Islamic Sufi Networks in the Western Indian Ocean (c.

1880–1940)
 Islam in Africa

Editorial Board

Rüdiger Seesemann
Knut Vikør

Founding Editor

John Hunwick

VOLUME 16

The titles published in this series are listed at brill.com/isaf

Islamic Sufi Networks in the
Western Indian Ocean
(c. 1880–1940)
Ripples of Reform

By

Anne K. Bang

LEIDEN | BOSTON
Cover illustration: Collection of receipts for the transfer of Waqf funding from Zanzibar to Mecca, 1927.
Zanzibar National Archives (ZA-HD10/5).

Library of Congress Cataloging-in-Publication Data

Bang, Anne K.
Islamic sufi networks in the western Indian Ocean (c. 1880–1940) : ripples of reform / By Anne K. Bang.
p cm. — (Islam in Africa ; v. 16)
Includes bibliographical references and index.
ISBN 978-90-04-25134-2 (hardback : alk. paper) — ISBN 978-90-04-27654-3 (e-book)
1. Sufism—Africa, Southern—History. 2. Sufism—Madagascar—History. I. Title.

BP188.8.A356B36 2014
297.409679—dc23

2014016837

This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see brill.com/brill-typeface.

issn 1570-3754
isbn 978 90 04 25134 2 (hardback)
isbn 978 90 04 27654 3 (e-book)

Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive,
Suite 910, Danvers, ma 01923, usa. Fees are subject to change.

This book is printed on acid-free paper.

Contents

Foreword and Acknowledgements ix

List of Illustrations xii
List of Figures xiii
Note on Transliteration, Quotes and Dates xiv

1 Introduction 1
The Ripple and the Reef: Perspectives and Objectives 2

2 The Luminescent Sun and Brilliant Rays of Light: Towards a

Geography of Reform 20
Towards a Geography of Reform: A Web of Centres 22
The Ḥaramayn: The Blessed and the Radiant 26
The Ḥaḍramawt: Home of the Luminescent, Encompassing
Mid-Day Sun 29
Zanzibar: The Brilliant Star of East Africa 32
Lamu and the Riyadha Mosque 35
The Comoro Islands: Moon Islands in a Sea of Sun 37
Rays of Light and Hierarchies 45

3 The Branches of the Qādiriyya and the Shādhiliyya in Northern

Mozambique: Silsilas to the South 47
The Ṭarīqa Qādiriyya in Zanzibar 48
The Qādiriyya in Mozambique: Multiple Routes South 54
Muḥammad Al-Maʿrūf and the Spread of the Shādhiliyya in Northern
Mozambique 67
The Emergence of Sufi Orders in Norhern Mozambique 70

4 The Shādhiliyya in Northern Madagascar c. 1890–1940: The Planting

of a Garden and the Growing of Malagasy Roots 72
Islam in Northern Madagascar 73
Family, Religion and Trade on Madagascar: East African-Comorian
Networks and the Shādhiliyya 74
Aḥmad al-Kabīr: The Great Shaykh of the Ṭarīqa Shādhiliyya of
Northern Madagascar 79
Reform and the Emergence of a Malagasy Sufi Order 88
vi contents

5 The Cape Town Muslim Community and East African Sufi Networks:
Beyond the Monsoon 90
Islam in South Africa and Cape Town 90
Muhammad Salih Hendricks: From Periphery to Centre to the
Network 93
Other Travellers – More Daʿwa 101

6 Travelling Texts: Arabic Literate Learning in Coastal East Africa,

c. 1860–1930 108
Textual Transmission and Religious Authority 109
Book Knowledge in the Age of Manuscripts: 1860s into the Twentieth
Century 111
From Manuscript to Print: Parallel or Converging Authorities? 130
Manuscripts, Printed Books and Religious Authority 139

7 Ritual of Reform – Reform of a Ritual: Rātib al-Ḥaddād in the

Southwestern Indian Ocean, c. 1880–1940 143
Rātib Al-Ḥaddād as Sufi Reform 143
Rātib al-Ḥaddād in East and South Africa 148
The Rātib in Writing: Textualization of Charisma 153
The Rātib Performed: Reform of a Ritual? 156
The Rātib al-Ḥaddād: New Reform of a Reformist Ritual? 161

8 Consolidating the Network: Waqf Distribution and New Organizations

in Zanzibar, c. 1900–1930 163
Scholarly Networks and the Zanzibari “Meccan Waqfs”,
c. 1880–1940 164
Waqf Distribution within Intellectual Networks:
Consolidating Reform through Waqf Funds 179
From Networks to Organizations: The Rise of the Jamʿiyya,
c. 1900–1930 181

9 Conclusions 191
On Ripples and Reefs: Agency in a Translocal World 191
Sufi Reform on the Move 192
The Ecumene that wasn’t – yet? 196

Appendix 1 199
The Zanzibari “Meccan Waqfs” Contained in ZA-HD10 199
contents vii

Sources and Bibliography 201

Arabic Textual Material 201
Arabic Manuscript Sources (Listed) 204
Archival Sources 211
Oral Sources/Interviews 211
Bibliography 212

Index 225
Foreword and Acknowledgements

This book is the result of several research projects and a series of collaborative part-
nerships in the period 2006–2012, as well as time shared with generous individuals in
diverse locations in the southwestern Indian Ocean. Thus, my thank-you list is long,
my gratitude deep.
I would never have contemplated writing this book without the many stimulat-
ing conversations I was fortunate to have with Maalim Muhammad Idris Muhammad
Saleh (1934–2012) in Zanzibar. From our first meeting in 1998, until his passing in
March 2012, Maalim Idris was a teacher, a research-partner, and a friend, also to my
family and my extended family as they came visiting. Maalim Idris was best known
for his great collection of Islamic textual remnants of the Zanzibari past, and I was
among the many researchers who were fortunate enough to benefit from this. When
we were in Zanzibar, he would insist that the whole family came over for what he
called “morning report”, to drink Africafe and talk about the latest news. Occasionally,
I would revisit a question from the day before, and Maalim would get up and say “Yes,
about that; I found something . . .” and hand me a document that I had thought unob-
tainable, residing somewhere in the deep corners of in his collection. In this way,
literally haba na haba or coffee-cup by coffee-cup, the ideas that make up this book
were formed, and the material that could support them located. However, there were
also more active searches, and Maalim participated happily. We travelled to northern
Madagascar together, to a world where people, to Maalim’s occasional consternation,
spoke no English, broken Swahili and much French. He also came to Bergen, braving
the June cold and the strange behaviour of the never-setting sun. Thank you for the
time we shared.
I am also very grateful to Aydaroos and Ahmad Badawi and the entire leadership
of the Riyadha Mosque in Lamu. They were excellent partners in the Endangered
Archives project for digitizing the Riyadha manuscript collection. However, my grati-
tude is first and foremost due to their deep knowledge of their own tradition, and their
generosity in sharing it.
In Ilha de Mozambique, I was fortunate to be introduced to Hafiz Jammu, and in
Diego Suarez my gratitude is due to the leaders of the Comorian and Shādhilī mosques,
as well as the leadership of the northern Madagascar Shadhiliyya.
My gratitude is also due to Seraj and Rawda Hendricks of the Azzawiya Mosque in
Cape Town, for many stimulating conversations since 2005, and for a series of nice out-
ings, both in Zanzibar and in the beautiful surroundings of Cape Town.
My interest in East Africa would never have been awakened without Professor
R. S. O’Fahey, who as my supervisor in 1997, brought me on my first trip to Zanzibar.
That, as they say, was it. Today, I am grateful to him for never ceasing to take an ­interest
x foreword and acknowledgements

in the Muslim intellectual history of East Africa, for donating to me his extensive
library of the region and for reading parts of this book in manuscript form.
In Zanzibar, I have also been fortunate to benefit from the now sadly closed Zanzibar
Indian Ocean Research Centre (ZIORI), and from long-term collaboration with
Professor Abdul Sherrif. ZIORI was a perfect combination of stimulating colleagues,
quiet workplace and a good library. The two ZIORI conferences were true meeting-
places for researchers, scholars and opinion-makers, and they are sadly missed as a
reference point for many.

I am grateful to the following agencies for funding this research: The Norwegian
Research Council for funding the research project “Linking Global Cities”; The
Norwegian Research Council and the National Research Association of South Africa
for funding the collaborative project “From transmission of tradition to global learn-
ing: African Islamic education, 1800–2000” and the Endangered Archives Project,
British Library for funding the digitizing of the Riyadha manuscript collection.
The bulk of the research for this book was conducted during the project “Linking
Global Cities”, which from 2007 to 2010 had its intellectual home in the research centre
named Uni Global. There, the input of colleagues from many disciplines, and the con-
stant coming-and-going of guest researchers and lecturers, made for a highly stimulat-
ing home. I am grateful to all who were there.
After the unfortunate dismantling of Uni Global in 2010, this project found a new
and welcoming home at the Chr. Michelsen Institute (CMI) in Bergen. There, the
results could be slowly put into written form, and the results discussed with stimu-
lating colleagues in the Politics of Faith cluster. Generously, CMI provided me with
funded time for the final revisions of this book, which gave me an unequivocal com-
mitment to finish the job.
Finishing the job is the hardest part. Dr. Shamil Jeppie and the team of the
Tombouctou Manuscript Project provided me with a perfect hideaway in the winter
of 2012 to write the final sections of this book; Thank you to Shamil and Gigi, Susana
and Rifqah for making my stay both pleasant and productive, with constant access to
swimming pools.
I have had the privilege to discuss the topics and themes of this book with many
good colleagues, including (in no particular order) Liazzat Bonate, Iain Walker,
Jeremy Prestholdt, Scott Reese, Nefissa Naguib, Abdul Sherrif, Knut S. Vikør, Preben
Kaarsholm, Kai Kresse, Elke Stockreiter, Susana Molins-Lliteras, Shamil Jeppie, Eirik
Hovden, Hassen Muhammed Kawo, Gerhard Bruinhorst, Chapane Mutuia, Leif
Manger, Samson Abebe Bezabeh, Roman Loimeier, Kirsten A. Kjerland, Tore Sætersdal,
Nefissa Naguib, Valerie Hoffmann, Eugeniusz Rzewuski, Sophie Blanchy, Amal Ghazal,
Kjersti Larsen and Ridder Samsom. Over the years, sections of this book have also been
presented as work-in-progress at a number of conferences and workshops and I have
foreword and acknowledgements xi

been privileged with feedback from colleagues too numerous to mention but who
nonetheless deserve gratitude.
Last, but not least, I am grateful to the anonymous reviewers of this book, who pro-
vided valuable comments that much improved the focus of the work.
So many have shared their knowledge that this book feels almost like the result of a
long, and fun-filled team-work. That does not absolve me from sole responsibility for
my interpretations and – particularly – for my mistakes.
Friends and family have made fieldwork truly rewarding, and I am grateful to all
friends in Zanzibar, Lamu, Ilha de Mozambique and Cape Town for the good time spent
together. Above all, I thank my family, Per and Nora, for enduring all those long and
intolerable holidays in Zanzibar, and for making our journeys all the more meaningful.

This book is dedicated to my parents, Tulle Bang (1927–2006) and Gisle Bang (1927–2011).
List of Illustrations

map caption

1 Map of East Africa 21

image caption

1 The Shādhilī mosque in Diego Suarez 87

2 Muhammed Saleh Hendricks 97
3 The Azzawiya Mosque in Cape Town 101
4 Sample of manuscript, devotional writing. From the collection of
Maalim Idris Muhammad Saleh 126
5 An early printed version of Rātib al-Ḥaddād, Cape Town 155
6 Receipts from Mecca with signatures and seals. Zanzibar National
Archives 180
List of Figures

Figure Caption

1 The Zanzibar Qādiriyya “Brawanese Branch” 49

2 Zanzibar Qādiriyya, “Mjana Kheri Branch” 51
3 Zanzibar Qādiriyya, “Comorian Branch” 53
4 Ilha Qādiriyya through ʿĪsā b. Aḥmad 59
5 Qādiri silsila of Palma, Mozambique 60
6 Shādhiliyya of Ilha de Mozambique 69
7 Shādhiliyya, Ambilobe, Madagascar 82
8 Shādhiliyya, Diego Suarez, Madagascar 83
9 Isnād of Abū Madyan/Shādiliyya/ʿAlawiyya of Muhammad Salih
Hendricks 95
10 Identified devotional and Sufi texts in the Riyadha
manuscript collection 129
11 The Islamic educational publications by the Zanzibar Government Print
1900–1940s 138
12 Distribution of the “Meccan Waqfs” of Zanzibar, 1927–1936 168
13 Receipts signed and returned to Zanzibar 12 Jumada I 1347/
26 October 1928 171–173
Note on Transliteration, Quotes and Dates

In this work, Arabic is transliterated when referring from written Arabic text and to
names of people who are or were most commonly known by their Arabic name. The
names of living people in East Africa, Madagascar, Mozambique and South Africa have
not been transliterated but rendered in the form most commonly known in the Latin
script. Similarly, place names are rendered in the form most commonly known, e.g.
Ingazija rather than Injazīja for the island of Grande Comore. Words that are known
in European languages have been rendered in English form, e.g. Quran rather than
Qurʾān.
The letters of the Arabic alphabet are transliterated as follows:

ʾ b t th j ḥ kh d dh r z s sh ṣ ḍ ṭ ẓ ʿ f q k l m n h w y

In quotations from the Quran, the translation of A. Yusuf Ali has been used.

Hijri/CE dates are given as follows:

11 Rabīʿ I 1433/3 February 2012
1433/2011–12
In cases where the hijra year fall entirely, or almost entirely within one CE year, only
one CE year is given.
 chapter 1

Introduction

The subject of this book is Shāfiʿī Sunni Islamic reformist impulses in the
coastal regions of the southwestern Indian Ocean in the period c. 1860–1940,
as transmitted by the three main Sufi orders in the region at the time; the
Shādhiliyya, Qādiriyya and the ʿAlawiyya.
The main protagonists of the narrative are the religious specialists, the
“learned men”, Sufi leaders, scholars, organizers of rituals, missionary travel-
lers, founders of mosques and schools, the readers, writers, copyists and buyers
and sellers of books, teachers and students. They were mainly men. Although
there are several indications of a network of female scholars, especially within
the field of education, the male network is by far the best documented. These
were also men of different backgrounds. Many, but not all, had a genealogical
connection to the Ḥaḍramawt in Southern Yemen, and many, but far from all,
were sāda – i.e. claiming descent from the Prophet Muḥammad. Irrespective of
their background, they were actors, both in the larger development of Islamic
reformist ideas and in the shifting ideological, social and political processes
taking place in several locations in the southwestern Indian Ocean. They were,
in other words, acting both globally and locally.
No less important protagonists are the locations themselves. If the men
described above were actors, they were not acting upon passive objects with
no agency of their own. Rather, the societies in which they worked – Lamu,
Zanzibar, northern Mozambique, Diego Suarez (Antsiranana) or Cape Town –
were communities with specific social, political and cultural structures, which
formed, transformed and localized people, knowledge, rituals and practices.
What is portrayed in this book is the interaction of these two protagonists
over a period of approximately 70 years, from c. 1860 and into the 1930s. Some
of these interactions took place simultaneously at different locations, and only
when viewed together can we detect the trajectory emerging. However, simul-
taneity, as all writers know, is a big problem for words, constrained as they
are by the struggle of the narrator to integrate events into order by using even
more words. In the following narrative, all notions of sequence will be first
abandoned in favour of glimpses of the action/interaction taking place in one
location after another. Then, the perspective shifts, to focus on specific aspects
of reform as found in more than one location.
The narrative that emerges is the story of a wave crossing the open sea, mak-
ing landfall at different places at different points in time. The wave represents

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_��2

2 chapter 1

Islam and its various reformist ideas. Sometimes this is a veritable tsunami, at
other times a mere ripple, and most often, a cross-current, pulling in several
directions, but nonetheless representing change, perceived on the ground. The
local, coastal communities, be it the big city of Cape Town or the small town of
Mozambique Island, can be represented as the reefs or shores, unmoving but
with specific features that impact the way in which the wave washes ashore
and which riptides are created.

The Ripple and the Reef: Perspectives and Objectives

The first aim of this work is to link the locations of the Ḥaramayn (Mecca
and Medina), the Ḥaḍramawt, Lamu, Zanzibar, the Comoro Islands, northern
Mozambique, Diego Suarez and Cape Town with one general movement of
Sufi-based Islamic reformist activity. A second aim is to place changes in this
region within the wider Islamic reformist discourse in the mid- to late nine-
teenth century and the early twentieth century.
This will be done first through three case studies that emphasize the agency
of daʿwa-oriented teacher-scholars in northern Mozambique, Diego Suarez in
northern Madagascar, and Cape Town. The aim is here to bring out the reach,
agency and conscious daʿwa impulse of the Sufi orders in an era of increased
facility of travel, as well as the different patterns of localization.
The analysis then turns to a thematic approach and discusses specific
aspects of change that reached beyond specific locations. The first theme is
the diffusion of books and textual material in the region. The second theme
is ritual change, exemplified by the diffusion of text-based ritual expressions
such as Rātib al-Ḥaddād. The network of scholars itself and its transformation
into new organizational forms is the third object of analysis. Finally, conclu-
sions will be drawn about the impact of Islamic Sufi reform in the region and
the impact of the region on Islamic Sufi reformist ideas.

Point of Departure: Translocality and the Southwestern

Indian Ocean
The study of Islamic societies in Indian Ocean lands has, in the past decade-
and-a-half, received renewed attention, specifically from the point of view of
networks, meaning that social and religious change is viewed as a function
of people who are not part of the same local community. Nevertheless, these
people are somehow in contact and able to exchange goods, ideas and – more
loosely defined – culture. An early example of such an approach is Martin van
Introduction 3

Bruinessen’s account of the Naqshbandī order in Indonesia,1 which traced,

amongst others, the movements of the nineteenth-century teacher Shaykh
Ismaʿīl Managkabawī who, around 1850, returned after a prolonged stay in
Mecca. As a consequence, new, previously unaffiliated individuals travelled
to Arabia and took various branches of the Naqshbandiyya directly in Mecca.
Upon return, they established themselves as khalīfas (Ar. lit: successors, spiri-
tual leaders drawing authority from their predecessors) their home regions. By
1900, when the phenomenon came to the attention of Dutch officials, it had
become an established network with a strong “Meccan connection”.
Ideas of “connections”, “transmissions” and – ultimately – studies of migra-
tion, re-migration and return, have, since the 1990s, become focal issues in
the study of Islam in the Indian Ocean. Despite the apparent unifying effect
of Islam in coastal communities along the southwestern Indian Ocean rim,
local knowledge, practice and social identities have changed with time and
place, depending on both local events and global inter-connections. In the past
decade, a series of studies have been produced which view the Indian Ocean
rim as a global world, linked through persistent cultural contact, the religion of
Islam and, later, the emergence of European colonialism.2
These studies all work on the basis of a (more or less explicitly formulated)
notion of translocality, as an object of inquiry and as a research perspective.
Furthermore, these works have analysed the movements of people, goods and
ideas between the port cities of the Indian Ocean, specifically with a view to
the inter-civilizational encounters and ensuing cultural change resulting from
this traffic. This book is based on the same premise of the Indian Ocean as a
social space where the movement of people has created specific patterns of
family, trade and religious links, as well as specific hierarchies of centres and
peripheries. This book also draws on the argument made by U. Freitag and
A. van Oppen, that the history of (for example) Indian Ocean locations must

1 M. van Bruinessen, “The Origins and Development of the Naqshbandi Order in Indonesia”,
Der Islam, 67, 1990, 150–179.
2 As examples may be mentioned U. Freitag, Indian Ocean Migrants and State Formation in
Hadhramaut, Leiden (Brill), 2003, E. Ho, The Graves of Tarim. Genealogy and Mobility across
the Indian Ocean, Berkeley (University of California Press), 2006 and A. Ghazal, Islamic
Reform and Arab Nationalism. Expanding the Crescent from the Mediterranean to the Indian
Ocean (1880s–1930s), London (Routledge), 2010.
4 chapter 1

take into account not only the diversity of spatial variation, but also the spatial
categories of the actors themselves.3
Within the study of the western Indian Ocean, much scholarly attention
has been devoted to the Swahili coastal cities of Kenya and Tanzania, partic-
ularly the cities of Zanzibar, Mombasa and Lamu. These studies have linked
the coastal cities to the wider Indian Ocean context through analyses of reli-
gious development, family links, material culture, legal structures, consumer
patterns etc.4 Despite our increasingly nuanced knowledge about the historic
Swahili economic and intellectual centres, the connections onwards to the
Muslim communities of Mozambique, Madagascar and southern Africa have
received comparatively scant attention.
While ground-breaking research has been forthcoming in recent years by
L.J.K. Bonate,5 the full scope and scale of the connections between northern
coastal Mozambique – or: more specifically: of the Muslims of Mozambique –
and the wider Indian Ocean tradition have yet to be fully mapped.6 Chapter 2
of this book is an attempt to trace some of these connections, and thus also a
step towards understanding northern Mozambique as a historically integral
part of the Swahili – and Islamic – world.
While not to the same extent part of the Swahili cultural continuum, the
same is true for the coastal towns of northern Madagascar as well as the South
African cities of Durban and Cape Town. In these locations, Muslim communi-
ties have long-standing, but differently founded historical ties to the Islamic
societies of the East African coast, to the ports of South Arabia and onwards
to the Indian Ocean. It is the aim of this book to expand both the scope and

3 U. Freitag and A. van Oppen, Translocality. The Study of Globalising Processes from a Southern
Perspective, Leiden (Brill), 2010.
4 See for example K. Kresse, Philosophising in Mombasa. Knowledge, Islam and Intellectual
Practice on the Swahili Coast, Edinburgh (Edinburgh University Press), 2007; J. Prestholdt,
Domesticating the World. African Consumerism and Genealogies of Globalization, Berkely
(University of California Press), 2008.
5 L.J.K. Bonate, Traditions and Transitions. Islam and chiefship in Northern Mozambique ca.
1850–1974, PhD Thesis, University of Cape Town, 2007; L.J.K. Bonate, “Roots of diversity in
Mozambican Islam,” Lusotopie, XIV (1), 2007, 129–149; L. J.K. Bonate, “The use of Arabic script
in Northern Mozambique”, Tydskrift vir Letterkunde, 45:1, 2008, 120–129; L.J.K. Bonate, “Islam
in Northern Mozambique: A Historical Overview” History Compass, 8, 2010. L.J.K. Bonate,
“Documents in Arabic Script at the Mozambique Historical Archives”, Islamic Africa, 1:2, 2010,
253–257; L.J.K. Bonate and C. Mutiua, “Duas Cartas de Farallahi”, Estudos Mocambicanos, 22:1,
2011, 91–106.
6 E.A. Alpers, “A complex relationship. Mozambique and the Comoro Islands in the nineteenth
and twentieth century”, Cahiers d’Etudes Africaines, 73–95.
Introduction 5

the scale of this research, to the region that was clearly “south of the border”,
at least from the point of view of the Bū Saʿīdī Sultanate of Zanzibar, and “west
of the sun”, at least from the point of view of dhows that travelled by the mon-
soon winds.

Travelling Sufi Teachers and Scholars: Agency in the Age of Reform

and Globalization
For every departure, there is an arrival (that is, unless something goes ter-
ribly wrong). A journey has (at least) two definite points, and whatever
happens between is progress, setbacks, detours, or – if the traveller is so-
inclined – a learning process. For every arrival, however, there is not necessar-
ily a new departure, or an onwards journey. The traveller may choose to settle,
to “go native” or to become what today would be called an expatriate. The
teachers and scholars in this book present the researcher (and then the hapless
reader) with countless departures and arrivals, endlessly veering somewhere
on spectrum between “going native” and becoming a person with hyphenated
identity in one form or another. This book is an attempt to make sense of their
restless activity. Two generations of Shādhilī, Qādirī and ʿAlawī teachers will be
viewed as mobile agents of a specific form of globalization, forming a network
through which both Islamization and Islamic reform could flow.
As stated at the outset, and as will be discussed further below, this focus
in no way implies that the people in the societies in which these teachers
worked, had no agency of their own. Furthermore, the focus on teachers and
religious leaders in no way implies that “commoners” did not play important
roles in introducing new beliefs and practices, reforming existing ones and
establishing or maintaining institutions. As has been pointed out most explic-
itly by Felicitas Becker, the role of Sufi orders and caravan traders has been
over-emphasized due to their tendency to leave behind written sources, and
that the role of local leaders and “commoners” in the Islamization process, was
equally – if not more – important than the translocal agent.7 Becker argues
instead for a closer study of indigenous religious practice as it developed after
conversion (or: as it might be added, after religious change of any type, includ-
ing reform of an existing Islamic belief system).
While research in all the locales under scrutiny in this book is likely to find
Becker’s statement correct, this work draws mainly – although not exclu-
sively – on written sources (Arabic and Swahili, as well as colonial), and thus

7 F. Becker, “Commoners in the process of Islamization: reassessing their role in the light of
evidence from southeastern Tanzania”, Journal of Global History, 3, 2008, 227–249; F. Becker,
Becoming Muslim in Mainland Tanzania, 1890–2000, Oxford (Oxford University Press), 2008.
6 chapter 1

focuses the lens more on the “expert” than on the “commoner”. It my hope that
the latter will be brought into his – and more particularly: her – rightful place
by studies to come.

Raising Anchor: Mobility

Behind the goings-and-coming of the itinerant shaykhs and scholars lie
basic questions of migration. Beyond the economic and political push/pull
factors, what actually facilitates migration in an Islamic context such as the
southwestern Indian Ocean? The answer to the first question has often been
that migration is facilitated by the presence of some sort of network, be
that based on family, trade, religion or any other group or organization of a
translocal or transnational nature.8 Earlier studies have pointed to the factors
that facilitated persistent migration to East Africa as well as the propensity
for inter-regional migration, and it is not the aim of this book to add to this
research.9 Rather, the object of this study is the knowledge that travelled with
migration, including the knowledge of migration itself, particularly knowledge
about the transnational networks created by migration. This approach takes
its cue from Engseng Ho,10 who has drawn attention to the reverse image of
mobility, namely absence – the person or persons who are not present. The
absent, however, are not absent so long as their place in the larger context is
known, in the sense of having a place in a context that is not only described
geographically, but also genealogically, or because he is the teacher of/
student of/author of/owner of/someone or something. He is, in other words,
still among us, but not here. This knowledge is formulated in the narratives
of migration that in turn shape and reshape, facilitate or – at times – object
to migration and implicitly to the formation of new contacts. Ho also demon-
strates clearly that this knowledge may be transmitted in the form of texts that
in different ways address the issue of migration. Finally, Ho also emphatically
draws our attention to the fact that this vision of a network incorporates those
still alive as well as those who have died, whether at home or away. The idea of
movement transcends each lifetime and perpetuates migration by the narra-
tives constructed by those moving and those staying behind.

8 See for example R. Cohen, Global Diasporas: An Introduction, London (UCL Press), 1997;
S. Vertovec and Cohen (eds.), Migration, diasporas and transnationalism, Cheltenham
(Elgar), 1999; R. Cohen, Migration and its enemies. Global Capital, migrant labour and the
nation state, Aldershot (Ashgate), 2006.
9 See for example A. Sheriff, Dhow Cultures of the Indian Ocean. Cosmopolitanism, Commerce
and Islam, London (Hurst), 2009.
10 E. Ho, The graves of Tarim.
Introduction 7

This type of study has no bounded field. As urged by Gupta and Ferguson,
this book focuses on several locations, or rather; a cluster of people in a clus-
ter of locations, in the sense that the phenomena under study are situated
in different locations, geographically but also linguistically, culturally and
hierarchically.11 Thus, the first framework of this study is the set of connec-
tions created by individuals and communities rather than the geographical,
linguistic, cultural or social sites where a phenomenon (in this case Islamic
reformist impulses) can be found. From this starting point, we may turn to how
knowledge (Islamic, but also knowledge about the Islamic networks that Sufi
orders constitute) was perpetuated, in space and time. How was this narrated
and how did this in turn affect potential or actual transmission within and
beyond the local?

Passage: Islamic Reform within the Sufi Episteme

This study argues that the activities of a certain set of scholars and intellec-
tuals, combined with locally-specific agency, produced “something new”,
something that was not there before. This “something new” can be sought on
the level of ideas and belief, i.e. in theology and doctrine as transmitted orally
and in writing, or it can be sought in the observable world in the form of prac-
tice. As outlined above, this study will focus on the emergence of three distinct
new developments in coastal East Africa, all observable in the social and cul-
tural realm:

1. An increased emphasis on writing in the transmission of learning, observ-

able in an increased circulation of manuscript and printed text, espe-
cially in Arabic but increasingly also in Swahili. This development is
closely linked to a growing emphasis on Islamic education (as discussed
in Chapter 6).
2. New ritual forms, observable in both the circulation of text-rituals, but
also in what rituals were actually performed (as discussed in Chapter 7)
3. New networks and new organizational forms, observable through the
forming of new (or: in some cases, the revitalization of old) networks,
observable in travel patterns and teacher-student networks. Also notable
were the formation of new institutions and organizations, especially
schools and associations (as discussed in Chapter 8).

11 A. Gupta and J. Ferguson, ‘Discipline and Practice: The “Field” as Site, Method and Location
in Anthropology,’ in: A. Gupta and J. Ferguson, Anthropological locations: Boundaries and
Grounds of a Field Science, Univ. California Press, 1997.
8 chapter 1

It is tempting to ascribe these changes directly to the emergence of reformist-

oriented Sufism in the region, represented most prominently by the Qādiriyya,
Shādhiliyya and the ʿAlawiyya. However, new organizational forms or new
modes of transmission are not necessarily the result only of fully formulated
ideological and/or religious doctrine. As R. Loimeier has pointed out, reform is
not only “change with a programme”, but also a set of responses to social, cul-
tural and political change.12 In East Africa, an increased emphasis on writing,
for example, can just as easily be ascribed to the nineteenth-century hegemony
of the Bū Saʿīdī state, and its need for qāḍīs (legal scholars), liwālis (governors),
scribes etc. Ritual changes can be ascribed to demographic change as Islam
came to be practiced by new groups. Finally, the new forms of organization can
be ascribed to a multitude of influences, including the colonial, be that British,
French or Portuguese.
Nonetheless, this book aims to show that these changes were primar-
ily the result of Islamic reform, operating entirely within the Sufi paradigm.
I also argue that these changes took different forms in different locations, but
that some similarities can be observed. This argument, however, can only be
made if we accept that such a thing as “reform” actually took place within
Sufi-oriented Islam, and if we assume that this was a process that played out
both globally and locally. What follows is a clarification of the understanding
reform, as applied in this study.

Something Old, Something New: What is Islamic Sufi Reform?

