slave labor dominated “are where we find little Arabic vocabulary, while the areas of life from which female slaves were excluded (such as the wage labor, religious, and political-legal domains), are heavily Arabized” (Eastman 1988: 13), and “childcare and bedroom vocabulary is relatively devoid of Arabism” (Eastman 1988: 14). The practice of marrying the local leader(s) daughter(s) and the high proportion of female slaves in the homes of the immigrating merchants from Arabia and the Gulf both contributed to the “Swahilization” of their families. “They bought slave-girls,” writes Burton (1859: 52), “built houses . . . and gave up the thought of returning to their home in the barren and burning North. Their children abandoned their father’s for their mother’s tongue.” Not only immigrants from the Arabic peninsula but also slaves brought from the interior of the continent were subject to acculturation processes. At least three such processes may be found behind the cultural transformation of “raw” slaves from the interior of the continent into coastal “civilized” members of the society. One was the hegemonic position of the “Swahili” way of life (see also Cooper 1981: 289–90); another was the slaves’ own strategies to become members of, rather than revolt against, the coastal societies; and a third was the slave owners’ willingness to convert material wealth into religious merit through the manumission of slaves. It is, however, more correct to understand the processes that took place in the coastal trade towns as processes of hybridization or creolization. The influences from up-country domestic slaves and servants on the development of a women’s world and the rapid “Swahilization” of immigrants from Arabia show that sociocultural processes are multidimensional and multifaceted and what is often perceived to be ethnic categories are rarely or never homogeneous and seldom have well-demarcated boundaries.

Resistance and Rebellion Despite the struggle for assimilation, resistance and rebellion against the slave status were also common in the coastal societies during the nineteenth century. Both “petite maroonage” and “grand maroonage” occurred frequently as slaves ran away as individuals and in groups (Alpers 2006).21 The former meant that the runaway slaves sooner or later returned to

172

•

Muted Memories

their masters. A punishment, o en corporal, was to be expected, but also new negotiation power to achieve be er treatment and position, since the owner now knew that the slave could run away again. “Grand maroonage” implied that the runaway slaves were determined to establish their own independent se lements. Such se lements existed along the coast and were targets for frequent a acks from the slave owners to recapture their slaves, who were o en met with a determined defense from the slaves who protected their achieved status not only as an equal among equals in the maroon se lements, but also aspired to be regarded as equals to their former masters. It would be a mistake to interpret this as the runaway slaves claiming or defending an abstract notion of freedom, as explained by Glassman (1995: 113): “Slavery and freedom—uhuru—were not clearly discrete categories in most peoples’ minds. Uhuru was not a status one was born with, but was rather a condition that was bestowed on the slave, by the owner or by some other powerful patron.” It has been noticed by Glassman (1995: passim) and Cooper (1981: 281) that the struggle of the slaves to improve their life situation was resistance against their slave status, rather than a revolution to change the entire system. Paul Lovejoy notes (1981: 30) that few argued against the slave system or even the trade in human beings in African societies. Thaddeus Sunseri, in contrast, finds that the establishment of maroon societies along the coast indicates that the runaway slaves not only “rejected captivity among Arab and Swahili overlords,” but also “rejected coastal culture” and “ex-slaves did not seek inclusion within Swahili society because they would not be accepted as anything but permanent, inferior underclass” (Sunseri 1993: 483–84) The discrepancies between his22 and Glassman’s views, Sunseri suggests, comes from his own focus on the ransoming of slaves on Mafia island who, in his words, were “among the least acculturated of East African slaves” (1993: 499) while Glassman mainly has his information from wazalia slaves or slaves born in the Swahili societies on mainland Africa. The plantation economy on Mafia Island developed relatively late and many slaves taken to the island had been brought up in up-country cultures and societies to which they still had an a achment when they came to Mafia while the wazalia slaves, as the name indicates, had been born and brought up on the coast. Another reason for the differences in their respective agency and identification is that some of the slaves in, for example, Bagamoyo had an opportunity to take part in the caravan trade or commercial cultivation and accumulate the resources needed to be ransomed and take part in the culture of conspicuous consumption, an option not open for plantation slaves on Mafia.

Pawned, Preyed Upon, Purchased, or Punished

•

173

The absence of class-consciousness among the slaves in the coastal societies is discussed by Glassman (1995). The struggle for improvements took the form of individual a empts to, first advance in the hierarchy of slave categories and, second, to pass from a “barbarian” to a “civilized” member of coastal society. The struggle to establish maroon se lements may be seen as a way to establish conditions that allowed the slave to live in a way that approached his master’s society and culture. Slaves in these se lements therefore both owned slaves and took part in the slave trade in a way that their former owners probably had. The hegemonic position of Muslim “Swahili” culture was underpinned by the strategy of slaves in a town like Bagamoyo to pass into the coastal culture and society. In his study on conflicts and consciousness in coastal societies during the end of the nineteenth century, Glassman (1995) repeatedly points out that the strategy of the slaves and poor segments was not to revolt and try to change the social order, but instead to try to acculturate into it. This strategy, also pointed out by Gwyn Campbell (2006b: xxf.), entailed the slave trying to establish relations with the slave owner rather than with other slaves. One reason was that only the slave owner could free a slave in many East African slave systems.

Manumission and the “Freeing” of Slaves A slave owner could free a slave for a number of reasons. The booming caravan trade made it increasingly common for slaves to accumulate the wealth needed to buy their freedom at the price of their own market value (Deutsch 2006: 73). Deutsch finds that this development was one of the main reasons for the decline of slavery during the German colonial period, in contrast to the colonial government’s failure to abolish slavery and slave trade (2006: 208). It may seem ironic that this procedure, to move from the status of slave to the status of a free person probably depended on a steady supply of “raw” slaves who could replace the manumi ed slaves. Slaves also provided a field of religious and social merit for their Muslim owners, since the freeing of a slave was a prestigious act, both in the eyes of Allah and other members of society. Cooper (1975: 244–45) states that “whosoever frees a slave who is a believer, God has freed such a one from every calamity—the calamity of hellfire—to happiness and comfort.” A slave owner who had sinned during his life could liberate himself from the sin through the manumission of a slave (Deutsch 2006: 73). The freeing of a slave who had converted to Islam was also based on the principle that a good Muslim should not hold other Muslims as slaves.

174

•

Muted Memories

Another reason can be sought in the social embedding of the slave institution in wider society. In the absence of institutions like the police, courts, and prison, one of the few ways a slave owner had to minimize the risk that his or her slave ran away was to develop social relations with the slave.23 This was particularly prominent, as explained above, in the “lineage production” type of society, while it was less common in more politically centralized and socially stratified societies. Gradually, these relations could develop to a point where the slave was freed and adopted as a member into the owner’s family or kinship group (see also Deutsch 2006: 76–77). It should be noted that a freed slave o en continued to live in a subordinate position under the former owner in a patron-client arrangement. Even the children born to a freed slave were, juridically, subjugated to the former owner. In some systems of slavery only the third generation, the grandchildren of a manumi ed female slave, was regarded as free (Deutsch 2006: 60; Nimtz 1980: 52). This is still an illustration of the gradual positions between being free or being a slave that existed in the East African societies involved in the caravan trade. A slave who was freed in Bagamoyo had come a long way in the process of acculturation into the coastal society. Regardless of place of origin, he or she now, most likely, adopted the ethnonym Mswahili, which indicated belonging in the coastal society, in contrast to the newcomers from up-country societies who arrived as traders, porters, or “raw” slaves. The term Mwungwana, probably originally used exclusively for the male part of the free-born population, carried similar connotations: a claim to belong to the “civilized” world, which meant the coastal way of life, in contrast to the up-country, “barbarian” people.

Notes  1. The coastal area was influenced by the Shari’s law that gave the slaves the right to hold property, including slaves, and to marry (Cooper 1981: 284).  2. Ghada Hashem Talhami found, in his study of the revolt, that slaves from East Africa only formed a minority of the rebels. Most of the African slaves did not come from Zanj, the East African coast, but from Sudan, the Libyan Desert, Ethiopia, and Maghrabi (1977: 461).  3. Sheriff (1987: 226ff.) estimates that approximately 25,000 slaves per year arrived to Zanzibar during the 1860s and 1870s.  4. A French slaver, Jean-Vincent Morice, who had bought 700 slaves in 1775 on Zanzibar and another 925 slaves there in 1776 managed the same year to “conclude a one hundred year treaty with the Sultan of Kilwa, Hasan bin Ibrahim, which granted the Frenchman primary rights in the European slave trade in

Pawned, Preyed Upon, Purchased, or Punished

 5.

 6.

 7.  8.

 9.

10. 11.

12.

13.

•

175

Kilwa. Morice contracted to purchase 1,000 slaves annually” (Alpers 1970: 103.) The agreement stipulated that the Sultan should have 10 percent of the value of the slaves bought by Morice. “Though Europeans negotiated the abolishment of the slave trade with the sultan of Zanzibar . . . even contemporaries had few illusions regarding the ultimate aim of antislavery campaigns: removal of the Arab-Swahili as economic (and cultural) competitors” (J. Fabian 2000: 34–35). One common genre during the colonization of the continent was the presentation of the life histories of slave children (Alpers 2009: 32). One example is the book Kiungani; or, Story and History from Central Africa: Wri en by the Boys in the School of the Universities’ Mission to Central Africa, published in 1887 by UNCA (The Universities’ Mission to Central Africa; Madan 1887). It contains thirteen life stories of slave boys who se led on Zanzibar a er they had been liberated by the British. Another example is found in a report wri en by Donald Mackenzie, one of the British abolitionists who visited East Africa in 1895 to investigate the prevalence of slavery. His report (Mackenzie 1895: 17), stating that it “is a curious fact that Slaves have but very few children, owing, it is said, to the manners in which very young girls are treated by the Arabs and others,” must have horrified the readership. It must, at the same time, be noted that slave children were in high demand in many societies as potential new members in the kinship groups. Under the “lineage production” system children were a source of growth of the labor force. Glassman (1991) criticizes the first notion of an almost compulsory transformation of slaves into kinsmen. The affinity between the master-slave relation and family relations was pointed out by Aristotle in his a empt to explain ancient Greek slave society. He argued that the dominance of the male household head and submission of wife and children was as “natural” as the dominance of the slave owner over the slave(s) (see Brion Davis 1966: 69–70). For a critical review of their approach, see Glassman (1991) who points out their difficulties in separating forms of slavery from other forms of relations and that it is a misconception to see it as a continuum from having the status of slave to becoming kin. A common component in definitions of slavery is that it is a labor regime (Deutsch 2006: 4). Goody specifies that “The acquisition of a human being, by outright purchase or directly by capture, involved the most extreme exploitation of human potential, the height of domination, and depth of subordination” (Goody 1980: 26). British colonial power failed to recognize many forms of slavery around the Indian Ocean such as pawnship, which could be transformed into slavery, and voluntary forms since they focused on the criteria of violence during the enslavement and making the slaves work (see Campbell 2006b: xiii). Gambling is another force behind slavery which will not be further elaborated in this text. Burton, in his characteristic language, reports that “many Wanyamwezi have been compelled by this indulgence to sell themselves into

176 • Muted Memories 14.

15. 16.

17. 18. 19. 20. 21. 22. 23.

slavery: after plying through their property, they even stake their mothers against a cow or a pair of goats” (Burton 1859: 363). This strategy was reported by Burton (1859: 350) with the following formulations: “In times of necessity, . . . a man will part with his parents, wives and children, and when they fail, he will sell himself without shame.” There are reasons to question the last statement since it has been shameful to sell relatives into slavery. The same type of means of exchange or compensation, i.e. human beings, was often used both as bride-wealth and “blood money” (Graeber 2011: passim). There seems to have been some ambiguities regarding debts incurred by slaves. On the one hand, “the master could be held legally responsible should the slave-debtor default” says Glassman (1994: 75). Mtoro writes, on the other hand, that “if a merchant advanced money to a slave he knows that if the slave takes the money and goes inland or dies or run away and does not return to the coast, his money has fallen into the sea. He cannot go to the master and claim it” (J. W. T. Allen 1981: 176). The term means someone who could “be sent” and to act on another person’s behalf and not by his or her own will. For references, see Eastman 1988, 13. Burton (1859: 356) estimates that the price of a female slave at the middle of the nineteenth century was one-third higher than that of a male slave. For the term “unculturedness,” see Spear 1984, 303. A Kiswahili term for runaway slaves is mtoroka from the verb kutoroka meaning “to run away.” Sunseri reports that “slaves of the same owner or region often appeared in groups asking to be ransomed, suggesting a collective consciousness” (1993: 499). Cooper (1981: 287) notes that “Not to manumit slaves could result in the formation of a slave class which not only lacked the hope of further integration into society but which might develop traditions and self-consciousness over generations.”

7  Conflicts and Clashes in the Competition over the Control of the Caravan Trade on the Central Routes The Nyamwezi porters who arrived in Bagamoyo in August– September during the 1888 caravan season stumbled into a town where the internal “tensions reached a white-hot pitch” (Glassman 1995: 179) caused by the sultan signing an agreement with a German company that awarded foreigners the right to administer Bagamoyo.1 When the Nyamwezi began bringing their ivory to the coast, this ignited escalating competition over the lucrative trade, which eventually contributed to the colonization not only of Bagamoyo but also of the whole area called Deutsch Ost-Africa (German East Africa). This chapter will explore the growing competition over the caravan trade to examine how different actors positioned themselves in the struggle that eventually erupted in open rebellion in most of the coastal towns a er the Zanzibari and German colonizers had entered the scene. We will begin by exploring the gradual marginalization of the Nyamwezi traders/porters, continue with the marginalization of the middlemen-merchants in the coastal towns, and end with the rebellion, primarily a reaction to the Zanzibari colonization of the trade towns, fu-

178

•

Muted Memories

eled by the establishment of the Deutsch-Ostafrikanische Gellsellscha Offices in Bagamoyo and Pangani, which erupted in 1888 and sped up the formal German colonization of the area. The frictions were caused by the a empts of various actors to control the caravan trade. This struggle became particularly intense along the central caravan routes from Bagamoyo to the eastern Congo, the most lucrative caravan routes. The main fractures developed between three categories of actors: those who had initiated the trade, the Nyamwezi; those who initially controlled the trade in Bagamoyo, the middlemen-merchants; and finally, the newcomers who tried to monopolize it, the Omani Sultanate and, eventually, the Germans.

New Fault Lines of Frictions and Fractures A previous chapter showed that the caravan trade along the central routes began when Nyamwezi traders/porters extended their long-established trading networks in the interior of the continent down to the coast. The prime trade item in this extension was ivory, but also slaves. In the coastal towns they met another category of traders, the middlemen-merchants, in what has been called the Swahili world. These traders had already established themselves as the interface between the networks in the coastal hinterlands and the Indian Ocean networks.

The Marginalization of the Original Entrepreneurs and Explorers The original and small rill of elephant tusks carried to the coast, mainly for consumers in India and China, rapidly increased to a steady seasonal stream when the demand for ivory increased in Western Europe and North America. The Industrial Revolution saw an emerging class of newly rich who sought new means to express their wealth and status. The demand for ivory increased as it was carved into cutlery handles, piano keys, billiard balls, statues, and more. This triggered the interest of the coastal traders to gain control over the lucrative trade by arranging their own caravans to the interior of the continent. Traders from the East African coast could have reached Lake Nyasa by the 1770s and possibly also reached the Atlantic coast in Angola from Mozambique. By the 1850s, coastal traders were established on the shore of Lake Tanganyika (Fee 2012: 2). Eventually the Sultan of Zanzibar,2 as well as the Europeans, notably German trading companies, entered the game in their efforts to gain control over the trade. Initially, the Nyamwezi traders/porters had at least three advantages in the ivory caravan trade over the coastal traders; they operated much

Conflicts and Clashes in the Competition over the Caravan Trade  • 

179

closer to the source of the ivory; they had better knowledge about the local demands, which was important since the ivory was not obtained in exchange for money but in exchange for trade goods locally accepted and demanded, and finally; they could press themselves to extremely tough marches and thereby minimize transactional and operational costs. Still, most of them were on the losing side when the coastal traders entered the business. The massive influx of Indian capital, used to organize large-scale caravans from the coast, the virtual Arabic occupation of the eastern part of Congo for slave-hunting and trade, and the establishment of Arabic or Swahili settlements at Tabora and Ujiji in western Tanzania gave a new character to the caravan trade along the central routes (W. T. Brown 1970; W. T. Brown 1971a; B. Brown and W. T. Brown 1976; Sheriff 1987; Unomah and Webster 1976). The competition for control over the caravan trade was toughest in Bagamoyo and along the central caravan lines. The main reason was that the Sultan of Zanzibar had his eyes on this caravan route, the most lucrative, but also the most heavily financed. The scale of the caravan trade resulted in the enrichment of the few but the pauperization of the many. For many Nyamwezi porters the meeting with the coastal traders must have been highly stressful, if not traumatic. A song reproduced by Mtoro (J. W. T. Allen 1981: 165) captures these experiences. “My mother, my mother, listen,” the caravanners sang as they marched, the rich coastal people3 are swindling us All is gone in dues and taxed O father, all is gone Go away, you coastal people.

The predatory character of the coastal traders was noticed already by Richard Francis Burton (1859: 57) who wrote “out of each Frasilah (thirty-five lbs. avoirdupois) of ivory, from eight to fourteen dollars are achieved as duties to the Government of Zanzibar; the headman then demands six dollars as their fee, under various technical names plus one dollar for ‘ugali’, or porridge—the ‘manche’—and one dollar for the use of water— the ‘bovoire’.” The number of Nyamwezi porters in Bagamoyo increased at the end of the nineteenth century as the caravan trade boomed. They were, as already reported, regarded as the strongest and most reliable of the porters. But for the majority of the Nyamwezi their role in the caravans changed at the end of the century. From having been free traders/entrepreneurs and porters, the majority was reduced to the role of employed porters. Abdul Sheriff, in his now classic study on Zanzibar as a mercantile empire and its trade in slaves, concludes that “[t]he Nyamwezi had become a nation

180

•

Muted Memories

of porters, and their country a labor reserve, a foretaste of the colonial situation” (Sheriff 1987: 182). This was probably an overstatement. Stephen Rockel, in his study on Nyamwezi caravan trade, counters by claiming that Nyamwezi strongmen continued to organize huge caravans to the coast until the 1890s (2000a: 191). For the majority of the Nyamwezi caravanners, however, the change must have been more painful, when they were reduced to paid porters. They became the members of a new class of paid laborers, which developed through the globalization of the caravan trade. Furthermore, the booming of the caravan trade, and its extension to more distant areas, increased both the demand for porters and the time they had to spend on the routes. This opened up for the development of a labor market in East Africa on which the much-demanded Nyamwezi porters could sell their services (Rockel 2006a: 66). They were also in high demand in the growing colonial economy as in the Pangani valley where they were recruited both to work with the German-built railway from Tanga to the Kilimanjaro area and on the sisal plantations (Wright 1976: 583). They were among the first to enter into the emerging labor market in Tanzania through both push (pauperization) and pull (employment opportunities) factors. The Nyamwezi traders/porters usually arrived between June and September and stayed for a couple of months or more to sell their trade items, but also to wait out the vuli, or short rainy season, which was expected from September to November. One reason to wait out the short rains was that they start earlier in up-country areas than on the coast. The main harvest of grain crops in March and April had then been taken when the caravans from the coast reached the Ugogo area in central Tanzania, a place with frequent periods of food shortage. While waiting for the start of the up-country march, they tried to find paid jobs as plantation laborers on the coast.4 The large influx of workers in the agricultural sector drastically changed the local labor market. The Fathers at Bagamoyo’s Catholic mission station noticed that the wages demanded by freemen and the vibarua slaves, who worked for a salary given to their owners, decreased from 40 centime per day to 10–15 centime when the Nyamwezi porters were in town (S. Fabian 2007b: 57f.). Not all Nyamwezi traders/porters turned into wage laborers, but those who did became increasingly dependent on caravans organized from coastal towns like Bagamoyo for their march back to the Nyamwezi area. They probably had a choice to either return with the merchandise they had obtained in Bagamoyo in exchange for their ivory and/or labor and se le in their up-country societies, or to temporarily or permanently se le in or around Bagamoyo. When this step was taken, they probably also carried on the aspiration to assimilate into the urban “civilized” Waswahili

Conflicts and Clashes in the Competition over the Caravan Trade

•

181

identity or waungwana status, associated with a life in Bagamoyo (Glassman 1995: passim).

The Marginalization of the Coastal Middlemen-Merchants Bagamoyo, the entrepôt to the heart of Africa, a racted the interest of the Omani/Zanzibari Arabs and their Indian financiers, as well as the German colonizers. The scale of the trade, which passed through Bagamoyo, must have helped both of these colonial powers in their decision to make Bagamoyo their stronghold on the coast. As explained elsewhere, the Zanzibari Sultanate’s presence was more heavily experienced in Bagamoyo than in, for example, Saadani, a few kilometers north of Bagamoyo, or Kunduchi, a port close to Dar es Salaam.5 The sultan’s invitation to Indian traders and financiers who se led in Bagamoyo was a highly effective strategy to marginalize the town’s original middlemen. Through their financial strength and skills, the Indians could o en successfully compete for the position of the sultan’s Custom Master in the ports. One example is Sewa Haji Paroo, who was Bagamoyo’s Custom Master for twenty years. He became Sultan Sayyid’s “chief Indian creditor” in 1896, and the same year he and the sultan signed an agreement that gave Sewa Haji the monopoly for ten years to sell the Zanzibari trade products (W. T. Brown 1971a: 192).6 This meant that the former local elite and “owners” of Bagamoyo were increasingly marginalized when new actors entered the arena. The newcomers to Bagamoyo, the Arabs or Indians, used somewhat different weapons to outmaneuver the original traders, the middlemenmerchants. The Arabs used political and military weapons while the Indians used economic tools. The Arabs, although not a homogeneous group as will be shown in another chapter, were o en allied with or worked for the Zanzibari Sultanate in its efforts to gain control over the lucrative caravan trade. Their struggle was backed by the sultan’s Baluchi soldiers in the Kaole garrison. The Indians deployed a different strategy to control the caravan trade. They had the advantage of being part of trading networks that dominated the economic activities around the Indian Ocean or, as expressed by Daniel Woulfin (2011: 60), “East Africa was held together by an Indian trade monopoly rather than the Sultan’s power.” They also monopolized the financial sector in the caravan trade. The patricians’ cultural obligations, in the form of the practice of conspicuous consumption, particularly at the time of festivals and dancing competitions (see chapter 11), in combination with the Indians’ practice of moneylending, turned out to be detrimental to many middlemen-merchants. Increasingly indebted, as already mentioned, many lost their land, houses, and control over the caravan trade.

182

•

Muted Memories

In April 1888, a er German gunboat diplomacy, the sultan sold the right to the German East African Company DOAG (Deutsch-Ostafrikanische Gesellscha ) to collect customs from the caravan trade and to administer the coastal societies in his name. The middlemen-merchants in Bagamoyo suspected that the two colonial powers had joined hands against them. This was not an unjustified suspicion. Jonathon Glassman, writing on the rebellions in Pangani and Bagamoyo, notes concerning the large landowners in Pangani, mainly Arabs, that “spli ing with most of the Shirazi majumbe [the leaders of the middlemen/patricians in Pangani town], they formed a capitulationist party which sought the restoration of the sultan’s authority at any cost, even if it meant that the Germans would have to exercise the authority on his behalf” (Glassman 1995: 200), and regarding Bagamoyo that “by the time of this crisis [the so-called Bushiri rebellion], the Company and the Omanis had become fully aware of their need for one another” (1995: 211). A ri had opened up between the former leaders of the coastal trade towns and the intruders from Arabia and Europe. The booming caravan trade and the scramble for Africa added new dimensions to the brewing conflicts. A contributing factor in the escalating conflicts was the sultanate’s expanding presence in, and control over, parts of the coast. This presence grew a er the sultanate had been relocated from Oman to Zanzibar and Sultan Seyyid Said (sultan between 1826 and 1856) initiated a period of activities to get control over the caravan trade. The military clash in 1844 at Kaole, 6 km south of Bagamoyo, between the local Zaramo people and the sultan’s soldiers, has been interpreted as a direct move from the sultan to establish firm control over the caravan trade along the central routes (Unomah and Webster 1976: 310).7 The control was also strengthened through the appointment of “governors,” liwali(s) in the most important trade towns. At the beginning of the nineteenth century only Kilwa on this part of the East African coast had such a representative for the sultan. At the death of Sultan Seyyid Said in 1856, only Saadani, a port town just north of Bagamoyo, was without a liwali who took care of the sultan’s interests. Another of the sultan’s strategies to get control over caravan trade was to stimulate Indian financiers and traders to se le in the expanding port towns. Their number on the Mrima coast grew rapidly from about 5,000 people by the 1850s (Unomah and Webster 1976: 275–76) to probably reach 8,000 in the 1870s. They soon dominated the financial sector in this part of the world and both the sultan and Arab merchants took on enormous debt through loans to set up caravans for the interior of the continent or to establish plantations along the coast or on the adjacent islands. Still a strategy to manifest the control of the sultanate was to stimulate the establishment of plantations along the mainland coast and construc-

Conflicts and Clashes in the Competition over the Caravan Trade

•

183

tion of customs houses and military garrisons in the trade towns. Sultan Seyyid Said encouraged the cultivation of cash crops like cloves on Unguja and Pemba, grain, sesame and coconuts in most coastal areas on the mainland, and sugar in Pangani, north of Bagamoyo. In the areas surrounding Bagamoyo coconut plantations dominated from around 1850, and rice plantation in the swampy areas along Ruvu or Kingani River. These were large-scale plantations on the mainland, which were owned by Omani landlords but based on slave labor (Achola 2002: 5).8 So, by the end of the nineteenth century, both a class of wealthy Indian financiers and newly rich Omani landowners had developed in the Bagamoyo area, a development that was never paralleled in any other coastal trading se lement. The arrival of the Germans during the 1880s added new fuel to the fault lines that had opened up in the political landscape; the local Omanis, the former colonizers along the coast, and the new colonizers, the Germans, found themselves on one side while the former leaders, the middlemen-merchants and their followers found themselves on the other. In the new political landscape, the former leaders questioned the sultan’s right to allocate land, and the sultanate’s administrative functions. Eventually the conflict erupted in open confrontations, when the population in most coastal towns in August and September of 1888 revolted against the colonizing powers. One of the ignition sparks was the accusation that the sultan, without being the rightful leader of the coastal towns, had sold the right to collect customs to the German company DOAG. The ri between the rival groups revealed itself during the midSeptember 1888 rebellion when Bagamoyo’s leading Indians and Arabs ran to the local representative for the DOAG to urge them to suppress the rebellion, which had erupted in Pangani. They feared that anti-Omani feelings and actions would spread along the coast (Glassman 1995: 210). We see here how the ri opened up between the three incoming groups, Arabs, Indians, and Europeans on the one side and the old, but increasingly marginalized, merchant elite on the other. The reaction from the middlemen-merchants must be seen in light of their struggle against their increasing marginalization, and as a struggle to regain their former influence. This is clearly demonstrated in a petition, which the former ruling part in Pangani sent on 11 September 1888 to Sultan Khalifa a er the German colonizers had been expelled from the town (see below). Their feeling of marginalization is clearly expressed in the opening statement of the petition “that only wageni,” or strangers, “Omanis, Hadramis, Indians and Comorians—had been informed of Khalifa’s instructions regarding the Germans” (Glassman 1995: 224).9 They, as rightful owners of Pangani, had not been informed about the

184

•

Muted Memories

sultan’s agreement with the Germans to hand over administrative control of the coastal societies to them.10 Their feeling of exclusion is captured in a contemporary Swahili poem by Makanda bin Mweny Mkuu that explains the actions a er the liwali or governor of Bagamoyo had received a le er from Sultan Khalifa which explained that the control over the Mrima coast had been handed over to the Germans: Liwali akabaini The Liwali explained everything akatoa insani And gave instructions kanitie madiwani To call his counselors, kulla ahli shauri Everyone should hear the news moyoni yametupate In our hearts we have the hatuna budi ya vita Feeling, we have no choice na mengine hatukiri But war, there’s no alternative kweli Seyidi yup eke Surely Seyyid is isolated kuyafanya pekeyake And acted all alone atufanya wanawake He treated us like women, ndipo alitushauri Not even consulting us (Transcription by Carl Velten, quoted in Pike 1986, translation from Pike 1986: 203–4)

Rivalry and Festivities Resulting in Riots and Rebellion The frictions and fractures described above show that the ri s in Bagamoyo’s social fabric probably did not need much to develop into social disorder and open confrontations. The town’s rapid population growth during the second half of the nineteenth century created stiff competition for citizenship versus exclusion. The thousands of porters who arrived each year in July to September, probably dissatisfied with the trading conditions and their vulnerability to exploitation, certainly added energy to this powder keg. Thus, when the Nyamwezi porters entered Bagamoyo in July and August 1888 they “stumbled into a conflict that had been brewing in their absence” (Glassman 1995: 213). Furthermore, they stumbled into two of the most important holidays of the year: the Muslim Idd-al Haji and the syncretistic Siku ya Mwaka or Solar New Year. Both implied a lot of festivities, and the la er also extensive dancing. Soon, armed dancers started to challenge the political order in the coastal towns.

Frictions and Fractures through Festivities In nineteenth-century coastal towns in East Africa, common wisdom suggested that wealth should not only be accumulated but also consumed,

Conflicts and Clashes in the Competition over the Caravan Trade

•

185

preferably given away as public acts of generosity. Although leaders such as the majumbe had the right to collect taxes and other tributes, much of this was claimed by relatives and clients. As has been shown elsewhere, many elite families ended up in debt and eventually lost their houses, plantations, and slaves. One of the major ways to acquire and maintain high rank and honor, but also to lose wealth, was to sponsor feasts and dances (see chapter 11). A jumbe (singular), the most highly ranked position in the society, was the leader of a specific quarter of the town. All majumbe (plural) of Bagamoyo were ideally coequals but processes of internal stratification operated to make some more renowned than others. A jumbe was assisted by his maakida (sing. akida) who were the midlevel authorities for a number of subordinate titleholders in the highly hierarchized coastal towns. These title holders, with the jumbe at the apex, constituted a chama (pl. vyama), most commonly translated as a “dance society” (see chapter 11 for further information). The members of a chama assisted their jumbe in public affairs, most notably in dance competitions (shindano ya ngoma), and could expect to receive a portion of the jumbe’s wealth in exchange for their support (Glassman 1995: 158). Ideally, only a person born into an elite family could become an akida or a jumbe. In reality, however, even outsiders could, if they had wealth enough to pay for the high costs associated with the rites performed when a person aspired for the akida title. It seems as even slaves, who had acquired wealth through the caravan trade, could become maakida (Glassman 1995: 159.) The expanding caravan trade, the stiff competition to take part in town life, and the escalating struggle to take control over the profit-generating caravan trade contributed to the disruption of the established pa erns to achieve and maintain respect and authority. The newly rich found that the festivals and rituals, which involved conspicuous consumption, provided avenues to rank, influence, and inclusion. The caravan trade and the wealth accumulated through it threatened the established order in a town like Bagamoyo in two ways. One was through inflation, which occurred when additional wealth was generated through the booming caravan trade, and the other was through the sudden increase in people who could sponsor the important rituals. Now even young people in the port towns, the vijana,11 could acquire the necessary wealth through the caravan trade, without the assistance from the elder generation and thus compete with them for titles, prestige, and power. From the other side the patricians came under a ack from the massive influx of people of up-country origin, some of whom aspired to and claimed inclusion and membership in coastal life. No wonder that Glassman, regarding the 1888 rebellions, concludes, “as ambitious parvenus flooded the coastal communities, and as long-established Shirazi

186

•

Muted Memories

[the elite] families became even more weakened by the rising power of the state [the sultanate] and the Indian creditors, the former élite soon lost exclusive control of the dance rituals which were so important to the public affirmation and enhancement of their authority” (Glassman 1995: 162–63). The costs involved in the competitive dance festivals became a curse for some, but a blessing for others. The sponsoring of the dance competitions and the conspicuous consumption became detrimental for many patricians, since wealth was redistributed among new actors involved in the caravan trade. The role of the Indians is captured in a subtle way in a Swahili poem that details how two competing patrons became involved in a competitive slaughtering of livestock, while the Indians benefi ed from their business. A couple of the verses, reproduced by Glassman (1995: 169–70): They came out to eat the meat the Indians were delighted Because of the endless profit coming to their shops . . . My brethren, let me give you the facts, the Indians were happy And they made a profit the profit was very substantial

Riots and Revolts The competitive character of the dance performances o en resulted in open hostility, including violence. The gradual establishment of the Zanzibari Sultanate, which undermined the former power balance established between the leaders of Bagamoyo and the leaders of Uzaramo, the area the caravans had to pass to reach either the coast or the interior of the continent, contributed to the escalation of riots and violence in Bagamoyo during the second half of the nineteenth century.12 But more important to the future development of the coastal towns was the appearance of European colonialists. Parallel to the growing tensions between the former owners of the trading towns, and the new traders under the Zanzibari Sultanate, was a growing competition among the European colonizing powers over the control of parts of Africa, which, on the east coast, also meant control over the caravan trade. Already in 1837, a few years a er Sultan Seyyid Said moved his court and capital from Oman to Zanzibar, a US Consul was established on the island, followed in 1841 by a British, and in 1844 by a French Consul. They were followed by four German trading companies (Russell 1980: 10).

Conflicts and Clashes in the Competition over the Caravan Trade

•

187

One step in the colonial partition process was the formation of a Boundary Commission in 1886, established by Germany, Britain, and France to determine the sultan’s influence over the coast of mainland East Africa (S. Fabian 2007b: 92–93). It finished its work in early 1887 with a suggestion that the sultanate’s dominion was a 10-nautical-mile wide coastal strip from Lamu to the Rovuma River and to split the rest of East Africa. This followed the negotiations at the Berlin Conference in 1884–85, which resulted in the partition of this part of East Africa into a German and a British area, and with the Zanzibar Sultanate as semi-independent. The German lust for a colony in the area is perhaps best illustrated by “the psychopathic, and, as it turned out later, also criminal adventurer” (Stoecker 1986: 95)13 Karl Peters’ frenetic treaty-signing activities in the near hinterland area at the coast. He is said to have “succumbed to the chauvinist Kolonialrausch or ‘colonial fever’ that gripped the imaginations of many Germans in the early 1880s” (Glassman 1995: 180) and to have expressed his desire “to obtain for myself personally an empire that was to my taste” (as cited in Stoecker 1986: 29). Karl Peters was the president of the Gesellscha für Deutsche Kolonisation (Society for German Colonization) and had managed to sign treaties with local leaders, which gave the Germans the right to administer the areas through which the central caravan routes passed. The German claims to these areas became formal when Kaiser Wilhelm proclaimed that the areas west of the ten nautical miles–wide area along the coast called “the Sultan’s dominions” were German areas. With this the imperial race to, and partitioning of, the Swahili world became official. That the control of the caravan trade was the primary colonial interest for the Germans, as it had been for the Zanzibar Sultanate, became evident when the German General Consul for Zanzibar, Gustav Michahellas, on 25 April 1885, informed Sultan Barghash bin Saíd (sultan between 1870 to 1888) that “Germany claims the main caravan route to Tabora and the Lakes” (Beck n.d.: 16). Five warships were sent from Germany to Zanzibar, as an act of gunboat diplomacy, to persuade the sultan to accept the Germans’ claims. On 7 August 1885, they arrived in the lagoon at the harbor of Zanzibar town and threatened to bombard the sultan’s palace (Woulfin 2011: 93). East Africa was thus divided into three domains. The sultanate would remain with 9,190 square miles including the three islands Pemba, Unguja, and Mafia plus a 10-nautical-mile wide section on the coast between the mouth of Tana River to the north and the southern bank of the Minengani River in the south. Germany was allocated a protectorate of 25,900 square miles in the coastal hinterland, plus 122,800 square miles in up-country societies described as “unclaimed territory.” Britain was given 72,000

188

•

Muted Memories

square miles as their “protectorate.” “These new borders were approved without consulting the people living within them, whether Europeans, Africans, and Swahilis” (Woulfin 2011: 101). The agreement also gave the Germans the right to administer the collection of customs duties on the Mrima coast, from Pangani in the north to Dar es Salaam in the south, including Bagamoyo. However, their ambition was not only to administer the area in the name of the sultan, but to get political control over the territory. The active conquest of what was to become German East Africa was achieved through a number of colonial wars that occurred between 1888 and 1900 (Wright 1976: 567). The Germans’ first interest in the area focused particularly on Bagamoyo since the customs duties earned on trade here was more than two times higher in 1885–86 than in the second most important town, Kilwa. The Germans had been very active under the reign of the new sultan, Sultan Khalifa bin Said bin Sultan (1888–90), to make him accept a new treaty called the East African Concession. This concession gave the German trade company: All the power which he [the sultan] possesses on the mainland Mrima [including the ports of Pangani, Bagamoyo and Dar es Salaam], and in all his territories and dependencies south of the Umba River, the whole administration of which he concedes and places in their hands to be carried out in His Highness’ name and under his flag and subject to his Highness’ sovereign rights. (Woulfin 2011: 107)

In reality, the German company acquired the right to administer the trade and to collect custom duties from the trade in the area and taxes from the inhabitants, and to occupy the public buildings in the area. The last right turned out to be sensitive since it gave the Germans the right to claim access to the houses inhabited by the sultan’s liwali or representatives on mainland East Africa. The German company also acquired the right to print its own bank notes and mint its coins, which was of strategic importance to get control over the economy, including the caravan trade. The uprising against the new political and administrative order occurred simultaneously in port towns along the coast, from Tanga in the north to Lindi and Mikindani in the south of what is today Tanzania. It spread along the central caravan routes as evidenced by an a ack by rebel troops on Mpwapwa, one of the most important trade and rest stations along the central routes (Rockel 2009) in central Tanzania in 1889. Two exceptions were Bagamoyo and Dar es Salaam, where the leaders objected to the burden of the sultanate and where freedom fighters were active but the rebellion never got a firm foothold. The fact that the rebellion process in Bagamoyo differed from what happened in some of the neighboring coastal towns like Saadani, Pangani, Tanga deserves further exploration.

Conflicts and Clashes in the Competition over the Caravan Trade

•

189

Illustration 7.1. Coins minted in 1890 by the Deutsch-Ostafrikanische Gesellscha . Photo by the author.

Particularly so since the initial ignition to revolts in Bagamoyo and Pangani seems to have been identical: the liwali of each place refused to hand over the sultan’s red flag to representatives from the German East Africa Company and they did not allow the company to hoist its own flag as a claim to represent the area or town.

The Expulsion of the Germans from Pangani To get a fuller understanding of the 1888–89 rebellion than what may be grasped from the Bagamoyo horizon, the focus will temporarily be moved 115 km to the north on the Mrima coast to the port town Pangani.14 There are contradicting interpretations regarding the causes behind the rebellion that spread among the port towns along the coast. One is that the revolt was against the intruding German trade company or, as Mervyn Hill (1957: 50) expresses it: “‘the rebellion against the Sultan’s authority which had broken out on the mainland under the influence of slave-dealers’—were nonsense, for the revolt was not against the Sultan but against the German company.” Jonathon Glassman (1995), in his elegant and convincing analysis, develops the argument that “the intensity of the rising of 1888–89 was inversely proportional to the degree of early German presence: most intense at Pangani and Saadani, least at Dar es Salaam and Bagamoyo. Clearly then the outbreak of rebellion cannot be explained simply in terms of the level of German provocation” (Glassman 1995: 196). It must also be remembered that the number of DOAG officers was low. Approximately

190

•

Muted Memories

fi een officers were stationed on Zanzibar in 1886–87, the same number in the coastal towns where the company had established offices and another fi een at its inland stations (Glassman 1995: 187n34). Glassman demonstrates that this presence was rather the catalyst, which caused maturing frictions and fractures to erupt in a violent revolt at that particular point in time. A central issue in this drama was the escalating competition over the control of the caravan trade. One ingredient in this brewing pot was the patricians in Pangani, who had experienced increasing marginalization at the hands of the Omani Sultanate. This not only concerned their control over the caravan trade on the northern or “Masaai” caravan routes and plantations established at Pangani, but also the status of the slaves. In 1873 and 1875, Sultan Barghas bin Said had reached agreements with the British Consul John Kirk, who acted on behalf of her Majesty, to stop the slave trade, which threatened not only the plantation economy around Pangani, but also the slave trade. An additional friction, Glassman writes (1995: 187n34) was the increasing stream of up-country people, washenzi or “barbarian,” who came to Pangani as traders and porters. When the DOAG representative Emil von Zelewski entered a mosque in Pangani with shoes and a dog, the powder keg exploded.15 There is probably still an explanation for the rebellion. Arabs had established themselves as plantation owners on the islands Pemba and Unguja (o en called Zanzibar), and along the coast of mainland Africa. These plantations were extremely labor-demanding and the US type of plantation slavery, the capitalistic form of slavery, developed here. People were captured both in up-country and coastal societies to be used as slaves on the plantations. Glassman (1995: 110) writes that: “Both wri en and oral sources suggest that the rebels were defending their status as autonomous clients against a empts by Arab masters to subject them to regimented plantation labor. A contemporary German wrote that many of the rebels were caravan traders resisting enslavement for debts incurred in the ivory trade.” Furthermore, a unique combination of specific cultural elements amalgamated in August 1888. One of the most important Islamic holidays, Idd al-Haji, the end of the pilgrimage to Mecca, occurred that year on 19 August. One day before, a more syncretistic holiday,16 the Siku ya Mwaka17 or the Solar New Year, occurred and was celebrated. The Siku ya Mwaka, also translated as “New Year’s Day,” was celebrated by all the people waking up early and going to the beach to have a ritual bath in the Indian Ocean. The Bagamoyo-born author Mtoro reports that when a man dipped himself in the ocean he prayed “God, take from me the sickness and give me health. Let me be clean this year and the year to come” (J. W. T. Allen 1981: 192).

Conflicts and Clashes in the Competition over the Caravan Trade

•

191

The Siku ya Mwaka is an anthropologically well-documented type of ritual of reversal or rebellion of the established social orders and hierarchies. For one day, clients and the poor could behave as if they were the patrons or lords, trans-dressing was common, and in other parts of Africa it has been reported that women that day could drink alcohol and sing the lewd songs usually sung by men. Mid-August 1888 was a time when people from the surrounding villages and up-country porters camping around the towns began to fill the streets to celebrate the Siku ya Mwaka. They had good reasons to do so since the established patron-client relations and the notions of “Arabic” hospitality meant that they could expect food and other gi s during the double holidays occurring on two consecutive days. Glassman (1995: 206) reports that “the carousing caravan personnel wandered the streets of the town shooting their guns in the air, and some spoke menacingly of looting the Indians’ shops. Such talk was o en heard during the carnival of the Solar New Year. But on this occasion it seemed particularly alarming.” Mid-August also marked the end of the caravan season. Many caravan porters who had set out two to three months earlier from their up-country societies eventually reached the coast, and an ever-increasing number settled in the camps around the coastal towns like Pangani and Bagamoyo. It is estimated that approximately eight thousand porters were present when the rebellion reached Bagamoyo. These three important cultural occasions; the celebration of the Siku ya Mwaka, the Idd al-Haji, and the arrival of thousands of porters in caravans from up-country societies, coincided with a growing discontent among the town’s elite who had seen their interests and independence increasingly circumscribed by the sultanate on Zanzibar and its henchmen, the Indians and the Arabs. The ignition, which exploded this powder keg, came when Emil von Zelewski, the representative for the Deutsch-Ostafrikanische Gesellscha , (German East African Company) in an a empt to be recognized as the sultan’s representative in Pangani, tried to usurp this position from liwali Abdulgawi bin Abdalla, called by Glassman the “governor” in the town. Assisting him were more than a hundred armed soldiers from the German warship Carola, which had arrived in Pangani’s port on the eve of the New Year’s Day. More than a hundred armed soldiers disembarked and occupied the town. The soldiers arrogantly broke into the house of liwali Abdulgawi and forced themselves into the female seclusion of the house in search for the liwali, a horrifying act in the eyes of the town dwellers who must have seen this as violations of both heshima, respect, and religious cleanliness, usafi.

192

•

Muted Memories

The intruding Germans learned that the liwali was at Friday18 prayers in the mosque. Von Zelewski and his men “forced their way into the mosque, disrupting the service by loudly demanding the governor’s whereabouts. Wearing shoes and accompanied by one of Zelewski’s hunting dogs, the Germans desecrated the mosque. Abdulgawi escaped but the holiday prayers had to be cancelled” (Glassman 1995: 5). Pangani was deserted during the days the Carola was in the harbor. Abdulgawi, with a German price on his head, escaped to Zanzibar (Glassman 1995: 217). When the Carola le on 23 August 1888, von Zelewski did not know that he had only ten days le in Pangani. These episodes are captured in a contemporary Swahili poem which explains: Walikuja wakangia They entered (the mosque) na majibwa yao pia With their great dogs too Liwali akakimbia And the governor had fled asihimili He could not bear to remain. Wakaikuna fahamu They struck down bendera ya muadhamu The flag of Islam na sasa wameazimu And now proposed kutweka yao sikia To raise their own. (Abdallah el Buhriy, translated by J. W. T. Allen 1960, quoted in Pike 1986: 206)

The cultural illiteracy or arrogance of von Zelewski must have been counterproductive to the wish of the DOAG and the German empire to rule the area through the sultan. The desecration of the mosque was a direct challenge to the ruling Omani dynasty’s authority. It is the ruling family or dynasty in a Muslim society that carries the duty to guarantee the religious institutions under its control. What happened in Pangani turned out to be a direct challenge to the sultan’s power, which also turned into an armed conflict between the sultanate and some of the patricians and other people in the coastal towns. It was probably cultural ignorance rather than empathy19 that lead von Zelewski to his next move. He opened the jail and released the inmates. The majority of them were either debtors, who could not repay their loans, or summoned slaves who had run away from their masters (Glassman 1995: 5, passim). The sultan’s power to imprison both categories was a fundamental corner stone in his politico-economic regime. Whatever the reason, the effect of von Zelewski’s actions was to undermine the authority of the sultan in the eyes of already rebellion-ripe coastal town people. Some of the decrees issued by von Zelewski during the few days he was in Pangani distanced him and the Germans from the people of the town, as well as from the caravan people. A highly sensitive issue was his

Conflicts and Clashes in the Competition over the Caravan Trade

•

193

a empts to prohibit the traders’ sale of arms and ammunition. Both items were among the most highly prioritized goods, sought by the caravan traders and porters in exchange for their ivory and other goods. Consequently, weapons and ammunition were among the most important items for the coastal traders in their dealings with the caravan traders/porters. Furthermore, weapons were essential for the protection of the caravans’ people and goods when they passed through hostile societies, and for the town traders to protect their shops and goods (Glassman 1995: 217). The third ignition spark in this powder keg came when the Germans tried to stop an incoming shipment of one thousand gunpowder kegs, and ordered the ship to return to Zanzibar. The following night, hordes of people from the surrounding camps and se lements streamed into the town, seized the gunpowder, and distributed the cargo among themselves (Glassman 1995: 220). The further escalation of the conflict is highly revealing regarding both the power structures and the political game in a coastal town like Pangani: In early September a crisis erupted in which armed crowds threatened Zelewski and his handful of colleagues. This alarmed the leading townsmen, who feared that Zelewski’s death would embarrass the Sultan and outrage his powerful European allies. Deciding that ge ing rid of the Germans would be preferable to killing them, the Shirazi and Omani leaders locked the Germans in the Company station house and set a guard of Arab mercenaries to protect them from the crowds outside. But it was not until a er several ugly altercations with the crowd that forces still loyal to the Omani state could send the Germans safely to Zanzibar, and then only under the close guard of additional troops sent by the Sultan (Glassman 1995: 5–6).

The rebels not only turned against the Germans but also against the sultanate. A clear and dramatic illustration of the sultan’s loss of power over Pangani was when the DOAG chief station officer on Zanzibar, Ernest Vohsen, traveled to the town in the sultan’s steamer Barawa, in an a empt to rescue his countrymen, who had, as mentioned above, been locked up in DOAG’s Customs house/office by people still loyal to the sultan. A er having entered a small boat at the mouth of the Pangani River, his oarsmen rowed him into the harbor. When he tried to disembark onto solid ground he was told by the people assembled there that “the Sultan of Zanzibar no longer ruled at Pangani. The rebels waded into the river, firing, and Vohsen made a frantic escape” (Glassman 1995: 221). In the weeks that followed, not only von Zelewski and the people of the DOAG were expelled from Pangani, but also the Omani state, which had granted the DOAG the right to collect customs from the caravan trade along the coast (Glassman 1995: 2). The expulsion of the Germans from Pangani was not the end of the German presence in East Africa. The hu-

194

•

Muted Memories

miliating events in Pangani triggered Reichkanzler Bismarck into fullfledged German colonization of the area.

Resistance to Colonization in Bagamoyo The first phase of the rebellion in Bagamoyo paralleled the one in Pangani (and Dar es Salaam)20 in the way it developed, although the underlying causes were slightly different. A er having received the right to collect the customs from the caravan trade, DOAG soon also showed its political aspirations. A few years a er the “Scramble for Africa” during the Berlin Conference, the German Trading Company sought political hegemony along the coast. One way to achieve this was to wrench the role as the sultan’s representative on mainland Africa out of the hands of his appointed liwali. Vohsen, the German administrator of the trading company on Zanzibar arrived to Bagamoyo on 15 August to oversee how the instructions he had issued regarding the inauguration of the company’s administration in Bagamoyo were followed. At lunchtime, 16 August 1888, the Germans were to fly the sultan’s flag over their administrative building. But Bagamoyo’s liwali refused to hand over the flag to the Germans (Glassman 1995: 201–2)21 as noticed by an eyewitness of the ceremony: “The place was magnificently decorated. The orchestra from the Warship H. M. Carohl performed German military melodies. From a platform Vohsen announced the first new regulations. The old Liwali of the town handed over the administration, but was by no means ready to lower the red flag of the Sultan who held according to the contract still the sovereignty” (Friedrich Ferdinand Mueller 1959, as quoted in Kajeri and Henschel n.d.: 4) The German trading company had managed to lease a house, the Old Fort or Arab Fort, which features in Tanzania’s a empt to nominate Bagamoyo for the World Heritage List, from the prominent trader Sewa Haji Paroo to be their new station office. The crux was that the same house had been leased to liwali Amir bin Suleiman to be his residence and he refused to move out. This forced the DOAG representatives to rent a house in another part of the town, a place they had to use for the ceremony and to which they now wanted the sultan’s flag to be moved.22. The liwali refused to hand over the flag and the temporary flying of two flags representing the sultanate in Bagamoyo made Vohsen hurry back to Zanzibar to complain directly to the sultan (S. Fabian 2007b: 107). The German takeover in Bagamoyo was orchestrated by the same lack of political correctness and cultural sensitivity that characterized their a empts to control Pangani. A missionary at the Catholic Mission in Bagamoyo reported that:

Conflicts and Clashes in the Competition over the Caravan Trade

•

195

In Bagamoyo the Germans ordered the Arab authorities to evacuate the administration buildings. They announced in public the new regulations, hoisted their own flag, informed about their laws, and introduced their own police force, their own customs and tax-duties: in short they manifested their intention to change the whole life. The townsmen listened first with big surprise. But the way the Germans approached them made all furious. Then rage burst out. (Report from Catholic missionary Ludwig Gommenginger, quoted in Kajeri and Henschel n.d.: 5)

In this insecure situation a group of Indian merchants in Bagamoyo, as British subjects, turned to the liwali to ask for protection. When liwali Amir bin Suleiman explained that if the Germans took his flag “the Warima [people from the coast] and Washenzi [here porters from up-country societies] people would ‘set fire to the town’” (S. Fabian 2007b: 108) and “if threatened to sack the town . . . he could do nothing to protect the merchants” (Glassman 1995: 206). As a consequence twenty-three of Bagamoyo’s Indian shop owners, both Muslim and Hindu, sent a petition to the British Consul-General on Zanzibar, their formal protector, explaining the situation and the threats, whereupon the British Consul-General approached his German counterpart and demanded that the Indians’ interests should be protected. In response to the request, the German gunboat SS Möwe was sent to Bagamoyo and had already le Zanzibar’s harbor when the sultan and the German Consul-General Michahellas “reached a face-saving agreement whereby DOAG would simply take over the liwali’s residence and then his flag would not have to be lowered” (S. Fabian 2007b: 109). The Germans first tried to blame the tense situation on the caravan porters who swarmed into the town during the caravan season. When Bagamoyo’s merchants questioned this explanation, the Bagamoyo DOAG station chief, Freiherr von Gravenreuth, threatened to ban the trade in firearms and ammunition. This ban seems to have come into effect on 22 August, but the subsequent development did not parallel the ban von Zelewski imposed in Pangani. The reason was that the ban in Bagamoyo was immediately li ed “in order to cause no injury to trade” (see S. Fabian 2007b: 111–12n38), which may indicate that von Gravenreuth was more culturally literate than von Zelewski. The situation in Bagamoyo, however, was not under control. There is a statement from a German source from the end of August 1888, reproduced by S. Fabian (2007b: 115) that explains: two chiefs of the town, Salim bin Abdallah and Kisoka, aware of the enormous sums to be made on the ivory coming in from the interior, on the crossing of the [Kingani] river, on the fishing, etc. have silently organized a general uprising. They were not the only ones. The ‘semi-Arabs’ [proba-

196

•

Muted Memories

bly Swahili and Shirazi merchants, see chapter 11] from the mosque in the Mangesani quarter and from the mosque in the Gongoni quarter . . . were always ready [at least verbally] to fall upon Gravenreuth and kill him.

The Germans were not ignorant of the situation, which must have motivated the involvement of SS Möwe and its soldiers in the development of the conflict. However, the move from the trade company to request use of the German state’s military machinery turned out to have wider political implications. The private company DOAG managed to receive help from the German state in the form of soldiers from SS Möwe, which had anchored on the roadstead outside Bagamoyo. Representatives from the German state, the Consul-General Michahellas on Zanzibar and Admiral Deinhard, disagreed with the confrontation with the sultanate orchestrated by Vohsen, the DOAG, and its representative in Bagamoyo. It was in the German colonizers’ interest to cultivate good relations with the sultanate (S. Fabian 2007b). Even Reichkanzler Bismarck reacted to the use of German naval forces to support the interests of a private company (Glassman 1995: 209). Eventually the Germans would be able to lay their hands on the sultan’s flag, when marine soldiers from the SS Möwe, unaware of the face-saving agreement reached on Zanzibar, surrounded Amir bin Suleiman’s house, and ordered him to hand over the flag with his own hands. When this was done they chopped up the flagstaff to hinder a new flag to fly from Amir’s residence. The scene was witnessed by a big crowd consisting of town dwellers and incoming porters. They did not react immediately, but one month later (Glassman 1995: 207–10), the Germans in Bagamoyo began to sense the presence of eight thousand newly arrived porters and the tensions between different segments in the town. The DOAG station chief in Bagamoyo, Freiherr von Gravenreuth, once again, requested that a German gunboat should be present in Bagamoyo, in case the situation became violent. A new German man-of-war, the frigate SS Leipzig, arrived on 20 September 1888.23 The uprising in Bagamoyo broke out early on 22 September.24 The leaders of the revolt were neither Arabs nor wealthy caravan traders reacting to the German intrusion but local leaders, the majumbe (Glassman 1995: 248). They had experienced increasing pressure and competition both from the sultanate and from German colonization and experienced a gradual marginalization in the caravan trade. Well-known Bagamoyo leaders at this time were majumbe Salim bin Abdallah, Kisoka, Bomboma, Marera, Simba Mbili, Korandi, Makanda bin Mwenyi Mkuu and Fimbo Mbili. When the rebellion broke out, the DOAG station chief von Gravenreuth and Admiral Deinhard had already been fetched the day before by frigate SS Leipzig (S. Fabian 2007b: 118).

Conflicts and Clashes in the Competition over the Caravan Trade

•

197

A crowd of people started to hack up a boat on the beach that belonged to DOAG. This act ignited the rebellion in Bagamoyo. When a DOAG officer tried to stop this he was threatened by the crowd and had to retire to the station house. The angry crowd surrounded the so-called Arab Fort on all sides and opened fire on it. The locked-in Germans mounted a canon on the flat roof of the house and began to fire on the people on the ground. They also signaled to the SS Leipzig, which was moored on the roadstead, and asked for help (S. Fabian 2007b: 121). In response, 150 to 200 soldiers (the sources disagree on the figure) were sent from the SS Leipzig to assist their countrymen in the surrounded house and martial law was declared by the Germans during the rebellion. They arrived with machine guns mounted on the small boats they used to reach land, and used the guns to fire on the people they met while the SS Leipzig lobbed shells over the marine soldiers into the town. It was noticed that the SS Leipzig “fired so many shells into the town that a firestorm broke out and decimated the northern quarters” (S. Fabian 2007b: 138). When the mariners landed they “overran the narrow streets in the town, shooting indiscriminately at any African they saw. . . . in reality it was massacre . . . They were finally the masters of Bagamoyo—but they were the masters of a deserted town” (Glassman 1995: 210–11). A contemporary Swahili poem (by Makanda bin Mwenyi Mkuu) captures the havoc: Basi ikawa kushika With the first sign of a ack na bunduki kualika Rifles spit fire mji mzima kiwaka The entire city was ablaze watu wakenda azari People stumbled about dead (Transcription by Carl Velten, quoted in Pike 1986, translation from Pike 1986: 24)

In a parallel movement the Germans in the surrounded Arab Fort had sent out their African soldiers, askari, to set fire by hand to the thatched palm roofs in the poorer parts of the town. A German eye witness reported that “the terror was general” (S. Fabian 2007b: 122). Four thousand of the denizens lost their homes and more than a hundred people lost their lives (S. Fabian 2007b: 123).

Bagamoyo under Blockade and Siege Local traditions, W. T. Brown writes, held that a message requesting support was sent from Bagamoyo’s leading jumbe(s) Bomboma and Marera to Bushiri25 bin Salim al-Harthi, one of the leading patricians in Pangani. Bushiri has been described as “the scion of the al-Harthi clan, whose long

198

•

Muted Memories

enmity with the Busaidi dynasty [the Sultan of Zanzibar] made his resistance to any intervention on the coast predictable” (Wright 1976: 568). The al-Harthi clan has been identified as the oldest Omani clan on Zanzibar and leading rivals of the al-Busaidi dynasty, who never became assimilated into the Swahili culture in the way the al-Harthi clan have been (Glassman 1995: 236). He came to personify the long-standing conflicts between the mainland middlemen-merchants and plantation owners and the intruding sultanate. A er his original success in expelling the Germans from Pangani in September 1888, he and his men started their advancement toward Bagamoyo in November the same year. When he reached the town he commanded a force of 3,000 well-equipped men and two canons that had been taken from the up-country DOAG offices in Dunda and Madimola. Along the march from Pangani to Bagamoyo, the forces of Ismael of Windi and Bwana Heri of Saadani had also joined Bushiri’s force, but under the command of their respective leaders (S. Fabian 2007b: 136; see also W. T. Brown 1971a: 279).26 Bushiri arrived on 3 December to Bagamoyo while it still was dark. He and his men occupied some of the stone houses deserted by Indian traders during the September riot. Bushiri himself se led in the building, later to become a school, built by the prominent trader and financier Sewa Haji Paroo. During the next few days, intense fighting broke out between the men from Pangani, Windi, and Saadani on one side and the Germans on the other. Bushiri’s men took whatever they could lay their hands on including slaves and some four hundred Nyamwezi porters were enslaved and a slave market was established at Bushiri’s residence “at which slaves can be bought in almost any numbers at very low price” (S. Fabian 2007b: 142–43). At the end of 1888, three-quarters of Bagamoyo had been destroyed and plundered, either by the Germans or the rebels. The havoc in the town in late September caused approximately five thousand of the townspeople and porters to take refuge at the Catholic mission station. Some Africans and Arabs were evacuated to Zanzibar. A new flow of refugees reached the mission under the December confrontations and destruction of parts of Bagamoyo. When the town was rebuilt again a er the expulsion of Bushiri and his troops, many house owners replaced the flat roofs with peaked roofs and many changed the palm leaves thatching the houses with corrugated iron sheets. The rebels withdrew from Bagamoyo and established a camp at Nzole 7 km southwest of Bagamoyo (W. T. Brown 1971a: 279) in December 1888. This move may be interpreted as a conscious measure to threaten the control the old arch-enemy to the coastal traders, the Sultan of Zanzibar, had over the caravan trade. It was also, however, a direct threat against

Conflicts and Clashes in the Competition over the Caravan Trade

•

199

the German interest to control the caravan trade. The campsite was strategically situated along or close to the caravan routes to Bagamoyo and Dar es Salaam and on 17 January 1889, Bushiri’s men a acked the German fort in Dar es Salaam. This development alerted the German Trade Company, which once again asked for military assistance from the German government. In the struggle over control of the caravan trade, the company saw it as urgent to have the Bushiri camp destroyed before the next caravan season started in June–August 1889. The Germans organized their fight against the rebels along two fronts: a military a ack against the camp and an economic blockade against the “Arabs” accused to be behind the revolt. As a response to the German trade company’s request for military assistance, Bismarck appointed Lieutenant Hermann Wissman27 as Reichkolonialkommissar (Imperial Colonial Commissioner) over the German areas in East Africa in February 1889. Wissman was a well-known “explorer” who had made two crossings of Central Africa. One was carried out for the German East African Company, DOAG, and the second for the Belgian King Leopold II (Stoecker 1986: 99). He arrived in East Africa as Major Wissman in April the same year with “600 Sudanese mercenaries [recruited] in Egypt, together with 350 Shangaans from Mozambique, 50 Somalis and some others” (Stoecker 1986: 99). They landed in Bagamoyo and Dar es Salaam, moved westward with cannons and ammunition transported by Nyamwezi porters, and soon stormed the heavily fortified Bushiri’s camp. The camp was conquered and demolished by the German forces on 8 May the same year,28 and in November 1889 an agreement was reached between the German government and the company to hand over the company’s concessions to the government. This implied the formal German colonization of East Africa, present day Tanzania, Rwanda, Burundi, and parts of Uganda. Bushiri escaped and fled westwards to find new allies. Lorne Larson (1977: 39) reports that “eventually he was able to recruit the Kisaki Mbunga as allies and advanced again on the coast. On 20 October 1899 [sic, should be 1889], four miles inland from Bagamoyo, a large Mbunga force for the first time encountered disciplined rifle fire. A er several fierce clashes with German columns, both sides withdraw with heavy casualties.” An African chief captured Bushiri and turned him over to the Germans for a ransom. Bushiri was taken to Pangani where he was tried and hanged at the age of thirty-six on 15 December 1889. Bomboma, Makambu, Marera, and other rebel leaders in Bagamoyo were also captured and hanged in Bagamoyo. To counteract the September 1888 rebellion, the Germans also opened up a second front against the rebels by proposing a blockade of the coast. They did this by playing the “Arab card” and accusing slave-trading Arabs

200 • Muted Memories for what they called the Araberaufstand (the Arab revolt).29 This was done on two fronts. One was as an argument in Germany for the colonization of parts of East Africa. In October 1888, the German Counsellor instructed the German Government to motivate the colonization as a measure to fight the slave trade (Stoecker 1986: 98). The second front was to seek support from the rival colonial power Great Britain for a blockade of the East African coastal ports. The Germans painted the picture of the “guileful Arab who by deceit and trickery enticed African chieftains to supply and sell slaves” (Koponen 1988: 100). The revolts, the Germans explained, were “brought forward principally by the Arabs who are interested in the slave trade” (S. Fabian 2007b: 124). The Germans tried to achieve two goals with the “Arab card”: to secure British support for a blockade against the rebels in the trade towns and raise support to an action that was against the interest not only of the traders in the coastal ports but also of the Sultan of Zanzibar and his control over the trade with mainland Africa. The joint German/British blockade came into force on 2 December 1888 (Stoecker 1986: 99). The purpose of the blockade was explained in an 1888 memorandum from the German Embassy: Joint action of Germany and England in support of the Sultan at first to maritime action, and for this purpose perhaps to establish a blockade of the coast of the mainland of Zanzibar between Kipini and the River Rovuma, by German and English ships in co-operation with the Sultan of Zanzibar. The object of such a blockade would be to cut off all traffic with the insurgent coastal districts, and especially that in slave-vessels, and the carriage of arms and ammunition. (Reproduced in Woulfin 2011: 125)

With this rhetorical trick to merge the proposed blockade with the antislavery issue, the Germans temporarily found allies in different quarters in favor of their proposal.30 The Secretary for the British Anti-Slavery Society added fuel to the issue when he asked, “how long will these Arab man-stealers be suffered to pour their slave cargoes into the island of Pemba, in order to cultivate the clove plantations of the Zanzibaris?” (as reproduced in Woulfin 2011: 127).31 The blockade proposed by the Germans, rhetorically motivated as a support to the sultan, was also a weapon in their struggle to control the trade in the coastal towns, since such a blockade would hit hard not only against the local traders in the coastal towns but also against the interests of the sultan and his supporters who controlled much of the lucrative caravan trade from East Africa. It has been estimated that about 200 tons of ivory annually passed the market on Zanzibar during the first decades of the nineteenth century. At the end of the century the volume had doubled to 400 tons (Beachey 1967: 287).

Conflicts and Clashes in the Competition over the Caravan Trade

•

201

Indian traders on Zanzibar were not late to send, as a reaction to the German/British blockade, a petition to Queen Victoria explaining: “There is no foundation for the reports recently circulated a ributing the native outbreak in the coast districts to religious fanaticism and the hostility of the Arab slave traders” (reproduced in Woulfin 2011: 128). As British subjects, one may assume that they were not free from the anti-German feelings, which characterized that society. They managed to hit the nail on the head when they gave their explanation to the riots in the mainland coastal towns: It was “due to the ignorance of the native character displayed by the European employés of the German East African Company, and especially their contemptuous treatment of the coast population and the disrespect shown to the Sultan’s officials and the Sultan’s flag” (as reproduced in Woulfin 2011: 128–29). Hill (1957: 50) is of the same opinion when he writes that when “Karl Peters as director-in-chief of the German company sought to hoist the German flag beside the Sultan’s flag at Pangani, the Arabs and Swahilis, led by Bushiri, revolted.” The reasons behind the Bagamoyo rebellion differ in some important aspects from the Pangani case. Some of the town’s leaders came together to challenge the increasing German presence, writes Walter T. Brown (1971a: 278). Glassman (1995: 199–200) goes one step further and points out the ri that ran between the town’s leaders, the majumbe on one side, and the sultanate and the Germans on the other, which became apparent in the September clashes of 1888. The local people believed, as already mentioned, that the sultan had already sold out the coastal area to the Germans for an enormous profit, and that the administration of the town had been handed over to the Germans without their consent. Steven Fabian, on the other hand, points to an interesting and important difference between the September uprising and the December invasion of Bagamoyo. The first conflict was led by Bagamoyo’s leaders against both the sultanate’s sell-out of the independence of the town and the Germans’ increasingly open a empts to take over the administration of the town. In the la er conflicts it seems like the people of Bagamoyo sided with the Germans against the intruders from Pangani, Windi, and Saadani, who had looted and plundered the town’s shops and deserted homes and who stole people le and right and enslaved them, particularly Nyamwezi porters. Glassman bluntly writes that Bushiri, a er he le Pangani and the support he had there, increasingly came to resemble a warlord who penalized the surrounding communities and passing caravans. This was also captured by a British diplomat with the words: “In his treatment of the Unyamwezi porters Bushiri, from the point of view of his own interests, made a grave mistake” (quoted in S. Fabian 2007b: 138). For a short period of time it seems like the porters, townspeople, and the

202

•

Muted Memories

Germans could join hands against a common enemy. A er the September massacre of townspeople and destruction of major parts of the town by the Germans, it seems ironic that a couple of months later, “without probably even realizing it then, the Germans in Bagamoyo had, from the time of Bushiri’s arrival, not only defended themselves, but actually defended the town of Bagamoyo from the aggression of the outsiders” (S. Fabian 2007b: 147). However, there are disagreements regarding the role of the caravan porters in the September uprising. There is no evidence, Glassman concludes, that the two major forces behind the uprising in Pangani, the up-country porters and the dissatisfied middlemen/patricians, acted in the same way in Bagamoyo. The Nyamwezi had experienced a gradual marginalization from the position they originally held in the caravan trade for years, he argues, from the side of the middlemen-merchants. “All the sources,” writes Glassman, “Swahili, Indian and European, agree that the unrest at Bagamoyo was led by majumbe and local villagers, not the Nyamwezi” (Glassman 1995: 213). It therefore seems safe to assume that the two groups, the middlemenmerchants of Bagamoyo and the arriving porters, reacted independently when they protested against their increasing marginalization in the caravan trade along the central routes. It may even seem probable, according to Glassman, that the Nyamwezi traders/porters may have welcomed the Germans as a substitute for the merchant-middlemen as well as the Arabs and Indian traders or bankers to whom they o en fell prey. Fabian, however, finds that the hypothesis of a growing fracture between the merchants-middlemen and the porters leading to the porters being either neutral or siding with the Germans has to be questioned. The fact that the caravans and porters continued to arrive in Bagamoyo year after year indicates that a symbiosis between traders and porters established when the Nyamwezi traders/porters extended their caravan routes to the coast continued to exist. This became apparent to the Germans when, a er the formal colonization of the area, they failed to wrench the control over the trade out of the hands of Bagamoyo’s traders. The fact that only a few hundred porters, many rescued from Bushiri and his men, sided with the Germans while a couple of thousand took refuge at the Catholic mission station is seen by Steven Fabian (2007b: 139–40) as evidence for his conclusion that the porters in Bagamoyo were not a major force in the rebellion. Hence, the rebellion did not develop in a uniform pa ern in the coastal towns, since the underlying frictions and fractures ran along different fault lines. One decisive factor was the sultanate’s differing involvement in the caravan trade. In Bagamoyo, by far the most important terminus and entrepôt, the sultanate sought and gained stricter control over the trade

Conflicts and Clashes in the Competition over the Caravan Trade

•

203

than it did in Saadani or Pangani. Another was the Germans’ determination to stay and take control over the lucrative caravan trade in Bagamoyo and to use the military resources available to do so. A third was Bushiri’s tactical mistake to alienate the Nyamwezi porters from the rebellion. The rebellion cracked along many different fault lines. While the Germans were expelled from Pangani, they managed to stay in Bagamoyo and destroy much of the town. A major reason for the limited success of the socalled Shirazi middlemen-merchant rebels in Bagamoyo was that the new elite, both Arabs and Indians and the caravan porters from up-country societies, withheld their support to the rebels, or even supported the Germans, as in the case of many Indian and Arab traders (Glassman 1995: 6).

Notes  1. Glassman writes about the porters who came to Pangani 115 km to the north of Bagamoyo, but his statement could be equally well applied to Bagamoyo.  2. The sultans who ruled during the epoch of the caravan trade were Seyyed Said bin Sultan (1828–56) who moved his capital from Oman to Zanzibar, Majid bin Sultan (1856–70) who invited traders and bankers from the South Asian communities to se le on the East African coast, Barghash bin Said (1870–88) who tried to move the trade from Bagamoyo and the capital from Zanzibar to Dar es Salaam, and Khalifa bin Said (1890–93) who signed the agreement with the Deutsch-Ostafrikanishe Gesellscha (DOAG).  3. Glassman (1995: 64) reproduces the song but with the term waungwana for the coastal people. The term waungwana underwent, as will be shown in chapter 10, many transformations of its meaning during the second half of the nineteenth century, both over time as more people se led in Bagamoyo and aspired to inclusion and to belong to the waungwana, but also when this identity label travelled along the caravan routes to the areas around Lake Tanganyika. In the song cited above it refers to the trading middlemen-merchants in, for example, Bagamoyo.  4. Some also tried to get hold of a piece of land to cultivate their own food crops.  5. One of the sultan’s means to take the control over the central caravan routes out of the hands of the town’s middlemen-merchants was to appoint a governor, Liwali, in Bagamoyo and his chief commander, Jamadar, who had a garrison with Baluchi mercenaries in Kaole 6 km south of the town, and a prison in Bagamoyo to his support.  6. This policy was also applied in the sultanate on Zanzibar where the Indian firm Wat Bani bought the right to collect taxes in exchange for an annual payment of $70,000, and Tania Topan, an Indian, was the Sultan’s Customs Master.  7. The Zaramo people controlled the area where Kaole, Mlingotini, and Bagamoyo were built.

204

•

Muted Memories

 8. It should be noticed that many slaves were kept on the plantations along the coast and were never exported to Zanzibar. Milcah Achola (2002), comparing Bagamoyo and Mombasa, finds that the Bagamoyo type of slavery was different from the Mombasa type. During the last decades of the nineteenth century, the area around Bagamoyo seems to have developed the “American” type of slavery, with its massive use of slave labor on the plantation, and considerable social distance between the slaves and the slave owners. The Mombasa type seems to have followed the more ancient East African type of slavery with “domestic slaves.” The “American type” also developed on the plantations at Pangani, approximately 115 km north of Bagamoyo (Glassman 1995).  9. When referring to the now ruling segments of the society as “strangers” they deployed the old classification pa ern in the coastal society, which separated the rightful owners, the wenyeji, from the strangers, the wageni, and they thereby identified themselves as the rightful “owners” and leaders of Pangani. 10. We here recognize the same form of protests as the poor townspeople deployed: to strive not for a change of the system, but to be included in it. The petition from the leading majumbe of Pangani breathes a similar aspiration— not only the “strangers” but also “we,” the rightful owners of the town, should have been informed. 11. In the gerontocratic society of the Swahili world the young people were o en dependent on and subjugated to the elders. 12. B. Brown and W. T. Brown (1976: 196) note that “social unrest was rooted in the very nature of these communities. The flood of immigrant se lers and transient porters . . . created explosive pressures that could not be contained by political deals between the new and the old elite.” 13. Glassman (1995: 180) describes him as “an apparent sadist and patent fraud, who in the following century would be elevated to the status of a minor god in the pantheon of the Thousand Year Reich.” 14. The rebellion that broke out in Pangani will be explored at some length since it provides a revealing illustration of the conflicts that developed in the coastal towns. 15. For a few months in late 1888 and early 1889, a er being expelled from Pangani, von Zelewski was stationed at the DOAG office in Bagamoyo before he was sent back to Germany. During this time he allowed “the Nyamwezi under his care free license to plunder some of the homes and shops belonging to Indian merchants. The Germans were also accused of pulling down stone houses and recycling the building materials and selling the elaborately carved wooden doors and window shu ers” (S. Fabian, 2007b: 143n123). 16. Syncretistic in the sense that the ritual has borrowed many elements from the surrounding cultures and societies, like the role of the myvale, a female ritual expert and the ritual of cleansing the area from evil spirits through the herding of a bull around the area (see Glassman 1995: 171). 17. Glassman (1995: 171) refers to the Swahili term Kuoga mwaka, “to wash the new year.” 18. It could also be that he was in the town’s Friday Mosque.

Conflicts and Clashes in the Competition over the Caravan Trade

•

205

19. Glassman speculates that von Zelewski probably believed that the demonstration of his power would frighten the population (1995: 5). 20. The DOAG takeover of the port town Dar es Salaam followed a similar process. The company’s representative, August Leue, insisted on taking over the house occupied by the sultan’s local representative, his liwali who, like the liwali in Pangani and Bagamoyo refused to follow the order from the Germans and evacuate their building (Brennan and Burton 2007: 21; S. Fabian 2007b: 107). 21. The liwali suggested that each party should have its own copy of the Sultan’s flag, which the German’s rejected. 22. In an Internet text from the Germany-based “Bagamoyo Friendship Society” it is reported that the company built its own house, called the Caravan Serai, outside the town; “In its centre it has a double-storey building for keeping ivory and other precious cargo and rooms for the porters along the inside wall” (www.bagamoyo.com/famous_sights.html). It is also mentioned on the same page that the Caravan Serai was built on the place that, according to a map from 1874/75 kept at the museum at the Catholic mission was a “camp for slaves.” This may explain why the Caravan Serai has mistakenly been seen as a building where slaves were kept and sold, a mistake which is further discussed in chapter 3. 23. A gunboat was once again used against Bagamoyo when the British, on 23 August 1914 during World War I, sent HMS Pegasus to bombard the town. 24. Another source, Ndunguru, Kadelya, and Henschel (2003: 5) reports that the uprising broke out on 27 September 1888. 25. Sometimes spelled (A) Bushiri, Abushiri, Abu Bushiri, or al-Bushiri. 26. The cooperation between Bushiri and Bwana Heri was fragile “and the terms of their defeat different. The Germans executed Abushiri and confirmed Bwana Heri in office” (Wright 1976: 568). 27. He was raised to the nobility first in 1890 with the new name Hermann von Wissman. 28. Two of the eight soldiers from Germany, Max Schelle and Heinrich Hoell, who were killed during the a ack, are buried at the German cemetery in Bagamoyo. 29. It could also be that they believed that only Arabs in their territory were capable of organizing political movements (see Glassman 1995: 207). 30. It has been suggested (Woulfin 2011: 132) that Britain’s support to the German blockade was motivated by their rivalry with France in the Indian Ocean. 31. Many disbelieved Germany’s explanation of the rebellions. One of them was Britain’s Prime Minister Salisbury who in a le er to Britain’s ambassador to Germany pointed out that the rebellion was “due to the errors which have been commi ed by the German Company themselves” (reproduced in Woulfin 2011: 125).

8  Bagamoyo and the Caravan Trade The Entrance to the Heart of Africa There are still many around who can remember how, during their youth, they were fascinated by the travelogues wri en by European and American “explorers” and colonizers about the penetration of Africa.1 Many, such as John Hanning Speke, J. A. Grant, L. V. Cameron, Charles New, John Morton Stanley, and Joseph Thomson (Koponen 1988: 26–30), used Bagamoyo, for good and for bad reasons, as their entry-point into the continent. When Stanley, three years a er his first visit to Bagamoyo, returned to the town in 1874, he arrived with porters recruited on Zanzibar. He paid these porters a ten-day ration sum or posho while he searched for more porters, but: within three hours Bagamoyo was in ferment. “The white man has brought all the robbers, ruffians, and murderers from Zanzibar to take possession of the town,” was the rumor that ran widely through all the streets, lanes, courts and bazaars. Men with bloody faces, wild, bloodshot eyes, bedraggled, rumpled and torn dresses, reeled up to our orderly and nearly silent quarters clamoring for rifles and ammunition. Arabs with drawn swords, and sinewy Baluchis with matchlocks and tinder ready to be ignited, came up threatening, and following them. (Stanley 1888: 57)

From Bagamoyo they followed the well-trodden caravan routes established by traders and porters from the interior of the continent and from the coast. One of the first of these caravans, The East African Expedition to

Bagamoyo and the Caravan Trade

•

207

the central African lakes, organized by Richard Burton and John Hanning Speke in 1857, relied on the already well-established system of transport and security for survival. “A spatial infrastructure of route networks, market towns, caravan stops, and provisioning systems provided a material foundation” (Rockel 2014: 172). Bagamoyo was not only the terminus for the long caravan marches on the central routes from the interior of the continent but the town and the routes also became the entry point and avenues for the spread of Swahili, Islam, and Christianity, but also a number of diseases, into the heart of Africa. Different interpretations have been given of the meaning of the name Bagamoyo, already mentioned in chapter 3, as to lay down the heart in despair or to lay down the cargo and enjoy. A third interpretation of the name Bagamoyo was given already at the end of the 1860s by a Catholic priest, who suggested that the name refers to Bagamoyo as a place which leads to the heart of Africa. Jèrôme Becker and George Revoil, as reported by Walter T. Brown (1971a: 4), support this interpretation of “Coeur de l’Afrique,” which derives from baga or bana, meaning the interior, and moyo, the heart. A similar explanation is provided by Florentine Mallya (2011: 23n2): “jusque dans le Coeur” or the way into the heart of the continent.2 Bagamoyo became the entrance for many of the nineteenth-century European travelers and traders who traveled from the town to up-country trading centers such as Tabora, Mwanza, and Ujiji. In their published texts they could report that Swahili was the language of trade. An example is John Hanning Speke, who wrote that when he reached Karagwe in 1860, a kingdom on the western side of Lake Victoria, he “was welcomed by Mukame [King] Rumanika . . . who spoke to Speke in Swahili” (Clerke 1960: 74, quoted in Malik 1999: 155). “Bagamoyo as the entrepôt for Islam, Christianity and Kiswahili” was a theme suggested during the 2002 International Conference in Bagamoyo during the search for the town’s “outstanding universal value.” It is, however, absent in the presentation of the town when it was placed on Tanzania’s Tentative List. The focus in this chapter is on the role of the town and the caravan trade in the spread of Islam and Christianity to “the heart of Africa,” and, second, on the spread of Kiswahili from Bagamoyo along the caravan routes to the interior of the continent. These processes, as will be shown, are interlinked.

Bagamoyo, the Entrance for Islam and Christianity into the Heart of Africa3 Bagamoyo became the entrepôt into the heart of Africa for what has been called two of the main world religions: Islam and Christianity. In addition, the spread of the two religions also meant the spread of Kiswahili—and

208

•

Muted Memories

vice versa. Kiswahili-speaking traders from the coast brought both their language and Islam with them to the interior of the continent and Christian missionaries used Kiswahili as the medium of communication when they established churches and congregations in the heart of Africa.4 In precolonial and early colonial times, Islam spread in different waves over East Africa. The first wave consisted of immigrants from the Arabic Peninsula, probably as early as the seventh century, who se led along the coast.5 Mwenda Mukuthuria (2009) suggests that the early se lements were a result of the first Hijira, the sca ering and persecution of the first Muslims caused by the non-Muslims on the peninsula (2009: 39). In the Bagamoyo area, it seems likely that Islam was established four to five hundred years later, since the first mosque in this part of East Africa was built in Kaole at the end of the thirteenth century. The first two mosques in Kaole are now in ruins (see Illustration 8.1). The oldest mosque in Bagamoyo still in use is the Friday mosque, probably built during the eighteenth century. The second wave began with the development of the caravan trade and accelerated with the colonization of present-day Tanzania. It has been estimated that the number of people from Arabia in this part of the coast rose from between a few hundred to a thousand at the beginning of the nineteenth century to seven or eight thousand at the end of the century (Koponen 1988: 61–62). Many came to port towns like Bagamoyo, which also a racted hundreds of Indian traders and financiers, many of whom were Muslims. The wave of immigrants from the Asian continent contrib-

Illustration 8.1. One of the mosques in ruins at Kaole. Photo by the author.

Bagamoyo and the Caravan Trade

•

209

uted to making Bagamoyo, for reasons discussed below, a prime engine in the spread of Islam from the coastal area to up-country societies. The wave in focus here, which may be called the third wave, began with the establishment of a branch of the Quadiriyya Brotherhood, a Sufi order within Islam, which was brought to Bagamoyo in 1905 (Nimtz 1980: 59). This order spread rapidly along the caravan routes during the last years of this institution for reasons that will be further explored below.

Bagamoyo—The Entrepôt for Islam into the Heart of Africa Bagamoyo as a town is younger than most other coastal towns, but has still been recognized as a leading center of Islam in East Africa (Nimtz 1980: 95). One reason was the town’s position in the caravan trade, which attracted Muslim traders from India and Arabia. They gradually penetrated the interior through the central caravan routes from Bagamoyo/Saadani to Lake Tanganyika and Lake Victoria and beyond into the Congo area, and they established strategic se lements to facilitate the caravan trade. On the central routes these were Morogoro, Mpwapwa, Kondoa, Kiomboi, Kazeh (Tabora), Ujiji, and, further north, Mwanza at Lake Victoria. Occasionally they also established madarasa schools to teach the population to read and write and the fundamentals of Islam (Mukuthuria 2009: 40). Islam spread slowly during the second half of the nineteenth century along the central caravan route.6 The caravan trade and the routes became the entrance for Islam for a variety of reasons. As mentioned above, one was the establishment of “Arabic” se lements in places like Tabora and Ujiji. W. T. Brown (1971: 619), has argued that Muslim traders from the coast colonized parts of western Tanzania “following a broad trunk route that stretched out from the Tabora area, . . . to Ujiji”, with the result that “Ujiji experienced such intensive colonization that both rural onlookers and local residents regarded it a Swahili town” (B. Brown 1971: 617).7 “Swahili town” is understood here as a Muslim society with a Friday Mosque. Another force behind the spread was the rhythm of the caravan trade. The porters stayed up to three to four months at the coast, waiting for the short rains to pass, and to sell their goods at the best price before they started the long march back to their homeland societies. Some stayed longer, o en cultivating a piece of land or trying to sell their labor as farm workers. During this time they were exposed to the “civilized,” uungwana, urban life in Bagamoyo, which included Islam. Some porters had become Muslims by the time they started their up-country march.8 A third force was the custom among some, certainly not all, slaveowning Muslims to encourage their slaves to adopt Islam. For example,

210

•

Muted Memories

the first to become Muslims in Ujiji on the eastern shore of Lake Tanganyika were Manyema9 slaves captured in eastern Congo. The first Arabs probably reached the area during the 1860s and in 1882 parts of eastern Congo came under direct “Arabic” occupation in their search for good slave trading and hunting grounds (Nimtz 1980: 9). Still a reason during the colonial era was that Islam provided an alternative worldview and social organization, which differed from the colonizers’ culture. The adoption of Islam among the colonized communicated a difference and distance from the colonizers and provided a partially alternative identity (Pawliková-Vilhanová 2010: 163), compared to that of being a “native” African, the lowest echelon on the racial scale. The role of the caravan trade and its routes for the spread of Islam is not very well researched (Pallaver 2012: 7).10 However, it seems that the Muslim influence during the heyday of the caravan trade primarily reached those who were directly involved in the caravan trade. The spread was probably limited, or absent, beyond these groups and the caravan routes. There is, however, a need to be cautious here, since the spread of the Brotherhood (for further details, see below) was unnoticed by many external observers. When Father Mènard, one of the White Fathers in Tabora, reported that “the Arabs in Tabora are totally absorbed in trade and they do not care at all about religion . . . ; fanaticism is not among their defects,” (quoted in Pallaver 2012: 7) he obviously had his focus on the Arabs and was probably ignorant of what happened among the underprivileged and marginalized African sections of the society which were usually a racted by the Brotherhood.11 One reason for the limited spread of Islam along the central routes during the early phase of the caravan trade could be that Islam was a highly protected part of the uungwana qualities or “civilized life” of the Swahili traders. J. Spencer Trimingham (1964: 58) even concludes that “the coastal Muslims have insulated their religious culture from the Africans.” The Arabs would not, with their worldview, which sharply divided between the “civilized persons” and the “barbarians,” be expected to “show any desire to raise the Africans to the status of the ‘civilized’ by converting them in large numbers” (Unomah and Webster 1976: 304f.). The coastal Muslim middlemen, in their position between the Indian Ocean trading networks and the African continental networks, depended on their abilities to translate values of the goods traded in both directions. An “Islamization” of the interior would have created a stiffened competition over the goods from India and China so highly valued by the Swahili merchants themselves (se also Horton and Middleton 2000: 89–90). The value added in the trade could most easily be absorbed by the Swahili middlemen if the barrier between “civilized” (Muslims) and “barbarian”

Bagamoyo and the Caravan Trade  • 

211

(pagans) was maintained, which restricted their interest in spreading Islam and other parts of their way of life to people in the interior of the continent. A further reason for the slow spread of Islam along the central routes may be found in the logic of trade. “There was a basic contradiction,” writes Viera Pawliková-Vilhanová (2010: 139), “between converting Africans and selling them as slaves. Islam forbids Muslims to enslave coreligionists; to convert too many Africans to Islam would have diminished the number of those they were permitted to enslave.” The call to contribute to the universal Muslim community, the Umma, may here have conflicted with the personal strife for wealth, prestige, and power.

The Role of the Brotherhood The spread of Islam was slow during the first decades of the caravan trade but gained in momentum when a new wave of Hadrami immigrants arrived with the growth of the caravan trade and the consolidation of the Zanzibari Sultanate as a mercantile empire in the western Indian Ocean region. Many of the immigrating Hadrami were religious scholars who began a period of active proselytization in the societies in the immediate hinterland of Bagamoyo where some of them settled as petty traders (Glassman 1995: 136). Islam was for decades confined to a narrow coastal belt on mainland Africa and the islands in the Indian Ocean. The urge to spread Islam to the infidels seems to have been low. It was only during the 1880s with colonization, first by the Zanzibar Sultanate and later by the Germans, that the spread of Islam to the interior gained momentum. Two separate Brotherhoods, the Shadhili and the Quadiriyya (or Quadiria), became active in the spread of Islam on mainland East Africa. Instrumental in the spread of the latter Brotherhood was Sultan Barghash bin Saidi’s invitation in the early 1880s of Shaykh Uways bin Muhammad al-Barawi, from Brava12 in southern Somalia, to Zanzibar.13 The Ibadiyya or Ibadi canon, to which Sultan Barghash belonged, with roots from the Khariyya secessions in 685, is characterized by its tolerance towards other canons (Bang 2008: 358n9). The Brotherhood has been characterized as an Islam revival movement, which became increasingly popular along the East African coast. The spread from Zanzibar to the mainland coast was helped by Shaykh Uway’s personal visits to port towns like Bagamoyo and Pangani on the Mrima coast and Lindi and Mikindani further south. The Brotherhood then spread along the caravan routes from Bagamoyo, via Tabora and Ujiji, to the Manyema area in eastern Congo. Shaykh Uways later became the most famous Quadiriyya leader in East Africa (Nimtz 1980: 72). A num​

212 • Muted Memories ber of learned persons, ulama, on Zanzibar also became members of the Quadiriyya Brotherhood (Martin 1971). The Brotherhood, in contrast to the ruling families on Zanzibar who belonged to the Ibadi canon of Islam, who in particular showed no interest in da’wa, or calling people to Islam, and the Shafi’ite Arabs from Hadramawt or the Shi’ite groups from the Middle East and Persia, who never or rarely propagated Islam among the original inhabitants of mainland East Africa (Trimingham 1964: 55), worked actively to assimilate new members into the organization and the Muslim faith. The Muslim Brotherhood or tariqa, meaning “the path,” builds on the Sunni mystical tradition known as Sufism, which has a long tradition within Islam. At the end of the twelfth century, “the followers of Sayyid Abd al-Quadir al-Jilani decided to perpetuate his ideas and memory; they organized the first tariqa, the Qadiriyya” (Nimtz 1980: 56). Different varieties of the Brotherhood have been present in East Africa at least since the end of the nineteenth century. The largest of them is the Qadiriyya, which is assumed to organize around 70 percent of the members of the various Brotherhoods. The Qadiriyya is organized into three independent branches, of which one became associated with Bagamoyo (see Trimingham 1964: 97). The Brotherhood spread from the ports on the coast along the caravan routes before and during the German colonization. It is possible to see a pattern in its spread. Most, if not all, of the major trading centers along the central caravan routes had contained a Muslim population segment since the second half of the nineteenth century. At the end of the century Islam began to spread, through traders and caravan porters who belonged to the newly established Brotherhood, first to these trading centers along the caravan routes, and then to the surrounding villages. In the work to spread the Brotherhood the members searched for aspirants, or murids.14 The members, both leaders and disciples, came together at least once a week for the dhikr (sometimes also spelled dhikiri; zikiri in Kiswahili), which is “the ‘remembrance’ of God by the repetition of His name and attributes; coordinated, when recited in congregation, with breathing techniques and physical movements” (Trimingham 1964: 96). Mark Horton and John Middleton characterize the dhikri as “self-induced trance” (2000: 189). During the dhikr rituals, prolonged hyperventilation was practiced to put the attendants in trance. Bodily movements were also used to reach this mental state (Franken 1991), which has been seen as a way to decrease the distance between God and his followers (Glassman 1995: 140). Still a feature in some of the Brotherhood organizations was the belief that the blessing of Islam’s saints can be obtained by visiting their tombs,

Bagamoyo and the Caravan Trade

•

213

especially on the saints’ birthdays. It was believed that Sufi leaders could a ain a state of sanctity and that they could become saints at death and “their graves became important centers for Sufi rituals” (Glassman 1995: 138). The ancient tombs at Kaole are still visited and gi s placed in or on them in hope for wealth, fertility, or prosperity. These are controversial issues for members of more orthodox canons within Islam. The beliefs were also controversial for Shaykh Uways, who became involved in a heated polemic with the leader, sometimes called the Mad Mullah, of another Brotherhood on these issues, which may have led to the assassination of Shaykh Uways in 1909 (Martin 1969: 472). From early on, Sufism was seen in East Africa as the Islam for the poor and marginalized. August Nimtz (1980: passim) points out that one reason for this was that the tariqa, rather than the “official” Islam reached the masses. He gives two reasons for this: one is that Sufism could more easily incorporate local customs than the more orthodox versions of Islam. In this sense the tariqa could be seen as a syncretistic religion, which could combine the tombs with a belief in a benevolent power of the spirits of deceased persons. Islam has been characterized as a rigid system of faith with an “unyielding nature of the Islamic institutions precluding” (Trimingham 1964: 68) synthesis between local religious beliefs and Islam. It may be assumed that Trimingham was ignorant about the role and character of the tariqa in societies beyond the more obvious influence of the Arabic segment of the population on the coast. Another example of the syncretistic character is that music, songs and dances, which are haram or “forbidden” in more orthodox versions of Islam, are frequently used by the Sufis. Another reason for the rapid spread of the Brotherhood was its Sufism tradition, which has been characterized as more egalitarian, in contrast to some of the more orthodox canons within Islam since status in the tariqa was not based on learning, which, in turn, was related to wealth. The highly prestigious schooling in Islam was o en only affordable for the wealthy elite in towns like Bagamoyo. As B. G. Martin (1971: 530), notes: “The best qualification for becoming a learned man was to be the son of another learned man,” which made piety the principle avenue to achieve religious credit for the poor and for those without the right kin relations or, in the words of Jonathon Glassman (1995: 140), “To Muslims with li le formal learning, zikiri seemed to offer an avenue for spiritual fulfillment that did not depend on refinements of literary accomplishment.” It opened an avenue to religious experience that was denied persons who lacked the formal religious training in the more orthodox canons of Islam. Sometimes the orthodox Sunni leaders had difficulties accepting the more popular Brotherhood version of Islam, particularly since it included

214

•

Muted Memories

Illustration 8.2. Sacrifices placed at an ancient tomb in Kaole. Photo by the author.

Bagamoyo and the Caravan Trade

•

215

musical instruments for songs and dances (Trimingham 1964: 101). It was even under consideration to put a ban—fatwa—on the dhikr performed by the Quadiriyya brotherhood as it was seen to be too intoxicating to be halal (permissible). At the same time we may here see an explanation for the rapid spread of the tariqa in the coastal societies with their long tradition of festivities and dancing and also in the societies along the caravan routes.15 Researchers disagree on the African impact on Islam in Africa. There is, as already mentioned, a long tradition of viewing the peoples and cultures of East Africa as only recipients of external influences from the Indian Ocean region. It has, for example, been said that “Islam in East Africa . . . bears many of the characteristics of a foreign religion” (Trimingham 1964: 54), or that “African culture was the passive element, and Islam brought the vital cohesive element” (Trimingham 1964: 66). Others, however, have pointed out that the conversion to Islam in many parts of Africa was a gradual process through which Islam was merged with pre-Islamic religions (Pawliková-Vilhanová 2010: 135). In the absence of formal instruction by Muslim masters, the slaves who aspired to a Swahili status adopted “a cultural identification with Islam in its most outward, communal features, such as ritual killings of meat, slaughtering of animals, and holy feasts and fasts” (Wright 1976: 552). It is today generally considered that the tariqa implied an “Africanization” of Islam in its forms but not in the schooling. The syncretistic character of the Brotherhood may also explain the integrative role of the organization in these societies, which may be seen as a parallel to the “dance societies” or vyama (see chapter 11). People of up-country origin, watu wa bara, o en stigmatized in a society that sharply divided between the “civilized,” the waungwana, and the “barbarian,” washenzi, free or slaves, could become members of a tariqa when they converted to Islam. With this they could initiate their own integration into the Swahili world. The tariqa seems to have been particularly a ractive, as already mentioned, to people in marginalized positions, like women, the young, or the impoverished. At least two reasons have been identified for its attractiveness among women. One is that women have been given only minor roles in more orthodox canons of Islam. Another is that the spirit possession cults found among women during the nineteenth century bore many resemblances with the zikiri rituals. Both gave the practitioners “an outlet for religious expressions that was denied them in dominant Muslim organizations” (Glassman 1995: 142). The most famous leader of the tariqa in Bagamoyo is a good illustration of this. Shaykh Ramiya, born under the name Mundu in the Manyema area in eastern Congo, was captured and sold as a child slave at the age

216

•

Muted Memories

of eight to an Arab slave caravan trader operating in the area during the second half of the nineteenth century.16 He was taken to Bagamoyo and became the domestic slave of a man named Amir.17 Ramiya was given the opportunity to engage in trade and could soon invest in coconut plantations. In 1912, he was the first with an African origin in Bagamoyo to own a stone house. A er ten years of religious studies, he was included in the class of the learned, ulama, under the name Mu’allim Ramiya. Shaykh Ramiya, who opened a madrassa, was regarded as the most prominent religious leader at the end of the nineteenth century and early twentieth century in Bagamoyo (Cooper 1981) and contributed to making Bagamoyo a leading center for Islam in East Africa (Nimtz 1980: 121) as children of wealthy Arab and Indian families in the interior were sent to the madrassa in Bagamoyo. The German colonialists, probably involuntarily, also contributed to the spread of Islam in what today is Tanzania.18 One way this occurred was von Wissman’s employment of six hundred Sudanese Muslim soldiers during the 1888–89 conflict with Bushiri and his rebels. This armed conflict centered on Bagamoyo as the headquarters for both Bushiri during his short control of the town and then as the headquarters for the German East Africa Company and, eventually, as the first German capital of colonial German East Africa. But the Germans’ contribution to the spread of Islam was also a result of their racial ideology, which identified the coastal peoples as more developed or advanced than the up-country peoples. “All the auxiliaries of the European penetration, pacification and government” writes Trimingham (1964: 57), “guides, interpreters, soldiers and servants—were Muslims. Each place where a European installed himself, military camp, government center, or plantation, was a center for Muslim influence.” Staff with a coastal origin were employed in low-level posts in the colonial administration and sent to the interior of the continent, and with them also Islam (Pawliková-Vilhanová 2010: 162). A third contribution of the Germans’ to the spread of Islam from Bagamoyo, their first colonial capital, was their adoption of Kiswahili as the language of administration. In all, these measures “contributed indirectly to the estimation of many Africans of Islam as the most prestigious and influential religion in the territory” (Nimtz 1980: 12). The Germans therefore, probably unintentionally, can be accredited both to the spread of Islam in their territory and to the fact that two-thirds of East Africa’s Muslims today live in Tanzania, their former colony (Nimtz 1980: 12). However, the Germans developed rather ambivalent relations with Islam and the Muslim part of the population.19 It may be hypothesized that their mixed feelings were a result of their worldview, which made them

Bagamoyo and the Caravan Trade

•

217

perceive the world in racial and ethnic categories. This lead the Germans, and other Europeans as well, to see the Arabs as superior to the Africans20 while, at the same time, the Arabs were Muslims while the Africans constituted uncultivated soil upon which the seeds of Christianity and “European civilization” could be sown.

Bagamoyo: The Entrance of Christianity Missionaries from the Holy Ghost Order, a Catholic order,21 established their first mission stations in East Africa during the middle of the nineteenth century with the mission to “fight slavery at the grass-roots level, to ransom as many slaves as possible, to train them in schools and agricultural se lements and to lead them to Christianity” (Mallya 2011: 2). Members of the order therefore conducted a survey along the East African coast in 1866 in order to find a suitable place on mainland Africa for a mission station with enough land and suitable soils for a large scale agricultural scheme. The choice was Bagamoyo because of the area’s fertile soils, the good harbor, the town’s dynamic development, and its closeness to Zanzibar (W. T. Brown 1970, 1971a; B. Brown and W. T. Brown 1976: 186). The Holy Ghost Fathers had been allocated a large piece of land to the north of the town by the majumbe, the leaders of Bagamoyo, a er pressure from both the Sultan of Zanzibar and the French Consul General for Zanzibar. One reason for the establishment of the mission was to ease the pressure on the already existing mission station in Zanzibar Stone Town. The Zanzibar mission received slaves freed by the British. The successful British chasing of Arab sailing slave vessels produced an ever-increasing flow of released slaves and the limits for what the mission station and its workshops in Zanzibar could absorb was soon reached. A new mission station on mainland Africa was seen as a valve reducing the pressure on the Zanzibar station (Kollman 2005: passim). Another practice through which slaves arrived at the mission stations was the owners’ strategy to hand over old and disabled slaves to the missionaries in order to cut their own costs (Wright 1976, 565). The missionaries looked for a place big enough to establish a colonie agriculture, a cultivation scheme where the freed slaves, particularly the children, could be put to work. This scheme had a double purpose. One was to function as a disciplinary mechanism. To let the children work for five hours in the fields and then five hours in the school was a means to both discipline and civilize them. The second aim was to let the children’s toiling on the fields assist the mission’s economy. The missionaries saw that Bagamoyo was very suitable to “cultivate the land and the hearts and the minds of the children” (Kollman 2005: 143).

218

•

Muted Memories

The second reason was to find a place for a mission station on mainland Africa, which could be the springboard for mission work in the interior of the continent (Kollman 2005: xvii). The establishment of new churches in the interior was essential to the missionaries for the establishment of Christian congregations, since they believed in a strict interpretation of extra ecclesiam, nulla salus or “outside the church, no salvation” (Kollman 2005: 63). The Spiritans, as they also are called, saw li le hope in mission work among Bagamoyo’s Muslim population. Adolescents and grownups were already influenced by Islam. Freed slave children22 and up-country people promised more fertile grounds for Christian missionaries. This strategy was explained in the late nineteenth century by the Father Superior of the mission station, who commented on the station’s relations with Bagamoyo townspeople: “Oh, our Mission is not for them at all, it is for the interior” (cited in W. T. Brown 1971a: 239). Even active colonizers saw the value in the mission’s activities in the interior of the continent. Henry Morton Stanley did not hide his expectations from the mission’s expansion into the interior: “It is evident that, if supported constantly by his friends in France, the Superior will extend his work still further into the interior, and it is, therefore, safe to predict that the road to Ujiji [at Lake Tanganyika] will in time possess a chain of mission stations affording the future European trader and traveler safe retreats with the convenience of civilized life” (Stanley 1888: 1:62f.). This dream never materialized for the Holy Ghost Order as the Bagamoyo mission station was extended to only five other places: Mhonda, Mandera, Morogoro, Tununguo, and Ilonga.23 However, Stanley’s dreams and the mission’s activities show that Bagamoyo was the entrance for both colonizers and missionaries in their efforts to reach the interior of the continent and that the central caravan route was, if not the highway, at least the gateway into the heart of Africa, as Protestant, Catholic and other churches were established in the interior of the continent. One example is the Church Missionary Society (CMS) and its strategically located mission station along the central caravan routes to Uganda (see below). The Spiritans’ main connection to the central caravan routes came through the slave trade. This engagement took the form of three main activities: to receive and/or buy slave children to “free” them, to convert the “freed” children to Christianity, and to establish new mission stations in the interior of the continent, as freed and baptized slaves were sent to up-country societies to find new proselytes. It should be noticed that most, if not all Christian mission orders operating in Africa as, for example, Methodists, Quakers, Lutherans, Presbyterians, Anglicans, and others. did the same—they worked with/through freed slaves and o en also bought children. The period from the purchase to the freeing of the

Bagamoyo and the Caravan Trade

•

219

children, however, could be prolonged since the children first had to be “educated by means of useful labour” (J. Fabian 2000: 30). This “made it possible for the staunchest opponents of slavery to profit from the trade” (J. Fabian 2000: 30). The Holy Ghost Fathers’ struggle to baptize slaves, in combination with their lukewarm interest in the life of the Muslims in Bagamoyo, and their search for the more fertile mission fields in the interior of the continent contributed, according to Paul Kollman (2005: xvi–xix), to the failure of their endeavor. He cites Vincent Donovan (1978), an American Holy Ghost Father, who characterized the early missionaries’ strategy with the following words: It is easy to understand the feelings of the missionaries who arrived on the scene in the last [i.e. nineteenth] century, their concern with doing something about the system of slavery, which was the cause of all the horrors. They did the only thing they could in the circumstances. They bought the slaves. They bought them le and right, with all the money they could get their hands on. They bought them in the hundreds and the thousands—and they Christianized all they bought. (Kollman 2005: xvi)

The Hamerton Treaty of 1847, signed between Sultan Said of Zanzibar and the British Government, gave the British navy the right to patrol Omani and Zanzibari ships sailing between the East Coast of Africa and Asia. The treaty also gave the British the right to liberate slaves found on the inspected ships. This became another important source of freed slaves who were handed over to the Fathers at the mission station in Bagamoyo. The first children to arrive in Bagamoyo were fi y boys sent from the Zanzibar mission station on 10 December 1868 (W. T. Brown 1971a, 211). They were followed the year a er by forty-six girls, who came under the supervision of the mission station’s Sisters (Mallya 2011: 3). They were placed under a similar work discipline as the boys. The girls were crucial for the missionaries’ plans to arrange marriages of Christians so families could be sent for mission work at new stations in up-country societies. The Fathers acted as matchmakers in the establishment of these Christian families. Although formally free, they were put in a tough school. Some of the Holy Ghost Fathers had a background in juvenile correction programs in Alsace and Lorraine in eastern France. In a Foucauldian manner they knew how to use time and space to impose discipline and to create good Christian converts. The mission station in Bagamoyo was planned, and the daily rhythm of the children organized, to instill a morale based on work and discipline (Kollman 2005: passim).24 By the missionaries, it was seen as a jardin d’acclimination, where both plants and freed Africans could be cultivated as targets for the missionaries’ interventions (Kollman 2005: 158–59).

220

•

Muted Memories

Thus, physical work and work schemes were actively used by the Fathers to instill “European values” in the children. The strategy came out very clearly in the views of Father Horner, the founder of the Bagamoyo mission station: For twenty years now I have been interested in the regeneration of the black races, and since this time I have been able to employ only one single effective means to raise them up from the abasement where they have been captured for centuries: work sanctified (and) ennobled through Christianity. For this reason I have founded the agricultural and industrial school at Bagamoyo. . . . I repeat: the regeneration of the blacks will never have any other source than work undertaken and endured within the spirit of Christianity. (Quoted in Koren 1958: 203, 456, and reproduced in W. T. Brown 1971a: 203)

However, work was not only a means to discipline or “form” the baptized African children, but also a source of wealth. The work performed by the mission’s children on the plantations and in the workshops also became a resource in the missionaries’ struggle to become financially independent. The time-scheme established by the missionaries let the children toil five hours per day on the mission’s fields and plantations. The second strategy in the colonie agriculture was to establish Christian villages around the mission station where liberated slaves could live a new and Christian life. These villages, supposed to be “best practice” societies, were filled with freed slaves, women and men, who should establish conjugal units. The first of these villages is known as St. Joseph’s which, when established in 1871, had eighteen families. Five years later the number of families had increased to forty.25 The focus on self-sustenance was not unique for the mission in Bagamoyo. It was also a prominent feature in the British mission stations established in East Africa. Sir Bartle Frere, one of the British architects for the handling of slaves liberated from Arab slave vessels suggested in a memorandum that it was important that the liberated slaves be able to sustain themselves to avoid becoming a burden on the British treasury. The villages at the mission station at Bagamoyo where slaves liberated by the British on the Indian Ocean were places meant to meet these aspirations. In order to guarantee the success of their missionary work in this Muslim environment, the Spiritans introduced strict codes of conduct for the Christians in their model villages. Life in these villages was organized by the missionaries. The tough labor regime was not only justified by the economic contributions the freed children and adults generated in the fields, but also by a conviction among the missionaries and the ruling elite in Europe about the importance of a good husbandry of time. The school developed into an organization to teach children, especially working-class children in Europe, and Africans to take proper care of their time.26 The

Bagamoyo and the Caravan Trade

•

221

clock announced the time for work, eating, prayers, and recreation. The villagers, former slaves now placed in the Christian villages, worked five days a week for the mission station in exchange for food and clothing (Kollman 2005: 152). The aim, according to Brown was to “preserve social peace and to isolate the villagers from traditional African cultures and the urban milieu of Bagamoyo” (W. T. Brown 1971a: 219). Many of the rules were meant to restrict the new Christians’ contact with the surrounding society and the hard labor regime was probably a means to limit their possibilities to mingle with non-Christians. The tough approach adopted by the Spiritans was a result of their view on conversion to Christianity, which they regarded as an individual act and decision. But, since the missionaries, like so many other Europeans, did not see the Africans as adults and therefore not fully responsible for their own actions, the missionaries imposed strict measures to control and discipline the converts. The missionaries did not always believe that the suddenly released slaves could handle their unexpected freedom. As in so many other colonial situations, both colonizers and the missionaries took on the role of being trustees for the “natives” (see Murray Li 2007). Those they placed under their supposedly benevolent trusteeship became their wardens, and as such they needed much help, guidance, and supervision. The missionaries, like the colonizers, acted on the basis of a strong conviction that they knew how the colonized people and their societies should be reformed and reorganized. The freed slaves in the Christian villages around the Bagamoyo mission station rejected this treatment and the hard labor regime. It has been pointed out that the “free” laborers on the mission’s land worked longer days without pay than the slaves on the surrounding plantations (Nimtz 1980: 108). It is therefore no surprise that Christian farm workers voted with their feet. “Thus the cruelest irony of this mission endeavor: the liberated slave became a runaway slave, subject to severe punishment and possible expulsion, but not free to leave the community at will” (W. T. Brown 1971a: 228).27 The third activity, besides taking care of liberated slaves and Christianizing them, was to provide social services to the population in and around Bagamoyo. The education efforts were primarily of an internal character and meant to meet the needs of the mission stations and their pastoral work. The provision of health care and the fight against slavery was, however, mainly externally oriented work. The mission provided medical care for the needy. This service was perhaps not without a “higher” baptizing purpose to save souls. It is reported that four hundred Africans were baptized in 1895, but more than half of

222

•

Muted Memories

them were Nyamwezi porters dying from diseases and distress (W. T. Brown 1971a: 240). An eye-witness reported: “One of our fathers, bustling with activity and initiative . . . goes each day to inspect the surrounding area and the alleyways of Bagamoyo . . . Accompanied by two or three young married men, his intelligent catechists and devotees, he tends to the most sick. They then carry the poor, dying Mnyamwezi back to the mission hospital to baptize him before death” (as quoted in S. Fabian 2007b: 69). The mission station soon found itself in a tense relationship with Bagamoyo’s leaders and the townspeople. The most severe conflict related to the land they claimed that the Sultan of Zanzibar had given to them, but which was actually allocated by Bagamoyo’s leaders, the town’s majumbe. At the core of the conflict lay the question of who could allocate land rights in Bagamoyo. The treaty negotiated between the townspeople and the surrounding Zaramo leaders, resulting in the town paying a rent—kanda la pazi—to the Zaramo leaders (see chapter 9), did not include the issue of who had the right to allocate land. When a newcomer on the scene, the Sultan of Zanzibar, intervened to assist other newcomers, French missionaries, this interrupted the earlier power balance and contributed to the processes that led to the so-called Bushiri rebellion in 1888–89 described in the previous chapter when most of the town was demolished. The missionaries, precisely like the colonial powers Britain and Germany, had mistakenly taken the sultan for the ruler of the area and taken for granted his right to allocate land. It seems probable that this contributed to their alienation from ordinary life in town.28 The church established by the Spiritans became, in the eyes of the coastal dwellers, associated with slaves. The flaws in their strategy may be characterized in the following ways: Christian faith was pressed upon reluctant or even hostile ex-slaves; the Church could not a ract new converts through its own faith or promises of salvation, but had to rely on receiving and/or buying slaves. The ex-slaves at the mission station, although released, were still regarded as slaves by many in the local population since only the owner through his or her free will could manumit a slave, not the colonizers, which made many local people want to distance themselves from the church. Kollman points out that this form of pastoral strategy is sometimes called kitumwa Christianity. The name comes from the Swahili term mtumwa (slave) and the expression carries the meaning of “slave like” or “slavish” Christianity (Kollman 2005: xvii). Christianity was for the slaves, not for the free people in Bagamoyo.29 The Catholic mission station has been in Bagamoyo since 1868. Its religious impact on the people of the town has been minimal. Only 5 percent of Bagamoyo’s population is estimated to be non-Muslim.30 Many of those

Bagamoyo and the Caravan Trade

•

223

employed by local or national authorities came from Christian up-country societies. Still, the mission station has contributed to making Bagamoyo and the caravan routes into an entrance for Christianity into the heart of Africa. Other missionary societies followed the caravan routes in their penetration of the continent. One of them, Church Missionary Society (CMS) established a mission station at Mpwapwa in 1878, the most important trade center and caravan camp site between Tabora in the interior and Bagamoyo, Saadani, and Winde at the coast. German sources as H. von Kanneberg (1900: 7) estimated that 100,000 porters passed Mpwapwa annually at the end of the nineteenth century. Caravans from the interior or the coast could arrive to Mpwapwa on different routes, but almost all passed this point in central Tanzania. As other missionary societies at this time in East Africa it focused on the conversion of ex-slaves and the rescue of slaves.31 The history of the Mpwapwa CMS mission station replicates that of the Catholic mission in Bagamoyo. The place was chosen for its good soil, which provided the ability to grow plentiful and cheap food for the porters who passed Mpwapwa’s trade and camp site, and the land could be used to support both the missionaries and the freed slaves. It also functioned as a place where missionaries traveling by foot to Uganda could rest and replenish their food supplies before they continued their long march to the Great Lakes area (Holway 1971–72: 209). As in Bagamoyo, the local political leader granted the missionaries the land they needed for their ex-slaves to support themselves and the mission. Over time it developed into “a substantial Christian community in Mpwapwa and surrounding areas, a community of extremely diverse origins but increasingly unified by religion [Christianity], a common experience, education and the Swahili language” (Rockel 2006b: 16).

Bagamoyo, the Caravan Trade, and the Spread of Kiswahili One of Bagamoyo’s “outstanding values” identified during the 2002 Conference on Bagamoyo as a potential World Heritage site concerned the town’s role in the spread of Kiswahili or the Swahili language.32 This language probably developed a millennium ago along the coastal area from southern Somalia to the southern part of Mozambique. Enough ink has been used up in a empts to trace the origin and crucial elements of the Swahili language and culture, so here only a few remarks will be made with focus on the role of the caravan trade and Bagamoyo. The name is derived from the Arabic term suahel, which is generally translated as “coast.” The term, however, has multiple meanings and can also stand for “bound-

224

•

Muted Memories

ary” and “margin” (Middleton 2003: 522n3). The term can therefore be interpreted as denoting people on the coast, people in a geographically marginal position, or people marginal to “Arabic” civilization. It is generally held that this language developed in the polylingual and multiethnic coastal communities where people from different origins met and merged through processes of creolization or hybridization. One reason for the rapid spread of Kiswahili, according to R. Reusch (1953: 22) is that the language “is rich, very elastic and has a comparatively simple grammar in comparison with the complicated Bantu grammars,” or in the words of Charles Pike: “The Swahili language . . . was likely perceived by most inland peoples for precisely what it was—an African language very similar to their own languages, relatively easy to learn and convenient to use” (Pike 1986: 211). Kiswahili is not a uniform language but has borrowed traits from other languages where it is spoken. The Kiswahili spoken on Madagascar, for instance, has been influenced by both the Portuguese and the Malay languages, which is not the case for mainland African versions (Reusch 1953: 27). The people associated with this language are o en called the Waswahili (sing. Mswahili), or simply Swahili. There are, however, uncertainties and disagreements regarding who they were and who they are today, or, as captured by Juhani Koponen: “It may not be unfair to say that the Swahili civilization is still be er known than understood” (1988: 59n31). Furthermore, as already mentioned, there is not a clearly demarcated “land of the Waswahili”—an Uswahili—and where they lived, people with different identities lived as well.33 This is taken as an indication of the diasporic character of the Waswahili and the multicultural character of their communities. The spread of the language took place during at least two different waves. The first was the “Swahilization” of the coastal areas when people who lived around the Swahili towns adopted Kiswahili as their first or second language. A German document (anonymous) from 1911, referred to by Carsten Legère (2002: 2) in his presentation at the 2002 International Conference in Bagamoyo, gave the following information on the ethnic composition and linguistic competence in the Bagamoyo area: 1. Swahili speaking people (Wasuaheli) who were said to be very numerous in Bagamoyo and Saadani with, of course, Swahili as their mother tongue. 2. Zaramo—Swahili is spoken and understood everywhere in Uzaramo. 3. Doe—assumed to be affiliated with Zaramo—“ein Unterstamme der Wazaramo” —Swahili is very much spoken. 4. Nghwele (Kwere) who mainly speaks Swahili.

Bagamoyo and the Caravan Trade

•

225

The Zaramo, the Doe, and the Nghwele controlled the area along the coast when Swahili-speaking se lements like Bagamoyo were established. This high degree of ethnic complexity probably facilitated the development of Kiswahili as a lingua franca along the coast and, as Legère shows, almost all people in and around Bagamoyo spoke the language, which illustrates how futile it is to use this language to identify what is assumed to be an ethnic group. Kiswahili spread in a second wave along the caravan trade routes to the interior of the continent. There can hardly be a place, it was suggested during the 2002 conference in Bagamoyo, on the East African coast more important than the area around Bagamoyo for this development.34 In a similar vein, there can hardly be an activity more important for the spread of this language than the caravan trade. Other places were probably more important for the development of the language, but Bagamoyo developed into an important place for the dissemination of this language to upcountry communities. A number of reasons can be traced behind Bagamoyo’s role in this development. One was that porters from up-country areas who arrived in Bagamoyo with their ivory met a highly “swahilisized” society when it came to the dominant language and culture. Although the ethnic and linguistic mixture must have been extreme and the creolization processes highly vibrant in the coastal communities, still most people also spoke Kiswahili. This meant that the caravan porters must have been inundated with Kiswahili during their stay in Bagamoyo. And, for the sake of the prolonged haggling, sometimes several weeks, which took place on the beach when they tried to sell the goods they had brought from the interior, it seems most plausible that they learned Kiswahili. In turn, these porters used and spread Kiswahili among the many different societies along the caravan routes when they returned to their up-country societies. Another reason for Bagamoyo’s role in the spread of the language was that the town’s merchants were not late to respond to the first up-country caravans that reached the coast. Around 1840, a massive penetration of the interior started from the coast as Kiswahili speaking coastal entrepreneurs organized their own caravans to trade with people in the heart of Africa. Bagamoyo was the most important place both for departure and as a terminal for the caravans. The German colonial administrator August Leue (1900/1901: 31), for example, reports that during the 1898–99 season no less than 1,305 caravans with 41,144 people le Bagamoyo for up-country destinations. The caravan trade contributed to the spread of Kiswahili along the trade corridors from the east to the west. One illustrative example is the spread of Kiswahili, and Islam, to the kingdom of Buganda to the west

226

•

Muted Memories

of Lake Victoria. It was Kiswahili, and Arabic speaking traders from the East African coast, who introduced Arabic and Swahili languages as well as Islam to the royal court, and Islam was announced to be the religion of the state by the king (Pawliková-Vilhanová 2010: 146; Trimingham 1964: 25). It is reported that Kabaka (King) Mutesa, who reigned between 1856 and 1884, was fond of Arabic poetry and that this language was spreading among the elite, but also the art of reading and writing in the Arabic script, which was also used for Kiswahili texts. “The concept of reading,” Pawliková-Vilhanová notices, “—okusoma—became a synonym for the adoption of a new religion, Islam, . . . The converts were called ‘readers’” [kusoma in Swahili means “to read”] (2010: 146). The language, however, as with Islam discussed above, probably rarely penetrated beyond the trade centers, routes, and home communities of the porters before the establishment of the German colonial administration centers (see Malik 1999: passim). One exception was the Kiswahili dialect spoken in northeastern Congo, kiNgwana, or Ki(u)ngwana), “language of the civilized,” which took its name from the caravan traders and slave hunters called waungwana, many of them slaves themselves, who came to the area through the caravan trade (Mukuthuria 2009, 40). It was spread during precolonial times in the northern and eastern parts of the Congo.35 The societies in the Manyema area collapsed, as already mentioned, under the pressure of the invading Ngoni from southern Africa and traders and slave hunters from the East African coast who colonized the area. It seems plausible that kiNgwana developed and spread to become a dominant language in this part of central Africa as a common means of communication from the rubble of what had been a number of independent societies.36 Although languages other than Kiswahili, like Arabic, may have been seen as superior in social life, particularly due to religious connotations, Kiswahili was still the most common language in trade and politics. One reason was that Arab traders who se led on the coast seem to have been quickly “swahilisized.” Even the Sultan of Zanzibar learned Kiswahili for the purpose of communication and commerce. As a consequence, Kiswahili became a status language (Horton and Middleton 2000). The swahilization of the Arabs and their life along the East African coast should not be understood as a loss of their Arabic identity. On the contrary, the division established on cultural prestige between the Wawanga or Shihiri (Arabs born on the Arab peninsula) and Waarabu, Arabs born on the East African Coast, was not only a question of place of birth or origin but also one of prestige. The categorization between Wawanga and Waarabu also came to indicate the speaker’s use of the Arabic or Kiswahili languages. It seems like Arabs on the coast rapidly became Kiswahilispeakers, as this was the language of the caravan trade. It has been no-

Bagamoyo and the Caravan Trade

•

227

ticed that, for example, the rebel leader Bushiri and the warlord and slave trader Tippu Tip both preferred Kiswahili (Legère 2002: 3). Another force behind the spread of Kiswahili from Bagamoyo to the interior of the continent was the German colonizers’ decision to make Bagamoyo their first colonial capital in German East Africa. From here the Germans began to extend and consolidate their control over this enormous territory, which included present day Tanzania (except Zanzibar), Rwanda, and Burundi. In this process the colonizers decided to make Kiswahili the administrative language of the colony. In the initial stage of the colonization they used coastal Kiswahilispeaking people to fill the lower posts in the administration. These civil servants had o en been to a Quran school where they had learned to read and write using Arabic script. Some also had direct experience from the people, conditions, and societies in the interior of the colony from having been there with caravans. Their use of Kiswahili in the colonial service increased the prestige of the language and it became a ractive for people outside the coastal zone and the caravan routes to learn it (Malik 1999: 155). The imperial race for colonies in East Africa inspired the German colonizers to adopt an education policy that stipulated, at least for two reasons, that Swahili was both the medium of instruction and a subject in the schools. One reason was to develop the language as a lingua franca over the vast colonial territory. Another was to restrict potential influences from competing colonial powers over the loyalties of the colonized. The German colonial language policy was designed to meet the fear of influences from the existing French and British missions and the fear of armed revolts from the colonized population. The major mission orders in the area were French, the Holy Ghost Fathers (from 1868) and the White Fathers (from 1878), both Catholic, or British, The Universities’ Missionary Society (from 1875) and the Church Missionary Society (from 1876), both Anglican. The Germans suspected that the languages of rivaling colonial powers were taught in the schools run by the missionaries, although each of them used Swahili both for education and proselytization (Pike 1986: 214). They also hesitated, on racial grounds, to promote the German language in the new schools established by the German administration. The colonial enterprise rested on the myth of the superiority of the colonizers. To use the German language for the development of a cadre of junior administrators would unite the colonized and the colonizers or, in the words of Carl Meinhof “I am fully convinced that the relationship between Germans and Colored People is best when they maintain their distinct languages” (Meinhof 1910: 736, quoted in Pike 1986: 231).