The “something new” that can be traced in East/Southeast Africa in the nine-
teenth century, has been traced in other parts of the Islamic world too. In the
last decades of the twentieth century, researchers started to attribute nine-
teenth-century changes to a new “form” of Sufism, labelled reformist, revival-
ist, or “neo”.13 The argument was that Sufism came out from its inward-looking,
reclusive mode some time in the seventeenth century and turned towards vari-
ous form of social activism, and that their religious outlook was sufficiently
“new” to merit a new label.
Several scholars later pointed out that there was “nothing “neo” about neo-
Sufism”, as O’Fahey and B. Radtke put it in their 1993 article.14 V. Hoffmann15

12 R. Loimeier, ‘Is there something like “Protestant Islam”?’, Die Welt des Islams, 45:2, 2005,
216–254.
13 For a retrospective account of the debate and a re-appraisal of the term, see J. Voll, “Neo-
Sufism reconsidered again”, Canadian Journal of African Studies, 2/3, 2008, 314–330.
14 R.S. O’Fahey and B. Radtke, “Neo-sufism reconsidered”, Der Islam, 1, 1993, 52–87.
15 V. Hoffmann, “Annihilation in the Messenger of God: The Development of a Sufi Practice”,
Int. J. of Middle East Studies, 31:3, 1999, 351–369.
Introduction 9

has demonstrated the long antecedents of the aims and aspirations of Sufism,
that were clearly part and parcel also of the writings of the so-called “neo-
Sufis”, including many discussed in this book.16 Among them is an explicit
devotional focus on the Prophet, and the quest for “annihilation” in his perfec-
tion ( fanāʾ fī-’l-rasūl). Hoffmann dates the belief in a pre-existing, primordial
“Muḥammadan light” back to the ninth century, and also points out that devo-
tion to the Prophet, expressed in mawlids and poetry can be found from the
thirteenth century. B. Radtke, focusing on the eighteenth century, has shown
a series of related continuities, both in aims, scope, method and terminology.17
E. Ho has pointed out that the distinction between the “classical” and the
post-eighteenth century is artificial also when it comes to the actual practices
of the Sufis themselves. As he demonstrates, law and mysticism has co-existed
since at least the 15th century, meaning that the Sufi shaykh might also be the
qāḍī or mufti of the city, or take on other legal or social functions.18 Rather than
constructing a bāṭin (esoteric)/ẓāhir (exoteric) dichotomy, he argues that the
Sufis themselves were scholars oriented towards both, long before the move-
ments described in academic scholarship as “reformist” or “neo”.
However, even if we accept that the theological content of the nineteenth
century Sufi orders (as they expanded in Eastern Africa) were still adhering to
a (more or less) unbroken chain of authority and unchanged spiritual content,
this does not preclude them from being instigators of change – nor, for that
matter; respondents to change. In other words: The will to instigate changes in
one’s community (or, importantly: in other communities), does not necessarily
have to be founded in theological breaks with a pre-existing tradition.
R. Loimeier has linked some specific features of social activism directly to
Islamic reformism, including reform coming from within the Sufi episteme;
Popularization of education, emphasis on independent access to text, a marked
scepticism towards “superstitions” (however defined), and a delegitimization
of political authority.19 These are changes in function, rather than in spiritual

16 For a good example, see the “ijāza and waṣiyya” of Aḥmad b. Sumayṭ, discussed in
A.K. Bang, “Zanzibari Islamic knowledge transmission revisited: loss, lament, legacy – and
transformation”, Social Dynamics, 38:3, 2012, 419–434. Although the “guide” from Aḥmad b.
Sumayṭ to his son ʿUmar is not discussed in this article as an expression of reformist ideas,
it perfectly confirms V. Hoffman’s point of continuity in the aims and aspirations of the
practicing Sufi. For example, Ibn Sumayṭ repeatedly recommends the texts of al-Ghazālī,
the ultimate aim of fanāʾ etc.
17 B. Radtke, “Sufism in the 18th century: An attempt at a Provisional Appraisal”, Die Welt des
Islams, 36:3, 1996, 326–364.
18 E. Ho, The Graves of Tarim, 164–165.
19 R. Loimeier, ‘ “Protestant Islam”?’. Loimeier further adds new forms of social organiza-
tion to the list, but links this more directly with the recent reform movement of the late
10 chapter 1

content, or a redefinition of what the social (and in some cases, political) role
of the Sufi order should be. In my view, the re-orientation charges the Sufi
leader, but also the regular murīd (student) and member, with agency. This, in
turn, can be applied to advocate change, or to respond to change. While agency
is certainly not “almighty”, but framed by varying set of contexts, this study
views the agency of Sufi leaders and regular ṭarīqa members as one among
several crucial factors for change.20
R. Loimeier has further defined Islamic reform as “change with a
programme.”21 This is an inclusive definition that to some extent disregards the
connection between spiritual belief and social action, and which thus allows
for the inclusion of Sufi-oriented reformist movements that operated entirely
within the esoteric episteme. A good example of the notion of “programme” is
a work by ʿAbd Allāh b. ʿAlawī al-Ḥaddād (1634–1719), the Imām and quṭb (spiri-
tual axis or pole) often credited for being the first and main reformer of the
ṭarīqa ʿAlawiyya. In his work al-Daʿwa al-tāmma,22 he outlined the roles and
responsibilities, as well as the concrete relationship between various social
groups, including the scholars, Sufis, rulers, traders, poor, women and children
and unbelievers. Behind the outline lies a clear notion of societal engage-
ment and a vision of change, firmly anchored in the ʿAlawī Sufi cosmology but
applied broadly to society.
It is also the definition that will be followed in this book, with one important
addition, namely the implicit understanding that change can be instigated not
only here (locally) but also potentially elsewhere, and even anywhere (glob-
ally). In other words: The Sufi reformist movements were prepared to insti-
gate change or respond to changes in several, and widely different locations,
as embodied in a formulated daʿwa impulse. The conclusive definition for the
purpose of this study is thus: Change with a programme and readiness to travel.

t­ wentieth century. However, as will be discussed in Chapter 8 of this book, new forms of
social organization was also part of the Sufi reform movement, particularly within the
field of daʿwa.
20 See also R. Loimeier, “Patterns and Peculiarities of Islamic Reform in Africa”, Journal of
Religion in Africa, 33:3, 2003, 237–262.
21 R. Loimeier, ‘ “Protestant Islam”?’, 219 and R. Loimeier, Between Social Skills and Marketable
Skills. The Politics of Islamic Education in twentieth century Zanzibar, Leiden (Brill), 2009,
62.
22 ʿAbd Allāh b. ʿAlawī al-Ḥaddād, al-Daʿwa al-tāmma wa-l-tadhkira al-ʿāmma, word press
online edition, 1461/2000: http://seekerofthesacred.wordpress.com/arabic-pdf-books/.
Introduction 11

Local Reform and Global

Evidence of reform-oriented Sufism may be found in the coastal regions of East
Africa from the mid/late 1800s. This, in turn, may be linked to a broader, more
widely oriented reform of Islamic Sufism, which has been dated back at least
to the eighteenth century. In Sufi reformist literature, the word daʿwa – literally
summon or call – by the late nineteenth century had taken on the distinctive
meaning of missionary or teaching activity. In some locations in the southwest-
ern Indian Ocean, the reformist impulse reached shore in communities that
were recognized as regional centres of learning (such as Lamu and Zanzibar).
In other places, such as in northern Mozambique, the process coincided with
or only briefly post-dated the arrival of Islam, or more specifically: With the
conversion of entire population groups to Islam. Finally, in Diego Suarez and
Cape Town, the social context was that of diversity, colonially administered
and locally defined.
One issue that will be revisited repeatedly throughout this book is the ques-
tion of reform as both a global and local discourse. In 2005, A. Tayob wrote: “A
global discourse among Muslims seems no longer to be questioned.”23 Tayob
went on to state that “very little has been documented about the nature of
this discourse, and how it is constructed alongside local Islamic discourses”.
While research has been forthcoming since, the latter part of Tayob’s state-
ment remains true when it comes to the southwestern Indian Ocean in the
late nineteenth and early twentieth century. In this work, Tayob’s argument
has been applied historically on Islamic discourse in the southwestern Indian
Ocean in the late nineteenth and early twentieth century. What was the nature
of Islamic Sufi reformist thought in the region, and how was it constructed
alongside – or merged with – local Islamic discourses? And, not least: should
we interpret this as part of a wider global discourse – or rather a local one that
may or may not be mistaken for global by the historian due to the presence of
themes or topics that prevail within the discourse now accepted as “global”?
The discussions takes its cue from Talal Asad24 and his notion of histori-
cally evolving discourses, whereby a given practice or opinion has both a past,
a present and a set of potential futures, all of which can be anchored in local
as well as translocal (or: global) discourses. The discussions in this book are
also inspired by Samira Haj’s understanding of reform within specific social

23 A.K. Tayob, “Muslims and Globalization in post-Apartheid South Africa”, paper presented
to the AEGIS European Conference on African Studies conference, London, 2005.
24 T. Asad, “The Idea of an Anthropology of Islam”, Occasional Paper Series, Georgetown
University (Center for Contemporary Arab Studies), 1986.
12 chapter 1

settings, both in place and time.25 As Haj points out, a tradition “is not sim-
ply the recapitulation of previous beliefs and practices; rather, each successive
generation confronts its particular problems via an engagement with a set of
on-going arguments.”26 The same can be said for people in different places,
who express belief in different languages, with reference to different social or
political realities.
Another way of approaching Islamic reformist thinking as a global phenom-
enon is to focus on what impact reformist impulses actually had on its political
and social surroundings – the “facts on the ground”. This was the approach of
Ahmad Dallal, who in his 1993 essay emphasized the wide range of opinions
and usages that came out of the late eighteenth and early nineteenth century
reformist drive.27 Focusing on four different thinkers, Dallal demonstrates
that these had widely different orientations: Shāh Walī Allāh (1703–1769) as
a synthesizer and mediator, Muḥammad b. ʿAbd al-Wahhāb (1703–1787) as
purely doctrinal, ʿUthmān dan Fodio (1754–1817) as primarily concerned with
the social and Muḥammad al-Sanūsī (1787–1859) as having a mainly commu-
nal focus. Beyond the fact that some of these reformist thinkers studied with
the same ḥadīth teachers in Mecca, their orientations were, in fact, divergent.
While Dallal’s close reading of the four scholars clearly brings out the differ-
ences between them, he makes little attempt to explain why they come out
with such a variety of vision, method and policy. The explanation may be
sought in the four ʿālim’s respective scholarly inclinations and temperaments,
but perhaps more likely: in the societies in which they worked.

Sufi Reform Versus Other Strands of Reform

In the period under study here (c. 1860–1930), Sufi-based reform was hardly
the only intellectual current that sought social, political or cultural change. In
addition to the various networks to be presented in this book, we may iden-
tify three main “strands” of global Islamic discourse present in East/Southeast
Africa. While research exists on all, the depth of knowledge is uneven:

25 S. Haj, Reconfiguring Islamic Tradition. Reform, Rationality and Modernity, San Francisco
(Stanford University Press), 2011. Haj approaches the work of Muḥammad ʿAbduh not,
as often portrayed, as a break with tradition, but rather as part and parcel of an Islamic
tradition.
26 S. Haj, Reconfiguring Islamic Tradition, 6.
27 A. Dallal, “The Origins and Objectives of Islamic Revivalist Thought, 1750–1850”, Journal of
the American Oriental Society, 113:3, 1993, 341–359.
Introduction 13

1. Pan-Islamism as propagated by the Ottoman Sultanate

Of all discourses that can potentially be labelled “global” by the late nine-
teenth century, this is by far the least studied when it comes to East/
Southeast Africa. For East Africa, the only work based on Ottoman archi-
val material is a BA thesis by H. Babavatan.28 There are no studies of the
impact of the pan-Islamist agents sent by the Ottomans to Zanzibar. It is
worth noting that Babavatan concludes unambiguously that the formal
relations between the Zanzibari Sultans and the Sublime Porte was part
of the Pan-Islamist drive of Abdul Hamid II (r. 1876–1909), whose reign
coincided with that of five Bū Saʿidī Sultans. While Pan-Islamism is not
the focus of this book, its presence and impact on Islamic discourse also
in East Africa is taken as given.

2. Islamic Modernism/Salafism
The past ten years has seen innovative research also on the emergence
of Islamic modernism in early twentieth-century East Africa and its
impact on Islamic institutions, practices and “culture”. As has been shown
most clearly by A. Ghazal, Islamic modernist discourses, as exemplified
by Muḥammad ʿAbduh and Rashīd Riḍāʾ and the flurry of journals pub-
lished in Beirut and Cairo, were known and discussed in East Africa
before the end of the nineteenth century.29 As Ghazal also demonstrates
for the Zanzibari Omani elite, their religious and linguistic ties predis-
posed them to follow – and participate in – the global Islamic discourse.
R. Loimeier has shown in great detail the impact of “globalized” modern-
ist thinking on Islamic education throughout twentieth century Zanzibar,
as well as the relationship between modernism and a colonially formed
vision of education.30 This book does not aim to describe the emergence
of modernist ideas in the southwestern Indian Ocean. Rather, it will point
towards, but not discuss in detail the way in which the Sufi mode of
reform both developed into – and served as a counterpoint for – Islamic
modernism and Salafism by the 1930s and 40s. As S. Reese has pointed out
with reference to Aden, Salafism too, in its local manifestation, was con-
stituted not only from the vast corpus of Islamic knowledge (including
Sufism) but also locally. In Aden, these local expressions often had Sufi

28 H. Babavatan, Understanding Afrikā-yi Osmani in the late Ottoman period. The case of
Zanzibar, BA thesis, Bogazici University, Turkey, 2000.
29 A. Ghazal, Islamic Reform and Arab Nationalism.
30 R. Loimeier, Between Social Skills.
14 chapter 1

antecedents.31 As the following chapters will show, a corresponding

development can be traced in Lamu, Mombasa and Zanzibar, if not fur-
ther afield. It is my hope that this work can form a background for a simi-
lar research on Salafi thought in eastern and southeastern Africa.

3. Nationalism
The studies that exist on nationalism and decolonization in Kenya,
Tanzania, Mozambique or Madagascar are more general, and less con-
cerned with nationalism expressed in Islamic terms (be that in the dis-
course of the Islamic umma, the khalīfate or the more general call for
decolonization).32 In the special case of Zanzibar (with its multi-ethnic
make-up and its experience with the revolution of 1964), the recent study
by J. Glassmann33 traces the origins and trajectory of different types of
nationalist discourse. J. Brennan’s study of racially based nationalism in
Dar es Salaam describes the twentieth century emergence of race as a
crucial denominator for identity.34 Both studies take note of the latent
nationalism in Islamic modernist ideas, but describe the shift towards
racial thinking as a consequence of colonialism. In this book, there will
be few references to fully developed arguments that can be labelled
nationalist. However, as in the case of modernist discourse, the embryos
of different form of nationalism can be traced in reformist intellectual
discourse. Within the field of Islamic education, and in the emphasis on
Arabic language we may see the roots of a wider, Islamic nationalism that
transcended ethnicity as well as colonial demarcation lines. However, in
the organized form of religious identity typified by the Sufi orders may
also be seen the kernels of ethnically, racially or even linguistically
defined nationalism. In other words: The localizing of Islam also created
new or cemented existing social groups or strata, which later found
expression in different types of nationalist discourse.

31 S. Reese, “Salafi transformations. Aden and the changing voices of religious reform in the
interwar Indian Ocean”, Int. Journal of Middle East Studies, 44, 2012, 71–92.
32 An exception is a thesis by J.W. Gonia, The Umma Ideal and Muslims in Madagascar.
Movement from Community to Community, Master of Theology Thesis, Luther Theological
Seminary, St.Paul, Minnesota, 1997.
33 J. Glassman, War of Words, War of Stones. Racial Thought and Violence in Colonial Zanzibar,
Bloomington (Indiana University Press), 2011.
34 J. Brennan, Taifa. Making Nation and Race in Urban Tanzania, Ohio University Press, 2012.
Introduction 15

In the context of the western and southwestern Indian Ocean, several ques-
tions can be raised. First: Who participated in these “global discourses” and
why? Secondly: When shaykhs and scholars in East/Southeast Africa engaged
in intellectual contact with the wider Islamic world, what was a nature of the
exchange, and how was its perceived by the interlocutors on both sides? For
example, when the Zanzibarī qāḍī and venerated Sufi Aḥmad b. Sumayṭ wrote
to the renowned Cairo scholar Muḥammad ʿAbduh in 1900, asking for a fatwā
(Ar: legal ruling, opinion) on some grammatical points in Ibn Ḥajar’s work Fatḥ
al-Jawād bi-sharḥ al-Irshād, what was he really doing?35 Participating in the
“global discourse” of Islam? This is not a far-fetched interpretation, given the
education of Ibn Sumayṭ in several of the centres of Islamic learning, includ-
ing Cairo.36 However, the correspondence may also be interpreted as that of
an established tradition within Islamic scholarship, whereby one ʿālim seeks
the opinion of another (Ar: istiftāʾ, request for a legal opinion), wherever he
may be. As historians, we cannot know how Ibn Sumayṭ or Muḥammad ʿAbduh
understood these exchanges. However, it is worth being aware that evidence of
such “global discourse” may also stem from older traditions that were not cre-
ated by nineteenth century modernity.37
Similarly, the near-contemporary, but more well-known istiftāʾ from the
Muslim community in Transvaal, South Africa, to Muḥammad ʿAbduh may
also be interpreted as participation in a global Islamic discourse.38 However,
the questions posed by the Transvaal Muslims (the wearing or non-wearing
of hats, the lawfulness of a Shāfiʿī praying behind a Ḥanafī Imām, and the eat-
ing of meat killed by Christians) clearly refer to concerns that were local, and
specific to the small Transvaal community. The fact that ʿAbduh’s fatwā caused
an uproar in Egypt (having declared lawful the eating of meat prepared by
Christians), was, in effect, not really the concern of the Transvaal Muslims.
On the more general level, globalization, however interpreted, is bound to
generate conflict, for the simple reason that “new” will meet “old”, or “reform”
meeting “tradition”, at some stage, in particular settings at particular times.

35 MORONI/1. Notebook in Sumayṭ family possession, containing notes and poetry by

Aḥmad and ʿUmar b. Sumayṭ. Copied, Moroni, June 1997, 22–23.
36 On the life of Aḥmad b. Sumayṭ, see A.K. Bang, Sufis and Scholars of the Sea. Family
Networks in East Africa, 1860–1925, London (RoutledgeCurzon), 2003.
37 See discussion by O. Scharbrodt, “The Salafiyya and Sufism: Muhammad ʿAbduh and his
Risalat al-Waridat (Treatise on Mystical Inspirations)”, Bulletin of the School of Oriental
and African Studies, 70:1, 2007, 89–115.
38 M. Sedgwick, Muhammad Abduh, Oxford (OneWorld), 2010, 97–99.
16 chapter 1

As has been pointed out by several authors,39 controversies over changing

practices takes place in a social context whereby established power-structures
(economic, political and relating to status) play into a debate that is phrased in
religious terms. This work will trace the conflicts (and take note of the lack of
conflict!) that arose when Sufism as an organized way of Islamic worship was
introduced into east/south-east Africa in the late nineteenth century.

Destinations: The Agency of Local Societies

We now turn to the reverse set of questions, posed from the point of view
of the reef, so to speak. If the ripples of reform influenced local knowledge
and practice, how did the reef absorb and influence the ripple? As has been
pointed out by F. Becker in the mainland Tanzanian context, the process of
Islamization was a process of innovation and adaptation – of ritual practice, of
teaching procedures and of geographical orientation. Becker’s understanding
of religious change is, in many mays, similar to that outlined above, whereby
religious change is “best understood as a response to a changing way of life”.40
The question then becomes how local beliefs, practices, ritual and social struc-
tures influenced the notion of what actually constitutes Islamic society and
notions of how this society could or should be changed.

Agency and Accommodation

Some of the locales under scrutiny in this book are situated explicitly on the
periphery of the Swahili world; Ilha de Mozambique and Diego Suarez in par-
ticular. Cape Town, on the other hand, is situated on the periphery of mul-
tiple worlds. Furthermore, the centres of the Swahili world (such as Lamu and
Zanzibar) were also peripheral to the Middle East heartlands.
A good starting point is the assertion made by several authors in the past
decade about the transitory and plural character of Swahili society and his-
tory. Swahili society, to borrow a phrase from R. Loimeier and R. Seesemann,41
is perceived here as multiple worlds, reaching across the Indian Ocean to the
Ḥaramayn, the Ḥaḍramawt and the Gulf and from there onwards to India,
into the African mainland, northwards to the Benadir coast and onwards to
Ethiopia and Eritrea, and finally southwards to the Comoro Islands, parts of

39 See for example R. Loimeier, “Zum sozialen Kontext eines religiösen Rituals”, Der Islam,
83, 2006, 170–186.
40 F. Becker, Becoming Muslim, 65.
41 R. Loimeier and R. Seesemann, The Global Worlds of the Swahili. Interfaces of Islam,
Identity and Space in nineteenth and twentieth century East Africa, Hamburg (LIT Verlag),
2006.
Introduction 17

Mozambique and the northern regions of Madagascar. The coastal rim of these
regions were, by the mid-nineteenth century, predominantly Muslim, influ-
enced over centuries by contact across the sea towards Arabia.
The cases presented in this book all address “travelling Islam”42 in the sense
that all the cases present some form of religious change which can be traced
to similar or related changes elsewhere. An essential question is how change
was legitimized locally in religious terms, and how it was related to a distant
(perceived) normative understanding of Islam or Muslim society.
Related to this is the question of how these new impulses came to be
accommodated to existing interpretations, rituals, practices, texts and orga-
nizational forms. What were, to borrow David Robinson’s words, the “paths
of accommodation?”43 Or to pose the question differently, focusing on social
structure: Where lay what C. Geertz called “the crucial juncture of synapses
of relationship which connect the local system with the larger whole”?44 Who
adopted a new ritual such as Rātib al-Ḥaddād or a practice such as ṭarīqa mem-
bership – and why? Who did not? Who sat down to painstakingly copy the Sufi
manuals of the Qādiriyya or the long silsilas (chain of authority) going back to
creation? What local concerns did such acts respond to? And how did these
concerns shape which aspects of Islamic discourse and practice was appropri-
ated locally? Finally: To what extent, and how did local concerns determine
what was transmitted onwards in space and time?

The Other Ports: Reform as a Network of Locations

In this book, the word network will appear time and time again, both with
reference to people, books, texts, rituals and institutions. The concept of net-
works first became a tool for the Manchester school of social anthropology45
and has since developed in several incarnations, most recently for the analysis

42 P. Mandaville used the phrase “travelling Islam” in his work Transnational Muslim Politics.
Re-imagining the Umma, London (Routledge), 2001. Although Mandaville’s work deals
primarily with Muslims in the West, the processes which he describes are equally appli-
cable to religious reform reaching other Muslim communities at other times by other
means. The “movable” nature of Islam was also emphasized by D. Eickelmann and
J. Piscatori (eds.), Muslim Travellers: Pilgrimage, Migration and the Religious Imagination,
Berkely (Univ. California Press), 1990.
43 See D. Robinson, Paths of Accommodation. Muslim Societies and French Colonial Authorities
in Senegal and Mauretania, 1880–1920, Oxford (James Currey), 2000.
44 C. Geertz, The Religion of Java, Chicago Univ. Press, 1960, 249.
45 See for example J.C. Mitchell, “The concept and use of social networks,” in: J.C. Mitchell,
Social Networks in Urban Situations, Manchester, 1989. For an applied use of the concept,
see for example F. Barth, Sohar. Culture and Society in an Omani town, Baltimore, 1983.
18 chapter 1

of cyber-networks and its social implications. R. Loimeier and S. Reichmuth46

(1996) were among the first to explicitly apply network analysis directly to
Islamic knowledge transmission. Their approach does seem ideally suited
to the study of Islamic religious networks, and in particular, the approach
lends itself to the study of traditional Islamic education, which was (and to
some extent, still is) based on a web of teacher-student relationships, docu-
mented in the form of an ijāza (Ar: certificate, proof of proficiency). This is
particularly pronounced in Islamic learning within the Sufi parameter, as these
tended to focus on personal instruction, of exoteric learning but also on the
more esoteric aspects of religion. For the historian of Indian Ocean Islamic
history, these networks often present themselves as a complex web of hagiog-
raphies that primarily lists the teachers of the individual portrayed, the various
educational journeys and pilgrimages undertaken, alongside his genealogical
origin.
This complexity poses a challenge not only to conceptualization on a theo-
retical level, but also organization on an empirical level. When tracing Islamic
networks, the researcher is, more often than not, confronted with massive
amounts of detail, which in turn threatens to grow into a blinding forest, over-
shadowing every tree in sight. This reflects a focus on the quantitative proper-
ties of a network – its extent and its density – and the comparatively scant
attention attention awarded to qualitative aspects such as intensity, direction
and content. The focus on links and transmissions run, in fact, the risk of col-
lapsing into “ijāza sociology” or “silsila history” whereby the most mundane
communications be construed as evidence for a network.
This criticism is related to that posed by F. Bourgoin,47 who stated that
the concept of network implies a lateral exchange that obscures other forms
of local hierarchy – political, social, or economic. Furthermore, as Bourgoin
notes, the focus on networks alone precludes a discussion of how any given
network is embedded in larger, overarching and hierarchical systems.
Taking note of these objections, this book will address Sufi networks as sys-
tems that transcend political, cultural and ethnic boundaries, and which thus
became embedded in different hierarchies at different times. In other words,
Sufi networks will be viewed as lateral, yet clearly hierarchical in their local

46 R. Loimeier and S. Reichmuth, “Zur Dynamik religiös-politischer Netzwerke in muslim-

ischen Gesellschaften”, Die Welt des Islams, New Series, 36/2, 1996, 145–185.
47 F. Bourgoin, “From Network to Class? Towards a More Complex Conception of Connection
and Sociability”, Working Paper of Danish Institute of International Studies. Copenhagen:
Danish Institute of International Studies, 31:2009.
Introduction 19

manifestations. They will also be viewed as informal, yet clearly imbued with
defined layers of authority that were constituted locally. Equally, the orders
were long-lasting in their essence, yet often limited in time to specific relation-
ships. From this starting point, answers will be sought to one specific ques-
tion: What was the nature of the connections “on the ground” and what type of
social change did they bring about?
 chapter 2

The Luminescent Sun and Brilliant Rays of Light

Towards a Geography of Reform

Some time in the late 1600s, a man named ʿAbd Allāh b. Aḥmad Jamal al-Layl
died at sea somewhere between the islands of Grande Comore and Nosy Be off
the northern coast of Madagascar. As was customary, ʿAbd Allāh was shrouded
for burial, and by dawn, the crew prepared to commit his body to the sea.
Just as the sun rose on the eastern horizon, a big flock of birds appeared and
landed on the body – one after the other, until they were so many that they
covered it entirely. To the astonishment of the assembled group, the birds flew
off with the body, out to the open sea and soon disappearing on the horizon.
The deceased ʿAbd Allāh b. Aḥmad was posthumously given the title Ṣāḥib
al-Ṭuyūr, the “Master of Birds”.
This narrative is related in several printed works, and is often retold by rep-
resentatives of the Jamal al-Layl family even today.1 It is in other words a well-
known element in a wider historical narrative of the Haḍramī ʿAlawīs in East
Africa, aiming to incorporate the “Master of Birds” into the web of the ʿAlawī
diaspora. The narrative locates him among the “climbers of the rock face”, to
use the phrase of E. Ho, whereby each generation is linked to the next through a
set of genealogical or intellectual connections.2 Portraying a miraculous event,
the narrative also makes a claim to sainthood, emphasizing the perceived posi-
tion of the deceased in his own time, and not least: in his own place.
Place being the key word here, we may also read the story as discussion of
location in the world of seaborne Islam. When he died, the “Master of Birds”
was nowhere in particular, literally in an intermediate position. If we assume
that the birds flocked to take ʿAbd Allāh’s body away to its rightful burial
ground, they would have many destinations to choose from, as the deceased

1 Al-Mashhūr, Shams al-ẓahīra al-ḍāḥiyya al-munīra fī nasab wa-silsila ahl al-bayt al-nabawī,
2nd ed., edited by Muḥammad Ḍiyāʾ Shihāb, Jiddah (ʿĀlam al-Maʿrifa), 1984, 492–493;
Interviews with Bin Sumayt Khitamy, Muscat 1999 and 2001–2002, Muhdar Khitamy,
Mombasa April 2010 and Ahmad and Aydaroos Badawi, Lamu, July 2010 and December 2011.
The versions of the story are somewhat different. In some versions the birds simply land on
the body, then fly off, in which case their presence can be interpreted as an homage to the
deceased. In the version where the birds fly off with the body, the narrative problematizes
the question of homeland, expressed through burial site.
2 E. Ho, The Graves of Tarim, 140–141.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_��3

the luminescent sun and brilliant rays of light 21

SOMALIA N
0 500 km
Brawa

KENYA

Lamu
Archipelago

Mombasa

Zanzibar

Dar-es-Salaam

TANZANIA

Grande
Comore
Comoro Diego
Islands Suarez
MOZAMBIQUE
Ilha de
Mayotte
Mozambique

map 1 Map of East Africa

had family connections all around the western Indian Ocean. ʿAbd Allāh’s
grandfather, Hārūn b. ʿAbd al-Raḥmān Āl Jamal al-Layl (known as Mwenye
Hasan or Mwenye Ba Hasan) had left the Ḥaḍramawt in the late sixteenth
century and settled in Pate. Hārūn did not migrate alone, but brought with him
his son Aḥmad, later to become the father of ʿAbd Allāh. The “Master of Birds”
was, in other words, the first of his lineage to be born in East Africa (prob-
ably to an East African mother), while his father and grandfather were both
Ḥaḍramī born.3

3 B.G. Martin, “Migrations from Hadramawt to East Africa and Indonesia, c. 1200–1900”,
Research Bulletin, Centre for Arabic Documentation, Ibadan, 7, III, 1971, 1–21; B.G. Martin,
“Arab Migrations to East Africa in Medieval Times,” Int. Journal of African Historical Studies,
22 chapter 2

Thus, if we were to follow the birds as they flew off with ʿAbd Allāh, the jour-
ney could end either in Pate (where he would join his father and grandfather in
their final resting places) or in Tarīm, the Ḥaḍramawt, where his would be one
of the many graves of the ʿAlawī sāda.
On his final journey, ʿAbd Allāh was in fact not far from his adopted home-
land of Grande Comore. Leaving behind his father and two sons, ʿAbd Allāh
had left Pate accompanied by his youngest son Aḥmad some time in the lat-
ter half of the seventeenth century, thus replicating the journey of his father
and grandfather from Ḥaḍramawt. He settled at Ntsujini, Grande Comore,
where Aḥmad was to continue the Jamal al-Layl lineage. On this background,
the birds that flew off with the Ṣāḥib al-Ṭuyūr might also have taken him to
Grande Comore, and to the village of Ntsujini which was to emerge as a centre
for the Jamal al-Layl lineage on the island.
Finally, as his death occurred en route to Madagascar, one could also envi-
sion the birds flying south, to what was to become the southern extension of
the Jamal al-Layl family presence in the Indian Ocean, and where many of his
family were to settle in the centuries to follow.
Ultimately, an oceanic grave seems the most fitting of all for the “Master
of Birds”. Following E. Ho and reading the many headstones mapping ʿAlawī
Indian Ocean migrations as “silent markers”, a burial at sea seems the loudest –
yet completely inaudible – sign symbolizing the expansion of the sāda clans in
East Africa. The location, too, is fitting, as it illustrates the onwards migration
of families such as the Jamal al-Layl to the Comoro Islands and onwards.

Towards a Geography of Reform: A Web of Centres

The immediate family history of Ṣāhib al-Ṭuyūr illustrates a pattern of migra-

tion starting in the seventeenth century that was to expand the Jamal al-Layl
family- and religious networks southwards from the Swahili coast, via the
Comoro Islands to Madagascar and Mozambique. By the nineteenth century,
reformist impulses were to be transmitted along the same routes, generating a
geography of reform that was both multi-local and interconnected.
While, as stated in Chapter 1, simultaneity is a big problem for words, dis-
tance, on the other hand, is nothing. A keystroke can transport the reader
across the sea, from the big city to the most remote village, from one social,
political and religious context to another. It is worth, therefore, before we

VII, 3, 1975, 367–390. As Martin points out, the dating of these migrations are notoriously
unreliable, as they are projected retrospectively from family lineages.
the luminescent sun and brilliant rays of light 23

c­ ontinue to outline simultaneous events in time, to establish an outline of the

geography of Islamic intellectual exchange in the southwestern Indian Ocean
in the late nineteenth and early twentieth century. The aim is to arrive at what
Freitag and van Oppen called “the spatial categories of the actors themselves”,4
i.e. a geography that would have been navigable by the actors in this story and
which identifies as a series of centres that stood in specific, and changing, rela-
tions to each other. In other words, what developed and changed over time
was a multi-focal network of “centres of gravity” whose specific relationship to
each other was determined by historical patterns.