228

•

Muted Memories

The German colonizers began organizing the education of the “natives” to meet their own needs. Beginning in the 1890s, schools were established by the German colonial administration to educate a cadre of Africans needed in the colonial administration since: “More of the indigenous population would have to become proficient in arithmetic and Swahili (using the Latin script) if the new needs for clerks and cra smen in administration, railways and plantations were to be met without resource to expensive immigrant labor” (Polomé and Hill 1980: 201, quoted in Malik 1999: 156). The first German Governor, von Soden, therefore introduced a government school system in East Africa aimed to supply the administration with low-level staff. The school taught subjects like typing, the dra ing of official le ers and forms and so forth, while most mission schools focused on practical skills like carpentry, gardening, and other handwork.37 We see a division of labor, although both were geared to supply trained “natives” to serve in the colonial administration and at the mission station. The pupils, at both government and mission schools, were taught “obedience, punctuality, consciousness and a sense of duty” (Cohen 1993: 130–31).38 Swahili became the medium of instruction when a number of government schools, Regierungsschule, were established in the 1890s. The Germans, however, saw a threat in the connection between Swahili and Islam. Their solution was to try to de-Islamize the language by adopting Latin script at the expense of the Arabic script in both government and mission schools. Bagamoyo served as the entrance of Islam, Christianity, and Kiswahili, which spread to the interior of the continent through the caravan trade along the central routes. When the German colonizers at the turn of the nineteenth century decided to move their colonial capital from Bagamoyo to Dar es Salaam, and built a railway from Dar es Salaam to Kigoma at Lake Tanganyika and Mwanza at Lake Victoria, the fate of Bagamoyo was sealed. This meant the end of Bagamoyo as the entrance to the heart of Africa and of a unique cultural institution, the caravan trade along the central routes.

The Spread of Diseases along the Caravan Routes An exploration of the roles Bagamoyo and the caravan trade played in the dissemination of people, goods, religions, and languages and as a cultural route would be incomplete without a few words on the spread of lethal epidemic diseases along the caravan routes. In 1859, a cholera epidemic developed in many towns on the Mrima coast. It is believed to

Bagamoyo and the Caravan Trade

•

229

have reached Zanzibar from Pangani, a trade port north of Bagamoyo. In Zanzibar town the disease killed a quarter of the island’s population or approximately 35,000 people. In Bagamoyo, 1,600 people lost their lives before “the disease made a secondary spread several hundred kilometers further inland over Ukwere and Uluguru to Ugogo” in central Tanzania (Koponen 1980: 162). Missionaries who traveled on the central caravan routes reported entirely depopulated villages and heaps of ro ing corpses along the paths (Koponen 1980: 162). An even more lethal disease that spread along the caravan routes was smallpox. Outbreaks seem to have occurred at intervals. The mission station in Bagamoyo reported severe outbreaks in 1872, a few years a er the station had been established. Another outbreak occurred in 1881 and 1882, and a year later the disease spread to the hinterlands. It has been observed that the disease was endemic in some places along the coast, where it mainly infected children and where the population knew how to inoculate them against the disease. When German colonizers during the 1890s reported on its spread they mentioned causalities among adults, which indicates that the disease now spread among people who neither had the knowledge of inoculation nor had been exposed to the disease as children (Kjekshus 1977: 132). When it spread to the interior with the caravan trade and the expeditions of travelers from Europe and America, “it invaded areas where it was not known . . . [as] is strongly suggested by the very destruction it le in its wake” (Koponen 1988: 164). The fact that it a acked adults as well as the children indicates that it was a new disease or that it reestablished itself a er many years of absence in the up-country societies. The disease was spread, probably in both directions, by traders and porters along the caravan routes and had soon become endemic in trade centers like Tabora and Ujiji and in the Ugogo area in central Tanzania (Koponen 1988: 165–66). The role of the caravans in the spread of this deadly disease made the caravans and their traders and porters dreaded in many up-country societies (Koponen 1988: 168), and many villagers rese led away from the caravan routes. Others tried to defend themselves in densely populated se lements behind strong palisades. This created a vicious circle for the people in the up-country societies. As they se led in nucleated villages with defense walls to protect themselves against the ruthless slave hunters, European caravan leaders and colonizers, and new lethal diseases, they also facilitated the spread of the deadly diseases brought by the caravan porters and traders. The result could be devastating. Stanley reports: “The bleached skulls of victims to this disease which lie along every caravan road, indicate but too clearly the havoc it makes annually, not only among the ranks of the several trading expeditions, but also among the villages of the respective tribes” (Stanley 1872: 533).

230

•

Muted Memories

Notes  1. Today, however, many reject and question their motivations, performance and results.  2. This interpretation of the name Bagamoyo as the entrepôt into the interior of the continent demands a change of perspective by 180 degrees from what has been applied so far in this work.  3. Focus will be on the role of the Catholic Church in the spread of Christianity since Bagamoyo is the place where the first Christian mission of the Holy Ghost order was built in the 1860s on mainland East Africa, with the ambition to expand to non-Muslim up-country societies.  4. Although Muslim scholars were the first to reproduce Kiswahili in wri en form, using the Arabic alphabet, a number of early English-Swahili or GermanSwahili dictionaries and grammars in Latin script were produced by missionaries and mission societies in order to make the language a medium for communication between the missionaries and the local population (see Mukuthuria 2009: 37).  5. Many of the first Muslims on the East African coast came from the important trading town Ibadiyya in the Basra area in today’s Iraq. They are known as the Ibadi subsection of Islam and disappeared from the coast during the twel h century. Ibadi Muslims from Oman returned to the East African coast during the eighteenth century to establish the sultanate on Zanzibar (Horton and Middleton 2000: 64–65).  6. A map in J. Spencer Trimingham (1964: at end) shows a partially broken corridor with a “Muslim minority” from the East Coast via Kondoa-Irangi, Iramba and Tabora Districts to Ujiji at Lake Tanganyika.  7. It should be noticed, however, that some coastal traders, when they se led in up-country communities, abandoned trade in favor of plantations. In Tabora, for example, “during the German occupation, the Muslims . . . split in two factions according to their main occupation: one faction was composed of plantation owners, who supported the Germans, and one was formed by traders, who allied with the local chief Isike, opposed the occupation fearing that the German presence could damage their business” (Pallaver 2012: 6).  8. Others decided to stay and take up a life on the coast. For them Islam was an avenue to integrate into their new social se ing (W. T. Brown 1970: 79).  9. Manyema was a collective name used for a number of groups who lived in the Manyema area in the eastern part of Congo. 10. Exceptions are B. Brown (1971), “Muslim Influence on Trade and Politics in the Lake Tanganyika Region”; C. F. Holmes (1971), “Zanzibari Influence at the Southern End of Lake Victoria: The Lake Route”; H. W. Langworthy (1971), “Swahili Influence in the Area between Lake Malawi and the Luangwa River”; K. Pallaver (2012),“Muslim Communities, Long-Distance Traders and Wage Labour along the Central Caravan Road”; M. Wright and P Lary (1971), “Swahili Se lements in Northern Zambia and Malawi.”

Bagamoyo and the Caravan Trade

•

231

11. Karin Pallaver is probably ignorant herself of the role of the tariqa when she states that “The limited number of coastal traders living in the interior was one of the reasons behind the limited influence of Islam in the area at this period” (2012: 7). 12. Also known as Barawa, Barawee, or al-Barawi. 13. This evidences the intricate relations among the various Muslims canons along the coast. The Sultan belonged to the Ibadi school; the majority of the population were Sunni-Shafí’ite Muslims, while Shaykh Uwaya bin Muhammad belonged to the Sufi subsection of Sunni Islam. 14. When the Brotherhood spread to eastern Congo the organization got the name Mulidi, or Muridi Movement (Martin 1969: 485). 15. Qadiriyya was, for example, established in 1908 in Mpwapwa in central Tanzania (see Nimtz 1980: 68). 16. There was a high demand for slave children in Arabic families where they were “groomed” into the households to become domestic slaves (Sheriff 1987: 37). 17. Amir adhered to the endogamous Ibadi sect of Islam, which made it difficult for him to incorporate his slave into this sect. Ramiya therefore became a Sunni Muslim of the Qadiriyya Brotherhood. A paradox may be noticed here since the Ibadi school is considered to contain more egalitarian tendencies, although this sect’s members became among the richest plantation owners in East Africa (Cooper 1981). See also Sheriff (1987: 18–19) on the early contradiction that developed in Oman between the Ibadhi ideology to elect their imam, who even had to be accepted by the “commoners” and the concentration of wealth in a few hands, which contravened the principles of an ascetic imam. “The growing secularization and tendency towards temporal power ran counter to Ibadhi principles of an elected Imam” (Sheriff 1987: 18–19). For the consequences of this contradiction, see Sheriff (1987: 19). 18. See, for example Kopf (2005: 2) on how the German colonial policies contributed to the spread of Islam. 19. Kopf (2005: 1) states: “Although Islam was concentrated on the coast of East Africa, anti-Islamic animus shaped colonial policy throughout the protectorate. The denigration of Islam was not a coincidental side effect to the emplacement of German values; colonial administrators intentionally sought to dismantle Islam as a viable economic, political, and cultural force in the colony.” 20. See the previous chapter on the Germans’ a empt to explain the Bushiri rebellion as an “Arabic uprising,” since they believed that only Arabs, not Africans, were capable of organizing such a movement. 21. Since the focus in this text is on the role of Bagamoyo as the entrance for Christianity, the text is mainly restricted to the Catholic faith for two main reasons: the first mission station (Catholic) on mainland Africa was built in Bagamoyo and other denominations, such as the Anglican and the Protestant churches, have always been very marginal in the town. 22. Father Henschel reports that one-quarter of the slaves received during the mission station’s first years were under the age of nine (Henschel 2001).

232

•

Muted Memories

23. Another Catholic order, The White Fathers, acquired in 1878 the right to the mission field 550 kilometers and further afield from the coast (W. T. Brown 1971a). 24. Time, space, and work were, in a Foucauldian manner, used by missionaries, “explorers” and colonizers to discipline and “civilize” the “Africans.” For the intruders “clock and calendar were the umbilical cord to civilization” (J. Fabian 2000: 56). Jérôme Becker (1887, 2:308, quoted in J. Fabian 2000: 54–55) maintains that “the only way to civilize Africa is through agriculture, commerce and industry. ‘But the element more indispensable than money and science is time, magister rerum.’” 25. Another Christian village at the station, Thomasville, had also forty families (S. Fabian 2007b: 98). 26. R. Baxter, in his A Christian Dictionary (1673: i. 285) declares: “A wise and well skilled Christian should bring his ma ers in such order, that every ordinary duty should know its place, and all should be . . . as the parts of a Clock and other Engine, which must be all conjunct, and each right placed” (as quoted in Thompson 1967: 87). The concern the European elite showed for the colonized resembles the concern they felt for disciplining the working-class men, women and children in Europe. 27. See also Kollman 2005, xx, on how the mission station kept the freed slaves in a state of continuous dependence. 28. The mission station and the missionaries were not always distanced from life in Bagamoyo, for instance, when more than three thousand denizens fled to the mission station during the 1888–89 Bushiri revolt and when the mission station and the Fathers played an active role in protecting the refugees (S. Fabian 2007b: 124). 29. This issue was also raised among the Spiritans themselves in the question “Wasn’t this policy [to focus on slaves and ex-slaves] an obstacle towards the evangelization of free Africans who never experienced slavery” (see Mallya 2011: 4). 30. Christian missionaries on the coast operated in “a barren area” (S. von Sicard 1970: 88, quoted in Brennan, Burton, and Lawi 2007: 27) 31. An agreement had been reached between Sir Bartle Frere, representing the British Imperial government and the CMS to establish a center for freed slaves on the East African coast. The agreement included a program for subsidizing the CMS for each slave liberated by the British and handed over to the mission. Frere, however, changed his mind, much to disappointment of the missionaries, since he saw The Holy Ghost Fathers and the Universities’ Mission to Central Africa as being more capable of taking care of the freed slaves (Githige 1986: 209f.). This episode indicates that lucrative trade in slave children could continue even a er they were freed. 32. I prefer the term Kiswahili for the “language of the Swahili people,” since the term Swahili also is used for the people, their culture, and society. 33. Chapter 10 discusses how this fact caused problems for the Europeans who thought in terms of tribes and saw the colonized world through a tribal grid.

Bagamoyo and the Caravan Trade

•

233

34. Author’s field notes from 2002. There was, however, another trading network developed before the Portuguese interference around 1500 CE—the one between Kilwa and Sofala and the Lakes and the empires in Zimbabwe and Zambia. Coastal traders se led along the trade routes and in the up-country political centers during the thirteenth and fourteenth century. It must be assumed that these se lements may have contributed to the spread of Kiswahili to the interior of the continent, but this skill probably receded when this trade declined as a result of the Portuguese impasse. 35. Different subdialects have been identified in the kiNgwana dialect, as the Lualaba variant spoken in the Kivu district and “the more pidginized in Oriental Province” (“Swahili: Kingwana and Other Simplified Forms” 1975: 677). 36. A variant of Kiswahili is also spoken in Katanga where it became the lingua franca of the workers in the mines. An impetus for the spread of Kiswahili in the Congo area was the Belgian colonizers language policy, which can be summarized as strife to favor a language different from the language spoken in the British colonies and also different from the French spoken by the colonizers themselves since, in the words of D’Hemtinne, adjoint supérieur at HautUele, “It would be a mistake to promote a European language in the Congo. We shall all be exposed to indiscretion by the blacks, who can be counted on spreading what we would like to be kept secret” (quoted in J. Fabian 1986: 57), a policy which resembles the language policy adopted by the Germans in Tanganyika. See also Goyvaerts (1987) for the different Swahili dialects and their distribution in the Congo. 37. It has been argued (Woulfin 2011: 102) that the focus on practical skills in some schools rather than intellectual training for the Africans was influenced by the racial belief of the intellectual inferiority of non-Europeans. 38. To maintain this moral order the Germans made frequent use of the Kiboko (Hippopotamus hide whip) and became known as the “people of the twentyfive,” since a standard punishment was twenty-five lashes with the whip.

9  Old Bagamoyo Central on the stage in this exploration of the East African caravan trade are the caravan porters and traders from the central parts of Africa. When they reached the coast they not only laid down their hearts a er the long journey, but also their ivory. Father Horner, one of the first Catholic missionaries to visit Bagamoyo during the 1860s, painted the following picture: Coming from very far away, these blacks carry on their heads and shoulders large quantities of ivory, copal, grain, animal hides, and other articles of trade. All these goods are spread out along the shore of the ocean. While waiting for the sale of their goods, which can take some two months, they stay on the coast to exchange them for cloth or goods from the coast. Nothing is more curious than to witness them barter the price for an elephant tusk. They begin in the morning at 6am, haggle until 6pm, and continue doing so for fi een days without ge ing tired and without concluding anything. (Quoted in S. Fabian 2007b: 182)

The ritualization of the transactions between coastal traders and the Nyamwezi traders and porters was also captured by Burton. The Nyamwezi who crossed the Zanzibar Channel to trade their ivory in Zanzibar town began their bargaining at night and: to a civilized man the work would be an impossible trial of patience. . . . Each article is laid upon the ground, and the purchaser begins by placing handsome cloths, technically called “pillows,” under the point and bamboo of the tusk, and by covering its whole length with a third; these form the first prerequisites of the seller. A er a few days, during which rice and ghee, sugar and sweetmeat must be freely supplied, commences the chaffering for the price. (Burton 1860, 1: 420)

Old Bagamoyo

•

235

What kind of society did they meet when they reached Bagamoyo after weeks of marching and carrying their heavy cargo? This chapter will present some characteristics of such a port town and try to find answers to why Bagamoyo developed into the most important nineteenth century port in the East African caravan trade. A second aim is to capture the extremely intricate economic and political life of this port town on mainland Africa. The Nyamwezi caravan traders/porters who arrived in Bagamoyo met a highly fragmented but vibrant port town. Old Bagamoyo (approx. 1800– 19121), as already mentioned, was the result of three converging human initiatives. One initiative was conducted by the entrepreneurs and explorers from the heart of the continent, who, in the absence of navigable rivers, extended their up-country trading networks and organized the first caravans on land that brought ivory to the coast on the heads and shoulders of the porters. The second was conducted by the traders and seafarers who learned to utilize the monsoon winds to sail, trade, and barter all over the Indian Ocean. The third was the active part played by the people of Bagamoyo in the booming caravan trade.2 The town developed into a highly vibrant sociocultural, economic, and political society. Its traders and financiers organized the trade in both directions. Imported goods from China, India, Europe, and North America were repacked and customized to meet the taste and demand of consumers in up-country societies and the traders of the town negotiated with caravan porters and traders over the purchase of their merchandise. The middlemen-merchants were the agents who could assure that the oceanic traders would find sufficient quantities of tradable goods waiting for them to make their travel profitable when they reached Bagamoyo. It was also the town’s financiers and traders who financed and equipped various types of caravans that set out from Bagamoyo into the interior of the continent. The wealth and cultural creolization generated through the caravan trade resulted in a unique sociocultural se ing different from other societies involved in the two interconnected trading networks. This chapter explores some components of this uniqueness.

Early Settlements in the Bagamoyo Area The East African coast has been occupied for a number of millennia. The concern here, however, is to sketch the decades when Bagamoyo was established and when the town bloomed. It has been shown in another part of this work that there are both academic and political controversies concerning the origin and spread of what has been called the Swahili culture.3

236

•

Muted Memories

The position taken in this research combines well with Deborah Fahy Bryceson’s statement that “the East African culture [Swahili culture] is neither African nor Arab. It constitutes a fusion culture evidencing all major distinguishing features of a creole society” (Fahy Bryeson 2009: 364).4 The movement of various population groups, goods, and ideas along the coast and from southwest Asia during the last two millennia (Middleton 1992: 14–15) caused the creolization processes that characterize the coastal towns and drove the development of the cosmopolitan character of parts of the population. Mlingotini, a village approximately 15 km south of Bagamoyo, and Kaole, 6 km to the south, were among the first trading and power nexuses on this part of the Mrima coast, but they lost their position as a new trading post, Bagamoyo, grew to prominence. Bagamoyo, originally a village subsisting on fishing and agriculture, probably called Pumbuji, began to expand at the end of the eighteenth century, which was the result of two not yet fully understood processes. One was the expansion of caravan trade on the African continent, which is in focus in other parts of this work. The other was the establishment of a trading community and culture in the fishing village, which was to become Bagamoyo town, which is in focus in this chapter. The development of Bagamoyo has been captured by Walter T. Brown in the following words: “Although Bagamoyo was younger than most of her coastal neighbors, her economic, cultural and political life was strongly rooted in the communal experiences of her older neighbors” (W. T. Brown 1971a: 79).

Illustration 9.1. Ancient Islamic tomb at Kaole.5 Photo by the author.

Old Bagamoyo

•

237

The history of the development of the Swahili se lements and culture may conveniently be divided into three major phases. The first was the early formation from around 800 CE to 1500 CE, which probably also included a movement of people from Shungwaya in present day Somalia and all the way to Chibuene in southern Mozambique.6 The second phase was the Portuguese intrusion7 between approximately 1500 and 1700. This impasse resulted in major changes in the Indian Ocean trade since the Portuguese brought violence, armed trading, and trade monopoly to a trade that so far had been free for traders from China, Java, India, Arabia and East Africa (Sheriff 2010: passim). The last phase coincided with the Omani control over the trade in the western Indian Ocean region from around 1700 to 1900. Since this text focuses on Bagamoyo, established at the end of the eighteenth century, the concern here will, except for a few references to the earlier phases, mainly be on the last phase of the development of the Swahili world. The Swahili se lements established before the Portuguese intrusion, which started around 1500 CE, were based on small-scale trade, agriculture, and fishing. The trade activities were restricted, since the coastal se lers mainly operated as middlemen in trading networks between the immediate interior and societies around the Indian Ocean. A specific natural feature that probably affected and restricted the coastal traders’ direct contact with the up-country societies was the Nyika, a semi-desert belt which runs parallel to the coast, making travel difficult (Sheriff 1987: 8). It may be assumed that the goods from the interior probably reached the coast a er having shi ed hands many times along the trade routes in a kind of relay trade mentioned above. The middlemen on the coast traded the goods received from the interior of the continent further eastwards when they sold or bartered them with merchants from the Arab world, Persia, and India who, in turn, had trading contacts with China and the Mediterranean region. The middlemen in the coastal towns, in their turn, traded with products from societies around the Indian Ocean such as goods from India, but also dried and salted fish from the coast. “Nineteenth century travelers especially acclaimed the popularity of salt fish as a food all along the coast and into the interior” (W. T. Brown 1970: 74). One of the main items of trade was cloth from India and/or the East African coast. Weaving has a long tradition in some of the port towns and the bulk of cargo carried from Bagamoyo to up-country consumers was bales or rolls with co on cloth. The local skills and cra s, however, began to suffer with the globalization of material production. The local weavers and blacksmiths were among the first to experience a stiffening competition from the growing import of cloth and metal ware. They were first transformed from be-

238

•

Muted Memories

ing producers of such items to become menders of damaged imported goods. Eventually much of their skills disappeared. An opposition politician in Tanzania captured this process when, just before the year 2000, he declared: “Tanzania entered the colonial period with locally produced hoes. She entered Independence with hoes produced in Sheffield. Now we enter the new Millennium with hoes produced in China and India. But [they are] still the same bloody hand hoes” (author’s field notes, December 1999), as a comment on the country’s development. It is also essential to note how the history of Bagamoyo differs from that of earlier trade towns along the coast in order to capture the role of Bagamoyo in the caravan trade. The “classical” Swahili trade ports, for example Kilwa, Pate, Mombasa, and Malindi, became centers for local merchants who developed what has been called Swahili culture or the Swahili world. This world has been called the “Swahili corridor,” the “Swahili rosary,” or the “Swahili wheel.”8 All the metaphors aim to give an impression of the Swahili world, but while the first two focus on what may be perceived as isolated se lements along the narrow coastal strip, the third has the advantage of also capturing how the trade towns were linked, or extended their tentacles in all directions to other places outside the specific town. The trade, which took place in these early towns, was a highly ritualized, culturally condensed phenomenon between traders belonging to different trading networks and with the middlemen-merchants as the interface. Middleton chose the term “merchants” rather than traders for these middlemen (see also Horton and Middleton 2000; Sheriff 1987). Their main trade activity in the coastal towns was to obtain goods from either the continental trading networks or the Indian Ocean networks and pass them on. Before the development and expansion of the long-distance caravan trade, they never, or rarely, met the consumers of the goods they traded, but dealt with visiting traders from the two networks. The trade, which took place when visiting traders appeared in the town, followed rules other than those dominating in a market economy. A visiting seaborne trader was expected to present an opening gi to one of the town’s middlemen-merchants, who then became his host. A er a number of ritualized exchanges, the main negotiations over the newly arrived goods took place in seclusion in the host’s stone house, where the goods obtained from the continental trading networks were, in turn, displayed to the guest. Other trade carried out by the visitor took place in the more public parts of the host’s house where other merchants than the host could also be present (Middleton 1992). The trade that developed in Bagamoyo during the nineteenth century was partly of a different character. As a result of the incoming caravans from the interior of the continent Bagamoyo was established not only on

Old Bagamoyo

•

239

the ritualized gi -and-exchange economy that took place in seclusion, but also on a market economy when the Nyamwezi traders and porters put up their ivory on the shore for public display and with open negotiations. It may be assumed that this kind of trade was the result of the new actors in the caravan trade, the Nyamwezi and the Indians. It was the Indians backed by the Sultanate, not the middlemen-merchants, who controlled much of the market-oriented trade, including the goods brought from the interior or imported from India via Zanzibar and sold to the Nyamwezi traders/porters. The development of Zanzibar Stone Town and Bagamoyo are interlinked since, in Walter T. Brown’s words, “without the mainland’s rich supply of export products, Zanzibar could not have emerged as a major economic power” (1971b: 45) and correspondingly Bagamoyo, without the access to the markets on Zanzibar, would probably not have become the dominating trading port on the mainland. But there are more answers to the question why Bagamoyo grew with the booming caravan trade than its proximity to Zanzibar Stone Town, which will be further developed below.

The Monsoons The direction of the winds over the western part of the Indian Ocean have been known and utilized by sailors for at least three thousand years (Horton and Middleton 2000: 9). The monsoons have transported traders/colonizers and cargo to and from coastal se lements all around the Indian Ocean. The revolt of African slaves in Basra in present day Iraq between 869 and 883 CE provides evidence of the early development of the trade. The “Black Sidis” in Gujarat in India, still performing African ngoma dances for their audience, is a similar testimony, as are the two giraffes presented in the year 1415 to the Chinese emperor in Beijing as a gi from the leaders in Malindi.9 The presence of a large number of glass beads and cowries in graves from the eleventh century in Ntusi, Uganda, is also evidence of the historical depth and spatial extension of the trading networks on the Indian Ocean, which utilized the trade or monsoon winds, and the continental networks that depended on the porters’ strength and skills. There is a seasonal reversal of the winds blowing over the Indian Ocean, which gives rhythm to the travels between the western, northern, and eastern parts of the Indian Ocean and has affected the structure and flow of trade in this part of the world (Sheriff 2010).10 The kusi monsoon blows from the southwest, i.e. from the East African coast toward India,

240

•

Muted Memories

Map 9.1. Kusi: The monsoon from the southwest. Created by the author.

between April to September/October (see Map 9.1). However, there is an interruption in the monsoon during the period from mid-May to midAugust, when the monsoon is unsuitable for smaller sailing vessels to sail in the northwestern part of the Indian Ocean. The dhows, wooden ships with a lateen sail, therefore tried to leave the East African coast in April or wait for the tail end of the monsoon in August to November. This strategy, Abdul Sheriff (1987: 10) remarks “becomes increasingly necessary if dhows have to proceed south of Zanzibar.” Then it turns and the kaskazi monsoon blows from the north and east, i.e. from the Bay of Bengal and India, the Persian Gulf and the Arabic Peninsula towards the East African coast from October/November to March. At the beginning it is steady and light in the western part of the Indian Ocean while the eastern part of the Arabic Sea o en experiences storms in October and November (Sheriff 1987: 10). The movement of the winds is accompanied by the currents in the Indian Ocean, which roughly follow the wind direction and facilitate navigation and sailing. The monsoons made it possible for sailing ships to set out from the Arabian coast to reach East Africa within thirty to forty days or, alternatively, within a time span of nine months, to set out from India and reach the East African coast and then wait for the wind direction to change in order to

Old Bagamoyo

•

241

Map 9.2. Kaskazi: The monsoon from the northeast. Created by the author.

set sail again and return (Horton and Middleton 2000: 10). If trading was undertaken along the coast of East Africa, it took approximately fi een months for a ship se ing out from the western shore of India to sail via the Persian Gulf and southern Arabia along the coast southward to Mozambique, and then back to India. Even greater distances were covered by those ships and traders who set out from Java to reach the East African coast (LaViole e 2008: 29–30). There are disagreements regarding the extension of the monsoon south of the line between Zanzibar and Bagamoyo/Dar es Salaam. These areas were more difficult to reach by sailing ships and were probably less involved in the early trading networks in items other than slaves until faster and larger vessels were developed during the thirteenth century (Horton and Middleton 2000: 9).11 The lacuna, created when the larger transoceanic sailing ships had difficulties reaching the southern parts of the East African coast, was certainly filled by the Swahili merchants, who acted as middlemen. With their o en smaller but slower sailing boats they could transport the cargo from Sofala in southern Mozambique (Datoo 1970), including slaves, northward to ports like Zanzibar and Pangani where transshipping took place when the larger vessels from Arabia, Persia, and India arrived.

242

•

Muted Memories

Zhao, on the other hand, reports that the monsoon wind made it possible for sailing ships to even reach Kilwa in southern Tanzania.12 It seems safe to assume, however, that Zanzibar, as the pivot in the trade, was strategically selected to be within the regimes of the monsoon winds. Even Bagamoyo, established in the late eighteenth century, was situated within the monsoons that favored this port town and provided it with the fresh breeze, which, among other things, gave the town its reputation as a pleasant place during the caravan trade.

Factors behind Bagamoyo’s Growth to Become the Most Important Mainland Port Despite its relatively recent birth, Bagamoyo’s origin is clouded in the mist of history. Most records link it to the demise of Kaole, a Swahili town approximately 6 km to the south of Bagamoyo, which flourished approximately between 1300 CE and 1800 CE. The demise and eventual death of Kaole, which today is in ruins, has in turn been ascribed to the fall of Kilwa, the dominant town in the south and a slave-trade based empire that also controlled Kaole. Another reason mentioned frequently in the literature is the re-growth of the mangrove swamp, which suffocated Kaole’s harbor. The first explanation is probably true. Slave trade was an important source of income for Kilwa and in extension possibly also for Kaole, and when the Portuguese a er the year 1500 CE destroyed Kilwa’s trade networks and subsequently, during the second half of the nineteenth century, the British enforced the abolition of the slave trade, the financial base for this empire collapsed. But why did Bagamoyo prosper while Kaole was deserted? One plausible explanation for the power shi from Kaole to Bagamoyo may be found in local politics and power relations in the growing presence and power of the sultanate on Zanzibar in the coastal societies. A clear clash of interests occurred, as already mentioned, in 1844 when a military confrontation took place between the sultan’s forces and the local Zaramo people around Kaole, a er which the sultan arranged for the establishment of a military garrison there. This garrison, with Baluchi soldiers, dramatically increased the sultan’s presence and pressure on the mainland (S. Fabian 2007b: 50–51). In 1874, the Sultan appointed his own representative, a liwali, whose prime duty was to secure the economic interests of the sultan in the area. This move caused considerable friction between the major trade actors involved, since it threatened the established balance between them (see below for details on the agreed division of power and interests). These

Old Bagamoyo

•

243

were the leaders of the dominating middlemen-merchant families of the town, called the majumbe, the leaders of the surrounding Zaramo society, the rightful owners of the land, called the pazi, and the Indian traders, who o en supported the interests of the sultan who had encouraged Indian businessmen to finance parts of the caravan trade. The two first groups now had to deal with a fourth major player, the sultan’s representative (S. Fabian 2007b: 52–54). The growth of Bagamoyo was parallel to and a result of the booming of the caravan trade along the central routes. When the first Nyamwezi traders/porters reached the coast, probably at the end of the eighteenth century, and established direct contact with the middlemen in the coastal towns they probably did so in two or three se lements to the south of present-day Bagamoyo: namely Mwbamaji, a few kilometers south of present-day Dar es Salaam; Mlingotini, in the lagoon between the mainland and the peninsula; Ras Luale, 15 km south of Bagamoyo; and Kaole, 6 km south of Bagamoyo. Bagamoyo replaced these trading centers when both the volume of the cargo and the numbers of porters grew. There are a number of physical conditions that contributed to the advantages Bagamoyo had over other places. One is its safe, although open, harbor (S. Fabian 2007b: 179). The coral reefs and sand banks, which run to the east of the beach, break the oceanic waves and calm the water’s surface toward the beach. Another is the continental shelf that widens between the Pangani and Rufiji rivers, the area where Bagamoyo is situated, which creates a current that is favorable for sailing ships when the monsoon wind is weak (Datoo 1974: 32). The soils, especially on the alluvial plains along the Kingani/Ruvo and Wami rivers to the northwest of the town are fertile and produce a large number of food crops. Other parts sustain large coconut plantations. The breeze, which blows into the town, provides a comfortable climate. In short, Bagamoyo gained a reputation as a place where you could live a “comfortable life” (W. T. Brown 1971a, v; see also S. Fabian 2007b, 45). When the caravan routes west of Bagamoyo were moved northward to avoid the Morogoro area (see Map 5.1) where the caravans frequently fell prey to a acks by the Hehe and Mbunga, Bagamoyo had still an advantage compared with the ports to the south, namely the proximity to the utilization of a newly developed route where it was possible to pass River Kingani (Ruvo) and Wami.13 This movement also had positive effects on the development of one of the most important camp sites and trade markets along the central routes: Mpwapwa, situated west of the Rubeho mountains and east of the food- and water-scarce Marenga Mkali plain (“Bi er Water,” see Map 5.1). Mpwapwa grew during the second half of the nineteenth century to become the most important stop for caravans

244

•

Muted Memories

in both directions from Tabora in the interior to Bagamoyo, Saadani, and Winde on the coast. Both Bagamoyo and Mpwapwa benefi ed from the shi of the main routes from its previous stretch further south between Ruaha through Ukimbu country to Isanga. Drought, low population density and food scarcity made this older route less a ractive (see Rockel 2006b: 9–10). Another advantage of the Bagamoyo area was that during the “slow season,” when few or no caravans reached the coast, other products and activities could fill the gap. One such gap-filler was the trade in gum copal. This resin-like product had long been exported to China and Japan where it was used to produce high quality varnish. Copal is produced by the msandarusi tree (Hymenaea verrucosa in Latin) and obtained by digging in the soil under such trees or in the soil where such a tree has stood.14 It has been reported that the Chinese, before 150 CE, were interested in obtaining copal from the East African coast (see W. T. Brown 1970: 74n46.). During the nineteenth century it was in high demand, especially in the United States where it was used for the production of carriage and furniture varnish. The copal from the Zaramo area surrounding Bagamoyo produced varnish of extremely high quality and the trade in this product became a seasonal complement to the caravan trade for Bagamoyo’s business people (W. T. Brown 1971a: 119–20).15 An indication of the role of copal export is given by W. F. Prideaux (reproduced in W. T. Brown 1971a: 248f.), who provides statistics for the period from August 1872 to August 1874 for the “Produce of the Zanzibar Dominion on the Coast and adjacent island, imported to Zanzibar.” The total value of the goods exported from Bagamoyo was $621,300, of which $200,000, or close to one third, came from copal, compared to $400,000 from the export of ivory. Copal contributed to the resilience of Bagamoyo as the major port for the caravan trade. Other products, which supplemented ivory, were copra, rubber, grain and other food crops that became “rainy season supplements” to the dry-season trade in ivory and slaves. The good access to food, including fish, was important for the a raction of porters. Unlike many other coastal societies depending on the dry-season caravan trade, the diversified commerce in Bagamoyo flourished the whole year (B. Brown and W. T. Brown 1976: 186). When the ivory trade started its decline at the end of the 1890s due to the depletion of the elephant herds in combination with the redirection, as a result from the scramble for Africa, of the tusks from eastern Congo to the Belgian ports on the Atlantic coast and ivory from the Lake Victoria area to ports controlled by the British in today’s Kenya, both copra and rubber increased as important export products (S. Fabian 2007b: 193).

Old Bagamoyo

•

245

The diversified commerce in Bagamoyo certainly also benefi ed from the proximity to Zanzibar, the main recipient of goods from mainland Africa.16 The intricate struggle over the control of the caravan trade, in combination with the symbiotic relations between the sultanate and the Indian traders and financiers, helped them monopolize the trade, to the frustration of American and European traders/companies. For example, Edward D. Ropes Jr., an American merchant in Zanzibar “expresses his disgust over the merchant community of Bagamoyo who had bought up all the ivory of July 1888 (approximately 140,000 kg worth) and artificially raised its price by keeping it in warehouses” (as reported in S. Fabian 2007b: 197).17 The competition, as analyzed in a previous chapter, increased with the boom of the caravan trade. The marked difference in economic cultures between the Indian and European/American traders made a German trader notice that “I don’t do the Indians injustice when I accuse them of building up an economic mistrust [among the Africans] against us. At least I can’t think of any other explanation that at present all natives prefer to trust Indians with their savings than the whites”(Deutsch Ostafrikanische Zeitung V/29, 18 July 1903, quoted in S. Fabian 2007b: 198). These factors also contributed to Bagamoyo’s resilience and capacity to overcome both the natural and man-made disasters that struck the town during the second half of the nineteenth century. Two of the more severe disasters were a cholera epidemic and a hurricane. The cholera epidemic struck many coastal towns in 1859 and again ten years later when 1,600 people died in Bagamoyo during a period of five months. This tragedy was followed by a hurricane in 1872 that destroyed gardens and plantations, leveled the mud-and-wa le houses to the ground and destroyed all fishing boats (W. T. Brown 1970: 75; B. Brown and W. T. Brown 1976). The man-made disasters came from the hands of the German colonizers and local rebels, who protested against the colonization of the coast by Zanzibaris and Germans. The open conflict in 1888–89, which erupted a er frantic struggles between various actors on the caravan trade scene when the Germans used marine soldiers and bombarded the town from ba le ships to suppress the rebellion, together with the looting that followed when normal life was disrupted, destroyed much of the town. Pictures taken a er the Germans’ use of their Maxim machine guns show a town in ruins.

Two Types of Coastal “Towns” The term “Swahili town” deserves further explanation. Urbanity, in the eyes of those called the Swahili, was the core essence of being “civilized” or having utamaduni, from the Arabic stem medina, which, according to

246 • Muted Memories James de Vere Allen (1981b: 322) referred to a town that was well-protected by a high wall, built with solid materials, and was meant to endure for a long time. On the East African coast, the term mji (singular, miji in plural), “town,” had long been used for the settlements, which were arranged around a central Friday Mosque (msikiti wa ijumaa), where all the free male members of the community could assemble for the Friday prayer. But such a settlement may, as is shown below, be of a primarily rural character.18 Thus, the mosque was central for a settlement to qualify as a town. The mosque and Islam, however, were introduced from the Arabic Peninsula and are sometimes used as evidence for the Swahili culture as a product of external influences. John Middleton on the other hand, has argued that: it may be asked whether these towns are uniquely Swahili or mere transplantations or copies of the towns and houses of the Arab world. A proper Near Eastern or North African town is traditionally defined by the possession of a congregational mosque (jami), a public bath (hammam), and a central market (suq): the Swahili towns all have the first, none the second, and the third only in a marginal sense. They must be regarded as sui generis. (1992: 60)

He also gives a vivid description of the vibrant complexity of these towns when he reports that the inhabitants included ship owners and traders from the Arab peninsula and the Gulf, financiers from India, Swahili middlemen-merchants, fishermen and manual workers, people from neighboring villages and upcountry communities and slave laborers. This portrait captures the cosmopolitan character of the port towns and the forces behind the creolization processes that took place in them. The maritime orientation of Swahili societies was interconnected with urbanity, particularly to live in a stone house, and Islam. However, as in many other highly hierarchized societies the majority of the population could not live according to the cultural ideals. They were Muslims, but most were farmers and lived in thatched mud houses in villages on the coast. Hence, life in these coastal settlements was not lived in a uniform way. Marjorie Franken has captured this in the following way: “wealthy merchants and humble slaves—all living cheek by yowl in these towns” (1991: 148). Furthermore, only some of them offered the possibility to live a more urban way of life, while the conditions in others were definitely more rural. Most Swahili people lived in more rural settings, which have been called commoners’ towns (Horton and Middleton 2000: 123), sustained with farming and fishing. These settings were “inclusive” (Middleton 1992: 60) and consisted of mud-and-wattle houses. The coastal stone towns, which Mark Horton and John Middleton call patricians’ towns were, by contrast, often surrounded by stone walls and

Old Bagamoyo

•

247

with houses, at least some of them, built with coral stone. This type of settlement stems from the tradition of the Mijikenda people in south-eastern Kenya to establish se lements called kaya, which was a fortified village where five hundred or more families could live. Cultural links between the kaya se lements and the early Swahili se lements are that both were o en fortified and that their deceased members were buried within the boundaries of the se lements. Both the houses and the walls expressed exclusion and they “were complex, economically differentiated trade emporia whose residents performed a wide range of different functions—hosting and provisioning foreign traders, bulking export cargo and breaking imports, warehousing trade goods and supplies, financing commercial transactions, and building and repairing ships” (Spear 2000: 276). These trade towns were much more than just the interface between two trading networks. They were, as Horton and Middleton describe it (2000: 201), a hive of cra smen who manufactured textiles, sandals and shoes, metal wares of different kinds, jewels from various materials, wooden boats, furniture and more. Slaves lived in the midst of the stone houses, as did cra smen, free people, and the middlemen-merchants. Solid stone houses were built side by side with houses of wa le-and daub.19 The mixture of people is captured by Henry Morton Stanley when he describes a part of Zanzibar town: “It is here we find the Waungwana, or Freemen of Zanzibar, whose service the explorer will require as escort on the continent. Here they live very happily with the well-to-do Coastman, or Mswahili, poor Banyans, Hindis, Persians, Arabs, and Baluchis, respectable slave artisans, and tradesmen” (Stanley 1888: 26). The major cause for these marked differences between the urban and rural types of towns spring from where and how a se lement was placed in the caravan trade links. A third of the societies on the coast were without a port or had a poor port (Glassman 1995: 35), but still, such a port could a ract foreign traders and financiers as well as caravans from the interior of the continent. A town without this asset became the supplier of food to the trade towns contributing to what Allen calls the urban-rural continuum in Swahili societies (1981b: 330). Thomas Spear (2000: 277) expresses this interdependence in the following way: “Each stone town served as the focal point of an exchange network comprising a number of country towns and hinterland areas . . . [that] produced basic food stuff and export commodities for a stone town entrepôt in exchange for locally produced cra s and imports.” For the porter from an up-country society, both types of towns were of importance; one for trade and the other for food. It should, however, be noted that many porters also looked for a piece of land where they

248

•

Muted Memories

could grow their own food while they waited for the right price for their goods, or for the vuli rains to end, or waited for job in a new caravan with an up-country destination. Some also worked temporarily for a salary on coastal plantations. Since the focus in this study is on Bagamoyo as a trade town, limited a ention has been paid to the more rural towns in its surroundings. It is enough here to notice that the thousands of porters who came to the coast seasonally and needed food must have fueled the economic exchange between these two different kinds of se ings. It may be claimed that one of the most probable reasons for the establishment of the new port town at its present site was the interdependence that developed with the surrounding villages and their fertile soils.

The Political Puzzle in Old Bagamoyo The existence of Bagamoyo was the outcome of the caravan trade. This economy a racted traders and colonizers who tried to control the flow of wealth from the interior of Africa and/or, alternatively, from the coast to the heart of the continent. An intricate power struggle developed for control over the caravan trade and over the town. One fault line developed between the original “owners” of the area, the Zaramo people, and the middlemen-merchants, many probably relocating from neighboring Mlingotini and Kaole. Another developed between these two groups of mainlanders and the new intruders from other parts of the world as Oman/Zanzibar and Europe.

The Bagamoyo–Zaramo Agreement At least three categories of major political actors emerged in Bagamoyo around the caravan trade. In the earlier stages, a power balance was established between the former owners, the Zaramo people who mainly subsisted from cultivation, and the growing number of immigrants in the town, who also cultivated their fields but also specialized in trade and maritime activities like fishing and boat building. It was from the Zaramo leaders, the Mapazi,20 that the town dwellers got permission to se le on the land of Bagamoyo. It was also a famous Pazi named Kilama who, a er Bagamoyo had been a acked by marauding Kamba people from present-day Kenya, came to rescue the townspeople by expelling the Kamba (W. T. Brown 1971a: 96–100). Narratives on the Kamba war specify that the victorious Pazi demanded compensation for his assistance and placed a pole (mlingoti) in the ground

Old Bagamoyo

•

249

and requested the people of Bagamoyo to pile cloth21 until the pole was no longer seen. An agreement was reached, when the people of Bagamoyo failed to meet this demand, that they as a compensation should pay an annual tribute to the Pazi called kanda la pazi (the Pazi’s raffia bag), which was paid until the 1870s (Owens 2012: 374). The legend about the pole and pile of cloth is a version of a common trope in the coastal societies that has become a cliché with different messages and meanings. One specifies the political relations between various groups in these se lements in terms of a theme that explains that the cultural hero arrives to the coast and establishes himself as ruler over a port town by means of cloth and strategic marriage with the former ruler’s daughter. This theme is prominent in the Kilwa Chronicle relating the tale about the Persian prince who arrived to Kilwa Kisiwani and civilized the town by covering it with cloth (Nurse and Spear 1985) or used cloth to cover the path from the islands to the interior to which the former ruler was asked to move. A common theme in this kind of foundation narrative is that it charts out the relations between distinct groups in an area, for instance, between the newcomers and the original se lers in the Kilwa Chronicle or the legends about the Kamba war. The Kamba incident is said to have crystallized the power relations in and around Bagamoyo. The agreement between the middlemen-merchants of Bagamoyo and the surrounding Zaramo defined the areas of influence of the two groups. A er an agreement was reached, the Zaramo leader was called Pazi haoni maji or the “Pazi/leader who would not see the water,” i.e. the Indian Ocean. In a similar vein, the leader(s) of Bagamoyo’s townspeople was called Shomvi haoni jua or the “Shomvi22 who would not see the sun.” It is assumed that the reference to the sun means the west where the sun sets (W. T. Brown 1971a: 105–6). The narratives also convey a message about the divide between the civilized and the barbarian worlds, a fundamental distinction in the world view of many people in the western part of the Indian Ocean. The coastal areas around Bagamoyo were, as already mentioned, divided due to the Kamba war into two areas of influence. The leaders of Bagamoyo, the jumbe or shomvi “who would not see the sun,” controlled the maritime world while the leaders of the Zaramo people, the pazi “who would not see the water” controlled the coastal area through which the caravans passed. This dichotomy is elaborated in local traditions that portray the pazi as an “inversion” of the shomvi leader. The pazi is sometimes portrayed as a fierce warrior, elephant hunter, and cannibal preparing and eating human flesh. He has no food taboos, lacks a religion, and, in the words of Geoffrey Ross Owens (2012: 368f.) “he has no knowledge of standardized weights and measures, and other markers of civilization.” He is, in other

250

•

Muted Memories

words, ill-fi ed to take part in the noble art of trade and has to resort to a robbery-like extraction of road tolls from the caravans passing his area. The coastal traders and their leaders the jumbe or shomvi are, by contrast, maritime and mercantile. They are civilized, live in a town, and follow Muslim food taboos and religious commands. The pazi and the people under them personify, in the views of the coastal population, the opposite of their own civilization. The orientation toward the Indian Ocean is central in the Swahili identity, as is implicit in the legend. Only the people of Bagamoyo should “see” the Indian Ocean. Archaeologists suggest that the Swahili developed a maritime worldview a thousand years ago.23 It may be hypothesized that part of the Swahili world from then on became involved in an active reorientation toward the Indian Ocean (Fleisher et al. 2015: 101–10). W. T. Brown expresses hesitation as to whether the agreement between the people of Bagamoyo and the Zaramo people actually happened or not, but the point is that even if it is a construction it still shows that the people of Bagamoyo and the Zaramo delineated and respected each other’s political and economic spheres. Bagamoyo’s leaders were in control of the Indian Ocean and its shore, while the Zaramo pazi controlled the region’s inland. This division of authority and influence also implied a division of the right to collect taxes, levies, and tolls from the caravan trade. The Zaramo milked the caravans through taxes for the right to pass their land in both directions, use water, and camp in the area. The traders in Bagamoyo made their income from the first hand position they had to buy the goods from the incoming caravans and then resell it to traders on Zanzibar, but also from equipping the caravans with up-country destinations with their trade goods. The most important source of income for the Zaramo leaders came from the tax, called hongo, they exhorted from the caravans passing through their territory. Even the kanda la pazi seems to have been directly linked to the caravan trade as one tradition among Bagamoyo’s merchants maintains that “one-third of our income . . . out of the sales of our ivory, slaves etc.” should go to the pazi while another specifies that one-ninth of all goods should be paid (W. T. Brown 1971a: 103). The political and economic life in Bagamoyo revolved, in short, around the caravan trade. The intimate but well-regulated power relations between the two groups were well illustrated when a new jumbe (political leader) in Bagamoyo was to be installed. The ceremony was announced in the following way: The palaver horn Mbiu ya mgambo—when it sounds there is something afoot—the horn of the Shomvi and of the townspeople—the inhabitants of Pumbuji24 and the Jumbe—the horn of the pazi and of his entourage—the home-born slaves and true slaves—the horn of Makame ya Shani and the

Old Bagamoyo

•

251

freemen of Winde—Ukasimu and the freemen of Shangea—The Shomvi horn proclaims to you—that on Monday—there will be kishina and mwanamana dances—and on Thursday—Mwenye so-and-so is to rule—his name in now jumbe Mkomatembo. (Harries 1965: 192, quoted in W. T. Brown 1971a: 138, my emphasis)25

In another chapter we will return to how ill-equipped the colonizers were to understand the kind of society they arrived in due to their preconceived worldview, according to which an area like Bagamoyo and its surroundings was divided into easily demarcated ethnic groups and cultures. In the text above we see how different peoples and polities come together to celebrate the installation of a new leader for one of the groups. Steven Fabian (2007b: 50), however, cautions us to make a clear distinction between the majumbe and wapazi. In the reports from the nineteenth century, the first title seems to be associated with urban milieus while the la er is associated with rural se ings. When the town expanded, the distinction was, according to him, sometimes blurred. One incident, which illustrates the power relations in Bagamoyo, occurred in the 1860s when the town’s leaders granted a large land concession to the French Holy Ghost Fathers to establish the first Catholic mission station on mainland East Africa. This move was questioned by the Zaramo pazi on the grounds that their agreement with the people of Bagamoyo did not include the right for the majumbe of the town to allocate land to foreigners (W. T. Brown 1971a).26

Internal Divisions of Authority and Status The local traditions described in W. T. Brown’s historical study of Bagamoyo (1971a) identified four quarters or mitaa in Bagamoyo when the town was established at the end of the eighteenth century and at the beginning of the nineteenth century. These sections were known as Pumbuji, Kongeni, Mangesani, and Segera. The booming caravan trade increased both the growth of the town’s population from about 3,000 people in the 1850s (S. Fabian 2007b: 46) to approximately 20,000 at the end of the century and the formation of new quarters. One of the first additional mtaa (sing.) was Nchi Pana, built on the coastal strip north of Pumbuji, followed by, among others, Gongoini. This development, in turn, a racted the interest of the sultan on Zanzibar. He strengthened his control over, and presence in, Bagamoyo by establishing his customs house in this new part of Bagamoyo (later to be moved to Saadani when the Germans built their new customs house). It is also where the Sultan’s jamadar, the Chief Commander over his Baluchi mercenaries, was stationed (W. T. Brown 1971a: 120–21).

252

•

Muted Memories

At the height of the caravan trade the town was divided into twelve jumbeates (W. T. Brown 1971a: 136). A jumbeate was a specific geographical portion, which, in a feudal sense, contained land, buildings, and people— both freemen and slaves. Such a portion of the town was governed by leaders with the title jumbe (or diwani), usually a merchant-middleman27 who also acted as a political representative for the people who lived in his area when the leaders from the town’s different quarters met. These leaders were, (according to B. Brown and W. T. Brown 1976: 193) the most prominent merchants, land owners, and slave owners in the town. A successful jumbe not only had his income from his position as a middleman between the two trading networks, but also from the tributes he could extract from his dependents and the compensations he could claim for his specific services. Tributes should, for example, be paid on each cow slaughtered for celebration or sale; each dugong or shark caught; and on the salt produced at Nungwe, approximately 3.5 km to the north of Bagamoyo (W. T. Brown 1971a: 137). The compensations claimed by a jumbe for the services he provided to the inhabitants or visitors to his mtaa were of different kinds. One was the mboni he expected to have from a slave owner when a runaway slave was caught and returned. Another was the ada or compensation for “protection” from the Indian businessmen in his part of the town. A third was the hongo, or tax claimed from the caravan leaders and traders (W. T. Brown 1971a: 137), although perhaps not the most important source of income from the caravan trade. As owners of buildings they also o en had an income from renting buildings or land to caravan leaders and traders and, later on, to German colonizers. It is evident that the caravan trade provided an excellent source of income for the leaders of Bagamoyo. A jumbe’s aides occupied the next step in the political hierarchy. The aides corps consisted of three assistants who were usually close relatives to the jumbe. One of them was his deputy or second hand, the shaha; another was the mwinyiamiri, the assistant who communicated the jumbe’s decisions or wishes, similar to the function of the town crier in medieval European towns; and his minister or adviser, the waziri was the third category of official aides (W. T. Brown 1971a: 138). This leadership was drawn from a particular kin group and a common feature in Swahili towns was that this group lived together in an identified part of the town. It seems safe to assume that much of their duties in one way or another were related to two prominent features in their middleman position. One was the arrival of seaborne traders from various areas around the Indian Ocean who needed local middlemen and lodging. The second was the handling of the tens of thousands of caravan porters who arrived seasonally, stayed for some months to sell their goods, occasion-

Old Bagamoyo

•

253

ally worked on the plantations while they also tried to obtain goods for future trade in the interior of the continent, a er which they began a new caravan march for their up-country destinations. Other leaders in Bagamoyo were drawn from the religious organization within Islam as the religious functionaries; the mu’allim (pl. mu’allimun) who taught the Islamic sciences at a mosque or a madrassa. Others in this prestigious category were the learned persons in Islamic law or shari’a law, the faqih (pl. fuqaha’), that is, those knowledgeable in jurisprudence (fiqh) or “judge” and those in charge of the mosque, especially the imam or leader of the prayers and the khatib, the deliverer of the Friday prayer (Nimtz 1980: 17). It is hard to determine their role and position in the town’s life during the heyday of the caravan trade, when the Nyamwezi porters or traders visited the town. The next position in the elaborately established hierarchy was filled by the free persons who could trace their kinship relations to known ancestors. This meant that they were descendants from a free man, although the mother could have been a slave.28 This condition is captured in the term used for them, wana wa watu,29 meaning “children of people” ( Jerman 1997: 102), i.e. people with known ancestors. Their efforts to be assimilated into and accepted by the local patricians must have provoked even more reactions from the elite who, as a result of the caravan trade, found themselves under a double a ack from the plebeians’ aspirations to belong to the “civilized” section of the population and the pressure from Omani and European colonizers to relegate the old middlemen-merchants to second or third class townspeople. Until the 1920s, slaves occupied the lowest position in this hierarchy. It has been estimated that around 15–30 percent of the population in Bagamoyo were slaves, a relatively low percentage compared to, for example, Dar es Salaam (S. Fabian 2007b: 174). The slaves, however, never constituted a homogeneous or stable category, as has been explained above. In Bagamoyo, they mainly functioned as domestic slaves, day laborers, caravan porters, fishermen, plantation or agricultural slaves, and concubines. Indebted free persons could easily, as shown above, be enslaved if they failed to repay their debt while prosperous slaves, who had acquired their own wealth through the caravan trade, could buy their own freedom at their market value. Such ex-slaves could, a er conversion to Islam and a er being accepted in a dance society, acquire the status of being a mungwana, “civilized.” It is difficult to estimate their percentage of the total of Bagamoyo’s population since the border line between poor but free clients and prosperous slaves was highly permutable during the peak of the caravan trade at the end of the nineteenth century. Slaves could accumulate wealth and

254

•

Muted Memories

own their own slaves while a free person could be heavily indebted to and highly dependent on his or her patron and, eventually, be enslaved and sold as a slave. Both processes illustrate the blurred demarcation line between free and slave in East African pre-colonial societies. This status system, or variations of it, was in operation until the Arabs and Europeans entered the arena during the middle part and second half of the nineteenth century and disrupted the power balance between the Zaramo and their pazi “who never saw the sea” and the Bagamoyo majumbe “who never saw the sun.” The class structure based on wealth, consumption, and descent was completely altered by the European colonizers when they introduced race and skin color as leading categories for hierarchization. Suddenly the shirazi (and shomvi), who claimed Asian origin, became Africans, while Arabs and Indians were considered “white.” A second source of disruption was the increasing number of up-country people who came with the caravans and strove for permanent residence and citizenship in Bagamoyo. Both Jonathon Glassman (1995: 13–15) and Paul Kollman (2005: 19) interpret the clients’ expectations of their masters in terms of a moral economy. It may be assumed that the carefully negotiated and long-established relations between a minority of rich patrons and the majority of poor clients, between old and young and free and slaves were challenged from all sides when thousands of up-country porters/traders came to the coast and Arabs and Indians se led here to skim off the profits from the caravan trade. With the accelerating competition over the profit from the trade, it must have been increasingly difficult for the old patrons to meet the public expectations of generosity during the society’s public festivals, which led to increasing frustration among the clients.30 This “moral economy” probably contributed to strengthening a particular aspect of the struggle for inclusion and exclusion in a coastal town like Bagamoyo. It has been shown that the aim of clients and poor people, including slaves, was not to radically change the hierarchical system, but to be accepted and recognized as civilized townspeople (Glassman 2007: passim). Kollman adds (2005: 75) that the vision of slaves revolting was not the achievement of a “Western” notion of “freedom,” but that the patrons should fulfill their obligations in accordance with the negotiated codes in the Swahili world. In short, the social structure in a precolonial coastal trade town contained three main slots: the patricians or the elite, who were merchants acting as middlemen between the continental trading networks and those on the Indian Ocean; the free persons; and finally, the slaves. Bagamoyo, established between the eighteenth and the nineteenth centuries, also contained, probably from its establishment, a fourth category: Indian and

Old Bagamoyo

•

255

Arab traders from the Indian Ocean trading networks. These newcomers, o en under the aegis of the Zanzibari Sultanate, established themselves above the Swahili structure and marginalized the middlemen in both the coastal towns where they se led and in the caravan trade along the central routes. At the end of the century, a new category of intruders arrived in Bagamoyo with the aspiration not only to take control over the caravan trade, but to colonize the entire area. They worked out sinister plans to crush both the town and the caravan trade, which will be further explored in the last chapter.

Notes  1. The opening of the railway line between Dar es Salaam and Kigoma/Mwanza marked the end of the period when Bagamoyo flourished from the caravan trade.  2. Henry Morton Stanley (1878: 55) captured this during a visit when he noticed that the people of Bagamoyo “are accustomed to have their normally languid and peaceful life invaded and startled by the bustle of foreigners arriving by sea and from the continent, Arab traders bound for the interior and lengthy native caravans from Unyamwezi.”  3. See next chapter.  4. See also Thomas Spear (2000: 281) and John Middleton (1992) for similar conclusions.  5. The monumental tombs with high pillars within the boundary of a stone town or village are endemic to the East African Islam societies. In southern Arabia, the cemeteries are placed outside the se lements and away from the mosques ( J. de V. Allen 1981b: 312).  6. See W. T. Brown 1971a, 38.  7. The devastating effects in the western part of the Indian Ocean were the results of their “mission.” The fleet which set out from Portugal on 9 March 1500 was commissioned to secure a firm base in the East, either by negotiation or if necessary by force. Should the infidels prove unwilling either to be converted or to engage in trade, then the spiritual weapon of the Gospel was to be augmented by the carnal weapons of fire and sword. All Muslim vessels, with the exception of those of Malindi, Cochin and Cananor, were to be sized. (Strandes [1899] 1961: 34)  8. Mark Horton launched the metaphor “The Swahili Corridor” in 1987. Abdul Sheriff saw the Swahili-states along the coast as the beads in a rosary (Sheriff 1987: 8). Paul Sinclair developed the idea of the “Swahili wheel” in a book chapter from 1995.  9. Actually, one was the gi from the King of Bengal who had received it from Malindi on the Kenyan coast and the other came directly from Malindi (see Wilson 1992).

256

•

Muted Memories

10. Actually, there are four winds which give rise to four seasons in the “Swahili” calendar (Parkin 2007: S39). Two of them are monsoons and of concern for this study. 11. See also Sheriff (1987: 10) who reports that the “African coast is at the fringe of the monsoon system, the constancy of the monsoon decreases dramatically as it encounters the south-easterlies blowing towards Mozambique.” 12. “In fact,” he writes, “Kilwa Bay is located at the southernmost part which can be reached by sailing from the north with the monsoon winds” (Zhao 2012: 49). 13. For information on the Mbunga confederacy see Lorne Larson, who writes: “By 1884 a large Mbunga se lement had established itself permanently just south of the Uluguru mountains at Kisaki, and was threatening the central Bagamoyo-Tabora caravan route” (1977: 38). 14. Sheriff reports that slaves were used to dig up the copal. 15. For the role of copal exports from the Mrima coast, see also Sunseri (2007). 16. Zanzibar developed, parallel to the growth of Bagamoyo, into a mercantile empire in a globalizing economy (see Sheriff 1987). 17. Ropes, who used to pay $115 per frasila (35 kg) now had to pay $130 (S. Fabian 2007b: 197) 18. A se lement without a mosque was not considered a “town” since it lacked both ustaarabu (civilization) and utamaduni (urbanity) (Horton and Middleton 2000: 129). 19. The knowledge about the nineteenth century wa le-and-daub houses in the Swahili towns is poor. One reason is that the spectacular stone houses have caught most of the a ention; another is that the building materials in wa leand-daub houses leave few testimonies when the house is deserted or demolished. Still, the majority in the Swahili world lived in this kind of structure, since only a few could afford to build or buy stone houses (Fleisher and LaViole e 1999). 20. The term was by Burton (1859: 12) translated as “chief.” Other sources suggest that it is be er translated as “sir” or “mister” since it was used as a title of respect for the elite among the population in the societies adjacent to the coastal se lements (Owens 2012: 369). 21. Some legends also mention salt and other manufactured commodities (Owens 2012: 353). 22. Here the term shomvi refers to a political position and a title, not to an ethnic group. Walter T. Brown, quoting Velten, finds that the titles jumbe and diwani have replaced the earlier title shomvi (W. T. Brown 1971a: 139). Or actually, jumbe replaced shomvi and diwani replaced jumbe. It may therefore be assumed that the label shomvi, at the time of the caravan trade, did not designate an ethnic group but rather a political community: the people under the leadership of a shomvi. 23. From this time, bones of sharks, bonito, and tuna begin to appear in archaeological excavations of Swahili se lements, which may indicate new fishing techniques in open waters. Other findings suggest a growth of the prosperity in some Swahili se lements during the twel h and thirteenth centuries,

Old Bagamoyo

24. 25. 26.

27.

28. 29. 30.

•

257

which indicates an involvement in the overseas trade which, in turn, may explain the appearance of monumental stone buildings, as the Husuni Kubwa palace at Kilwa Kisiwani, at the end of the thirteenth century (Fleisher et al. 2015: 103). This was probably an earlier name of Bagamoyo. We see here that the terms pazi and shomvi are used for the leaders, not for specific ethnic groups. However, the fact that it was Bagamoyo’s leaders, and not the Sultan of Zanzibar, who gave the Holy Ghost Fathers the land, shows the limited control the sultan had over the societies on the mainland. All information available on the political structure in Bagamoyo suggests that the leaders were males. I therefore use the pronoun “he” or “his” when I write about the town’s leadership. Information from other places as Zanzibar and some coastal towns show that women, o en titled “Queens,” were leaders as well (see Askew 1999; Horton and Middleton 2000). Many manumi ed slaves, in a sense “free,” still o en had a liminal position as explained in chapter 6. Implicit in this term is the divide between free and slave. Slaves are sometimes called “socially dead” or non-persons without ancestors—see chapter 6. It should be noted that the term “moral economy” is here used to describe two different sets of relations. Glassman talks about the relations between patrons and clients in “Swahili society,” while Kollman talks about the relations between missionaries and baptized Africans. The new Christians entered into a relationship with the missionaries in a junior position, as “wardens” while the Holy Ghost Fathers took the position of the “trustees.” As “wardens” the newly baptized expected generosity and favors from their patrons, the Fathers. For the terms “trustee” and “wardens,” see Murray Li (2007).

10  Fluid Identities Politics of Identity in Multicultural Bagamoyo The Nyamwezi porters/traders who brought their ivory and other goods to the coastal trading towns in the 1880s stumbled into brewing pots of identity creation and communication, hierarchy-making, and claims and struggles for inclusion versus exclusion. Processes of creolization or hybridization took place when subordinated sections of the town’s population as well as newcomers from up-country communities strove to be included in the dominant group’s society and culture rather than trying to challenge it or defend and promote their own identities. The struggle for inclusion triggered new mechanisms for exclusion invented and applied by the local elites since, in the words of Helle V. Goldman (1996: 6), the “Swahili evince a deep commitment to hierarchy based on membership in ethnically defined social categories.” As will be shown below, the long-distance caravan trade not only created an economic boom but also ignited a remarkable sociocultural dynamic, which is part of Bagamoyo’s outstanding historical and cultural values almost forgo en today. Many tried to profit from what the porters brought to the coast, and to profit from what was sold to them, including women, as noticed by Mtoro (J. W. T. Allen 1981: 114): “When the Nyamwezi comes to the coast, he greatly likes to drink beer. Women make a profit out of it.” This chapter will focus on who the Nyamwezi porters/traders encountered when they reached the palms at the beach where they could lay down their hearts and burdens, and the sociocultural processes triggered by this mixture of people with different backgrounds “with the result that what constitutes

Fluid Identities •

259

the Swahili cultural and linguistic complex is always in a state of development and of flux” (Parkin 1994: 2–3). And, as in the words of Laura Fair: “Ethnic identities, like all social identities, are multiple, mutable and situational” (Fair 1994: 195). Identities have a tendency to be fluid, unstable, and relative since they are constructed by both the observer and the observed. “The identity of the Swahili,” wrote Mark Horton and John Middleton, “represents a classic case of a society constantly redefining itself, through myth and tradition, in language and dress, names and genealogy, in response to the demands of the observer—be they Asian or African traders, European or Arab overlords, or western tourists” (2000: 199).

Multicultural Bagamoyo During the first half of the nineteenth century, the Nyamwezi traders/ porters came into contact with people who o en proclaimed their own identities, but equally o en were ascribed an identity by external actors. At the beginning of the nineteenth century, their counterparts probably identified themselves as “Swahili” and “Arab”1 traders, or belonged to the local “Zarmo” population or slaves from societies both at the coast and from upland areas. During the la er half of the century they also encountered people from various parts, traditions, and religions from South Asia, Europe, and the Americas. They were all, in one way or another, involved in the caravan trade, which connected the interior of Africa with other continents and/or the struggle to conquer and colonize this part of Africa. Some in the coastal societies and their surroundings produced the food for the porters while they stayed to bargain at the coast, others extracted tolls from the caravans, still others tried to purchase their ivory and other goods, some provided the merchandise for the porters and caravan leaders when they returned to their upcountry destinations, still others financed the business of the coastal traders and some came from very far afield to follow the well-trodden caravan paths when they tried to “discover” the interior of the continent.

Intersectional Processes in Identity Formation With reference to the branding of Bagamoyo as a slave port, today’s observers may believe that the most drastic divide between people in nineteenth-century Bagamoyo was the one between those who were free and those who were slaves. The society was strictly hierarchical in terms of relations of superiority versus inferiority, where differences were cre-

260

•

Muted Memories

ated and articulated through ethnic belonging, kinship, gender, age, conformity to different canons of Islam, the time a person/family had lived in the town, wealth and degree of “civilization” versus “barbarism,” categorizations we will return to below. In many cases, the patron-client relations established through such status and power differences resembled master-slave relations. The border between the free and the un-free was not always clear in Old Bagamoyo, particularly since slaves, as already mentioned, could own slaves, take part in the caravan trade, and raise the money needed to buy their own freedom at their market value (Deutsch 2006). It seems more appropriate, on the basis of current research results, to argue that the most fundamental divide between people in and around Bagamoyo was between those who considered themselves “civilized,” possessing udu, which was to be real humans, and those they regarded as “barbarian.” This dichotomy was not unique for the Swahili world. It has been suggested that this cleavage, which also paints the barbarians as physically different from the civilized, and therefore inferior to those who regarded themselves as superior, was or still is common in world views and legends in societies south of the Sahara (Kopytoff 1989: 49–61). Some of the conditions required to aspire for inclusion in the “civilized” group were descent, place of birth and way of life, to be a Muslim, to be able to afford prestigious goods and food, and take part in the town’s socio-religious life. This categorization was subjective depending on the perspective of the town dwellers. Unfortunately, we know relatively little about how the uplanders categorized the people they met in Bagamoyo. The available sources, however, show that many up-country people, whether free or slaves, who se led more permanently on the coast tried to assimilate into the “civilized” sectors of the society. The porters who came to Bagamoyo and other coastal towns met the importance of the a achment to a specific place, which separated them as “barbarian” with an up-country residence or origin from the “civilized” coastal dwellers. They must have sensed and learned that the place, as a space filled with cultural meaning, featured in at least two fundamental manners in the creation and re-creation of identities among the inhabitants of these coastal towns. One was the importance ascribed to a place, o en putative, of origin, preferably located in the Islamic world and in Asia for elite groups and individuals; the second was the place of habitation, which meant to live in Bagamoyo in order to be a rightful citizen of the town. The toponymical identity, from your geographical place, however, was combined with the ethnonynic identity or your ethnic belonging. A Mbagamoyo, a person belonging to Bagamoyo, was also either a Zaramo, the people surrounding the town; a “Swahili”; a “Shirazi”; an “Arab”; an

Fluid Identities •

261

“Indian”; or another ethic identity. What further complicated the situation was that a person could shi identity according to the situation or claim an identity, which was accepted or refused by others.

Those Trapped into Thinking in Terms of Tribes The understanding of the cultural dynamics and creolization that took place during the decades before colonization was disturbed by the colonizers’ rigid worldview, which restricted them to think in racial and ethnic terms. Even early anthropological “studies of Africa took it as axiomatic that tribes existed; this was what anthropologists studied and what colonial government dealt with” (Caplan 2007: 306).2 The early travelers and colonizers simplified the cultural landscape, which facilitated their rule. In their efforts to demarcate clear boundaries and identify cultural divides, they blatantly disregarded the local complexities and instead applied preconceived views of stereotypical ethnic differences, which disabled the observers from seeing the cultural complexities and hybrid identities. The European colonizers came to East Africa with a cognitive map telling them that the world was chopped up into distinct territories and peoples, much like a chessboard. Their worldview was based on the nationalistic slogan one language, one culture, one people, one territory, and one political system; in the nationalistic version this implied independence, in the colonial system it meant indirect rule. No wonder the European colonizers ran into difficulties when they tried to understand East African coastal societies. There was no “Swahililand” or territory that belonged to the Swahili people. Instead, the intruding Europeans met a number of independent towns and villages with their own governing bodies, o en surrounded by other people claiming different ethnic identities. There was certainly a Swahili language, but it was spoken by many different peoples along the coast, including immigrants from the Arabic peninsula. There was a “Swahili culture” but many of the main traits of this creole culture were shared with many other groups in their surroundings and also with immigrants from the Arabic peninsula. And, as pointed out by Steven Fabian (2007b), the “Swahili” se lements harbored many other groups, like the Arabs and Indians, and in Bagamoyo also the Zaramo and the Nyamwezi. An additional flavor to the colonizers’ problem of identifying and demarcating the Swahili came, according to Steven Fabian, from their negative a itudes towards this “tribe,” which challenged their preconceived view of the world; “because the Swahili were largely an urban population of mixed racial and ethnic heritage, it created problems of governance for the British based on their conceptions of ‘tribal’ rule” (S. Fabian 2007b:

262

•

Muted Memories

13).3 At this time the acceptance for and interest in creolization and hybridization was minimal, since the “pure forms” in culture and nature were acclaimed and appreciated. This made it difficult for the colonizers to understand what they encountered in Bagamoyo. Considering all this, it appears futile to try to grasp the sociocultural and political dynamics in Bagamoyo through the focus on ethnic belonging and ethnic groups. When the processes of hierarchization and the slave-based economy are taken into consideration, it seems more promising to approach the dynamics that the caravan trade brought into the town from a class perspective (see also Fahy Bryceson 2009; Jerman 1997: 101), in combination with approaches based on the politics of identity and agency. This is not, however, neglecting the importance of ethnic (or kabila) belonging of the town dwellers or a support for the argument that the colonizers brought ethnicity and thinking in tribal terms to African soil. The argument is only that, while the Europeans saw the area through their tribal grid, other splits than just those based on ethnic belonging could have been employed in order to be er understand the struggle for inclusion and exclusion, status and power, which accompanied the caravan trade in a town like Bagamoyo. The complexity of the processes of identity formation and the problems they caused external observers is captured by David Parkin and François Constantin with the following words: “At one level this is a self-sealed, indigenous system of stratification, whose private intricacies, prejudices, resentments and judgments are specific to and perhaps only ever really understood by the Swahili” (Parkin and Constantin 1989: 143). Our understanding of the local life before and during the first decades of colonization has been influenced and distorted by the early travelers and colonizers’ constant a empts to press the local population into stable ethnic or tribal categories.4 Their worldview is epitomized in the tribal maps they produced with their clear-cut boundaries between the ethnic groups. The groups lived, it was believed, side-by-side on a horizontal map, while the concept of race was used to order the groups in vertical hierarchies. This is a worldview and approach ill-fi ed for any a empt to understand and present a place like Bagamoyo. Hence, a different and more emic approach will be applied here that draws on the fundamental dualistic categorization of being either a civilized town dweller or a barbarian from the rural areas, and the strategies for inclusion and exclusion used by different classes that resulted in a categorization that was fluid as were other classification schemes in Old Bagamoyo. This split was geographical along an east–west divide. People at the coast, free or enslaved, saw themselves as civilized in contrast to the peo-

Fluid Identities •

263

ple from up-country societies to the west. Since this dichotomization depended on one’s point of reference, the split was not fixed but fluid. From a reference point further east, Zanzibar, even the mainland coast and its people were, and sometimes still are referred to as “barbaric,” while people from Zanzibar, free or enslaved, saw themselves as civilized. It may also be hypothesized that the name Nyamwezi, meaning, “people from the west,” was a synonym for the term washenzi, a term with an Arabic root, meaning barbarian. In a previous chapter it was shown that up-country traders tried to establish joking relationships with a trader or a trader’s representative in Bagamoyo, who was the one who guided the trader and his or her goods to a merchant. This partner was most certainly what the porter called a Mswahili, a person from the coast, a mwaraabu an “Arab-like” person, a mwungwana, a term of Bantu origin that denoted a civilized person living in a town, or a mwenjeyi, a person who saw himself as an owner of the town. It is probable that this person called himself a shirazi with a putative origin from Persia and, if a leader of the town also a shomvi. None of these identifications are purely ethnic denominations. Most of these labels of identification, however, have been used by travelers and colonizers, and in the literature on the East African coast, to delineate and categorize what were believed to be ethnic identities. Very li le, unfortunately, is known about how people categorized themselves, except for the views of the elite, which was o en reproduced as “Arab” in the travelogues and other texts from Western visitors. This makes it difficult to refer to some of the identities without using these o en very broad, sometimes misleading and overlapping labels. The aim of the following pages will be neither to present each group’s specific cultural traits, nor to provide a cultural catalog. Rather, the focus is on the sociocultural interaction between those involved, who came together through the caravan trade in a town like Bagamoyo. In particular, the focus will be on how the wealth generated through the booming caravan trade affected the fluidity of these labels, processes of cultural hybridization, politics of identities, and strategies for inclusion and exclusion in identity-making processes.

The Owners of Bagamoyo Bagamoyo developed as a response to the establishment and development of the caravan trade along the central routes. The growth of trade a racted an increasing number of foreigners to the scene, which meant that the original middlemen-merchants became, as already noted, increasingly marginalized. The first part of this chapter will present some of these groups.

264

•

Muted Memories

The processes of marginalization and hybridization may have been particularly prominent in Bagamoyo for two reasons: first, the young age of the town at the time of the booming caravan trade meant that the tight social ties, which forged together a solid kernel of “owners” of the town, as in for example Mombasa, Lamu, or Kilwa, were less cemented in Bagamoyo, and second, because of the town’s proximity to Zanzibar, the hub in the Western Indian Ocean trading networks. The second part of this chapter will give a brief presentation of the “Indians” and “Arabs” who took over much of the business in Bagamoyo. The intrusion of the German colonial power, which eventually terminated the caravan trade with the move of the colonial capital from Bagamoyo to Dar es Salaam and the construction of a railway from the coast to the Great Lakes area, will be examined in the last chapter.

Those Called the “Swahili” Bagamoyo belongs to what has been called “The Swahili Corridor” (Horton 1987), the coastal area stretching from central Somalia to southern Mozambique, including the islands along the coast. The academic world has for a long time been divided regarding the interpretation of the origin and development of the Swahili language (Kiswahili) and the culture and people called the “Swahili.” The debates o en revolved around the politically sensitive issue concerning the extent to which a foreign, mainly Arabic, influence shaped the development of the Swahili language and culture. The name comes from Arabic Suahil or Sawahili (sometimes Suwaail) for “coast,” “edge,” or “border”5 and was used by the famous traveler Ibn Ba uta from Morocco, who visited the East Coast of Africa early during the fourteenth century. When he had passed Mogadishu he came to the “Sawahil country” (Gibb 1939: 112, quoted in Chami 2002c: 3), or Swahilini, in the land of the Swahili (Spear 1984: 301). It has been suggested that the term Sawahil or Sawahili emerged as a label for a group of people at the same time as the terms Shirazi and Arabu (see below) came into use among people who met and mingled along the coast. A similar association with locality is found in early Portuguese reports from the sixteenth century, which refers to the people of Kilwa as “Suaili,” in contrast to the identity “Arab,” who are “those whose home was, or has been, Arabia” (J. de V. Allen 1993: 3). In the hierarchical order, reproduced by external European and US observers of the three partly overlapping groups, the Swahili occupied the lowest social stratum, since they did not claim an origin in either Persia or on the Arabic peninsula. They were, however, as free people, above the slaves.

Fluid Identities •

265

In the search for common characteristics of the “Swahili,” historians, archaeologists, linguistics, architects, anthropologists, among others, usually identify three criteria to identify “Swahili” people, culture, and society. These are the geographical area they inhabit, the corridor along the East African coast; the language they speak, Kiswahili; and their religion, which is the Shafi’i canon of Sunni Islam. It was not always fully understood that a part of this population, the merchants, could live a mobile life as they moved back and forth between their places of residence in the coastal towns and trading posts in the interior of the continent. The “Swahili people” never developed a unified political system and their se lement pa erns were far from uniform.6 Some lived in rural villages, many depended on agriculture and fishing, while others lived in urban milieus in the trade towns and were traders or artisans. Both forms of se lement were referred to as “towns” since they contained at least one mosque. The term indicated, during the nineteenth century, aspirations to live the “civilized” way of life lived in a “Swahili” town, a topic we will have reasons to return to below. They also spoke a common language, Kiswahili, among other languages they probably spoke. For small-scale traders or middlemen it was an advantage to master more than one language. It is o en said that Kiswahili is a mixture of words from the Arabic and Bantu languages but with the grammar of Bantu languages. The incorporation of Arabic vocabulary, however, was relatively late and may be related to the expanding trading networks in the western parts of the Indian Ocean, beginning at the end of the eighteenth century, when the Arabic traders and plantation owners received a hegemonic position over the trade in the Indian Ocean (Spear 2000: 259; Chami 2002b). Early interpretations of the origin of the Swahili language and culture were influenced by colonial and partly racist perspectives. It was held for a long time that the language and culture were developed through an Arabic influence in what Chami (2002c: 2) calls the “Orientalist school of thought.” Their perspective was essentially diffusionist “in assuming that cultural innovations and historical development in Africa only could have come from elsewhere” (Spear 1984: 291). It was also racist since Africans were deemed incapable of developing what many Europeans saw as a cultural refinement in the Swahili culture such as stone buildings, carved doors and frames, art, poetry, dress styles and more. It was assumed that influences behind this development must have come through external sources, particularly “Arabs” who seasonally or permanently lived on the East African coast. Earlier researchers, who o en fell into the trap of thinking in terms of tribes, had difficulties grasping what Beverly Brown and Walter T. Brown

266

•

Muted Memories

call Swahili culture’s greatest strength: its openness. “A meeting of old lifestyles and a reaching for new answers lay at the vital core of community life in East African trade towns” (B. Brown and W. T. Brown 1976: 198). The creolizing character of this culture also became a headache for the colonial administration since it was impossible to demarcate Swahililand. The “Swahili” corridor fi ed, very badly into the colonizers’ worldview of monolithic tribal territories.

Reasons to Be a “Swahili” Two main reasons may be traced behind the a ractiveness at the turn of the nineteenth century, when Bagamoyo came into existence and began to grow, to adopt the Swahili identity. One was directly connected with the caravan trade and the processes of creolization it triggered. The other was a result of the abolition of the slave trade and the freeing of slaves. Most freed women and men, many of whom had been brought from the interior of the continent to the coast, began to identify themselves as “Swahili,” which at that time signaled that a person was a free Muslim member of the coastal society (Fair 1994: 208). The sociocultural processes developed through the abolition of slavery must have been particularly prominent on Zanzibar where roughly three-quarters of the island’s population at the time of abolition were slaves. The fluidity of the “Swahili” identity made it a ractive to the slaves liberated through the banning of the slave trade and slavery. In contrast to their former slave identity, without connotations to an ethnic identity in a society otherwise populated by free people who claimed various ethnic identities, the “Swahili” identity was open for appropriation. The “Swahili” identity had, before it was appropriated by freed slaves, been a ractive to people from the interior of the continent who were brought to the coast through the caravan trade. It represented coastal urban life, which was associated with a high degree of civilization and refinement. It seems to have been a common strategy for people from up-country societies to call themselves “Swahili” when they came to the coast through the caravan trade and se led there. They had at least two reasons for this: in the dominant worldview on the coast, this meant both an a achment to the “civilized” world and people on the coast, and at the same time, dissociation from their “barbarian” up-country background. If the view is shi ed further west, people along the central caravan route could also identify themselves as “Swahili,” particularly if they were involved in the caravan trade. An example is the “Swahili” people in Ujiji town at Lake Tanganyika, many of whom originally were Wamanyema from central Congo (see Arens 1975: 428). Through their involvement in

Fluid Identities •

267

the caravan trade and residence in Ujiji, which was characterized by visitors as a “Swahili town,” they themselves also became swahilisized. Another example of the strategic use of the label “Swahili” may be drawn from William Arens’ study, a decade a er Tanzania’s independence, of the multicultural society Mto wa Mbu at the foothills of the Ngorongoro Crater. Residents of Mto wa Mbu, coming from many different parts of Tanzania referred to themselves as “Swahili” and contrasted their way of life with that of the “tribesmen” surrounding them. Swahili-ness stood for “sophistication and modernity. They consider themselves to be bearers of the new culture of independence” (Arens 1975: 435). Here, to be “Swahili” has become a collective identity very similar to the one some porters, almost one hundred years before, strived for when they tried to be swahilisized or civilized and also identified themselves as “Swahili.” It is no surprise that the “Swahili” identity became increasingly popular a er the independence of both Tanzania and Kenya, since the “Africanization” that took place also affected this identity. It became politically opportune to display “a specific ancestry with their own clans and lineages, share a common language and literature, live in the same geographical area and have traditions which can distinguish them from others” (Caplan 2007: 309). The political process of the “Africanization” of many identities appeared with Independence in Tanzania and Kenya, and illustrates how identities may change in order to meet new demands regarding political correctness. As captured by Patricia Ann Caplan: “It is paradoxical that people who have spent much of their history seeking to be identified as ‘Arabs’ or ‘Persians’ . . . should now be emphasizing not only their African roots, but even their claims to ‘tribal’ status” (Caplan 2007: 312). This, however, happened a century a er the boom of the caravan trade.

Reasons Not to Be a “Swahili” A peculiar feature of the term Swahili, as noticed by Steven Fabian (2007b) and others, is that few in the coastal se lements today use it to identify themselves, since the term is essentially “an externally applied appellation” (S. Fabian 2007b: 8; see also Goldman 1996: 6; Spear 2000). Today, “Swahili” as a term for self-reference remains largely out of use for most of the inhabitants of the East African coast. In the Tanzanian coastal societies, it is even considered a derogatory term, implying a thief or somebody who is not to be trusted. John Middleton (1992) adds: “It has been held that there is no society that can be called Swahili, there being only coastal se lements whose population has been called so by others, and in which those of higher rank have referred to those of lower rank as Swahili while refusing the name for themselves” (Middleton 1992: 1).

268

•

Muted Memories

As thousands of former slaves began to redefine themselves as Swahili, writes Fair (1994: 190) “the meaning of being ‘Swahili’ began to change.” Positive connotations associated with the status of a free person began to change in favor of a pejorative connotation of being a former slave. The “Swahili” identity soon lost its popularity, which is seen in population censuses. In the 1920s, roughly 34,000 persons on Zanzibar identified themselves as “Swahili” as compared with 2,000 by 1931, a decline of 86 percent (Fair 1994: 190). The number continued to decline to only 290 in the 1948 Census (Arens 1975: 434). Furthermore, it was noticed fi y years ago that the term, particularly on Zanzibar, was used “as a euphemism for descendants of slaves from inland areas” (Prins 1967: 11), which made the label even more una ractive for self-identification, and many in the coastal se ings “do not particularly like to call themselves, or be called, Swahili” (Hino 1980: 111).7 François Constantin (1989) finds that: the Swahili refused to declare themselves as such as long as they were despised by the dominant Arab group (who regarded them as pretentious subordinates) and by the subordinate Africans (who thought of them as traitors), and so were unable to act collectively as a politically conscious group. It is also why the Swahili, despite figuring so much in writings and discussions, numbered so few in census returns. (Constantin 1989: 150)

Thus, there is a general consensus among researchers that the term gradually acquired an increasingly negative value. It may be assumed that this occurred when European travelers, tainted by the status scheme of the “Arabs” they met on Zanzibar, on their way to mainland Africa developed their prerogatives against the “Swahilis” who, in Zanzibari slave society, were either slaves or descendants from people from mainland Africa, which implied being “barbarian.” They constituted the lowest class in a society ruled by the “Arabs” (Arens 1975: 432). The “Swahili” were degraded to lower-class citizens or a slave-like status through their association with mainland Africa. The wealth of the early “Swahili” towns was not created through control over a certain area or population and the ability that such control gave to extract tributes and levy taxes. Rather, as middlemen in the interface between the continental and Indian Oceanic trade networks, their wealth was based on commissions and profit (Middleton 1992: 20), the conversion of values between different trading networks, and the customization of goods that passed through their houses and hands. There was no need for the ruling elite to establish control over a geographical territory and its people and recourses beyond the stone walls of their towns and houses. But the colonial image of Africa held by the Germans and the British was of a continent made up of distinct ethnic groups, each one with its lan-

Fluid Identities •

269

guage, culture, political organization, and territory. Indirect rule was built on indigenous political organization linked together with the colonial administration. Nomads, be they pastoralists or hunter-gatherers on land, or mercantile societies that depended on the flow of people and goods on the sea, or along caravan routes like the “Swahili,” provided a headache and threat to their established image. A multicultural polyglot society like colonial Bagamoyo therefore posed a challenge to the established view that partly came from the tendency among the colonizers to use the term “Swahili” for a narrow segment of the people who lived in the coastal towns and were supposed to be of mixed “African” and “Arabic” origins, but not to use the term for the majority of the townspeople who were regarded to be of “Bantu” origin, nor for people with an “Indian” or “Arab” origin living in the “Swahili” se lements. The problem must have been particularly prominent in Bagamoyo with its rapid a raction of people from different origins and its high number of immigrants from India.

Those Who Called Themselves the “Shirazi” Still a supposedly ethnonymic identity that has caused much confusion among colonial administrators, researchers, and visitors is the term “Shirazi.” An essential feature of both the “Shirazi” and the “Swahili” identity was that they, during the heyday of the caravan trade and a erward, have functioned as self-identification in the creation of “we” and “them” categories. The shi of the “Swahili” identity from free person to former slave made it less appetizing for coastal people to use this label to identify themselves. This was a process that continued into the twentieth century since “in Zanzibar and Pemba, for instance, it seems that people who once identified themselves as Swahili in about 1930 began to call themselves Shirazi” (Nimtz 1980: 30), a process which culminated with the establishment of the Shirazi Association during the 1940s on Pemba. In the multicultural towns on the African mainland, the patricians in the Swahili societies used to call themselves “Shirazi” to suggest an origin from Shiraz in Persia. A long-lived myth from a number of chronicles from the fi eenth century in some of the Swahili towns has survived as oral tradition. Some were recorded by the early colonialists in, for example, Kilwa, Pate, and Vumba Kuu. “Rather than reflecting a historical narrative to be read literally,” writes Stephanie Wynne-Jones (2010b: 412), they “are more o en now constructed as thematic charters for the towns establishing longevity for issues of importance for urban Swahilis of the moment.” The texts tell about the arrival of Persian princes to the East African coast, who each established himself as ruler through marriage with the daugh-

270

•

Muted Memories

ters of the chiefs in the areas they came to (see, for example, Spear 2000: 259–60).8 It may be assumed that these legends were not only of actual extra-African origin, but also a result of a policy of identity, which built on a manipulation of history where descent from the Arab-Persian world was prestigious.9 Persian traders came to the East African coast during the early development of the Indian Ocean trading networks. The Persian conquest of Yemen (570 CE) and Egypt (616 CE) opened up increased trade with East Africa in gold, ivory, mangrove poles for house construction, amber and more, and this trade stimulated the development of the first trade towns on the East African coast between the eighth and tenth centuries (Spear 2000: 280). The Persians’ empire had, however, declined before the Portuguese entered the scene early in the sixteenth century. Today there are few Persian cultural elements detectible in Swahili culture, except for the stories about the Persian princes in the chronicles mentioned above. Another argument for the claim of an extra-African origin of the segment of the coastal population that identified itself as the “Shirazi” concerns the semantic similarities between the ethnic denomination “Shirazi” and the geographical name for a land-locked place 200 km from the coast in Persia/Iran—Shiraz. The issue here is not if this alleged Persian origin is true or not, since there is ample evidence that people from Persia have been on the coast during the last millennium. Rather, the focus is on what such legends tell us about the social dynamics and politics of identities in these coastal se ings, and the character of the elements used to claim exclusivity and prestige. This association with a place in Persia was a local adaptation among some segments on the East African coast of a practice common in the Arabic world to adopt the name of a prestigious place as one’s family name. This practice, called nisbas (Spear 2000: 259) was a common strategy to create identity among the Arabs from Oman who came to the east coast of Africa during the nineteenth century, a tradition that resembles the strategy among the aristocrats in Europe to ennoble themselves by adding a proposition to their names. It is here argued that the strategic toponymical self-identification was part of the status competition game played in the multicultural coastal trading towns. Deborah Fahy Bryceson (2009) suggests that this putative claim of Persian origin equals the putative claims of European royal and aristocratic families that they differ from the rest of the population by their blue blood. There is another stubborn feature in processes of identity-making that also pops up here. A common theme in many political ideologies explains that the rulers are of a foreign origin. “Such claims are recognizable,” writes Jonathon Glassman, “as variants of a deep political tradition, wide-

Fluid Identities •

271

spread among speakers of Bantu languages, whereby leading families preserved myths linking them to exogenous conqueror-heroes who had introduced the technologies or forms of social organization by which civilized life was defined” (2004: 735). The Nyamwezi-Sukuma traditions may illustrate this, since they “tell the story of extraordinary strangers, coming from other places and establishing their rule over the autochthons, not through belligerent conquest, but by means of guile, cunning and magical powers” (Brandström 1990: 181). Immigrants from Asia who came as cultural heroes to the East African coast brought Islam, the art to construct mosques, elaborate dress codes, and other prestigious goods. The messages in the legends of an exogenous conqueror-hero are intertwined with another legend according to which civilization comes from abroad. Civilization is, as in the dichotomy civilization-barbarism along the East African coast and its hinterland, o en associated with the Arabian Peninsula and Islam. Before and during colonial times die-hard myths spread among both the local elite and the colonizers in the coastal societies that people on mainland Africa, the Africans, were not capable of developing civilization. For the colonizers, the Arab-Persian heritage was “more easily reconciled with colonial assumptions than if these urban people had claimed full-blood African descent” (Wynne-Jones 2010b: 411). Civilization had to be brought from abroad. This worldview, present among both the local elite and the European colonizers, could even find a defense for the “Arabs’” enslavement of the “Africans” since this helped the la er to become “civilized” (see also chapter 6). In the intellectual circles on Zanzibar during the 1930s this argument was taken to its furthest limits in the dichotomy between islanders, the people from Zanzibar and mainlanders, where the original islanders, the Hadimu, who were conquered by the “Arabs,” “had become ‘sufficiently Arabianized’ to have lost most of the qualities of African barbarism, mainlanders are but ‘primitive natives . . . whose culture and history are in process of formation only now’” (Glassman 2004: 746).10 The vibrant sociocultural and economic processes in multicultural Bagamoyo, triggered by the growing caravan trade, created fertile ground for processes of inclusion and exclusion. The “Shirazi,” in Spear’s view (1984: 299–300) developed less as “a people than as a historical phenomenon as people along the coast traded, se led, interacted, and developed more complex social structures to form the distinctive culture that we know as Swahili today. The Shirazi are the prototypical Swahili, conceived in the intercourse between different peoples and economies, born into a society in tension, and raised in inequality.” Other cultural and identity markers usually communicated at the end of the nineteenth century by the “Shirazi” were that they were Muslims

272

•

Muted Memories

and urbanites who claimed to live a civilized life style called ustaarabu.11 Ustaarabu, meaning “Arabic-like,” style was, according to them, the watershed between civilized people from the coast and up-country barbaric pagans and slaves (Glassman 1995: 117). That is why they were the rightful “owners” and citizens of their coastal town, the wenyeji, still another self-denomination used to express inclusion and exclusion while immigrants from the interior or surrounding societies were wageni meaning “strangers” or “guests.” The legends nurtured by members of the coastal elite not only functioned to create a boundary between themselves and other people in the societies where they lived. There are reasons to assume that the popularity of the “Shirazi” identity during the end of the nineteenth century also was a reaction among the owners of the Swahili towns against the upstarts who arrived from the Arabic peninsula (Glassman 2004: 736). The gradual marginalization of the middlemen-merchants may have increased the attractiveness of legends, which could be used as evidence of an origin in a prestigious Asian area. It was shown above that the fluidity and strategic rejection or adoption of the “Swahili” identity continued a er the era of the caravan trade. The same applies to the “Shirazi” identity under the colonial era when the British colonialists came up with food-rationing formulas on Zanzibar during World War II, which were based on the differentiation between the “races,” “Europeans,” “Asians,” and “Africans.” “Europeans” were qualified to receive meat, bread, rice, sugar, and other items in high demand. “Asians,” from Arabia or the Indian subcontinent were qualified for rice, a status food in Swahili society. “Africans” or “Swahilis” could have maize meal and beans (Mbwiliza 2000: 4–5), the food distributed as “posho,” or daily rationing, to the porters or caravan slaves. “Under colonialism,” writes de Vere Allen (1981c: 223) “all those who called themselves Swahili were systematically discriminated against in terms of education, participation in political process, labor laws, wages in government service, food rationing in wartime, and so on”. The colonial racial policy and ideology made it preferable and profitable to claim the “Shirazi” identity, which was believed to be of Asian origin. It should be noted that the label “Shirazi” in mainland coastal societies, in contrast to the label “Swahili,” which is more open for usurpation even by people from up-country societies, was mainly used by people originating from the coastal societies. Glassman hypothesizes that the term became more popular when the patrician families in the coastal se lement came under increased pressure from two opposite directions—upland newcomers in the trade caravans who sought membership in the coastal communities and pressure from the Arabs and especially the sultanate

Fluid Identities •

273

in Oman, later on Zanzibar, which sought to extend its control over the mainland communities (Glassman 1995: 2–4). In this hybridization process, the owners of the towns, the wenyeji, used “many idioms with which to distance themselves from those, whom they regarded as socially inferior, who, although not wenyeji, are nonetheless members of Swahili society” (Middleton 1992: 1).

Those Who Called Themselves the “Shomvi” and “Shaha” There are two other terms used for some of Bagamoyo’s elite: “Shomvi” and “Shaha.” The term “Shomvi” is associated with the legend of the arrival during the eighteenth century of a group of “Shomvi” from the coastal town Barawa of present-day Somalia, which may be another example of the conqueror/cultural hero as an immigrating stranger. Other sources say that they came from Kaole to the south and se led in Bagamoyo (W. T. Brown 1970: 72). It has not yet been established if the “Shomvi” was an ethnic group who expelled, conquered, or mingled with the original inhabitants of the area or if it was a new category of incoming leaders who gained power in, for example, Bagamoyo (S. Fabian 2007b: 46). Geoffrey Ross Owens concludes that “there has been considerable confusion about what the name Shomvi referred to” (Owens 2006: 718). Some of present-day Shomvi descendants use the term as an ethnonym to present themselves as an elite group with a putative origin in Yemen or Persia. The term appears in other occasions, as in inscriptions on graves in, for example, Kunduchi north of Dar es Salaam as a title or rank. Other sources combine the two interpretations and suggest that the term may best be understood both as “a claim to status and title and as membership in a descent group” (Brennan, Burton, and Lawi 2007: 14). It was shown in another chapter that the term “Shomvi” was used to designate the leaders of Bagamoyo during the nineteenth century. Some of the sources available indicate that it was a political title and it is here proposed that the term “Shomvi” was used for these leaders, and in extension, also for their dependents, who were tied to their rulers through patron-client relationships. The term “Shaha” is as elusive as the term “Shomvi.” Some sources (W. T. Brown 1971a; Prins 1967, 96) see the “Shaha” as the “jumbe’s double,” as one of the leaders of a coastal village or town. This relates well with the translation of Shaha with “First Minister” (see Owens 2006: 743) and “deputy” or “second hand” of the political leader in Bagamoyo, the jumbe (W. T. Brown 1971a: 137f.) This indicates that the “Shaha” was a title holder in the hierarchized coastal community. One interpretation of the

274

•

Muted Memories

terms “Shomvi” and “Shaha” is that they both referred to elite positions in these societies. The oral tradition that explains the differences between the “Shomvi,” “Shirazi,” and “Shaha” follow one of the major fault lines in the hierarchization and segregation of the coastal societies. Most of the “Shomvi” claim an origin in Hadramawt on the Arabic peninsula and see themselves as “Arabs.” Many believe that they first migrated to Barwa or Brava on the Somali coast and then gradually infiltrated the coastal areas further south (Owens 2006: 719). They gradually came to constitute one of the elite groups in the societies between Bagamoyo and Dar es Salaam. Legends explain that they reached agreements with the original owners of the land, which made it possible for them to stay in the area. The “Shaha,” on their hand, were intimately intertwined with the “Shomvi,” but performed a separate political function. The “Shaha” portray themselves as the original owners of the land.12 The “Shaha,” in contrast to the “Shomvi” who see themselves as immigrants coming with civilization, cultivate the idea of being those who originally established the coastal se lements since “they have always lived there”; they are “wenyewe wenye mji wao (first people, with their own se lement)” (Owens 2006: 739). Two fundamental principles in the world view of the people in the coastal towns and villages here collide with each other. One builds on the cultural heroes theme that o en inflates the importance of an origin external to Africa. The other emphasizes the importance of being the “original owner” of the place. The two themes seem to have established equilibrium in the oral traditions nurtured by the “Shomvi” and the “Shaha.” The “Shomvi,” the new leader with his followers ruled, while the “Shaha,” with his followers, became the deputy but received an annual compensation from the immigrants called the kanda la pazi. The political and socioeconomic dynamics triggered by the booming caravan trade in nineteenth-century Bagamoyo could be summarized in the following way: the “Shirazi” and “Shomvi” tried to distinguish themselves from other coastal dwellers by claiming a putative Asian origin while, at the same time, the “Shaha” emphasized their originality while the “Swahili” increased in number as newly arrived “barbarian” porters arrived with the caravans and freed slaves tried to assimilate into the urban life. The “Shirazi” and “Shomvi” were the political and certainly also the economic elite and leaders in the town before the “Indians” and “Arabs,” supported by the Zanzibari Sultanate, took over the control of the caravan trade. And, regardless of the truth in the legends and claims of an origin outside of Africa, many members of the groups who made such claims certainly believed them.

Fluid Identities •

275

Immigrants from the Northeast The booming caravan trade also a racted actors with different origins outside East Africa, first from Asia but later also from Europe and the United States. Backed by the sultanate on Zanzibar, the “Indians” managed to take over much of the economic power in the caravan trade in Bagamoyo.

Those Called the “Indians” Another category of people the porters met in Bagamoyo was the people from South Asia, particularly India, who were divided into different segments. Already during the 1870s Bagamoyo had the largest “Indian” population on mainland East Africa. It has been estimated that they numbered between 800 to 1,000 persons at the time of the German colonization, at the end of the nineteenth century. One reason was that many “Indians” who lived in Kilwa regrouped in Bagamoyo a er the 1873 ZanzibarBritish agreement on the abolition of the slave trade, a er which they could no longer as British citizens take part in the slave trade at Kilwa, the most important slave port on this part of the East African coast. Instead, they moved to Bagamoyo where other trade goods dominated the caravan trade. Another reason was that the Sultan of Zanzibar actively encouraged Indian financiers and traders to establish themselves both on Zanzibar and in Bagamoyo. They featured in the lives of Nyamwezi traders and porters in at least three different capacities. They bought the ivory and other goods, which the porters had carried to the coast from the interior of the continent. They were the traders who provided the porters, and also large-scale caravan traders, with the goods they brought back to the up-country customers. These wealthy “Indians,” as rich traders and money lenders became a stereotype produced and reproduced in many travelogues wri en by visiting European and American travelers. The third class of “Indians,” which is less well known and documented, worked as artisans and handymen who serviced and repaired all sorts of items or produced goods. They worked as smiths, carpenters, tailors, leather workers, and po ers in the societies that grew up around the caravan trade. This is o en a hidden category of “Indians” who came to this part of East Africa; poor people, sometimes with low-caste backgrounds who carried out manual work for a living. The focus here will be on their two first roles or capacities. As skilled businessmen, the “Indians” o en took advantage of their financial position when they extended loans or advanced payments both to caravan

276

•

Muted Memories

traders and to porters, but also to local middlemen-merchants. Many porters o en found themselves trapped by the interest on the loans granted by which the financiers could skim off the profits the porters hoped to make. In this sense, there was li le difference between the slave porter who was rented out by his owner to a caravanner and who had to give some or all of his salary to his owner, and an indebted but free porter. Some of the rebels in the 1888–89 revolt were indebted porters who refused to be turned into slaves as was explained in a previous chapter. Mtoro (J. W. T. Allen 1981: 158) informs us about the relationships between the local merchants-middlemen and the Indian traders. “When a Swahili plans a journey,” writes Mtoro, “he goes to talk to his Indian merchant, saying I have come to you, Mukki, to borrow trade goods for a journey.” A er extended negotiations the Swahili caravaneer is allowed to ask for the merchandise he wants to borrow: “twenty bolts of Bombay, thirty bolts of sun, fi een bolts of majigam, and ten bolts of gami. All these are varieties of white cloth. Then, ‘Cloth for turbans.’ Kreati, buraa, rehani, sturbadi, barawaji, kikoi mzinga, pasua moyo” (J. W. T. Allen 1981: 159). Many traders came to East Africa from different parts of the world to trade on the ivory, copal, and textile markets. No one, however, seem to have known how to exploit the caravan trade be er than the “Indian” traders who had access to the Indian textile and hardware production that contributed to providing the Indian traders with a competitive advantage over traders from other parts of the world (Machado 2009). The “Indian” trading houses were well stocked with goods from India but also from other parts of the world and could provide the goods on demand. This gave them a monopolistic position in the caravan trade, as noticed by a contemporary visitor: “it is hardly an exaggeration to say that all trade passes through Indian hands; African, Arab, and European, all use the Indian agent or Banian to manage the details of buying and selling, and without the intervention of an Indian, either capitalist or pe y trader, very li le business is done” (Protocols and General Act of the West African Conference, C. 4361, 1885: 38, as quoted in Woulfin 2011: 59–60). The people from South Asia, however, never formed a uniform group but belonged to different religious traditions, came from different parts of the Indian subcontinent, and o en established themselves either as rich traders and financiers or as poor artisans and laborers. This means the label “Indian” is of li le relevance or value since it only carries the information that a person or his or her ancestors came from the British colony of India. Today that colony constitutes three independent states; Pakistan, India, and Bangladesh. Furthermore, some were Hindus, during the time of colonization called Banians or Banyans, who through religious restrictions were not allowed

Fluid Identities •

277

to bring their wives to East Africa. Children born from a relation between a Hindu father and an African mother were called “chotaras” or “halfcaste” and were rarely accepted in the way children from “Arab” fathers and African mothers were (Fair 1994: 206). The number of the Hindus in Bagamoyo is not known, but they were numerous enough to establish their own Hindu temple in town at the end of the nineteenth century. Others were Christians, like the Goans, Catholics from the west coast of India. Still others belonged to different Muslim canons as the Sunni Menons or Shiite Isma’ilis or Bohras, a Gujarati-speaking group. The Isma’ilis or Khojas, as they also were called at the time of colonization, was the largest “Indian” group in Bagamoyo (W. T. Brown 1970: 77). Their first house of prayer in Bagamoyo, the Jamatkhana or Jamàatkhana, where the believers in the jammat congregation could gather, was built during the 1880s and is still in use. The town was also the first place the Aga Khan visited on mainland Africa in 1899 when he toured the region. The most prominent trader in Bagamoyo Sewa Poroo Haji, who established the first multiracial school on the coast, constructed wells in town, let the Sultan of Zanzibar rent the fort for his representative (liwali) and, finally, donated the building to the German colonizers, belonged to this religious group. Sewa Haji Paroo was also a prominent provider of porters to the “explorers” who set up their caravans in Bagamoyo, porters who were probably his own slaves.

Those Called the “Arabs” Identities are o en as much claimed by appropriators as they are constructed by people in their surroundings. It was shown above that the term “Swahili,” occasionally designating a person of slave origin, was used by people who lacked this derogative ancestry. The same holds true for the label “WaArabu” or “Arabic people,” which was not only used by people claiming a specific identity but also by Kiswahili-speaking people along the coast to designate people who claimed an origin from the Arabic Peninsula. “The WaArabu, however,” Marjorie Franken writes, “were born in Africa, do not speak Arabic, have more ancestors of African than extra-African origin” (Franken 1991: 156). The confusion about the “Arabic” identity was to a large extent created by early European and American travelers. They assumed, regarding the urban coastal culture they met when they came to the East African societies, that it must have been imported from other societies around the western Indian Ocean, and particularly from the Arabic Peninsula. Their o en racist ideology, which made them believe in pure races or “pedigree” caused them considerable confusion when people with obvious African

278

•

Muted Memories

“racial” features presented themselves as “Arabs” (for the frequent comments of the “degeneration of the Arabs” in their texts, see Fair 1994: 198). The “Arabs,” in contrast to the “Indians,” were o en caravan traders who traveled up and down along the central caravan routes. They have come to the East African coast since the eighth century in different waves with the help of the monsoon winds blowing between the East African coast and the Arabian Peninsula. Most of them originally se led in the coastal areas, but some also in the trading posts along the caravan routes when the caravan trade grew. The “Arabs” had their own internal categorization mainly based on when they came to East Africa, and depending on where on the Arab peninsula they originated from and the canon of Islam they adhered to. A clearly demarcated category is the Sharifs or Macharifu in Kiswahili, who claim descent from Prophet Mohammad and therefore are particularly favored by Allah. Some came to East Africa with the first wave of immigrants from Arabia, while others have followed during the centuries. They o en se led in already established Muslim communities where their relations with the prophet rendered them a privileged position as to be exempted from taxes, since they “were regarded by contemporaries as highly desirable additions to Islamized societies” (Martin 1971: 529). This is evidence of a strictly hierarchical religious order within the category “Arabs.” This makes this category equally imprecise as the term “Indian” presented above. There is, however, a difference between these two presumably ethnic designations. The “Arabic” identity has been much more porous and inclusive and many on the East African coast have been “Arabized” in their struggle for prestige and power. But the opposite process has also worked since many “Arabs” have been integrated in the local coastal communities. Neither of these two processes has interfered in the same way or degree with the various “Indian” identities, which in this sense seem to have been much more stable, closed, and excluding, although marriages between “Indian” men and “African” women were frequent. One split among the “Arabs” was between the orthodox Sunni Muslims, the o en rather poor Hadrami Arabs, who came to East Africa from the Hadramaut (Aden) area in southwestern Yemen and the o en wealthy Omanis who belonged to the Ibadi canon of the Kharijites13 tradition and monopolized much of the military, economic, and political power. While the rich Omanis o en established themselves as plantation owners, the Hadrami “did what was open to them, mainly retailing everyday consumer goods” (Horton and Middleton 2000: 185). They came by the thousands with the trade winds to the East African coast to seasonally or permanently work with the trade in the port towns.

Fluid Identities •

279

The Hadrami constituted a heterogeneous segment of the population since they both se led as small scale traders and poor workers (Le Cour Grandmaison 1989). The wave of poor Hadrami immigrants for East Africa seems to have coincided with the growth of the caravan trade. They took up all kinds of jobs open for them as “sailors, fishermen, servants, water-bearers, and o en dockers” (Le Guennec-Coppens 1989: 189) others turned to small-scale retail trade. Their extended social networks made it possible for them to arrange long-distance trade, which in turn made it possible for them to compete at the local level with the Indian traders. This cleavage in occupation became particularly prominent on the markets on Zanzibar where the be er-off Hadramis soon controlled the transports between the boats at the harbor and the storehouses in town while the less well-off worked for them (Fair 1994: 30). They also benefi ed from the abolition of slavery and the slave trade, since they were willing to trade with and sell to slaves and ex-slaves.14 “Within a few years the Hadrami had replaced the patricians in many coastal towns as wealthy retailers, although only in local merchandise and not in intercontinental commerce” (Horton and Middleton 2000: 185f.). Although only a small minority in the region, probably no more than 2 percent of the total population, the Hadrami still came to have a major impact on the coastal culture. One reason was that they managed to provide the coastal societies with religious leaders both before and a er colonization (Nimtz 1980: 20–21) and contributed to the spread of the Shafi’ite version, one of the four Sunni schools of religious law. Their dominance in East African Islam was already well established when the Omanis, belonging to the Ibadi canon of Islam established their sultanate on Zanzibar (Le Guennec-Coppens 1989: 188). Another reason for their dominating position in religious education was, August Nimtz explains, obtained through the wealth they accumulated by gradually becoming the largest slave owners in the region (Nimtz 1980: 26). But this only holds true for Zanzibar. In precolonial Bagamoyo it was the Barawi, immigrants from Somalia, and “Shomvi-Shirazi” who produced the religious leaders (Nimtz 1980: 105).15 In his list of religious leaders in Bagamoyo, 54 percent were of “Shomvi/Shirazi” origin, 20 percent of Hadrami origin, and 20 percent were from the local “African” population. This should be compared with the proportion these groups constituted in the total population: 10 percent “Shomvi/Shirazi,” 3 percent “Arabs,” and 84 percent “Africans” (Nimtz 1980: 105–6). In this respect it seems like the ruling families of Bagamoyo could compete well with the intruding “Arabs” when it came to the religious leadership in Bagamoyo town. The financial and military strength in combination with the symbolic resources of the “Arabs” resulted in rapid Islamization and Arabization

280

•

Muted Memories

among some groups in the societies involved in the caravan trade. In particular it was their “idealized . . . ustaarabu as it is significantly called in Kiswahili” (Constantin 1989: 148), and their degree of “civilization” that a racted other segments of the society. One avenue for those who aspired to become members of the “civilized” group or encompass the ustaarabu way of life was to exchange the wealth generated through the caravan trade into religious, political, social, and cultural capital. Different groups in the urban se ings along the coast strived to be “Arabized.” One reason for the “Arabization” of other groups was the ideological hegemony (Goldman 1996: 7) the “Arabs” held over socioeconomic and religious life in the trade towns. Another force was that the British colonial regime in 1910 re-classified the “Arab” population as a non-native population. This meant that they suddenly were exempted from the hut tax, which all “natives” had to pay (Arens 1975: 433). Furthermore, in the racial thinking of the colonizers the “Arabs” were classified as “whites,” which, for some in the colonial society, must have been a further a raction to be “Arabized” and be identified as an “Arab.” One strategy to secure a prominent position in the coastal societies was to keep education and knowledge held like cards tight to one’s chest. The “Arabic” elite seem to have used the control over knowledge, even religious knowledge, as a strategy to monopolize the top position on the status ladder. One of the consequences of this was that “Arabs” never established a total hegemony over the other groups in the coastal towns, despite some of the other segments’ struggle to become “Arabized” (Constantin 1989: 152). Since the “Arab” identity was broad and vague because children from an “Arabic” father and an “African” mother could usually claim an “Arabic” identity, “on the coast they [those who called themselves “Arabs”] can scarcely be distinguished from the Aborigines” and there has for a long time been “a significant number of black ‘Arabs’” (Stanley 1899: 35). This was noticed early on by the European and American travelers who interpreted and presented the dynamics through their racial worldview. It has, for example, been stated of the “Arabs” who accompanied Sultan Said when the Sultanate was established on Zanzibar that “almost all of them purchased negro concubines, the result of which we trace today in the various complexions of those who call themselves Arabs, while the descendants of the first Arabs of the first migration are now deteriorated so much that on the coast they can scarcely be distinguished from the Aborigines” (Stanley 1888: 35). However, this process of identity-making and creolization occurred in multiple directions, since

Fluid Identities •

281

the trend on the Coast has always been the Swahilization of the Arabs rather than the Arabization of the African Muslim inhabitants, in such facets of culture as language and mode of living in general. This trend has been so forceful that most Arabs on the Coast, who have se led here within the last two centuries or so, have lost their original culture and language and completely adopted Swahili culture. (Abdulaziz 1979: 8, quoted in K. A. Askew 1999: 72)

The “Arabic” identity was like a sieve with the processes of osmosis leaking in both directions. The immigrating Omani “Arabs” were regarded as members of the Swahili societies a er having been socially and culturally integrated in the coastal societies (Middleton 1992: 1–2). “Arabic” traders became “swahilisized” while “Swahili/Shirazi” traders became “Arabized” to an extent, which would not have happened without the caravan trade. The wealth generated through the caravan trade fueled dynamic processes of creolization or hybridization in Bagamoyo.

The Waungwana Identity on the Move along the Caravan Routes The waungwana identity must have been a most challenging experience for European visitors and colonizers of Eastern and Central Africa. Originally the term was used as a term of self-identification by the males of the middlemen-merchants or leading families,16 as the first families in the coastal towns had referred to themselves as mungwana (sing.) or waungwana (pl.), an identity which is associated with uungwana or “civilization” as to possess kiungwana or the ability to speak and behave in a proper way. The term, however, has caused much bewilderment since it has been used in different ways and in different contexts on the coast and along the caravan trade routes. One of the most detailed accounts of the fluidity of the term was provided by C. T. Wilson and R. W. Felkin who arrived to Zanzibar in 1876 to organize their expedition from Zanzibar, via Bagamoyo to Uganda. They report that: The word Waun’guána means gentlemen, and has been appropriated by these negroes [who work with the caravans] to distinguish themselves from the slaves who work on the mashamba or plantations. These Waun’guána are not, as a rule, natives of Zanzibar, but slaves who have been brought from the interior when young, and have become to a great extent naturalized, and who have learnt Kiswahili. . . . They are all professedly Mohammedans, having been circumcised by their Arab masters that they may be “clean” to slaughter animals for food; it being unlawful for a Mohammedan to eat meat killed by an “infidel.” (Wilson and Felkin 1882: 1:14f.)

On Zanzibar, the term seems to gradually have lost its connotation to slaves and was used for the group of people who have entered the status

282

•

Muted Memories

of free men as in Henry Morton Stanley’s account: “When the Arabs conquered Zanzibar, they found the black subjects of the Portuguese to be of two classes, Watumwa (slaves) and Wangwana (freemen). The freemen were very probably black people who had either purchased their freedom . . . or were made free upon the death of their masters” (Stanley 1888: 37). Tippu Tip adds still a characteristic as he translates waungwana as “freemen, as distinct from both slaves and Arabs” ([1974] 1958–1959, 21). The categorization of the original population into free, waungwana, and slaves, watumwa, lost much of its importance when people from mainland Africa se led in Zanzibar to take part in the boom of the caravan trade. A new split developed between the waungwana or natives of Zanzibar and the washenzi, pagans and barbarian porters from mainland Africa. With this shi of identity, the term waungwana now stood for both freemen and slaves with a Zanzibar origin (see Stanley [1899] 1988: 37). This is still an illustration of the importance paid to the geographical origin, putative or not, in the construction of hierarchical schemes in coastal societies. On the other side of the Zanzibar Channel, in the coastal towns of mainland Africa, the term shi ed to carry the meaning of “urban gentility” (Glassman 1995: 62), or “gentleman” or “civilized.” Here the sociopolitical setup differed from the situation created by the direct Arabic occupation and colonization on Zanzibar. However, the fortune that the newcomers, freemen or slaves, could make from the hectic caravan trade gave them the possibility to approach the lifestyle of the waungwana and make claims that even they belonged to the “civilized” people. Tippu Tip used the term in his autobiography for people from the coast who took part in the caravan trade in the interior of the continent. Many of them were slaves, which indicate that the term was used along the caravan routes to designate people of a coastal background or with aspirations to belong to a coastal society rather than persons from a certain social strata or ethnic background. A similar explanation is given by Marcia Wright (1993) for the term alungwana, a derivation from waungwana used in the Lake Nyasa, Tanganyika, and Mweru areas. According to her alungwana is “a term suggesting a Muslim religious identity and linkages with coastaloriented commerce, but not necessarily with ancestry external to Africa” (Wright 1993: 8). In the coastal se ings the identity waungwana made a downward status movement as a result of an increasing number of up-country people who se led on the coast and aspired to the identity. Furthermore, some slaves owned by coastal traders could, as already reported, not only take part in the caravan trade as businessmen but could also own their own slaves who worked for them as porters. These wealthy slaves from the

Fluid Identities •

283

coast adopted the waungwana identity to differentiate themselves from the up-country “barbaric” (uncircumcised; Stanley [1899] 1988: 37) Nyamwezi newcomers. Thus, the term gradually came to denominate free or slave persons with a coastal origin or residence, where wealth from the caravan trade could be used to counterbalance their slave status. The waungwana, free or slaves, acted as porters, owners of other slave-porters, military guards, or caravan leaders (Rockel 2009: 93f., 96). They could even enroll in caravans set up in the coastal ports without their owners’ consent. European travelers, traders, and missionaries from the time of the East African Exploration of Richard Francis Burton and John Hanning Speke into the colonial conquest of the 1880s recruited waungwana slaves for their caravans as if they were free people (Rockel 2009: 98–99; 2014: 183). The ability of the waungwana to roam from the highly hierarchical coastal society and spend months and years away in caravan trade provided them with opportunities to advance in the social hierarchies (Rockel 2009: 90). But the waungwana identity also made a geographical movement as it traveled along the caravan routes and got a new meaning in the up-country societies. This was particularly prominent in the area around Lake Tanganyika and especially in the Manyema area in eastern Congo, which suffered from the invasion of Ngoni people from southern Africa and an occupation by “Arabs” and their henchmen from the East African coast. Their subordination under the intruding slave-hunting “Arabs” led, as already mentioned, to the collapse of their society and enslavement of a large portion of the population. As a result many tried to adopt the “Arabic” slave hunters/traders way of life, including Islam, giving them the label “petits Arabes” by the Belgian colonizers (see Page 1974: 70). Those who could claim connections with the coastal societies and traders through the caravan trade began to claim a waungwana identity. Some of the most ruthless slave hunters, as already mentioned, o en slaves themselves, and owned by “Arab” traders and slave traders, became identified as waungwana (B. Brown 1971; Page 1974). This shi in meaning Melvin Page (1974: 71) notes, “confounded the Europeans who could not understand how freemen [waungwana] could be slaves.” It has, however, been pointed out that many early European missionaries, travelers, and colonizers made a mistake when they translated waungwana or alungwana with “Arabs” or “petits-Arabes.” Wright interprets this mistake as “an example of how the antislavery rhetoric has sometimes invaded the English and German versions of vernacular understandings” (Wright 1993: 8). The cruelty of the waungwana slave hunters has been explained by their own fate of being slave-napped when young and brought

284

•

Muted Memories

up by an “Arab” slave trader to whom they developed an unrestricted loyalty. It may be assumed that the increased pressure from the newly rich at the bo om of the hierarchical scale led the original waungwana to adopt new identity markers to monopolize their leading position in society. It may also be assumed that the presence of the Zanzibar Sultanate and the establishment of “Arab”-owned plantations along the coast gave new role models and that the identification for those who tried to present themselves as “civilized” shi ed from uungwana to ustaarabu (Rockel 2009: 102n2; see also Spear 2000: 282) communicating connotations with the Arabic cultural and social spheres. Ustaarabu is a quality you possess, and not an ethnic label of being an “Arab.” Rather, you live and act like an “Arab.” When the pressure increased on the waungwana identity of the elites, some adopted the wastaraabu identity in the a empt to distance themselves from the upstarts.

The Power of Boundaries and Categorization It is increasingly apparent that an alternative approach to rigid tribal thinking is needed to understand and present the cultural dynamics that followed in the wake of the booming caravan trade. The last part of this chapter will present such an alternative approach. The fluidity in processes of identity-making and communication in nineteenth century Bagamoyo may be grasped through the application of anthropological approaches and perspectives regarding the construction of sociocultural categories, identities, and boundaries. In this tradition it is recognized that socially constructed identities are not permanent but relative and depend both on the identifier and the identified. An identity “is merely a name or term, an idiom or representation of role, used to describe a person or group in relation to the wider society. It is an a ribute, not an innate quality, and is changeable and renewable” (Horton and Middleton 2000: 199). Such an approach is needed to grasp the changes that occurred regarding the “Swahili” and “waungwana” identities at the end of the nineteenth century. In his “Introduction” to the book Ethnic Groups and Boundaries, Fredrik Barth (1969) suggests that we should not concern ourselves with the “cultural stuff ” of different ethnic groups in a multicultural se ing. Rather, we should concentrate on how the boundaries between the groups are created and communicated. In a later development during the thirtieth anniversary of the publication of the book, he redefined the argument that

Fluid Identities •

285

we, as human beings, always try to construct categories based on differences, but not always set up boundaries and barriers (Barth 2000). The “Swahili” identity and way of life, as the Nyamwezi porters met it, is an example of an identity that for them must have been both very different from their own but also, for some, highly a ractive. It was a porous identity without or with only thin boundaries. The minimum demands to qualify as a person belonging to the Swahili world were to live in, or close to, a town on the coast, speak Swahili and also be a Muslim. No clear gatekeepers controlled those who tried to pass into this identity in Bagamoyo during the peak of the caravan trade. This contrasted markedly with the borders to the higher echelon, the “Shirazi,” the middlemen-merchants of the town, whose borders were jealously patrolled. The “Swahili”/“Shirazi” divide was very much a split based on class, wealth, prestige and power, rather than an ethnic division. In a similar vein Thomas Hylland Eriksen (1993) clarifies that ethnicity appears as a dimension of social relations when importance is inscribed, or given, to cultural differences—when differences make a difference. We may well observe differences without paying further a ention to them. We may even enjoy being part of a multicultural se ing. It is first when we give meaning to these cultural differences between various groups that ethnicity, as one kind of identity of politics and a specific dimension of social relations, appear. And this is what we see in the cultural brew in Bagamoyo when the middlemen-merchants played out one card a er the other to mark their difference or demarcate the border between themselves and the newly rich plebeians, even of slave or up-country origin, who demanded inclusion and belonging. For some up-country porters, the difference between her/his identity of origin and the Swahili identity must have seemed great enough to seek the rewards of crossing the divide. This is not to say that all porters tried to pass into the coastal urban culture or that they abandoned their original identities. To claim Swahili-ness while on the coast implied a claim of citizenship in a town like Bagamoyo, but this new identity could be added on top of other identities. It may even be assumed that the identity, rather than being fixed and bound, as in the colonizers’ worldview, was highly fluid and varied as the porters/traders marched up and down the central caravan routes. A crucial element in the understanding of the politics of identity in nineteenth century coastal towns is the middlemen-merchants readiness to categorize people in binary oppositions, or what Nimtz calls “cleavages,” as the “Swahili” versus the “Arabs” or the “civilized” versus the “uncivilized” barbarian. It may be assumed that the precarious position

286

•

Muted Memories

of the “Swahili”/“Shirazi” along the coast contributed to forming their worldview, particularly in relation to the areas and peoples to the west. It has been noted that “the traditional Swahili a itude towards the bush (nyika) . . . that being savage, lost, uncouth, astray, or without religion is associated with being on the mainland, outside coastal society” (Pouwels 1987: 36). And “Swahili townsmen traditionally have divided their universe in true structuralist fashion between the town—representing order, civilization (uungwana), and predictability—and the world outside it— representing chaos, barbarity (ushenzi) and the fear-inspiring unknown” (Pouwels 1987: 30). The Nyamwezi porters came to a society that was highly hierarchical, and where they represented the “fear-inspiring unknown.” However, beneath the dichotomy “civilized-barbaric” was a finer grading of the town citizens into numerous ranked categories. A prominent companion to peoples’ obsession with constructing categories is, under certain circumstances, to also arrange the categories in hierarchical order.17 The communities along the East African coast were far from the stable, cultural chessboard the colonizers expected to meet and turned instead into brewing pots with a number of ingredients: structures and agency, processes of inclusion and exclusion, manipulation of genealogies, boundary protection, and claims for acceptance. This proposition should not be understood as a suggestion that a clear distinction in ethnic categories was absent in these societies, only that they existed together with other forms of categorization that may have been more important for those who lived in precolonial Bagamoyo. The historical dynamics with various, often internally competing powers and internal competitive consumption in order to claim or confirm prestige and power gave the flavor to this brew or, rather, powder keg.