The View from Sea Level: Simultaneous Sunrises

In his book, A Hundred Horizons,5 S. Bose traces the cultural and histori-
cal interconnectedness of the Indian Ocean from 1850 until decolonization.
Instead of A Hundred Horizons, this chapter will navigate a set of simultaneous
sunrises across the southwestern Indian Ocean. It will do so by imagining a set
of beacons, beaming light across the dark sea, while at the same time drawing
travellers to shore.
The imagery of light (and implicitly, the sun) is a central one in Islamic
literature in general, where light is connected to knowledge, religion and –
ultimately – access to Paradise. Quranic references to light are manifold, and
refer to God, the believers and the religion of Islam. Furthermore, on the Day
of Judgment, a light will guide the righteous on their path to eternity, and the
community of believers are described as people of light.
Light is also a core concept in Sufism, especially the notion of the
Muḥammadan light (nūr Muḥammadī). This is a cosmological understand-
ing, much of it derived from Ibn al-ʿArabī but extrapolated upon by genera-
tions of Sufis, of the primordial creation of the light (nūr), passed on through
history until it became manifest in the Prophet.6 Renowned Sufis and saints
(sing: walī, pl.: awliyāʾ, “friends of God”) are also often portrayed as possess-
ing a “light”, which in turn is connected to the ability of the walī to perform
miracles (karamāt).7 Even the self-perception of the walī could be expressed

4 U. Freitag and A. van Oppen, Translocality. See above, Chapter 1.

5 S. Bose, A Hundred Horizons. The Indian Ocean in the age of Global Empire, Harvard (Harvard
University Press), 2009.
6 See A. Knysh, Ibn ʿArabi in the later Islamic Tradition. The Making of a Polemical Image in
Medieval Islam, New York (SUNY Series in Islam), 1999, V. Hoffmann, “Annihilation in the
Messenger of God” and E. Ho, The Graves of Tarim.
7 See for example the hagiographies in: J. Renard (ed.), Tales of God’s Friends. Islamic
Hagiography in Translation, Berkely (University of California Press, 2009).
24 chapter 2

in a vocabulary of light. For example, the Ḥaḍramī ʿAlawī saint and poet ʿAbd
Allāh b. ʿAlawī al-Ḥaddād (to be discussed in detail in Chapter 7) is reported
to have said: “We are the sun for the people, those who open their doors will
receive as much of it as they allow their doors to let in.”8 The two best-known
mawlid texts recited on the East African coast (Mawlid al-Barzanjī and Mawlid
al-Ḥibshī)9 both highlight the creation of the Muḥammadan light leading up
to the birth of the Prophet.
When it comes to nineteenth and twentieth century Islamic reformism in
general, the imagery of light is also encountered repeatedly. Perhaps the best
known is the journal al-Manār (The lighthouse) published by Rashīd Riḍāʾ and
which even featured an image of a lighthouse in its logo, spreading “rays of
light” to the Muslim world. As will be demonstrated in the following chapters,
the writings of reformist-oriented Sufis also frequently employed the imagery
of light to denote the “purification” of Islamic practice from older (or locally
generated) accretions. The light here represents the missionary ideal, formu-
lated as the call (daʿwa) to bring knowledge to “dark corners”. As new practices
were implemented, scholarship grew and saintly figures lived, died and were
buried, so places too came to be imbued with light – described in flourishing
terms, and with suitable use of luminescent imagery – by hagiographers and
authors of riḥlāt (travel accounts).

The View from the Shore: Hierarchies of Purity and

Horizontal Relationships
As outlined in Chapter 1, Sufi networks are here not only viewed as lateral enti-
ties, but should also be understood as locally constituted hierarchies, while
at the same time being embedded in larger structures such as cultural zones
(for example the Swahili) or political bodies (such as the British or French
empires). Discussing the relation between anthropological fields and the
perceived “home”, A. Gupta and J. Ferguson evokes the notion of a hierarchy
of purity,10 whereby the most isolated place, from the point of view of classi-
cal anthropology, will be considered the most pure. For the Muslim scholar-
traveller in the southwestern Indian Ocean, the situation was almost the exact
reverse, as the most isolated place would be considered impure, wild (Swahili:
shenzi), uncivilized lands. A case in point is the portrayal of Madagascar
in classical Arabic sources. The Arab view was that settling in Madagascar

8 M. al-Badawi, Sufi Sage of Arabia. Imām ʿAbdallah ibn ʿAlawi al-Haddad, Louisville (Fons
Vitae), 2005, 72.
9 See below, Chapter 6.
10 A. Gupta and J. Ferguson, Anthropological locations, 13.
the luminescent sun and brilliant rays of light 25

(but similar arguments could be made for other locations) was somehow
bordering on being “lost”, or assimilated into an un-Islamic culture and thus
becoming mutawaḥḥish (lit: savage, barbarian), the Arabic equivalent of “going
native”.11
However, when focusing on the mid/late nineteenth century, it is becom-
ing increasingly clear that the migrants to Madagascar (whether they came
straight from Arabia, via East Africa or the Comoros) were closely tied in with
the wider-ranging Indian Ocean networks, both in terms of genealogy, trade,
and organized religious practice. The hierarchy of impurity, thus, was increas-
ingly becoming “flattened”, as the Indian Ocean was reaching high tide of com-
mercial activity by the time of the colonial take-over. The “flattening” of the
landscape, however, does not necessarily imply that locally constituted “hubs”
were not hierarchical or that new hierarchies did not form.
In Chapter 1, Islamic reformist impulses were represented as a series of rip-
ples travelling the Southwestern Indian Ocean in the period c. 1880–1940. If
we are to be consistent to the imagery, we must acknowledge that ripples does
not appear out of nothing, but from a “first movement”, be that a small stone
dropped in the ocean or a massive subsea earthquake. In either case, we would
have an epicentre, with ripple forming as concentric circles and washing up
on near and far shores. The question is, in other words: Where did Islamic
reformist impulses come from in this period? Earlier research has pointed
out various locales (Cairo, Beirut, Istanbul, India, Mecca, Muscat, Mzaab
and the Ḥaḍramawt, but also regional centres such as Lamu and Zanzibar),
time frames (eighteenth century, nineteenth century), formal institutional
bases and regional or local shaykh-student relations. Others have pointed to
increased travel, the emergence of printed media in the Muslim world and the
role of individual, influential scholars.
Rather than engaging with this discussion, the aim here is to point not to
any one epicentre but to analyse a series of centres and establish an outline
of the relationship between them. Below will be introduced five major “suns”,
or locations from which different (and sometimes colliding) rays of light were
thought to be emanating; the Ḥaramayn, the Ḥaḍramawt, Lamu, Zanzibar and
the Comoro Islands.12

11 See R. Kent, Early Kingdoms of Madagascar, 1500–1700, New York, 1970. A similar outlook
on Arab settlements in Madagascar is found in G. Ferrand, Les Musulmans a Madagascar
et aux Iles Comores. Vol 3: Antankarana, Sakalava, Migration Arabes, Paris, 1902.
12 Of course, Lamu, Zanzibar and the Comoros were also centres for transmission to the
African mainland and onwards into the interior. This trajectory, although still very much
in need of research, is beyond the scope of this study.
26 chapter 2

The Ḥaramayn: The Blessed and the Radiant

As in the rest of the Islamic world, Muslims on the East African coast viewed
Mecca as the undisputed religious centre, a place of pilgrimage, but also a cen-
tre of learning, i.e. a source of light unrivalled by any other. Among those most
actively travelling in the search of knowledge were the Sufis, seeking enlight-
enment or their true spiritual guide. For them, Mecca came to be of immense
importance; to the point where the city has been described as the “strategic
centre” of Sufism.13 K.S. Vikør, in his study of the Sanūsiyya order, called Mecca
“the centre of the peripheries”,14 a description that, as we shall see, fits well also
with regard to the southwestern Indian Ocean. However, it is important to note
that in the Indian Ocean, a journey to Mecca was not for everyone, especially
not before the emergence of regular steam ship passage in the 1860s–1870s.
Those who went to study in Mecca were usually men of some standing in their
own communities, either because they had the means to do so, or because they
had achieved a relatively high level of Islamic education.
Towards the end of the nineteenth century, the number of travellers
increased from Indian Ocean locations in south-east Asia to Mecca. The sheer
number of people making the journey increased radically, although absolute
figures still remained low compared to the population at large. For example, a
mere 2,000 pilgrims left Dutch East India in 1850. By the 1880s, the figure had
tripled to an average of 6,000. In the first years of the twentieth century, the
figure more than doubled again, to an average of 15,000. This increase is all the
more significant when we consider that the Dutch colonial authorities actively
limited the number of pilgrims – fearful of whatever anti-colonial ideas travel-
lers might pick up in the Holy Land.
For the Ḥijāz in 1870s and 80s, our most comprehensive source is that of
the Dutch orientalist Christian Snouck Hurgronje who stayed in Mecca in the
early 1880s, posing as a Muslim.15 Despite his evident sympathies, it should be
noted that Snouck was writing during a time when Pan-Islamism was making
significant headway into Dutch East India. He is thus likely to project tenden-
cies that were not necessarily there, and see pan-Islamism in every Javanese
awakened to the global nature of Islam. Snouck is probably more correct in
his observation that the Javanese visitors to Mecca come back imbued with a

13 M. Sedgwick, Saints and Sons. The making and remaking of the Rashīdī Sufi order, 1799–
2000, Leiden (Brill), 2005.
14 K.S. Vikør, Sufi and Scholar on the Desert Edge. Muḥammad b. ʿAlī al-Sanūṣī and his
Brotherhood, London, 1995.
15 C. Snouck Hurgronje, Mecca in the latter half of the nineteenth century, Leiden (Brill), 1931.
the luminescent sun and brilliant rays of light 27

sense of Islam as a faith which transcended local stratification, customs and

traditions, and with the sense that Islam stretched back in time to known and
recognised authorities. Despite his concerns, Snouck made few attempts to
investigate how (or even whether or not) this enlarged sense of Islam actually
was perceived locally and what form it took when appropriated by villagers
throughout the archipelago.
The history of the Ḥijāz in the nineteenth century is one of political upheaval
and endless rivalries, but also of deep intellectual re-orientation and the con-
vergence of ideas in the great madrasas of the two Holy Cities. The century
began with the Wahhābī invasion in 1803. The restoration of the Ḥaramayn
to Sharīfian-Ottoman sovereignty in 1814 did not necessarily mean an end to
political strife in the city. Rivalry between the Sharīf and the Ottoman gover-
nor continued throughout the century. Nevertheless, practicalities were well
established in Mecca. Each major Sufi order would have their zāwiyas. These
were normally large houses that functioned as a combination of boarding
house, canteen and spiritual centre. Snouck Hurgronje points specifically to
the Sanūsiyya, Qādiriyya and Shādhiliyya, ṭarīqas that all maintained grand
houses in Mecca.16
It would, however, be incorrect to view Mecca as a place filled with Arab
scholars imparting their Arab knowledge to Swahilis, Malays and other non-
Arabs. As travel became more widespread, it became a place where scholars
of all languages and ethnicities converged and debated, using the scholarly lin-
gua franca, Arabic. In total, both Mecca and Medina were truly cosmopolitan,
to an extent which led Snouck Hurgronje to note that: “Mecca is partly a town
of foreigners” and a “many-tongued mass of humanity”.17

The “Scholars of the Age”

In Mecca and Medina were also the “great scholars of the age”, most notably
the theologians and jurists of the different law-schools. These would teach in
the Ḥaram mosque and in the Nabawī (Prophet’s) mosque in Medina accord-
ing to fixed schedules.
To the Shāfiʿīs of East Africa, by far the most influential scholar of the
period was Aḥmad Zaynī Daḥlān (1817–1886),18 who was Shāfiʿī mufti from 1871.

16 C. Snouck Hurgronje, Mecca, 205–206.

17 C. Snouck Hurgronje, Mecca, 208.
18 On him, see J. Schacht, “Sayyid Aḥmad Zaynī Daḥlān”, Encyclopedia of Islam; 2,
H. Sharkey, “Aḥmad Zaynī Daḥlān’s al-Futuḥāt al-Islāmiyya: A Contemporary view of the
Sudanese Mahdī”, Sudanic Africa, 5, 1994, 67–75; U. Freitag, “Der Orientalist und der Mufti:
Kulturkontakt im Mekka des 19. Jahrhunderts”, Die Welt des Islams, 43, 1, 2003, 37–60.
28 chapter 2

Daḥlān’s connection with the Indian Ocean world was close, and his impact
on East African Sufi practices and Islamic scholarship was to be long-standing.
Firstly, he himself studied with a number of Ḥaḍramī ʿAlawīs, many of whom
had family branches in East Africa as well as in the wider Indian Ocean. Then,
he became a teacher for new generations of ʿulamāʾ from Indian Ocean lands.
Daḥlān’s theological outlook was very much in line with reformist Sufism. In
his treatise against the Wahhābīs, Daḥlān clearly views Sufism as a legal and
integral part of Islamic practice, including such aspects as the visitation of
tombs.19 From Daḥlān’s perspective, these practices fulfil rather than trans-
gress the Sharia. He views grave-visitation or the recitation of dhikr as devo-
tional acts, rather than ones with magical overtones. At the same time, Daḥlān
also accepted the need for ijtihād (reinterpretation) and clearly claimed the
right to reinterpret the revelation.
This view was shared by Daḥlān’s “second-in-command”, Muḥammad Saʿīd
Bāb Ṣayl or Bā Buṣayl (d. 1912), who was of Ḥaḍramī origin. Like his mentor, Bāb
Ṣayl wrote a treatise in defence of Sufi practices.20 He was also a well-known
teacher. Amongst others he is reported to have held special sessions for women
on Friday mornings.21 A third important teacher was Abū Bakr b. Muḥammad
Shaṭṭā (d. 1893), known in Mecca as Sayyid Bakrī.22 The Shaṭṭā family was orig-
inally from Damietta in Egypt, but came to be very influential in Mecca. In
their lifetimes, these three men were important teachers for a generation of
East African scholars. As we shall see in Chapter 8, their connection with East
Africa was not only intellectual and it did not end with their deaths.

Returning from the Ḥaramayn

Before the full implementation of Dutch, British or French colonial administra-
tion, Indian Ocean returnees with extended periods in Mecca were employed
in some kind of official capacity, usually that of qāḍī in the service of a ruler.
However, recruitment to official positions increasingly came to be regulated
by colonial powers as most of the Indian Ocean came under colonial rule. This
meant that the Meccan-educated scholars often had to adapt to political real-
ity, or seek other careers.

19 Daḥlān, Aḥmad Zaynī, Al-Durr al-Saniyya fī ’l-radd ʿalā al-Wahhābiyya, 4th ed., Cairo,
1980.
20 Bāb Ṣayl, Muḥammad Saʿīd, Al-Qawl al-majdī fī ’l-radd ʿalā ʿAbd Allāh b. ʿAbd al-Raḥmān
al-Sindī, litograph, Batavia (Jakarta), 1309/1891–92.
21 C. Snouck Hurgronje, Mecca, 200.
22 C. Snouck Hurgronje, Mecca, 184.
the luminescent sun and brilliant rays of light 29

Evidence suggest that a large proportion of the Meccan-educated segment

engaged in teaching in one form or another, regardless of their affiliation with
political power. Often, their teaching activities also included propagation of a
Sufi order or a specific approach to Sufism irrespective of order. Teaching could
also be in the form of writing, that is anything from didactic pamphlets in the
vernacular, to massive learned volumes in Arabic.
A central question is what impact the teachings of scholars like Aḥmad Zaynī
Daḥlān had on local communities in east- and southeast Africa? Robert Hefner
has described the process by which “local” Islam, as played out at the “village
square” becomes linked to the wider diversity of Islamic religious expressions.23
In other words: local religion may – at some point in time – be “caught up” in
the wider social and political currents. This is the chain of events described
in the following chapters; the ways in which local understandings became
caught up with the wide scope of possible interpretations of Islam, as medi-
ated through a network of scholars who themselves were part of this wider
scope. The scale was enlarged, and Mecca and Medina were shining beacons
at the centre of this enlargement.

The Ḥaḍramawt: Home of the Luminescent, Encompassing

Mid-Day Sun

The “blessed homeland” to the ʿAlawī sāda, and a perceived centre of learning
from the nineteenth century until the present, several authors have evoked
the imagery of light about the Ḥaḍramawt. This imagery is even more specifi-
cally used about its religious capital, Tarim. The most widely quoted ʿAlawī
genealogy, for example, employs the image of the luminescent, encompassing
mid-day sun from which life-giving warmth and light radiates.24 Another, later
example, refers to Tarim as a centre from which “rays of light” spread to the far
corners of the Indian Ocean.25
Much of the imagery of light is derived from the specific understanding of
the ṭarīqa ʿAlawiyya as perpetuated through the bloodline of the Prophet. As
has been discussed in detail in earlier research, the ʿAlawī ṭarīqa originated in

23 R.W. Hefner, “The Political Economy of Islamic Conversion in Modern East Java”, in:
William R. Roff (ed.), Islam and the Political Economy of Meaning, London/Los Angeles,
1987, 53–78.
24 ʿAbd al-Raḥmān Al-Mashhūr, Shams al-ẓahīra, 294–295.
25 Abū Bakr al-ʿAdanī b. ʿAlī b. Abī Bakr al-Mashhūr, Lawāmiʿ al-nūr. Nubdha min aʿlām
Ḥaḍramawt, 2 vols., Ṣanʿāʾ (Dār al-Muhājir), 1412/1991–92.
30 chapter 2

the Ḥaḍramawt in the Yemen and with the descendants of the Prophet, through
one Aḥmad b. ʿĪsā al-Muhājir (the emigrant) who arrived from Iraq around
950 AD. This group has since formed a special stratum in the Ḥaḍramawt. Over
the centuries, the sāda have kept unity and cohesion primarily based on their
understanding of their own bloodline combined with Sufi spirituality. ʿAlawī
Sufism as it came to be formulated around the 15th century rests on the claim
that both their bloodline and their transmission chain of spiritual methods
and secrets, goes back to the Prophet. This dual emphasis on both spiritual
and physical descent made for a tight-knit stratum with a very strong tendency
towards intermarriage.26
At the same time, the sāda of the Ḥaḍramawt, like the rest of the Ḥaḍramī
population, had a strong tendency towards migration, mostly as traders or reli-
gious experts or both. To read a family tree of a Ḥaḍramī ʿAlawī lineage is like
reading a map of the Indian Ocean. From about the 13th century one will find
the notes on each individual: died in Java, died in Pate, died in Lamu, died
in Calicut etc., thus forming what E. Ho has called a “tissue of names across
the ocean”.27
By the nineteenth century, young Alawī son in the Indian Ocean diaspora
tended to be brought up with the idea of the ancestral homeland as a place
where spiritual boons were to be found, an idea of “origin” expressed both
spiritually and in terms of lineage. The homeland was “poor but pure”. The
diaspora, on the other hand, was rich but potentially corrupting and the only
remedy was to go “home” to Ḥaḍramawt and Arabia for periods of learning.

The ʿAlawī Reform Case

Upon arrival in the ancestral land, returnees in the latter half of the nineteenth
century would meet a Sufi order in reform. This reform can be said to have
started in the eighteenth century, with scholars such as ʿAbd Allāh b. ʿAlawī
al-Ḥaddād (1634–1719) and his student, Aḥmad b. Zayn al-Ḥibshī (1658–1733).28
The shift was not so much in doctrine as in action. An increasing number of
sāda took on the roles as teachers, some even offering Islamic teaching to
women, like the scholar Aḥmad b. ʿUmar b. Sumayṭ (1769–1842). They wrote
devotional texts meant for all and engaged in politics to promote Islamic rule in
matters of state, as expressed in the treatise al-Daʿwa al-tāmma by al-Ḥaddād

26 On the marriage patterns of Ḥaḍramīs in diaspora, see F. Le Guennec-Coppens, “Social

and Cultural Integration: A Case Study of the East African Hadramis”, Africa, 59 (2), 1989,
185–195.
27 E. Ho, The Graves of Tarim, 321.
28 On the works and impact of al-Ḥaddād and his students, see below, Chapter 7.
the luminescent sun and brilliant rays of light 31

mentioned in Chapter 1. The ʿAlawī ṭarīqa remained a Sufi order, but its ori-
entation turned towards the world beyond its genealogical demarcation lines.
The missionary emphasis was clear, focusing mostly on “inner mission”, i.e.
teaching Islam to people who were already Muslims. Aḥmad b. Zayn al-Ḥibshī,
for example, wrote a short treatise called al-Risāla al-Jāmiʿa, a basic instruction
on how to live a Muslim life through prayer, fasting, zakāt (Ar: almsgiving) etc.
The booklet was used to teach Bedouin youth in improvised lessons when they
came to town for trading. The word daʿwa is used explicitly to denote educa-
tional efforts in the writings of ʿAbd Allāh b. ʿAlawī al-Ḥaddād, Aḥmad b. ʿUmar
b. Sumayṭ and Aḥmad b. Zayn al-Ḥibshī.
In the nineteenth century, al-Risāla al-Jāmiʿa was copied by hand and –
later – reprinted repeatedly, precisely for educational purposes in other parts
of the Indian Ocean where Ḥaḍramī ʿAlawīs lived, and where the pupils were
not bedu but young Malays, Swahilis or “Cape Malays”. A Malay translation
was first printed in Batavia in 1875 and has since been reprinted numerous
times. An Arabic/Swahili/English translation was printed in Zanzibar in 1925
by Aḥmad b. Abī Bakr b. Sumayṭ.29 In fact, tracing the translations of al-Risāla
al-Jāmiʿa gives an indication of the extent of the daʿwa-oriented ʿAlawī educa-
tional efforts, but also of the ways in which their teachings had become local-
ized before the arrival of printing presses.
The full consolidation of the ʿAlawī reform process came in the nineteenth
century, when ʿAlawī scholars in Ḥaḍramawt established schools with hous-
ing for students (ribāṭ pl. arbiṭāʾ) designed precisely to teach Islam in accor-
dance with Sufi tenets. These schools offered teaching in a structured and
organized manner. The new system differed from the classical tradition by
which a student sought out a teacher and sat with him at his “circle” for an
indefinite period. The most famous was al-Riyāḍ mosque college in Sayʾūn in
the Ḥaḍramawt, established in 1878 by ʿAlī b. Muḥammad al-Ḥibshī (d. 1915).
He was to become a teacher to a whole generation of East African scholars.
Also important was the ribāṭ founded in Tarim by ʿAbd al-Raḥmān al-Mashhūr,
the author of the ʿAlawī genealogy entitled Shams al-Ẓahīra.30 He, too, was to
become a much-sought authority by East African ʿulamāʾ, especially ʿAlawīs.
The emerging daʿwa orientation of the ʿAlawīs also had a strong Meccan
connection. From the mid-nineteenth century, many ʿAlawī scholars from
the Ḥaḍramawt spent prolonged periods of study in Mecca, mostly with each
other but also with other teachers. By the late 1800s, there was a community

29 See below, Chapter 6.

30 U. Freitag, “Hadhramaut: a religious centre for the Indian Ocean in the late nineteenth
and early twentieth centuries?”, Studia Islamica, 89, 1999, 165–183.
32 chapter 2

of established ʿAlawī teachers in Mecca, closely connected with the ongoing

tendency towards reform and religious activism. Muḥammad Saʿīd Bāb Ṣayl
has already been mentioned. Another important figure was ʿUmar b. Abī Bakr
Bā Junayd (1270–1334/1853–54 – 1915–16).31 He was a student of Daḥlān and Bāb
Ṣayl, and was to become an important node for East African and – as we shall
see – South African scholars in Mecca.
Finally, the ʿAlawīs of Ḥaḍramawt also developed the habit of sending out
emissaries to the lands of diaspora. One of them was ʿAlawī b. ʿAbd al-Raḥmān
al-Mashhūr who visited Zanzibar in 1911–1912.32

Zanzibar: The Brilliant Star of East Africa

The reformist activities of Shāfiʿīs and Ibāḍīs in Zanzibar have been the topic
of several studies in the past decade and will not be recounted at length
here. Rather, the purpose of this section is to point to the impact of both the
Ḥaramayn and the Ḥaḍramawt on Islamic scholarship in Zanzibar. East African
scholars of two successive generations spent significant periods of study in
the Ḥaḍramawt and the Ḥaramayn. Moreover, they studied with a handful of
teachers whose influence on East African and Indian Ocean Islamic tenets was
to remain strong, well into the twentieth century.
The first generation of Shāfiʿī scholars33 emerging under Bū Saʿīdī aus-
pices in East Africa included such men as Muḥyī al-Dīn al-Qaḥṭānī (ca. 1790–
1869), ʿAbd al-ʿAzīz al-Amawī (1838–1896), ʿAlī b. ʿAbd Allāh b. Nāfiʿ al-Mazrūʿī
(1825–1894) and ʿAlī b. Khamīs al-Barwānī (1852–1885). This generation trav-
elled widely, and studied with numerous scholars, although only al-Mazrūʿī
is known for certain to have spent time in Mecca with Aḥmad Zaynī Daḥlān.
One of Al-Qaḥṭānī’s students, the Comorian-born scholar Muḥammad b.
Aḥmad b. Ḥasan al-Murūni/al-Moroni (d. 1890)34 spent a period of study in

31 Al-Mashhūr, Lawāmiʿ al-Nūr I, 277–278.

32 On the life and works of ʿAlawī b. ʿAbd al-Raḥmān al-Mashhūr, see al-Mashhūr, Lawāmiʿ
al-Nūr, which essentially is a biography on him. He was a teacher in the ribāṭ in Tarim.
33 The biographies of most of the scholars described here can be gained from A.S. Farsy,
Baadhi ya Wanavyoni wa Kishafi wa Mashariki ya Afrika/The Shafiʿi Ulama of East Africa,
ca. 1830–1970, Translated, edited and annotated by R.L. Pouwels, University of Wisconsin,
African Primary Text Series, III, 1989. See also al-Mashhūr, Lawāmiʿ al-nūr, and al-Mashhūr,
Shams al-ẓahīra; R. Loimeier, Between Social Skills.
34 See Farsy/Pouwels, The Shafiʿi ʿUlama, 78–80 and passim; MKELLE/ZNZ-2, 50–52.
Muḥammad b. Aḥmad al-Murūnī (al-Moroni in the Farsy hagiography) was a qāḍī dur-
ing the reign of Sayyid Barghash, and departed for Mecca without the permission of the
the luminescent sun and brilliant rays of light 33

Mecca during the reign of Sayyid Barghash (1870–1888). The history of his
period in Mecca was recounted some 40 years later by his fellow Comorian-
Zanzibari Burhān Mkelle. The narrative offers a glimpse of the system of teach-
ing, but first and foremost it demonstrates the importance attached to this
group of teachers as late as the 1930s:

He (al-Murūnī) stayed in the Bayt Allāh al-Ḥarām (Mecca) where he

encountered all the sāda and ʿulamāʾ there as well as other learned
people. The Shāfiʿī muftī at that time was Sayyid Abū Bakr Muḥammad
Shaṭṭā. It was the habit of Muḥammad al-Murūnī when attending Sayyid
Abū Bakr’s darsa in the Ḥaram al-Sharīf, to sit at the edge of the majlis
( jalasa haythu yantahā al-majlis). If Abū Bakr Shaṭṭā saw him (al-Murūnī)
among the students, he would motion him to sit nearer. Eventually, he
would be asked to arrive before the others, and to sit (with Shaṭṭā) in
front of the people.35

The Ḥaramayn and the Ḥaḍramawt become even more noticeably influential
in the second generation. This cohort of men came into influential positions
in the early colonial period.36 Here we find, most notably, the later Chief Sunni
Qāḍī of Zanzibar, Aḥmad b. Abī Bakr b. Sumayṭ (1861–1925). Although he did not
study directly with Aḥmad Zaynī Daḥlān (who had died by the time Ibn Sumayṭ
arrived), he spent time with Bāb Ṣayl and Shaṭṭā. Ibn Sumayṭ’s disciple and
close companion in Zanzibar, ʿAbd Allāh Bā Kathīr (1860–1925) made the same
journey in 1897, seeking out the teachers of his master and returning to write
a travel account of his experiences.37 Bā Kathīr’s erstwhile mentor in Lamu,
Sayyid Manṣab b. ʿAbd al-Raḥmān (1829–1922), spent two periods in Mecca,
both times studying with Daḥlān, Bāb Ṣayl and Shaṭṭā. In the early twentieth
century, Ibn Sumayṭ’s son ʿUmar followed suit, and spent a period in Mecca,
amongst others with Daḥlān’s son.38 All these men came to be extremely influ-
ential in the East African scholarly community, and in their writings they make

Sulṭān. Al-Murūnī was also a central teacher to ʿAbd Allāh Bā Kathīr, which makes him a
representative of the “first generation”.
35 MKELLE/ZNZ-2, 52.
36 For a discussion of the intellectual formation of this generation, see A.K. Bang, Sufis and
Scholars.
37 Bā Kathīr al-Kindī, ʿAbd Allāh b. Muhammad, Riḥlat al-Ashwāq al-Qawiyya ilā Mawāṭin
al-Sāda al-ʿAlawiyya, Reprint, Cairo, 1405/1984 (first ed. Cairo, 1934).
38 Sumayṭ, ʿUmar b. Aḥmad, Al-Nafaḥāt al-Shadhdhiyya min al-Diyār al-Ḥaḍramiyya wa-
talbiyyat al-ṣawt min al-Ḥijāz wa-Ḥaḍramawt, Privately printed, Tarim/Aden, 1955.
34 chapter 2

numerous references to Daḥlān, Bāb Ṣayl and Shaṭṭā. Within the framework of
the ʿAlawī ṭarīqa, Sumayṭ, Bā Kathīr and Sayyid Manṣab were instrumental in
propagating a Sharia-based Sufism, and focal points for this instruction were
teaching institutions, like the Madrasa Bā Kathīr, founded in Zanzibar by ʿAbd
Allāh Bā Kathīr.
As has been described in earlier research, two other Sufi orders entered
Zanzibar in the late nineteenth century: The Qādiriyya and the Shādhiliyya.
The Shādhiliyya will be discussed below (under the section about the Comoro
Islands); suffice it here to say that by 1900, it had also made substantial head-
way in Zanzibar.
Zanzibar was one of the centres for the diffusion of the Qādiriyya on the
East African coast. It was popularized39 in Zanzibar in the mid-1880s by
Shaykh Uways b. Muḥammad al-Barāwī (1847–1909).40 Born in Brawa, he
journeyed to Baghdad where he studied the Islamic sciences and was initi-
ated into the Qādiriyya by its principal shaykh, Sayyid Muṣṭafā b. Sulaymān
al-Jīlānī. Upon his return to East Africa, he travelled widely, including frequent
visits to Zanzibar, reportedly upon the invitation of Sayyid Barghash. There,
the Qādiriyya spread through the khalīfas appointed by Shaykh Uways, most
notably his fellow Brawanese ʿAbd al-ʿAzīz al-Amawī (1838–1896)41 and Sayyid
ʿUmar Qullatayn (d. 1926).42

39 Although there are some indications that the Qādiriyya already had been introduced in
Zanzibar by the time it was propagated by Shaykh Uways, he was certainly the one who
made it an established order. An earlier propagator may have been one Ḥusayn b. ʿAbd
Allāh al-Murnī al-Qādirī, who was reportedly also a Rifāʿī. The major impact of Shaykh
Uways, however, illustrates the importance of local networks for proliferation. Interview,
Maalim Muhammad Idris, 22.07.2006.
40 On the history of Shaykh Uways, see among others, C.C. Ahmed, God, Anti-Colonialism
and Dance. Sheykh Uways and the Uwaysiyya, in: G. Maddox (ed.), Conquest and Resistance
to Colonialism in Africa. New York (Garland Publishing), 1993, 145–167; S. Reese, “The death
of Shaykh Uways of Somalia”, in: J. Renard (ed.), Tales of God’s Friends. Islamic Hagiography
in translation, Berkely (Univ. California Press), 2009. See also the hagiographic account of
his life given by ʿAbd al-Raḥmān b. ʿUmar, al-Jawhar al-Nafīs fī khawāṣṣ al-Shaykh Uways,
Cairo (Maṭbaʿat al-Mashhad al-Ḥusaynī), 1964.
41 See V. Hoffman, “In his (Arab) Majesty’s Service: The Career of a Somali Scholar and
Diplomat in Nineteenth-Century Zanzibar”, in R. Loimeier and R. Seesemann (eds.), The
Global Worlds, 251–273; R. Loimeier, Between Social Skills, 75–87.
42 His full name was ʿUmar b. Qullatayn al-Naẓīrī Bā ʿAlawī; Copy of the ijāza given to ʿUmar
Qullatayn (ZANZIBAR/QADIRI1). He died on 17 Rabīʿ I 1345/25 September 1926; Burhān
Mkelle, BURHANIYAT, Elegy on the death of ʿUmar Qullatayn, 60–62.
the luminescent sun and brilliant rays of light 35

The network was close, cemented by marriage. ʿAbd al-ʿAzīz al-Amawī was
married to Shaykh Uways’ daughter Dédé43 and a close confidante of both
Sayyid Khalīfa and Sayyid Barghash. The importance of Shaykh Uways as a
religious teacher to the leading Zanzibari scholars of Brawanese origin, can be
gained from the entries in the diary of ʿAbd al-ʿAzīz al-Amawī. He recorded the
comings-and-goings of Shaykh Uways, and how they co-incided with his own
travels.44
Al-Amawī has been described by Abdallah Saleh Farsy as a “reformist”, or
even something of a proto-modernist in his denunciation of the “worship of
coughing”, or loud dhikr.45 This portrayal may have been a projection of Farsy’s
own modernist views and his scepticism towards Sufi practices and should
not be construed to understand al-Amawī as an early modernist. As a Qādirī
shaykh (who even founded his own sub-branch, the Nūraniyya), al-Amawī was
not opposed to dhikr per se, nor to the practices associated with Sufism. As
V. Hoffmann has pointed out based on al-Amawī’s own writing, he was neither
adverse to entering trance, nor prescribing it to others.46 The reformist aspect
of his career lies in his writing of educational material, and in his role as a
teacher for a generation of daʿwa-oriented scholars.
ʿUmar Qullatayn, too, was to become an important teacher to a generation
of Qādirī scholars. In his house in Mkunazini, he held dhikr sessions every
Thursday.47
As will be shown in Chapter 3, the Qādiriyya quickly developed several
branches in Zanzibar, with different networks that in turn made for different
extensions beyond the island.

Lamu and the Riyadha Mosque

In Lamu, Islamic practice and intellectual traditions in the late nineteenth

century became closely identified with the Riyadha mosque, founded by Ṣāliḥ
b. ʿAlawī Jamal al-Layl, better known in East Africa as Habib Saleh (1853–1936).
He was a descendant of the “Master of Birds” whose death was depicted at the

43 V. Hoffman, “In his (Arab) Majesty’s Service”, 264.

44 Al-Amawī, QADHI DIARY, photocopy, unpaginated, several references. For example,
al-Amawī records on 15 Rabīʿ II 1312/15th October 1894 that Uways left Zanzibar for Brawa,
and then, on 7 Rajab 1312/3rd January 1895 notes the return of Shaykh Uways.
45 Farsy/Pouwels, The Shafiʿi ʿUlama, 44–48.
46 V. Hoffmann, “In his (Arab) Majesty’s Service”.
47 Interview, Ahmad b. Abi Bakr Qullatayn, 22.07.2006.
36 chapter 2

outset of this chapter, and yet another example of the back-and-forth migra-
tion pattern of his family. He left his native Grande Comore for Lamu, then
returned to Grande Comore, and finally settled in Lamu some time in the late
1870s or early 1880s. His biography is well-known from earlier studies, as is his
religious and intellectual formation.48 The spiritual connection to his Sufi mas-
ter in Ḥaḍramawt, the aforementioned ʿAlī b. Muḥammad al-Ḥibshī is espe-
cially emphasized in the family history of the Riyadha, and its teaching was
explicitly modelled on that of al-Riyāḍ in Sayʾūn, the Ḥaḍramawt.49 It is also
clear from some of the manuscripts in the Riyadha Mosque that the two cor-
responded and exchanged books.
The Riyadha Mosque as it stands today came into being through the assis-
tance of Sayyid Manṣab b. ʿAbd al-Raḥmān (1828–1922),50 whom we encoun-
tered above as one of the scholars who spent time in Mecca with Daḥlān, Bāb
Ṣayl and Shaṭṭā. He transformed some of his land in Lamu into a waqf (pious
endowment) for the purpose of building the Riyadha.51 However, there is every
reason to believe that Habib Saleh started his teaching activities in Langoni
significantly earlier, probably shortly after he settled in Lamu between 1875
and 1885.52
The main hallmark of the Riyadha, from the very beginning, was the incor-
poration of ritual traditions derived from ʿAlī b. Muḥammad al-Ḥibshī (nota-
bly the Simṭ al-Durar, also known as the Mawlid al-Ḥibshī). However, the most
enduring reformist agenda of the Riyadha was inclusion of people (Oromo,

48 Ṣāliḥ Muḥammad ʿAlī Badawī, Al-Riyāḍ bayna māḍīhi wa-ḥādirihi, Transcript, NP,
1410/1989; A.H.M. el-Zein, The Sacred Meadows. A Structural Analysis of Religious
Symbolism in an East African Town, Evanston, 1974; P. Lienhardt, “The Mosque College of
Lamu and its Social Background”, Tanzania Notes and Records, 1959; BinSumeit Kitamy,
“The role of the Riyadha mosque college in enhancing Islamic identity in Kenya”, in:
M. Bakari and S.S. Yahya (eds.), Islam in Kenya, Nairobi (MEWA), 1995, 269–276; P. Romero,
Lamu. History, Society and Family in an East African Port City, Princeton, 1997; A.K. Bang,
Sufis and Scholars.
49 Although it unclear if Habib Saleh and ʿAlī al-Ḥibshī ever met in the physical world, sev-
eral stories are recounted about their encounters in the spiritual realm. In one example,
Habib Saleh was “with ʿAlī al-Ḥibshī in Sayʾūn” when a fire broke out. Later, when Habib
Saleh came to perform his prayers and hold lectures in the Riyadha in Lamu, his clothes
were covered in ash and he smelled of smoke and fire. Interview, Aydaroos Badawi, Lamu,
December 2011.
50 Farsy/Pouwels, The Shafiʿi Ulama, 66–68 and passim.
51 Waqfiyya dated 1320/1903 (both years are actually given in the waqfiyya), and stamped by
the East Africa Protectorate Lamu Registry, 21 Feb. 1903: In the possession of the Riyadha
Mosque.
52 This is also the conclusion arrived at by P. Lienhardt, “The Mosque College”, 228–242, 230.
the luminescent sun and brilliant rays of light 37

Pokomo and others) who were considered “outsiders” by the traditional Lamu
aristocracy.
The Riyadha has since imparted Islamic education to children and higher
learning (Islamic law, Quranic exegesis, Sufism etc.) to advanced students. The
Riyadha, in the course of its history, and with its “branches” in other parts of
the region, has come to epitomize East African Islam. In the latter half of the
twentieth century, the Riyadha developed into a bastion of conservatism. For
example, they opposed Abdallah Saleh Farsy’s translation of the Quran into
Swahili.53 However, as will be discussed in Chapter 6, it also developed into
a bastion of tradition, in the most physical sense, as a keeper of texts (manu-
scripts and printed books) read by scholars in the past.

The Comoro Islands: Moon Islands in a Sea of Sun

Many studies of the Comoro Islands open with a statement of how small the
islands are, creating an impression that these four islands in the Mozambique
Channel are notable first and foremost for being tiny. As pointed out in an
opening section by A. Bourde in 1965, the total territory of the archipelago
is a quarter the size of Corsica and at the time of writing, the population on
neither island exceeded 100,000 souls with a total of about 200,000.54 In 1988,
A.S. Bakar gave another comparison in his introduction, noting that the
Comoros in total are in fact smaller than Reunion and only slightly bigger than
Mauritius.55 However, as also pointed out by many authors, including Bourde
and Bakar, the islands have seen persistent migration both in and out, as well
as a long-standing overseas trade, and they are consequently tied to neigh-
bourly coastal regions through family- and trade networks.
As noted by several authors, and demonstrated most poignantly in the
account given by Abdallah Saleh Farsy, Comorian trade – and family connec-
tions in the period immediately before and in the decades after French colo-
nization were solid in the northerly direction, towards Zanzibar and Lamu.
However, as this book aims to demonstrate, these links were also strong, and
growing increasingly stronger in the southern direction, towards Madagascar
and Mozambique. In this sense, the Comoro Islands, and ­particularly Grande

53 K. Kresse, Philosophising, 117.

54 A. Bourde, “The Comoro Islands: Problems of a Microcosm”, Journal of Modern African
Studies, 3,1, 1965, 91–102.
55 Abdourahim Saïd Bakar, “Small Island Systems: A case study of the Comoro Islands”,
Comparative Education, 24,2, 1988, 181–191.
38 chapter 2

Comore and Anjouan, should be viewed as centres – transit hubs in the middle
of a string of ports – rather than as peripheries.56

Comorian Migration Patterns and Indian Ocean Networks

Some time in the late 1920s or 30s, Burhān Mkelle wrote about his forefather’s
native Ngazija (Grande Comore):

The people of Injazīja are known as travellers, both in the past and pres-
ent, to such an extent that they can be found in most parts of Africa and
even in the cities of Asia.
Many travelled (hājirūn) to Madagascar, because of its proximity to
their island. Their number there (in Madagascar) is 30,000 souls, and this
(number) has caused some who don’t know to say that their number in
Madagascar is higher than in Grande Comore. This is untrue. [. . .]
In Madagascar, Comorians engaged in trade, while others joined the
army or the civil service. Many of them joined the army to assist the
French in their war against the Hova57 and Sakalava tribes. In those days,
the Comorians showed great bravery which pleased the French leaders.
That was because they fought not only the enemies of the French, but
also their own enemies. It was revenge on behalf of their ancestors, for
the attack which the Malagasy tribes (qabāʾil) carried out in 1790, without
any declaration of war. [. . .]
The second country to which the Comorians emigrated was Zanzibar.
Their number there and in neighbouring countries was 10,000 according
to some census.
The third country of emigration was Delagoa Bay (Maputo Bay), and
from there they proliferated to the Cape. Their exact number is not fully
known, as most of them settled in the city of Delagoa Bay (Maputo). This
is a Portuguese colony located in the East African coast, and the only
means of communication was by rail to Transvaal, whose capital is
Pretoria.58

56 This in line with the view suggested by Alpers, that the history of the islands is integral
to several aspects of East African history E.A. Alpers, “Indian Ocean Africa: The Island
Factor”, in: E.A. Alpers, East Africa and the Indian Ocean, Princeton, 2009, 39–54.
57 Note that Mkelle (MKELLE/ZNZ-2, 16) writes “Hova” in transliteration, but uses the term
qabāʾil Ambalāmbū in Arabic.
58 MKELLE/ZNZ-2, 16–17.
the luminescent sun and brilliant rays of light 39

What Burhān Mkelle emphasizes here is the notorious Comorian passion for
travel that caused people from the islands to form important and at times very
influential minorities in such established centres as Zanzibar and in locations
such as Diego Suarez, Mahjunga, Nosy Be and coastal Mozambique.
Burhān Mkelle is not the only writer or observer to have noted the diffu-
sion of Comorians in the Western Indian Ocean.59 Writing from the perspec-
tive of Zanzibar, Saʿīd al-Mughayrī, in the 1950s, devoted much space in his
book to the influx of Comorians to Zanzibar from the time of Sayyid Barghash
(r. 1870–1888) and onwards.60 His figures are at times highly unreliable
(amongst others, he states that 15,000 Comorians migrated to Zanzibar in 1899
alone, which is hard to believe), but his emphasis on this group as contributors
to the Zanzibari community shows the importance attached to the Comorian
presence over a long period of time. This is also emphasized by Ibuni Saleh in
A Short History of the Comorians in Zanzibar, which in laudatory term describes
the impact of Comorians on the British-Bū Saʿīdī state, aimed to underpin the
author’s argument that Comorians in Zanzibar be granted special treatment
under the British system, and not be classified as “natives”.61
In his account, Mkelle devotes much space to the “racial” background of
the Comorians, i.e. the incoming journeys that formed Comorian society as
part and parcel of an Indian Ocean network.62 In a terminology that is remi-
niscent of colonial racial discourse, Mkelle categorizes the ethnic background
of Comorians as consisting of south-east Asian/Malagasy, Arab and Israelite,
while the influx from Africa is given comparatively little attention. The same
pattern is noticeable in the history of Zanzibar by Saʿīd al-Mughayrī, who
devoted two pages to the connections between the Comoro Islands, South
Arabia and the Middle East:

The inhabitants of Grande Comore are a mixture of Arabs, Javanese,

Shirāzī and Africans. The Javanese came before anyone lived there.
They named the place (of their arrival) after their place of origin in
Java, and that name has remained until today. The name is Mṭumhūlī
(Mitsamilouli). After then came the Arabs of the Munādhira. After them

59 See for example M. Said, “Le Périple d’un Comorien de Zanzibar a Mascate et Retour,”
Études Océan Indien, 10, 1988, 103–122.
60 al-Mughayrī, Saʿīd b. ʿAlī, Juhaynat al-Akhbār fī ta⁠ʾrīkh Zinjibār, 4th ed., Ministry of
National Heritage and Culture, Oman, 1994, 424–427;
61 I. Saleh in A Short History of the Comorians in Zanzibar, Dar es Salaam (Tanganyika
Standard), 1936.
62 MKELLE/ZNZ-1, 27–34; MKELLE/ZNZ-2, 27–36.
40 chapter 2

came other Arab tribes of Oman. They were followed, year after year, by
Arabs from Yemen and the Ḥaḍramawt.
Concerning the Arabs of the Ḥaḍramawt, most of them are of the
Awlād Muḥammad b. Ṭāhir BāWazīr. There also came people from Persia
( fārs bi ’l-ʿajam), led by Muḥammad b. ʿĪsā al-Shahīr, and they were there
for many years, but then their migration ceased.
The Africans were brought by the Arabs and Javanese from the coast to
help in the agricultural and other work. This race (the Africans) mixed
with the other races until they were one race of one colour. Among the
Arabs who went to the Comoros and multiplied there were the people of
Ṣūr in Oman who came every year and then returned. Their offspring can
be found in the islands.63

Al-Mughayrī’s racial discourse is clearly skewed, with a strong emphasis on the

“Asian/Middle Eastern connections” and their role as “civilizers” (not to men-
tion his version of how Africans were brought “to help in agricultural and other
work”). That, said, it is interesting to note that al-Mughayrī singles out the
Ḥaḍramīs as being among the most influential arrivals, a point that has also
been supported by later research. As pointed out by I. Walker,64 the Ḥaḍramīs
who arrived in Grande Comore in the nineteenth century reached shores that
were beyond both Omani or British control, while not yet under French con-
trol. This meant that they could achieve higher positions. At the same time it
meant that status derived from being “Arab” was not automatically assigned by
a ruling power, and the newly arrived would have to employ various strategies
to enter Comorian society. One such strategy was marriage into the uxorilocal,
matrilineal clans, while retaining both patrilineality and claims to “Arabness”.65
Finally, as has been pointed out by S. Blanchy, matrilineal transmission was
linked to place and affiliation, while patrilineal transmission was linked to net-
work and overseas connections. As we shall see in the following chapters, this
made for conflicts when it came to succession, for example within Sufi orders
where rivalling networks would strive for control.
In the period before the French colonial expansion in the late the nine-
teenth century, immigrants from Ḥaḍramawt specifically came to play impor-
tant roles in the political and religious life of the islands. According to one

63 al-Mughayrī, Saʿīd b. ʿAlī, Juhaynat al-Akhbār, 526–527.

64 I. Walker, “From Moroni to Mukalla: Hadhramis on the island of Ngazidja (Comoros) and
in the Hadhramhout”, Journal des Africanistes, 72:2, 2002, 111–121.
65 I. Walker, “From Moroni to Mukalla”. See also F. le Guennec-Coppens, “Social and Cultural
Integration”.
the luminescent sun and brilliant rays of light 41

chronicler, the following sharīfian lineages were present in Grande Comore

by the early twentieth century: Āl Shaykh Abū Bakr b. Sālim (the author’s own
lineage), Āl ʿAydarūs, Āl Jamal al-Layl, Āl bin Sumayṭ, Āl Saqqāf, Āl Bal Faqīh,
Āl al-Ḥaddād, Āl Ismaʿīl and Āl al-Baytī.66
What can be stated on the basis of Arabic and Swahili chronicles, combined
with Ḥaḍramī ʿAlawī sources, is that the religious and political leadership of
mid- to late eighteenth century Comoro Islands to some extent was dominated
by families of Ḥaḍramī-ʿAlawī descent, mainly the Āl Jamal al-Layl and the Āl
Shaykh Abī Bakr b. Sālim. As we saw in the immediate family history of the
“Master of Birds”, the Jamal al-Layl lineage reached the Comoros some time in
the late seventeenth or early eighteenth century. They later came to be centred
on the village of Ntsujini on the west coast. Displaying a similar migration pat-
tern, the Āl Shaykh Abī Bakr b. Sālim family had been present on the Comoros
since the late seventeenth century, i.e. within two generations of the original
migrant to East Africa.67 Their foremost representative in the nineteenth cen-
tury was Mwinyi Mkuu Sultan Aḥmad (1792–1875) who was the paramount
sultan on Grande Comore from the 1830s. According to the ʿAlawī genealogical
“central office”68 in Singapore, he fathered 63 children, with the result that by
1933, one third of the population of Moroni were his descendants.69
One of Sultan Aḥmad’s daughters was married to the Sultan of Anjouan,
and she became the mother of the later sultan Sayyid ʿAlī. In the course of
the 1880s, Sayyid ʿAlī brought Grande Comore under his tentative control, by a
mixture of trickery, outright confrontations and French support. From 1890, he

66 S. Chouzour, “Histoire et Sociologie de Ngazidja. Le Manuscrit de Said Hussein.

Présentation, traduction et notes”, Études Océan Indien, 1:1982, 15–54, 38. The author of
the manuscript, Said Husayn (1889–1978) was the son of Sayyid ʿAlī (of the Āl Shaykh
Abū Bakr b. Sālim) who resigned Grande Comore to French rule (see below). He should
therefore be considered a reliable informant on this issue.
67 B.G. Martin, “Arab Migrations”, 381.
68 The Rābiṭa ʿAlawiyya was the ʿAlawī league in Singapore, founded in 1927. One of the main
activities of the league was to collect and publish information on the ʿAlawīs in diaspora.
See U. Freitag, Indian Ocean Migrants, 252 onwards.
69 Muḥammad b. ʿAbd al-Raḥman b. Shihāb, “Al-Sāda al-ʿAlawiyya fī jazāʾir al-Qamar wa-
Madaghaskar”, al-Nahḍa al-Ḥaḍramiyya, Singapore, 10/1933, 24–25. Bin Shihāb gives
Sulṭān Aḥmad’s full nisba as follows: Aḥmad b. Ṣāliḥ b. ʿAlī b. Ṣāliḥ b. Aḥmad b. al-Ḥusayn
b. Shaykh Abī Bakr b. Sālim. The career of Sultan Aḥmad is also the main topic of L.
Harries (ed.), The Swahili Chronicle of Ngazija by Said Bakari bin Sultani Ahmed, African
Studies Program, Bloomington, 1977, 124. Harries here bases his information on the docu-
ment by Rev. Charles Sacleux who originally transcribed the chronicle into Latin script in
the early twentieth century.
42 chapter 2

ruled with the assistance of a series of French Residents, with varying degree of
mutual sympathy. In 1909, he resigned Grande Comore entirely to French colo-
nial rule and in 1912 the entire archipelago was incorporated into the French
colonial system as a province of Madagascar.70 In French colonial literature
from the 1890s/1910s, Sayyid ʿAlī features as an “Arab Sultan”, complete with
Arab dress and calligraphic letterheads, as a leader who had inherited his posi-
tion from a long line of paternal forefathers.71
More interesting are Sayyid ʿAlī’s activities after his resignation from the
intricacies of Comorian power-politics. Around 1910, he embarked on a tour
of his family networks, spending time in Mozambique and Madagascar before
visiting Zanzibar in 1913. He then proceeded to the ancestral homeland of
Ḥaḍramawt where he, like the East African Islamic scholars described above,
sought out the aforementioned Sufi shaykh ʿAlī b. Muḥammad al-Ḥibshī.72

Muḥammad al-Maʿrūf: The Quṭb of the Shādhiliyya in East Africa73

Although the importance of Muḥammad al-Maʿrūf in the diffusion of the
Shādhiliyya in East Africa has been addressed in several previous studies, it
will be recounted here for the sake of tying the network he created in with
the history of the order in Mozambique and Madagascar. His life history also
demonstrates the importance of family networks in the diffusion of organized
Sufism southwards to Madagascar.

70 L. Harries (ed.), The Swahili Chronicle of Ngazija.

71 S. Blanchy, “Images Coloniale de la Société Comorienne. Les Raisons d’une méconnai-
sance durable”, in: Images coloniales de sociétés de l’Ocean indien, Presses Universitaires
Paris-Sorbonne, 2009. This misrepresentation, according to Blanchy, derives from the
reports and descriptions of the French residents on Grande Comore.
72 Al-Mashhūr, Shams al-Ẓahīra, 294–295.
73 On the life of Muḥammad al-Maʿrūf, see P. Guy and A. Cheih Amin, La Vie et l’Oeuvre du
Grand Marabout des Comores, Said Mohamed ben Ahmed Al-Ma’arouf, Tananarive, 1949.
See also G. Boulinier, “Le Rôle de Said Mohamed el-Maarouf dans le Developpement de
la Confrerie Shadhili”, Bulletin CENDERES (ND); C. Ahmed, “Networks of the Shādhiliyya
Yashruṭiyya Sufi Order in East Africa”, in: R. Loimeier and R. Seesemann, The Global Worlds
of the Swahili; J.-C. Penrad, “La Shâdhiliyya-Yashrûṭiyya en Afrique orientale et dans
l’océan Indien occidentale,” in: É. Geoffroy, Une voie soufie dans le monde. La Shâdhiliyya,
Paris, 2005, and B.G. Martin, Muslim Brotherhoods in nineteenth Century Africa, Cambridge
University Press, 1976, 152–158. See also MKELLE/ZNZ-2, 40–41.
the luminescent sun and brilliant rays of light 43

Muḥammad al-Maʿrūf was from a Ḥaḍramī ʿAlawī family; the Āl Shaykh Abī
Bakr bin Sālim.74 His mother, too, is reported to have been of the Āl Shaykh Abī
Bakr bin Sālim, a daughter of Sultan Aḥmad (Mwinyi Mkuu) in Moroni.
Muḥammad al-Maʿrūf first studied in his native Moroni. He was then sent
to Zanzibar for further studies, accompanied by his uncle Aḥmad, known as
Aḥmad al-Kabīr, whom we shall meet again in Chapter 4 as a Shādhilī daʿwa
scholar in Madagascar. According to the oral histories recounted by the
shayhks of the order in Madagascar, this was in 1876.75
Muḥammad al-Maʿrūf then proceeded to Mecca, although it is unclear
if he was still accompanied by his uncle. Upon his return from the Ḥijāz,
Muḥammad al-Maʿrūf met ʿAbd Allāh b. Saʿīd Darwīsh, a fellow Comorian
who had studied directly under ʿAlī al-Yashruṭī in Palestine, and who initiated
Muḥammad al-Maʿrūf into the Shādhliyya-Yashruṭiyya.76
The manāqib does not discuss why Muḥammad al-Maʿrūf chose to propa-
gate the Shādhiliyya-Yashruṭiyya rather than his own “family order”, the ṭarīqa
ʿAlawiyya which at this point in time has expanded beyond the genealogical
confines of the sāda. Given his family background, Muḥammad al-Maʿrūf
had undoubtedly been affiliated with the ʿAlawiyya, and he was certainly also
aware of the common links in the silsila of the two orders. Several authors here
indicate al-Maʿrūf’s encounter and close relationship to ʿAbd Allāh Darwīsh as
essential for his intellectual formation and orientation.77 Burhān Mkelle also
notes that the principles of the Shādhiliyya itself formed part of its appeal:
“[The reasons why it pleased us] was the dhikr and the devotedness to love

74 His manāqib gives the nisba of Muḥammad al-Maʿrūf to be: Muḥammad b. Aḥmad b. Abī
Bakr b. Aḥmad b. Abī Bakr b. ʿAbd Allāh b. Sālim b. Aḥmad b. ʿAbd Allāh b. ʿAlī b. Shaykh
Abī Bakr bin Sālim, P. Guy and A. Cheih Amin, La Vie et l’Oeuvre, 6.
75 Interviews, Antsiranana/Ambilobe, Madagascar, July 2008. Also supported by brief his-
tory (in Arabic) compiled by Sayyid Momad Aydarus of Moroni, in the possession of
Shaykh Abdu Samad, Ambilobe. P. Guy and A. Cheich Amin do not mention a year, but
1876 fits well with the general chronology.
76 The Yashruṭiyya branch of the Shādhiliyya was founded by the Tunisian ʿAlī al-Yashruṭī
(d. 1891 in Palestine). It became particularly influential in Syria, Jordan and Palestine. ʿAlī
al-Yashruṭī was a student of the mid-nineteenth century Shādhilī shaykh Muḥammad
b. Hamza al-Madanī who in turn was a student of al-ʿArabī al-Darqawī, the founder of
the Darqawiyya, a sub-branch of the Shādhiliyya. See B.G. Martin, Muslim Brotherhoods,
155–156 and Enclyclopedia of Islam: Shādhiliyya.
77 Notably J.-C. Penrad, “La Shâdhiliyya-Yashrûtiyya”. See also MKELLE/ZNZ-2, 41.
44 chapter 2

and solidarity (taḍāmuniyya) among the disciples.”78 B.G. Martin explained the
vast success of the Shādhiliyya-Yashruṭiyya in the Comoros as part of a broader
wave, “buoyed up by a wave of religious enthusiasm, of Islamic revivalism”.79
By all accounts, al-Maʿrūf was a very active propagator of his ṭarīqa. It spread
rapidly on all the Comoro Islands, sometimes to the irritation of local powers.
Eventually, al-Maʿrūf’s activism and his differences with the ruling houses led
to a warrant for his arrest, and he was forced to flee (the manāqib gives the year
1893 for his departure). He settled in Zanzibar, where he continued to propa-
gate the Shādhiliyya which became firmly established on the island. Some time
in 1903 or 1904, al-Maʿrūf was permitted by the French authorities to return to
the Comoros, and he died there after having established a zāwiya in Moroni.
The Shādhiliyya-Yashruṭiyya became so successful that by 1980 it was esti-
mated that 70% of the adult male population in the Comoro Islands had some
type of Shādhilī affiliation. Due to the non-exclusiveness of the Shādhiliyya,
many of its members were sāda who also maintained their ʿAlawī affilia-
tion. One prominent example is ʿUmar b. Sumayṭ, who, by all accounts, was
an accomplished shaykh of the ʿAlawiyya. He was nonetheless participant in
the Shādhiliyya dhikr-sessions in Moroni, and he even composed poetry to be
recited in for the ḥawliyas commemorating the life of al-Maʿrūf.

Islamic Education in Grande Comore

In 1903, the Palestinian writer Ruhi b. Muḥammad al-Khālidī (Ruhi Bey, 1864–
1913) published an account of the Comoro Island in al-Hilāl, most likely based
on contact with Comorians in France. Here, he had little praise for the knowl-
edge of Arabic and Islam among the Comorians. Expressing faith in the French
presence on the islands, he expressed: “. . . let them [the Comorians] be busied
with reading newspapers and magazines in Arabic, let the barbarisms in their
speech disappear, and let them grasp easily the Book of God and the Sunna of
His Prophet . . .”80
Ruhi Bey was plain wrong. The “Book of God and the Sunna of His Prophet”
was, by the 1880s widely known and embedded in a scholarly tradition in the
Comoros, as has been shown by A.M. Toibibou, who also emphasize the influ-
ence of Zanzibar in this period.81 In terms of legal traditions, the Comoro Islands

78 MKELLE/ZNZ-2, 41.
79 B.G. Martin, Muslim Brotherhoods, 155.
80 B.G. Martin, “A Palestinian Arab and Writer on the Comoros, 1903”, Études Océan Indien, 6,
1985, 71–124, 102.
81 T.A. Mohamed, Ahmad Qamardine (1895–1974). Un intellectuel Comorien et ses réseaux,
PhD Thesis, University of Paris Diderot, 2010, 53–76 and passim.
the luminescent sun and brilliant rays of light 45

relied on much the same corpus of texts as Zanzibar and other Swahili centres.
Al-Nawwawī’s Minhāj al-Ṭālibīn formed the basis for legal deliberations,82 and
copies of the Minhāj were imported from Zanzibar to the Comoros.83 Contrary
to the misgivings of Ruhi Bey, Comorian children had long been educated in
traditional madrasas known as shio/shioni (Swahili: chuo).84
One important figure in the introduction of systematic Islamic educa-
tion in Grande Comore was Abū ’l-Ḥasan b. Aḥmad Jamal al-Layl, known as
Mwinyi Bahasani (1801–1883). Like Habib Saleh in Lamu, he was a descendant
of the “Master of Birds” presented at the outset of this chapter.85 According to
Toibibou, he was vital in establishing madrasas and higher education facilities
in Grande Comore in the nineteenth century. He is also known as the main
teacher of two of East Africa’s best known religious scholars in the early twen-
tieth century, Aḥmad b. Sumayṭ and Habib Saleh.
No exact figures exist for exactly how many Quranic schools existed in
Grande Comore in the latter half of the nineteenth century, but according to
M. Pechmarty, there were no less than 247 primary Quranic schools in 1917,
attended by 4,738 pupils (2,880 boys and 1,858 girls) as well as 27 advanced
schools attended by 262 persons.86 Most likely, these were not all new estab-
lishments in the twentieth century, but it is not known how many dated back
to the 1880s or earlier.

Rays of Light and Hierarchies

The descriptions above have outlined the emergence of reformist ideas in five
locations that all can be considered scholarly centres and that were to func-
tion (in various ways) as intermediate points towards the locations further
south. The fact that I have ordered them in a certain sequence (the Ḥaramayn,

82 G. Ferrand, Les Musulmans a Madagascar et aux Iles Comores.

83 T.A. Mohamed, Ahmad Qamardine, 63.
84 E. de Brye, “L’enfant comorien et l’apprentissage de l’Islam. Quelques observations”,
Études Océan Indien, 6, 1985, 41–51.
85 T.A. Mohamed, Ahmad Qamardine, 53–54 and passim., A.K. Bang, Sufis and Scholars,
50–51, and al-Mashhūr, Shams al-ẓahīra, 485–498. This Abū ’l-Ḥasan is also known as Abū
’l-Ḥasan al-Kabīr, not to be confused with Abū ’l-Ḥasan b. Aḥmad Jamal al-Layl (1888–
1959), known as Abū ’l-Ḥasan al-Ṣaghīr (See below). Interview with Aydaroos and Ahmad
Jamal al-Layl, Lamu, 06.12.11.
86 The figures derive from the detailed report authored in 1916/17 by M. Pechmarty on Islamic
education on Grande Comore. P. Verin, ‘Le Rapport Pechmarty sur les “écoles koraniques”
à la Grande Comore (1917)’, Études Océan Indien, 16, 1993, 105–123.
46 chapter 2

the Ḥaḍramawt, Zanzibar, Lamu, Comoro Islands) should not be construed to

mean that I imply a “hierarchy of purity” as outlined by Gupta and Ferguson. In
fact, my choice has been mainly with a view to narrative; to the sequence that
best orders a long list of names and connections.
Beyond the central position of the Ḥaramayn to all Muslims, the under-
standing of hierarchy would of course depend on point of view; A Zanzibari
scholar would view this quite differently from, for example, Abū ’l-Ḥasan b.
Aḥmad Jamal al-Layl in Grande Comore. Similarly, a person with extensive
family networks to the Ḥaḍramawt would view the region as a “blessed land”
whereas one with little or no connection would have a different outlook. All,
however, have in common that they were viewed as religious and scholarly
centres by surrounding peoples – the only difference being how far their “radi-
ance” reached.87
These centres were societies with specific social, political and cultural tra-
ditions, as well as contemporary concerns of their own. These shaped pat-
terns of translocal interaction, including transmission of religious ideas. We
can assume that this was particularly so when it came to reform, understood
as agendas for change. What the agenda should be, and who should have the
power to interpret and implement it, was continuously influenced by the inter-
ests of local elites and commoners.
The most striking feature was the fact that these locations were inter-linked
by a pattern of migration that had been in place since at least the seventeenth
century. By the mid-1800s, they were also increasingly linked through a net-
work of scholarly exchange.
Finally, it is worth pointing out what these locations did not have in com-
mon. Firstly, while East Africa was increasingly coming under European influ-
ence, and by the 1890s incorporated into British or French colonial rule, this
was not the case in the Ḥaramayn or Ḥaḍramawt. While the British had gov-
erned Aden since 1837, the Ḥaḍramawt remained beyond British control. The
Ḥaramayn, as we have seen, had different power-struggles going on, mainly
involving the Sharīf and the Ottoman governor. These differences should be
kept in mind as we move towards the first southern expansion of the Sufi
orders: Mozambique.