Avenues for Status Manipulation and Recognition The booming caravan trade brought together different social and ethnic categories in Bagamoyo and triggered both processes of hybridization and hierarchization. In these processes Goldman (1996: 7, writing on Pemba) finds that “in a nutshell, it is ‘be er’ to be a Swahili than a Mainlander, but being a Swahili is ‘worse’ than being an Arab.” Such views gave a specific flavor to the life in Bagamoyo during the caravan trade. Two main avenues were open for those who sought status advancement and inclusion in coastal society. One was through consumerism and conspicuous consumption, which is further explored in the next chapter.18 The other was through the establishment of strategic kinship ties and marriage.

Fluid Identities •

287

Status Manipulation through Kinship and Marriage Both kinship systems and preferred forms of marriage lend themselves to processes of stratification and ranking that characterized life in coastal towns and were frequently used as means for inclusion and exclusion. The “Arabic” population tried to follow strict patrilineal kinship rules, o en based on the idea to keep “the blood” (dam) clean, and practiced endogamous marriages. This is a kinship and marriage system that promotes exclusion and monopolization, which may be essential since a daughter inherited half the value of what a son inherited. The ideal form of marriage was the so-called patrilateral parallel-cousin marriage (a man marries his father’s brother’s daughter), which is assumed to keep the blood “clean,” prevent outsiders from “marrying-in,” and property to “go-away,” i.e. be lost to the family or clan. The preferred endogamous system did not prevent elite men, even the Sultan Seyyid Said, to father children with women other than his official wife. One of the sultan’s daughters, Princess Salme bint Said bin Sultan al-Bu Said, writes in her autobiography that her father had only one horme or legitimate wife while he kept seventyfive sariri (sing. surîe, in Kiswahili suria) or concubines, and that he had fathered thirty-seven children with them while none with his legitimate wife (Said-Reute [1980] 1994: 5). The elite section in the Swahili towns as those who saw themselves as the “Shomvi” was in a more ambivalent position since they too favored endogamous marriage alliances in the form of patrilateral parallel cousin marriages, but they also strove to form strategic marriages either between their daughters and high status traders or immigrants from the Arabic peninsula or between their sons and daughters from the elite lineages of the original rulers in the coastal societies. Their life breathed seclusion and distance from the ordinary townspeople, as seen in their houses, which o en turned inward and were built to offer a life in seclusion, with the rooms o en arranged around a central open court. Some of the old coastal towns were also surrounded by stone walls, as may have been the case in the early years of Bagamoyo (see Chami et al. 2004: 40), which separated the insiders from the outsiders and public affairs, with the jealously patrolled boundaries being in the hands of the representatives from the dominating group or clan(s). The “Swahili” practiced, in contrast to the unilineal system o en practiced by the “Arab” families and the old ruling families, was also a system which has been referred to as an “unrestricted cognatic descent groups” (Glassman 1995: 121; Horton and Middleton 2000: 154f.). This gave rise to loosely composed networks in which: “membership could be traced through either the male or female lines. This meant that a person could

288

•

Muted Memories

choose to claim membership in any or several of the descent groups of his or her parents or grandparents. It also meant that if a person married into a local ukoo [kingroup] the children would have the full rights as members” (Glassman 1995: 121.) Typically, this kind of descent system was not suitable for corporate ownership since it had an almost open-ended recruitment of members, which resulted in the establishment of wide-ranging networks. The dominating descent system among the middlemen-merchants, classified as cognatic and bilateral, was of strategic importance for them in their role as middlemen between the Indian Ocean and the African continental trading networks. Strategic marriages of their daughters to incoming rich traders were a prominent theme in the legends and chronicles of the old Swahili towns, which harmonized with the strategy of immigrating trades or rulers to marry the local leader’s daughter, as was the message in the Kilwa Chronicle about the arriving Persian prince referred to above. This system promoted inclusion and absorption (see Chi ick 1963). The exogamous system and strategies, however, created a much more porous system than the endogamous marriage rules. Glassman (1995: 121–22) provides an interesting account of the strategic play with kinship and marriage among the inhabitants of the coastal towns. A young couple’s first marriage was carefully planned and organized by their parents and should ideally take place between first cousins. However, divorces were exceptionally frequent in this society and the young bride could suddenly become available for marriage to another man. One particular condition may have contributed to the high degree of divorces in Swahili society. It was reported early during colonial times from Kenya that “a ‘large proportion’ of Swahili women were sterile” (cited in Eastman 1988: 13). This may have contributed to the absorption of a high number of slave concubines who could give birth to the offspring needed to perpetuate the patriclans. It is also plausible that a high degree of infertility blamed on the freeborn women resulted in a high degree of divorce. A divorced elite woman or widow in Swahili society was very much le to herself when it came to finding a new husband (Glassman 1995: 122–33). It may be hypothesized that the deteriorating economic situation the patricians faced in the competition with the newly arrived “Indians” and “Arabs” made it increasingly difficult to live up to their cultural ideals. This made it possible even for a low status man, given he was Muslim and not a slave, to marry a divorced daughter even from an elite family. The children of such marriages, granted that it was her former husband and not the woman herself who was infertile, were full members of the father’s or the mother’s clan and hence of the society. By a strategic marriage a man could complete his process of integration into the society through

Fluid Identities •

289

his wife and children. In this sense, the exclusivity, which the patricians tried to uphold, was conquered at its weakest point—divorced women and widows had li le to win by defending the exclusivity (Glassman 1995: 133).

From Fluid to Frozen Identities Precolonial nineteenth century Bagamoyo was a cultural brew with sometimes clear but more o en unstable boundaries among the people who came to live there for shorter or longer periods of time. The processes of hybridization created a cosmopolitan society. The processes behind this are captured by Randall Pouwels in the following way: A crucial aspect of the development of many coastal se lements was the persistent, frequent necessity of integrating groups of . . . newcomers (wageni) with the established social order within them. One thing, then, which characterized the coastal town was that it institutionalized change introduced by such immigrants and furthermore, the internal structures created to institutionalize such change reflected the fundamental ambivalence townspeople felt towards the outside world. (1987: 33)

It was only through colonization that some boundaries became fixed, as experienced by Mtoro bin Mwinyi Bakari, a teacher from Bagamoyo who was taken to Germany to teach Swahili in Berlin and Hamburg to the German colonizers. When he tried in 1905 to return with his German wife Bertha Bakari to German East Africa, at the time when the German colonizers introduced more strict rules for racial segregation (Brennan, Burton, and Lawi 2007: 24), the couple was not allowed to disembark from the ship and take up a life in his country of origin. The reasons are evident in the memoires of Graf von Götzen who writes that: Our laws have endorsed the civil marriage of a woman from Berlin and an East African Negro . . . This docile and orderly, but due to the legislation, completely misguided Negro arrived today with the intension of spending his life from now on with his wife in a Negro dwelling. . . . the undermining of the repute of the white race, which would inevitably follow the appearance of a mixed white and black couple in the streets of Dar es Salaam, would have worsened our situation the more, at a time when, almost daily, the belief in our superiority is already contested. (As quoted by Wimmelbücker 2009: 46)

Both the Germans and the British built their colonial regimes on rigid racial classifications, segregation and discrimination, very different from the dynamic processes of identity formation, inclusion, exclusion, and creolization that characterized precolonial Bagamoyo. This change will be further explored in the last chapter.

290

•

Muted Memories

Notes  1. In this chapter (only) the ethnonyms are wri en between quotation marks to indicate that the labels are both highly questionable and extremely fluid, as will be shown. Quotation marks are not used for the Swahili language, Kiswahili, or the Swahili culture or society.  2. Anthropological and ethnographic work contributed to the creation of this worldview through its a empt to detect and describe differences rather than similarities. See, for example, Tania Murray Li (2007) regarding the Dutch scientists’ work in Indonesia, and David van Reybrouck ([2010] 2014) on the Belgian colonial state’s work in the Congo.  3. This statement also represents the world view of the German colonizers.  4. Or, as expressed by James de Vere Allen (1981b: 325) “our image of Swahili society . . . is based on documents whose authors seldom went outside the large se lements.”  5. John Middleton (1992: 1) suggests that it can be interpreted as a term for those who were at the border of the Arabic civilization.  6. The term should therefore not be understood in its political dimension as a specific country of the “Swahili” but rather the “land where the Swahili live.”  7. The redefinition was probably most dramatic on the eastern part of Zanzibar among those called the “Hadimu.” The name means “servant” or “slave.” In connection with a strike among workers on the clove plantations and activities from the Shirazi Association (SA) “virtually overnight, villagers throughout the region embraced the SA’s language, saying to the Arab elites, in effect: We are not slaves or servants, not ‘Hadimu’, but civilized people, ‘Shirazi’, with an inheritance of ustaarabu as deep as your own” (Glassman 2004: 749).  8. Randall Pouwels (1987: 10–37) believes that the “Shirazi” identity became fashionable first during the thirteenth to seventeenth centuries as a result of the increased competition over the trade between the patricians in the East African trading towns and the Arab intruders.  9. Some prefer to call this group by the combined term “the Shomvi/Shirazi” (i.e. Nimtz 1980). According to legend, the “Shomvi” probably moved from the coastal area in central Somalia during the eighteenth century. Some se led in and around Bagamoyo. J. Spencer Trimingham sees Swahili as the overarching category, which embraces, for example, the Shirazi (1964: 31–32). 10. For Glassman’s references, see 2004, 746n107. 11. The root of the verb staarabu means “to become civilized.” It derives, according to Nimtz, from “the Arabic verb ista’arab, meaning “to become an Arab” (Nimtz 1980: 33). 12. There is, however, a legend that tells the tale of a trader, Mzee Shaha, who arrived on the coast and bought land by means of cloth (Owens 2006: 725), which may be still a version of the cultural hero who comes to civilize Africans by means of cloth. 13. The Kharijites are strong defenders of an egalitarian ideal since they claim that the position as a leader of other Muslims should be achieved through election

Fluid Identities •

14.

15.

16. 17.

18.

291

rather than ascribed through class or kinship connections. The Kharijites fled to Oman a er conflicts with other Muslim groups in Arabia who supported other interpretations and schools of law (Cooper 1975: 49–50). The “Swahili” patricians and middlemen tried to uphold a clearer distance to the slaves and ex-slaves and, thus, le the field open for the Hadrami traders (Horton and Middleton 2000: 185–86). John Middleton reports that the ruling al-Busaidi dynasty on Zanzibar belonged to the Ibadhi and not the Shafe’i version of Islam and “however powerful, was not accepted as socially or morally equal to the patricians” (1992: 34). Carol Eastman (1984) sees a split between the male waungwana part and the female wanawake part of the Swahili society. A well-known example is René Dumont’s ([1970] 1980) classical study Homo Hierarchicus: The Caste System and Its Implications, which documents and analyzes the Hindu tradition to arrange everything, be it people, plants, places, rivers, etc. in hierarchical order. It should be noticed that this obsession with hierarchical arrangements is not found in all societies. Some, mainly based on hunting and gathering, do very well without elaborate hierarchies. Ritualized systems of conspicuous consumption, like the fiesta system in Mesoamerica (see Chance and Taylor 1985), are said to have redistributive effects in the society when wealth is exchanged for prestige.

11  Conspicuous Competitive Consumption and Communication by Means of Cloth The wealth generated through the caravan trade fuelled one of the most conspicuous features of the Swahili world, namely the acquisition and maintenance of status and power by means of consumption and display of material wealth. This cultural practice was behind the fluidity in identity formation, manipulation and communication presented in the previous chapter. The booming trade meant an increasing flow of wealth and a growing number of the newly rich, which created tensions, fractures, and riots in many coastal towns. The conflicts and clashes were channelled through and fuelled by two of the most central institutions in the coastal societies: competitive dancing and conspicuous consumption.1 Both ruined some of the old middlemen-merchants but, at the same time, opened avenues for wealthy status-seekers to gain acceptance and prestige. Occasionally it took the form of the destruction of wealth, as when valuables were thrown into rivers, when bowls were destroyed a er they had been used to serve food to guests and then discarded a er the feast (Fleisher 2010: 211) or when the patricians moved out of their old stone houses to other houses and let the first stand and deteriorate as a sign

Conspicuous Competitive Consumption and Communication

•

293

of their wealth (Horton and Middleton 2000: 125). The increased flow of wealth in combination with the culture of competitive consumption made Bagamoyo a most vibrant society during the caravan trade. “The defining trait for inclusion is consumerism,” as stated by Steven Fabian (2007b: 12). The spending took two main forms: the inwardly oriented display of prestige and wealth, such as when Chinese porcelain was shown to invited guests in the stone houses, and the outwardly oriented display when the patron(s) sponsored feasts and publicly displayed exclusive imported items like cloth, clocks, umbrellas and more at festivals and rituals. Festivities were a characteristic feature of the Swahili world through which prestige and power were achieved, reconfirmed, or lost and through which patron-client relations were created and maintained. The caravan trade, which expanded along the central routes in the middle of the nineteenth century, brought together religious, ethnic, and statusbased social groups from the African continent, Indian Ocean, and eventually also from Europe and the Americas in Bagamoyo and triggered both processes of hybridization and hierarchization. The cosmopolitan character of the town during the second half of the nineteenth century fed, as explained in a previous chapter, numerous processes of identitymaking and communication whereby people tried, on the one hand, to gain access to highly ranked status identities and, on the other, to block or exclude the upstarts. The energy behind these processes came from the wealth and prosperity created through trade between the interior of the continent and the wider world. The Nyamwezi porters/traders who reached Bagamoyo with their primary goods—ivory, bee’s wax, livestock, and slaves—entered into an urban milieu. The central part of the town, Dunda or Stone Town, was made up of solid stone houses, two to three stories high. They were of two main types: one made up of residential houses with few or no windows to the public spaces and with more narrow, although elaborately carved, wooden doors. The other type was made up of houses with shops on the ground floor with wider doors, which could display the goods for sale. It is understandable that many porters felt uncomfortable entering the stone town and preferred to display their goods on the beach. Still, they had an influence on what took place in the shops in town since their taste, demands, and knowledge about up-country consumption pa erns influenced what was for sale in Bagamoyo.

Status Manipulation through Wealth and Conspicuous Consumption The previous chapter explored the extreme fluidity in the identities presented, manipulated, and controlled by various segments of Bagamoyo’s

294

•

Muted Memories

population. The socioeconomic, political, and economic fabric of coastal trading towns was built up by different networks of patron-client relations (Glassman 1995: 22). The clients expected generosity from their patrons and the patrons expected obedience and support from their clients. Such networks were not based on religious or ethnic belonging, but crossed over all population segments in Bagamoyo and beyond.

Competitive Dancing Public rituals became weapons in the political brewing pot when people outside the narrow segment of middlemen-merchants got access to the wealth, which the caravan trade brought to the coast. One of the elements in these rituals that served as a ba lefield for conspicuous consumption was the public dancing arena. Each coastal town had its own set of ngoma nkuu or “great drums,” representing the political independence of the society.2 They were used during public dances called ngoma or ngoma ya shindano (competitive dances) during the most important rituals (Nimtz 1980: 98). Swahili ngoma are entertaining, but they are also “serious business from a standpoint of social relations and political influence: reputations, alliances, political factions, and fortunes could all be confirmed (or possibly lost) by the relative success or failure of a huge public ngoma” (Franken 1991: 149). Mtoro (J. W. T. Allen 1981: 82) gives further evidence on the important of dances: “From ancient times the great expression of joy at weddings or on other occasions is the dance. If there is rivalry, they dance. All love the dance, the very old and the very young.” Music and dances in a Muslim society may be considered halal or legitimate and even beneficial but they may also, or at least some forms of songs and dances, be considered haram and should therefore be avoided. In the Swahili world, however, the ngoma, both music and dances, developed into a cultural expression with strong political overtones as they created and expressed political influence and alliances as well as personal power and reputation. They were at the core of life in the nineteenth century towns on the East African coast and conveyed a number of central codes of that culture and society. Individuals who sponsored the events, a team of dancers representing their patron or own moiety or section of the town3 or the team of the town against the team of another town, could either gain reputation and power or lose it. Every important ritual occasion in the Mrima communities, writes Jonathon Glassman (1995: 161) was marked by festive dances or ngoma. The term carries different meanings. One is the drum, an instrument that also stands for political power within a coastal town. The big drum was

Conspicuous Competitive Consumption and Communication

•

295

taken out and used at important occasions in society and publically displayed for those assembled. The existence of the “big drums” in a number of port towns signals their independence from each other which in turn evidences both the absence of an overarching Swahili political organization or territory and the diasporic character of their world. This political setup paralleled the internal organization of each trade town where the leaders of the hamlets or quarters (mitaa in Kiswahili) were, ideally, coequals to each other. Another meaning of the term ngoma is dance or an occasion when dances are performed. In his explanatory text, “Ngoma: Music and Dance,” which accompanies Mtoro’s The Customs of the Swahili People (J. W. T. Allen 1981), James de Vere Allen explains that ngoma “mark every important moment in the traditional Swahili life: every rite of passage, every occasion for happiness or sorrow, even such events—which must have been common enough in a place like Bagamoyo—as the departure of a caravan” (J. de V. Allen 1981a: 241). It may be added that also the arrival of caravans from up-country areas was celebrated with songs and dances. This makes the ngoma a central institution in the coastal world. Allen adds that “to a very large extent they [these dances] actually encapsulate that society and its culture: they form its paradigm, its skeleton, and a good deal of its flesh as well” (Allen 1981a: 246). The same goes for the “dance societies,” the vyama (plural) that organized the important dance performances. The competitive character, which developed when different chama (singular) came together, set the stage for the acquisition of power through the disposal of wealth. It seems like the competition was modeled on a specific feature in the organization of the Swahili societies or towns (Swartz 1991: 30–31). These societies were o en organized into two halves or moieties or demes (see Horton and Middleton 2000: 120f.).4 This fundamental principle of social organization stimulated both competition between the moieties and cooperation within them. A variety of this kind of social organization is explained by Mtoro ( J. W. T. Allen 1981: 84) when he writes that a dance society could be formed by “one quarter of the town to challenge another” (1981: 83) and that “they could dance all night for six or seven nights of continuous dancing. They spent a great deal of money . . . On the last night of the dance there was a party, and the visitors and the local people were told” (J. W. T. Allen 1981: 84). The competitive character of the ngoma can be seen in the characteristics of this cultural event; a ngoma is an occasion when individuals, dancing teams, the two moieties (halves) of a town or even different villages or towns competed for the a ention of the audience; the performance was public and o en took the form of a procession through the society; the

296

•

Muted Memories

texts of the songs sung had both open and more secret messages, o en ridiculing the opponents; and the performance of the competing teams were evaluated by the audience and the team which a racted most, or all, followers was victorious (Franken 1991; see also Campbell and Eastman 1984). A crucial element in the evolution of the dance competitions was the public estimation of the costs that a team spent on costumes, performance, and feasting, which indicated the wealth the sponsor spent on public entertainment and food. The public was both the audience and the judge in this competition. The supposed spending was thereby converted into prestige and power, or alternatively the loss of them. Music, dances, and songs all expressed inclusion and exclusion, similarity but also difference (K. A. Askew 1999: 76).5 It was not only the competitive sponsoring of music and dances that distributed prestige and reputation among the sponsors. The performance itself was also a political weapon. There is a tradition in Swahili poetry and texts, as the texts printed on the colorful co on cloth called kanga (see below), to compose messages with multiple interpretations. This opened up a way for linguistic assaults on and gossiping about an opponent without direct confrontation. It was in the ears of the listeners to interpret the message. “In ngoma performance,” Askew explains, “participants use song lyrics to transmit gossip, cast public accusations, and stake out political positions in an indirect and socially-sanctioned manner. . . . performance is both product and producer of social relations” (K. A. Askew 1999: 77). Thus, the conflicts in Bagamoyo between the original merchants-middlemen and the intruding Omani colonizers were reproduced in this specific but central cultural phenomenon in the coastal societies. This may be seen in the establishment of two new societies during the German time: the gaboreni, associated with the middlemen-merchants o en called the shomvi/shirazi and the sereda, associated with the Arabs (Nimtz 1980: 98).6 The culture of competition and consumerism resulted in a situation in which each group came to sponsor and organize its own competing dance societies during the peak of the caravan trade (Nimtz 1980: 98). These dance societies not only enhanced the status and reputation of their sponsor(s), but also opened up an avenue for poor people, newcomers, and women, to establish a relationship with a powerful patron and also to take part in the ritual life of the society. Glassman (1995: 162) notes that the majority of the dancers had a background as newcomers from surrounding or up-country societies and, that the dances they performed had o en been brought by them to the coastal societies. Rich urban dwellers sponsored the dance societies while up-country newcomers did much of the groundwork. To become a member of a dance society

Conspicuous Competitive Consumption and Communication

•

297

became a stepping-stone away from the “barbaric” status into “civilized” urban society.

Hosting Feasts The economic boom that followed a er the Nyamwezi started to bring their ivory to the coast and manufactured merchandise was traded in the opposite direction, particularly helped young and o en marginalized men, ambitious upstarts, acquire the wealth necessary to sponsor their own festivities and demonstrate generosity during the festivals and rituals. Through the culture of consumerism, the newly rich could engage in the competitive display of wealth, so central in the Swahili world, as a means of gaining access to a higher status in society (Glassman 1995: 165). Hosting a feast was a central cultural phenomenon in these societies through which, in Michael Dietler’s words (2001: 66), “people actually negotiate relationships, pursue economic and political goals, compete for power, and reproduce and contest ideological relationships of social order and authority.” Feasts and feast-hosting became vehicles to claim and demonstrate inclusion and exclusion as well as superiority and inferiority. The increasing flow of wealth from the caravan trade must have had a dramatic effect on the elaborately established hierarchical orders in the coastal societies. The large feasts in the Swahili world included different categories of participants, which gave the feasts their own dynamics. A patron could throw a feast to reinforce the patron-client relations in accordance to the expectations within the “moral economy.” Expectations were high among the clients, who demanded that their patron show generosity, while the clients’ a endance and participation in the feast put their patron in the limelight. To summon a large number of followers is a classical way to demonstrate and reaffirm already established prestige and power. Another reason in the Swahili world for hosting a feast was the aspiration to acquire higher positions, an elevated status, and more power. These feasts, or certain aspects of them, drew a different kind of a endee than the client, namely those who could affirm the host’s aspiration of higher status or, in other words, other middlemen-merchants from the same or other stone towns. This form of feast-hosting has been called empowering feasts (Dietler 2001: 76) and involved “the manipulation of a commensal hospitality toward the acquisition and maintenance of certain forms of symbolic capital, and sometimes economic capital as well.” The characteristic form of these feasts in the Swahili world combined commensality with competition. Archaeologists have demonstrated that a shi occurred in the old Swahili se lements during the twel h and thirteenth centuries regarding the taste for imported Chinese ceramics, from

298

•

Muted Memories

jars to wide-brimmed bowls with flat bo oms. This is interpreted as a new preference for ceramics that could contain a large amount of food around which a number of people could sit and share the meal offered by the host. These serving vessels and the amount of food they could contain, in combination with the number of selected guests who could share the meal from the same bowl, made the host’s wealth visible to the invited guests (Fleisher 2010: 205–10). The competitive character of the feast could contain elements of accusations and insults, as described by Glassman in the following way: Sometimes, if one of the guests is a well-known person who has not himself sponsored a large feast, the host will take a bowl of curry . . . and pour it over the guest openly, intended for all to see. He will empty the entire bowl, ruining the guest’s clothes, and then will present the guest with a suit of beautiful new clothes. Now the guest will take those clothes which had been ruined with the curry and keep them until the day that he gives a feast of his own. (Glassman 1995: 166)

The competitive element in this form of conspicuous consumption, with insults and a empts to deflate the status of other participants led to hostilities, but also to a reordering of the status positions. Many old leaders became heavily indebted and saw their status dwindle, while new successful actors in the caravan trade saw their status rise. New actors pressed on the borders of the middlemen-merchants’ society to gain access and new moves were made by the elite to keep the upstarts out.

Communication through Material Objects Consumption is a highly cherished research subject in social anthropology (see, for example, Appadurai 1986; Douglas and Isherwood 1979; McCracken 1988; McLaren 2002). The reason is that many objects not only represent their use or barter-value but also, and o en above all, are valued for what they can express, communicate, and claim when it comes to identities, wealth, status, and power (see Fleisher 2010: 200). One type of status objects in the Swahili world was watches, cherished by the owners as a help to know the time for the daily prayers, a sign of piety.

Illustration 11.1. Pocket watch from 1890, manufactured in Switzerland and owned by one of Bagamoyo’s traders. Photo by the author.

Conspicuous Competitive Consumption and Communication

•

299

Many material objects, such as the cloth discussed below, are formidable communicators of agency and aspirations. In the Swahili world some of these objects were “treated as treasures that would in time become heirlooms and placed in the wall niches [in the houses]. They included manuscripts and copies of Islamic sacred books, rare silks, jewels, personal arms such as ornamented daggers, and accoutrements of gold and silver such as containers and silver used for sandals and bed-ladders” (Horton and Middleton 2000: 113). The wealth should not always be displayed openly and was therefore kept inside the houses in special niches in the walls. Only those invited into the house could see these treasures. Other items were ostensibly demonstrated publicly as when a feast was sponsored. Control over certain objects was used to create exclusion and inclusion and express similarities and differences. Furthermore, objects, like stone houses, demonstrate the age of the family in that particular town and were used as evidence of the owner’s claim to belong to the ruling class of the town.

Stone Houses A most prominent piece of evidence of Bagamoyo’s involvement in the caravan trade is the presence of the stone houses in Dunda or Stone Town neighborhood. It is reported that 225 such houses, the vast majority owned by merchants, existed at the time of the British takeover of the former German colony. More than a third of them were owned by three prominent traders while approximately 150 were owned by other people (Nimtz 1980: 109).7 At the time of the 2002 International Conference only fi y-four stone houses remained, which is approximately one quarter of the number of house at the end of the nineteenth century, some of them in different stages of decay (Knippel 2002). Material objects like houses are widely used to express status and wealth. A classical study on aspirations, wealth, and status is Torstein Veblen’s The Theory of the Leisure Class ([1899] 1994), where he shows that material objects and performative acts are visual evidence of the owner’s/ agent’s status and wealth. Viewed from a constructivist position the objects do not carry a specific status in themselves, but are given it. It seems safe to assume that objects are used to create or express an identity and that one way to exclude others is to get control over the objects that are scarce but may be presented as in high demand and, by extension, control other people’s access to these items. In the Swahili world, stone houses, Chinese ceramics, and expensive textiles were such objects (Fleisher 2010; Wynne-Jones 2007; Zhao 2012). Different types of dwellings have been distinguished in the Swahili world, which is not a surprise since its se lements varied from villages

300

•

Muted Memories

based on agriculture to stone cities oriented towards mercantile activities and the maritime world. Furthermore, wa le-and-daub houses and stone houses were o en built side by side in the towns (Fleisher and LaViole e 1999; Sheriff 2001). A house built of stone (or actually coral) was called jumba in Kiswahili. A multi-story stone house was called kiunga and a mud house thatched with palm leaves, makuti, was called nyumba ya udongo or “earth house.” The stone houses were usually grouped and formed, as in Bagamoyo, in the town’s center. In the coastal trade towns, they were constructed at the sea front, which must have been strategic for the trade activities but also caught the cooling sea breeze. A town’s trade kernel also contained wa le-and-daub houses, although most of these houses surrounded the center of the stone towns as did the camps for the caravan porters. A nyumba ya udungu, sometimes called a Swahili-house, is o en rectangular and built with a central corridor that runs from the front to the back with doors on both sides that lead to four to six rooms. Such a house is o en occupied by more than one family. Facing the street or lane some of these houses have a veranda, called baraza, which is an extension of the house into the public sphere. On the opposite side is a yard fenced with palm-leaves with a space for cooking, washing, and bathing. Here is also the latrine, a deep pit covered with logs.

Figure 11.1. Layout of a nyumba ya udongo or wa le-and-daub house and its yard. Created by the author, a er Tanzania: Bagamoyo township survey, 1969.

Conspicuous Competitive Consumption and Communication

•

301

The stone houses in Bagamoyo were essential in this mercantile cosmopolitan society. A stone house announced the creditworthiness of the owner. It also signalled the wealth and permanent residence of the owner, which were important qualities for visiting traders when they sought trade partners. The stone houses also provided accommodation and store facilities for such traders, which, in turn, facilitated the local merchant-middleman’s chances to monopolize trade with the visitor. The stone houses in Bagamoyo are usually two- or three-story rectangular buildings. In the past they were occupied by one family and their domestic slaves. Today, when many of the traders, and with them the trade, have moved to Dar es Salaam, a number of tenant families may occupy such a house. The front of the house facing the street is usually one-third to one-quarter of the length of the house. A central corridor usually runs from the front door to the door at the back of the house. The rooms on this floor are arranged on each side of the corridor. The latrines are found at the rear of the house. The rooms on the second floor are more o en arranged as a labyrinth with increasing degrees of privacy. The backyard is surrounded by a stone wall and contains separate structures which function as kitchen, stores, and washing place. The yard may also contain a well, which reduces the need to enter the public space to collect water, which o en was female duty. The yard is constructed to

Figure 11.2. Layout of a two-story stone house and its yard. Created by the author, a er Tanzania: Bagamoyo township survey, 1969.

302

•

Muted Memories

provide seclusion and privacy. In front, the stone house had a solid stone bench, called baraza, which provided a liminal space between the privacy inside the house and yard and the public street. The first stone houses on the East African coast were probably built at the end of the fourteenth century at places like Shanga, Manda, Gedi, and Kilwa. This assumption is supported by Moroccan traveller Ibn Ba uta’s report that the houses in Mombasa and Kilwa, two towns he visited in 1331, were built by wood, so the first stone houses were probably built soon a er his visit. The use of coral stone in house construction is, however, at least two hundred years older and was used for the construction of tombs and mosques (Horton and Middleton 2000: 116–17), and it seems probable that this building technique soon a er Ibn Ba uta’s visit was transferred to the construction of higher status dwelling houses. It has been shown in a previous chapter that diffusionist and racist arguments have been used to explain the development of the Swahili language and culture, including the know-how to build stone houses. More recent research has, however, shown that the East African tradition to build stone houses found in the trade towns has no stylistic analogs in Arabia or Persia (Spear 1984: 291), but is a local development rooted in the tradition to build wa le-and-daub houses. This skill underwent a dramatic development when the wealth generated through the expanding caravan trade was accumulated in the hands of a few traders and was intergraded in the processes of stratification and inclusion versus exclusion that characterize the Swahili culture. The elite of the classical Swahili towns “jealously guarded their status through not allowing others to live in the stone town or even to build in stone elsewhere” (Spear 1984: 294–95). The new building technique implied the use of coral stones bounded by lime rather than mud, which had been used before. Lime created a much more solid construction that could support a flat coral/lime mangrove roof, while the mud technique had only allowed thatched roofs (Horton and Middleton 2000: 119). The flat roof soon became a ractive and had symbolic significance since it allowed the construction of additional floors when, for example, a daughter was married. Such an arrangement, in turn, was proof of a family’s long stay in the town and strengthened its claim to belong to the “owners” of the town. The solid walls and roof and the massive doors also acted as signals of the host’s integrity and purity as opposed to the dirt and impurity of the town outside the house (Horton and Middleton 2000: 124). A stone house not only functioned as a dwelling place, but if the owner was a middleman-merchant, the house was intimately integrated in the occupant’s trading activities, created an aura of respect and honorability, and functioned as evidence of wealth and power (uwezo) or capability.

Conspicuous Competitive Consumption and Communication

•

303

Stone houses, as Stephanie Wynne-Jones writes, were “deliberate statements designed for display” (2007: 332). The layout of the stone houses was built on the contradictory principles of hospitality and seclusion and it was designed to communicate different degrees of seclusion between the public streets and the strictly private rooms at the back of the house. One of their functions was to receive visitors and guests, for example visiting traders. O en the houses had a special guest room close to the entrance, the sebule (also spelled sabule), which could house visiting traders (Horton and Middleton 2000: 91). However, the houses were also designed to provide privacy and seclusion, especially for the female part of the household, but also seclusion for the trade negotiations between host and guest. It had an outer porch, where trade goods could be placed for public inspection. The most private part was the ndani or bedroom and possibly rooms behind the bedroom8 (Middleton 1992: 66). It has been argued (Horton 1994: 164–65) that the stone houses provided a key se ing for the conduct of trade and for visiting traders, by providing both a familiar and formalized structure. The visiting trader, most o en from Asia, would adopt a local middleman-merchant with whom he would stay and trade in his house and from the outer room of whose house he would conduct his trade with other traders than the host. Through this arrangement, the middleman became identified with the visiting trader and could demonstrate his own cosmopolitanism and ability to mediate between the different trading networks and worlds. A stone house was also the ultimate expression of wealth. Only a few could afford to build such a house. The concentration of stone houses in a few hands was also the result of restrictions on who could own such a house (Middleton 2003: 518). In the “classical” Swahili world only the elite could. Kilwa provides an example of the effect of restrictions on stone house ownership. The rarity of stone buildings in the Kilwa area, compared to the many buildings further north, has been interpreted by archaeologists as a result of a different political organization. In the south one dominant king probably restricted the construction and use of this prestigious type of dwelling, while further north each town was its own independent political unit with fewer restrictions on who could build a stone house. In the more classical Swahili towns such as Kilwa and Lamu, the merchants’ houses acted as a filter. The most prestigious and precious of a visiting trader’s goods were not openly displayed but kept in the house. That is why Chinese porcelain or items made of glass, except for small beads, are rarely found in archaeological investigations outside the narrow coastal area. The merchant middlemen mainly traded locally manufactured goods made from textiles, metal, wood, and leather with visiting

304

•

Muted Memories

traders from the interior of the continent, and ivory, gold, and slaves with the visiting traders from Asia. The situation developed slightly differently in the more recently established but rapidly expanding Bagamoyo. The wealth generated through the booming caravan trade was partly invested in stone houses. Already in the 1860s, the town was known to have had twenty stone houses (Koponen 1988: 347) and the number grew, but the owners had difficulties in claiming the generations-old ownership of the town. Furthermore, the trade created a class of newly rich and influential people, also from outside the Swahili world, who wanted to rent or purchase such houses, and thereby approach the cultural ideal to live in a stone house. One example was the sultan’s representative in Bagamoyo, his liwali, who rented the oldest now existing stone house in the town, the Arab Fort or Old Fort from the Indian trader Sewa Haji Paroo. Sewa Haji later rented this prestigious building to the German colonizers. Another example was Shaykh Ramiya, captured at the age of eight in the Manyema area in eastern Congo and brought as a slave to Bagamoyo. As a domestic slave he was given the opportunity to engage in the caravan trade, which gave him a profit to invest in coconut plantations. He converted to Islam and eventually bought, as the first person with an up-country origin, his own stone house, as explored in a previous chapter. Another function of the stone houses was to display the second material associated with wealth and power; Chinese ceramics. These objects were probably brought along numerous trading links to the East African coast. The most valuable were used for the display and consumption of food. They were o en kept in special niches in the walls of the stone houses and displayed for visitors and guests. Even shreds of broken ceramics could be plastered into the walls.9

Swahili Doors A conspicuous part of a wealthy stone house was the decorated and carved wooden doors, frames, and lintels, which had caught the interest of the British colonialists (see chapter 1).10 These doors usually consist of two parts or blades. Over the door is a lintel, which can be either rectangular or a half circle. The lintel is o en decorated with rose es and other motifs from the plant world and sometimes also with an inscription from the Quran. The frames are o en decorated with geometrical or floral motifs. The center post, which covers the gap between the two doors, is o en decorated in a similar way as the frames. The center post is always fixed on the le door panel and is therefore not part of the frame of the door (Aldrick 1988: 99). The threshold under the door is undecorated.

Conspicuous Competitive Consumption and Communication

•

305

The door is a separate structure in the house and not integrated in the carrying construction of the wall. This means that it can be removed from one house and inserted in another without a reduction of the carrying capacity of the first house’s wall. Today the door blades are usually fixed to the frame by hinges. In the past it was more common to fix the blades by the use of pivots that fi ed into holes in the lintel and the threshold. A round wooden pivot was used for the top fixation while round iron pivots were used to fix the door blades in the threshold. Iron spikes or studs were used for fi ing together the boards in the blades, but they also became decorative details on the outer surface of the blades. Swahili doors have been constructed in different styles and for different purposes, o en following the identity, taste, or occupation of the person who ordered the door and frame. Two main types of doors have been identified in Bagamoyo; residential and commercial doors (Areskough and Persson 1999: 29). The residential doors are o en more elaborately decorated but narrower than the commercial. Their prime function is to provide privacy and seclusion since they mark the liminal zone between the public outside, the street or open place and the secluded private space inside of the house. The space between the outside and the inside constituted a transgressional zone, which could be imbued with different meaning and practices depending on the use of the room immediately inside the door. Different gradients of private/public zones are passed when a house is entered or le through the door. The two opposites are the public space on the street versus the ndani or bedroom at the back of the house. Only wealthy traders and plantation owners could afford to meet the cultural and religious demands to keep parts of the family and household secluded from the public world. The commercial doors are o en wider and the two blades are divided horizontally in two halves, which makes it possible to have the lower pair shut while the upper pair is open. This makes it possible to display the goods in the shop inside and also make an integrated display between the show room with the goods put on display outside on the façade of the house or on the elevated bench, the baraza, connected to the house just at the door. Merchants from Oman, o en adhering to the puritan ethics of the Ibadi canon, to which the Sultan of Zanzibar belonged, o en preferred modestly decorated door blades with elaborately carved frames. A highly appreciated motif on approximately 650 “Omani doors,” original or copied, along the East African coast was the rose e, waridi, which was placed at the center of the lintel and from which motifs like chains, ropes, waves, or plant leaves extended toward the ends of the lintel.

306

•

Muted Memories

Illustration 11.2. The popular rose e motif with foliage and waves. Photo by the author.

Another highly appreciated carving style in the southern part of the Swahili world has been called the Zanzibar (Indian-Persian) style, and was probably introduced by Sultan Seyyid Barghash (sultan between 1870 and 1888). A common belief has it that he was a racted by the IndoIslamic carving style during a visit to Bombay (Athman 1996: 20).11 Another popular style, the more massively decorated doors, o en with brass knobs,12 are called Gujarat doors and are associated with traders from India. This style resembles the Zanzibar style but is can also be considered an independent style. The decorations differ from the first two described through carvings of leafy foliage and other motifs from the plant world. The more free-floating pa ern of the motifs, compared to the other two styles, Athman H. Athman notes (1996: 23), makes this style suitable to apply also on furniture and balconies. The motifs carved on the doors were influenced by the owner’s profession, taste, and religious orientation. Geometrical figures are common, plants and flowers are frequently seen, fishes, birds, and other animals may be seen, while human motifs are absent (Athman 1996: 12–13) in Islamic art. A special feature is a motif interpreted as a chain, said to communicate that a slave trader built the house.13 Motifs representing fish or scales of fish or lobsters at the bo om of many Omani doors are interpreted to communicate an association with the Indian Ocean, an indica-

Conspicuous Competitive Consumption and Communication

•

307

tion that the house owner is a sea-borne trader. The association with the Indian Ocean is further interpreted to indicate both plenty and fertility, drawn from the ocean, in this particular house (Athman 1996: 27). Today, however, both carvers and house owners seem to have lost many of the former interpretations of the motifs.14 The names used to characterize the different styles of carving the doors and lintels indicate an origin outside of Africa. An earlier style, called Bajuni, dates back to the eleventh century. Its name derives from the chip carving technique used to carve out circles and straight lines on the wood (Athman 1996: 13). The cosmopolitan character of the port towns that followed on the booming caravan trade and the introduction of a plantation economy resulted in the introduction and development of new styles and preferences. The doors fi ed into the facades of the houses became public manifestations both of the owner’s wealth and status and the society’s integration in a globalizing economy.

Textiles Talk: Communication by Means of Cloth Cloth was one of the most important trade items, as mentioned above, and it was also a central element in the construction of a worldview that carved a sharp divide between the civilized and the barbarian. Cloth was also excellent material both for demonstrations of wealth and acquisition of power and reputation. Cloth can be used to publicly announce and display a person’s claims or qualities. Thus, the amount of cloth carried on the body and its quality were public manifestations, in contrast to some of the Chinese ceramics mentioned above, which were kept in the private sphere in the stone houses and only visible to invited guests. Furthermore, cloth can be used for the display of the shape of the human body and its parts or for the hiding them. Clothing practices may be used by the actors to shape identities and may be used by scholars to study boundaries between social categories. The use of cloth as a boundary marker was particularly common in the highly hierarchical coastal societies involved in the caravan trade. The three main classes—the elite traders and bankers, the free-born population, and the slaves—were marked by their specific codes of dress. The members of the wealthy stratum of the population who o en saw themselves as Arabs or living an Arab-like way of life used their wealth to acquire status textiles as “cloth with names” and to create dress styles that ordinary people could not afford to follow and the slaves were not allowed to adopt. Textiles called “cloth with names,” indicating that a number of specific textiles had their own identities, constituted a small, but luxury, part of the

308

•

Muted Memories

textiles imported to eastern Africa. Sarah Fee (2012: 50) reports that “these luxury textiles made up a small part of cloth imports, from 6 per cent to 25 per cent of value depending on time and place; yet they were indispensable to the social relations upon which the trade depended.” While the monochrome white co on sheet called merikani and the blue sheet called kaniki were the grease and fuel for the caravan porters and people along the caravan routes, the more luxury “cloth with names” energized the interface between the various trading networks that met in coastal ports like Bagamoyo and between the leaders of the caravans and the leaders of the societies that a caravan passed through. The label “cloth with names” refers both to co on and silk textiles handwoven in western India and Oman. These usually striped textiles came to replace the cloth that had been imported to eastern Africa from Gujarat during the eighteenth century. During the nineteenth century, much of the production of the monochrome crude types of textiles was taken over by mechanized mills in Europe and the United States, which cut down the prices of such textiles and made them accessible to an increasing segment of the population in eastern Africa. It may be assumed that this increased the interest among the wealthier sections of the society to acquire handwoven expensive materials in bright colors, and intricate pa erns and design. British industrially manufactured co on sheeting were early used throughout the world to print textiles in local demand. Much of the printing and dyeing was made in western India, as was the case during the 1840s and 1850s with the cloth used by Swahili women to cover their bodies. A decade later, writes Fee (2012: 61), “there is convincing evidence that some printing was also done in situ in eastern Africa by resident artisans— either African, Indian or both.” Africa was not just a passive recipient of imported goods, but an active participant in global trade. The result was the kanga, which will be presented below. Cloth therefore played a most important commercial and symbolic role in the Swahili world (Middleton 1992: 32). They were used to mark the fundamental divide in their worldview, which existed, and still exists, on both sides of the Zanzibar Channel, between the civilized, the wastaraabu, who ideally lived an urban life, utamaduni, and the barbarian, the washenzi. The cloth played, for example, a highly symbolic role in the Kilwa Chronicle15 when the “Persian Prince” arrived to this trade town. He encircled the whole town with cloth, according to the legend, to “civilize” it (see Strandes [1899] 1961: 73–74). This symbolic change from the state of washenzi to the state of wastaraabu, as enabled through the use of cloth, made it possible for him to marry the town ruler’s daughter, now civilized, and through this marriage he became the new ruler of the town (Horton and Middleton 2000: 52).

Conspicuous Competitive Consumption and Communication

•

309

The role of cloth also appears in legends that explain the origin of Bagamoyo as a trade town. One legend specifies that Mzee Shaha arrived from overseas to the area around Bagamoyo and met the local leader Mzee Mazongera. The trader asked for a piece of land and received the answer that the local population was willing to sell a piece of land to the immigrant and that the local people “will go up,” which meant that they were willing to move further inland (cf. the legend in chapter 9 about “the pazi who will not see the water,” which is a variant on the Mzee Shaha/Mzee Mazongera legend). Pazi Mazongera “stuck a pole (mlingoti) in the ground and told Mzee Shaha to place cloth [in a pile] up to the top of the pole” (Owens 2006: 725). Shaha failed, however, to cover the whole pole and therefore had to bring a yearly tribute called kanda la pazi to the local leaders.16 A variant of these lines of thought can be found among the nineteenth century coastal slave owners who believed that a slave, especially a “raw slave” brought from the interior to the coast would become loyal to her or his new owner if given a piece of cloth (Glassman 1991). It was not only the a raction of the cloth which lured the slave to stay with the owner, but also the symbolic significance of the gi , which represented the slave’s embryonic integration into the “civilized” world. Cloth here played a central role in the coastalists’ construction of a civilized and a barbarian world. This divide ran through the Swahili world, between the people from the coast and those from up-country societies. By extension this also marked the divide between “the African” and the “non-African” where the Swahili and Shirazi placed themselves on the “non-African” side.17 Cloth, as mentioned, played a significant role in the demarcation of this divide (Horton and Middleton 2000: 111). Cloth, furthermore, marked the division between free people and slaves. Men from the elite carried, as witnessed by the first Portuguese who visited the Swahili coast during the sixteenth century, a dress, which has hardly changed during the centuries, with the exception of the dropping of the turban. The Sheikh “wore a long garment reaching to his ankles, a braided velvet jacket and a multicoloured silk turban worked with gold thread. Tucked into his silken sash was an engraved silver dagger . . . Today [the end of the nineteenth century] the dress of the prosperous East African Arab would be exactly the same” (Strandes [1899] 1961: 18). The influence from Oman over the prestigious types of textiles was considerable during the first part of the nineteenth century. Documents from Zanzibar shows that 100,000 pieces of cloth were annually imported from Muscat in the late 1840s. “As a general rule,” writes Sarah Fee (2012: 74), “eastern Africa retained the cloth pa ern names used in Oman (subaya, Ismaili, sabuni, rehani), where most of these pa erns are still woven to this day.”

310

•

Muted Memories

More precisely, a male member of the elite, and particularly Omani men, wore a kilembe or kilamb, a turban of several meters of cloth wrapped around the head. The turban was a symbol of honor and prestige and men of status never le their houses without a kilembe covering their hair (Fair 1994: 214–16). Wealthy Omani men also wore a joho, a richly embroidered jacket which reached to the floor or a short jambia jacket both carried over a foot-long kanzu gown. A most important detail in their outfit was the jambia dagger with a handle in ivory, gold and silver and stuck in the waist cloth (Fair 1998: 72). A wealthy woman was dressed to be shielded from public view in a shirt that fell below the knees. She also wore tight fi ing pants called marinda, embroidered with gold and silver threads and with frills around the ankles. Over her head she wore a veil that included several articles. One was an embroidered silk mask covering her forehead and mouth but with holes for the eyes. This a ire is known as a barakoa, the swahilisized version of the Omani burqa (Fair 1998: 70). On her feet, an elite woman wore sandals. A free ordinary man wore a kanzu, a white gown reaching to the ankles. The simplest type was untailored and just thrown over the shoulders on each side of the head, which le the arms bare. More elaborated versions were sewn from white calico cloth in a style that covered the arms down to the wrists. A Muslim free man covered his head with a kofia, an embroidered cap. The dress style for free women depended on their financial resources. If poor she was dressed in a kaniki, a dark blue or black piece of cloth usually made from the unbleached merikani co on cloth or a thin grey shirting called satini. This textile from England entered the market a er the 1860s when the merikani disappeared. It became popular since it was cheaper than the kaniki because of its lower quality (Biginagwa 2012: 63). If the economic circumstances permi ed, a free woman could instead of the kaniki wear a kitambi, imported colorful co on cloth. Unlike an elite woman, an ordinary free woman was rarely veiled, but unlike a slave woman she was not prevented from covering her head. It seems like the tradition among free women to veil their heads grew stronger during the early twentieth century with the introduction of the buibui. The buibui seems to have appeared among wealthy women on Zanzibar at the turn of the nineteenth century. It is a long piece of dark cloth sewn in a circle and tied around the head. If combined with an ukaya mask the whole head could be draped. The a ire was first seen on elite women on Zanzibar but was later adopted among free women, also in the towns on mainland East Africa. Completely different ideals and prescriptions applied to the slaves, who could not decide how to dress on their own. Swahili cultural codes,

Conspicuous Competitive Consumption and Communication

•

311

as already mentioned, said that a slave man should only carry a winde or winda, a loin cloth tied around the waist. A slave man was forbidden to cover his head with a cap or turban and to carry shoes. A female slave should not be veiled in contrast to free women and, preferably, carry a single piece of cloth. During the first part of the nineteenth century, both slave women and men were dressed in a white cloth that later became dark blue or black and called kaniki.18 Women tied it under their armpits while men carried it as a loincloth. As explained by Mtoro Bakari: “A slave was known by his dress, for never in his life he wear a cap . . . He never wore sandals not a kanzu long enough to cover his legs. . . . He did not protect himself from the rain with an umbrella. . . . Nor did he ever in the house wear clogs” (J. W. T. Allen 1981: 173). The restriction on slaves not to carry headdresses, even if they had converted to Islam suggests that, according to Laura Fair (1998: 74), “the class interest of the dominant members of society overrode religious prescriptions.” Clothing, as seen, became a way to express one’s status as free or enslaved. Slave women, especially wapambe or concubines, could be dressed in extravagant clothes and expensive bodily ornaments and exposed publicly to express their owner’s wealth. This sharply contrasted with the seclusion of his wives. Not surprisingly, former slave women revolted against this dress code when freed and used colourful and eye-catching textiles to communicate their new status. Some of the trade towns on the East African coast (Sheriff 1987: 13, mentions Mogadishu, Pate, and Kilwa), as well as some up-country peoples, had their own long tradition of weaving. It was shown in chapter 5 that imported cloth from India, Britain, and the United States were customized or re-fabricated in the ports or along the caravan routes to be er suit the tastes and demands of up-country consumers, which illustrates the active role these trade ports held between different trading networks. This was a cra that had existed in the ports for many centuries. Precious materials like silk or gold threads had been disassembled from imported cloth from India and China only to be rewoven into locally manufactured textiles.

The Kanga An illustration of the extent of the trading networks, the hybridization that occurred around the caravan trade in East Africa, and the rapid processes of change that took place in the coastal societies is the kanga, a rectangular cloth, which was used mainly by women in pairs and wrapped both around the body and the head/hair, which came into use around 1875 on the coast. The trading networks on and around the Indian Ocean

312

•

Muted Memories

have been captured in the expression the “Dhow Cultures,”19 which emphasizes the role of communication by means of sailing boats and, by extension, the role of the monsoons. The monsoons or “trade winds” carried people, trade goods, ideas, and cultural elements between Africa and Asia. As expressed by Phyllis Ressler (2012: 2), “The Kanga is best understood as a product of rich cultural exchange over centuries on the east African coast which has been an active partner in the world of the Indian Ocean for more than two millennia, and throughout its history has maintained deep cultural and commercial roots in the African interior.” Textiles used in East Africa have, as already mentioned, been produced both locally for centuries but were also imported, mainly from India, but also from Europe and America. The Portuguese, when they arrived on the East African coast brought with them textiles that would have a determining influence on local tastes and demands. One such textile was the colorfully printed rectangular handkerchief, or lemco in Portuguese. They were sewn together in East African ports three by two to produce a rectangular cloth locally called leso, occasionally also peso (Hamid 1996: 105). In a later development, twelve similar handkerchiefs were used to make two similar cloths since they were used in pairs, sometimes called a doti, which were used in the same way a kanga is used today. Originally the leso only had a central print and a printed frame on only two sides but a er a while the frame ran along all four sides. Where, why, and how this development occurred remains uncertain (Hamid 1996: 106). It may be assumed that the local demand for this kind of textile encouraged the local development of the manufacturing of the leso or kanga,20 which is a thin co on rectangular cloth with a central print and a symmetrical frame around the four sides. Gradually the name leso disappeared in the southern parts of the Swahili world and was replaced by the name kanga. Other influences in the development of the Swahili dress code came from central and northwestern Europe and places like the United Kingdom, the Netherlands, France, and Switzerland. The first textiles were block printed but with mechanization most were machine-printed, as the brightly colored cloth imported from Manchester, but with local designs to meet local demand. MacKenzie Moon Ryan (2013: passim) traces the genealogy of the kanga and finds influences endemic to East Africa as the a empts to improve the appearance of the sub-standard cloth imported in the wake of the disappearance of the high-quality merikani cloth during the US Civil War. This was fused with external influences that came through the import from the Netherlands of “Java print,” British Indian supply of colourful “Sari” textile, and cloth from central and Western Europe printed with a paisley motif. The influence from the eastern parts of the Indian Ocean li oral is also pointed out by David Parkin

Conspicuous Competitive Consumption and Communication

•

313

(2005: 48) when he writes that a kanga has the size of a sarong and is worn in a similar way. One possible explanation for the development of this dress code is that female slaves from the Manyema area in eastern Congo were brought by Stanley’s caravan westward to the Atlantic Ocean and sailed all the way around South Africa to Zanzibar. When they arrived they a racted much a ention because of their elaborate dresses, which they, being slave women as opposed to free women, could display in this Muslim society. Since the heyday of the caravan trade, East African women have used cloths for communication. One of the first statements made through textiles was, as mentioned, when former female slaves a er abolition started to carry colorful kanga textiles to communicate their new, free, status. They were also worn by women from mainland Africa to communicate their aspiration for inclusion in what Rose Marie Beck calls the “Swahili-Muslim East African society” (2000: 105). The kanga became instrumental in the creation of a female Swahili identity. On Zanzibar, where the kanga first seems to have gained general popularity, it served at least two purposes in this process, which coincided with the freeing of slaves. One was the conscious adoption of new dress codes to communicate the new status as free. The other was to also communicate a personal quality of uwezo or capability, a term used so separate the “‘haves’ from the ‘have nots’” (Fair 1998: 80). Many of the symbols printed on a kanga are borrowed from the worlds of plants and flowers. The prohibition for Muslims to reproduce humans or animals, since the task of creation, particularly of the human being, was reserved only for God, restricted what could be reproduced. A highly popular symbol printed on the textiles is the Persian or Kashmiri boteh or Sco ish paisley (named a er the shawls produced in Paisley, Scotland) motif. In India as well as in East Africa it is seen as a fertility symbol and occasionally referred to as a mango (Ressler 2012: 7). In East Africa it is also interpreted as a germinating bean or pea, a most powerful symbol of fertility. It can also be interpreted as a calabash with water or a cashew nut, which has given the symbol its name in Swahili, korosho. A kanga is usually designed with a frame running on all four sides and a central motif called mji (Hamid 1996: 106). Close to the frame on a long side, o en the lower side, a saying or mo o is o en printed. The first texts on textiles from the 1880s (Ryan 2013: 120) were reprints from the Holy Quran and in the Arabic script. One interpretation relates that this innovation was invented by women from Zanzibar or Mombasa. Another claims that a trader in Mombasa, Kaderina Hajee Essak or “Abdullah” Essak, a Sunni entrepreneur from what today is Pakistan (Parkin 2005: 41), introduced this novelty, which later developed into texts in Kiswahili

314

•

Muted Memories

in the Latin script. This development, which is an indigenous East African invention, the use of the script of the infidels, opened up a new kind of message printed on the kanga. The texts printed on a kanga usually took the form of a proverb, mo o, or riddle, which could be interpreted in different ways. It is very much in the eyes of the reader to interpret the message. It allowed, and still allows, women to express what otherwise would be impossible for them to speak about. One appreciated form of poetry was called mafumbo or veiled speech (R. M. Beck 2000: 108). Like the poems used by Bedouin women (see AbuLughod 1986), it is neither the specific women who opened their mouths to recite a well-known poem, nor the women in East Africa who carried a piece of cloth with a preprinted text who could be blamed for sending the message. Usually, one of the interpretations of a kanga’s text had clear erotic overtones or ambiguities (Horton and Middleton 2000: 197). The kanga became a multifunctional cloth o en used in a pair. One part could be used, in the public sphere, to cover the hair as a veil and the upper part of the body. The second cloth could cover the rest of the body. Thus, the kanga “simultaneously asserts both Islamic identity and local identity while enacting a paradox between Islamic demands for female modesty and Swahili traditions of sexual expressiveness” (Davidson 2012: 15f.). Women’s use of the kanga to communicate sentiments or comments on what took place around them was not uncontroversial. The cloth was arranged so the text appeared on the bu ock of the carrier. If a swaying walking style was added, many eyes would be a racted to the text, which, by extension, drew a ention to the female body. This was shameful in the eyes of many orthodox and elite Muslims. Prudent Muslim women should prove piety by covering their entire body with a buibui, which, a er the freeing of the slaves was permi ed for all women, when they le the domestic sphere. The kanga symbolized the cosmopolitan character of the coastal life at the time of the caravan trade. The textiles usually came from India or the United States, the printing techniques from India and Europe, the symbols from the Indian Ocean and the printed text from East Africa, and finally the message of the text was generated by the gender relations existing in the Swahili world.

Notes  1. The competitive character of social relations, power and wealth existed in the coastal towns long before the kind of caravan trade which this text focus on developed at the end of the eighteenth century. Leif Landberg (1975) shows

Conspicuous Competitive Consumption and Communication

 2.

 3.  4.  5.

 6.  7.  8.  9. 10. 11.

12.

13. 14.

•

315

that Kigombe, a small coastal community in northern Tanzania during the thirteenth to fi eenth century “was divided between proprietors (wenjeji) and strangers (wageni), men and women, kin and non-kin, and free and ex-slaves, all of whom competed with one another for status and power in rounds of weddings, funerals, dance competitions, and maulid throughout the year” (Landberg 1975: 353ff., as quoted in Spear 2000: 278). The big drums and side-blown horn, siwa, the most important regalia, are endemic to the Swahili world and “unknown as appurtenances of royalty in the Arab world but are typical of many, probably most, parts of sub-Saharan Africa” (de Vere Allen 1981: 315). This indicates, according to him, the African influence on the development of the Swahili culture. Classical Swahili towns were o en organized into two halves or moieties, but there is no strong evidence of this kind of organization in Bagamoyo. Also the Sukuma/Nyamwezi society was divided in this way in the two dance societies Bagika and Bagalu. While much of the political structure crumbled under the colonial system, the dance institutions seem to have survived and occasionally took on a resistance character. This central institution in the coastal societies also adapted to the colonial situation and borrowed elements from European military drilling and brass bands. Terrance Ranger has demonstrated how the development of the beni dance tradition in East Africa borrowed its name from the English “band” (Ranger 1975). Members of the beni groups usually dressed like British officials during their performances. The most famous groups, according to Mark Horton and John Middleton were the Kingi and Scotchi (2000, 190). Otherwise, he writes (Nimtz 1980: 98), very li le is known about the two dance societies. It is difficult to estimate their number thirty to forty years earlier during the peak of the caravan trade. Middleton (1992: 66) writes that pre-pubertal children were not allowed into the ndani, but slaves were—as non-persons. An apt illustration of Arjun Appadurai’s (1986) finding that material objects move in and out of the commodity status. Three hundred carved doors were found in Bagamoyo and the surrounding villages at the time of the conference. The sultan, however, married two women from Persia and, Judith Aldrick hypothesizes, probably also imported doors from that region, which may have influenced the door carvers to move away from the strict Ibadi ideals (Aldrick 1988: 88). The knobs on doors in India are said to prevent elephants from rubbing themselves against the doors, a preventive measure not needed on Zanzibar. Here the knobs became purely aesthetic. An alternative interpretation is that the chains, sometimes ropes, surrounding the doors represent safety (Areskough and Persson 1999: 29). Athman H. Athman (1996: 27) reports: “In all Swahili towns which are intensively engaged in wood carving, it is difficult to get people who can give convincing answers as to the interpretation of motifs. A curious researcher

316 15. 16. 17.

18. 19.

20.

•

Muted Memories

on Swahili carving will be forced to give his own judgement in as far as the meaning of motifs are concerned.” Actually, there were two versions of this chronicle; one is wri en in the Arabic and the other in the Portuguese language (Chi ick 1965: 276). Owens adds that the “purchase of land with cloth is a cliché in many coastal narratives” (727). From the horizon of old Bagamoyo the trade towns belonged to the “civilized” world, while, from the horizon of Zanzibar, all of mainland Africa was populated by the washenzi or barbaric people. This caused confusion when a new dress code moved through the Muslim world, which favored black clothes for free women. The title of a book by Abdul Sheriff (1988) Dhow Cultures of the Indian Ocean, Cosmopolitanism, Commerce and Islam captures the role of this sailing boat for the communication of taste, ideas, and identities and the transportation of people and goods. The dhow is a large wooden boat with one single sail arranged on a lateen rig. Kanga is the Kiswahili name for the colorfully do ed Guinea fowl and the term that is used in this text.

12  Intruders and Terminators The End of the Story The Nyamwezi traders/porters who brought their ivory to the coast met not only traders interested in buying their ivory and selling cloth and other trade items to them, but also more sinister visitors with the intention of changing the existing economic, political, sociocultural, and religious systems in the region.