87 It should be noted that the emergence of of Shādhiliyya and Qādiriyya also brought Accra
and Baghdad into focus as centres of learning.
 chapter 3

The Branches of the Qādiriyya and the Shādhiliyya

in Northern Mozambique
Silsilas to the South

I instructed that he open the doors of the

zāwiya for the service of the poor . . .1

∵
Contrary to the East African towns of Zanzibar and Lamu which only came
under European control at the end of the nineteenth century, Mozambique
had long experience with European rule by the time of the arrival of organized
Sufism. This was especially so in the former Portuguese naval base of Ilha de
Mozambique, which had been under Portuguese control since the early 1500s.
The diffusion of the Qādiriyya and Shādhiliyya to the south was, in other words,
not only a transmission beyond the immediate Swahili cultural zone, but also
into another political structure with a different set of hierarchies.
Much has been written about the history of the Qādiriyya in East Africa and
the role the order played in the so-called “Meccan letter affair” in German-
held Tanganyika in 1908,2 but also its role as a social movement throughout
East Africa. As S. Reese has pointed out, by the 1920s, the order “easily cut
across most major social, cultural and economic boundaries.”3 Focusing on the
Benadir coast, Reese demonstrates how the Qādirī shaykhs aimed their teach-
ing not only at their murīds, but at the general population. He also demon-
strates how the shaykhs increasingly emphasized daily praxis (observance of

1 Quote: Qādirī initiation document from Palma, Mozambique. See below.

2 See J.S. Trimingham, Islam in East Africa, London (Oxford University Press), 1964;
B.G. Martin, “Muslim Politics and Resistance to Colonial Rule: Shaykh Uways b. Muhammad
al-Barawi and the Qadiriya Brotherhood in East Africa”, The Journal of African History, 10: 3,
1969, 471–486; B.G. Martin, Muslim Brotherhoods, 152–177; A.H. Nimtz, Islam and Politics in
East Africa; R.L. Pouwels, Horn and Crescent: Cultural Change and Traditional Islam on the
East African Coast 800–1900, New York (Cambridge University Press), 1987; A.A. Issa, “The
legacy of Qadiri scholars in Zanzibar”, in: R. Loimeier and R. Seesemann (eds.), The Global
Worlds of the Swahili; R. Loimeier, Between Social Skills, 66–95 and passim.
3 S. Reese, Renewers of the Age. Holy Men and Social Discourse in Colonial Benaadir, Leiden
(Brill), 2008, 214.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_��4

48 chapter 3

the precepts of the law) as the way to obtain paradise, while de-emphasizing
the importance of mystical exercises to obtain fanāʾ (“annihilation in God”).
More recent research has also deepened our knowledge about the role of
this order in northern Mozambique, notably E.A. Alpers4 and L.J.K. Bonate.5
Especially the work by Bonate has expanded the history of Sufi orders in north-
ern Mozambique as well as the complex patterns of integration into local com-
munities. Bonate’s research also echoes the findings of Reese at the northern
end of the Swahili coast; an emphasis on normative faith, preached by teachers
who clearly saw themselves as “on a mission”. As late as in 1972, the Portuguese
A.P. Carvalho noted that the Mozambican Qādiriyya was “under the tutelage
of Zanzibar”.6
This chapter will trace some of the routes of the Qādiriyya and Shādhiliyya
from Zanzibar to Mozambique, and identify networks that, in different ways,
took southwards the reformist impulses embedded in the ṭarīqas. The pur-
pose is thus to contextualize and nuance earlier research, to point to paral-
lel paths of knowledge transmission, and indicate corresponding processes of
localization.

The Ṭarīqa Qādiriyya in Zanzibar

As outlined above (Chapter 2), the Qādiriyya in Zanzibar quickly acquired

followers from different backgrounds in Zanzibar after the arrival of Shaykh
Uways in 1884. By 1900 it was a well-established and expanding ṭarīqa, which
recruited from different segments of the population, in turn generating differ-
ent networks.

The “Brawanese” Qādiriyya

The first network is that which we can call “Brawanese”, propagated by a seg-
ment that was Arabic-literate educated and urban. This was a class of schol-
ars with a clear “bookish” orientation, consisting of men who were scholars
in their own right, as well as religious leaders. An indication of this orienta-
tion can be drawn from the document whereby ʿUmar Qullatayn was initi-
ated into the Qādiriyya in Zanzibar, some time in the late nineteenth/early

4 E.A. Alpers, “A Complex Relationship”.

5 L.J.K. Bonate, “Roots of diversity”, and L.J.K. Bonate, Traditions and Transitions.
6 A.P. de Carvalho, “Confraries Muculmanes nativas da Cidade de Mocambique”, Voz Africana,
August 29, 1972; Here quoted from L.J.K. Bonate, Traditions and Transitions, 80.
the branches of the qādiriyya and the shādhiliyya 49

twentieth century. This text is in highly literate Arabic, referring to textual

sources beyond the local context, and with a formalized hierarchy of power
that claims authority from the Qādirī centre in Baghdad. This type of silsilsa,
in other words, establishes authority as something that comes with written,
authenticated words in Arabic, and with an ultimate authority situated beyond
the local context.

This is the silsila from its origin (shajara aṣlihā) and its branches. Its car-
rier is a noble man. [. . .]
I authorized the servant of the fuqarāʾ, the Sayyid ʿUmar b. al-Sayyid
Qullatayn Sayyid Maẓhar al-Naẓīrī al-Bā ʿAlawī in the knowledge of my
shaykh Sayyid and our master, ʿAbd al-Qādir al-Jīlānī, may God hallow his
secret.
In the name of God, most merciful, most compassionate. This is to
confirm, in knowing and doing (ʿilman wa-ʿamalan) to all our Muslim
brothers that the bearer of this letter is one of the wandering seekers
(al-darāwīsh al-sālikīn) who came to Zanzibar and entered the assembly
of the pole of knowers (quṭb al-ʿārifīn), the guide of seekers, my shaykh
ʿAbd al-Qādir al-Jīlānī, may God hallow his secret [. . .]
It may be that you achieve entry into the group affiliated with this
ḥaḍrat al-Sunniya, by honouring him and showing generosity towards
him and protecting him from encroachments, as did the Ayat al-Sharīf
[here used as a reference to the Prophet]. He (SAWS) said: “The highest
reward will bestowed on him who share with a stranger from his posses-
sions, or eases his distress by a drink of water or food or clothing or a
smile on his face, for him is the garden of Paradise”.7

ZANZIBAR/QADIRI 1: “Brawanese Branch”

Silsila back to ʿAbd al-Qādir al-Jīlānī and the Prophet

Sayyid Muṣṭafā b. Sayyid Sulaymān al-Jīlānī (Baghdad)
Uways b. Muhammad al-Barāwī
ʿUmar b. Qullatayn al-Naẓīrī

Figure 1 The Zanzibar Qādiriyya “Brawanese Branch”

7 ZANZIBAR/QADIRI 1. Silsila given to ʿUmar Qullatayn.

50 chapter 3

As outlined in Chapter 1, this urban, educated branch of the Qādiriyya was

associated in Zanzibar first and foremost with Muḥyī al-Dīn al-Qaḥṭānī, ʿAbd
al-ʿAzīz al-Amawī and Sayyid ʿUmar Qullatayn, to whom the above ijāza was
issued. Like Shaykh Uways, all three men originated from the Brawa coast. All
were, as we have seen in Chapter 2, also part of the wider network of Zanzibari
ʿulamāʾ and held various official positions in the Bū Saʿidī Sultanate. Shaykh
Uways, too, had close connection with the Bū Saʿīdīs, and appointed both Sayyid
Barghash and Sayyid Khalīfa as his khalīfas. It should also be noted that this
group of scholars also had family and/or intellectual links with the Ḥaḍramawt
and Mecca. The Qullatayn family, for example, claim descent from the Naẓīrī
branch of the ʿAlawī sayyids, and family history narrates that ʿUmar’s father
came to Brawa from Tarim. The family is reported to have its own mosque in
Tarim, and contact has been kept up in the twentieth century via return visits
to the city.8 This was the segment of the Zanzibari Qādiriyya that tended to be
most critical of “ecstatic” rituals, which, as R. Loimeier has pointed out, the
Qādiriyya itself had introduced to beyond the learned classes.9
The Qādirī shaykhs held teaching sessions in that capacity, teaching the
Qādirī rituals. Teaching seems initially to have been confined to adults, and
conducted in private homes. Qullatayn family history remembers teaching
in the home of ʿUmar Qullatayn in the Mkunazini quarter of Stone Town,
and dhikr sessions held every Thursday. Until the house collapsed some time
around 2000, the house also served as the Qādirī zāwiya.10 However, as the
order spread, teaching was also taking place in mosques as the order diffused
to the countryside and the mainland.

The “African” Qādiriyya

A second, and very important network from the point of view of the southward
expansion of the order was the initiation of what A. Issa has called the “African
scholars” of the Qādiriyya.11 Among the foremost scholars in this network was
ʿAbd Allāh Mjānakhayr, known in Zanzibar as Shaykh Mjana Kheri. He was
born ca. 1870 in Zanzibar to parents of mainland origin.12 In the late 1800s, he
introduced the Qādiriyya among plantations slaves in Zanzibar. Around the
turn of the century, he moved to Ngambo, where he was influential in s­ preading

8 Interview, Ahmad b. Abi Bakr b. Umar Badawi Qullatayn, Zanzibar, 21.07.2006.

9 R. Loimeier, Between Social Skills, 76–85 and passim.
10 Interview, Ahmad b. Abi Bakr b. Umar Qullatayn, grandson of Umar Qullatayn, Zanzibar,
21.07.2006.
11 A.A. Issa, “The legacy”, 349. See also A.H. Nimtz, Islam and Politics, 58.
12 Loimeier, Between Social Skills, 76–78.
the branches of the qādiriyya and the shādhiliyya 51

ZANZIBAR/QADIRI 2: “Mjana Kheri Branch”

Silsila back to ʿAbd al-Qādir al-Jīlānī and the Prophet

Sayyid Muṣṭafā b. Sayyid Sulaymān al-Jīlānī (Baghdad)
Uways b. Muhammad al-Barāwī
Shaykh Mjana Kheri
Ḥasan b. Ameir and several khalīfas following in Zanzibar and beyond

Figure 2 Zanzibar Qādiriyya, “Mjana Kheri Branch”

the Qādiriyya, and established a zāwiya there. Through the activities of Shaykh
Mjana Kheri, the Qādiriyya also spread to the mainland, and (according to
A.H. Nimtz’s account) also southwards. A. Issa has stated that “. . . many people
regarded Shaykh Mjanakheri as the successor of Shaykh Uways in the region,
as he had managed to attract many followers to the Qādiriyya in Tanganyika,
Nyasaland (present-day Malawi) and even Congo by the middle of the twen-
tieth century.”13 It is important to note that although fewer written sources
exist from this network, it is clear that their activities were similar to those of
the other branches (and indeed, the other ṭarīqas). A central component was
education; as pointed out by Loimeier, Shaykh Mjana Kheri, his wife and his
students all opened darsas and zāwiyas for educational purposes.

The “Comorian” Qādiriyya

A third, and less researched network was that of Comorian scholars (or schol-
ars of Comorian origin) propagating the Qādiriyya. As the Comorian learned
class has been associated more en bloc with the Shādhiliyya, less research has
been conducted on the role of Comorians in the proliferation of the Qādiriyya.
A central node in this segment was Muḥammad “Mkelle” b. Adam, the father of
Burhān Mkelle.14 In his History of Grande Comore, Burhān Mkelle includes his

13 A.A. Issa, “The legacy”, 351. Issa here bases her statement on A.H. Nimtz, Islam and
Politics, 57.
14 This identification is based primarily on the silsila text, which indicates that Mkelle is
from Ingazija, combined with the biography given by Burhān. It is also based on the fact
that this silsila was found in the collection of Maalim Muhammad Idris, where much of
the material derived from the archives of the Comorian Association. Both Burhān and his
sons were central figures in this association, and are likely to have deposited some of their
material there. As has been discussed elsewhere, the same collection also contains much
of the private books of Burhān Mkelle. See A.K. Bang, “Authority and Piety, Writing and
Print: A preliminary Study of the Circulation of Islamic Texts in late nineteenth and early
52 chapter 3

father among the Comorians who joined the service of the Bū Saʿīdīs, and also
gives his biography. Mkelle, whose full name was Muḥammad b. Adam, was
born in Zanzibar in Rabīʿ I 1272/ November 1855.15 Burhān also gives the origin
of the nickname Mkelle:

He was known by this name because his father used to travel for trade
between Grande Comore, Zanzibar and Mukalla and during one of his
trips he left for Mukalla on the third day after the birth of his son, so the
people in his house jokingly named the son Mkelle.16

Mkelle studied the Quran, and later the Islamic sciences with Shaykh
Muḥammad Mlomrī (d. 1897).17 He then took up occupation as a tailor, and
soon entered the service of Sayyid Barghash. Then, some time after 1880, i.e.
when he was between 30 and 40 years of age, he was appointed khalīfa of the
Qādiriyya:

In the name of God, most merciful and most compassionate. This is to

confirm to all our Muslim brothers, in knowledge and deed (ʿilman
wa-ʿamalan) that the carrier of this letter, Mkelle b. Adam b. Nūḥ al-Shirāzī
al-Qādirī, is (one) of the wandering seekers (al-darāwīsh al-sālikīn). He
came to us, and he entered the assembly of the pole of the knowers (quṭb
al-ʿarifīn), the guide of seekers, my shaykh ʿAbd al-Qādir al-Jīlānī, may
God hallow his secret [etc.]. It may be that you achieve entry the group
affiliated with this ḥaḍrat al-Sunniya, by honouring him and showing
generosity towards him and protecting him from encroachments, as did
the Ayat al-Sharīf. He (SAWS) said: “The highest reward will bestowed on
him who share with a stranger from his possessions, or eases his distress
by a drink of water or food or clothing or a smile on his face, for him is the
garden of Paradise”.
He has true intentions and I have chosen him as my khalīfa that he
may increase it. [. . .]
This is the silsila from its origin (shajara aṣlihā) and its branches. Its
carrier is a noble man. [. . .] I am the poor in God’s sight and the servant of

twentieth Century Zanzibar”, Africa, 81:1, 2011, 89–107. For the biography of the Mkelle
father-and-son, see also A.K. Bang, “When there are no foreign lands and all lands are for-
eign”, in: N. Naguib and B. de Vries (eds.), Heureux qui comme Ulysses a fait un beau voyage.
Movements of People in Time and Space, Bergen (Bric), 2010, 151–165.
15 MKELLE/ZNZ-2, 24.
16 MKELLE/ZNZ-2, 24.
17 On him, see R. Loimeier, Between Social Skills, 583.
the branches of the qādiriyya and the shādhiliyya 53

the fuqarāʾ, Shaykh al-Ḥājj Uways b. Muḥammad al-Qādirī, servant of my

shaykh ʿAbd al-Qādir al-Jīlānī, may God hallow his secret.18

The silsila then gives an ijāza for the prayers and texts in which Mkelle b. Adam
has received instruction, including a series of specified Qādirī awrād and their
silsilas. The document also gives the bearer ijāza in texts that are not exclu-
sively Qādirī and which has a wider resonance in East Africa (and beyond),
such as Mawlid al-Barzanjī 19 and Qaṣīdat al-Burda.20 Finally, the text moves
on to the long silsila of the Qādiriyya, back via ʿAbd al-Qādir al-Jīlānī to the
Prophet and from him back to creation. The link from Baghdad via Shaykh
Uways is universal to all the main branches of the East African Qādiriyya:

ZANZIBAR/QADIRI 2: Comorian Branch

Silsila back to ʿAbd al-Qādir al-Jīlānī and the Prophet

Sayyid Muṣṭafā b. Sayyid Sulaymān al-Jīlānī (Baghdad)
Uways b. Muhammad al-Barāwī
Mkelle b. Adam b. Nūḥ al-Shirāzī.

Figure 3 Zanzibar Qādiriyya, “Comorian Branch”

The role of Mkelle in the Qādiriyya was clearly that of a khalīfa, and his con-
tinuous contact with Comorians in Zanzibar made for a diffusion of the order
to Grande Comore.21 What is also clear is that this network partly overlapped
with the first one, and shared many of its characteristics, including the lean-
ing towards “bookishness” and an emphasis on literary (written) learning and
education, as most clearly exemplified by Mkelle’s son Burhān.

18 ZANZIBAR/QADIRI 2. Qādirī silsila, scroll, 77 lines, undated, signed by Shaykh Uways. The
scroll is badly torn, and parts of the text impossible to read.
19 On the position of the Mawlid al-Barzanjī in East African Muslim society, see
C. Ahmed, Ngoma et Mission Islamique (Da⁠ʾwa) aux Comores et en Afrique Orientale, Paris
(L’Harmattan), 2002, 13 onwards. On its diffusion in East Africa, see below, Chapter 6.
20 Kawākib al-Durriya fī madḥ khayr al-bāriyya, a poem of praise for the Prophet, known as
Qāṣīdat al-Burda, by Sharāf al-Dīn Muḥammad al-Busīrī (for its diffusion in East Africa,
see below, Chapter 6).
21 It also explains the role of Burhān Mkelle as adept of the Qādiriyya in Zanzibar in the
twentieth century.
54 chapter 3

The Qādiriyya in Mozambique: Multiple Routes South

The history of the Islamic sultanates of northern Mozambique has been out-
lined in earlier research, and will not be recounted here.22 Suffice it to say
that Islamization took place in a pattern similar to that further north, with
Muslim sultanates being established on the coast. After the arrival of the
Portuguese, the Sultanate of Angoche came to hold a dominant position, as
the Portuguese gradually took control of Ilha de Mozambique, Quelimane and
other Shirazi settlements.23 The Angoche sultanate went into decline follow-
ing the Portuguese takeover of the Zambezi in the mid-1500s, but experienced
a revival in the nineteenth century.24
By the nineteenth century, there had emerged a coastal culture, sharing
many of the traits of the Swahili to the north – notably Shāfiʿī Sunni Islamic
religious practice.25 As has been emphasized by L.J.K. Bonate, inhabitants
of the coast also held close kinship ties with, and shared many cultural traits
with the peoples of the interior.26 The Muslim leaders in the region maintained
relations with the Sultans of the Comoro Islands, in matters of trade in general
and slave trade in particular.27
Nineteenth century Muslim leaders also maintained religious links, some-
times overlapping with family links, and sometimes with an explicit daʿwa
orientation. The life of Musa Muhammad Sahib Quanto (d. 1879) is instruc-
tive in this regard. He was the maternal brother of the Sultan of Angoche, and
on the latter’s behalf he carried out a military operation that expanded the
power of the Sultanate along the coast and into the interior. Before embark-
ing on his conquests in the 1850s, Musa studied Islam in Zanzibar, and he also
travelled to Madagascar and the Comoro Island.28 Hafkin maintains that Musa
accompanied “a relative” on a proselytizing mission in Mozambique, and that
the entire military expansion was (at least in part) motivated by furthering

22 N.J. Hafkin, Trade, Society and Politics in Northern Mozambique c. 1753–1913, PhD
Dissertation, Boston University, 1973, 1–30 and passim. By north here is meant north of
the Rovuma river. See also M. Newitt, A History of Mozambique, London (Hurst), 1995.
23 L.J.K. Bonate, “Matriliny, Islam and Gender in Northern Mozambique,” Journal of Religion
in Africa, 36:2, 2006, 139–166.
24 M. Newitt, “The Early History of the Sultanate of Angoche,” Journal of African History,
XIII:3, 1972, 397–406.
25 N. Hafkin, Trade, Society and Politics, 38–46.
26 L.J.K. Bonate, “The Ascendance of Angoche. The Politics of Kinship and Territory in
Nineteenth Century Northern Mozambique”, Lusotopie, 2003, 115–140.
27 N. Hafkin, Trade, Society and Politics, 50–58.
28 L.J.K. Bonate, “The Ascendance of Angoche”, 124.
the branches of the qādiriyya and the shādhiliyya 55

Islam. However, as L.J.K. Bonate has pointed out, the sources are inconclusive
whether Musa’s operation actually led to any mass conversions in the interior.
We do not know who this “relative” was, as he is not named by Eduardo do
Couto Lupi, who chronicled the life of Musa.29 However, it is not unlikely that
he was the person described as follows by Burhān Mkelle:

Among the renowned ʿulamāʾ of this island [Grande Comore] was the
ʿālim, the pious shaykh Aḥmad Mrūzī.30 He was learned in Arabic and in
the Islamic Sharīʿa. Among his venerable works was his call for Islamic
daʿwa in East Africa and his calling (daʿwa) of the people of Unguja
(Angoche)31 to Islam. They responded to his call and embraced Islam.
Even if he had no more good deeds to his name, this alone suffices for his
religion, righteousness and honour. The aforementioned Unguja is a land
(balad) on the coast of Portuguese Africa, only a short distance from the
island of Mzumbījī (Ilha de Mozambique). The shaykh settled there and
died there (may God grant him mercy).32

Little is known about this Aḥmad Mrūzi. He is stated by Burhān Mkelle to

have been one of the teachers of Abū ’l-Ḥasan Jamal al-Layl known as Mwinyi
Bahasani (1801–1883), who was introduced in Chapter 2 as a driving force for
Islamic education in Grande Comore.33 If this is correct, it places Aḥmad
Mrūzī’s year of birth some time in the late eighteenth century, and his most

29 E.C. Lupi, Angoche. Breve Memoria Sobre uma das capitanias-móres no Districto de
Mozambique, Lisboa, 1907. http://ia600404.us.archive.org/3/items/angochebrevemem0­
0lupigoog/angochebrevemem00lupigoog.pdf
30 Note that Mkelle spells the name of this shaykh differently in the two versions: MKELLE/
ZNZ-2 gives the name “Mrūzī” (MKELLE/ZNZ-2, 50), while MKELLE/ZNZ-1 uses the
name “Amrwāzī” (MKELLE/ZNZ-1, 58), with an alif at the start, and an alif following
the waw. In existing literature, Shaykh Mrūzī is known under the spelling “Mruzi”. D.E.
Gaba gives the transliteration “Mruzi” on the authority of B.G. Martin, who in turn took
his information from MKELLE/ZNZ-2 (See discussion of MKELLE/ZNZ-1 and MKELLE/
ZNZ-2 in the Sources section of this book). L.J.K. Bonate also uses the spelling “Mruzi”,
again on the authority of B.G. Martin. D.E. Gaba, “Le Manuscrit de Burhan Mkelle sur
la Grande Comore”, Asie du Sud-Est et Monde Insulindien: Islam et Littératures dans
l’Archipel Comores, Vol. XII: 3–4, 1981, 43–80, 68; Martin, B.G., Muslim Brotherhoods, 156;
L.J.K. Bonate, Traditions and Transitions, 81.
31 Mkelle gives the transliteration “Ngoji” for Unguja/Angoche.
32 MKELLE/ZNZ-2, 50.
33 MKELLE/ZNZ-2, 53. Mkelle does qualify his statement slightly, saying “(Mwinyi Bahasani)
took his knowledge from Shaykh Aḥmad Mrūzī, according to what we have heard.”
56 chapter 3

likely period of activity in Angoche in the mid-nineteenth century, i.e. more

or less as a contemporary of Musa Quanto. It also places him well before the
arrival of the Shādhiliyya or the Qādriyya in East Africa, and his activities must
therefore be understood as more generally daʿwa-oriented, most likely linked
to the general move towards organized Islamic education as it emerged in
Ḥaḍramawt, Zanzibar and the Comoro Islands itself.
In other words, Aḥmad Mrūzī could well be the “relative” accompanying
Musa Quanto. Whether or not the efforts of Musa and his “relative” actually
resulted in mass conversion, it seems clear that daʿwa oriented Islamization
in northern Mozambique was a combined result of the variously motivated
efforts of coastal Muslim leaders, paramount chiefs of the interior, and itiner-
ant preachers from overseas.

The “Comorian Network” and the Introduction of the Qādiriyya in

Northern Mozambique
As was described by B.G. Martin and thoroughly discussed by L.J.K. Bonate,
the Qādiriyya was brought to Mozambique Island around 1905–06 by one
Comorian-Zanzibari named ʿĪsā b Aḥmad al-Injazījī (“The Ingazijan”), also fig-
uring as “Issa d’Itsandra”. The movements, motives, and indeed the identity of
this shaykh ʿĪsā is surrounded by some confusion. B.G. Martin noted that ʿĪsa b.
Aḥmad, Qādirī khalīfa in Lindi, but normally resident in Zanzibar, had become
involved in the so-called “Meccan letter affair”34 around 1908.35 Another
observation is made by C. Ahmed, who state that the Qādiriyya arrived in
the Comoros only around 1910, by “Issa d’Itsandra”. The most likely sequence
of events is thus that ʿĪsā b. Aḥmad had visited Ilha de Mozambique before
becoming embroiled in the Meccan letter affair, and that he returned there
upon departing from Lindi following the German intervention. In other words,
he brought the Qādiriyya first to Ilha de Mozambique and then travelled to the
Comoros.36 B.G. Martin does indeed relate that ʿĪsā fled to “Palma and Ibo”, in

34 A letter that purportedly was written in Mecca started circulating among the Muslim
populations in the Lindi and Makonde region, but also as far inland as Ujiji and Tabora.
The letter was apocalyptical in nature, foretelling the imminent end of the world, but also
the end of European rule and eventual Muslim power. The Germans, with the Maji-maji
rebellion fresh in mind, took the potential for unrest and Muslim resistance of their rule
very seriously. Potential troublemakers, among them Qādirī khalifas were sought out as
leaders of the conspiracy.
35 B.G. Martin, “Muslim Politics and Resistance”; B.G. Martin, Muslim Brotherhoods, 156–157.
Note that Martin bases a substantial part of his account on that of Burhān Mkelle. See also
E.A. Alpers, “A Complex Relationship,” 163.
36 According to C. Ahmad, the Qādiriyya arrived in Grande Comore around 1910, introduced
by “Issa d’Itsandra”, who had been initiated in Zanzibar, also by Shaykh Uways. C. Ahmad,
the branches of the qādiriyya and the shādhiliyya 57

other words a direct escape southwards following the “Meccan letter affair”.
As L.J.K. Bonate also points out, the present-day shaykhs of the Qādiriyya in
Mozambique Island are unanimous that the order was introduced by ʿĪsā b.
Aḥmad, but have little information beyond basic biographical facts.
A more direct account of ʿĪsā’s proselytization efforts in Mozambique is
given by Burhān Mkelle, who was a contemporary of ʿĪsā and a fellow Qādirī in
Zanzibar. He is likely to have known ʿĪsā in Zanzibar, either before his depar-
ture to the mainland, after his return, or both. Burhān Mkelle, not unexpect-
edly, places ʿĪsā b. Aḥmad squarely as a Comorian, giving him a nisba name of
“al-Msujini” (referring to his place of origin in the village of Ntsujini on Grande
Comore):

I have omitted in this book the narratives of the many ʿulamāʾ of Grande
Comore who went to Madagscar and Ilha de Mozambique, because I
have little certain information about them.
The exception is that of shaykh ʿĪsā b. Aḥmad al- Mtsujini. He studied
the sciences of the Sharʿīyya in his native land, Grande Comore, and then
in Zanzibar, until he reached the level of the people of true knowledge.
Then he drank the drink of the ṭarīqa Qādiryya (i.e. was initiated) and
became one of its leaders (murshids). He travelled to Mozambique and
stayed there for years teaching the principles of religion. He travelled
around, calling the people to the religion of Islam, until many unbeliev-
ing natives (zunūj) had converted. He caught a serious illness and
returned to Zanzibar for treatment, and he died there in 1344/1925–26.
May God have mercy on him.37

From Abdallah Saleh Farsy, we learn that ʿĪsā b. Aḥmad upon one of his
return visits to Zanzibar, was accompanied by one Said b. Abdallah Lindy
Mmakonde (c. 1890–1956), probably from Lindi where ʿĪsā had functioned as
Qādirī khalīfa.38 It is not clear if this was one of ʿĪsā’s converts, a fellow Qādirī
or a leader of the Makonde Muslim community. What is clear, however, is that
Ahmad entered the household of ʿAbd Allāh Bā Kathīr, and travelled with him
on many journeys. From 1914, Said taught in the Vuga mosque in Zanzibar,
before departing for Tabora in 1953. The story shows that students were sent

Religion et Politique aux Comores. Evolution de l’autorite spirituelle depuis le protectorate

francais (1886) jusqu’a nos jours, Paris (L’Harmattan), 2000, 118. See also an alternative
chain of transmission into the Comoros discussed by T.A. Mohamed, La Transmission de
l’Islam aux Comores (1933–2000), Paris (L’Harmattan), 2008, 88–90.
37 MKELLE/ZNZ-2, 55.
38 Farsy/Pouwels, The Shafiʿi Ulama, 116.
58 chapter 3

from southern Tanzania and probably also Mozambique to Zanzibar, and that
they entered a into scholarly network that opened for teaching positions.
The introduction of the Qādiriyya to Mozambique by Shaykh ʿĪsā is further
substantiated by the silsila manuscripts held by the present-day Qādiriyya
in Ilha. The silsila39 examined for this analysis was given by ʿĪsā to one Kunā
ʿAhad b. Kuʾil and is dated 1322/1904–05. It is also notable, although not unex-
pected, that the text contains the exact same wording as the ijāza from Shaykh
Uways to Mkelle b. Adam, except that the order of the two first paragraphs has
been changed.40

This is the silsila from its origin (shajara aṣlihā) and its branches. Its car-
rier is a noble man.
I am the poor in God’s sight and the servant of the fuqarāʾ, Shaykh
ʿĪsā b. Aḥmad al-Qādirī, servant of my shaykh ʿAbd al-Qādir al-Jīlānī, may
God hallow his secret.
This is to confirm to all our Muslim brothers that the carrier of this let-
ter, Kunā ʿAhad b. Kuʾil al-Qādirī, is (one) of the wandering seekers
(al-darāwīsh al-sālikīn). He came to us, and he entered the assembly of
the pole of the knowers (quṭb al-ʿarifīn), the guide of seekers, my shaykh
ʿAbd al-Qādir al-Jīlānī, may God hallow his secret [etc.].
It may be that you achieve entry into the group affiliated with this
ḥaḍrat al-Sunniya, by honouring him and showing generosity towards
him and protecting him from encroachments, as did the Ayat al-Sharīf.
He (SAWS) said: “The highest reward will bestowed on him who share
with a guest from his estate, or eases his distress by a drink of water or
food or clothing or a smile on his face, for him is the garden of Paradise”.
I have chosen him as my khalīfa [. . .] ʿĪsā b. Aḥmad al-Qādirī.
1322/1904–05.

The text then goes on to say that the author (ʿĪsā b. Aḥmad) was a student of
Shaykh Uways, thus establishing the following chain of transmission:

39 QADIRI SILSILA/ILHA. Scroll, 74 lines. This silsila was photographed by Dr. Elke
Stockreiter in January 2007. I am grateful to her for allowing me to use the manuscript for
this discussion.
40 There are also some minor discrepancies where a word has been omitted, probably by
mistake. Furthermore, contrary to the Zanzibar ijāza issued by Shaykh Uways, the sil-
sila and ijāza given by Shaykh ʿĪsā does not mention any specific texts, such as Mawlid
al-Barzanjī.
the branches of the qādiriyya and the shādhiliyya 59

QADIRI SILSILA/ILHA

Silsila back to ʿAbd al-Qādir al-Jīlānī and the Prophet

Sayyid Muṣṭafā b. Sayyid Sulaymān al-Jīlānī (Baghdad)
Uways b. Muhammad al-Barawī
ʿĪsā b. Aḥmad al-Qādirī al-Ngazījī
Kunā ʿAhad b. Kuʾil al-Qādirī

Figure 4 Ilha Qādiriyya through ʿĪsā b. Aḥmad

Noticeably absent from the silsila is ʿUmar Qullatayn. Here, the chain of author-
ity goes from Shaykh Uways to ʿĪsā directly. This leads to the conclusion that
the introduction of the Qādiriyya in 1904/05 came through what was called
above the “Comorian network”, and that ʿĪsā b. Aḥmad was a khalīfa initiated
by Uways alongside Mkelle b. Adam and others. However, as will be discussed
below, the later fractioning of the Qādiriyya in Mozambique Island may have
led to some “re-orientations” of the silsilas, linking the chain of authority to the
“Brawanese” (and sharīfian) branch instead, represented by ʿUmar Qullatayn.
It is also likely that a number of Comorians circulated the route Grande
Comore-Zanzibar-Mozambique, including ʿĪsā b. Aḥmad. As was pointed out
by Maalim Idris Muhammad Saleh in one interview, his own grandfather had
left Grande Comore for Mozambique where he settled and died.41 Parts of the
family then moved onwards to Zanzibar, through well-established migration
channels. According to Maalim Idris’ family history, this was a “common pat-
tern” in the mid/late nineteenth century.