The Urge to Discover, Document, and Discipline From the mid nineteenth century both the Nyamwezi and coastal traders and caravans met competition from an increasing number of caravans now also led by European and North American explorers and colonizers, some of them under the cover of being “scientific expeditions,” particularly in the search for the source of the Nile. One of the first explorers was Johann Rebmann, a German missionary, who in 1848 visited the slopes of Mount Kilimanjaro and could report on the snow at the top. This marked the beginning of what has been called the classical period of the European/American exploration of East Africa1 (Bridges 1973: 220). Among the first were also Richard Francis Burton and John Hanning Speke who started their expedition from the Bagamoyo area to the Great Lakes region. Other expeditions at the same time were led by Speke, again, who together with James Grant set out from the coast in 1858 and David Livingstone who, together with the German K. K. von der Decken

318

•

Muted Memories

marched to Lake Nyasa (Lake Malawi), and some time later, Henry Morton Stanley, who, in November 1874, set out from Bagamoyo with a caravan of three hundred people and eventually, in August 1877, reached Boma at the estuary of the Congo River at the Atlantic Ocean. Some undertook their expeditions during the second half of the nineteenth century under the label of being scientific expeditions or exploratory travels;2 others were missionaries, such as Livingstone and the French White Fathers who established their inland headquarter in 1878 in the Great Lakes region. Still others were employed by colonial agencies, as the German Karl Peters and Henry Morton Stanley, who worked for the Belgian King Leopold II and the organization International African Society.3 R. C. Bridges (1973: 220) notes that “the late 1870s were, in fact, a well-marked turning point in the story of European concern with East Africa. Instead of isolated explorers there arrived carefully organized parties of missionaries, treaty makers, concession hunters and traders.” One example of such a “scientific geographical” expedition was the “East African Expedition” of 1856–58, led by Richard Francis Burton and John Hanning Speke. They originally planned to set out from Bagamoyo, but changed the starting point of their venture to Kaole, probably because of their dependence on the Sultan of Zanzibar. A er the military clash in 1844 with the Zaramo people at Kaole, the Sultan had a much firmer grip on Kaole than on Bagamoyo. Officially, Burton and Speke departed with eighty to ninety employed staff in their caravan. The total number, however, was probably twice as high (see Wisnicki 2008: 130n10, who refers to D. Kennedy 2005: 108) since wives and children accompanied their husbands and fathers. The caravan was led by Said bin Salim, an experienced trader who had been appointed and supplied by the sultan, as had a dozen soldiers from his home area, the southern part of the Arabic peninsula. Two other prominent members of the caravan were Sidi Mubarak Bombay, a former slave, who a er a sojourn to India became the perhaps most famous caravanner in East Africa, and the Indian merchant Ramji and his ten “sons” or slaves (Wisnicki 2008: 116).4 Regardless of the reason behind their expeditions into the interior of the African continent, almost all intruders passed Zanzibar and came to Bagamoyo before they started their up-country march. During the race for colonies, the time before, during, and immediately a er the Berlin Conference in 1884/85, German, Belgian, and British “explorers” used Bagamoyo as their means of entry to the continent.5 One of them was Jérôme Becker, a lieutenant in the Belgian army, who came to East Africa in 1880 to serve the Association Internationale Africaine, at its station at Karema east of Lake Tanganyika, which was a territory that, a er the Berlin Conference, was

Intruders and Terminators

•

319

claimed by Germany. Becker was also tasked with finding a suitable place for a new Belgian post to the west of the lake in the Congo.6 His caravan may exemplify the kind of caravans whose main aim was colonization. He contracted 198 men, women, and children to work as porters, service staff, including four women who prepared his food, and soldiers.7 This figure did not include other people who followed the caravan, like all the women and children who accompanied their husbands and fathers. The women played many crucial roles as they o en obtained and provided the food for the porters and assisted in se ing up the night camps. Becker (1887: 1:147) found that “you can be certain that a traveler who returns with a rich harvest of information and of studies of customs owes most of them to the women in his escort” (quoted in J. Fabian 1986: 27). A caravan, hence, was a society on the road—a complex sociocultural institution during an early phase of globalization. The development of the industrialized modern world in Europe and the United States built on the heritage from the Enlightenment and the belief of the role of science in the organization and administration of the modern society (Bridges 1973; J. Fabian 1986; Foucault 1978; Rabinow 1984; Sco 1998). At this time there was a firm belief in the importance of observing, classifying, and documenting the world. It is therefore no surprise that we meet endless descriptions, classificatory schemes, maps and illustrations in the travelogues of the early travelers who visited the African continent. The information needed by the modern state, its statistics, was collected everywhere. “One is tempted,” write Christoph Brumann and David Berliner (2016: 12), “to see this through a framework of Foucauldian governmentality (Foucault 1991) where only what is known, measured and rationally processed can be governed.” These “explorers” contributed to what Johannes Fabian (1986: 24) calls the in-scription of these o en remote areas and peoples, from a European point of view, in the expanding colonial enterprises. Many of the British explorers, such as Burton, Speke, Livingstone, and Cameron, had financial support from the Royal Geographical Society, but o en also from the British Government (Bridges 1973: 222). It may be assumed that the purpose was not only to fill in the “blank spots,” from the perspective of the Europeans/Americans, on the continent, or to bring light to the Dark Continent but also to provide “statistics,” first for an economic colonization in terms of suitable markets for industrial products and valuable export products from the continent, and later for political colonizing enterprises. This is clearly spelled out by one of the most wellknown “explorers” who described his travel from the Bagamoyo area with the following words: “We were also opening for Europeans a new road into the heart of Africa, a region of boundless commercial resources”

320

•

Muted Memories

(Burton 1872: 2:295). In a later phase the “explorers” also began to search for potential areas suitable for “white se lement.” The increased number of caravans marching up and down the territory and the multitude of negotiations over hongo or tributes led to inflation in the taxes the caravans had to pay. Sarah Fee, quoting Becker (1887: 2:221, 470), reports that: In the 1880s, it was advised to plan for gi ing 4000 yards of merikani and colored cloth for the entire passage of a caravan of 28 from Zanzibar to Ujiji. . . . Stanley (1891: 443) blamed missionaries and Arab traders for having turned the trail to Lake Victoria into a “highway” by the 1880s. Missionaries he particularly castigated for inflating hongo by naively paying out 1200 yards to a single chief where 20 years previously he had paid closer to 200 yards. (Fee 2012: 6)

Other changes also took place along the caravan routes due to militarization, which followed as a result of the passing caravans. This was partly the result of a growing resistance from the people along the routes against the extra burdens on food, water, and other resources caused by the thousands of passing people. Fabian, quoting Becker (1887: 1:312), who reported that: The time is gone when Arabs went from the coast to Ujiji [on Lake Tanganyika] with nothing more than a walking stick in their hands and when the Mouzoungou [the white man] was taken for a supernatural being and inspired fear and respect merely by being present in a caravan . . . Most of the Negroes along the road know today that the only advantage we have are our manufactured goods and the superiority of our weapons. (J. Fabian 1986: 28)

Stereotypes in the Intruders’ Caravan Culture The iconography of the “white” man/men struggling to penetrate the hostile and unknown interior of the continent is highly misleading. Many of the “explorers” traveled under the sultan’s red flag, with his guides and military escort and followed, rather than led, experienced guides from Eastern and Central Africa, marching on the well-trodden paths, or what Burton termed “a beaten path” from the coast to the interior of the continent (Burton 1872: 2:292). Fee (2012: 3) writes that: “Instead of hacking through African wilderness, Europeans were guided along well-worn paths, some deemed ‘highways’ by the 1880s.” Further south the “explorers” could also use early Portuguese sources on the trade between Sofala in Mozambique and Zimbabwe and the Lake Nyassa (Malawi) and Lake Tanganyika area. Still, the European/American caravan leaders developed a jargon in their travelogues that elevated their own achievements in “the opening up of Africa” through the contrasts they painted with regard to the trou-

Intruders and Terminators

•

321

bles they encountered in their relations with the caravan porters. Burton (1872: 2:320) glorified his own role in the following words: “I led the most disorderly of caravans into the heart of Intertropical Africa, and succeeded in discovering the Tanganyika, and the southern portion of what is now called the Victoria Nyanza Lake. The road was thus thoroughly laid open.” It has been shown, regarding the stereotype of the lonely hero who struggled onward in a hostile environment, as in the case of Burton and Speke’s expedition, that the “explorers,” traders and missionaries were heavily dependent both on the support and supplies of Indian traders who operated on the markets in the sultanate on Zanzibar, and on the already established caravan routes and stations along it. “Solitary explorers,” writes Johannes Fabian (2000: 29) “never traveled alone” and Fee adds: “Before se ing out from Zanzibar, Europeans invariably solicited information and advice from Arab (and Indian) merchants. Their dependency increased thousand fold once they le the coast” (Fee 2012: 2–3). In many cases they were in the hands of the political leaders through whose countries they had to pass. Many reports from these “explorers” clearly indicate that they, and the rest of their caravan staff were temporarily hostages while the local leaders waited for the gi s and the payment of the compulsory tax, hongo, demanded for the caravan’s right to pass, camp on, take water and wood for fuel, purchase food, and eventually leave the territory. Harry Johnston provides a dramatic, but probably over-exaggerated account when he writes: “Our progress was repeatedly hindered and occasionally stopped by the ever-increasing and turbulent force of savages who pressed upon our flanks, claiming for a division of our goods . . . ‘Ngubo!’ [cloth] was their sole cry” (Johnston 1886: 288)8 and Joseph Thomson lamented: “It will be useless to recount the next day’s proceedings. We certainly had to give, give, give, until we had barely a cloth le , or an article of any description” (Thomson [1881] 1968: 161). The road toll became, through the jargon of the European and North American travelers, the spoil extracted by blood-sucking local African leaders who robbed the traders and “explorers” of their valuables. A careful examination carried out by Carol Jane Sissons (1984) of the reports from the caravan leaders who passed through Gogoland in central Tanzania during the second half of the nineteenth century gives a different picture. Between 1 to 2 percent of the total value of the caravans’ merchandise was paid as road tax to the five to eight chiefdoms in Gogoland through which the caravans passed. The travelers rarely understood that the hongo was part of a ritualized system of exchanges that opened up subsequent trade in food and other items and that the mechanisms for redistribution in the local economies meant that the whole village rather than the individual chief benefi ed from the system.

322

•

Muted Memories

To organize a caravan from the coast was a time-consuming, arduous, and expensive task.9 Many times the European “explorers” and traders were inadequately equipped by their home governments, organizations, or companies and had to rely on loans from Indian financiers, and traders as well as local instructors and advisors. The advisors’ information was essential, since up-country rulers had different tastes and their people had various demands regarding the type, quality, and color of the cloths (see chapter 5 for further information on the “cloth economy”). A specific subculture developed among the caravan leaders. The “explorers” and colonizers came with different nationalities during the scramble for Africa, but they o en came as officers with a military background and they “produced remarkably similar accounts of their travels” (J. Fabian 2000: 37). It seems as the caravan business pulled out some of the most negative sides of humankind in the form of open racism, violence, and arrogance. The military background of the European caravan leaders probably contributed to the common presentation of the caravan as a large number of people marching in a file. In reality, however, the expeditions o en took the form of a swarm, which passed through the landscape (J. Fabian 2000: 40). One part of the explorers’ subculture was their concern, or perhaps rather obsession, concerning the material culture of caravaneering. A rather stereotypical picture emerged on what was considered as essential or necessary to bring along with the caravan into the “Dark continent.” For example, Jérôme Becker, already referred to above, listed in his text more than two hundred different items necessary for his expedition of two hundred men. Even the missionaries, who o en came to East Africa in order to save slaves and souls easily, sometimes to their own disappointment and despair, were trapped in this subculture. A newly arrived missionary soon realized that “the most dramatic and demanding of these secular tasks was also the one that first confronted each newly arrived missionary, the organization and running of a caravan into the interior” (Beidelman 1982: 601). A text that contains some of the most outspoken statements of the “missionaries’ ordeals” is In the Heart of Savagedom: Reminiscence of Life and Adventure during a Quarter of a Century of Pioneering Missionary Labours in the Wilds of East Equatorial Africa (n.d.), wri en by Mrs. Rachel Stuart Wa and edited by her husband. Under the subtitle “Difficulty of Conquest and Problem of Punishment” she writes: There is no doubt whatever that disobedience and desertion on the part of the natives, when on the march through Africa, may not only frustrate the object of the expedition, but also endanger the life of the European and those

Intruders and Terminators

•

323

of the caravan. Hence it is a ma er of vital importance to know how best to treat an aggressive spirit of disobedience, when it is first manifested, and nip it at the bud. (Wa n.d.: 97)

The most frequent punishment to nip disobedience at the bud was flogging. Mrs. Wa reports: “with regard to the ma er of flogging the natives, the most eminent travelers in Africa agree that corporal punishment is inevitable in any caravan” (Wa n.d.: 100). The widespread use of flogging and shackling porters in front of their peers as a means of disciplining the whole troop made T. O. Beidelman conclude that “the task of caravanning proved to be highly threatening to the missionaries’ self-image, an idealized view that they had to maintain at all costs in order to endure the difficult lives they led” (Beidelman 1982: 601). The frequent use of extreme violence against the porters is well documented in Johannes Fabian’s book Out of Our Minds (2000). The development of a specific subculture among the European/American caravan leaders in relation to their employed or slave porters probably limited their potential to lead caravans beyond the well-established caravan routes. The European and North American intruders seem to have developed a firm belief in the use of physical violence as a means to prevent desertion, rather than to find out why desertion was a ractive. All missionaries could not subscribe to the subculture developed among the caravan leaders. Mrs. Wa reports: I knew one Missionary, who, when he entered Africa for the first time, was so compassionate towards the natives, that he could not bear to see them sweltering under the burdens which they were carrying for him, and alternately relieved one porter a er another by carrying their load on his own shoulder. That man was so unsuccessful in his work as a Missionary, that a er some years he gave it up altogether. (Wa n.d.: 99, my emphasis)

Intruders and Colonizers from Asia The delicate power balance negotiated between the people of Bagamoyo town, the Wabagamoyo, and the surrounding Zaramo people described in chapter 9, was disturbed when external actors began to show interest in the town and the trade. Among the first to do so were traders from the Indian subcontinent. Their influence, however, was mainly restricted to the economic life of the town,10 since they rarely became involved in local political life. This business class, however, was quick to react on developments that were negative to their trade, but their reactions were primarily channeled through the British Consul for Zanzibar. As British subjects

324

•

Muted Memories

they used this resource and the Consul’s close contact with the Sultan of Zanzibar to defend or promote their own interests.11 Only rarely did they involve themselves in the local political life.

In the Shadow of the Sultanate More direct interference with life in town came through the gradual establishment of the Sultanate of Zanzibar. A first direct confrontation between people on mainland East Africa and the expanding Sultanate took place in 1844 at Kaole when the Zaramo people met the Sultan’s soldiers in an armed clash. The subsequent establishment of a garrison with Baluchi mercenaries at Kaole was meant to tighten the Sultan’s grip over this frontier area. This move may have helped the Kaole traders, as already mentioned, in their decision to move to the expanding Bagamoyo town. The booming caravan trade in Bagamoyo caught the sultan’s eye and the sultanate tried to strengthen its presence and domination over the town and the caravan trade. The established power balance, however, exhibits the characteristics of a frontier area with vaguely demarcated or defined fields of influence and responsibility (see Tsing 2005: chap. 1). The sultan’s claim, for example, that all caravan trade should pass his customs houses on the mainland, could be circumvented by the caravan traders and porters by directing their trade to Saadani, 45 km north of Bagamoyo, which never came under the strict control of the Zanzibari Sultanate12 (Glassman 1995: 64–68). It was shown in a previous chapter that the former leaders of the coastal towns, the merchant-middlemen between the continental caravan trade and the transoceanic trade on the Indian Ocean challenged the sultan’s power over the coastal strip. A clear indication of the weakness of the sultanate was when Sultan Majid, in an a empt to take control over the lucrative trade, tried to establish his own port and capital in Dar es Salaam and a ract the caravan trade to this port. The a empt failed, which also illustrated the limited power he held over life in the coastal communities. When Sultan Majid died in the late 1870s, the succeeding sultans did not repeat his efforts, but had other means at their disposal to try to control the caravan trade in Bagamoyo, like placing their own representatives in or near the troublesome town. Since Bagamoyo’s traders refused to move to the sultan’s new port in Dar es Salaam, then the sultan had to move his presence closer to Bagamoyo’s traders. During the second half of the nineteenth century the sultanate managed to exert increasingly stronger control over Bagamoyo. This was achieved through the Indian financiers and traders, who controlled the resources and goods needed to set up and equip large-scale trade and “exploration” caravans.

Intruders and Terminators

•

325

However, not all actors on the scene wanted to see or to admit the rather limited extent of control the Sultan of Zanzibar had over areas on mainland Africa beyond a few specific port towns. European colonialists, at an early stage of the scramble for Africa, o en exaggerated his influence when they needed the sultan as a negotiation partner or his signature on an agreement (see S. Fabian 2007b: 41–42). The Zanzibari Empire, as it was seen by the Europeans at the end of the 1870s, is described by Frederic Elton in the following way: About one thousand miles of this coast, from Cape Delgado in the south to Warsheikh in the north, is Zanzibar territory. In many places the Sultan’s jurisdiction does not extend very far inland; but his influence is great even on the shores of the Tanganyika and Victoria Nyanza lakes, and his governors are recognized in more than one large Arab station far in the interior. (Elton 1879: 15)

The actual control that the sultan held over the area, including Bagamoyo, is questioned by Steven Fabian (2007b) in his doctoral thesis “Wabagamoyo: Redefining Identity in a Swahili Town, 1860s–1960s.” He shows that the orders and decrees issued by the Sultan were o en disobeyed or ignored by the inhabitants of Bagamoyo. In fact, even the sultan’s representatives, like his jemadar (military leader) formed close bonds of mutual respect and interest with the people of the town. This forced the sultan to appoint another representative, his liwali, as a superior to the jemadar, but it seems like the liwali soon also forged his own bonds with the denizens of Bagamoyo (see S. Fabian 2007b: chap. 2). His control was also questioned by some of the contemporary visitors to East Africa. Sir Bartol Frere, a British colonizer and diplomat, undertook a short journey in the early 1880s on mainland East Africa to investigate the sultan’s control over the area and found that “I knew his authority did not extend far inland, but I was not prepared to find it so entirely confined to a few ports on the coast, and that even at some of the more important of these ports his garrisons are hemmed in by the pe y chiefs of neighbouring tribes” (quoted in Woulfin 2011: 59). However, it may be misleading to analyze and judge the sphere of influence of the sultan through the prism of a territorially bounded nation-state. We ought to look at this entity as a more amorphous polity resembling the medieval feudal state. The control over a territorially demarcated piece of land was only secondary to the control over some of its inhabitants and, in particular, to some of the economic activities, which may be subject to levies, taxes and duties. The loose economic, political, and cultural networks that radiated from Zanzibar and out over the Indian Ocean as well as over the coastal zone of mainland Africa lacked the characteristics of a modern state, such as political cohesion, population control, disciplined bureau-

326

•

Muted Memories

cracy and so forth (Woulfin 2011: 3) or, what Michel Foucault has called “governmentality.” The political system in the port towns was much more a network based on personal master and client relations. Thus, interpreting the construction of the sultan’s customhouses in the major ports from Somalia to Mozambique as evidence of his political control over this vast area can be misleading (S. Fabian 2007b: 41). Still, the customhouses were absolutely crucial in a mercantile economy based on trade rather than production. They were the foundation of the sultanate’s economy during the la er half of the nineteenth century. In Bagamoyo, there was perhaps no urgent need for the sultan to extend his control far beyond the coast and the customhouse. Still, diplomats and traders from Britain, Germany, and France all swarmed around the Sultan’s court in Zanzibar Stone Town in their efforts to persuade him into entering agreements that favored their states, traders, and citizens. For the British it was also an issue of having the sultan sign agreements on the abolition of the slave trade. The sources disagree on the sultan’s control over Bagamoyo. Walter Brown (1971a) finds that it was minimal and that the town’s traders continued to control much of the trade. Jonathon Glassman (1995), on the other hand, argues that the sultan had considerable control over trade. Steven Fabian (2007a) suggests that Glassman’s conclusion may depend on his focus on the role of Sultan Bargash who, in contrast to the other sultans may have developed a certain amount of control through his personal relationship with individual members of the town’s elite. The role of the sultanate seems, however, to have been minimal. In an a empt to strengthen his control, the sultan established a garrison with Baluchi soldiers at Kaole. Many rulers have preferred to employ foreigners or slaves, who are without kinship relations, in their armies. People without kin or patron-client connections with the rest of the population are believed to be more reliable when it comes to upholding or implementing the ruler’s will. However, this did not prevent the Baluchi men in Kaole to be involved in local politics and economic activities, as the caravan trade, or to intermarry with local women and establish themselves in the local communities (W. T. Brown 1971a). The prime role of the militarization of the trade towns was to protect the Sultan’s economic interests, which were of different kinds but two are of particular interest for this work.13 One concerned his business and to optimize the trade in the port towns under his control. The other was to protect the loans and business in the hands of the Indian financiers and especially punish those who could not pay their debts or who broke their contracts. The role of his soldiers and their officers was to capture and sentence them either to jail or slavery, an issue which has been further explored in a previous chapter.

Intruders and Terminators

•

327

The second intervention from the sultanate to try to control the caravan trade, besides through the establishment of customs houses at the terminal or entry port of Bagamoyo, was a grandiose plan to establish military posts every three leagues, or every 15 km, along the central caravan route from Bagamoyo to Ujiji. To further improve the service and strengthen the control of the sultanate over the central caravan routes he also planned to construct wells at the same intervals (W. T. Brown 1971b: 267; S. Fabian 2007b: 77). Very li le actually came of this, but his a empts to control trade must have been seen as a threat, not only against the caravan porters/traders, but also the business community in Bagamoyo. One of Sultan Bargash’s further moves to increase his control over both Bagamoyo town and the caravan trade was to install his own representative in Bagamoyo, his liwali or “Governor.” Most o en the governor was a member of the sultan’s own clan, but foreigners could be also appointed, for example, Sabri bin Safir, a Baluchi, was appointed by Sultan Majid to be both the Liwali wa Waswahili, the governor of the Swahili people, and Jemadar Mkuu or the supreme commander in chief (W. T. Brown 1971a: 259). The sudden appearance of a representative of the sultanate disturbed the former power balance in Bagamoyo and other coastal towns. The formerly self-governing towns suddenly felt the influence of the Sultan of Zanzibar. Jemadar/Liwali Sabri bin Safir became instrumental in the strengthening of the sultan’s grip of Bagamoyo. When the sultan, a er a fatal incident when a prisoner was a acked by a leopard, learned that Sabri kept the criminals in devices called stocks, which like chains immobilize the prisoner, under open sky in a compound near the barracks in Kaole, he summoned Sabri and asked for an explanation. “Mimi sina gereza”—I have no prison, he answered. Shortly a er that, sometime during the 1860s, the first prison on mainland Africa was built, not in Kaole but in Bagamoyo on a spot close to where the present District Office is located, and a Baluchi Jemadar called Isa bin Kunari was assigned to it (W. T. Brown 1971a: 261– 62). The prison was a strategic institution in the protection of the interests, particularly economic, not only of the sultan but also, for example, of the Indian bankers and businessmen in Bagamoyo he had invited to support the most heavily financed caravan routes. Mtoro (J. W. T. Allen 1981: 201) described how the prisons built by the Arabs had two sections: one for debtors, who a er punishment by the stick were locked up in the “debtors’ room”; and another room for criminals who had commi ed serious offences. Those locked up in this chamber were also chained. It is possible that the difference in punishment was based on class and status. Freemen traders, although in debt, may not have been chained but put in custody to remind others to repay their loans. Porters were probably chained, perhaps as a first step to be enslaved and sold.

328

•

Muted Memories

The ties between Bagamoyo and Zanzibar were not very tight. Bagamoyo was visited only once by a sultan during the precolonial period. It seems safe to assume that the terrible tropical cyclone, which hit both sides of the Zanzibar Channel on 15 April 1872 (W. T. Brown 1971a: 266; S. Fabian 2007b: 77; Kollman 2005: 154), helped the sultan in his decision to visit what he must have considered his economically most important place on mainland Africa. The cyclone destroyed two-thirds of the clove and coconut plantations on Zanzibar. About 150 Arab and Indian boats, many of them full of cargo, had either sunk or were wrecked on the shore. Sultan Sayyid Said’s navy—three steamers and a corve e warship—were stranded and wrecked. It has been suggested (Woulfin 2011) that this caused an economic disaster of the sultan’s economy, which contributed to his willingness to sign the 1873 anti-slavery agreement with the British, an agreement that became a first step in the colonization of Zanzibar. In Bagamoyo, the cyclone destroyed fi y houses and the agricultural scheme that the Catholic Fathers had tried to establish through the labor of the freed slave children. The sultan arrived in Bagamoyo six months a er the cyclone with parts of his court. The visit, however, turned out to be a suitable illustration of the frictions that had developed between the Zanzibari Sultanate and the leadership in Bagamoyo. The sultan’s representative in the town, Liwali Sabri bin Safir was expected to host the sultan and his entourage. Sabri, however, did not welcome the sultan with the expected “traditional Muslim—and especially Baluchi—hospitality” (W. T. Brown 1971a: 268; see also S. Fabian 2007b: 77–78). The visit turned out to be a clear demonstration of the tensions, which had developed between the two centers of power, and an open reaction to, what must have been interpreted as, an intrusion by the sultan in Bagamoyo’s economic and political affairs. The liwali had to balance the interests of his superior against the resistance from his subjects.

German Colonization We have seen in a previous chapter that Karl Peters, the Director first of the Gesellscha für Deutsche Kolonisation or Society for German Colonization, later of the Deutsch Ostafrikanische Gesellscha DOAG or German East Africa Company had visited the Bagamoyo hinterlands in 1884 to sign treaties le , right, and center with local leaders, which formed the basis on which the Germans claimed the right to govern these territories. It is established that Peters and his assistants had accumulated 140,000 square kilometers as their territories. A crux for the realization of their

Intruders and Terminators

•

329

colonial dreams, however, was that an agreement from 1887, based on the work carried out by a Boundary Commission, with members from Britain, Germany, and the Zanzibari Sultanate, granted the sultan a 10-nauticalmile wide strip along the coast, in which Bagamoyo was situated. The claim of the sultanate on this part of the coast was not accepted by the German colonizers (Woulfin 2011: 95). Bagamoyo, the pearl of the coast, was strategically selected first by the DOAG to serve as its headquarters and a er the formal colonization of German East Africa also the colony’s first capital. In both capacities the town became the entrepôt for European colonial aspirations in this part of Africa. The formal German colonization took effect from 1 January 1891, when the German Reich declared, a er the chaos caused by the German trading company, that the area should be the Schutzgebiet Deutsch-Ostafrika or the Protectorate of German East Africa. Racial distinctions became more prominent in the administration once the German state took over from the German trading company. Theirs was the distinction between die Weissen or white and die Farben or colored persons. This was made into the overriding distinction during colonial time; the one between watu wa weupe (white people) and watu wa weusi (black people). Both the Germans and the British drew a sharp line between the “white”—a term applied to Europeans, Indians, and Arabs—and the “black Africans.” A consequence was that the middlemen/patricians in

Illustration 12.1. Participants at the 2002 International Conference viewing the German Boma. Photo by the author.

330

•

Muted Memories

Bagamoyo, o en calling themselves the “Shirazi,” who had long nurtured the idea of a non-African origin, now became “black Africans” (Nimtz 1980: 50). Kelly A. Askew (1999: 72) finds that the “colonial-era scholars conceptualized the coast largely in Arab terms due to their inability to reconcile the sophistication and cosmopolitanism of coastal cultures—its mosques and coral architecture, women in purdah, elaborate carving and poetic traditions—with their evolutionist, racist framework.” It may be assumed that their worldview was tainted by evolutionary ideas about development in stages, which painted the Arabs as more advanced or developed than the Africans, a worldview which resembled the view nurtured by the Arabs and other coastal people on whom the Germans depended for the administration of their vast colony. The new colonial regime was certainly felt by those affected, as the people who lived close to the administrative and military centers established by the Germans. Although the influx of Europeans to the new colony was slow; the number reached 1,473 in 1904 and had doubled to 2,845 in 1908, the colonial structure affected a growing number of inhabitants in the territory. In 1908, the Colonial Undersecretary (or Colonial Minister) Bernhard Dernburg, explained the situation in words that now provide a vivid illustration of Steven Luke’s third dimension of power:14 We must introduce and keep up an energetic, just, trustworthy administration, and we have to teach the people that they gain an advantage from German rule. It is very difficult to make them understand this, because the advantages they had, so far, have been small in proportion, compared to the disadvantages the German administration has imposed on them, according to their own perception, concerning the alteration of their traditions, taxation, control etc. (The World at War, n.d: 8)

The decision to make Bagamoyo the first German colonial capital meant a militarization of the town. The German colonizer Richelmann redesigned the physical town plan when Bagamoyo was rebuilt a er the 1888–89 rebellion (S. Fabian 2007b: 164–65). The militarization took the form of the construction of five defense towers constructed by von Wissman to protect Bagamoyo’s entrances. The “German Block House” on the western side of the Old Stone Town, is one of the five block houses mistakenly identified in the presentation of Bagamoyo on the (World Heritage) Tentative List as a structure remaining from the slave trade. The German occupation of Bagamoyo was sensed in the cityscape. Some of the prestigious stone houses were taken over by the Germans. The old Arab Fort15 constituted a central scene in the 1888 rebellion. The sultan’s representative, his liwali, refused to evacuate the house in favor of

Intruders and Terminators

•

331

the Germans and hand over the sultan’s red flag to the representatives of the German East Africa Company, but the Fort was eventually evacuated and taken over by the German administration. Other previous trade/store houses in Bagamoyo, such as the Liku House on what the Germans called Kaiser Strasse and the Arab Tea House on the same street were taken over by the Germans and used as administrative offices and/or residential buildings. Still others, like the two-storied building at the Caravan Serai, the Customs House at the shore, and the Post Office, the first on mainland East Africa, were built by the Germans to meet their economic and political needs.

The Abolition of the Slave Trade In the last decades of the nineteenth century the freeing of slaves took a new turn and dimension when colonial powers, primarily Great Britain, used both diplomatic and military means to negotiate agreements with local leaders. In the Indian Ocean region, they focused on negotiations with two of the leading political powers; the king of Merina on Madagascar and the Sultan of Oman, from the early 1830s located on Zanzibar. In the year 1822, the first agreements were signed with both King Radama I of Merina and Sultan Seyyid Said of Oman. The agreement between the British and the Sultan of Zanzibar forbade the export of slaves to “Christians” south of Cape Delgado, a place close to the border between present-day Tanzania and Mozambique, that is, to the French and Portuguese colonies in the southern part of the Indian Ocean region (Koponen 1988: 85; Ryan 2013: 73). The 1822 Morseby Agreement with the sultan banned the exportation of slaves from the areas controlled by the sultanate. It did not ban slavery as an institution, which, paradoxically, grew in the sultanate with the banning of the exportation of slaves. It has been estimated that this ban deprived the sultan of annual revenue of more than $50,000. He compensated this loss with a feverish involvement in clove plantations on the two major islands of Zanzibar: Unguja and Pemba. These plantations depended on slave labor, which increased the internal trade and use of slaves. Another agreement, the Hamerton Treaty, which was signed in 1845 but came into effect in 1847, made the export of slaves illegal north of Lamu, a port town in northern Kenya, which meant to the northern parts of the Indian Ocean region. The sultan’s own “dominions” on Zanzibar and along the coast were, however, exempted. Here a total abolition of slave labor and slavery would mean a disaster to the growing plantation

332

•

Muted Memories

economy. The effectiveness of this ban has also been questioned and Frederick Cooper (1975: 91) writes: “In the first season a er the treaty, Arabs took few slaves to Arabia, but by the next year they were carrying slaves again.” This development caused friction in the Zanzibar-British relations, which resulted in the 1873 agreement on banning the trade in slaves, which “was signed into law only a er the British naval force threatened to blockade Zanzibar” (Ryan 2013: 73). The agreement was perhaps not based on a mutual understanding of principles of human rights and freedom. Stanley, in jargon well-nurtured among the “explorers,” wrote about Sultan Bargash, who was the one to eventually sign an agreement with the British, that: “here we see an Arab prince, educated in the strictest school of Islam, and accustomed to regard the black natives of Africa as the lawful prey of conquest and lust, and fair objects of barter” (Stanley 1899: 32).16 His signature on the agreement to abolish the slave trade was probably not wholehearted but the result of British gunboat diplomacy and Stanley could report that “when the Prince’s reluctance to sign became known, the fleet under Admiral Cumming made its appearance before Zanzibar, and by process of gentle coercion, or rather quiet demonstration, the signature of the Prince was at last obtained” (Stanley 1899: 32). It has been suggested that this controversy over slavery as an institution (R. B. Allen 2012: 14) contributed to the establishment of Zanzibar as a British protectorate in 1890. It has also been suggested that the prohibition on the transportation of slaves from mainland Africa to the Zanzibar islands fueled the expansion of the plantation economy along the coast. The highly labor-demanding plantations were based on slave labor. One consequence of the agreement was that the slave economy moved from the islands to the coastal areas where slaves were in high demand on the coconut and grain plantations around Bagamoyo and the sugar plantations at Pangani (Koponen 1988: 92). Another was, as already mentioned, that slaves were marched northward on land from Kilwa, the most important slave trade town on the East African coast, some all the way to Barwa in present-day Somalia. The changes in the transportation of slaves were noticed by the British who pressed Sultan Barghash to issue a new decree which stated that: “In disobedience of our order, and in violation of the terms of our Treaties with Great Britain, slaves are being constantly conveyed by land from Kilwa, for the purpose of being taken to the island of Pemba. Be it known that we have determined to stop, and by this order prohibit, all conveyances of slaves by land under any conditions” (quoted in Woulfin 2011: 77). It has also been suggested that the inspections carried out by British ships of the dhows changed the means of transportation of slaves over the

Intruders and Terminators

•

333

Zanzibar Channel from mainland Africa to the islands Unguja and Pemba (Mackenzie 1895: 17). The small, but fast sailing outrigged canoes with a lateen rig called ngalawa were now used for this purpose. The abolition of the slave trade is usually associated with the treaty signed on 5 June 1873, by his Highness Seyyid Barghash and the British Consul to Zanzibar, Dr. Kirk.17 The treaty directed the slave market on Zanzibar to close and stipulated that the shipping of slaves was to be illegal. The Proclamation, affixed upon the doors of the Custom House on 8 June read: Proclamation (Translation) From Barghash bin Saéed. To all our subjects who may see this, and also to others, may God save you. Know that we have prohibited the transport of raw slaves18 by sea in our harbours, and have closed the markets which are for the sale of slaves through all our dominions. Whosoever, therefore, shall ship a raw slave a er this date, will render himself liable to punishment, and this he will bring upon himself. Be this known. Dated 12 Rabi el Akher, 1290 (June 8, 1873) (in Elton 1879: 59)

This interference in the local economy and politics was not well received in all parts of what the sultan and the colonizers regarded as “the Sultan’s dominions.” It was a threat against the plantation economy and part of the caravan trade, which either used slaves as porters and/or as the main trade item. It is therefore no surprise that the sultan’s decree caused resistance and further challenges regarding his right to interfere in and decide over societies on the mainland, especially from the plantation owners and slave traders on the coast of East Africa who had long since questioned the sultans’ claims of authority over these areas. A contemporary observer notices that: The Coast Arabs have told me that they do not care a bit for the proclamation. Sayyid Barghash has no power to enforce it; and if it were not for the English support he has, they would soon turn him out, and get rid of the infidels altogether. His power is only nominal, and he is only obeyed as far as his people choose to obey him. The slave-trade is as rampant as ever now. (Anti-Slavery Report 1877: 232, quoted in Woulfin 2011: 78)

The progress of the implementation of the treaty was, for various reasons, slow and the export of slaves from mainland Africa to Zanzibar continued. The British pressed for stricter measures to be taken which made the newly seated Sultan Sayyid Ali bin Said publish a new Decree on 1 August 1890. The first paragraph stated: “We hereby confirm all former decrees and ordinances made by our predecessors against slavery and the slave trade, and declare, that, whether such decrees have hitherto been

334

•

Muted Memories

put in force or not, they shall in future be binding on ourselves and our subjects” (Hill 1957: 22). One month later the Sultan’s own representative in Bagamoyo, Liwali Suleiman bin Naser, challenged the Sultan’s Decree through his own Decree of 6 September 1890, probably with the intention to support the plantation economy and stimulate the rese lement of the devastated town, which stated that: Be it known to all that we give permission to everybody who has got land in Bagamoyo and Shenzi (three or four days’ march around Bagamoyo) to recover and retain their slaves, and everybody who possesses slaves has permission to sell his slaves to the people of Bagamoyo, but it is forbidden to ship the slaves by sea. We desire that the shamba [plantation or field] owners should begin working their shambas without delay, because it will be good for the people and the town. (As quoted in S. Fabian 2007b: 172n52)

The decree caused an international scandal which forced the Germans, since Bagamoyo was now a town under the German sphere of interests in East Africa, to react through a quick denouncement of the proclamation. However, they “abstained from posting copies of Sultan Ali’s anti-slavery decree, and carrying out its measures in the German colony” (S. Fabian: 2007b, 174–75). Their reasons for reacting in this way seem obvious. Well aware of the profit generated through slavery on the slave plantations, they acted to protect the economy.19 The German Foreign Officer asked representatives on Zanzibar “to calm down the slave owners [by informing them that] we do not intend by our anti-slavery measures to interfere with current relationships” (S. Fabian 2007b: 172n51). This may seem cynical since one of the arguments the Germans used when they sought British support for the proposed blockade of the ports on the Mrima coast during the Bushiri rebellion 1888 and 1889 was to end the Arabic slave trade. The German colonial administration was “very reluctant” (S. Fabian 2007b: 167) to abolish slavery, although they, like many other European powers, used the suppression of slavery argument to justify their colonization efforts. Slavery in German East Africa was never abolished. A Parliamentary Act was signed by Kaiser Wilhelm II in 1895, which outlawed both slavery and the slave trade, but this act only affected Germans in the colony. It did not apply to the “natives,” Africans, Indians, and Arabs. It was first a er the British had taken over the former German colony a er World War I as a Mandate under the League of Nations that slavery was abolished in 1922.20 In other words, slavery was still legal in the German colony a er twenty-five years of colonial rule (Deutsch 2006: 97). The Germans, however, were not passive observers of the slave trade and slavery. Instead of imposing a ban on slavery, they tried to reform the institution by integrating it into the colonial structure. Slavery was

Intruders and Terminators

•

335

thought best regulated and controlled by the colonial state in a governmental manner, which also collected taxes from the slave trade (Deutsch 2006: 97–101). A reason for this a itude was the efforts that had gone, by apologists, into the painting of the African “domestic” form of slavery as more benevolent than the US plantation system, making it more acceptable. It has been hypothesized that the portrayal in early colonial accounts of the African forms of slavery as more human was the result of a broad understanding of the importance of slavery for the production systems in the African colonies (see, for example, Deutsch 2006: part 2). To avoid a total abolition, as demanded by the 1890 Brussels agreement on the abolition of the slave trade, the German colonial power tried to reform slavery in its colony. They only imposed control over the trade with slaves, not a ban on slavery (Deutsch 2006: 98–101.). The economic importance of slavery may explain the Germans’ adoption of what Jan-Georg Deutsch (2006: 98) observes as a piecemeal approach toward the abolition of slavery. The Germans imposed an ordinance in 1901 that allowed slaves who wished to escape their status to seek help from the third party, for example a German plantation owner, to pay their ransom fee. Many slaves ransomed through this arrangement became tied to the plantation owner who, in turn, got access to the former slave’s labor until the debt to him or her was worked off. This arrangement protected the interests of the planters by providing the cheap supply of workers needed to build up the colonial economy. It was of importance that the first plantation owners made good profit in order to a ract further European investments and the slave-like working conditions contributed to meeting these demands (Sunseri 1993). The ransom fee was set by local colonial officials and “based on local demand for labor, the slave’s gender, age, skills, or ability to work, and the price the master paid for the slave” (Sunseri 1993: 488–89). The demand to be ransomed could generally not be refused by the slave owner if it was considered a reasonable offer “unless they could prove that they were dependent on their slaves for prosperity” (Sunseri 1993: 506). The offers were generally 15 to 20 percent higher for female slaves than for male slaves, indicating the productive and reproductive value of women. A ransomed slave received a booklet with work cards, each card stipulating the number of days per month the former slave should work for the plantation owner. A former slave could, according to Deutsch (2006) work off such a debt within a time period of two years. When the days of debt had passed the former slave was entitled to receive a certificate of freedom (Sunseri 1993). This process indicates that the Germans were more interested in reforming and regulating slavery rather than abol-

336

•

Muted Memories

ishing the institution and to meet the colonizers interest to access cheap labor. Thus, the Germans developed a split vision on slavery and the slave trade. Their efforts went into suppression, which aimed at control by the colonial state, since this aim justified German intervention against Arab slave traders and eventually the colonization of the area. Slavery as an institution was to remain, since it was the foundation of the colonial economy (Deutsch 2006: 104). Especially the plantation economy along the coast, the early backbone in the colonial economy, depended on slave labor or ransomed slaves. A few decrees were issued but rarely enforced. It is estimated that approximately 25 to 30 percent of Bagamoyo’s population in 1891, a few years a er colonization, was slaves. The town contained a similar proportion of slaves at the time of the British takeover in 1918 (S. Fabian 2007b: 174). It was only with the British Involuntary Servitude Ordinance in 1922 that slavery was officially abolished, although not so in reality.21 The freeing of slaves under the prevailing socioeconomic, cultural, and religious conditions was more complicated than just rescuing them from an owner and telling them “you are free now, you can go.” A British vice-consul, reporting in 1908 on the porters in the Congo Free State finds that “in some instances these porters are literally slaves, and are considered so by their masters and themselves. They formerly belonged to Bangwana [waungwana] raiders. When the slaves were emancipated they were supposed to leave their masters, though hardly any did so, and they immediately returned” (Northrup 2007: 117).22 Gwyn Campbell (2006a) reports on slaves who refused to be freed in the Merina society on Madagascar. As free individuals they would lose their protection and could be forced to perform corvée labor. It has also been noted that runaway slaves o en sought new patrons for protection (Glassman 1995: 107). There are also cases when slaves liberated by Europeans insisted on paying ransom money to their former owners. Only through this act could they be free in their own eyes and in the eyes of their former owners. In Bagamoyo, the slaves liberated by the British were placed in the Catholic mission station and in its “freedom villages.” In the eyes of many of the people in Bagamoyo, however, they were seen as the mission’s slaves or the British Consul’s slaves. Such views were based on two convictions in many East African societies. One, found in the Muslim context along the coast, was that only the owner, according to the Shari’a law (Horton and Middleton 2000: 88), could free his or her slave. The other was that it was only the kinspeople who had sold or pawned another kinsperson into slavehood who could redeem such a person. Neither the colonizers nor the missionaries could grant a slave his or her freedom.

Intruders and Terminators

•

337

Terminators: Attempts to Destroy Bagamoyo Victorious in the conflicts with the rebels in 1888–89 and with the power transferred from the DOAG to the German state, the German colonizers started to consolidate their authority in two, as it turned out, mutually counterproductive ways regarding their own interests. The rebuilding of Bagamoyo as the colonial capital was the first, and the efforts to wrench the control over the caravan trade from the Indians was the second. The Indians, however, had a century-old monopoly to supply East Africa with textiles and other goods from the Gujarat area in northwest India. These textiles were “in great demand as exchange commodities in the African trade of the interior and were regarded as the ‘currency’” (Machado 2009: 56–57). Indian traders had managed to establish a monopoly over much of the trade on the East African coast. They knew the rules in the ritualized exchange of trade items and had access to the goods from India that were in high demand in East Africa. They o en operated as agents for Bombay trade houses, which gave them access to a variety of trade goods and financial support. All trade, wrote the British envoy Sir Bartle Frere in 1873, “passes through Indian hands. Their silent occupation of the coast . . . is one of the more curious things of the kind I know. It has been going on for forty years” (as reported in Koponen 1988: 62). It is therefore not a surprise that things never went the way the Germans expected regarding the control over the caravan trade. It soon became clear to them, as it had to previous external powers who had tried to gain control over this trade that they, for various reasons, could not compete with the traders from the Indian subcontinent.23 One reason, as mentioned, concerned the transaction costs. It has been estimated that these costs were three times higher for European (or North American) traders compared to the costs for Indian traders (see below). Another of their advantages was the well-established contact with local agents and traders who could inform them of the changes in local demands in up-country societies. Since the traders/porters from the interior of the continent came to the coast to exchange their ivory, slaves and other goods for textiles, it was an advantage in the competition over the caravan wealth to nurture close relations with these traders in order to learn as early as possible about changes in consumption pa erns and order goods from India or Britain according to these demands (Machado 2009: 64–65). This became even more important since “African consumers were able to negotiate the terms of trade and their engagement in relations of exchange of which they formed an integral part” (Machado 2009: 57). The intimate

338

•

Muted Memories

relations established through, for example, the joking relationship (utani) institution between Bagamoyo’s traders and their agents on the one hand, and up-country traders on the other, provided a reliable source of information. Mtoro (J. W. T. Allen 1981: 163) explains the relations between the Nyamwezi caravan traders and the merchants in Bagamoyo in the following way: “None came to the coast unless he had a contact in the town, a jumbe or a citizen whom he could inform of his arrival . . . and ask for a house to be put ready for him and plenty of beer. . . . When he reached the town he went into his house, and the ivory was sent to the customs. For the next six or seven days he would be drunk.” The competitiveness was elegantly explained in a le er wri en by representatives for two German trading companies, both from Hamburg, operating on Zanzibar. The le er, directed to the German colonial administration, was a response to the Germans’ plans to close the port in Bagamoyo in an a empt to direct the caravan trade to Dar es Salaam. The two representatives explained that: Bagamoyo’s trade . . . is in no way the kind of trade that has its roots in buying and selling; rather, it is for a good part dependent in consultation, power, incentive, enterprising spirit and personal effort. It cannot be undertaken by Europeans on account of its high expenses, since the available goods are sca ered across the various regions. This trade is carried out by Indians in each region. It demands experience and bonds that can be trusted . . . Furthermore, trade is, of course, entirely dependent upon the porters. These people have been used to coming to Bagamoyo annually for almost a century. The cheap food provided from the rich hinterland of Bagamoyo offer them a greater a raction than the poor countryside around Dar es Salaam. (Reproduced in S. Fabian 2007a: 443; 2007b: 200)24

Swahili trade is, as explained in chapter 5, an ancient cultural institution, built around a prolonged exchange of gi s and counter-gi s. This system, very different from the market system which European travelers and traders expected to meet, was also practiced in up-country societies, as evidenced by Hermenegildo Carlos de Brito Capelo and Roberto Ivens with the following words: “Following the vicious system in operation throughout Africa, of not selling anything to the Europeans but making him a present of it, they extort from him in turn all his goods and effects, bit by bit, until the unhappy man finds himself under the necessity of refusing all presents” (as quoted in J. Fabian 2000: 57).25 Mtoro (J. W. T. Allen 1981: 163) described the negotiations between the Nyamwezi and the Indian trader in the following way: the Nyamwezi “goes with his friend to the Indian shop . . . They start in the morning, and by noon they have not agreed over a single task. First he [the Nyamwezi] wants this color and then that, and all the time they must be brought dates and coconuts.”

Intruders and Terminators

•

339

The Indians and Arabs had at least three additional advantages over the Europeans in the caravan trade. One was that many Nyamwezi/Sukuma caravans were built up by individual traders/porters. The coastal merchant who wanted to purchase their ivory had to negotiate with each one of them, which was extremely time-consuming. Second, the porters refused to take money in exchange for their goods. Instead, they demanded goods like co on cloths (kaniki or merikani), weapons and ammunition, and metal goods like iron tools, copper wire etc. (S. Fabian 2007b: 163). The trader who wanted to deal with the porters had to keep these goods in stock. This initially favored the Indian traders, who kept a variety of goods imported from India and the UK in their stores and shops in Bagamoyo, while, at the same time, it disqualified German traders since German goods were neither known nor in demand. Their third advantage was the frugality of Indians who “could live on 25$ per year. They were also willing to sell at the lowest possible price and be content with the smallest possible margin of profit. They used family and communal ties to set up links with older firms in Zanzibar, with trades on the mainland coast, and with firms in India” (Cooper 1975: 78). The art of trading with the up-country caravan traders/porters is captured by J. Sturtz and J. Wangemann (1890: 37–38) in the following way: Trade with the Nyamwezi is above all arduous since they have much experience and are as knowledgeable about the value of goods as they are with ivory and they have many, many needs; for example, they, too, have their fashion, and they wouldn’t accept fabric with a rooster and hen pa ern when they’ve heard that the ship pa ern is now “in.” The Indian trader must therefore have extraordinary patience if they are to cheat these strangers. It wouldn’t be possible for a European to get involved in a trading relationship with the Wanyamwezi. That’s why all the ivory is bought up in Zanzibar through the mediation of the Indian. (S. Fabian 2007b: 197)

Even German merchant companies such as DOAG, Hansing, and O’Swalds continued to use Bagamoyo as the main place for their operations for strategic commercial reasons. They profited more from a continuation of their ties with the town’s merchant community than they would have done if they moved to Dar es Salaam in order to come closer to the colonial headquarters.

The Final Blows It has been repeatedly claimed in this work that the caravan trade was “in no way the kind of trade that has its roots in buying and selling.” It was a social institution with its own culture, rules, institutions, and

340

•

Muted Memories

office holders. Many new aspirants on the scene who a empted to enter the game in Bagamoyo met closed doors since they never managed to out-maneuver the strong trading networks that already existed between the town’s traders and the caravan personnel.26 During the second half of the nineteenth century, three colonial powers proposed a similar solution in their a empts to wrench the control over the caravan trade out of the hands of the Bagamoyo traders. They all tried to move the caravan routes from Bagamoyo to Dar es Salaam.27 The first a empt was made by Sultan Majid (Sultan between 1856 and 1870) when he tried to make Dar es Salaam his administrative capital in the sultanate. His dream was to a ract both traders from Bagamoyo and porters from up-country societies to carry out their business in his port. The choice to establish a new administrative capital was strategic, Steven Fabian (2007a: 450) writes, “if he was to capture a greater share of the coastal trade, he would have to undermine the longer standing socio-economic relations of Bagamoyo and the interior by building his own town.” The sultan even asked the Holy Ghost fathers, who searched for a suitable place in the early 1860s to establish the first mission station on the east coast of Africa, to consider Dar es Salaam. This would have strengthened his control over the mission and possibly also have weakened Bagamoyo’s control over the caravan trade on the central routes. If he had managed to establish his own caravan port, which he never did, then the control over the caravan trade would have been removed from the hands of the Bagamoyo traders and come under his direct control. A somewhat different approach to destroy Bagamoyo in order to take control of the caravan trade was launched by the British. During the 1870s, before the Berlin Conference when the area became a “German territory,” the British started to construct a road, the so-called Mackinnon Road, from Dar es Salaam to up-country societies, even to Lake Victoria (Nyanza). It would pass through Unyamwezi—the homeland of the Nyamwezi caravan traders/porters. The road was named for Sir Willian Mackinnon who was one of Britain’s leading ship owners with ships all over the Indian Ocean where his ships contributed to the British a empts to stop the maritime slave trade. The proposed road was part of a plan to gain control not only over the caravan trade between the interior and the coast but also over the territory through a lease of the Sultan of Zanzibar’s mainland dominions (Munroe 1987: 210). The British plan included gaining control over the port in Dar es Salaam, which was rejected by Sultan Bargash who saw the importance of the port for his own plans to gain control over the trade. Their plan to divert the profitable caravan trade away from Bagamoyo to Dar es Salaam faded away when the Brits were denied the right to control the port and

Intruders and Terminators

•

341

probably also because the progress in the construction of the road was extremely slow. The construction commenced in 1877 and was abandoned in 1881. Only a li le more than 160 km had been constructed during the four years of the project (Brennan, Burton, and Lawi 2007: 18n10). Another factor was that the planned road passed areas with less favorable conditions for the porters compared to the Bagamoyo routes (S. Fabian 2007a: 455). Even the missionaries of the Church Missionary Society embarked on a project to improve communications to reduce the dependence on porters between the coast and their inland mission stations. They began to construct a road with the intent to drive ox-charts up to their station at Mpwapwa in present-day central Tanzania (Beidelman 1982: 605). Mpwapwa was one of the major rest, market, and camp sites between Lake Tanganyika and the Indian Ocean. However, the missionaries’ project was a failure. The oxen could not survive the unhealthy area and because the rains repeatedly washed out the road grade. O en local inhabitants felled trees or burned grass so that the tracks were difficult to follow; worse still, when missionaries cleared some areas to facilitate passage they found that Africans then planted crops in these open areas. Such setbacks forced the missionaries to resign themselves to the use of porters rather than wagons. (Beidelman 1982: 605)

The fourth a empt to move trade from Bagamoyo to Dar es Salaam was planned and implemented by the Germans during the 1890s. Faced with the almost total monopoly that the Indians in Bagamoyo had developed over trade, the Germans adopted a two-pronged strategy to gain control over the caravan trade. The first was a parallel to Sultan Majid’s a empt in the 1860s to move the main caravan port from Bagamoyo to Dar es Salaam by making Dar es Salaam the new administrative center and capital. The second was to build a railway between Dar es Salaam to Mwanza on the southern tip of Lake Victoria and Kigoma, just north of Ujiji, on the eastern shore of Lake Tanganyika. The Germans had an additional motivation to try this; not only to get the trade out of the hands of Bagamoyo’s traders, but also out of the hands of the Zanzibari Sultan, now dominated by the British, since all the cargo in and out of Bagamoyo passed through Zanzibar. If the Indians were eliminated from the trading networks so would the Sultan be. Their strategy to take control over the caravan trade was clear: From the moment that steamers run to the best port, Dar es Salaam, and deliver a large assortment of goods to the natives at a cheaper price than could be delivered from Zanzibar—and this will occur on account of the direct import—as soon as the independent caravans from the interior begin going to Dar on account of this and are fi ed out there for the return trip as

342

•

Muted Memories

soon as the Indians who supply the caravans follow the natives to Dar; then Bagamoyo will fall. (reproduced in S. Fabian 2007a: 462)

A branch between Dar es Salaam and Bagamoyo was discussed but never implemented, probably for strategic reasons; Bagamoyo was intended to fall. A branching off at Tabora to Mwanza at Lake Victoria made sure that the railway would tap the resources from the Nyamwezi/Sukuma area, the heartland of the central caravan routes. The railway would also make the porters superfluous and break the bonds between the caravan porters and the Bagamoyo merchants. That would mean the end of a more than a century-long cultural institution of the caravan trade along the East African central routes. The initial progress was, however, slow. A commi ee had in 1896 recommended: “The immediate construction of a 75-cm.-gauge railway from Dar es Salaam and Bagamoyo to Morogoro” (Hill 1957: 84). Three years later the Colonial Council (Kolonial Rath) urged that the German Government set aside adequate resources in its 1900 budget for the survey of the proposed railway. Another four years passed until the Ost Afrikanische Eisenbahn Gesellscha (East African Railway Association) was founded with a budget for the survey (Hill 1957: 85). The construction of the railway, once it began in 1905 as a meter-gauge railway, was o en ahead of its contractual time. Its railhead reached Tabora at the center of the colony in 1912 and Kigoma at Lake Tanganyika in 1914 (Hill 1957: 94).28 The opening of the railway eventually meant the end of the heyday of the caravan trade. During the fourteen years that passed from the Germans’ decision to move the capital and trade to Dar es Salaam, and the railway reaching Tabora, trade statistics show that Bagamoyo continued to be the most important trading center on mainland Tanzania.29 Population figures from the turn of the century show that Bagamoyo and Dar es Salaam both had around 18,000 inhabitants. During the caravan season the population of Bagamoyo increased to 50,000, which indicated the a ractiveness the town had for the caravanners (Brennan, Burton, and Lawi 2007: 26). One reason for the continued booming of the town was that traders from the South Asian community had grown close relations to the town and its caravan trade and that the financially powerful headquarters were located here while only pe y trade was conducted by Indians in Dar es Salaam. Another reason to remain may have been to avoid the strict control the German colonial administration could establish in Dar es Salaam. But when the first part of the railway was ready, the caravan trade on the central routes became history. Bagamoyo town saw a dramatic decrease in trade and the number of residents reduced from approximately 30,000 to

Intruders and Terminators

•

343

3,000. The German colonizers’ efforts to crush the trading community in Bagamoyo had dramatic and devastating effects on the future of the town. This means that Bagamoyo has a most peculiar history. It was the first capital in German East Africa and in 1929 it was deprived of its status as a township under British rule by Government Notice number 120 which meant that Bagamoyo from now on consisted of a number of villages. In the words of Alphonce Kyessi (2002: 3) “its socio-economic strength and status deteriorated by the collapse of slave trade in the early 1930s. It lost its town status, which it has not managed to regain. At the moment it does not have the official status of a town that can be administered by a Town Council.” One may question Kyessi’s statement that the slave trade collapsed during the 1930s. However, the decline of Bagamoyo is beyond doubt. Suffice it here to reiterate that the population was around 30,000 in the year 1890, and it was probably once again 30,000 in the year 2000. The 1967 National Population Census established, however, that only 5,772 people lived in Bagamoyo. It was not surprising that the once German capital lost its status as a township during the years of drastic decline. J. M. Mghweno (2002: 5) notes in a biting statement that Bagamoyo “was reduced into a fishing and subsistence-farming village as business people immigrated to Dar es Salaam.” Kajeri and Henschel go even further when they draw their picture of Bagamoyo: Bagamoyo, once a flourishing trade center, became something like a “ghosttown” with its huge, but empty boma [the German colonial headquarter], with the empty Customs House where no more customs were collected, with its now useless Dunda tower. The beautifully built Arab houses were le empty, without inhabitants, and collapsed as well. Sea waves destroyed the quai at the Customs House. Soon broken houses, sand covering the ancient and famous Bagamoyo streets and lonely palms moving up and down in the fresh breeze of the ocean became symbols of the town’s decline. (Kajeri and Henschel n.d.: 30)

Notes  1. The classical period ended, according to R. C. Bridges, with Stanley’s expedition from the east to the west coast of Africa in 1874–77 (Bridges 1973: 221).  2. The leaders of these expeditions had different nationalities, but they were o en officers from the army/navy (J. Fabian 2000: 37).  3. Some had multiple roles. Livingstone, for example, said in his famous speech in 1857 at the Senator House in Cambridge that he hoped to open a path into the interior of the continent “for commerce and Christianity.” He held the idea that with the expansion of trade and markets, African rulers should

344  4.

 5.

 6.  7.

 8.  9.

10.

11. 12.

13.

•

Muted Memories

stop slave-napping their own and neighboring people and selling them to the Arabs. The use of the epithet “son” or “child” for a slave was common and used as a means to point out the slave’s dependence on and subjugation to his owner or “father.” It was not a sign of familiarity but an expressed relation of seniority or dominance versus dependence, as already mentioned in chapter 6. It was only a er the construction of a communication link between Boma where the Congo River meets the Atlantic Ocean and Stanley Pool, todays Brazzaville, and the Berlin Conference, that the Belgians ceased to use Bagamoyo as their entry point into the Congo in favor of the ports on coast of the Atlantic Ocean. Hardly surprising since the Belgian King Leopold II threw his eyes on the coast in his desire to acquire his private colony—see Hochschild 1998. Children were recruited to the caravans, o en slave children, when they were bought by the caravan leader(s) (J. Fabian 2000: 30). Slave children were also a trade item and sought a er as domestic servants among the Arabs. As quoted in Fee 2012, 3. An example is Burton and Speke’s expedition. They spent six months in 1856 on Zanzibar while they organized and equipped their caravan (see also Stanley [1872: chap. 1] for the organization of the caravan which brought him to Ujiji and Dr. Livingstone). Domesticated elephants were once suggested in an a empt to minimize the dependence on human porters. Colonel Maximilien Strauch, in charge of King Leopold II’s colonial project, suggested in 1880 that if the experiment succeeded  one would soon be able to see groups of domestic elephants rapidly traveling the routes on which today negro caravans arduously drag along under their heavy loads. Then our trade goods, no longer subject to such exorbitant expenses, can be offered to the natives at a price that will not dampen their desire, their use will spread and, from luxury articles which they are at the moment, they will change into necessities. (As quoted in J. Fabian 2000: 41; see also Rankin 1882) Four domestic elephants with their mahouts were shipped from India to East Africa. Two of them died before their caravan reached Tabora in central Tanzania, one at the Karema station, and the last seven months later. However, they soon dominated the economic sphere around the caravan trade, and managed to marginalize the original owners of the town, the middlemen-merchants. But their status also forced them (ideally) to follow the British anti-slavery laws. The massive influx of caravans into Bagamoyo during the caravan season reveals, however, that for most of the caravanners Saadani was not an option. The main reason for choosing Bagamoyo was the bonds that had been established between the town’s traders and the caravan traders and porters and what the many well-stocked shops in Bagamoyo could offer. On the islands of Pemba and Unguja, the sultan’s economic interests also

Intruders and Terminators

14.