From Zanzibar via the “African Network”: The Students of Shaykh

Mjana Kheri
A second, and clearly distinct route of the Qādiriyya into Mozambique was
that opened by the many students of Shaykh Mjana Kheri in Zanzibar. An
example of this can be found in a Qādirī silsila document photographed in
Palma in Northern Mozambique in 1988.42
As has been noted by E. Rzewuski, little remains of direct scriptural sources
of the Tungi sultanate, which was located around what is today the village of

41 Interview, Maalim Muhammad Idris Muhammad Saleh, 16.07.2006.

42 QADIRI SILSILA/PALMA. I am grateful to Prof. Eugeniusz Rzewuski, University of Warsaw,
for providing me with a copy of this manuscript.
60 chapter 3

The silsila of the Palma branch of the Qādiriyya

QADIRI SILSILA/PALMA

Silsila back to ʿAbd al-Qādir al-Jīlānī, onwards to the Prophet Muḥammad, and back
to Adam and creation.
Sayyid Muṣṭafā b. Sayyid Sulaymān al-Jīlānī (Baghdad)
Uways b. Muhammad al-Barāwī (1847–1909)
ʿAbd Allāh Mjana Kheri (Zanzibar)
ʿAbd al-Mushadil (Zanzibar?)
Ḥusayn b. Ramaḍān (Zanzibar?)
Ismāʿīl b. Ḥamīd (Zanzibar or Palma?)
ʿAlī b. Ḥamīd b. ʿAlī (Zanzibar or Palma?)
Muḥammad b. Thābit (d. Palma, 2005/2006)

Figure 5 Qādirī silsila of Palma, Mozambique

Kiwiya on the peninsula of Cabo Delgado.43 The sultanate is, however, men-
tioned in several Swahili and Comorian chronicles, either as one of the set-
tlement of the mythical seven (or nine) princes arriving from Shiraz to East
Africa, or even as the point of origin for the early inhabitants of the Comoros.
In the latter half of the nineteenth century, Tungi became a source of con-
troversy between the Zanzibari sultanate and the Portuguese who eventually
annexed the area in 1877.44 Equally little is known about the various connec-
tions which existed between Zanzibar and the Tungi region, except that these
resulted in the specific linguistic situation whereby Swahili was used among
men, as a marker of “civilization” while women continued to speak the ver-
nacular Kimwani. From the point of view of Islamic history, equally little is
known, and the following analysis of the introduction of the Qādiriyya is thus
a small contribution to fill this lacuna.

43 On what little is known about the history of the Tungi sultanate, see E. Rzewuski, “Origins
of the Tungi Sultanate (Northern Mozambique) in the light of local traditions”, in:
S. Pilaszewicz and E. Rzewuski (eds), Unwritten Testimonies of the African Past. Proceedings
of the International Symposium held in Ojzan, Warsaw 7–8 November 1989, University of
Warsaw (Orientalia Varsovensia), 1991.
44 On this controversy, see N. Bennet, “Zanzibar, Portugal and Mozambique: Relations from
the late eighteenth century to 1890,” Working Papers in African Studies, Boston University,
1987.
the branches of the qādiriyya and the shādhiliyya 61

At the time of photographing, the document was owned by shehe

Muḥammad b. Thābit, who had studied in Zanzibar, and who was a respected
leader of the Muslims in Palma. He is the last person in the silsila, and died
some time in 2005/6.
Based on the dates of birth and death of the shaykhs, this would date the
arrival of this branch of the Qādiriyya in Palma to some time in the early twen-
tieth century, probably before or around 1920.
However, this dating is uncertain, as the individuals between Shaykh Mjana
Kheri and Muḥammad b. Thābit unfortunately remain obscure. While it is fair
to assume that ʿAbd al-Mushadil (who follows directly upon Shaykh Mjana
Kheri) was initiated in Zanzibar, we know nothing about the three names that
follow. One likely interpretation is that both ʿAbd al-Mushadhil and Ḥusayn
b. Ramaḍān were initiated in Zanzibar – based on the reasonable assumption
that Ḥusayn b. Ramaḍān is identical to the Ḥusayn named by Amina Issa as a
Zanzibari Qādirī scholar of mainland origin.45 One possible interpretation is
that Muḥammad b. Thābit was initiated in Zanzibar, and that all the scholars
preceding him in the silsilsa were in fact Zanzibar-based. Another possibility
is that either one of Ismāʿīl b. Ḥamīd or ʿAlī b. Ḥamīd travelled from Palma to
Zanzibar, or vice versa.
Based on information given in the silsila itself, a third, and most likely
sequence of events is that ʿAlī b. Ḥamīd was the first local Palma scholar,
as his ijāza (from his shaykh Ismāʿīl) is given in more detail than the others in
the text:

Then, when I saw (that) the pious disciple, the successful seeker and for-
tunate believer, the khalīfa, shaykh ʿAlī b. Ḥamīd b. ʿAlī al-Qādirī (may his
blessings (baraka) last long) was suited for it [Qādirī khalīfaship] and
that he merited what is in it of knowledge.
I instructed him in the Kalimat al-Tawḥīd. I clothed him in the blessed
robe of initiation (khirqa) and made him my khalīfa in the ṭarīqa
Qādiriyya and raised him to my own level.
I placed him on my seat, and gave him an ijāza to teach the noble dhikr
and to place on others the mantle of initiation. [I furthermore authorized
him] to appoint as khalīfa whoever he wanted among the Muslims who
merit it. [I also instructed] that he open the doors of the zāwiya for the
service of the poor, and that he collect alms and distribute them among

45 A.A. Issa, “The legacy”, 353. According to Issa, Ḥusayn Ramaḍānī (1880–1978) was was born
to Zigua parents in Mwera outside Stone Town. In 1910, Shaykh Ḥusayn opened his own
Qurʾanic school in Ngambo, Zanzibar. See also R. Loimeier, Between Social Skills, 80.
62 chapter 3

some of those in need and among the poor who come to him. [I further
instructed that] he pass the basket around [distribute food among the
poor] according to his own volition and dispose freely of it among them.
I authorized him to give the aforementioned well supported and trans-
mitted khirqa which links back to the Prophetic presence to those who
were capable and genuine from among his students and friends, high and
low, male and female, so that the wearer (of the khirqa) could rejoice.
I also gave him (May his blessings last) an ijāza to take the pledge (nadhr)
and particularly the pledge of our Sayyid and Shaykh Sultan of Sultans,
and the guide of the seekers, head (raʾs) of the beloved of God, the Quṭb
of the knowers, Sayyid Shaykh ʿAbd al-Qādir al-Jīlānī, may God hallow his
secret and illumine his tomb.46

The detailed instructions, combined with the extensive chain of transmission

that precedes the actual ijāza, indicates the need for a local scholar (in this
case ʿAlī b. Ḥamīd) to show his credentials in order to spread the order further
in his community. It should also be noted that the text in this silsila is different
from the three others presented above (where the links are directly to Shaykh
Uways). First of all, it does not have the form of a letter of introduction, but
rather a long silsila with narratives included. It starts with the sentence “This is
the tree from its origin and with its branches” but continues with the immedi-
ate chain of transmission of the present carrier, Muḥammad b. Thābit.
What the silsila also indicates is the importance of charitable work con-
ducted on the ground – locally – by the order. In other words, the document
establishes, through reference to “facts on the ground” (redistribution of alms,
the opening of the zāwiya to the poor), a new type of “truth on the ground” that
had the potential to supersede previous lines of social organization. In other
words, whether initiated in Zanzibar or in Palma, ʿAlī b. Ḥamīd followed a well-
established pattern of recruiting the lower levels of society into the brother-
hood, thus establishing a new form of community based on ṭarīqa affiliation.

From Zanzibar Directly via the Scholarly Network:

Abū ’l-Ḥasan Jamal al-Layl (1888–1959)
The first of the Zanzibari Qādirī silsila lines outlined above was that propa-
gated by ʿAbd al-Azīz al-Amawī and Sayyid ʿUmar Qullatayn. Among the fore-
most initiates of ʿUmar Qullatayn – and indeed among the foremost public
intellectuals in Zanzibar Town during the first half of the twentieth century –

46 QADIRI SILSILA/PALMA, line 85–103.

the branches of the qādiriyya and the shādhiliyya 63

was Abū ’l-Ḥasan Jamal al-Layl.47 He was born in 1888, in an unknown location
in Madagascar. At age four, he was sent to Zanzibar along with his brothers,
and there he started his education with Aḥmad b. Sumayṭ and ʿAbd Allāh Bā
Kathīr. In other words: He was placed into the network of ʿAlawī scholars, very
much in line with his own family background. However, some time in the early
twentieth century, he was initiated into the Qādiriyya by ʿUmar Qullatayn,
who later also appointed him khalīfa in Zanzibar Town. In this capacity, he
led dhikr and prayers and also recited during the mawlid celebrations in Stone
Town mosques. As a scholar, Abū ’l-Ḥasan mastered the Islamic sciences, but
excelled particularly in the Arabic language and poetry. He is also reported to
have authored a history of the ʿulamāʾ of East Africa, which was later to be built
upon and expanded by Abdallah Saleh Farsy. Abū ’l-Ḥasan then started teach-
ing in various mosques as well as in his home. Interestingly, he also started
teaching in the Zanzibar Government Schools (opened in 1905), and he is
stated to have been the one to recruit Burhān Mkelle and the Egyptian teacher
Shaykh ʿAbd al-Bārī. The image of a “public intellectual” is projected also by his
other activities, for example as a fund-raiser for the Jamʿiyyat al-Sunna estab-
lished in Zanzibar in 1927 (See further below, Chapter 8).
The image of a socially active, daʿwa-oriented scholar is strengthened by the
fact that Abū ’l-Ḥasan Jamal al-Layl sometimes wrote in newspapers concern-
ing religious matters. However, his most enduring legacy is as a poet. His poetry
was recited during the mawlid celebrations and his diwān was republished as
late as the 1990s.48
According to his obituary, he travelled widely, to Kenya, Tanganyika and
Mozambique – “to spread Islam and to educate the people”. It is stated explic-
itly that he, during his daʿwa missions in Mozambique, initiated people into
the Qādiriyya.

47 The following information concerning Abū ’l-Ḥasan Jamal al-Layl derives from his
obituary printed in Mwongozi, (undated, but printed shortly after his death in August
1959) and al-Mashhūr, Lawāmiʿ al-Nūr, 234. Abū ’l-Ḥasan is known also as Abū ’l-Ḥasan
al-Ṣaghīr (the small one), with reference to his nineteenth predecessor by the same name
(see above, Chapter 2). It should be noted that a somewhat miraculous event is associ-
ated with Abū ’l-Ḥasan al-Ṣaghīr: He is reported to have stayed in his mother’s womb for
11 months. (al-Mashhūr, Lawāmiʿ al-Nūr, 234).
48 Al-Ḥabīb Abū ’l-Ḥasan b. Aḥmad Jamal al-Layl, Diwān al-Fatḥ wa’l-Imdād, printed in
Dubai, ND.
64 chapter 3

The Qādiriyya: Different Silsilas – Different Routes to Mozambique

Viewed together, the three different paths of the Qādiriyya demonstrate dif-
ferent geographical orientations and distinct “flows” of Islamic learning in the
southwestern Indian Ocean. Whether or not these actually represented differ-
ent practices “on the ground” can only be determined by future research.
By the 1920s, Mozambique Island had emerged as a centre for the Qādiriyya.
However, the order was heading for a split after the return of ʿĪsā b. Aḥmad
to Zanzibar. L.J.K. Bonate has given a detailed analysis of the divide which
was had its roots in ethnic divisions and revolved around Islamic authority.49
Around 1929, leadership of the Qādriyya in Ilha de Mozambique and the sur-
rounding region passed to Sayyid ʿAbd al-Raḥmān Bā Ḥasan who, like ʿĪsā b.
Aḥmad Mtsujini, originated from Grande Comore and who was of Ḥaḍramī
ʿAlawī descent. He had been the student of ʿĪsā from an early age, as ʿĪsā, upon
his arrival in Ilha, had lived in the house of the Bā Ḥasan family. However,
Bā Ḥaṣan’s authority was challenged around 1930 by the so-called “moors”
(a group of mixed Indian-African origin) from the mainland who had also taken
up the Qādiriyya. In response to the challenge, Bā Ḥasan provided a silsila from
Zanzibar, detailing his authority as a khalīfa through Sayyid ʿUmar Qullatayn.
The moor’s leader, in turn, visited Baghdad where he obtained another silsila
which was dismissed by Bā Ḥasan. Consequently the Qādiriyya, by 1934, was
split in two; one faction with its chain of transmission passing through Shaykh
Uways and ʿUmar Qullatayn, and one branch with a spiritual chain of trans-
mission coming from Baghdad.
As we saw above, the initial chain of initiation (from ʿĪsā b. Aḥmad) did not
at all go via ʿUmar Qullatayn, but to Uways directly. The most likely interpreta-
tion is that Shaykh Bā Ḥasan, during his journey to Zanzibar, had his Qādirī
affiliation “re-confirmed” through Abū ’l-Ḥasan Jamal al-Layl, who, as we have
seen, had his initiation through ʿUmar Qullatayn. There are two possible ways
in which this could happen. The first is that BāHasan sought out Qullatayn
before the latter’s death in 1926, which would push the challenge from the
moors somewhat back in time from the account given by L.J.K. Bonate, to the
first half of the 1920s. The second is that he sought out Abū ’l-Ḥasan Jamal al-
Layl (who, as we have seen, had been initiated by Qullatayn), and in this way
re-oriented his Qādirī affiliation towards the urban branch in Zanzibar. There,
Abū ’l-Ḥasan was at the height of his career, propagating the Qādiriyya while
teaching and composing poetry.
It should be noted here that this was not Abū ’l-Ḥasan’s only intervention
into quarrels among ṭarīqa leaders in the region. Later, in 1949, he was called

49 For detailed account of this split, see L.J.K. Bonate, Traditions and transitions, 80–95.
the branches of the qādiriyya and the shādhiliyya 65

upon by the British colonial authorities in Malawi (Nyasaland) to settle a dis-

pute between Yao ṭarīqa leaders concerning proper behaviour during funerals.
This time, the disagreement was between the so-called “zukutis” (“quietists”),
who favoured a sober, quiet tone, and other ṭarīqa leaders who permitted
music, dancing, singing and the flying of banners during funerals.50
The trajectory outlined above of the re-orientation of the Qādriyya in Ilha
de Mozambique makes sense for several reasons. Firstly, as L.J.K. Bonate has
described, by the 1920s, the order was being brought under the influence of
Muslim ruling clans, who in turn were becoming increasingly connected to
their clansmen and fellow Muslims to the north.51 Secondly, the very fact that
he was an ʿAlawī sayyid is likely to have made Bā Ḥasan more prone (by fam-
ily networks) to attach himself to the branch propagated by his fellow ʿAlawī
Abū ’l-Ḥasan Jamal al-Layl. It also fits well with the time of the establishment
of the Jamʿiyyat al-Sunna in Zanzibar (see below, Chapter 8) that may have
taken the idea of taking the daʿwa to Mozambique, as well as his later involve-
ment with the Yao dispute, precisely from the contact with Bā Ḥasan. Finally, it
is worth noting that in his own Qādirī writing, Abū ’l-Ḥasan refers to the com-
munity of Qādirīs as al-Sādāt al-Qādiriyya.52
This interpretation is also substantiated by later (1940s) Qādirī text collec-
tions from Mozambique Island, where ʿUmar Qullatayn features among the
Qādirī shaykhs for whom fātiḥas are read.53 The recital of the fātiḥa for the
Qādirī shaykhs was and is an integral part of Qādirī ritual:

A Fātiḥa for all the shaykhs of the ṭarīqa Qādiriyya. A Fātiḥa for the soul
of our shaykh and ustādh [teacher] ʿUmar Qullatayn, may God have
mercy upon him. A Fātiḥa for the soul of our shaykh and ustādh Sayyid
Muḥammad Zuʿūb Jīlānī, may God have mercy upon him. A Fātiḥa for
the soul of our shaykh and ustādh ʿAbd al-Shakūr b. ʿAbd al-Razzāq, may
God have mercy upon him. A Fātiḥa for the soul of our shaykh and ustādh
shaykh ʿAbd Allāh b. Sālim, may God have mercy upon him. A Fātiḥa for

50 J.N.D. Anderson, Islamic Law in Africa, London (Frank Cass), 1970, 169–70.
51 L.J.K. Bonate, “Muslim Religious Leadership in Post-Colonial Mozambique”, South African
Historical Journal, 60 (4) 2008, 637–654.
52 Maalim Idris Collection, ABU AL-HASAN/ZANZIBAR 1. Manuscript, undated, by internal
evidence by Abū ’l-Ḥasan Jamal al-Layl. The same reference is used in QADIRI DHIKR/
ZANZIBAR1.
53 QADIRI QASIDA/ILHA 1 (notebook, handwritten, 83 pages) and QADRI QASIDA/ILHA 2
(notebook, hadwritten, 135 pages). These are notebooks, both in more than one hand,
dating from around the 1940s.
66 chapter 3

the soul of our shaykh and ustādh, the guide of the murīdīn [students]
and leader of the sālikīn [seekers], through whom forgiveness is granted,
ʿĪsā b. Aḥmad al-Qādirī, may God have mercy upon him. A Fātiḥa for the
souls of all the people of the Qādirī and Shādhilī silsila, from the East to
the West and from the West to the East. May they have (God’s) forgive-
ness and mercy.54

The selection of shaykhs singled out for prayer in this 1940s document indi-
cate, as Bonate also has pointed out, a re-orientation some time in the 1920s or
30s towards the “learned class” network in Zanzibar and the formation of the
Qādiriyya sādāt. Although not all the names in the prayer list can be identified,
the “ʿAbd al-Shakūr b. ʿAbd al-Razzāk” mentioned is possibly identical to the
scholar by that name in Zanzibar.55
From the presently available evidence, it remains unclear whether the branch
of the northern Mozambican Qādiriyya stemming from the line of Shaykh
Mjana Kheri was confined to Cabo Delgado (Palma, Quissanga, Querimba and
other locations in the interior) or if it actually proliferated further south. What
the sources do indicate is that this particular line had a distinct orientation dif-
ferent from the other two, recruiting from different social stratas and/or ethnic
groups. The text of the Qādiriyya initiation document deriving from Shaykh
Mjana Kheri and his followers contain a much stronger orientation towards
the poor, and towards practical alleviation of poverty, such as distribution of
food etc. What may be deduced is that this branch recruited more extensively
from the lower, less educated classes of northern Mozambique, and that the
diffusion of this branch is the one that may be directly compared to that in
Tanzania, as described by F. Becker.56 Although more research is needed to
needed to further substantiate this claim, the evidence presented here indi-
cates that the diffusion of the Qādiriyya in northern Mozambique was drawn
between two poles; that of the combined Comorian/Zanzibarī leadership and
that of an “Africanized”, and by the mid-twentieth century, localized leader-
ship. Some corresponding traits can also be detected in the diffusion of the
Shādhiliyya, as will be discussed below.

54 QADRIRI QASIDA/ILHA 1, 73–74.

55 Farsy/Pouwels, The Shafiʿi Ulama, 36. ʿAbd al-Shakūr was a scholar of Indian origin who
studied with Aḥmad b. Sumayṭ and ʿAbd Allāh Bā Kathīr in Zanzibar. He died in 1921.
56 F. Becker, Becoming Muslim, passim.
the branches of the qādiriyya and the shādhiliyya 67

Muḥammad Al-Maʿrūf and the Spread of the Shādhiliyya in

Northern Mozambique

According to E. Alpers, the groundwork for the spreading of Sufi orders into
northern Mozambique was laid by long-standing commercial networks and
personal relations between the Mozambican coastal peoples (in particular
Angoche) and the Comoro Islands.57 The most prominent example of such
contacts is the journey of Muḥammad al-Maʿrūf to Ilha de Mozambique in
1897 – a journey undertaken with the explicit purpose of spreading the ṭarīqa
Shādhiliyya on the island. According to L.J.K. Bonate, Muḥammad al-Maʿrūf
gave ijāzas to “local Swahili or Muslim chiefs”, rather than to fellow Comorians,
of whom there were many in Ilha de Mozambique.58
One such Comorian was ʿĀmur b. Jimba who already had preached the
Shādhiliyya-Yashruṭiyya in Mozambique for a year before the arrival of
al-Maʿrūf. ʿĀmur b. Jimba was born in Moroni, and Burhān Mkelle, in his
account, list him among the “ʿulamāʾ of Grande Comore” and gives him the
nisba “al-Murūnī al-Shādhilī.”59 He is said to have studied in his native town
for some time before travelling to Zanzibar, where he applied himself to the
“search of knowledge”. Apparently, ʿĀmur b. Jimba was somewhat lacking in
educational background when he arrived in Zanzibar in 1311/1893–94. Mkelle,
who knew him then, state that he “knew only a few principles of the religious
sciences.” However, so clever was ʿĀmur that by devoting himself wholly to his
studies, he (according to Mkelle) became a fully-fledged scholar within a very
short time. Armed with new knowledge, he proceeded to Ilha de Mozambique,
where he settled and had children. There, he took his time to “settle into its
ways, to know its people, and the extent of their knowledge and their need of
religious teaching”. Mkelle states further that people in Mozambique recog-
nized the value of his knowledge, his gracious manner, and his will to teach.
Specifically, ʿĀmur was teaching the rules of worship (ʿibāda) in addition to
guiding the people to the Shādhiliyya. He died in Ilha de Mozambique in
1339/1920–21.
For reasons unknown, al-Maʿrūf, upon his departure, decided to pass on the
leadership of the Shādhiliyya not to his fellow Comorian ʿĀmur b. Jimba, but
to two local leaders, Haji Muhammad Ahmad Gulamo and Naiman b. Haji Ali
(see their inclusion in silsila documents below). Based on Portuguese colonial
sources, L.J.K. Bonate interprets this as an attempt by al-Maʿrūf to create ties

57 E.A. Alpers, “A complex relationship.”

58 L.J.K. Bonate, Traditions and Transitions, 76.
59 MKELLE/ZNZ-2, 2, 55–56.
68 chapter 3

to the local population, by choosing established local leaders as khalīfas. In

contrast to the rosy picture drawn by Mkelle, she argues that ʿĀmur b. Jimba
would have been unacceptable to the people of Ilha, due to his lack of kin-
ship, property or credentials. Nonetheless, soon after Al-Maʿrūf’s departure,
ʿĀmur b. Jimba claimed to be the “real” khalīfa of the Shādhiliyya in Ilha de
Mozambique. As a response, Al-Maʿrūf sent his brother, ʿAlī b. Shaykh Aḥmad
(also known as Ali wa Shehi), to settle the dispute, with the result that the
order – for the first few years – remained as a sub-branch of the Comorian one.
Then, some time in the early years of the twentieth century, leadership eventu-
ally passed on to ʿĀmur b. Jimba, despite his position as “outsider” in the Ilha
community. He held this position until his death, whereupon the struggle for
leadership resumed. Again, the power-struggle may be interpreted as a rivalry
over the right to define the legitimacy of religious praxis, but also as a struggle
for rights to intervene and settle disputes in general – the traditional marker
of chiefdom.
B.G. Martin noted that the further proliferation of the Shādhiliyya to
Mozambique “remains obscure.”60 A survey of four existing silsila documents
in Ilha de Mozambique, dated at different points in the twentieth century,
does little to clarify the trajectory. The first is clearly a handwritten copy of an
earlier version. It shows Muḥammad Aḥmad Ghulām being initiated into the
Shādhiliyya in 1906 with no particular reference to any local initiation.61
A later document shows the same entry of the Shādhiliyya via Muḥammad
Aḥmad Ghulām, initiated by ʿĀmur b. Jimba, then ʿAlī b. Shaykh Aḥmad via
Muḥammad al-Maʿrūf and ʿAbd Allāh al-Darwīsh to Abū ’l-Ḥasan al-Shādhilī
and from him back to the Prophet.62 This is repeated in two further silsilas,
both deriving from some time in the first half of the twentieth century:63

I authorize him in all [the above mentioned dhikr, given at the outset]
which I took from my shaykh Nuʿmān b. Hājj al-Matāmī and he from his

60 B. G. Martin, Muslim Brotherhoods, 156.

61 SHADHILI/ILHA 1. Scroll, 43 lines, dated 12 Dhū ’l-Ḥijja or Dhū ’l-Qaʿda 1324/Dec 1906/Jan
1907, Arabic. Photographed Mozambique Island, 2007 by Dr. Elke Stockreiter.
62 SHADHILI/ILHA 2. Scroll, 59 lines, undated, Arabic. Photographed Mozambique Island,
2007 by Dr. Elke Stockreiter. By internal evidence (the person being initiated, Ṣāliḥ b.
ʿUbūd, is 3 links removed from Muḥammad Ghulām), this document is more recent than
SHADHILI/ILHA 1. This is also indicated by the appearance of the paper and ink.
63 SHADHILI/ILHA 3. Scroll, undated, 48 lines, Arabic. Photographed by Chapane Mutiua,
line 13–17. This document appoints as khalīfa a student of Nuʿmān b. Ḥājj, and should thus
be dated around the 1920s. The same chain of transmission is given in SHADHILI/ILHA 4.
Scroll, undated, 46 lines, Arabic; Photographed by Chapane Mutiua.
the branches of the qādiriyya and the shādhiliyya 69

shaykh Hājj Ghulām b. Muḥammad b. Aḥmad, and he from his shaykh,

ʿĀmur b. Jimba, and he from his shaykh, ʿAlī b. Shaykh Aḥmad, and he
from his shaykh, Muḥammad b. Shaykh Aḥmad al-Maʿrūf, and he from
his shaykh, ʿAbd Allāh al-Darwīsh, and he from his shaykh, ʿAlī b. ʿAbd
Allāh al-Yashruṭī.

Thus, we have two parallel narratives on the initiation of Muḥammad Ghulam,

one with a clear “Comorian connection” and one without:

SHADHILI SILSILA/ILHA 1 SHADHILI SILSILA/ILHA 2–3–4

Silsila back to Abū ’l-Ḥasan al-Shādhilī and

the Prophet
ʿAbd Allāh al-Darwīsh
Muḥammad al-Maʿrūf
ʿAlī b. Shaykh Aḥmad (brother of al-Maʿrūf)
ʿĀmur b. Jimba
Muḥammad b. Aḥmad Ghulām Muḥammad b. Aḥmad Ghulām
The students of Muḥammad b. Aḥmad Nuʿmān b. Hājj

Figure 6 Shādhiliyya of Ilha de Mozambique

L.J.K. Bonate has given a detailed account on the activities of ʿAlī b. Shaykh
Aḥmad as the middleman in his 1898 efforts to settle the controversy over
khalīfaship there,64 notably the claim of Muḥammad Aḥmad Ghulām to spiri-
tual leadership in the order. As Bonate has described, Aḥmad b. Shaykh and
ʿĀmur b. Jimba’s lack of local, mainland connection through marriage meant
that they were unacceptable as local leaders of the Shādhiliyya, and that
spiritual leadership was maintained by the khalīfas of Muḥammad b. Aḥmad
Ghulām and Nuʿmān b. Hājj. The two versions must thus be understood as
post-fact attempts to either exclude or integrate the Comorian chain of author-
ity, and implicitly also different ways of rooting authority locally.

64 Settling disputes seems to have been somewhat a specialty of ʿAlī b. Shaykh. He was sent
by al-Maʿrūf to Anjouan to reconcile with antagonistic factions there. However, he was
unable to finally settle the dispute over control of the order in Moroni, where his descen-
dants still claim control over the zāwiya, while another faction claim spiritual leadership.
C. Ahmed, “Networks”, 326–328.
70 chapter 3

The Emergence of Sufi Orders in Norhern Mozambique

The main impact of the Qādiriyya and Shādhiliyya Sufi orders seems to have
been an institutionalization of Islamic learning, whether in formalized schools
(madāris) or in study circles. This was especially so in the branch that traced
its spiritual authority back to the “learned” network, but also for the branch
spread by Shaykh Mjana Kheri’s disciples. This process took place in northern
Mozambique somewhat later than in Zanzibar and the Comoros. However, it is
worth noting that the delay was no more, and probably significantly less, than
8–10 years, which in turn indicates the close connections, especially with the
Comoro Islands. However, it is also clear that more research is needed to trace
in detail the local impact of the Sufi orders on Islamic learning in twentieth
century Mozambique.
More research is also needed on the impact of the Sufi brotherhoods on new
forms of social organization. The initiation into the Qādiriyya (as with the other
orders) implied a new belonging that in principle transcended earlier affilia-
tions, including kinship and family ties. In reality, however, it often became
an additional component of social identity. The emphasis on a new form of
loyalty was clearly expressed in the Qādirī ritual of initiation, as transmitted by
Abū ’l-Ḥasan Jamal al-Layl (and certainly also by all the other branches):

This is how to pass on the ijāza to a murīd among our sādat al-Qādiriyya:
First, the shaykh motions the murīd to sit in front of him. The shaykh
then gives the dhikr, while the murīd supports his weight [on the shaykh]
and covers his eyes. Then he asks God’s forgiveness three times, together
with the shaykh. The shaykh then takes the hand of the murīd and reads
the prescribed words: “In the name of God, most merciful and most
compassionate. Praise be to God and our Prophet Muḥammad [SAWS]”.
Then, the shaykh gives the silsila of the ṭarīqa back to the Prophet SAWS.
Then, the shaykh says: “I authorize you, oh murīd, in the reading of the
Qādirī wird “Lā ilah ilā Allāh” 165 times after every prayer. [Text omitted,
more prayers to the Prophet and Jibrīl.] The shaykh then recites Qurʾān
48:10: “Verily, those who plight their fealty to thee, do not less than plight
their fealty to God: The hand of God is over their hands: Then any one
who violates his oath, does so to the harm of his own soul, and anyone
who fulfils what he has covenanted with God, God will soon grant him a
great reward. Then, the shaykh says “Lā ilah ilā Allāh” three times and the
murīd says it with him, and when the shaykh says “Lā ilah” he takes that
hand of the murīd in his right, and when he says “ilā Allāh” he takes the
hand of the murīd in his left, and repeats this tree times. Then, the shaykh
the branches of the qādiriyya and the shādhiliyya 71

takes water sweetened with sugar or honey, and says “In the name of God,
most merciful, most compassionate”. He then hands it to the disciple who
drinks it all, and while drinking he opens his eyes. After that he prays
two rakʿas, giving thanks to God and our Prophet Muḥammad and His
Companions until the day of Judgement.65

The oath of loyalty to the Qādirī shaykh repeats that of the early Muslims
pledging their loyalty to the Prophet, thus renouncing old allegiances. It is
clear that this was not truly the case with all Qādirī converts, as was also shown
by P. Caplan in her 1975 study from a village on Mafia Island. A long-standing
rivalry between two Qādirī leaders tended to (but did not exclusively) organize
ṭarīqa membership along descent-group lines. At the same time, events such
as the annual ziyāra united the entire village (across descent group lines) in the
efforts to prepare the grounds, the food etc.66 Kinship ties remained strong, in
other words, but, as this chapter has shown, so did overseas family ties – nota-
bly among the Comorians. A clear example is Burhān Mkelle, who in the 1920s
still recounted the role of “Comorian” scholars in Mozambique. This also led to
tendency towards “double book keeping” when it came to the records of spiri-
tual authority among the Qādirīs and Shādhilīs. This “double book-keeping” in
turn, indicates that the arrival of Sufi orders, despite its inherent potential, did
not override existing forms of social organization, but rather proliferated as
parallel networks with parallel chains of authority.
However, the reverse tendency is also clear. Like in the Benadir, the Sufi
orders in Mozambique did cut across social, cultural and economic boundar-
ies in the sense that they recruited from all levels of society and from a range
of ethnic bakgrounds. As we shall see, a similar pattern emerged in the next
location where Comorian scholars played an important role: Madagascar.

65 ABU AL-HASAN/ZANZIBAR 1. Manuscript, undated, by internal evidence by Abū ’l-Ḥasan

Jamal al-Layl. The instruction is repeated in several Qādirī prayer books, such as for exam-
ple another collection from Zanzibar: QADIRI DHIKR/ZANZIBAR 1. In the handlist of the
Zanzibar National Archives collection, the same text is attributed to Shaykh Uways him-
self. I have been unable to verify this. L. Declich, The Arabic Manuscripts of the Zanzibar
National Archives: A Checklist, Supplemento No. 2 Alla rivista degli Studi Orientali, Nouva
Serie, Vol. LXXVIII, Pisa/Roma (Instituti Editoriali e Poligrafica Internazionali), 2006, 81.
66 P.A. Caplan, Choice and Constraint in a Swahili Community. Property, Hierarchy and
Cognatic Descent on the East African Coast, London (Oxford University Press), 1975, 84–99.
 chapter 4

The Shādhiliyya in Northern Madagascar

c. 1890–1940
The Planting of a Garden and the Growing of Malagasy Roots

I have planted a small garden in the Antankarana mountains.

If it is not tended, it will wither away.1

∵
The diffusion of the ṭarīqa Shādhiliyya in northern Madagascar is closely con-
nected to the general Islamization of the Antankarana people (“the people
of the [coral] stones”, meaning the Ankarana mountains with its remarkable
tsingy rock formations)2 and with Comorian migration to the region. The
spread of the Shādhiliyya to Madagascar is also closely connected to religious
change taking place in Zanzibar and the Comoros, propagated by existing and
long-standing networks of scholars.
Previous research by M. Lambek and A. Walsh has addressed the
Islamization of the Antankarana and this process will not be reiterated in
detail here.3 Rather, this chapter shows how the spread of the Shādhiliyya ties
in with the emerging daʿwa orientation of the Sufi orders, whose activities in
northern Mozambique was outlined in the previous chapter. Members of the
same network also figure largely in the history of the Shādhiliyya in northern
Madagascar.

1 Quote: ʿUthmān b. ʿAbd al-Laṭīf (see below).

2 For an early ethnography of the Antankarana, see G. Ferrand, Les Musulmans a Madagascar.
In line with colonial notions of Islam, Ferrand saw neither the Antankarana nor their Sakalva
neighbours as “real Muslims”. In his view, they were “unconvertible” to any of the world reli-
gions. The only “real” Muslims in north/west Madagascar, according to Ferrand, were the
resident Arabs, Swahilis and Indians.
3 A. Walsh, Constructing “Antankarana”. History, Ritual and Identity in Northern Madagascar,
PhD Dissertation, University of Toronto, 1998; M. Lambek and A. Walsh, “The Imagined
Community of the Antankarana: Identity, History and Ritual in Northern Madagascar”,
J. of Religion in Africa, 27, 3 (Aug. 1997), 308–333; A. Walsh, “What makes (the) Antankarana
Antankarana? Reckoning Group Indentity in Madagascar”, Ethnos, 66:1, 2001, 1–22.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_��5

the shādhiliyya in northern madagascar 73

Islam in Northern Madagascar

The Islamization of northern Madagascar has been described in earlier research

as a result of consecutive “waves” of immigrants, including Arab immigration
that led to the development of the sorabe (“big writing”) script.4 In the first half
of the nineteenth century, the kings of the Antankarana were in constant con-
flict with the Merina of the plateau.5 By this time, the Antankarana, through
contact with trading Arabs, Comorians and Zanzibaris, were well aware of
Islam, and in some cases also converted. The majority of the population, how-
ever, remained with their native Malagasy customs and traditions, including
the practice of reburial.6 The Antankarana king Tsialana I accepted Merina
overlordship, but his son and successor Tsimiaro revolted against Merina rule.
In the ensuing conflict, Tsimiaro and his people took refuge in one of the many
caves in the Ankarana mountain range. According to popular tradition, the
cave was beleaguered by the Merina, and at this point the king made a promise
that if they made it out of there, he would embrace Islam and make his people
Muslim too.
This pledge, according to tradition, was made on the 27 August 1840, when
a complete solar eclipse passed over Madagascar. As the king and his family
looked out from the cave, the Merina soldiers, believing that it was night, were
asleep, and the king and his retinue escaped unharmed. After their miracu-
lous escape, the royal family retreated to Nosy Mtsio (an island south-west of
Diego Suarez, north-east of Nosy Be), where the king remained until his death
in 1882.7
Prior to these events, the Sakalava had appealed to Sayyid Saʿīd b. Sulṭān in
Zanzibar for assistance against the Merina. In 1838, Sayyid Saʿīd sent a small

4 K. Versteegh, “Arabic in Madagascar”, Bulletin of the School of Oriental and African Studies,
Vol. 64, 2, 2001, 177–187. See also L. Munthe: La tradition Arabico-Malgache vue à travers le
manuscript A-6 de Oslo et d’autres manuscript disponibles, Antananarivo, 1982.
5 Cassam Aly, “Islam dans le Nord de Madagascar”. Unpublished paper, ND. It is unclear where
Mr. Cassam Aly got this information from: presumably partly oral history and partly general
French history writing.
6 G. Ferrand, Les Musulmans, Vol. 3.
7 This narrative differs somewhat from that related by P. Verin, also based on oral sources.
According to Verin, Tsimiaro took refuge in a mountain cave with his retinue, including the
8-year old queen of the Sakalava. He stayed there for more than a year, but was eventually
betrayed by an Antankarana from a rival clan, and then settled at Nosy Mtsio. P. Verin, The
History of Civilization in North Madagascar, A.A. Balemka/ Rotterdam/Boston, 1986, 134. The
same story is recounted by Gueunier, Les chemins de l’Islam, 54.
74 chapter 4

naval force and even established a fort at Ambavatobe facing Nosy Be,8 but
the troops were insufficient to make any difference in the military situation in
either Sakalava or Antankarana territory. However, a handful of students were
sent to Zanzibar to be trained in Quranic schools, and to return to northern
Madagascar as Imāms. The question of what type of Islam these kings con-
verted to is not addressed, although it is likely that it was the Islamic practice
of the seafaring Comorian, Arabs and Zanzibaris, i.e. Shāfiʿī-Sunni in orienta-
tion. The Diego Suarez local historian C. Aly writes that the Islamic practice to
which the Antankarana now turned was “formés à Zanzibar,”9 in other words a
classic Indian Ocean pattern.
The French colony of Madagascar was declared in 1896, and the joint gover-
nance of Madagascar with the Comoros led to even closer contact between the
Comoros and the coastal cities of northern Madagascar. As has been pointed
out by earlier research, the mass conversion of Antankarana only took place
at this point in time, i.e. after 1896.10 Comorians who arrived after coloniza-
tion to teach Islam were generally referred to as fondys, teachers of the Quran.
Furthermore, Antankarana people who had taken refuge in the Comoros (par-
ticularly in Moheli) during the Merina wars, now returned with knowledge of
Islamic practice gained there.

Family, Religion and Trade on Madagascar: East African-Comorian

Networks and the Shādhiliyya

As M. Lambek has aptly noted, the Comoro Islands form the “bridge between
the coastal East African and the Malagasy world.”11 The main Comorian set-
tlements in northern Madagascar were Diego Suarez (Antsiranana)12 in the
north, and the island of Nosy Be further southwest. The city of Ambilobe some
120 kilometers south of Diego Suarez also saw a significant Comorian

8 R. Decary, L’Ile de Nosy Be de Madagascar. Histoire d’une colonization, Paris (Editions

Maritime et d’Outre-Mer), 1960.
9 Cassam Aly, “Islam dans le Nord”, 4.
10 I. Said, M.A. Chamanga, N.J. Gueunier, “Un Qasida Arabe en l’honneur de Sayyid Ahmad
al-Kabir, missionaire de l’Islam a Madagascar de 1896 a 1919”, Études Océan Indien, 1, 1982,
137–148.
11 M. Lambek, Knowledge and Practice in Mayotte. Local Discourses of Islam, Sorcery and
Spirit Posession, Univ. of Toronto Press, 1999, 34.
12 The city is today known both by its colonial name (Diego Suarez) and its Malagasy name
(Antsiranana). Here, I have chosen to use the colonial name, as this is the one that appears
in the Arabic sources.
the shādhiliyya in northern madagascar 75

­settlement.13 Finally, the city of Majunga on the west coast had the high-
est number of Comorian immigrants before the massacres and expulsion of
Comorians in 1976. By the time of the riots, Comorians constituted about 1/3
of the population in Majunga.14 Conversely, a substantial Malagasy popula-
tion existed – and still exists – on the islands of Mayotte and Mohéli, mostly
the result of immigration from Malagasy speakers from northern Madagascar.
Only a few Malagasy settled permanently on Anjouan or Grande Comore.15
The pattern of Comorian migration and settlement in Madagascar was
already well established by the mid-nineteenth century. From the Comorian
side, we have already seen the account by Burhān Mkelle, who estimated some
30,000 Comorians in Madagascar by the late 1920s (See above, Chapter 2).
Mkelle gives further details on the Comorian traders on Madagascar who
traded goods from the “Big Island” onwards to Zanzibar and beyond:

The Comorians in Madagascar were traders, starting small from that early
date and growing little by little, until it (the trade) was big. From about a
hundred years ago, it expanded much. Among the traders were the shaykhs
Ṣāliḥ b. Muḥammad Laytī, Ījhād b. Bakr and Ṣāliḥ b. Muḥammad Manūbī,
Sayyid Bakr, Sāʾib Mitshayū (?), ʿAbd Allāh Falāmbū who had posessions
in Tamatave and Mājid b. Saʿīd b. ʿAbd Allāh al-Mwinyi al-Murūnī and
others trading in small goods. Due to their business, they accumulated
great wealth (tharwāt ʿaẓīma). Very little remains of this (wealth) today.
That is because the Comorians were diligent in their work and energetic,
but they did not know the skill of managing their fortunes. Rather, they
gained (wealth) through hard work, but spent it extravagantly until their
businesses were depleted.
In this trade they used big sailing ships (sufun shirāʿiyya) which carried
products from Madagascar to Zanzibar, Lamu, Pate and other places in
East Africa, and from there to the Comoros and back to Madagascar.
When their trade deteriorated, they sold the ships and that was the end of
their trading empire in Madagascar.16

13 For figures on Comorian settlements in these locations, see N.J. Gueunier, Les Chemins de
l’Islam à Madagascar, Paris (L’Harmattan), 1994, 37–39. Based on the census from 1955,
Gueunier lists 5,500 Comorians in Diego Suarez, 3,000 in Nosy Be, 27,000 in Majunga and
2,900 in Ambilobe.
14 On the Comorian community in Majunga, see contributions in special issue of Études
Océan Indien, 28/39, 2007.
15 M. Lambek, Knowledge and Practice in Mayotte, 36.
16 MKELLE/ZNZ-2, 17.
76 chapter 4

Mkelle is noticeably unforthcoming with dates in his exposition of Comorian

trade in Madagascar, but from the general context of his narrative, it seems clear
that this took place in the first half of the nineteenth century. He is also silent
on the nature of their trade. From the Comorian perspective, Madagascar was
the place to buy meat (dried zebu meat, known as nyama la Bukini), a sweet
known as lado, and vanilla.17 Unfortunately, Mkelle is equally silent concerning
the Comorian ʿulamāʾ who went to Madagascar. Among the names omitted by
Mkelle is ʿAbd Allāh b. Aḥmad b. Hārūn Jamal al-Layl, who was introduced in
Chapter 2 under the name Ṣāḥib al-Ṭuyūr (Master of Birds). Although he may
primarily have gone to Madagascar for trade, the miraculous event narrated
about his death suggests that the journey also had a religious dimension.
Several other Comorians of Ḥaḍramī ʿAlawī background followed the same
pattern and ventured on to Madagascar. According to the ʿAlawī “head office”
in Singapore, the offspring of the Āl Shaykh Abī Bakr b. Sālim, Jamal al-Layl and
other ʿAlawī lineages by 1933 could be found in Diego Suarez, Nosy Be, Tulear,
Mahjunga and other places.18 As we saw in the previous chapter, Abū ’l-Ḥasan
Jamal al-Layl, who was active in spreading the Qādririyya to Mozambique,
was born in Madagascar. He is thus yet another example of ʿAlawī migration
to Madagascar, although we do not know if his family came directly from
Ḥaḍramawt or via the Jamal al-Layl pattern of Pate-Lamu-Comoro Islands.
What we can say is that Arab (mostly Ḥaḍramī) migrants to Madagascar in
the latter half of the nineteenth century arrived via the Comoros, and that
these were more closely tied in with the general Indian Ocean network, both
in terms of genealogy, trade and organized religious practice.
Overall, the outline of these connections indicate a network which held
both political and religious power in the Comoros, and which operated net-
works that extended onwards to East Africa and Arabia. The connection to
Madagascar should be viewed as a southern extension of an established trade-
and religious pattern operated not only by Ḥaḍramī ʿAlawīs but also by East
African traders and religious scholars in general.
Contact between the Antankarana royal family and the Sultanic families of
Grande Comore can also be evidenced. For example, around 1885, Tsimaninyi,
the son of Tsialana II (who had succeeded his father Tsimiaro in 1883) arrived
in Grande Comore as a guest of Sultan Sayyid ʿAlī. The latter was then in the
midst of his struggle for paramount power on the island, and Tsimaninyi even
fought with him on one occasion.

17 Interview, Maalim Muhammad Idris Muhammad Saleh, 17.11.2007.

18 Muḥammad b. ʿAbd al-Raḥman b. Shihāb, “Al-Sādat al-ʿAlawiyya”, 24.
the shādhiliyya in northern madagascar 77

Saidina b. Umar in Madagascar

The research by N.J. Gueunier describes the arrival of a Comorian “aristo-
crat” referred to as Saidina who introduced the ṭarīqa Shādhiliyya among the
Antankarana people some time in the late nineteenth century.19 On the back-
ground of Shādhilī expansion in Zanzibar, the Comoros and Mozambique,
this offshoot to Madagascar is not surprising. More surprising, however, is the
information that Saidina, upon his arrival in Madagascar, was given a tempo-
rary position as “local governor” by the French administration. The prominent
position given to Saidina indicates that this was no ordinary travelling teacher,
but someone known to the French. Most likely, he was the brother of Sultan
Sayyid ʿAlī in Grande Comore.20
Nothing is known of exactly what Saidina taught in northern Madagascar,
nor of how long he remained. What can be said, though, is that his presence
alone must have boosted Comorian-Antankarana relations at the royal level.
It is also possible that his presence added prestige to the Shādhiliyya. What
seems certain is that the spreading of the Shādhiliyya started in Madagascar
fairly soon after its introduction into the Comoros by Muḥammad al-Maʿrūf.
The history of Saidinia also indicates that the process was led by a network of
ʿAlawī sāda originating from Ḥaḍramawt.

ʿUthmān b. ʿAbd al-Laṭīf: The “First Flower” Planted in the

Antankarana
According to the manāqib written about him, al-Maʿrūf was at one point forced
to flee Anjouan following a confrontation with the less-than-sympathetic
chief qāḍī of the island who explicitly forbade the population to participate in
the Shādhilī dhikr. At this point, al-Maʿrūf escaped to Nosy Be on board an

19 N.J. Gueunier, Les Chemins de l’Islam, 55–56.

20 As mentioned above Sayyid ʿAlī had received the Antankarana king Tsimaninyi in Grande
Comore and it is likely that the arrival of Saidina came out of this contact, possibly sup-
ported by the French authorities. However, a more direct identification of Saidina is
more difficult to obtain, especially if we treat the name as a title rather than as a proper
name. The Swahili Chronicle of Ngazija does mention a brother of ʿAlī, namely ʿUmar
who later became ruler of Anjouan but it does not say anything about him having spent
time in Madagascar. Another possible candidate for having been the “Saidina” is Sayyid
ʿAlī’s maternal uncle, Hāshim, who eventually turned against his nephew in the quest
for supremacy and was exiled to Madagascar in the mid-1880s. He was given a French
pension of 6000 Francs but eventually made his way back to Grande Comore where he
was killed in battle with his nephew. S. Blanchy, “Images Coloniales”; L. Harries (ed.), The
Swahili Chronicle of Ngazija, 65, 71–77, 123–125.
78 chapter 4

Indian ship.21 Nosy Be had at that time been under French control since
1841, governed by military commanders and reporting to the government of
Reunion.22 However, the hagiographic account says nothing about the dura-
tion of al-Maʿrūf’s sojourn on Madagascar, nor does it say anything about
whether he spread the ṭarīqa there (or even if he ventured beyond Nosy Be).
According to the formal silsila of the present-day ṭarīqa Shādhiliyya of
Northern Madagascar, the first teacher to seriously start spreading the order
among the Antankarana was not Muḥammad al-Maʿrūf, nor Saidina, but a
certain ʿUthmān b. ʿAbd al-Laṭīf who arrived from Anjouan. According to the
present-day head of the order in Ambilobe, ʿUthmān had been initiated into
the order in Anjouan and been given an ijāza by al-Maʿrūf himself.23 No date
is given for his arrival, but based on what we know about the movements and
activities of al-Maʿrūf, we may assume this to have been some time in the early
1890s. No further identification is given for ʿUthmān b. ʿAbd al-Laṭīf, except that
he was one of Muḥammad al-Maʿrūf’s students or companions there, and that
he came to Madagascar on a daʿwa mission. He may even have been a relative.
According to the history of the Malagasy Shādhiliyya, ʿUthmān did not stay
long, but returned to Anjouan. Upon departure, he reportedly stated that he
had “planted a flower” [in some versions: “a small garden”] in the Ankarana
(mountains, metaphorically referring to the people). He added that if this gar-
den was not tended to, it would “wither away.”24
It should be emphasized here that ʿUthmān b. ʿAbd al-Laṭīf is referred to as
the first to spread the Shādhiliyya among the Antankarana – not in Madagascar
in general, nor among the Comorian communities in the region. Thus, one
interpretation is that Muḥammad al-Maʿrūf, on his previous visit, limited him-
self to the established Comorian communities in northern Madagascar, most
likely even only Nosy Be. This has implications for the interpretation of the
ways in which the order became structured into ethnic branches, Comorian
and Malagasy. Today, the Shādhiliyya as practiced in northern Madagascar is
perceived as being a Malagasy (specifically: Antankarana) brotherhood, whose
khalīfas are of Malagasy origin. The need to emphasize the first ­diffusion
among the Antankarana themselves, is probably the reason why ʿUthmān b.
ʿAbd al-Laṭīf is presented as the first khalīfa in Madagascar, rather than
al-Maʿrūf himself, or the Comorian royal Saidina.

21 P. Guy and A. Cheich Amin, La Vie et l’Oeuvre, 37.

22 R. Decary, L’Île de Nosy Bé de Madagascar, 21.
23 Interview with Abduh Samad, 27.07.08.
24 Interview with Abduh Samad, 27.07.08.
the shādhiliyya in northern madagascar 79

Aḥmad al-Kabīr: The Great Shaykh of the Ṭarīqa Shādhiliyya of

Northern Madagascar

All the leaders interviewed on Madagascar (whether of Malagasy or Comorian

background) universally agreed that the next khalīfa was Sayyid Aḥmad
al-Kabīr, the maternal uncle of Muḥammad al-Maʿrūf. Aḥmad al-Kabīr was one
of the sons of Mwinyi Mkuu Sultan Aḥmad, and is recorded in the manāqib as
Aḥmad b. Mwinyi Mkuu Aḥmad al-Saqqāf, also known as Abū al-ʿArab.25 He
was thus the maternal uncle of Muḥammad al-Maʿrūf, and according to Jean-
Claude Penrad, he had previously accompanied al-Maʿrūf during the latter’s
studies in Zanzibar in the 1870s.26 According to the manāqib of al-Maʿrūf, he
was given the nickname “al-Kabīr” on account of his physical size.27
According to the manāqib, Aḥmad al-Kabīr was specifically instructed by
al-Maʿrūf to spread the order in northern Madagascar. This contrasts with the
information given by Said, Chamanga and Gueunier who state that Aḥmad
al-Kabīr was brought to Madagascar by the Antankarana royal family and
specifically by the king Tsialana II,28 in order to lead the new Shādhiliyya
­movement.29 This, in turn, contrasts with the oral tradition mentioned above
that Aḥmad al-Kabīr was entrusted the “tending of the garden” in Madagascar
by ʿUthmān b. ʿAbd al-Lāṭif. Again, the different narratives are best interpreted
as expressions of the parallel diffusion of the order; a Comorian angle would
emphasize the role of al-Maʿrūf whereas the role of Tsialana II, or of ʿUthmān b.
ʿAbd al-Laṭīf would mark the Shādhiliyya as a Malagasy order.
Wherever the instruction came from, it is certain that throughout northern
Madagascar Muslim communities, Aḥmad al-Kabīr is the one who is univer-
sally remembered as the main proponent of Islam in general and of the ṭarīqa
Shādhiliyya in particular. To his name are attached stories of karamāts (mira-
cles) and his name is the one best known also to people outside of the order.

25 On the missionary activities of Aḥmad al-Kabīr among the Antankarana, see Chanfi
Ahmed, “Networks of the Shādhiliyya”, 328–329.
26 See Penrad, “La Shâdhiliyya-Yashrûṭiyya”, 388, for family tree. Penrad based his compi-
lation on conversations with the great-grandson of Aḥmad al-Kabīr. Note that Aḥmad
al-Kabīr was also the paternal uncle of Sayyid Manṣab b. ʿAlī, a well-known scholar and
reformist figure in Zanzibar (see below, Chapter 8).
27 P. Guy and A. Cheih Amin, La Vie et l’Oeuvre, 90.
28 Tsialana II reigned 1883–1924. He allied himself to the French against the Merina.
29 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 138.
80 chapter 4

A qaṣīda in Arabic his praise was published as a leaflet in 1966, in Arabic in the
Latin script, and with a Malagasy and French translation.30
According to the brief biography prefacing the qaṣīda in his praise, Aḥmad
al-Kabīr arrived in Madagascar in 1896.31 At first, he is said to have stayed in
Ambatoaranana, the place for residence of the Antankarana kings, which would
support the theory that he came at the invitation of the king. He then returned
to the Comoros after a brief period, whereupon he returned to Madagascar in
1906. On this second arrival, he was accompanied by his son ʿAbd al-Raḥmān
al-Saqqāf, who was then about 15 years of age.32 Again, according to the biogra-
phy prefacing his qaṣīda, Aḥmad al-Kabīr remained close to the Antankarana
royal family.33
On his second arrival, Aḥmad al-Kabīr was reportedly accompanied by
about twenty fondys, i.e. Islamic teachers.34 Some of these are named in the
biography; Sayyid ʿAbd Allāh b. Sultan, Shaykh ʿUthmān b. Shaykh Muḥammad
Bakari from Anjouan and Sayyid Muḥammad Jamal al-Layl.35 Although no
other biographical information is given, the collection of names of names indi-
cate a network of ʿAlawī sāda of the Shaykh Abū Bakr b. Sālim and the Jamal
al-Layl clans, originating in the Comoros but almost certainly with strong fam-
ily and Shādhilī links onwards to the East African coast.
According to the present-day shaykhs of the order, Aḥmad al-Kabīr did not
stay in any one place, but moved about from Nosy Be to Ambilobe to Diego
Suarez. However, the shaykhs all state that Aḥmad al-Kabīr spent much of his
time with the royalty of the Antankarana, as a teacher of Islamic practice and
recruiting members of the royal family to the ṭarīqa.
Aḥmad al-Kabīr returned to Anjouan in 1917 and died there 1919.36 Before
departing, he installed his companion Sayyid ʿAbd Allāh b. Sulṭān as khalīfa.

30 Said, Chamanga, Gueunier, “Un Qasida Arabe”. It should the noted that the qaṣīda itself
was rendered in the Latin script, as there existed at that time no printing facilities for
Arabic letters in Madagascar. The poem consists of 18 lines ending in nā. It starts: “Aḥmad
al-Kabīr shaykhunā/murshidunā wa-ghawthunā”. A full translation into French is given
by Said, Chamanga and Gueunier.
31 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141. This is supported by Casam Ali,
Islam, 6.
32 Interview, Abd al-Rasul, Ambilobe, 27.07.08.
33 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141.
34 Cassam Aly, Islam, 4.
35 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141.
36 J.-C. Penrad, “La Shâdhiliyya-Yashrûtiyya”,
the shādhiliyya in northern madagascar 81

However, Sayyid ʿAbd Allāh passed the powers directly on to ʿAbd al-Raḥmān
al-Saqqāf who “seized the powers his father had left behind.”37
By all accounts, Aḥmad al-Kabīr was a daʿwa scholar, mainly preoccupied
with teaching. Exactly how this process took place, is more difficult to assess,
as I could not obtain any written sources with the present-day order. What is
certain is that the adoption of Islam by Antankarana royalty took place within
a ṭarīqa framework. It is also relevant to compare the process of Islamization
in Northern Madagascar with that of the rest of East Africa, whereby reference
to scriptural sources partially came to replace authority based on inherited
position and kinship. Said, Chamanga and Gueunier, has described a related
process in Madagascar, for example with regard to the pre-existing Malagasy
burial rites. These came to be integrated into the 7-day and 40-day Islamic
prayer, thus integrating old beliefs with new ones and opening for new groups
to self-identify as Muslims.38 However, as was the case in Ilha de Mozambique,
and as P. Caplan showed for Mafia Island,39 this did not mean the creation
of a Muslim Shādhilī community that entirely superseded older kinship and
linguistic lines.

ʿAbd al-Raḥmān al-Saqqāf (d. 8 des. 1965): Successor to His Father

As mentioned above, ʿAbd al-Raḥmān al-Saqqāf was around 15–16 years of age
when he arrived in Madagascar together with his father, which should imply
that he was born some time around 1870.40 His mother was Comorian, and she
remained there. That meant that ʿAbd al-Raḥmān al-Saqqāf grew up to become
a Malagasy speaker of Comorian origin, what in Madagascar is referred to as a
zanatany, the term used for anyone born or raised in Madagascar.
Following the departure of his father, ʿAbd al-Raḥmān al-Saqqāf, known in
the region as “Said Sakafo”, took over as khalīfa of the Shādhiliyya in Northern
Madagascar. It was ʿAbd al-Rahmān al-Saqqāf who established the Shādhilī
headquarter in Ambilobe, and it was also he who firmly established the order
as a Malagasy organization, recruiting mostly from the surrounding popula-
tion of Malagasy-speaking peoples. Casam Aly further states that “Said Sakafo”
used “les methods soufis” to recruit Malagasys to the order, and that he opened
for a level of tolerance of local practices within the order. Among these were
for example the consumption of locally produced rum on ritual occasions such

37 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141.

38 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 138.
39 A.P. Caplan, Choice and Constraint.
40 Interview, Sh. Abd al-Rasul, Ambiliobe, 27.07.08. Sh. Abd al-Rasul was himself a student of
ʿAbd al-Raḥmān al-Saqqāf, until the latter’s death in 1965.
82 chapter 4

as weddings and funerals. According to some informants, he even insisted that

his own fondys drink rum in order to integrate with the people. However, on
the issue of local Malagasy burial rites (which are specific to the Antankarana
but similar to that of other Malagasy peoples and which include prolonged
funeral rites and reburial) he seems to have been less tolerant, and insisted that
the body be buried within the next day.
Al-Saqqāf died in Ambilobe on December 8th 1965,41 reportedly more than
90 years of age. The final years of his life were spent in Anjouan with his family
and brethren of the Shādhiliyya and he returned to Madagascar only shortly
before his death.42 His grave in Ambilobe is unmarked, and there is neither a
qubba (dome) nor a name to mark it. Nonetheless, a ḥawliyya (recitation) is
held every year in November to commemorate him.43 For this occasion, mem-
bers of the Shādhiliyya arrive from throughout Northern Madagascar and the
Comoro Islands.

A Malagasy order with Local Branches and Comorian Connections

After ʿAbd al-Raḥmān al-Saqqāf, the Shādhiliyya became “localized” in the
sense that the khalīfas that followed him were either Antankarana or, in some
cases, Comorians born in Madagascar.
The khalīfas of the Shādhiliyya of Northern Madagascar from 1965 to the
present, show the integration of the order with the Malagasy population:

SHĀDHILIYYA AMBILOBE, MADAGASCAR

ʿUthmān b. ʿAbd al-Laṭīf

Aḥmad al-Kabīr
ʿAbd al-Rahmān al-Saqqāf (d. 1965)
ʿUthmān Miyarana (Malagasy).
Muḥammad b. ʿAbd al-Rahmān (Father came from Anjouan).
Abduh Samad, Ambilobe (present-day khalīfa)

Figure 7 Shādhiliyya, Ambilobe, Madagascar

41 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141.

42 Interview, Shaykh Abduh Samad, 27.07.2008. According to Shaykh Samad, ʿAbd al-Raḥmān
al-Saqqāf spent the years 1961–1965 in Anjouan.
43 Interview, Shaykh Yusuf Abud.
the shādhiliyya in northern madagascar 83

Although the headquarter located in Ambilobe, and the main khalīfas after
1965, seem to indicate a firm “Malagasyzation” of the ṭarīqa, this does not mean
that all branches were equally deeply rooted as Malagasy. In a development
similar to that in Mozambique Island, Comorian control remained firm on
the coast. In Diego Suarez, the ṭarīqa has until the present been dominated by
people of Comorian origin, as indicated by the recent silsila of the local Diego
Suarez branch:

SHĀDHILIYYA DIEGO SUAREZ, MADAGASCAR

ʿAbd al-Rahmān al-Saqqāf (d. 1965)

Ali Nur. Comorian from Ngazija who returned to the Comoros.
Ali Daira. Comorian who returned to the Comoros
Abduh Nur. Comorian who returned.
Isa. Comorian born in Diego Suarez.
Yusuf Abud and Abdu Samash; both born in Madagascar of Comorian origin.

Figure 8 Shādhiliyya, Diego Suarez, Madagascar

The growing of parallel branches – “Comorian” and Malagasy – mirrors the

findings of L.J.K. Bonate in Ilha de Mozambique outlined in the previous
chapter. While L.J.K. Bonate’s findings indicate a gradual integration between
the Comorian “ṭarīqa agents” and local chieftains, the process in northern
Madagascar seems to have had the added dimension of a urban/rural divide
which also corresponded to a Comorian-Malagasy divide. One reason for the
persistence of Comorian leadership in the Diego Suarez branch may have been
the re-enforcement of Comorian scholarship through a stronger element of re-
migration than has been the case in Mozambique, which remained off-limits
for much of the civil war period.
The tendency towards parallel branches can also be detected in the biog-
raphy of Aḥmad al-Kabīr published in 1966. Recounting the history of Aḥmad
al-Kabīr’s relation to king Tsialana, it gives the names of the tree aforemen-
tioned scholars who accompanied him from the Comoros (Sayyid ʿAbd Allāh b.
Sulṭān, Shaykh ʿUthmān b. Muḥammad Bakari and Sayyid Muḥammad Jamal
al-Layl). The biography then goes on to list, with great care, the names of those
who attached themselves to the Shādhiliyya:
84 chapter 4

And these are the men who were with Sayyid Ahmad al-Kabir: Sayyd
Alid Faida, Sheikh Kasimu from Ambanja, shaykh Abdallah Pei, shaykh
Zakaria Mordady, Sayyid Ahmed Darwesh and shaykh Musa Ntsay from
Betamboho, shaykh Abdallah Sin from Betamboho, shaykh Muslim,
Sayyid Shafi from Ambohiboro. All these had taken the hand [taken the
Shādhilī initiation] and kept firmly to the word of Ahmad al-Kabir, and
were with him spreading the Shadhliyya brotherhood in this country of
the Ankarana. And whosoever was not attached to the silsila of Sayyid
Ahmad is lost.
Nuruddin Ali Sayyid said: “It is Sayyid Ahmad al-Kabir who was the
master of the Shādhilī brotherhood in this country of the Ankarana.”
Those whose names are written here are those who were with him,
and who remained in this country upon the return of Ahmad al-Kabir
to the Comoros in 1919, when power was given to Sayyid Abd Allah b.
Sultan. [. . .].44

It should be noted here that the author is at pains to emphasize the local
diffusion of the Shādhiliyya, and to ascribe authority only through the local
transmissions that originated from Aḥmad al-Kabīr. The listing of Malagasy
names, stressing their local belonging, indicate a need to root the history of the
Shādhiliyya post-Aḥmad al-Kabīr directly with the Antankarana themselves.
However, we do not know if this poem was actually written around 1966 or
if it had circulated for much of the twentieth century before reaching print.
The exact time of the Comorian/Malagasy divide can thus not be accurately
pinpointed.
The question of ethnic identity in the Sufi context is also related to the ques-
tion of Islamic identity in the Malagasy context, as indicated by the finding of
M. Lambek in among the Sakalava in Majunga.45 As M. Lambek has discussed,
Islamic identity (being “Silamo”) may in Madagascar just as well be interpreted
as a descent-based category rather than an exclusivist religious faith. The ṭarīqa
level of organization may be interpreted as yet another identity-defining crite-
ria which co-incides with either Malagasiness or Comorianness, depending of
branch belonging.