15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

•

345

included his plantations, cultivated by his slaves, and the slave market in Zanzibar Stone Town. Luke’s third dimension implies that the Germans had to work with the local population to change “their own perceptions” and teach the colonized people that German colonization was good for them, so they could become their own oppressors. The original building has been fortified several times and was used as a garrison by the Germans before the British turned it into a prison. It is sometimes mistaken for a building where slaves were kept and sold before they were shipped from Africa as discussed in chapter 3. Stanley here echoes the view of the relations between Arabs and Africans broadcasted by the British anti-slavery movements. One of its front figures, Donald Mackenzie could report that “this [slavery] is the cheapest form of labor the Arabs could find, and they grew rich, and flourished in Oriental luxury on the lives of the poor blacks whom they looked down upon as animals created for their special purpose” (Mackenzie 1895: 14). A er the treaty, Dr. Kirk traveled along the coast in an a empt to stop the slave trade and reported a er his journey that “as the practical result of my journey, I registered as free 1408 slaves, held by Indians, British subjects. Of these 920 remained with their masters, and 488 began life again, working for pay, or in many instances uniting in small bodies and se lings near the villages” (Elton 1879: 106). It was particularly disturbing that Indians, as British subjects, held slaves. Dr. Kirk immediately objected to the Arabic term bagham, in Kiswahili mjinga, translated as “raw slave” in English referring to slaves recently brought from the interior of the continent, and a second proclamation was affixed to the doors only talking about “slaves” (Elton 1879: 59–60). The effect, however, of this protection of the economy during the first years a er the 1888–89 rebellion in Bagamoyo was that the lucrative trade remained in the hands of the Indians (S. Fabian 2007b). This happened twelve years a er slavery was abolished in Mozambique, fifteen years a er coastal Kenya, twenty-two years a er Uganda, and twentyfive years a er it was abolished on Zanzibar (Deutsch 2006: 2–3). There are different dates for the abolition of slavery in the area. In 1890, European governments met in Brussels for the Anti-Slavery Conference, a er which the German Colonial Administration adopted the 1890 Act of AntiSlavery, which granted the slaves certain rights. However, slavery was totally abolished in Tanganyika first in the year 1922. For references, see his endnote no. 27 on 122. It may be of interest to notice that the Indian rupee was the most preferred medium of exchange when money was involved in the transactions. The DOAG therefore introduced their own German silver rupees—not German Marks. Their understanding of what took place is an ample illustration of Marcel Mauss’ finding that an exchange of goods sometimes was a social interaction rather than an economic transaction (Mauss [1950] 1966).

346

•

Muted Memories

25. Fabian is quoting Ivens and Capelo (1881: 1:116–17). 26. A German governor of the colony used the term Krebsschaden or “cancerous disease” to characterize the Asian traders’ control over the caravan trade in Bagamoyo (see S. Fabian 2007a; 2007b: 201). 27. A parallel a empt to marginalize the Zanzibar Omanis’ control over the prosperous slave and ivory trade at Kilwa was launched by the French slave trader Morice when he tried to establish his own direct link with India to supply Kilwa with textiles and, thus, circumvent Zanzibar in this trade (see Sheriff 1987: 44). 28. It may seem ironic that the bulk of the labor force in the construction of the railway came from the Wanyamwezi (Hill 1957: 85) who, consequently, helped to seal the fate of the caravan trade. 29. S. Fabian (2007b: 192) reports that “From 1891 to 1907, various commentators remarked upon Dar’s lack of commerce while Bagamoyo remained a thriving town.”

 Epilogue On 30 May 2013, it became official that Tanzania had signed an agreement a couple of months earlier with the state-owned Chinese Merchants Holding International Co. Ltd. during Chinese President Xi Jinping’s first visit to Tanzania and Africa. The agreement focused on the development of a new port at Mbegani, an area of 800 hectares, a few kilometers south of Bagamoyo. Other parts of the agreement concerned the establishment of a special economic zone (or Export Development Zone) in the area and the construction of a railway that would link the new port with both the TAZARA railway to Zambia and the Tanzanian Central Railway to Kigoma in the western part of Tanzania. A network of roads would also link the port with large parts of eastern and central Africa. The total budget for these projects was $11 billion coming from the state-owned Chinese company, with the State Government Reserve Fund from Oman as a co-financer. Bagamoyo was once again the focus of the global media a er it had been launched as a potential World Heritage site. Reuters reported that “Tanzania Dreams Big with Port Project at former slave harbor” (Reuters, 15 March 2015) and BBC added “Bagamoyo Port: Tanzania Begins Construction of Mega Project” (BBC, 16 October 2015). Bagamoyo could once again be a pivot in global trade. The Chinese investors saw this as a part of their Twenty-First Century Maritime Silk Road (The Diplomat, 1 December 2015) while Tanzanian politicians and investors saw it as a means to trigger both local development in a poor part of the country and to channel export and import to and from eastern and central Africa through Tanzania and thereby challenge the position of the port at Mombasa, Kenya. The proposed port shall be able to handle 20 million containers a year in 2045, compared to 500,000 containers a year today through the port of Dar es Salaam. The expected catchment area for the new harbor is as ex-

348

•

Muted Memories

tensive as the area once covered during the heydays of the caravan trade, but the resources to be extracted today are mainly natural resources such as minerals. The part of mainland Africa in focus consists of the area between Zambia and Malawi, eastern parts of the Democratic Republic of Congo, Burundi, Rwanda, Uganda, and Tanzania. The construction of the project was expected to begin in October 2015 and the first parts were to be completed in 2017. The hope is, reports BBC Alice Muthengi in Dar es Salaam that “the new port will not only transform the economy of the sleepy tourist town, but also put the country in good stead to compete for regional business, especially with neighboring Kenya” (BBC, 16 October 2015). Or, as expressed by The Diplomat (1 December 2015) “Bagamoyo will go from a sleepy ‘do oir’ for Dar es Salaam to a major port with a Pan-African and a global role along the Maritime Silk Road.” Soon a er the election of a new president for Tanzania in October 2016, it was announced that the project with a new port was suspended (Port News, 1 November 2016). Critical commentators believed that this project was meant to be the legacy of the outgoing president who originated from Bagamoyo District. Experts explained that the planned port was too big in comparison with the expected growth of the Tanzanian economy. It was pointed out that it was more strategic to improve the port in Dar es Salaam, which already had railway connections with Zambia and western Tanzania. It was also suggested that the port in Tanga in the northern part of Tanzania could be developed to handle the oil from Uganda and South Sudan and that the port in Mtwara in southern Tanzania, in the vicinity of the gas field in this part of the country should be developed. It seemed like Bagamoyo in the near future will continue to be a “sleepy tourist town” or a “sleepy ‘do oir’” for Dar es Salaam. Rumors received in August 2018 from colleagues in Bagamoyo suggested that the Chinese and Omani companies had continued to pay compensation to local residents to access their land. Various sources (see R!SK 2018; Servant 2019a; Servant 2019b) reveal that preparation for the projects has continued and the construction of the new port soon will start. Powerful global actors once again orchestrate the development of Bagamoyo. It remains to be seen what will happen to its cultural heritage.

 Glossary Ada: commissions or fees paid to leaders, for example the majumbe of Bagamoyo. Aibu: shame, disgrace. Akida (pl. maakida): a title used both for the leader of a dance society and for advisors to the jumbe. Alungwana: see waungwana. Arabu/WaArabu: Arab(s) born in East Africa. Askari: armed guard, later also used for the soldiers in at the German colonial army. Baga (from bwaga): to lay down; one of the parts of the name Bagamoyo. Bangwana: see waungwana. Baraza: a concrete bench outside a stone house. A liminal zone between the private and the public spheres. Buibui: a Muslim dress which became popular among free and former slave women at the end of the nineteenth century. Busaid: the Omani dynasty which moved the Sultanate from Oman to Zanzibar. Bwaga: to lay down or relieve oneself of a burden. Chama (pl. vyama): a term used for an association or society as a dance society. Dhows: large wooden sailing boats with a lateen rig also called jahazi. Diwani: a leader in a coastal society, oĞen with ritual/religious power. Doti: a 12-foot-long coĴon cloth. A standard measure in the caravan trade.

350

•

Glossary

Frasila: a measure of weight; about 13 kg. Used for example to measure the weight of elephant tusks. Fundi: a term used for a trained or otherwise skillful person as artisans, oĞen slaves, in the coastal towns, or a skilled elephant hunter. Halal: permissible, according to Muslim laws and ethics. Haram: forbidden, to be avoided according to Muslim laws and ethics. Heshima: respect, dignity, honor. Hongo: tribute paid by caravans to the owners of the land, which gave the caravan personnel rights to camp, take water and fuelwood, and purchase food. Ibadiyya: the Islamic canon of the Omani who came to East Africa. Said to practice a modest, ascetic, and egalitarian life style. Idd-al Haji: the end of the pilgrimage to Mecca. Jemadar/Jemedari: a military leader appointed by the Sultan to be in command for one of his garrisons. Jumba: a house made of stone (or actually coral). Jumbe (pl. majumbe): the highest position among leaders in Swahili towns, but also used for village headmen. Kanda la pazi: compensation paid by the Wabagamoyo, the people of Bagamoyo, to the Zaramo leaders who allocated land to the Wabagamoyo to build the town. Kanga: a rectangular printed coĴon sheet which became popular among women on the Mrima coast at the end of the nineteenth century. Kaniki: plain coĴon sheeting, usually black or dark blue, oĞen made in India. Kanzu: a long-sleeved, usually white calico gown used by men to cover the whole body from neck to ankles. Kaskazi: the monsoon which usually blows from November/December to March used to sail to the northeast. Kibarua (pl. Vibarua): a slave who could be hired to work for other people than the owner. Kabila: ethnic belonging; tribe. Kiboko: strip of hippopotamus hide used for flogging. kiNgwana: a Swahili dialect spoken in northeastern Congo. Kirangozi (pl. virangozi): guide or caravan leader. Kiunga: a multi-storied, flat-roofed stone house, also called nyumba ya sakafu or nyunba ya ghorofa.

Glossary •

351

Kusi: the monsoon which blows from May to October used to sail to the southwest. Kutoroka: to run away. Leso: handkerchief, but also another name for the kanga still in use in the northern part of East Africa. Liwali: a local governor appointed by the Sultan to protect his interests. The next in rank in the Sultanate aĞer the Sultan. Maakida: see akida. Magunda Mkali: “fierce field,” west of Ugogo. Majumbe: see jumbe. Makuti: palm leaves used for thatching. Manga: recently arrived Arab from Oman or an Arab whose mother tongue is Arabic. Mapazi: see pazi. Marenga Mkali: “biĴer/brackish water,” east of Ugogo. Mashamba: see shamba. Mbagamoyo: a person from Bagamoyo. Mbuga: plains with black coĴon soil, seasonally flooded and used to grow rice among the Nyamwezi. Merikani: unbleached coĴon sheeting usually from Salem or Boston in the United States. Mganga: a medicine man oĞen consulted in times of distress or for supernatural protection. Mgeni (pl. wageni): a stranger or person excluded from the group of owners of a coastal seĴlement. Miji: see mji. Miombo: woodland covered with Brachystegia-Julbernardia species. Common in central/western parts of Tanzania. Mitaa: see mtaa. Mji: town. Mjinga: ignorant, newcomer, excluded from the “civilized.” Mlingoti: pole or mast. Moyo: heart. Mpambe (pl. wapambe): girls/women dressed up in finery and publically displayed; oĞen slaves. Mrima: the coastal area from the Kenyan border to Dar es Salaam. Mshenzi (pl. Washenzi ): uncivilized, barbarian, pagan, people of up-country origin. Mtoro (pl. watoro): a runaway slave.

352

•

Glossary

Msandarusi: Hymenaea verrucosa (Latin). Msikiti wa ijumaa: Friday Mosque. Mswahili: a Swahili person. Mtaa (pl. mitaa): quarter in a town usually dominated by one family or kinship group. Mtoroka: runaway slave. Mtumwa (pl. watumwa): slave. Mungwana: see waungwana. Murids: neophytes in the Quadiriyya Brotherhood. Mwenjeyi: here “owner” of a Swahili seĴlement; original inhabitants. Mzalia (pl. wazalia): a slave born on the coast; implying acculturation into the Swahili world. Ndani: the secluded inner private part of a house. Ngoma: dance or drum. Ngoma nkuu: the drums of a town or village which symbolized the independence of the seĴlement. Ngoma ya shindano: public competitive dancing oĞen carried out by two teams. Nisbas: the tradition to include a geographical name in a family or clan name. Nyika: the dry, open landscape which runs parallel to the coastal strip. Nyimbo (sing. wimbo): song. Nyumba ya udongo: a waĴle-and-daub house. Pazi (pl. mapazi): a political and ritual leader among the Zarama; an equivalent to the jumbe in Bagamoyo. Pazi haoni maji: “the pazi who did not see the sea”; probably a metaphor for the pazi’s sphere of influence which was in the inland area of the coast. Posho: rations or provisions; a part of a porter’s salary. Sawahili: another name for Swahili. Sebule: a guest room close to the entrance of a stone house. Shafii: one of the principal variants of Sunni Islam, which became a dominating canon in East Africa. Shaha: an honorary title for a leader in the Swahili world. Sometimes also understood as an ethnonyme. Shamba: a plot of land used for cultivation and establishment of homesteads. Shia: a form of Islam in East Africa oĞen associated with Muslim immigrants from India.

Glossary •

353

Shihiri: Arabs born on the Arabic Peninsula or Arabs who had immigrated from the Hadramaut port el-Shihr. Shirazi: both an ethnonym and a status title used by the elite in the coastal seĴlements. Those who use this term for themselves usually claim an origin of Shiraz in Persia. Shomvi: an honorary title for a leader in the Swahili world. Sometimes also understood as an ethnonym. Shomvi haoni jua: “the shomvi who did not see the sun (or to the west)”; probably a metaphor for the shomvi’s sphere of influence which was in the shore of the Indian Ocean. Shuka: a 6-foot-long coĴon cloth or half a doti. A standard measure in the caravan trade. Siku ya mwaka: Solar New Year. Suahel: coast or edge, another way to spell Sawahil. Suahil: another way to spell Sawahil. Sufism: a syncretistic form of Islam that in East Africa also includes dances, music instruments, techniques for trance, and sacrifices to spirits. Sunni: the canon of Islam that spread in East Africa to become the dominant form. Suria (pl. mashuria): a slave concubine. Suwaail: another way to spell Swahili. Tariqa: from tariki, for path or way; a brotherhood within Sufi Islam. Udu: being civilized; being a real human being. Ugogo: the land of the Gogo people in central Tanzania through which the caravans on the central routes passed. Ulama: the learned people. Upande: see shuka. Usafi: cleanness or purity, especially in religious terms. Ushenzi: unculturedness, uncivilized, uncouthness. Ustaarabu: to possess the qualities of being a civilized person; e.g. to live an “Arab-like way of life.” Utamaduni: the essence of being “civilized.” Utani: joking relationships. Uungwana: the way of life that the elite in the coastal societies see as civilized and honorable. Uwezo: the capability to live, for example, according to the uungwana standards. Vibarua: see kibarua.

354 • Glossary Vijana (sing. kijana): a term indicating to be junior to and dependent on a senior person. Vuli: the “small rains” from November to December/January in Bagamoyo. Vyama: see chama. WaArabu (sing. Mwarabu) Arabs. Wageni: see mgeni. Wana wa watu: literally “children of people”; a metaphor for being born of free parents. Wanawake: women. Wanyampara: caravan leader. A corruption of banamhala, the Nyamwezi term for elder. Washenzi: see Mshenzi. Waswahili: see Mswahili. Watu wa bara: people from mainland Africa. Often a term of abuse from those who regard themselves as people from the Indian Ocean world. Waungwana: first used as a term for free people in contrast to slaves, but with the increased contacts that followed on the caravan trade, the term was adopted by people from the coast, free or slave. Wawanga (sing. Manga): Arabs, usually from the region Muscat in Arabia or Arabs whose mother tongue is Arabic; see also Shihiri. Wimbo: see nyimbo.

 References Abdalla el Buhriy, Hamedi bin. 1960. Utenzi wa Vita vya Wadachi Kutamalaki Mrima. Translated by J. W. T Allen. Abdulaziz, M. H. 1979. Mukaya: 19th-Century Swahili Popular Poetry. Nairobi: Kenyan Literature Bureau. Abrahams, R. G. 1967. The Peoples of Greater Unyamwezi, Tanzania (Nyamwezi, Sukuma, Sumbwa, Kimbu, Konongo). London: International African Institute. Abu-Lughod, L. 1986. Veiled Sentiments: Honor and Poetry in a Bedouin Society. Oakland: University of California Press. Achola, M. 2002. “A Comparative Study between Bagamoyo and Mombasa.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. ADHT (African Diaspora Heritage Trail). 2006. Newsle er. Retrieved 5 April 2009 from http://www.tanzaniatouristboard.com/news/Document/Africa_ DIAPORA. Africa Travel Magazine. 2006. “Bagamoyo, Tanzania: UNESCO World Heritage Site on the Slave Route.” Retrieved 14 February 2013 from h p://www.africaata.org/tz_bagamoyo_sculptures.htm. Aldrick, J. S. 1988. “The Nineteenth Century Carved Doors of Mombasa and the East African Coast in Two Volumes.” Masters thesis, Durham University, Durham, UK. Retrieved from h p://ethesis.dur.ac.uk/6619/. Allen, J. de V. 1981a. “Ngoma: Music and Dance.” In The Customs of the Swahili People: The Desturi Za Waswahili of Mtoro Bin Mwinyi Bakari and other Swahili Persons. Edited and translated by J. W. T. Allen. Berkeley: University of California Press. ———. 1981b. “Swahili Culture and the Nature of East Coast Se lement.” International Journal of African Historical Studies 14(2): 306–34. ———. 1981c. “The Swahili World of Mtoro bin Mwinyi Bakari.” In The Customs of the Swahili People: The Desturi Za Waswahili of Mtoro Bin Mwinyi Bakari and other Swahili Persons. Edited and translated by J. W. T. Allen, Berkeley: University of California Press. ———. 1993. Swahili Origins: Swahili Culture and the Shungwaya Phenomenon. Eastern African Studies. Melton: James Currey Publishers.

356

•

References

Allen, J. W. T. 1981. The Customs of the Swahili People: The Desturi Za Waswahili of Mtoro Bin Mwinyi Bakari and other Swahili Persons. Berkeley: University of California Press. Allen, R. B. 2012. “European Slave Trading, Abolitionism, and ‘New Systems of Slavery’ in the Indian Ocean.” PORTAL Journal of Multidisciplinary International Studies 9(1): 1–21. Alpers, E. A. [1967] 1974. The East African Slave Trade. Historical Association of Tanzania Papers, No. 3. Nairobi: East African Publishing House. ———. 1970. “The French Slave Trade in East Africa (1721–1810).” Cahiers d’Études Africaines 10: 80–124. ———. 1975. Ivory and Slaves in East Central Africa: Changing Pa erns of Internal Trade to the Late Nineteenth Century. London: Heinemann. ———. 1983. “The Story of Swema: Female Vulnerability in Nineteenth-Century East Africa.” In Women and Slavery in Africa, ed. C. C. Robertson and M. A. Klein. Madison: University of Wisconsin Press. ———. 2006. “Flight to Freedom: Escape from Slavery among Bonded Africans in the Indian Ocean World, c. 1750–1962.” In The Structure of Slavery in Indian Ocean Africa and Asia (Slave Societies), ed. G. Campbell. New York: Frank Cass Publishers. ———. 2007. “The Other Middle Passage: The African Slave Trade in the Indian Ocean.” In Many Middle Passages: Forced Migration and the Making of the Modern World, ed. E. Christopher, C. Pybus, and M. Rediker. Berkeley: University of California Press. ———. 2009. “Representations of Children in the East African Slave Trade.” Slavery and Abolition: A Journal of Slave and Post-Slave Studies 30(1): 27–40. Amiji, H. M. 1983. “The Asiatic Bias in the Historiography of the East African Coast.” Journal of African Studies 10(2): 66–72. Anderson, B. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York: Verso. Anheier, H. K., and Y. R. Isar, eds. Cultures and Globalization: Heritage, Memory and Identity. London: Sage. Anti-Slavery Report. 1877. “Revival of the Slave Trade on the East Coast of Africa, May 1877,” 232. Appadurai, A. 1986. “Introduction: Commodities and the Politics of Value.” In The Social Life of Things: Commodities in Cultural Perspective, ed. A. Appadurai. Cambridge: Cambridge University Press. Arens, W. 1975. “The Waswahili: The Social History of an Ethnic Group.” Africa: Journal of the International African Institute 45: 426–37. ———, ed. 1976. A Century of Change in East Africa. The Hague: Mouton Publishers. Areskough, A., and H. Persson, 1999. In the Heart of Bagamoyo: The Decoding of a Coastal Town in Tanzania. Department of Architecture and Development Studies, Lund Institute of Technology, Lund. Arizpe, L. A. 2000. “Cultural Heritage and Globalization.” In Values and Heritage Conservation, Research Report. Los Angeles: The Gerry Conservation Institute. Askew, K. A. 1999. “Female Circles and Male Lines: Gender Dynamics along the Swahili Coast.” Africa Today 46(3): 67–99.

References •

357

Askew, M. 2010. “The Magic List of Global Status: UNESCO, World Heritage and the Agenda of the States.” In Heritage and Globalization (Key Issues in Cultural Heritage), ed. S. Labadi and C. Long. London: Routledge. Athman, A. H. 1996. “Styles of Swahili Carving.” Afrikanistische Arbeitspapiere: Schri enreihe des Kölner Institute für Afrikanistik 7: 11–29. Austen, R. A. 1988. “The 19th Century Islamic Slave Trade from East Africa (Swahili and Red Sea Coast): A Tentative Census.” Slavery and Abolition 9(3): 21–44. “The Black Sidis of Gujarati.” 2008. The Azania Sea, 2 December. Retrieved 3 December 2014 from h ps://azaniansea.wordpress.com/category/slavery/. Bachrach, P., and M. S. Baratz. 1962. “Two Faces of Power.” American Political Science Review 56(4): 947–52. “Bagamoyo Aims for Heritage Status.” 2014. East African Business Week. Retrieved from h p://www.busiweek.com/473/. Bagamoyo Friendship Society (Freundekreis Bagamoyo). n.d. Famous Sights. Retrieved 8 December 2014 from www.bagamoyo.com/famous_sights.html. “Bagamoyo Set to Be Promoted.” 2002. The Guardian, 9 September. Bancro , G. 1862. History of the United States from the Discovery of the American Continent to Present Time, vol. 3. Boston: Charles Bowen. Bang, K. A. 2008. “Textual Sources on an Islamic African Past: Arabic Material in Zanzibar’s National Archive.” In The Meanings of Timbuktu, ed. S. Jeppie and B. Diagnes. Cape Town: CODESRIA/HSRC. Barnes, R. 2005. Textiles in Indian Ocean Societies. London: Routledge Curson. Barth, F. 1969. Ethnic Groups and Boundaries: The Social Organization of Cultural Difference. Waveland, MS: Waveland Press. ———. 2000. “Boundary and Connections.” In Signifying Identities: Anthropological Perspectives on Boundaries and Contested Values, ed. A. P. Cohen. London: Routledge. Baumann, O. 1890. In Deutsch-Ostafrika während des Aufstandes: Reis eder Dr. Hans Meyer’s Expedition in Usambara. Vienna: Eduard Hölzel. ———. [1896] 1957. Der Sansibar-Archipel. Leipzig [translated as O. Baumann 1957: “Mafia”]. Tanganyika Notes and Records 46: 1–24. Baxter, R. 1673. A Christian Dictionary. London. BBC. 2015. “Bagamoyo Port: Tanzania Begins Construction of Mega Port.” 16 October. Retrieved 24 November 2015 from http://www.bbc.com/news/ world-africa-34554524. Beachey, R. 1967. “The East African Ivory Trade in the Nineteenth Century.” Journal of African History 3(2): 269–90. Beck, R-M. 2000. “Aesthetics of Communication: Texts on Textiles (Leso) from the East African Coast (Swahili).” Research in African Literatures 31(4): 104–24. Beck. S. n.d. “East Africa 1700–1950.” Retrieved 14 January 2015 from h p://www .San.beck.org/16–12–EastAfrica.html. Becker, J. 1887. Vie en Afrique, ou Trois Ans dans l’Afrique Central. Vols. 1 and 2. 2nd edition. Paris. Beidelman, T. O. 1982. “The Organization and Maintenance of Caravans by the Church Missionary Society in Tanzania in the Nineteenth Century.” International Journal of African Historical Studies 15(4): 601–23.

358

•

References

Bendix, F. R., and G. Hasan-Rokem, eds. 2012. A Companion to Folklore. Oxford: Wiley-Blackwell. Benne , N. R. [1968] 1974. “The Arab Impact.” In Zamani: A Survey of East African History, new edition, ed. B. A. Ogot. Nairobi: East African Publishing House. Biginagwa, T. J. 2012. “Historical Archaeology of the Nineteenth-Century Caravan Trade in Northwestern Tanzania.” PhD diss., University of York. Boswell, R. 2008. Challenges to Identifying and Managing Intangible Cultural Heritage in Mauritius, Zanzibar and Seychelles. Council for the Development of Social Science Research in Africa (CODESRIA) Monograph Series. Senegal. Bothe, T. 2003. “Number of Tourists Will Increase If . . .” Bagamoyo Friendship Society. October. Retrieved from h p://www.bagamoyo.com/466.html. Brady, C. T. 1950. Commerce and Conquest in East Africa with Particular Reference to Salem Trade with Zanzibar. Salem, MA: Essex Institute. Brandström, P. 1986. “Who Is a Sukuma and Who Is a Nyamwezi? Ethnic Identity in West-Central Tanzania.” Working Papers in African Studies, no. 27. Uppsala: African Studies Programme, Department of Cultural Anthropology. ———. 1990. “Seeds and Soil: The Quest for Life and the Domestication of Fertility in Sukuma-Nyamwezi Thought and Social Reality.” In The Creative Communion: African Folk Models of Fertility and the Regeneration of Life, ed. A. JacobsonWidding and W. can Beck, 167–86. Uppsala: Almqvist and Wiksell. Breidenbach, J., and P. Nyíri. 2007. “‘Our Common Heritage’: New Tourist Nations, Post-‘Socialist’ Pedagogy, and the Globalization of Nature.” Current Anthropology 48(2): 322–30. Brennan, J. R., and A. Burton. 2007. “The Emerging Metropolis: A History of Dar es Salaam, circa 1862–2000.” In Dar es Salaam: Histories from an Emerging African Metropolis, ed. J. R. Brennan, A. Burton, and Y. Lawi. Dar es Salaam, Nairobi: Mkuki na Nyota/British Institute in East Africa. Brennan, J. R., A. Burton, and Y. Lawi. eds. 2007. Dar es Salaa: Histories from an Emerging African Metropolis. Dar es Salaam, Nairobi: Mkuki na Nyota/British Institute in East Africa. Bridges, R. C. 1973. “Europeans and East Africans in the Age of Exploration.” Geographical Journal 139(2): 220–32. Brion Davis, D. [1966] 1988. The Problem of Slavery in Western Culture. Oxford: Oxford University Press. Brown, B. 1971. “Muslim Influence on Trade and Politics in the Lake Tanganyika Region.” African Historical Studies 4(3): 617–29. Brown, B., and W. T. Brown. 1976. “East African Trade Towns: A Shared Growth.” In A Century of Change in Eastern Africa, ed. W. Arens. The Hague and Paris: Mouton Publishers. Brown, W. T. 1970. “Bagamoyo: An Historical Introduction.” Tanzania Notes and Records 71: 69–83. ———. 1971a. “A Pre-Colonial History of Bagamoyo: Aspects of the Growth of an East African Coastal Town.” PhD diss., Boston University. ———. 1971b. “The Politics of Business: Relations between Zanzibar and Bagamoyo in the Late Nineteenth Century.” African Historical Studies 4(3): 631–43. Brumann, C. 2016. “Conclusion: Imagining the Ground from Afar: Why the Sites

References •

359

Are So Remote in World Heritage Commi ee Sessions.” In World Heritage on the Ground: Ethnographic Perspectives, ed. C. Brumann and D. Berliner. New York: Berghahn Books. Brumann, C., and D. Berliner. 2016. “Introduction: UNESCO World Heritage— Grounded?” In World Heritage on the Ground: Ethnographic Perspectives, ed. C. Brumann and D. Berliner. New York: Berghahn Books. ———, eds. 2016. World Heritage on the Ground: Ethnographic Perspectives. New York: Berghahn Books. Burton, R. F. 1859. “The Lake Regions of Central Equatorial Africa, with Notices of the Lunar Mountains and the Sources of the White Nile: Being the Results of an Expedition Undertakes under the Patronage of her Majesty’s Government and the Royal Geographical Society of London, in the Years 1857–1859.” Journal of the Royal Geographical Society 29: 1–454. ———. [1860] 1869. The Lake Regions of Central Africa. Vols. 1 and 2. Longman, Green. ———. 1872: Zanzibar; City, Island, and Coast. Vols. 1 and 2. London: Tinsley Brothers. Cameron, C., and M. Rössler 2013. Many Voices, One Vision: The Early Years of the World Heritage Convention. Oxford: Ashgate Publ. Ltd. Campbell, C. A., and C. M. Eastman. 1984. “Ngoma: Swahili Adult Song Performance in Context.” Ethnomusicology 28(3): 467–93. Campbell, G. 2006a. The Structure of Slavery in Indian Ocean Africa and Asia (Slave Societies). New York: Frank Cass Publishers. ———. 2006b. “Introduction: Slavery and Other Forms of Unfree Labour in the Indian Ocean World.” In The Structure of Slavery in Indian Ocean Africa and Asia (Slave Societies), ed. G. Campbell. New York: Frank Cass Publishers. ———, ed. 2005. Abolition and Its A ermath in Indian Ocean Africa and Asia. London: Routledge. Caplan, P. A. 1975. Choice and Constraint in a Swahili Community: Property, Hierarchy, and Cognatic Descent on the East African Coast. New York: The International African Institute. ———. 2007. “‘But the Coast, of Course, Is Quite Different’: Academic and Local Ideas about the East African Li oral.” Journal of Eastern African Studies 1(2): 305–20. Chalcra , J. 2016. “Decolonizing the Site: The Problems and Pragmatics of World Heritage in Italy, Libya and Tanzania.” In World Heritage on the Ground: Ethnographic Perspectives, ed. C. Brumann and D. Berliner. New York: Berghahn Books. Chami, F. 1999. “Roman Beads from the Rufiji Delta, Tanzania: First Incontrovertible Archaeological Link with Periplus.” Current Anthropology 40(2): 239–41. ———. 2002a. “Bagamoyo and the Early Exploration of Africa.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. ———. 2002b. “Kaole and the Swahili World.” In Southern Africa and the Swahili World: Studies in the African Past 2. F. Chami and G. Pwiti. Dar es Salaam: University Press. ———. 2002c. “The Rise and Fall of Bagamoyo.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September.

360

•

References

Chami, F., E. Maro, J. Kessy, and S. Odunga. 2004. Historical Archaeology of Bagamoyo: Excavations at the Caravan-serai. Dar es Salaam: University of Dar es Salaam Press. Chami, F., and G. Pwiti. 2002. Southern Africa and the Swahili World: Studies in the African Past 2. Dar es Salaam: University Press. Chance, J. K., and W. B. Taylor. 1985. “Cofradías and Cargos: An Historical Perspective on the Mesoamerican Civil-Religious Hierarchy.” American Ethnologist 12(1): 1–26. Chemillier-Gendreau, M. 2002. “The Idea of the Common Heritage of Humankind and Its Political Uses.” Constellation 9(2). Chi ick, N. 1963. “Kilwa and the Arab Se lement on the East African Coast.” Journal of African History 4(2): 179–90. ———. 1965. “The ‘Shirazi’ Colonization of East Africa.” Journal of African History 6(3): 275–94. Chrétien, J-P. 2007. “The Slave Trade in Burundi and Rwanda at the Beginning of German Colonization 1890–1906.” In Slavery in the Great Lakes Region of East Africa, ed. H. Médard and S. Doyle, 210–30. Oxford: James Curr. Christopher, E., C. Pybus, and M. Rediker, eds. 2007. Many Middle Passages: Forced Migration in the Making of the Modern World. Berkeley: University of California Press. Clarke, P. H. C. 1960. A Short History of Tanganyika. London: Longmans. Cleere, H., 2002. “Africa on the World Heritage List: Correcting the Imbalance.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Cohen, C. 1993. “‘The Natives Must First Become Good Workmen’: Formal Education Provision in German South West and East Africa Compared.” In “Namibia: Africa’s Youngest Nation.” Special issue, Journal of Southern African Studies 19(1): 115–34. Constante, B. 1804. De lésprit de conguête et de lúsarpation, dans leurs rapports avec la civilisation Européenne: Troisième E´dition. Paris: Le Normant and H. Nicolle. Constantin, F. 1989. “Social Stratification on the Swahili Coast: From Race to Class?” Social Stratification in Swahili Society. Africa 59(2). Cooper, F. 1975. “Plantation Slavery on the East Coast of Africa in the Nineteenth Century.” PhD diss., Yale University, New Haven, CT. ———. 1981. “Islam and Cultural Hegemony: The Ideology of Slave-Owners on the East African Coast.” In The Ideology of Slavery in Africa. Sage Series on African Modernization and Development, no. 6, ed. P. E. Lovejoy. Beverly Hills: Sage Publishers. Cory, H. 1954. The Indigenous Political System of the Sukuma and Proposals for Political Reforms. The East African Institute of Social Research. Dar es Salaam: The Eagle Press. Coupland, R. 1938. East Africa and Its Invaders: From the Earliest Times to the Death of Seyyid Said in 1856. Oxford: Oxford University Press. ———. 1939. The Exploitation of East Africa, 1856–1890: The Slave Trade and the Scramble. London: Faber and Faber.

References •

361

Datoo, B. A. 1970. “Misconceptions about the Use of Monsoons by Dhows in East African Waters.” East African Geographical Review 8: 1–10. ———. 1974. “Influences of Monsoons on Movement of Dhows along the East African Coast.” East African Geographical Review 12: 23–33. Daugbjerg, M., and T. Fibiger. 2011. “Introduction: Heritage Gone Global; Investigating the Production and Problematic of Globalized Pasts.” History and Anthropology 22(2): 135–47. Davidson, L. S. 2012. “Woven Webs: Trading Textiles around the Indian Ocean.” PORTAL: Journal of Multidisciplinary International Studies 9(1). Deutsch, J-G. 2006. Emancipation without Abolition in German East Africa, c. 1884– 1914. Oxford: James Currey. ———. 2007. “Notes on the Rise of Slavery and Social Change in Unyamwezi, c. 1860–1900.” In Slavery in the Great Lakes Region of East Africa, ed. H. Médard and S. Doyle, 76–110. London: James Curry. Deutsch Ostafrikanische Zeitung V/29, 18 July 1903. Dietler, M. 2001. “Theorizing of Feats: Rituals of Consumption, Commensal Politics, and Power in African Contexts.” In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, ed. M. Dietler and B. Hayden, 65–114. Washington, DC: Smithsonian Institution Press. Donovan, V. 1978. Christianity Rediscovered. Notre Dame, IN: Fides Press. Douglas, M., and B. Isherwood. 1979. The World of Goods: Towards an Anthropology of Consumption. New York: Basic Books. Dumont, L. [1970] 1980. Homo Hierarchicus: The Caste System and Its Implications. Revised edition. Chicago: Chicago University Press. Eastman, C. M. 1984. “Waungwana na Wanawake: Muslim Ethnicity and Sexual Segregation in Coastal Kenya.” Journal of Multilingual and Multicultural Development 5(2): 97–112. ———. 1988. “Women, Slaves, and Foreigners: African Cultural Influences and Group Processes in the Formation of Northern Swahili Coastal Society.” International Journal of African Historical Studies 21(1): 1–20. Ellio , A. M. and V. Schmutz. 2012. “World Heritage: Constructing a Universal Cultural Order.” Poetics 20: 256–77. Elton, J. F. 1879. Travels and Research among the Lakes and Mountains of Eastern & Central Africa: From the Journals of late J. Frederic Elton. Edited and compiled by H. B. Co erill. London: John Murray. Eriksen, T. H. 1993. Ethnicity and Nationalism: Anthropological Perspectives. New York: Pluto Press. ———. 2001. “Between Universalism and Relativism: A Critique of the UNESCO Concepts of Culture.” In Culture and Rights: Anthropological Perspectives, ed. Jane Cowan, Marie-Bénédicte Dembour, and Richard Wilson, 127–48. Cambridge: Cambridge University Press. ———. 2007. Globalization: The Key Concepts. Oxford: Berg. eTN Global Travel Industry News. 2009. “Tanzania to Host 5th International African Diaspora Heritage Trail Conference.” Retrieved from h ps://www.eturbonews .com/16536/tanzania-host-5th-international-african-diaspora-heritage-trail-c.

362

•

References

Fabian, J. 1986. Language and Colonial Power: The Appropriation of Swahili in the Former Belgian Congo 1880–1938. Cambridge: Cambridge University Press. ———. 2000. Out of Our Minds: Reason and Madness in the Exploration of Central Africa. Berkeley: University of California Press. Fabian, S. 2007a. “Curing the Cancer of the Colony: Bagamoyo, Dar es Salaam, and Socioeconomic Struggle in German East Africa.” International Journal of African Historical Studies 40(3): 441–69. ———. 2007b. “Wabagamoyo: Redefining Identity in a Swahili Town, 1860s–1960s.” PhD diss., Dalhousie University, Halifax, Nova Scotia. ———. 2013. “East Africa’s Gorée: Slave Trade and Slave Tourism in Bagamoyo, Tanzania.” Canadian Journal of African Studies 47(1): 95–114. Fahy Bryceson, D. 2009. “Swahili Creolization: The Case of Dar es Salaam.” In The Creolization Reader: Studies in Mixed Identities and Cultures, ed. R. Cohen and P. Toninato, 364–75. London: Routledge. Fair, L. 1994. Pastime and Politics: A Social History of Zanzibar’s Ng’ambo Community. PhD diss., University of Minnesota. ———. 1998. “Dressing Up: Clothing, Class and Gender in Post-Abolition Zanzibar.” Journal of African History 39(1): 63–94. ———. 2001. Pastime and Politics: Culture, Community, and Identity in Post-Abolition Urban Zanzibar, 1890–1945. Athens: Ohio University Press. Feilden, B. M. 1979. “An Introduction to Conservation of Cultural Property.” ICCROM. Retrieved from http://unesdoc.unesco.org/images/0003/000378/ 037868eb.pdf. Fee, S. 2012. “Hostage to Cloth: European Explorers in East Africa, 1850–1890.” Textile Society of America Symposium Proceedings, 9 January 2012. Ferguson, J. 1994. The Anti-Politics Machine: “Development,” Depoliticization, and Bureaucratic Power in Lesotho. Minneapolis: University of Minnesota Press. Finley, M. I. 1980. Ancient Slavery and Modern Ideology. London: Cha e. Fleisher, J. 2010. “Rituals of Consumption and the Politics of Feasting on the Eastern African Coast, AD 700–1500.” Journal of World Pre History 23(4): 195–217. Fleisher, J., P. Lane, A. LaViole e, M. Horton, E. Q. Morales, T. Vernet, A. Christie, and S. Wynne-Jones. 2015. “When Did the Swahili Become Maritime?” American Anthropologist 117(1): 100–115. Fleisher, J., and A. LaViole e, A. 1999. “Elusive Wa le-and-Daub: Finding the Hidden Majority in the Archaeology of the Swahili.” Azania: Archaeological Research in Africa, 34(1): 87–108. Flint, J. E., ed. 1976. The Cambridge History of Africa. Vol. 5: c. 1790 to c. 1870. Cambridge: Cambridge University Press. Foucault, M. 1978. The History of Sexuality. Vol. 1: An Introduction. New York: Pantheon Books. Franken, M. 1991. “Islamic Cultural Codes in Swahili Dances.” Journal for the Anthropological Study of Human Movement 6(4): 146–58. Garlake, P. S. 1966. The Early Islamic Architecture on the East African Coast. Oxford: Oxford University Press. Gibb, A. 1939. Ibn Ba uta: Travels in Asia and Africa, 1325–1354. London: Routledge and Sons.

References •

363

Giddens, A. 1990. The Consequences of Modernity. Stanford, CA: Stanford University Press. Githige, R. M. 1986. “The Issue of Slavery: Relations between the CMS and the State on the East African Coast Prior to 1895.” Journal of Religion in Africa 16(3): 209–25. Glassman, J. 1991. “The Bondsman’s New Clothes: The Contradictory Consciousness of Slave Resistance on the Swahili Coast.” Journal of African History 32(2): 277–312. ———. 1995. Feasts and Riot: Revelry, Rebellion, and Popular Consciousness on the Swahili Coast, 1856–1888. Portsmouth: Heinemann. ———. 2004. “Slower Than a Massacre: The Multiple Sources of Racial Thought in Colonial Africa.” American Historical Review 109(3): 720–54. Goldman, H. V. 1996. “A Comparative Study of Swahili in Two Rural Communities in Pemba, Zanzibar, Tanzania.” Thesis, Department of Anthropology, New York University. Goody, J. 1980. “Slavery in Time and Space.” In Asian and African Systems of Slavery, ed. J. L. Watson. Berkeley: University of California. Goyvaerts, D. 1987. “Review of Johannes Fabian, Languages and Colonial Power: The Appropriation of Swahili in the Former Belgian Congo 1880–1938.” Journal of Linguistics 32(2). Graeber, D. 2011. Debt: The First 5000 Years. New York: Melville House. Gray, R., and D. Birmingham, eds. 1970. Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa before 1900. London: Oxford University Press. Grey, Sir J. M. 1957. “Trading Expeditions from the Coast to Lake Tanganyika and Victoria before 1957.” Tanganyika Notes and Records 49: 226–46. The Habitat Agenda. 1996. Istanbul Declaration on Human Se lements. Retrieved from hdrnet.org/533/1/2072_61331_ist-dec.pdf. Hafstein, V. T. 2012. “Cultural Heritage.” In A Companion to Folklore, ed. F. R. Bendix and G. Hasan-Rokem. Oxford: Wiley-Blackwell. Hamid, M. A. 1996. “Kanga: It Is More Than What Meets the Eye—A Medium of Communication.” African Journal of Political Science 1(1): 103–9. Hannerz, U. 1991. “Scenarios for Peripheral Cultures.” In Culture, Globalization and World-System, ed. A. D. King. Binghamton: State University of New York. ———. 1992. “The Global Ecumene as a Network of Networks.” In Conceptualizing Society, ed. A. Kuper. London: Routledge. ———. 1996. Transnational Connections: Culture, People, Places. London: Routledge. Harries, L., ed. and trans. 1965. Swahili Prose Texts: A Selection from the Material Collected by Carl Velten from 1893 to 1896. London: Oxford University Press. Havnevik, K. J. 1994. Tanzania: The Limits to Development from Above. Uppsala: The Nordic African Institute. Henschel, J. 2011. 19th Century Humans as Merchandise: Slaves in Bagamoyo. Dar es Salaam: DeskTop Productions Limited. Henschel, J., F. Ndunguru, and W. Kadelya. 2000. The Two Worlds: Bagamoyo and Slavery in the 19th Century. Occasional Paper 2, Department for Antiquities/ Catholic Museum. Bagamoyo: Peramiho Printing Press.

364

•

References

———. 2001. Descendant of Former Slaves and Slave Owners Tell about Slavery in Bagamoyo. Department for Antiquities/Catholic Museum. Bagamoyo: Peramilo Printing Press. Hernàndez I Martí, G-M. 2006. “The Deterritorialization of Cultural Heritage in a Globalized Modernity.” Transfer: Journal of Contemporary Culture 1: 92–107. Hill, M. F. 1957. Permanent Way: The Story of the Tanganyika Railways. Vol 2: East African Railways and Harbours. Nairobi: East African Railways and Harbours. Hino, S. 1980. “Territorial Structure of the Swahili Concept and Social Function of the Swahili Group.” Senri Ethnological Studies 6. Hochschild, A. 1998. King Leopold’s Ghost: A Story of Greed, Terror, and Heroism in Colonial Africa. Boston: Houghton Mifflin Company. ———. 2005. Bury the Chains: The British Struggle to Abolish Slavery. London: Pan Books. Holmes, C. F. 1971. “Zanzibari Influence at the Southern End of Lake Victoria: The Lake Route.” African Historical Studies 4(3): 477–503. Holway, J. D. 1971–72. “C.M.S. Contacts with Islam in East Africa before 1914.” Journal of Religion in Africa 4(3): 200–12. Honan, E. 2015. “Tanzania Dreams Big with Port Project at Former Slave Harbor.” Reuters, 15 March. Retrieved 11 December 2015 from h ps://www.reuters.com/ article/us-tanzania-ports-bagamoyo/tanzania-dreams-big-with-port-project-atformer-slave-harbor-idUSKBN0MB08H20150315. Horton, M. 1987. “The Swahili Corridor.” Scientific American 257(3): 86–93. ———. 1994. “Closing the Corridor: Archaeological and Architectural Evidence for Emerging Swahili Regional Autonomy.” In Continuity and Autonomy in Swahili Communities: Inland Influences and Strategies of Self-Determination, ed. D. Parkin, 15–21. London: SOAS. Horton, M. C., and J. Middleton. 2000. The Swahili: The Social Landscape of a Mercantile Society. Oxford: Blackwell Publishers Ltd. Håkansson, T. 2004. “The Human Ecology of World Systems in East Africa: The Impact of the Ivory Trade.” Human Ecology 32(5): 561–91. Håkansson, T., M. Widgren, and L. Börjesson. 2008. “Introduction: Historical and Regional Perspectives on Landscape Transformation in Northeastern Tanzania 1850–2000.” International Journal of African Historical Studies 41(3): 369–83. ICOMOS. 1975. “European Charter for Architectural Heritage.” Retrieved from h ps://www.icomos.org/en/charters-and-texts/179-articles-en-francais/ressou rces/charters-and-standards/170-european-charter-of-the-architectural-heritage. ———. 1994. Nara Document on Authenticity. Nara, Japan. Retrieved from h ps:// www.icomos.org/charters/nara-e.pdf. ———. 2005. 5th Dra of the ICOMOS Charter on Cultural Routes (September 2005): Working document presented for discussion at the 15th GA in China in view of further comments and eventual adoption at the 16th GA in Canada. Retrieved from h ps://www.icomos.org/xian2005/culturalroutes-dra .pdf. ———. 2008a. “The ICOMOS Charter on Cultural Routes: Ratified by the 16th General Assembly of ICOMOS, Québec (Canada), on 4 October 2008.” Retrieved from h ps://www.icomos.org/images/DOCUMENTS/Charters/cultural routes_e.pdf.

References •

365

———. 2008b. “Quebec Declaration on the Preservation of the Spirit of Place.” Quebec, Canada. Retrieved from h ps://www.icomos.org/quebec2008/que bec_declaration/pdf/GA16_Quebec_Declaration_Final_EN.pdf. ———. 2016. “Cultural Heritage, the UN Sustainability Development Goals, and New Urban Agenda.” Concept note. Retrieved from h p://www.usicomos.org/ wp-content/uploads/2016/05/Final-Concept-Note.pdf. ICOMOS-CIIC. 2003. “Meeting of Experts on Cultural Routes.” Madrid, 30–31 May 2003. Retrieved 21 September 2016 from h p://www.icomos-ciic.org/ciic/ NOTICIAS_reunionexpertos.htm. Iliffe, J. 1979. A Modern History of Tanganyika. African Studies Series 25. Cambridge: Cambridge University Press. IRA (Institute for Resource Assessment). 2007. Annual Report July 2006—June 2007. Retrieved from h ps://www.udsm.ac.tz/web/index.php/institutes/ira/ ira-annual-reports. Isaacman, B., and A. Isaacman. 1977. “Slavery and Social Stratification among the Sena of Mozambique: A Study of the Kaporo System.” In Slavery in Africa: Historical and Anthropological Perspectives, ed. S. Miers and I. Kopytoff. Madison: University of Wisconsin Press. Isar, Y. R. 2011. “UNESCO and Heritage: Global Doctrine, Global Practice.” In Cultures and Globalization: Heritage, Memory and Identity, ed. H. K. Anheier and Y. R. Isar, 39–52. London: Sage. Ivens, R., and H. Capelo. [1889] 1996. De Benguela às Terras de Laca [From Benguela to the Lands of the Lakes]. Publicacões Europa-América. Jensen, U. J. 2000. “Cultural Heritage, Liberal Education, and Human Flourishing.” In Values and Heritage Conservation, Research Report. Los Angeles: The Gerry Conservation Institute. Jerman, H. 1997. Between Five Lines: The Development of Ethnicity in Tanzania with Special Reference to Western Bagamoyo District (Jipemoyo). Uppsala: Nordic Africa Institute. Johnston, H. H. 1886. Kilimanjaro Expedition. London: Kegan Paul. Jokilehto, J. 2007. “International Charters on Urban Conservation: Some Thoughts on the Principles Expressed in Current International Doctrine.” Retrieved from h p://www.ceci-br.org/novo/revista/docs2008/CT-2008-119.pdf. Joy, C. 2016. “‘UNESCO Is What?’ World Heritage, Militant Islam and the Search for a Common Humanity in Mali.” In World Heritage on the Ground: Ethnographic Perspectives, ed. C. Brumann and D. Berliner. New York: Berghahn Books. Kajeri, S. M., and J. Henschel. n. d. Bagamoyo: The Beauty at the Beach. Tabora: T. M. P. Printing Department. Kamamba, D. M. 1983. “The Essence of Conservation Architecture with Bagamoyo, Kilwa Kivinje and Dar es Salaam in Tanzania as Case Studies.” Proceedings of the Seminar on Conservation of Historic Towns and Monuments along the Coast of Kenya, Mozambique, Somalia and Tanzania. Paris: UNESCO. ———. 2002. “The Conservation of Bagamoyo Historic Town.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Kanneberg, H. von. 1900. “Durch die Marenga Makali.” Mi heilungen aus dem Deutschen Schutzgebieten 13.

366

•

References

Karlsson, K., and S. Tenga 2002: “Bagamoyo Strategic Urban Development Framework: Appraisal Project.” First dra , July 15. Kennedy, D. 2005. The Highly Civilized Man: Richard Burton and the Victorian World. Cambridge: Harvard University Press. ———. ed. 2014. Reinterpreting Exploration: The West in the World. Oxford: Oxford University Press. Kirknaes, J., and J. Wembah-Rashid. 1993. Bagamoyo: A Pictorial Essay from Tanzania. n.p.: Bega kwa Bega. Kirkman, J. S. 1964. Men and Monuments on the East African Coast. Bu erworth. Kirshenbla -Gimble , B. 2004. “Intangible Heritage as Metacultural Production.” Museum 56(1–2): 52–65. Kjekshus, H. 1977. Ecology Control and Economic Development in East African History: The Case of Tanganyika 1850–1950. London: Heinemann. Knippel, V. 2002. “Bagamoyo’s Old Town Today: A Survey of Buildings and Their Inhabitants in Dunda 2002.” Bagamoyo Cultural Heritage and Community Media Project. International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Kollman, P. V. 2005. The Evangelization of Slaves and Catholic Origins in Eastern Africa. American Society of Missiology Series, no. 38. New York: Maryknoll. Kombe, W. J. 2002. “Socio-Economic Characteristics and Poverty Questions in Bagamoyo.” International Conference: “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Kopf, J. 2005. “Space of Hegemony and Resistance in German East Africa.” PhD diss., University of Kentucky, Lexington. Koponen, J. 1988. People and Production in Late Precolonial Tanzania: History and Structures. Jyväskylä. Finnish Society for Development Studies in cooperation with Scandinavian Institute of African Studies. ———. 1994. Development for Exploitation: German Colonial Policies in Mainland Tanzania, 1884–1914. Helsinki: Tiedekirja. Kopytoff, I. 1982. “Slavery.” Annual Review of Anthropology 11: 207–30. ———. 1989. The African Frontier: The Reproduction of Traditional African Societies. Bloomington: Indiana University Press. Koren, H. J. 1958. The Spiritans: A History of the Congregation of the Holy Ghost. Pi sburgh, PA: Duquesne University. Krapf, J. L. [1968] 1860. Travels, Research and Missionary Labours during an Eighteen Years Residence in Eastern Africa. New York: Johnson Reprint Group. Kurlansky, M. 2002. Salt: A World History. New York: Penguin Books. Kyessi, A. G. 2002. “Bagamoyo: Infrastructure and Communication.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Labadi, S. 2013. UNESCO, Culture Heritage, and Outstanding Universal Values: ValueBased Analyses of the World Heritage and Intangible Cultural Heritage Conventions. Plymouth: AltaMira Press. Labadi, S., and C. Long. 2010. Heritage and Globalization. Key Issues in Cultural Heritage. London: Routledge. Landberg, L. C. W. 1975. Men of Kigombe: Ngalawa Fishermen of Northwestern Tanzania. Davis: University of California.

References •

367

Langworthy, H. W. 1971. “Swahili Influence in the Area between Lake Malawi and the Luangwa River.” African Historical Studies 4(3): 575–602. Larson, L. E. 1977. “A History of the Mbunga Confederacy, ca. 1860–1907.” Tanzania Notes and Records 81/82: 35–42. LaViole e, A. 2008. “Swahili Cosmopolitanism in Africa and the Indian Ocean World, A.D. 600–1500.” Archaeologies: Journal for the Archaeological Congress 4(1): 24–49. Léautier, F. A., M. Schaefer, and W. Shen. 2015. “The Port of Bagamoyo: A Test for China’s New Maritime Silk Road in Africa.” The Diplomat, 1 December. Retrieved 14 April from h ps://thediplomat.com/2015/12/-the-port-of-bagamoyoa-test-for-chinas-new-maritime-silk-road-in-africa/. Le Cour Grandmaison, C. 1989. “Rich Cousins, Poor Cousins: Hidden Stratification among the Omani Arabs in Eastern Africa.” Africa: Journal of the International African Institute 59(2): 176–84. Le Guennec-Coppens, F. 1989. “Social and Cultural Integration: A Case Study of the East African Hadramis.” Africa: Journal of the International African Institute 59(2): 185–95. Legère, K. 2002. “Kiswahili and Other Languages: The Impact of Bagamoyo.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Lennon, J. J., and M. Foley. 2000. Dark Tourism: The A raction of Death and Disaster. London: Continuum. Leslie, J. A. K. 1963. A Survey of Dar es Salaam. Oxford: Oxford University Press. Leue, A. 1900/1901. “Bagamoyo.” In Beiträge zur Kolonialpolitik und Kolonialwirtscha . Vol. 2: 1900–1901, 11–31. Berlin: Deutsche Kolonialgesellscha . Light, D. 2015. “Heritage and Tourism.” In The Palgrave Handbook of Contemporary Heritage Research, ed. E. Waterton et al., 144–58. Basingstoke: Palgrave Macmillan. Lindström, J. 2002. “The 2002 Bagamoyo World Heritage Conference: Prospects Regarding a Partnership for Conservation and Poverty Reduction.” PADRIGU, Department for Peace and Development Research, University of Gothenburg Sweden. Logan, W. S. 2002. “Globalizing Heritage: World Heritage as a Manifestation of Modernism and Challenges from the Periphery.” In Twentieth Century Heritage: Our Recent Cultural Legacy; Proceedings of the Australia ICOMOS National Conference 2001, 28 November–1 December 2001, University of Adelaide, Australia, ed. D. Jones. Adelaide: University of Adelaide and Australia, ICOMOS. Lovejoy, P. E., ed. 1981. The Ideology of Slavery in Africa. Toronto: Sage Publication. ———. 1983. Transformations in Slavery: A History of Slavery in Africa. Cambridge: Cambridge University Press. Lukes, Steven. 1974. Power: A Radical View. Hampshire: Palsgrave Macmillan. Machado, P. 2009. “Cloths of a New Fashion: Indian Ocean Networks of Exchange and Cloth Zones of Contact in Africa and India in the Eighteenth and Nineteenth Centuries.” In How India Clothed the World: The World of South Asian Textiles, 1500–1850, ed. G. Riello and T. Roy. Leiden: Brill. Mackenzie, D. 1895. “A Report on Slavery and Slave Trade in Zanzibar, Pemba,

368

•

References

and the Mainland of the British Protectorate in East Africa.” LSE Selected Pamphlets. Madan, A. C., ed. and trans. 1887. Kiungani; or Story and History from Central Africa: Wri en by Boys in the School of the Universities’ Mission to Central Africa. London: Bell and Son. Mafuru, J. K. 2002. “Land Use Plan in Bagamoyo.” Workshop on the Conservation and Development of Bagamoyo Township, Bagamoyo, 20–21 March. Maghimbi, S. 1999. “Some Problems in Bagamoyo District.” Workshop on Coastal Resources of Bagamoyo, Faculty of Science, University of Dar es Salaam, Bagamoyo. 18–19 December. “Making Bagamoyo a World Heritage Site.” 2002. The Guardian, 5 September. Malik, N. 1999. “Extension of Kiswahili during the German Colonial Administration in Continental Tanzania (Former Tanganyika), 1885–1917.” AAP 47: 155–59. Mallya, F. 2011. “Bagamoyo and the Spiritans.” In Spiritan Horizons: A Journal of the Congregation of the Holy Spirit 6. Retrieved from h ps://www.duq.edu/assets/ Documents/spiritans/_pdf/Spiritan%20Horizons%206.pdf. Mapunda, B. 2002. “East African Entrepots in a Period of Transition: Bagamoyo and the Slave and the Ivory Trade.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Martin, B. G. 1969. “Muslim Politics and Resistance to Colonial Rule: Shaykh Uways B. Muhammad Al-Barawi and the Qadiriya Brotherhood in East Africa.” Journal of African History 10(3): 471–86. ———. 1971. “Notes on Some Members of the Learned Classes of Zanzibar and East Africa in the Nineteenth Century.” African Historical Studies 4(3): 525–45. ———. 1976. Muslim Brotherhoods in Nineteenth-Century Africa. African Studies Series 18. Cambridge: Cambridge University Press. Mauss, M. [1950] 1966. The Gi : Form and Reasoning for Exchange in Archaic Societies. Translated from French by Ian Cunnison. London: Cohen and West Ltd. M’bow, A-M. 1981. “Preface.” In General History of Africa, vol. 1, ed. J. Ki-Zerbo for UNESCO International Commi ee. London: Heineman Educational Books. Mbwiliza, J. 2000. “The Birth of the Political Dilemma and the Challenges of the Quest for New Politics on Zanzibar.” In The Political Plight of Zanzibar, ed. T. L. Maliyamkono. Dar es Salaam: Tanzania Publishing Company Ltd. McCracken, G. 1988. Culture and Consumption: New Approaches to the Symbolic Character of Consumer Goods and Activities. Bloomington: Indiana University Press. McCurdy, S. 2006. “Fashioning Sexuality: Desire, Manyema Ethnicity, and the Creation of the ‘Kanga,’ ca. 1880–1900.” International Journal of African Historical Studies 39(3): 441–69. McLaren, C. 2002. “Shopping Spies.” In The Applied Anthropology Reader, ed. J. McDonald, 419–21. Boston: Allyn and Bacon. McNair, L. L. 1998. “West Africa Tour Brings Visitors to Tears: Urban League Celebrates 80th Year with African Pilgrimage.” New Pi sburgh Courier, 8 August 1998. Retrieved 13 August 2014 from h ps://www.highbeam.com/ doc/1P3-480820161.html.

References •

369

Médard, H., and S. Doyle, eds. 2007. Slavery in the Great Lakes Region of East Africa. Oxford: James Curry. Meillassoux, C. [1986] 1991. The Anthropology of Slavery: The Womb of Iron and Gold. Chicago: University of Chicago Press. Mercer, G. 2007. Bagamoyo: Town of Palms. Dar es Salaam: Graham Mercer. Merryman, J. H. 2005. “Cultural Property Internationalism.” International Journal of Cultural Property 12(11): 11–39. Mesaki, S. 1999. “Explaining Underdevelopment in the Coast Region of Tanzania: With Illustrations from Bagamoyo District.” Workshop on Coastal Resources of Bagamoyo, Faculty of Science, University of Dar es Salaam, Bagamoyo, 18–19 December. Meskell, L. 2012. “The Rush to Inscribe: Reflections on the 35th Session of the World Heritage Commi ee, UNESCO Paris, 2011.” Journal of Field Archaeology 37(2): 145–51. ———. 2013. “UNESCO’s World Heritage Convention at 40: Challenging the Economic and Political Order of International Heritage Conservation.” Current Anthropology 54(4): 483–94. ———. 2014. “States of Conservation: Protection, Politics, and Pacting within UNESCO’s World Heritage Commi ee.” Anthropological Quarterly 87(1): 217–43. ———. 2015. “Transacting UNESCO World Heritage: Gi s and Exchanges on a Global Stage.” Social Anthropology/Anthropologie Sociale 23(1): 3–21. Mghweno, J. M. 2002: “Bagamoyo Institutional and Administrative: From Village to Township.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Middleton, J. 1992. The World of the Swahili: An African Mercantile Civilization. New Haven, CT: Yale University Press. ———. 2003. “Merchants: An Essay in Historical Ethnography.” Journal of the Royal Anthropological Institute 9(3): 509–26. Meinhof, C. 1910. “Die praktische Bedeutung der Einheitssprache für die Kolonien.” In Verhandlungen des Deutschen Kolonialkongresses 1910. Berlin: Verlag von Dietrich Reimer. Miers, S. 2006. “Slavery: A Question of Definition.” In The Structure of Slavery in Indian Ocean Africa and Asia, Slave Societies, ed. G. Campbell. New York: Frank Cass Publishers. ———. 2008. “Reviewed Work: Slavery in the Great Lakes Region of East Africa by Henri Médard, Shane Doyle.” African Studies Review 51(3): 175–77. Miers, S., and I. Kopytoff, eds. 1979. Slavery in Africa: Historical and Anthropological Perspectives. Madison: University of Wisconsin Press. Mturi, A. A. 1983a. “Introduction.” In Proceedings of the Seminar on Conservation of Historic Towns and Monuments along the Coast of Kenya, Mozambique, Somalia and Tanzania (CLT-85/IS/15). Paris. UNESCO. ———. 1983b. “The Planning and Management of the Conservation of Historical Towns and Their Integration with the Social and Economic Planning and Development Process.” In Proceedings of the Seminar on Conservation of Historic Towns and Monuments along the Coast of Kenya, Mozambique, Somalia and Tanzania, (CLT-85/IS/15). Paris: UNESCO.

370

•

References

———. 1983c. “The Conservation of Swahili Monuments and Sites in Tanzania.” In Proceedings of the Seminar on Conservation of Historic Towns and Monuments along the Coast of Kenya, Mozambique, Somalia and Tanzania (CLT-85/IS/15). Paris: UNESCO. ———. 2000. “Conservation and Development of Old Bagamoyo Town.” Workshop on the Conservation and Development of Bagamoyo Township, Bagamoyo, 20–21 March. ———. 2002. “Cultural and Architectural Heritage of Bagamoyo and World Heritage Values.” International Conference, “Bagamoyo a World Heritage,” Bagamoyo, 9–12 September. Mueller, F. F. 1959. Deutschland–Sansibar–Ostafrika: Geschichte einer deutschen Kolonialeroberung, 1884–1890. Berlin. Mukuthuria, M. 2009. “Islam and the Development of Kiswahili.” Journal of Pan African Studies 2(8): 36–45. Munjeri, D. 2000. “Cultural Landscapes in Africa: An Overview.” In The World Heritage Convention and Cultural Landscapes in Africa: Expert Meeting—Tiwi, Kenya—9/14 March 1991. Paris: UNESCO. Munroe, F. J. 1987. “Shipping Subsidies and Railway Guarantees: William Mackinnon, East Africa and the Indian Ocean, 1860–93.” Journal of African History 28(2): 209–30. Murray Li, T. 2007. The Will to Improve: Governmentality, Development, and the Practice of Politics. Durham, NC: Duke University Press. Musitelli, J. 2002. “World Heritage, between Universalism and Globalization.” International Journal of Cultural Property 11(2): 323–36. Ndagala, D. 2000. “The Need and Urgency for a Planned and Coordinated Development of Bagamoyo.” Workshop on the Conservation and Development of Bagamoyo Township, Bagamoyo, 20–21 March. Ndunguru, F., W. L. Kadelya, and J. Henschel. 2003. Bagamoyo during the Bushiri War 1888–1889. Bagamoyo: Department for Antiquities, Catholic Museum. Nilsson, M. 2001. “Daily Life in Magomeni: An Improvement of Public Space in an Informal Part of Magomeni, Bagamoyo.” Diploma work at Department of Spatial Planning, Blekinge Institute of Technology, Karlskrona, Sweden. Nimtz, A. H. 1980. Islam and Politics in East Africa: The Sufi Order in Tanzania. Minneapolis: University of Minnesota Press. Nolan, F. P. 1977. “Christianity in Unyamwezi, 1878–1928.” PhD diss., University of Cambridge. Northrup, D. 2007. “Slavery and Forced Labour in the Eastern Congo: 1850–1900.” In Slavery in the Great Lakes Region of East Africa, ed. H. Médard and S. Doyle. Oxford: James Curry. Nurse, D., and T. Spear. 1985. The Swahili: Reconstructing the History and Language of an African Society, 800–1500. Philadelphia: University of Pennsylvania Press. Ogot, B. A. ed. [1968] 1973. Zamani: A Survey of East African History. Nairobi: EAPH/Longman. Owens, G. R. 2006. “The Shomvi: A Precursor to Global Ethnoscapes and Indigenization in Precolonial East Africa.” Ethnohistory 53(4): 716–52.

References •

371

———. 2012. “The Kamba War: Foundation Narratives, Ethnogenesis, and the Invention of the Zaramo in Precolonial Africa.” Ethnohistory 59(2): 353–85. Page, M. E. 1974. “The Manyema Hordes of Tippu Tip: A Case Study of Social Stratification and the Slave Trade in East Africa.” International Journal of African Historical Studies 7(1): 69–84. Pallaver, K. (n. d.). “‘A Second Zanzibar’: Some Notes on the History of Precolonial and Early Colonial Tabora, Tanzania.” University of Bologna. ———. 2006. “Nyamwezi Participation in Nineteenth-Century East African Trade: Some Evidence from Missionary Sources.” Africa 61(3–4): 513–31. ———. 2009. “‘A Recognized Currency in Beads’: Glass Beads as Money in 19th-Century East African Central Caravan Road.” In Money in Africa, ed. E. Catherine, H. Fuller, and J. Perkins, 20–29. London: The British Museum. ———. 2012. “Muslim Communities, Long-Distance Traders and Wage Labour along the Central Caravan Road.” Storicamente 8. Parkin, D. 1994. Continuity and Autonomy in Swahili Communities: Inland Influences and Strategies of Self-Determination. London: School of Oriental and African Studies. ———. 2005. “Textile as Commodity: Dress as Text: Swahili kanga and Women’s Statements.” In Textiles in Indian Ocean Societies, ed. R. Barnes. London: Routledge Curson. ———. 2007. “Wa ing on the Wind: Smell and the Cycle of Spirit and Ma er.” Journal of the Royal Anthropological Institute 13: S39–S53. Parkin, D., and F. Constantin. 1989. “Preface: Social Stratification in Swahili Society.” Africa: Journal of the International African Institute 59(2): 143. Pawliková-Vilhanová, V. 2010. “Rethinking the Spread of Islam in Eastern and Southern Africa.” Asian and African Studies 19(1): 134–67. Pearce, F. B. 1920. Zanzibar: The Island Metropolis of Eastern Africa. London: T. F Unwin. Peters, C. 1891. New Light on Dark Africa: Being the Narrative of the German Emin Pasha Expedition. Trans. H. W. Dulcken. London: Ward, Lock and Co. Pike, C. 1986. “History and Imagination: Swahili Literature and Resistance to German Language Imperialism in Tanzania, 1885–1910.” International Journal of African Historical Studies 19(2): 201–33. Polomé, E. C., and C. P. Hill. 1980. Languages in Tanzania. Ford Foundation Language Surveys. Oxford: Oxford University Press. Pomeranz, K., and S. Topik. 1999. The World That Trade Created: Society, Culture and the World Economy 1400 to the Present. New York: M. E. Sharpe. Pouwels, R. L. 1987. Horn and Crescent: Cultural Change and Traditional Islam on the East African Coast, 800–1900. African Studies Series 53. Cambridge: Cambridge University Press. Prestholdt, J. 2004. “On the Global Repercussions of East African Consumerism.” American Historical Review 109(3): 755–81. Prins, A. H. J. 1967. The Swahili-Speaking Peoples of Zanzibar and the East African Coast. East Central Africa, Part 12. Ethnographic Survey of Africa. London: International African Institute. Rabinow, P. ed. 1984. The Foucault Reader. New York: Pantheon Books.

372

•

References

Ranger, T. 1975. Dance and Society in East Africa 1890–1970: The Beni Ngoma. Berkeley: University of California Press. Rankin, L. K. 1882. “The Elephant Experiment in Africa: A Brief Account of the Belgian Elephant Expedition on the March from Dar-es-Salaam to Mpwapwa.” Proceedings of the Royal Geographical Society and Monthly Records of Geography 4(5): 273–89. Reichard, P. 1890. “Die Wanyamwezi.” Deutsche Kolonialzeitung 3:228–30, 239–41, 263–65, and 278–78. Resch, P. 1992. Althusser and the Renewal of Marxist Social Theory. Berkeley: University of California Press. Ressler, P. 2012. “The Kanga: A Cloth That Reveals—Co-Production of Culture in Africa and the Indian Ocean Region.” Textile Society of America Symposium Proceedings, Paper 736. Reusch, R. 1953. “How the Swahili People and Language Came into Existence.” Tanzania Notes and Records 34: 20–27. Riello, G., and T. Roy, eds. 2009. How India Clothed the World: The World of South Asian Textiles, 1500–1850. Leiden: Brill. Rigby, P. 1969. Ca le and Kinship among the Gogo: A Semi-pastoral Society of Central Tanzania. Ithaca, NY: Cornell University Press. R!SK. 2018. “Bagamoyo: The Largest Construction Project in Tanzania,” 28 December 2018. Retrieved 9 January 2019 from https://riskmagazine.nl/ article/2018-12-28-bagamoyo-the-largest-construction-project-in-tanzania. Roberts, A. 1970. “Nyamwezi Trade.” In Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa before 1900, ed. B. Grey and D. Birmingham. London: Oxford University Press. Robertson, C. C., and M. A. Klein, eds. 1983. Women and Slavery in Africa. Madison: University of Wisconsin Press. Robertson, R. 1992. Globalization: Social Theory and Global Culture. London: Sage. Rockel, S. J. 2000a. “‘A Nation of Porters’: The Nyamwezi and the Labour Market in Nineteenth-Century Tanzania.” Journal of African History 41(2): 173–95. ———. 2000b. “Enterprising Partners: Caravan Women in Nineteenth Century Tanzania.” Canadian Journal of African Studies 34(3): 748–78. ———. 2006a. Carriers of Culture: Labor on the Road in Nineteenth-Century East Africa. Portsmouth: Heinemann. ———. 2006b. “Forgo en Caravan Towns in 19th Century Tanzania: Mbwamaji and Mpwapwa.” Azania: Archaeological Research in Africa 41(1): 1–25. ———. 2009. “Slavery and Freedom in Nineteenth-Century East Africa: The Case of Waungwana Caravan Porters.” African Studies 68(1): 87–109. ———. 2014. “Decentering Exploration in East Africa.” In Reinterpreting Exploration: The West in the World, ed. D. Kennedy. Oxford: Oxford University Press. Roscoe, J. 1921. Twenty-Five Years in East Africa. Cambridge: Cambridge University Press. Russell, J. 1980. “Proposed Center for the Conservation and Restoration of Cultural Property at Bagamoyo.” UNESCO Restricted Technical Report RP/1979– 80/4/7.6/04, Paris.

References •

373

Ryan, M. M. 2013. “The Global Reach of a Fashionable Commodity: A Manufacturing and Design History of Kanga Textiles.” PhD diss., University of Florida, Gainesville. Said-Reute, E. [1980] 1994. Memoires of an Arabian Princess. Edited and with an introduction by G. S. P. Freeman-Grenville London: East-West Publications. Salazar, N. B. 2010. “The Globalisation of Heritage through Tourism: Balancing Standardisation and Differentiation.” In Heritage and Globalization, ed. S. Labadi and C. Long. Key Issues in Cultural Heritage. London: Routledge. Saouma-Forero, G., R. Hö , and M. Rössler, eds. 2000. “Introduction.” The World Heritage Convention and Cultural Landscapes in Africa: Expert Meeting—Tiwi, Kenya—9–14 March 1999, ed. M. Rössler and G. Saouma-Forero, 5. UNESCO. Retrieved from h ps://whc.unesco.org/en/events/596/. Sassen, S. 2006. Territoriality, Authority, Rights: From Medieval to Global Assemblages. Princeton, NJ: Princeton University Press. Scholte, J. 2005. Globalization: A Critical Introduction. New York: Palgrave-Macmillan. Schweinitz, H. H. von. 1893. “Wanyamwezi, Wassukuma, Wasiba und Wakara.” Zeitschri für Ethnologi 25: 477–84. Sco , J. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven, CT: Yale University Press. Schoenbrun, D. 2007. “Violence, Marginality, Scorn and Honour: Language Evidence of Slavery to the Eighteenth Century.” In Slavery in the Great Lakes Region of East Africa, ed. H. Médard and S. Doyle. Oxford: James Curry. Servant, Jean-Christophe. 2019a. “Bagamoyo [sic.], from Fish to 20m Containers: Tanzania’s Port out of Africa.” Le Monde diplomatique, February. Retrieved 15 March 2019 from h ps://mondediplo.com/2019/02/11tanzania-port. Servant, Jean-Christophe. 2019b. “Will a New Port Make Tanzania ‘Africa’s Dubai’?” The Nation, 19 February. Retrieved 15 March 2019 from h ps://www .thenation.com/article/tanzania-china-bagamoyo-port/. Shaw, R. 1996. “The Politician and the Diviner: ‘Big Men’ and the Ritual Consumption of Power in Sierra Leone.” Journal of Religion in Africa 26: 30–55. ———. 1997. “The Production of Witchcra /Witchcra as Production: Memory, Modernity, and the Slave Trade in Sierra Leone.” American Ethnologist 24(4): 856–76. ———. 2002. Memories of the Slave Trade: Ritual and the Historical Imagination in Sierra Leone. Chicago: Chicago University Press. Sheriff, A. 1987. Slaves, Species, and Ivory in Zanzibar: Integration of an East African Commercial Empire into the World Economy, 1770–1873. Nairobi: EAEP. ———. 2001. “The Spatial Dichotomy of Swahili Towns: The Case of Zanzibar in the Nineteenth Century.” Azania: Archaeological Research in Africa 36–37(1): 63–81. ———. 2005. “The Slave Trade and Its Fallout in the Persian Gulf.” In Abolition and Its A ermath in Indian Ocean Africa and Asia, ed. C. Campbell, 103–19. London: Routledge. ———. 2010. Dhow Cultures of the Indian Ocean: Cosmopolitanism, Commerce and Islam. New York: Columbia University Press.

374

•

References

Shorter, A. 1972. Chiefship in Western Tanzania: A Political History of the Kimbu. Oxford: Clarendon Press. Seaton, A. V. 1996. “Guided by the Dark: From Thanatopsis to Thanatoursim.” International Journal of Heritage Studies 2(4): 234–44. Sida. 2000. “Report from a Workshop in the Conservation and Development of Bagamoyo Township, March 20–21, 2000.” Embassy of Sweden, Dar es Salaam. ———. 2004. A Future for the Past: Historic Cities in Development. Retrieved from h ps://www.sida.se/contentassets/1f0f488bd4c7424ea29a94062b8a3394/14411 .pdf. Simpson, D. 1975. Dark Companions: The African Contribution to the European Exploration of East Africa. London: Paul Elek. Sinclair, P. 1995. “The Origin of Urbanism in Eastern and Southern Africa: A Diachronic Perspective.” In Islamic Art and Culture in Sub-Saharan Africa, ed. K. Ådahl and B. Sahlström. Uppsala: Almqvist & Wiksell International. Sissons, C. J. 1984. “Economic Prosperity in Ugogo, East Africa, 1860–1890.” PhD diss., University of Toronto. Smith, A. [1776] 1974. The Wealth of Nations: Books I–III. London: Penguin Books. Smith, L. 2006. Uses of Heritage. Abingdon: Routledge. Spear, T. 1978. The Kaya Complex: A History of the Mijikenda Peoples of the Kenya Coast to 1900. Nairobi: Kenya Literature Bureau. ———. 1984. “The Shirazi in Swahili Traditions, Culture, and History.” History in Africa 11: 291–305. ———. 2000. “Early Swahili History Reconsidered.” International Journal of African Historical Studies 33(2): 257–90. Stairs, W. G. 1891. “From the Albert Nyanza to the Indian Ocean.” Nineteenth Century: A Monthly Review 29: 953–68. Stanley, H. M. 1872. How I Found Livingstone in Central Africa. London: Sampson, Low. ———. [1888] 1988. Through the Dark Continent, vols. 1 and 2. New York: Dover Publications. ———. 1970. Stanley’s Despatches to the New York Herald, 1871–1872, 1874–1877, ed. N. R. Benne . Boston: Boston University Press. Stern, R. 1910. “Überlieferungen der -Nyamwezi über die Eisengewinnung.” In Handwerk und Industrie in Ostafrika: Kulturgeschichtliche Betrachtungen, ed. F. Stuhlmann. Hamburg: L. Friederichsen and Co. Stoecker, H., ed. 1986. German Imperialism in Africa: From the Beginnings until the Second World War. Berlin: Akademie-Verlag. Strandes, J. [1899] 1961. The Portuguese Period in East Africa. Nairobi: East African Literature Bureau. Strobel, M. 1983. “Slavery and Reproductive Labor in Mombasa.” In Women and Slavery in Africa, ed. C. C. Robertson and M. S. Klein. Madison: Wisconsin University Press. Stuhlmann, F. 1894. “Dr. F. Stuhlmanns Forschungsreisen in Usaramu.” Mi eilungen von Forschungsreisenden und Gelehrte aus den Deutschen Schutgebieten, VII. ———, ed. 1910. Handwerk und Industrie in Ostafrika: Kulturgeschichtliche Betrachtungen. Hamburg: L. Friederichsen and Co.

References •

375

Sturtz, J., and J. Wangemann. 1890. Land und Leute in Deutsch Ostafrika: Erinnerungen aus der ersten Zeit des Aufstandes und der Blockade. Berlin: Königliche Hofbuchnadelung von Siegfries Mi ler u. Sohn. Sunseri, T. 1993. “Slave Ransoming in German East Africa, 1885–1922.” International Journal of African Historical Studies 26(3): 481–511. ———. 2007. “The Political Ecology of the Copal Trade in the Tanzanian Coastal Hinterland, c. 1820–1905.” Journal of African History 48(2): 201–20. Suzuki, H. 2012. “Enslaved Populations and Indian Owners along the East African Coast: Exploring the Rigby Manumission List, 1860–1861.” History in Africa 39: 209–39. “Swahili: Kingwana and Other Simplified Forms.” 1975. Oceanic Linguistics Special Publications, No. 14, “A Bibliography of Pidgin and Creole Languages,” 676–90. Swantz, L. 1956. The Zaramo of Tanzania: An Ethnographic Study. Syracuse, NY: Syracuse University Press. Swartz, M. 1991. The Way the World Is: Cultural Processes and Social Relations among the Mombasa Swahili. Berkeley: University of California Press. Syversen, I. L. 2007. “Intentions and Reality in Architectural Heritage Management: In Search for the Influence of International Policy Documents on Contemporary Sustainable Local Heritage Management; Case: Zanzibar Stone Town, Tanzania.” PhD diss., Department of Architecture, Chalmers University of Technology, University of Gothenburg, Sweden. Talhami, G. H. 1977. “The Zanj Rebellion Reconsidered.” International Journal of African Historical Studies10(3): 443–61. Tanzania: Bagamoyo Township Survey 1969: Housing Building and Planning in Developing Countries. 1969. University of Lund, Department of Architecture II B, Sweden. “Tanzania Suspends Construction of U.S. $10 Million Bagamoyo Port.” Port News, 1 November 2016. Retrieved 11 January 2017 from h ps://www.hellenicshipping news.com/tanzania-suspends-construction-of-u-s-10-billion-bagamoyo-port/. Terray, E. [1970] 1974. Le matérialisme historique devaut les sociétés segmentaires et lignagères. Paris: Francois Maspero. Thompson, E. P. 1967. “Time, Work-Discipline, and Industrial Capitalism.” Past and Present 38: 56–97. Thomson, J. [1881] 1968. To the Central African Lakes and Back: The Narrative of the Royal Geographical Society’s East Central African Expedition, 1878–80. 2 vols. London: Frank Cass Co. Tippu Tip (Hamed bin Muhammed el Murjebi) [1958/1959] 1974. Maisha ya Hamed bin Muhammed el Murjebi yanni Tippu Tip. Edited and translated by W. H. Whitely. Nairobi: East African Literature Bureau. Torell, E. A. Mmochi, and P. Spiering. 2006. “Bagamoyo Governance Baseline.” Coastal Research Center. Retrieved from h ps://www.crc.uri.edu/download/ Bagamoyo_baseline.pdf. Trimingham, J. S. 1964. Islam in East Africa. Oxford: Clarendon Press. ———. 1968. The Influence of Islam upon Africa. London: Longman. Tsing, A. Lowenhaupt. 2005. Friction: An Ethnography of Global Connection. Princeton, NJ: Princeton University Press.

376

•

References

Tucker, A. R. [1911] 1970. Eighteen Years in Uganda and East Africa. 2 vols. Westpoint, CT: Negroe University Press. Turnbridge, J. E., and G. J. Ashworth. 1996. Dissonant Heritage: The Management of the Past as a Resource in Conflict. Toronto: John Wiley and Sons. Turtinen, J. 2006. Världsarvets villkor: Intressen, förhandlingar och bruk i internationell politik. Stockholm: Almqvist and Wiksell International. University College of Lands and Architectural Studies (UCLAS). 1999–2000. Bagamoyo Town Environmental Profile 2000. Prepared by Semester IV Students at UCLAS, Dar es Salaam. UNESCO. 1945. “UNESCO in Brief: Mission and Mandate.” Retrieved from h p:// www.unesco.org/new/en/unesco/about-us/who-we-are/history/constitution/. ———. [1954] 1999. “Convention for the Protection of Cultural Property in the Event of Armed Conflict with Regulations for the Execution of the Convention of 1954.” 14 May. Retrieved from h ps://ihl-databases.icrc.org/ihl/INTRO/400. ———. 1983. Proceedings of the Seminar on Conservation of Historic Towns and Monuments along the Coast of Kenya, Mozambique, Somalia and Tanzania (CLT-85/IS/15). Bagamoyo, Tanzania, 5–10 December. Retrieved from h ps://unesdoc.unesco .org/ark:/48223/pf0000065145. ———. 1994. “Global Strategy for a Representative, Balanced, and Credible World Heritage List.” WHC-94/CONF.003/INF.6. Retrieved from h p://whc.unesco .org/en/documents/101993. ———. 1996. Our Creative Diversity: Report of the World Commission on Culture and Development, Summary Version. Retrieved from h p://unesdoc.unesco.org/imag es/0010/001055/105586e.pdf. ———. 1998. Action Plan on Cultural Policies for Development: Adopted in by the Intergovernmental Conference on Cultural Policies for Development. Stockholm, Sweden, 2 April. Retrieved from h p://www.lacult.unesco.org/docc/1998_Action_Plan_ Cultural_Policies_for_Dev_UNESCO.pdf. ———. 2005. Evaluation of the Slave Route Project 1994–2002. Internal Oversight Service, Evaluation Section. IOS/EVS/PI/34. Retrieved from h p://unesdoc.unesco .org/images/0014/001473/147352e.pdf. ———. 2006. “The Central Slave and Ivory Trade Route.” Retrieved from h p:// whc.unesco.org/en/tentativelist/2095. ———. 2007. The Slave Route Project: Slavery and Slave Trade. Paris. BPI/EPP/2007/ PI/60M/19. ———. 2012. Operational Guidelines for the Implementation of the World Heritage Convention. UNESCO World Heritage Centre, Paris. Retrieved from h ps://whc .unesco.org/archive/opguide12–en.pdf. ———. 2013. Status of Implementation of the Action Plan 2012–2017 for the African Region. Internet source. Retrieved from whc.unesco.org/document/116975. ———. 2014. The Slave Route: 1994–2014. The Road Travelled. Slave Route Project. Retrieved from h p://www.unesco.org/culture/pdf/slave/the-slave-route-theroad-travelled-1994-2014-en.pdf. Unomah, A. C., and J. B. Webster. 1976. “East Africa: The Expansion of Commerce.” In The Cambridge History of Africa. Vol. 5: c. 1790 to c. 1870, ed. J. E. Flint, 270–318. Cambridge: Cambridge University Press.

References •

377

van Reybrouck, D. [2010] 2014. Congo: The Epic History of a People. Translated from the Dutch by Sam Garre . New York: HarperCollins. Veblen, T. [1899] 1994. The Theory of the Leisure Class: An Economy Study of Institutions. Dover Thri Editions. New York: Dover Publications. Vernet, T. 2009. “Slave Trade and Slavery on the Swahili Coast (1500–1700).” In Slavery, Islam and Diaspora, ed. B. A. Mirzai, I. M. Montana, and P. Lovejoy, 37–76. Trenton, NJ: Africa World Press. Retrieved from h ps://halshs .archives-ouvertes.fr/file/index/docid/671040/filename/Slave_trade_and_ slavery_on_the_Swahili_coast-T._Vernet_.pdf. Von Sicard, S. 1970. The Lutheran Church on the Coast of Tanzania 1887–1914: With Spherical Reference to the Evangelical Lutheran Church in Tanzania, Synod of Uzaramu-Uluguru. Studia Missionalia Upsaliensia 12. Uppsala: Gleerup. Wallace, M. 1996. Mickey Mouse and Other Essays on American Memory. Philadelphia: Temple University Press. Watson, T. N. 1982. “Areas of Concern: Bagamoyo, Tanzania.” Monumentum 25(1): 29–42. Wa , Mrs. S. n.d. In the Heart of Savagedom: Reminiscence of Life and Adventure during a Quarter of a Century of Pioneering Missionary Labours in the Wilds of East Equatorial Africa, ed. Mr. S. Wa . London: Marshal Brother. Wilson, C. T., and R. W. Felkin. 1882. Uganda and the Egyptian Sudan. Vol 1. London: Samson Low. Wilson, S. M. 1992. “The Emperor’s Giraffe: Not Every Powerful Country Chooses World Domination.” Natural History (December): 24–25. Wimmelbücker, L. 2009. Mtoro bin Mwinyi Bakari (c. 1869–1927): Swahili Lecturer and Author in Germany. Dar es Salaam: Mkuki na Nyota publishers. Wisnicki, A. S. 2008. “Chartering the Frontier: Indigenous Geography, ArabNyamwezi Caravans, and the East African Expedition of 1856–59.” Victorian Studies 51(1): 103–37. The World at War. n.d. German East Africa 1884–1922. Retrieved from www .schudak.se/timelines/germaneastafrica 1884–1922.html. World Monuments Fund. n.d. “Bagamoyo Historic Town.” Retrieved from h p:// www.wmf.org/project/bagamoyo-historic-town. Woulfin, D. S. 2011. “Slaves, Trains, and Missionaries: British Moral Imperialism and the Development of Precolonial East Africa, 1873–1901.” PhD diss., Stony Brook University, NY. Wright, M. 1976. “East Africa 1870–1905.” In The Cambridge History of Africa. Vol. 6: 1870–1905, ed. R. Oliver and G. N. Sanderson, 539–91. Cambridge: Cambridge University Press. Retrieved from h ps://doi.org/10.1017/ CHOL9780521228039.014. ———. 1993. Strategies of Slaves and Women: Life-Stories from East/Central Africa. New York: Lilian Barber Press and James Currey. Wright, M., and P. Lary. 1971. “Swahili Se lements in Northern Zambia and Malawi.” African Historical Studies 4(3): 547–73. Wynne-Jones, S. 2007. “It’s What You Do with It That Counts: Performed Identities in the East African Coastal Landscape.” Journal of Social Archaeology 7(3): 325–45.

378

•

References

———. 2010a. “Lines of Desire: Power and Materiality along a Tanzanian Caravan Route.” Journal of World Prehistory 23(4): 219–37. ———. 2010b. “Remembering and Reworking the Swahili Diwanate: The Role of Objects and Places at Vumba Kuu.” International Journal of African Historical Studies 43(3): 407–27. ———. 2011. “Recovering and Remembering the Slave Route in Central Tanzania.” Proceedings of the British Academy 168: 317–42. Wynne-Jones, S., and M. Walsh. 2010. “Heritage, Tourism and Slavery at Shimoni: Narrative and Metanarrative on the East African Coast.” History in Africa 37: 247–73. Zhao, Bing. 2012. “Global Trade and Swahili Cosmopolitan Material Culture: Chinese-Style Ceramic Shards from Sanje ya Kati and Songo Mnara (Kilwa, Tanzania).” Journal of World History 23(1): 41–85.

 Index abolition of slave trade/slavery, 23, 68, 70, 82, 140, 145, 147, 242, 266, 275, 279, 313, 326, 331, 333–35, 345n21; the 1873 agreement, 59, 70, 190, 275, 326, 332; abolitionists, 70, 82, 160, 165, 175 African Diaspora Heritage Trail (ADHT). See Bagamoyo alungwana. See waungwana ammunition, 6, 7, 11, 74, 76, 120, 130, 139, 165, 193, 195, 200, 339 Arab Fort/Old Fort, 28, 37, 80, 194, 197, 304, 330 Arabs: architecture, 89, 302, 343; “Arabization,” 48, 226, 279, 280; caravan trade, 48, 49, 51, 52, 65, 67, 74, 75, 116, 122, 126, 129, 134, 182, 202, 320, 321, 339; colonization, 12, 53, 82, 159, 179, 181, 208, 209, 210, 255, 274, 282; cultural influences, 22, 24, 46, 53, 60n16, 66, 80, 171, 213, 226, 236, 245; as an elite, 51, 203, 271, 280, 287, 310; Germans, 183, 195, 199, 217, 228, 334, 336; politics of identity, 267, 270, 271, 272, 274, 285, 286; racial stereotypes (see racial/colonial stereotypes); slave trade, 144, 148, 282, 333; slaves/slavery, 65, 66, 67, 74, 80, 117, 159, 172; Swahilization (see Swahili civilization); 1888/1889 rebellion (see rebellion 1888/1889) Arabu/Waarabu, 226, 264, 277

Bagamoyo, 1, 115; 1888/1889 rebellion (see rebellion 1888/1889); African Diaspora Heritage Trail, 79, 80; blockade, 12, 19, 200, 201, 334; Caravan Serai (see Caravan Serai); caravan trade (see caravan trade); Catholic mission station (see Christianity); cosmopolitan character, 50–53; Customs House (see Customs House); development cooperation (see development cooperation; partnership); DOAG (see DOAG); German colonial capital, 12, 14, 43, 216, 227, 228, 264, 329, 330, 337, 343; German colonization (see Germans); heritage-making (see heritagemaking); Indian traders/financiers (see Indians); International Conference (see International Conference 2002); multicultural character, 224, 258, 259, 269, 272; name, 10, 18, 62, 63; Nyamwezi porters/trades (see Nyamwezi); Old Fort (see Arab/Old Fort); origin of, 50, 242, 248, 309; pazi (see pazi); Sewa Haji Paroo (see Sewa Haji Paroo); siege, 12, 197, 8; as a slave port, 5, 67–74; Slave Route Project (see UNESCO); slavery (see slavery); slaves (see slaves); Stone Town (see Dunda/Stone Town); the Sultan’s

380

•

Index

liwali (see liwali); Swahilization (see Swahili civilization); Tentative List (see UNESCO); tourism (see tourism); World Heritage List (see UNESCO); Zanzibar intrusion (see Zanzibar) Bagamoyo Stone Town. See Dunda/ Stone Town Bagamoyo Township Survey, 23, 24 Bakari, Mtoro bin Mwinyi, 49, 126, 133, 141n10, 148, 149, 157, 160, 258, 276, 289, 295, 311, 327, 338 bangwana. See waungwana baraza. See stone houses barbarian. See washenzi beads, 135, 136, 139, 163, 239; caravan trade, 75, 141n5; means of payment, 135, 136 Becker, Jèrôme, 318, 319, 322 big drum. See drums blockade. See Bagamoyo bombardment. See Germans: gunboat diplomacy Bombay, 90, 140, 276, 337 boom in caravan trade, 120, 144, 159, 245, 282, 297 Boston, 76, 90, 135, 140 Brotherhood. See Quadiria/ Quadiriyya/tariqa Brussels Agreement, 147, 335 brute force. See slave-making buibui, 310, 314 Burton, Richard, 149, 317, 319, 321, 344n9; caravans, 163, 179, 207, 283, 318, 320 Busaid/al-Busaid, 159, 198; Sultan Barghash bin Said, 187, 211, 332; Sultan Khalifa bin Barghash, 183, 184, 188; Sultan Majid bin Said, 324, 327, 340, 341; Sultan Sayyid Ali bin Said, 181, 328, 333 Bushiri, 53, 197, 199, 201; 1888/1889 rebellion. See rebellion 1888/1989 Caravan Serai/Caravanserai, 18, 70, 73, 205n22, 311

caravan trade: beads (see beads); brutalization of trade, 156, 157, 159; caravan culture, 6, 11, 75, 126, 127, 133; competition over, 117, 177–203, 210, 254, 337; creolization (see creolization); as a cultural route, 12, 47, 65, 76, 104, 113, 126, 228; end of, 9, 14, 112, 342; German colonizers (see Germans); globalization (see globalization); hybridization (see hybridization); as an institution, 6, 7, 11, 13, 14, 117, 228, 319, 338, 339, 342; Swahilization (see Swahili civilization); wealth (see wealth) carved (Swahili) doors, 21, 22, 88, 89, 293, 304–7 central caravan routes, 12, 48, 64, 65, 74, 75, 81, 82, 104, 113, 114, 115, 117, 121, 188, 212, 218, 229, 278, 342; control over, 178, 187, 209, 327; Gogoland, 120, 153, 161 chama/vyama, 185, 215, 295 Christianity: Bagamoyo, 9, 12, 47, 55, 81, 104, 153, 161, 207, 217–21, 223, 228; Anglican Church, 53, 227; Church Missionaries Society (CMS), 218, 223, 227, 341; freedom villages, 55, 68, 336; Holy Ghost Order/Spiritans, 217, 218, 220, 221, 222; spread of, 9, 12, 55–57, 104, 207; White Fathers, 210, 227, 232n23, 318 cloth, 8, 9, 112, 115, 116, 140, 237, 293, 299; American (see textiles); caravan trade, 6, 7, 11, 13, 74, 76, 116, 120, 163; as “civilization,” 308, 309; cloth economy, 132–135, 136, 249; cloth with names, 132, 307, 308; doti, 133, 134, 142n13, 312; dress codes, 166, 169, 170, 310, 311; identity making, 307, 308; kanga (see kanga); kaniki (see kaniki); in Kilwa Chronicle (see Kilwa Chronicle); means of exchange, 132, 139, 339; means of payment,

Index •

135, 139; merikani, 76, 132, 133, 135, 139, 140, 308, 310, 312, 320, 339. See also textiles concubines. See slaves conspicuous consumption, 13, 112, 172, 181, 185, 186, 286, 292, 293–98, 299. See also carved (Swahili) doors; dances; feasts; stone houses consumerism. See conspicuous consumption copal, 6, 7, 13, 74, 76, 113, 169, 244, 276 courts, 154, 159, 160, 174 creolization, 7, 8, 47, 112, 171, 224, 225, 235, 236, 246, 258, 261, 262, 266, 280, 281, 289 cultural landscapes. See UNESCO Customs House, 70, 71, 159, 183, 193, 251, 324, 327, 331 dances, 63, 294, 295, 315n5; competitive dancing (ngoma ya machindano), 13, 185, 186, 296 development cooperation, 15, 18, 30, 33, 35 dhikiri. See Quadiria/Quadiriyya/tariqa dhow, 140, 332 disembedding, 6, 7, 13, 74, 76, 113, 169, 244, 276; processes of, 2, 3, 10, 80, 95, 107 DOAG (Deutsch Ost-Afrikanische Gesellscha ), 189, 198, 328; agreements with the Sultan, 182, 183, 193; Bagamoyo, 194, 195, 196, 197, 329, 337; cultural illiteracy, 190, 192; Pangani, 190, 193, 194; the Sultan’s red flag (see Sultan’s red flag). See also Germans domestic forms of slavery. See slavery domestic slaves. See slaves door frames. See carved (Swahili) doors doors. See carved (Swahili) doors doti. See cloth drums, 63, 137, 294 Dunda/Stone Town, 1, 9, 21, 25, 30, 32, 33, 36, 45, 56, 59, 70, 87, 88, 97, 293, 299, 330

381

elephants, 81, 114, 121, 125, 141, 244; domesticated elephants, 344n9; hunters, 130–32; tusks, 18, 50, 115, 116, 132, 161, 175, 234 emancipation, 150, 151, 167 enslavement. See slave-making exclusion, 1, 13, 112, 147, 184, 245, 247, 258, 262, 296, 302; of newcomers, 184, 272; processes of, 1, 93, 166, 271, 286, 287, 279, 299; of slaves, 149, 166 explorers. See travelers feast-hosting. See conspicuous consumption; dances; feasts feasts, 13, 185, 293, 297 female slaves. See slaves firearms. See weapons food, 23, 32, 128, 130, 164, 247, 259, 272; and Bagamoyo, 223, 243, 244, 248, 338; beads exchanged for, 135, 136; and caravan trade, 82, 117, 119; cloth exchanged for, 132, 133, 134, 136; and Gogo/Gogoland, 120, 153, 154, 180; and hongo (see hongo); and porters, 117, 119, 120, 127, 319, 320, 321; as prestige objects, 13, 191, 260, 269, 298, 304; production, 123, 124, 138; self-enslavement (see slave-making); slaves exchanged for, 11, 75, 161 generosity, 13, 185, 254, 294, 297 German East African Company. See DOAG Germans, 195, 202; 1888/1889 rebellion (see rebellion 1888/1889); agreement with the Sultan, 117, 182, 184, 188; and Bagamoyo, 153, 188, 189, 192, 194, 216; blockade of Bagamoyo (see Bagamoyo); caravan trade, 178, 187, 199; as colonizers, 6, 9, 11, 12, 14, 19, 43, 65, 70, 71, 131, 177, 178, 183, 187, 188, 194, 284, 328–32; destruction of Bagamoyo, 191, 192, 197, 245,

382

•

Index

339; expulsion from Pangani (see Pangani); gunboat diplomacy, 12, 182, 195, 196; racial/colonial stereotypes (see racial/colonial stereotypes); and railways (see railways); and slave trade (see slave trade); spread of Islam (see Islam); spread of Kiswahili (see Kiswahili); see also DOAG global ecumene of modernity. See modernity/modernization “Global Strategy for a Representative, Balanced, and Credible World Heritage List.” See UNESCO globalization, 2, 3, 6, 7, 9, 10, 43, 46, 64, 75, 84, 89, 90, 95, 98, 99, 101, 106, 107, 114, 137, 140, 180, 237, 319; and UNESCO, 1, 8, 9, 10, 19, 47, 48, 50, 53, 55, 59, 69, 91, 92, 94, 96, 98–100, 102, 207, 223. See also heritage-making Gogo, 119 121, 154; people (Wagogo), 120, 122, 152, 153 Gujarat, 22, 134, 135, 142n16, 239, 306, 308, 337 Habitat Agenda, 34, 135 Hadramis. See Muslims heritage-making, 1, 2, 3, 7, 9, 10, 17, 18, 19, 21, 27, 41, 45, 48, 55, 58, 62, 80, 87, 90, 93, 95, 98, 102, 105, 107, 108 heshima, 167, 191 hindus, 16, 276, 277 hoes. See iron Holy Ghost Order. See Christianity hongo (road tolls), 120, 128, 132, 133, 250, 252, 320, 321 hybridization, 47, 171, 224, 258, 262, 263, 264, 273, 281, 286, 289, 293, 311 Hymenaea verrucosa. See copal Ibadi/Ibadhi/Ibadiyya. See Islam ICOMOS, 31, 96, 97 identity, 83, 89, 93, 203n3, 210, 259, 261, 262, 264, 267, 271, 277, 284, 285; ethnonymical, 250, 260, 269;

formation of, 1, 7, 13, 84, 166, 167, 258, 259, 262, 266, 269, 280, 281, 282, 284, 292, 305, 313, 314; toponymical, 260, 270, 272. See also Arabs; Indians; Manyema; Nyamwezi; waungwana; Zaramo inclusion, 13, 166, 185, 254, 258, 262, 263, 272; processes of, 262, 286, 287, 288, 293, 296, 297, 299; in Swahili civilization (see Swahili civilization) India, 6, 50, 64, 74, 76, 81, 90, 114, 116, 117, 118, 132, 135, 139, 140, 178, 209, 210, 235, 237, 238, 239, 240, 241, 246, 269, 275, 276, 308, 311, 314, 315n12, 337, 339 Indians, 15, 51, 183, 191, 203, 261, 275; as British subjects, 168, 195, 239, 275; as caravan financiers/traders, 51, 139, 181, 186, 245, 338; caravan trade, 181, 245, 288, 337, 339, 341, 342, 345n19; and the Sultanate, 191, 275, 345n17 International Conference 2002, 32, 33, 38, 41, 43, 46, 100, 101, 229; organization, 42, 45, 58, 64; opening speech, 4, 57; route theme, 103, 104, 207; slave route theme, 57, 59, 64, 65, 79, 80 iron, 130, 138, 139; chains/necklaces, 5, 68, 80; hoes, 119, 120; trade in, 11, 50, 79, 118, 119, 121, 339 Islam, 9, 12, 54, 55, 81, 104, 147, 207, 209, 216, 254; and slaves, 211, 311; Brotherhood (see Quadiria/Quadiriyya/tariqa); and caravan trade, 12, 209, 210; and “civilization,” 13, 210, 246, 278; conversion to, 173, 209, 210, 253, 283, 304; dhikri/dhikiri/zikir (see Quadiria/Quadiriyya/tariqa); in Central Africa, 209, 225, 226; in East Africa, 54, 66, 208, 230n5, 246, 271; spread by the Germans, 212, 216; spread through Bagamoyo, 12, 54, 55, 81, 207, 209–16; spread

Index •

through caravan trade, 12, 54, 81, 209; Sufism, 12, 209, 212–15; syncretism, 148, 190, 213, 215; tombs, 212, 213, 255n5 ivory, 115, 125, 130, 131, 132, 200, 244, 245, 270, 297; Bagamoyo, 7, 13, 50, 52, 67, 74, 76, 113, 118, 337; global demand, 50, 114, 178; globalization, 134, 137, 178; Gogo, 120, 121, 154, 161; Nyamwezi, 6, 11, 12, 48, 112, 116, 119, 121, 122, 123, 125, 136, 154, 177, 178, 179, 234, 235, 239, 339; slaves, 75, 124, 138, 144, 178, 190; trade, 48, 50, 65, 76, 81, 84, 104, 118, 129, 133, 137, 180, 193; trade routes, 38, 42, 84, 104; traders from the coast, 275, 276, 338, 339 jemadar, 325, 327 joking relationships. See kinship jumbe/majumbe, 243, 249, 250, 251, 252, 273, 338 kanda la pazi, 222, 249, 250, 274, 309. See also pazi kanga, 64, 113, 140, 296, 308, 311–14 kaniki, 132, 308, 310, 311, 339 kanzu, 310, 311 Kaole, 23, 54, 118, 181, 242, 248; ancient mosques, 54, 208; armed conflict with Zanzibar, 50, 182, 318, 324; garrison, 50, 181, 203n5, 242, 324, 326; tombs and pillars, 213, 214, 236 Khoja Ismailis/Ismáilis. See Muslims kibarua (pl. vibarua). See slaves Kilwa, 2, 3, 54, 73, 76, 134, 165, 182, 238, 242, 264, 275, 302, 303; as the main slave port, 59, 67, 73, 144, 165, 275; French slave traders (see slaves); Kilwa Chronicle, 249, 288, 308; and slaves, 18, 68, 161, 164; slaves marched from, 59, 74, 332 Kilwa Chronicle. See Kilwa kinship, 62, 127, 131, 144, 149, 151, 152, 153, 253; as an idiom, 150, 151,

383

154; as a means of integration, 286, 287–89, joking-relationship/utani, 126, 127, 338; security through, 152, 157; slavery-to-kinship continuum, 150, 152; slaves, lack of (see slaves) kirangozi (pl. virangozi), 127, 128 Kirk, John, 190, 233, 264, 265, 277 Kiswahili, 170, 223, 264, 265, 345n17; Bagamoyo and the spread of, 9, 12, 47, 53, 54, 207, 228; as a lingua franca, 225, 226; spread through caravan trade, 12, 53, 54, 104, 207, 208, 224, 225, 226; spread through German colonization, 12, 216, 227 leso. See kanga lineage/domestic mode of production, 150, 153, 154 lintels, 304, 305. See also carved (Swahili) doors Livingstone, David, 52, 53, 65, 80, 81, 149, 317, 318, 319, 343n3 liwali: in Bagamoyo, 189, 195, 304, 327, 328; the flag incidences (see Sultan’s red flag); and the Germans, 188, 191, 195, 330; as the Sultan’s representative, 70, 182, 194, 203n5, 242, 325, 327, 344 Magomeni, 32, 33 Magunda Mkali, 120, 130 majumbe. See jumbe/majumbe manumission, 78, 154, 171, 173–74 Manyema, 210, 211, 304, 313: Area, 81, 82, 122, 129, 161, 211, 215, 226, 283; people, 81, 82, 83, 84, 104, 116 mapazi. See pazi Marenga Mkali, 120, 130, 243 marginalization: of middlemenmerchants, 181–84, 190, 196, 202, 272; of Nyamwezi porters/traders, 81, 82, 122, 129, 161, 177, 178–182, 202, 211, 215, 226, 283 mashamba. See shamba mbagamoyo/wabagamoyo, 100, 260, 323 mbuga, 124, 162

384

•

Index

Mediterranean world, 101, 237 merikani. See cloth Merina, 331, 33 metastory/metanarrative. See slave trade mgeni/wageni, 183, 204n9, 272, 289 middlemen-merchants, 66, 178, 235, 238, 246, 247, 248, 249, 253, 281, 285, 288, 297; struggle for control over caravan trade, 12, 198, 202, 263 missionaries, 8, 55, 68, 115, 116, 137, 165, 218, 318, 321; in Bagamoyo, 12, 13, 217, 219, 222; as caravan leaders, 52, 126, 137, 160, 283, 322, 323; and discipline, 220, 221, 232n24; and slaves/slaver, 82, 217–233, 322 missions. See Christianity mjinga. See slaves: “raw” slaves Mlingotini, 236, 243, 248 modernity/modernization, 30, 89, 93, 98; global ecumene of, 46, 98, 99, 102 monsoons, 12, 235, 239, 240, 241, 242, 278 moral economy, 163, 254, 297 Mpwapwa, 24, 244; mission station, 223, 341; rest place, 120, 188, 209, 243 Mrima coast, 6, 11, 112, 182, 184, 188, 211, 228, 334 msandarusi tree. See copal mshenzi/washenzi, 78, 190, 195, 215, 236, 282, 308 Mtoro bin Mwinyi Bakari. See Bakari mungwana. See waungwana murids. See Quadiria/Quadiriyya/tariqa museum in Bagamoyo, 68, 70, 73, 82, 205n22 Muslims, 51, 219, 246n22; as the “civilized,” 147, 271; conversion to Islam, 209, 210; and Germans, 216, 217; Hadramis, 211, 278, 279; Ibadi/Ibadhi/Ibadiyya, 211, 212, 230n5, 231n17, 278, 279, 305; Khoja Ismailis/Ismáilis, 51, 52; migration

to East Africa, 208; Shafíit, 212, 265, 279; and slaves/slavery, 173, 211; Sunni, 51, 212, 213, 265, 278, 279 muted, 3, 5, 7, 8, 105. See also silencing mwaarabu/waarabu. See Arabu/Waarabu mzalia/wazalia. See slaves Nassir Virji House, 37, 45 ndani. See seclusion ngoma. See dances ngoma ya shindano. See dances Ngoni, 124, 156, 226, 283 Nyamwezi, 11, 14, 82, 114, 119, 120, 129, 131, 140, 141, 143, 160, 161, 162, 179, 180, 238; agriculture, 122, 123, 124, 162, 180; and Bushiri (see Bushiri); caravan culture (see caravan trade); caravan organization, 117, 127–29, 130, 180; and caravan trade, 49, 74, 178, 338, 339; ivory (see ivory); joking relationships/utani (see kinship); marginalization of (see marginalization); as porters (see porters); slaves (see slaves); as traders, 59, 64, 75, 234, 239, 243, 338, 339; as “uncivilized,” 263, 283, 286 nyika, 237, 286 Old Fort. See Arab Fort/Old Fort Oman, 22, 54, 132, 144, 182, 186, 305, 308, 309, 331, 337 Operational Guidelines. See UNESCO Outstanding Universal Value (OUV). See UNESCO Pangani, 74, 188; 1888/1889 rebellion (see rebellion 1888/1889); and DOAG, 188, 189; expulsion of Germans, 189–94; plantation economy, 180, 182, 183, 331; as a trade port, 76, 118, 124; Zelewski, Emil von (see Zelewski) partnership, 30, 33, 35, 36. See also development cooperation Pawning/pawns. See slave-making pazi, 243, 248, 249, 250, 251, 254, 309;

Index •

kanda la pazi, 222, 249, 250, 274, 309 Persia: doors, 22, 306; immigrants from, 249, 263, 269, 270, 273, 288; influences, 53, 54, 270, 313; monsoons, 240, 241; slave trade, 74, 114, 165 Peters, Karl, 187, 201, 318, 328 pillars. See tombs plantation economy, 145, 148, 149, 157, 161, 164, 165, 172, 190, 307, 332, 333, 334, 336 porters, 64, 90, 115, 123, 125, 133, 149, 191, 201, 223, 229, 276, 285; 1888/1889 rebellion (see rebellion 1888/1889); Bagamoyo, 50, 62, 63, 117, 168, 179, 196, 206, 235, 244, 258; as “barbarians,” 13, 190, 215, 260, 274, 282, 286; caravan culture (see caravan trade: caravan culture); caravan routes (see central caravan routes); caravan trade (see caravan trade); female caravanners, 129, 130; from the interior, 6, 18, 50, 52, 67, 115, 119, 234, 337; food (see food); joking relationship/ utani (see kinship); Manyema (see Manyema); means of payment/ exchange, 132, 133, 134, 135, 136, 138, 193, 339; Nyamwezi (see Nyamwezi); ritualized trade (see ritualized trade/exchange); slaves/ slavery, 11, 116, 149, 156, 158, 160, 163, 167, 282; spread of Kiswahili (see Kiswahili); spread of Islam (see Islam); violence against, 8, 322, 323, 327, 336 Portuguese, 12, 49, 111, 145, 224, 237, 242, 270, 282, 312, 320, 331 posho, 132, 160, 206, 272 poverty, 30, 31, 33, 34, 35, 36, 42, 56, 89, 152 prison(s), 70, 159, 203n5, 327 privacy. See seclusion Quadiria/Quadiriyya/tariqa, 221; in Bagamoyo, 215, 216; and the

385

marginalized, 213, 215; murid, 212; spread of, 215; syncretistic character, 213, 215; tariqa, 211–15, 221 racial/colonial stereotypes, 77, 321; Arabs, 46, 65, 66, 67, 80, 147, 165, 200, 210, 217, 254, 268, 272, 278, 283, 329, 330; colonial ideology, 66, 210, 217, 227, 261, 272, 280, 289, 329; among “explorers,” 320–323; Germans, 216, 227, 289, 271; and slavery, 77, 146; Swahili, 261, 307, 308, 309; Zanzibar, 77, 271, 272 railways, 9, 14, 112, 141, 180, 228, 284, 341, 342 rebellion 1888/1889: Arabs, 201, 231n20; Bagamoyo, 183, 191, 194–203, 245; Bushiri, 53, 157, 177, 182, 188, 189, 203, 222, 330, 334; Zanzibar, 177, 189, 190, 245. See also Bushiri ritualized trade/exchange, 177, 189, 190, 245 Saadani, 12, 18, 71, 74, 181, 182, 189, 198, 221, 229, 244, 251, 324, 344n12 Sabri bin Safir, 327, 328 Salem, 76, 90, 135, 139, 140 salt, 11, 49, 50, 118, 119, 120, 121, 123, 138, 141n6, 252 scramble for Africa, 11, 147, 182, 194, 244, 322, 325 seclusion, 170, 191, 238, 239, 287, 301, 302, 303, 305, 311 Sewa Haji Paroo, 52, 160, 181, 194, 198, 277, 304 shaha: as an ethnonyme, 273, 273; Mzee Shaha, 290n12, 309; as a title, 252, 273, 274 shamba, 23, 168 Shihiri, 51, 226 shomvi, 249, 273, 259 siege. See Bagamoyo Siku ya Mwaka, 184, 190, 191 silencing, 6, 19, 48, 62, 65, 74, 105, 113. See also muted slave hunters, 15, 65, 80, 83, 129, 147, 165, 226, 229, 283

386

•

Index

slave-making: brute force, 155–57; debts, 149, 158, 159, 160, 163, 190, 253; pawning, 149, 157–59; punishment, 159–60, 326; selfenslavement, 155, 157, 158, 159, 176n14. See also slave hunters slave markets: Bagamoyo absence of/ existence, 68, 70, 73, 80, 157, 198; Zanzibar, 11, 59, 68, 74, 81, 83, 144, 164, 167, 333 Slave Route Project. See UNESCO slavery, 4, 57, 64, 66, 77, 172, 219, 326; abolition (see abolition of slavery/slave trade); in Africa, 78, 147, 149, 151; and Arabs (see Arabs); and Bagamoyo, 23, 68, 166; “civilizing” effect of, 146, 147; as a crime against humanity, 4, 78; in East Africa, 80, 123, 151, 155, 169; defense of, 77, 78, 105, 146, 147; domestic forms, 147, 149, 151, 335; fight against, 55, 65, 149, 217, 221; and the Germans, 173, 334, 335, 336; as an institution, 144, 145, 148, 150, 165, 174, 332, 336; kinship idiom, 127, 150, 151, 152–55; plantation/American form, 10, 80, 144, 145, 149, 190 slaves, 5, 65, 70, 74, 78, 79, 105, 108, 113, 115, 118, 120, 132, 137, 144, 148, 163, 168, 190, 253, 254, 256, 260, 268, 276, 283; acquisition of (see slave-making); and Arabs, 65, 66, 74, 78, 79, 200, 283, 284, 318, 326; and Bagamoyo, 9, 10, 18, 19, 50, 62, 66, 67, 68, 70, 71, 73, 74, 113, 161, 166–73, 198, 252, 275, 304; and caravan trade, 52, 75, 116, 129; categories of, 150, 166, 167, 169; and Christian missionaries (see missionaries); from the coast, 116, 117; concubines, 146, 152, 170, 253, 287, 288, 311; and cultivation, 23, 149, 217; dehumanization of, 143, 144, 155, 164, 166, 167, 169, 268; demand for, 6, 84, 122, 144, 148, 153, 159, 165; domestic slaves,

149, 151, 171, 304; dress codes, 166, 307, 309, 310, 311, 313, 314; emancipation, 150, 151; female slaves, 123, 146, 162, 166, 168, 169, 170, 171, 288, 313; French traders, 145, 174n4, 346n27; integration of, 149, 150, 151, 154, 171, 173, 215, 266, 285; kibarua/vibarua, 168, 169, 189; Kilwa, 18, 59, 67, 73, 144, 164, 165, 242, 275; lineage mode of production (see lineage/domestic mode of production); manumission (see manumission); maroonage, 171, 172; as means of exchange, 11, 75, 119, 123, 124, 161; mzalia/ wazalia, 167, 168, 172; Nyamwezi, 11, 48, 123, 138, 161; plantation economy (see plantation economy); ransoming, 55, 217, 331–36; “raw” slaves, 150, 155, 166, 167, 168, 171, 173, 174, 309; rights of, 147, 148, 149, 174n1, 185; self-enslavement (see slave-making); as slave owners, 148, 163, 167, 173; as “socially dead”/lack of kinship ties, 143, 152, 153, 154, 159, 163, 167, 326; trading slaves, 162–64; treatment of, 164, 151; as waungwana (see waungwana) slave-to-kinship continuum. See kinship slave trade, 4, 5, 67, 70, 77, 78, 80, 82, 83, 84, 101, 105, 138, 156, 162, 165, 283, 334; abolition of (see abolition of slavery/slave trade); and Arabs (see Arabs); expansion of, 59, 73, 145, 148, 149, 152; Germans, 173, 200, 334, 335, 336; metastory/ metanarrative); 4, 10, 80, 81, 84, 92, 145–47; waungwana (see waungwana); and Zanzibar (see Zanzibar) socio-economic/economic development, 17, 28, 29, 36, 38, 42, 91 Speke, Johan Hanning, 134, 206, 207, 283, 317, 318 standardization, 2, 3, 10, 19, 62, 75, 94–99

Index •

stone houses, 27, 204n15, 238, 247, 299–304; in Bagamoyo, 18, 25, 27, 45, 198, 216, 299, 301, 304; baraza, 300, 302, 305; conservation and protection, 27, 29, 56, 88; construction/layout, 164, 300, 301, 303; as status objects, 13, 167, 246, 292, 293, 299, 302, 303, 304, 307; Swahili doors (see carved [Swahili] doors) Strategic Development Urban Plan (SUDP), 33–38, 56 Sufism. See Islam Sultan Khalid bin Barhgash. See Busaid Sultan Majid bin Said. See Busaid Sultan Sayyid Ali bin Said. See Busaid Sultan Seyyid Bargash bin Said. See Busaid Sultan’s red flag, 189, 201; in Bagamoyo, 194, 195, 196 Sunni. See Muslims Suria. See slaves Swahili civilization, 12, 21, 22, 23, 32, 48, 51, 52, 54, 66, 67, 144, 162, 171, 178, 187, 210, 224, 241, 250, 260, 268, 270, 286, 287, 307, 308, 338; Bagamoyo, 66, 242, 259; cloth (see cloth); colonizers’ misconceptions, 22, 265, 269, 272; conspicuous consumption (see conspicuous consumption); costumes/dresses, 310, 311, 312, 313, 314; dances (see dances); hierarchies, 225, 258, 164, 273, 286, 303, 309; inclusion into, 112, 172, 173, 179, 185, 198, 215, 223, 266, 267, 272, 285, 288, 302, 304, 313; in the interior, 82, 156, 207, 209, 266, 267; Kiswahili (see Kiswahili); maritime orientation, 246, 250, 256n23; poetry, 184, 185, 192, 197, 296; political organization/structure, 252, 258, 265, 295, 327; origin/development of, 46, 49, 66, 164, 236, 237, 238, 246, 247, 259, 270, 271, 302; slaves/ slavery, 162,n167, 170, 171, 254, 277; stone houses (see stone houses);

387

Swahili doors (see carved [Swahili] doors); Swahilization, 53, 171, 215, 224, 225, 226, 265, 276 (see also Ujiji); urban-rural continuum, 245, 246, 262, 265; worldview of (see worldviews) Swahili doors. See carved (Swahili) doors syncretism, 184, 190, 204n16, 213, 215 Tabora, 106, 120, 123, 131, 134, 139, 179, 187, 207, 209, 210, 211, 223, 229, 244, 342 tariqa. See Quadiria/Quadiriyya/tariqa Tentative List. See UNESCO textiles, 6, 137, 247, 303, 307, 308, 309, 313, 313; American, 76, 135, 140, 314; Indian, 134, 139, 308, 314, 337; monochrome, 132, 308; multicolored, 132, 133, 134, 166, 296, 310, 313. See also cloth Tippu Tip, 54, 83, 122, 156, 227, 282 tombs, 212, 213, 255n5, 302 toponymical identity. See identity tourism, 5, 45, 77, 89, 93; Bagamoyo, 17, 24, 26, 29, 45, 56, 57, 76, 79, 100; and development, 17, 19, 26, 29, 56, 91; “roots tourism,” 5, 57, 79, 88 tourists, 1, 3, 5, 24, 26, 27, 36, 56, 87, 95, 105 trade. See caravan trade; slave trade trade routes, 65, 74, 78, 80, 101, 103, 119, 225, 237, 281 trading slaves. See slaves travelers, 48, 52, 65, 67, 115, 116, 121, 126, 127, 136, 137, 149, 160, 207, 229, 261, 262, 263, 268, 275, 280, 283, 319, 321, 338; and Arabs, 67, 268, 277, 283; and Bagamoyo, 52, 68, 73, 207; and road tolls (see hongo); Sewa Haji Paroo (see Sewa Haji Paroo) travelogues, 65, 275, 319 tusks. See ivory Ujiji, 38, 81, 229, 341; Bagamoyo-Ujiji route, 81, 82, 84, 106, 327; caravan

388

•

Index

trade, 134, 207, 218; Swahilization/ Arabization, 179, 209, 210, 211, 266, 267 UNESCO, 2, 21, 56, 58, 77, 93, 95, 96, 97, 105; Bagamoyo, 26, 27, 31, 36, 56, 61, 84; cultural landscapes, 101, 102; “Global Strategy for a Representative, Balanced, and Credible World Heritage List,” 101, 102; nomination procedure, 9, 19, 41; Operational Guidelines, 91, 101, 102; Outstanding Universal Value (OUV), 1, 2, 3, 9, 87, 41, 91, 92, 94, 102, 108n3, 207; power of, 95–100; race to inscription, 2, 10; routes theme, 36, 79, 103; Slave Route Project, 4, 10, 57, 59, 64, 77, 79, 105; Tentative List, 1, 3, 4, 8, 11, 18, 19, 21, 59, 61, 62, 67, 79, 84, 87, 88, 97, 207, 330; World Heritage List, 2, 3, 4, 7, 9, 10, 18, 31, 32, 36, 41, 42, 43, 45, 46, 47, 55, 57, 58, 62, 79, 84, 87, 88, 90–94, 95, 98, 99, 101, 102, 103, 107, 194 Unguja, 74, 165, 183, 187, 190, 331, 333 ushenzi, 170, 171, 286 ustaarabu, 272, 280, 284, 290n7 utamadumi, 245, 308 utani. See kinship uungwana, 209, 210, 281, 284, 286; kiNgwana, 212 vibarua. 224. See slaves vyama. See chama/vyama wageni, 183, 272, 289 washenzi, 78, 190, 195, 215, 262, 282, 308 waswahili, 53, 180. See also Swahili civilization wa le-and-daub house/mud-andwa le house, 245, 246, 247, 300, 302 waungwana, 281, 283, 284; as an elite, 170, 171, 215, 282, 283; as freemen, 247, 282; as slaves, 129, 282, 283, 336; as slave hunter, 82, 83, 226, 283

wealth, 7, 8, 13, 133, 136, 148, 150, 171, 185, 235, 253, 263, 280, 281, 283, 292, 293, 294, 297, 304, 337 weapons, 6, 7, 11, 74, 117, 120, 124, 139, 181, 193, 339 Wissman, Herrman, 70, 199, 330 World Heritage List. See UNESCO worldviews, 77, 210; Arabs, 210, 271, 308; Colonizers, 216, 251, 261, 262, 266, 271, 280, 285, 307, 308, 330; Swahili, 249, 250, 260, 266, 274, 286, 307 Zanzibar, 36, 53, 81, 83, 84, 139, 147, 179, 182, 186, 187, 247, 263, 268, 271, 279, 286, 282, 310, 313, 321; 1888/1889 rebellion (see rebellion 1888/1889); Bagamoyo, 12, 51, 206, 239, 245, 250, 275, 328, 341; and the British, 59, 70, 73, 145, 162, 164, 272, 275, 328, 331, 332, 333; caravan trade, 64, 76, 104, 117, 122, 135, 140, 178, 179, 200, 239, 308; control over the coast, 11, 50, 67, 159, 193, 242, 251, 318, 324, 325; dehumanization of slaves, 77, 146; economic empire, 48, 76, 140, 179; garrisons (see Kaole: garrison); monsoons (see monsoons); plantations (see plantation economy); racial stereotypes (see racial/colonial stereotypes); slave markets (see slave markets); slaves/slavery, 83, 146, 158, 165, 266, 280, 333; slave trade, 59, 73, 83, 138, 144; Swahili doors (see carved [Swahili] doors); US traders, 76, 139, 245 Zaramo, 244, 262; and Bagamoyo, 224, 254; Bagamoyo agreement, 222, 248–51; clashes with the Sultanate, 182, 242, 318, 324; original inhabitants/owners, 83, 225, 243, 248. See also pazi Zelewski, Emil von, 192; Bagamoyo, 204n15; Pangani, 190, 191, 192, 193