44 Said, Chamanga, Gueunier, “Un Qasida Arabe”, 141. Translated from the French version
prepared by Said, Chamanga, Gueunier.
45 M. Lambek, The Weight of the Past. Living with History in Mahjanga, Madagascar, Palgrave,
NY, 2002.
the shādhiliyya in northern madagascar 85

ʿUmar b. Sumayṭ in Diego Suarez in the 1920s: The ʿAlawī-Shādhilī

Connection and the Comorian Community in Northern Madagascar
The life of the Ḥaḍrami-Comorian-Zanzibari scholar and Chief Qāḍī ʿUmar b.
Aḥmad b. Sumayṭ (1886–1973) has been discussed in earlier research and
will not be discussed at length here.46 However, his sojourn in Madagascar
in the 1920s and 30s is a good example of the integration of Madagascar into
Ḥaḍramī-ʿAlawī-Zanzibari-Comorian networks. The part of ʿUmar b. Sumayṭ’s
career that was spent in Madagascar, is also yet another example of how
Ḥaḍramī-Zanzibari-Comorian-Madagascar connections built and reinforced
ṭarīqa-based Islam, and how these networks were kept up well beyond the
colonial era.
ʿUmar b. Sumayṭ first arrived in Diego Suarez most likely in 1926, having
returned to the Grand Comore to find his business there ruined. According
to one of his biographers, who clearly based his information on ʿUmar’s own
narrative of his life, Umar decided to go to Madagascar to perform daʿwa.
Although this may be a retrospective addition by his biographer, the addi-
tional comment makes sense for someone whose business had recently been
frauded: He would trade in a proper, Islamic way as a good, Muslim example.47
The choice of Diego Suarez, as should be clear from the history above, was not
altogether unexpected. He followed in a long and steady line of Comorians
(ʿAlawī or otherwise) who had ventured to Madagascar. By 1926, there were
also several of his fellow Comorians and also fellow ʿAlawī-Ḥaḍramī in the city,
among them the above-mentioned ʿAbd al-Raḥmān al-Saqqāf, son of Aḥmad
al-Kabīr, by then the khalīfa of the Shādhiliyya. The two reportedly spent much

46 ʿUmar b. Aḥmad b. Sumayṭ was the son the leading ʿAlawī scholar and qāḍī Aḥmad b.
Sumayṭ (1861–1925). He was educated by the scholars in Zanzibar, primarily in the
Madrasa Bā Kathīr, and in the Ḥaḍramawt. In the early 1920s, he spent time in the fam-
ily home in Grande Comore, and then went to Madagascar. In 1936, he became a qāḍī of
Zanzibar, and he was appointed Chief Qāḍī in 1943. He held this position until 1960. After
the Zanzibar revolution of 1964, he left with his family to Ḥaḍramawt, where he settled
with the Sumayṭ family there. Following the 1967 revolution of Yemen, he again relocated,
this time to the family home in Grande Comore. He died there in 1973. A.K. Bang, “My
generation. Umar b. Ahmad b. Sumayt (1886–1973): Inter-generational Network transmis-
sion in a trans-oceanic Hadrami Alawi family, ca. 1925–1973”, in L. Manger and M. Assal,
Diasporas within and without Africa – Dynamism, hetereogeneity, variation, Nordic Africa
Institute, Uppsala, 2006.
47 Ṭāhir Muḥammad ʿAlawī, Tarjama ʿUmar b. Sumayṭ, Typescript, ND, in family possession,
Moroni, Grande Comore. The staff that ʿUmar had left behind in Grande Comore had
mortgaged much of his property to pay off merchant bills.
86 chapter 4

time together.48 We can also assume that many of the fondys who had accom-
panied Aḥmad al-Kabīr were also still in the city.
In Diego Suarez, ʿUmar built a mosque on the site just above the present-day
harbour, on the site where the Mosqué Jacob stands today. According to oral
history,49 the mosque was built of wood and corrugated iron, and it had a living
quarter attached to it where ʿUmar b. Sumayṭ lived and from which he con-
ducted trade.50 He is said to have stayed together with two fellow Comorians,
Muḥammad al-ʿArīs and Aḥmad al-Ṭabīb. However, it was emphasized that
despite his close connections with the Shādhiliyya on the Comoros and his
family connections with several of its leaders in Madagascar, the mosque of
ʿUmar was not the centre for the Shādhiliyya order in Diego Suarez.51
Rather, a separate Shādhilī mosque in Diego Suarez was constructed some
time in the 1920s or early 1930s, on a piece of land (“a big hole” as it was related)
which was originally owned by one “Ashraf” (said to have been the name of
a man from Grande Comore). According to some informants, it was ʿUmar
himself who owned this land, and who gave it to the Comorian community,
but this cannot be verified from any written sources.52 What all agree though,
is that this happened later, after ʿUmar had built his own mosque in Diego
Suarez. The Shādhilī mosque, all informants agreed, was built by the entire
Comorian community of Diego Suarez, including ʿUmar b. Sumayṭ who was
also known to teach in the mosque. However, as presented above he is explic-
itly not included in the silsila of the Shādhiliyya of Diego Suarez.
In 1936, ʿUmar b. Sumayṭ left Madagascar to take up a qāḍīship in Zanzibar.
In other words, he spent almost 10 years of his life in Madagascar, and during
this time he also visited Antananarivo and “other parts of the interior”.53

48 Interview, Shaykh Abduh Samad, 27.07.2008

49 Interview, Bakar Momad, Diego Suarez, 26.07.2008 and Isa Aboudou Hussen, Diego
Suarez, 26.07.2008.
50 Interview, Ismail Hasan Jacob, 21.07.2008 and conversations with the Jacob family. The
site where ʿUmar’s mosque stood was taken over by Ismail Jacob in 1951 (the grandfather
of Ismail Hasan Jacob); hence the name of the mosque. The actual construction of the
present-day mosque took place in 1968, when in was built in concrete. The family origi-
nates from Kutch, and immigrated to Madagascar some time in the nineteenth century.
The present-day mosque is Ḥanafī.
51 T.A. Mohamed also state that ʿUmar b. Sumayṭ served as qāḍī of Diego Suarez:
T.A. Mohamed, Ahmad Qamardine, 338. Although this is not confirmed anywhere else, it
is certainly likely.
52 Interview, Bakar Momad, Diego Suarez, 26.07.2008.
53 Ṭāhir Muḥammad ʿAlawī, Tarjama ʿUmar b. Sumayṭ, 3.
the shādhiliyya in northern madagascar 87

Image 1 The Shādhilī mosque in Diego Suarez.

photo by the author.

ʿUmar b. Sumayṭ returned again one last time, as an old man. On this journey,
he was accompanied by his son-in-law, Muḥammad ʿAlawī Bin Amiyya (also
known as Bu Numay).54 The informants disagree as to when this last visit took
place, but it was certainly after ʿUmar’s departure from Zanzibar in 1965. It
could also have been after his departure from Yemen, i.e. in the early 70s, after
ʿUmar had settled in his final home in Grand Comore. A very likely guess is that
the visit came about in conjunction with an arbitration role that ʿUmar took on
in Ilha de Mozambique in 1972.55 The process was organized by the Portuguese,
and concerned ritual disputes between the traditional ṭarīqa leaders and the

54 See A.K. Bang, “My Generation”. ʿUmar had one daughter who died in childbirth. Her
daughter was raised in ʿUmar’s household. The granddaughter and her husband were
among those who left from Zanzibar with ʿUmar. Also: Personal communication,
Muhammad Idris Muhammad Saleh, 21.07.2008.
55 E.A. Alpers, “East Central Africa”, in: N. Levtzion and R.L. Pouwels, The History of Islam
in Africa, Oxford (J. Currey), 2000, 318. The parallel is striking to the 1949 intervention of
ʿUmar’s friend and colleague Abū ’l-Ḥasan Jamal al-Layl between the “Zukutis” and local
Qādriī and Shādhilī leaders in Malawi in 1949 (see above, Chapter 3).
88 chapter 4

new, emerging modernists. Not surprisingly, ʿUmar b. Sumayṭ favoured the

ṭarīqa leaders, and thus gave the Portuguese an “alibi” to support traditional
leaders in an effort to stem emerging, and potentially explicitly anti-colonial
radicalization.
ʿUmar b. Sumayṭ was then in his eighties, and already ill. Upon his last visit
to Diego Suarez, he stayed with the Shādhilī mosque and then subsequently
with the Jacob family. According to informants who were there, “all of Diego”
turned out to greet him, visit and talk to him. It was by all accounts, a last
return, a last meeting with his friends and companions in Madagascar.
From the point of view of scholarly history, his biographers seem to agree
that ʿUmar b. Sumayṭ’s sojourn in Diego Suarez was not an important one in
the sense of intellectual formation. The period is not even mentioned by his
main biographer, ʿAbd al-Qādir al-Junayd, who predictably focuses on ʿUmar’s
education in Ḥaḍramawt and Zanzibar.56 It is also clear that ʿUmar’s period
in Madagascar did not leave behind any direct traces, in the sense of a ṭarīqa
silsila or established teacher-students relationships. He was primarily there
for trade, and it is likely that this took up most of his time in Diego Suarez.
However, it is also clear that his presence was occasioned by the long-standing
connections between the Comoros and northern Madagascar, as well as by
family- and religious links.

Reform and the Emergence of a Malagasy Sufi Order

By the early twentieth century, the Antankarana people of northern Madagascar

region, through a combination of political and religious networks, had become
closely intertwined with religious developments taking place in the Comoro
Islands, Zanzibar the rest of coastal East Africa and even Ḥaḍramawt and
the Ḥaramayn. Implicit in this is the impact of Comorians (ʿAlawīs and non-
ʿAlawīs) on Islamic practice in the region, as agents who held the potential
to transcend the boundaries kinship and ethnicity. However, the process of
diffusion of the Shādhiliyya also shows a clear tendency whereby the social
organization of the order, in a manner similar to in Ilha de Mozambique, fol-
lowed kinship and ethnic lines. However, contrary to in Ilha de Mozambique,

56 ʿAbd al-Qādir b. ʿAbd al-Raḥmān b. ʿUmar al-Junayd, Nubdha ʿan ḥayāt al-Imām al-ʿārif
bi-llāh al-Ḥabīb ʿUmar b. Aḥmad b. Sumayṭ, MS in the collection of Maalim Muhammad
Idris. The Malagasy interlude is also omitted in the section on ʿUmar b. Sumayṭ in ʿAbd
al-Qādir b. ʿAbd al-Raḥmān b. ʿUmar al-Junayd, al-ʿUqūd al-Jāhiza wa ’l-wuʿūd al-nājiza,
Printed in India, ND (2009?).
the shādhiliyya in northern madagascar 89

the “rooting” of the Shādhiliyya among the Antankarana, and the creation of
an Antankarana silsila of authority, was carried through by a locally raised
Comorian, ʿAbd al-Raḥmān al-Saqqāf.
The inauguration of the Zanatany Mosque in the Soafeno quarter of Diego
Suarez (inaugurated on 23 August 2008) may be understood as a culmination
of this process. The name “Zanatany” indicates that this is to be the mosque for
all Malagasy Muslims (or at least all Sunni Muslim Malagasy), i.e. all Muslims
residing in Madagascar irrespective of origin, be that Comorian or otherwise.
As has been noted by N.J. Gueunier, the term zanatany literally means “child of
the land” (zanaka=child, tany=land), thus distinguishing Madagascar-born res-
idents of foreign (mainly Comorian) origin from immigrants. However, as also
indicated by Gueunier, the term in combination with the word “Muslim” has
come to denote a Muslim of any lineage who is born or raised in Madagascar –
thus something akin to a “Malagasy Muslim Community”.57 Indeed, one infor-
mant translated the term zanatany simply into the Arabic word jāmiʿa; general,
communal, for all Muslims. In a small city that boasts no less than three dif-
ferent “Comorian mosques”, in addition to the Shādhilī (mostly Comorian)
mosque (let alone the Ḥanafī and the different Shīʿa mosques), a zanatany
mosque may be interpreted as the final physical construction of a northern
Malagasy Shāfiʿī Islam – shaped by the Indian Ocean but in the 21st century
anchored firmly on Malagasy terra firma.

57 N.J. Gueunier, Les Chemins de l’Islam, 50.

 chapter 5

The Cape Town Muslim Community and

East African Sufi Networks
Beyond the Monsoon

This chapter traces the late nineteenth and early twentieth century daʿwa-
based, ocean-borne Islamic reform beyond the orbit of dhow shipping to its
southernmost point, that is to the Muslim community of Cape Town. In other
words, focus here is on the links between Cape Town and the Indian Ocean
networks of Islamic scholarship.

Islam in South Africa and Cape Town1

The first influx of Sufi teachings in the Cape region came from the Dutch
possessions in south-east Asia, notably in the form of exiles who had been
involved in resistance against colonial rule. Yūsuf al-Tāj al-Khalwatī was the
first such shaykh, deported from Batavia to the Cape in 1694, often referred
to as the year of Islam’s arrival to South Africa.2 Shaykh Yūsuf was a Sufi of

1 On what follows on the background of Sufism in the Cape region and in South Africa, see
J.A. Naude, “A Historical Survey of Opposition to Sufism in South Africa”, in: F. de Jong and
B. Radtke, Islamic Mysticism Contested. Thirteen Centuries of Controversies and Polemics,
Leiden (Brill), 1999; M. Haron, “Islamic Dynamism in South Africa’s Western Cape”, Journal of
Muslim Minority Affairs, 9:2, 1998, 366–372; M. Haron, “Da⁠ʾwah movements and Sufi tarīqahs:
Competing for spiritual spaces in contemporary south(ern) Africa”, Journal of Muslim Minority
Affairs, 25:2, 261–285; Y. da Costa, Islam in Greater Cape Town. A Study in the Geography
of Religion, PhD Thesis, University of South Africa, 1989: Y. da Costa and A. Davids, Pages
from Cape Muslim History, Shooter and Shuter, Pietermaritzburg, 1994, A.K. Tayob, Islamic
Resurgence in South Africa, Cape Town: University of Cape Town Press, 1995; A.K. Tayob, Islam
in South Africa. Mosques, Imams and Sermons, University Press of Florida, 1999; S. Hendricks,
Taṣawwuf (Ṣūfism): Its role and impact on the Culture of Cape Islam, MA Thesis, University
of South Africa, 2005; N. Green, Bombay Islam. The Religious Economy of the Western Indian
Ocean, 1840–1915, Cambridge (Cambridge University Press), 2011.
2 A celebration was held in South Africa in 1994 to mark the “300 years anniversary” of Islam
in the country. The career of Yusuf al-Maqassarī/Tuan Salamanka/Sjech Joesop/Yusuf of
Macassar is analysed from an Indian Ocean perspective by M. Feener, “Hybridity and the
“Hadrami Diaspora” in the Indian Ocean Muslim Networks,” Asian Journal of Social Science,
32:3, 2004, 353–372.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_��6

the cape town muslim community 91

the Khalwatiyya ṭarīqa, and before arriving in Cape Town he had studied in
Mecca. In Cape Town, Shaykh Yūsuf gave Islamic instruction to his followers
and a handful of others, all the while operating under strict laws that banned
Islamic practices completely. What he founded was in effect “a rudimen-
tary Muslim community”.3 Shaykh Yūsuf died in Cape Town in 1699, and his
grave soon became – and continues to be – a pilgrimage site for Muslims in the
Cape region.
Another influential early teacher was ʿAbd Allāh Qāḍī ʿAbd al-Salām, also
known as Tuan Guru (“Mister Teacher”), who died in 1806. He spent time
imprisoned on Robben Island before going on to found the first mosque in
Cape Town. That was the Awwal Mosque (“The First Mosque”), founded in
1798 in what was to become BoKaap, the Muslim quarter of central Cape Town.
The mosque also had a madrasa, where the still tiny Muslim community could
study Quran and religious texts.
The years following the death of Tuan Guru saw a dramatic expansion of the
Muslim community of Cape Town. By 1842 it had reached 6,400 people; by 1854
about 8,000. By 1867, the estimate was between 12,000 and 14,000.4 The num-
ber of mosques grew accordingly, and the Imāms connected to these mosques
became public figures. However, as has been pointed out by S. Jeppie, no formal
organization of Imāms took place in the nineteenth century.5 Furthermore, the
community was predominantly of south-east Asian origin, giving rise to the
notion of the “Cape Malay” as a distinct and bounded community.6 It was also

3 Fahmi Gamieldien, The History of the Claremont Main Road Mosque. Its People and their
Contribution to Islam in South Africa, Published by the Claremont Main Road Mosque, 2004.
4 S. Jeppie, “Leadership and Loyalties: The Imāms of Nineteenth Century Colonial Cape Town,
South Africa”, J. of Religion in Africa, XXVI, 2, 1996, 139–162.
5 S. Jeppie, “Leadership and Loyalties.” The Muslim Judiciary Council was only founded in 1945.
6 The notion of the Cape Muslim community as being in nature “diasporic” has undergone a
number of revisions. During the apartheid era, the “foreignness” of the Cape Muslims was
invoked to differentiate the Muslims from other communities, such as the local Khoisan.
The concept of “Malay” came under criticism by the 1980s and 90s, as the notion of “for-
eignness” came to be seen as obsolete and counterproductive to the anti-apartheid struggle
and in turn to the building of the new, non-racial South-Africa. See for example S. Jeppie,
“Re-classifications: Coloured, Malay, Muslim”, in: Z. Erasmus (ed.), Coloured by History,
Shaped by Place. New Perspectives on Coloured Identities in Cape Town, Cape Town (Kwela),
2001. However, diasporic consciousness resurfaced as a strategic resource in the later 1990s
as ties with Malaysia and Indonesia were re-activated. S. Bangstad, “Diasporic Consciousness
as a Strategic Resource: A Case Study from a Cape Muslim Community,” in: L. Manger and
M. Azzal, Diasporas within and without Africa – Dynamism, hetereogeneity, variation, forth-
coming, Uppsala, 2006.
92 chapter 5

a community destined to expand under colonial rule. Following the British re-
occupation of the Cape in 1806, restrictions on religious practice were lifted.
This, however, did not mean that the relationship between the Muslim com-
munity and the colonial authorities was not complex. From the colonial per-
spective, the Muslim clergy were conceived of as something akin to priests.
European descriptions are filled with words like the “chief priests” and “Malay-
Mohametan Church.”7 Aside from misrepresentations in an orientalist sense,
the vocabulary and tropes used to describe the Cape Muslim community over
time served to integrate the community into a colonial framework. However,
representations by others say little about the internal structure of power and
authority within the community.
Here, it seems that travel – and specifically travel in the search of knowl-
edge – came to be a marker of religious authority. The first known Cape Muslim
to perform the pilgrimage was one Gastordien who stayed in Mecca between
1834 and 1837.8 The early ḥājjis were held in great esteem upon return, although
their authority must have diminished as the number of pilgrims grew by the
1870s and 1880s. According to J.A. Naude,9 the returned students strengthened
taṣawwuf practices in the Cape region, manifested by the building of shaykhly
tombs and grave visitations. One such traveller was Ḥājji Manuel Bakar who
performed the pilgrimage in 1903–04, and received instruction from ʿAlawī
teachers both in Mecca and Zanzibar.10 Upon his return to South Africa, Bakar
remained in touch with his erstwhile teachers and maintained links with other
ʿAlawī centres of learning, particularly Grande Comore.
Teachers arriving from other places seem not to have been a very frequent
phenomena in the nineteenth century Cape Muslim Community. The best-
known teacher from overseas was Abū Bakr Effendī who was sent to the colony
in 1862 by the Ottoman sultan. Being a Ḥanafī of madhhab, Effendī attracted
a following to that school, which in turn caused rupture in the Cape Muslim
Community.
However, by the late 1800s a new breed of Muslim itinerant scholars had
found their way to the southernmost point of Africa. These were the daʿwa-
oriented scholars of different Sufi orders whom we have already seen at work
in Mozambique and Madagascar, preaching to both the converted and the
unconverted, and focusing especially on education. One Zanzibar scholar who

7 S. Jeppie, “Leadership and Loyalties”.

8 S. Jeppie, “Leadership and Loyalties”.
9 J.A. Naude, “A Historical Survey”.
10 Oral information: Muhammad Haroun, lecture in Bergen, 12.05.03.
the cape town muslim community 93

is reported to have visited South Africa was Muḥammad b. Aḥmad al-Murūnī

(al-Moroni). We have already encountered him in this book as the diligent
Zanzibari student who used to sit in the darsa of Abū Bakr Shaṭṭā in Mecca and
who was promoted to sit in the front with his teacher (See above, Chapter 2).
According to Farsy, al-Moroni spent a period in an unknown location in South
Africa, in combination with a sojourn in the Comoros.11
Another itinerant traveller came to Swellendam where he encountered a
young Muslim man named Muhammad Salih Hendricks.

Muhammad Salih Hendricks: From Periphery to Centre

to the Network

Muhammad Salih Hendricks was born in 1871 in Swellendam into a family of

recent converts to Islam.12 Swellendam was then a prosperous town, located
in the southwestern Cape. It was also a significant Christian mission station.
Hendricks’ grandfather had embraced Islam, and his father was relatively
wealthy, partly from operating the Swellendam-Cape Town postal route.13
According to family history, the young Muhammad Hendricks was originally
scheduled to study medicine in London, and money had been put aside for
this purpose.
However, in 1884/85, when Muhammad was about 15 years of age, an itiner-
ant Sufi teacher came to Swellendam, having learnt that there were a handful
of Muslim families there. This man was ʿAbd al-ʿAzīz al-Mālikī, of Moroccan
origin and member of a prominent family in Mecca. The life of ʿAbd al-ʿAzīz
al-Mālikī display all the typical features of the “travelling daʿwa scholar” of
the late nineteenth century; a Sufi orientation and an educational outlook
especially focused on the more peripheral regions. His pattern was to travel
between Port Elizabeth and Cape Town, preaching Islam to the Muslim com-
munities in the region.
It was ʿAbd al-ʿAzīz al-Mālikī who persuaded Muhammad Hendricks’ father
to give up the planned medicinal studies for his son, and instead have the
young man study the “medicine of the heart”, i.e. religion.

11 Farsy/Pouwels, The Shafiʿi Ulama, 78.

12 On the scholarly biography of Muhammad Salih Hendricks, see da Costa, Y. and Davids, A.,
Pages and Davids, A., The Mosques of Bo-Kaap, The South African Institute of Arabic and
Islamic Research, Cape Town, 1980.
13 Interview with Seraj Hendricks, Azzawiya Mosque, Cape Town, 10.03.05.
94 chapter 5

In 1888, at age seventeen, Hendricks travelled to Mecca, together with his

father. While there, his father died and was buried in Mecca. Muhammad
Hendricks then stayed on for no less than fourteen years, pursuing Islamic
learning with the same “scholars of the age” whom we have encountered in the
previous chapters.

Muhammad Salih Hendricks in Mecca: 1888–1902

The two men named as the main teachers of Muhammad Salih Hendricks
in Mecca are the same who played important roles for the ʿAlawī reformers
of East African Islam; ʿUmar b. Abī Bakr Bā Junayd and Muḥammad Saʿīd
BabṢayl (See above, Chapter 2). Hendricks studied with both for several years,
amongst others together with ʿAbd Allāh Bā Kathīr. Hendricks took the adhkār
and rātibs of the ṭariqa ʿAlawiyya from them. Specifically, he was taught Rātib
al-Ḥaddād, which by that time was becoming an integral part of ʿAlawī Sufi
practice throughout the Indian Ocean (See below, Chapter 7). He also became
familiarized with the by then substantial body of ʿAlawī pedagogic literature,
such as al-Risāla al-Jāmiʿa.
It is unclear who was his main shaykh in the Sufi sense. From the isnād
(chain of spiritual transmission) of his sons, we may gain a sense of affilia-
tion, if not actual transmission. The isnād is that of Abū Madyan (d. 1197), the
spiritual origin of the ṭarīqa Shādhiliyya.14 One derivate of the Abū Madyan
isnād spread to the Ḥaḍramawt, where it was passed on via Muḥammad b. ʿAlī,
known as al-Faqīh al-Muqaddam (see below, Chapter 7). This is the chain that
came to be known as the ṭarīqa ʿAlawiyya. The strong presence of the ṭarīqa
ʿAlawiyya is not surprising given that Hendricks clearly took most of his educa-
tion from the ʿAlawī shaykhs of Mecca. Some differences can be noted, how-
ever, between the isnād as passed on to the East African scholars (Sumayṭ, Bā
Kathīr) and Hendricks. The most noticeable absence in the Hendricks isnād
is ʿAlī b. Muḥammad al-Ḥibsḥī, the founder of the ribāṭ in Sayʾūn. Instead, a
prominent place is awarded to Aḥmad b. ʿUmar b. Sumayṭ, who, as outlined
in Chapter 2, was a central reformist teacher and also known for imparting
Islamic knowledge to women. The slight difference may reflect a difference in
emphasis as the lineage grew, especially over issues of who should and who
should not be given Islamic instruction.15

14 On the influence of Abū Madyan, see V.J. Cornell, The Way of Abū Madyan. The Works of
Abū Madyan Shuʿayb, Cambridge (Islamic Texts Society), 1996.
15 Interview, Seraj Hendricks, Azzawiya Mosque, Cape Town, 10.03.2005.
the cape town muslim community 95

Muḥammad
ʿAlī b. Abī Ṭālib
Abū Saʿīd al-Ḥasan
Abū Muḥammad Saʿīd al-ʿAjmī
Abū Sulaymān Dāwūd b. Naṣīr
Abū Maḥfūẓ Maʿrūf b. Fayrūz al-Karikhī
Abū ’l-Ḥasan al-Sirrī
Abū ’l-Qāsim al-Junayd d. 910
Abū Bakr
Abū Ṭālib al-Makkī Muḥammad b. ʿAlī
Abū Muḥammad ʿAbd Allāh b. Yūsuf al-Jawīnī (or Juwaynī)
ʿAbd al-Malik al-Jawīnī (or Juwaynī), Imām of the Ḥaramayn
Abū Ḥāmid al-Ghazālī
Qāḍī Abū Bakr Ibn al-ʿArabī
Abū ’l-Ḥasan ʿAlī b. Ḥarzham
Abū Yaʿazzā d. 1176
Shuʿayb Abū Madyan d. 1197
MUḤAMMAD B. ʿALĪ AL-FAQĪH MUQADDAM d. 1255
ʿAlawī b. al-Faqīh al-Muqaddam
ʿAlī b. ʿAlawī b. al-Faqīh al-Muqaddam
Muḥammad b. ʿAlī Mawlā al-Dawīla
ʿAbd al-Raḥmān al-Saqqāf
Abū Bakr known as al-Sakrān
ʿAbd Allāh b. ʿAydarūs b. Abī Bakr
Abū Bakr al-ʿAydarūs
Shihāb al-Dīn Aḥmad b. ʿAbd al-Raḥmān
Shaykh Abū Bakr b. Sālim, Ṣāḥib ʿInāt
Ḥusayn b. Abī Bakr b. Sālim
ʿUmar b. ʿAbd al-Raḥmān al-ʿAṭṭās
ʿAbd Allāh b. ʿAlawī al-Ḥaddād
Ḥāmid b. ʿUmar b. Ḥāmid
Aḥmad b. ʿUmar b. Sumayṭ
ʿAbd Allāh b. Ḥusayn b. Ṭāhir Ḥasan b. Ṣāliḥ al-Baḥr
ʿAydarūs b. ʿUmar al-Ḥibshī Aḥmad b. Abī Bakr b. Sumayṭ
Ḥusayn al-Ḥibshī ʿUmar b. Aḥmad b. Sumayṭ
ʿAbbās al-Malikī Muḥammad al-Malikī
ʿAlawī al-Malikī

Figure 9 Isnād of Abū Madyan/Shādiliyya/ʿAlawiyya of Muhammad Salih Hendricks.

96 chapter 5

Finally it must also be assumed that Muhammad Salih Hendricks became

familiar (if only by hearsay) with the ribāṭs in the Ḥaḍramawt and the formal,
structured way in which Sufism was taught there. In other words: There is
every reason to assume that Hendricks left Mecca filled with the same ideas of
reform – and specifically educational reform – that was current in the ʿAlawī
network of which he had become part.
To what extent he had been incorporated into the network is illustrated by
the fact that on his way back from Mecca he stopped for a year in Zanzibar.
According to family history, Muhammad Hendricks found himself somewhat
in a “vacuum” following his years in Mecca. His parents were dead and what
was left of his family and associates in the Cape region had turned against
him.16 He had no particular incentives to return to Swellendam, and when the
opportunity arose to take a one-year temporary qāḍīship in Zanzibar, he took
it. It is likely that this came about through the mediation of Hendricks’ teacher
in Mecca, ʿUmar b. Abī Bakr Bā Junayd and his fellow student, ʿAbd Allāh Bā
Kathīr or his teacher in turn, Aḥmad b. Sumayṭ.
Interestingly, Hendricks is also said to have taken part in the shaping of the
mawlid celebrations in Zanzibar. Mass events such as the mawlid, played a vital
role in propagating a more scriptural based Islam to a wider, sometimes only
nominally Muslim community, strengthening organizational lines according
to Sufi orders and/or specific mosques. It is at this point that Muhammad Salih
Hendriks enter the historiography of East African Islam. He is referred to by
Abdallah Saleh Farsy17 simply as a “shaykh” of the ʿAlawī tradition, a teacher
and a scholar. The fact that he was from Cape Town is not mentioned as any-
thing particular or peculiar, indicating yet again that in this kind of network,
distances were of little importance.

Return to Cape Town: Controversy, Continued Links and

Educational Efforts
Muhammad Salih Hendricks returned to Cape Town in 1903, together with his
wife Khadīja, a Ḥaḍramī lady whom he had married in Mecca. His marriage
to a Ḥaḍramī woman further demonstrates his incorporation into the ʿAlawī
network; marriage as a strategy to strengthen teacher-student and/or family
relations can be found in abundance in East Africa. However, as it turned out,
Khadīja did not adapt to Cape Town life and the marriage ended in divorce.
Hendricks’ initial reception in Cape Town was apparently cordial. When a
reporter from the South African News went to seek him out in his residence
in Wandel Street, he was guided to the house by a huge welcome-banner and

16 Interview with Seraj Hendricks, Azzawiya Mosque, Cape Town, 10.03.05.

17 Farsy/Pouwels, The Shafiʿi Ulama of East Africa.
the cape town muslim community 97

Image 2 Muhammed Saleh Hendricks.

courtesy of the azzawiya mosque, cape town.

bright, coloured lamps. According to the report, Hendricks was holding recep-
tion wearing “a rose colored brocaded gown and white turban”, surrounded
by “gray haired priests.”18 Continuing the Colonial Cape Town tradition of

18 South Africa News, ND (photocopied extracts) Approximately 02.07.1903.

98 chapter 5

d­ escribing the Muslim community in Christian terms, the interviewer went

on portray Hendricks as “a bishop” on the basis of his qualifications from
Mecca. Here was a new, relatively young leader of the Cape Muslim com-
munity. When asked whether he was really the first Islamic “bishop” of Cape
Town, Hendricks – according to the South Africa News – answered: “The first in
South Africa”.
The reactions were fast and furious. Less than a week later, a reader who
called himself “A Mohammedan” pointed out in a letter that in Islam there is
no such office as that of a bishop.19 The author added that he had “no quarrels”
with Hendricks, and that his remarks were “devoid of the personal element.”
Later contributors to the South African News were less apologetic. One reader
invoked the “disgracing of Islam by false Mahdis”, and called for Hendricks to
make clear the basis of his claim to authority.20 Another challenged him to
“make public his authority at a meeting of Imāms and other men of knowl-
edge so that we who follow the school of Shafee may be assured and stop the
rumour.”21 The element of rivalry soon came out. A few days later, “A Muslim”
wrote: “We Cape Muslims who hold professorships on the same lines as
Mr. M.S. Hendricks shall wait with great interest and anxiety for him to prove
his title.”22 Accusations were the next level: “We are told that Mr. Hendricks
is the first Mohammedan bishop in South Africa. This we contradict; to
answer this question affirmatively illustrates a tendency towards falsehood.”23
Whether the controversy was due to bad press-handling by Hendricks or by a
latent rivalry in the Cape Muslim community, his career as a teacher in Cape
Town did not get off to a good start.
Hendricks, without family and with few friends, started teaching the reli-
gious sciences in different mosques in the city, particularly in the Bo Kaap,
by then an established Muslim quarter. According to family history, Hendricks
was met with opposition regarding some of his teaching methods, such as his
program to provide Islamic learning to women.24 Further newspaper cam-
paigns were mounted against him and only a few people – notably converts
and the handful of families of Arab origin – stood by him in this period. It
is unclear what exactly were the charges against Hendricks, and detailed
research is needed to find if this was based in religious/ideological outlook or

19 South Africa News, 13.07.1903.

20 South Africa News, 14.07.1903.
21 South Africa News, 14.07.1903.
22 South Africa News, 17.07.1903.
23 South Africa News, 20.07.1903.
24 Interview with Seraj Hendricks, Azzawiya Mosque, Cape Town, 10.03.05.
the cape town muslim community 99

in more mundane conflicts. Family history simply remembers the period after
his return as an unhappy one for Hendricks as he moved from one mosque to
the next without much support.
Both Y. da Costa and A. Davids describe the Cape Muslim community of
the early twentieth century as one marked by strife and internal conflicts,
although the actual reason for the discord is unclear. From what can be gath-
ered, rivalries were based on he long-standing Ḥanafī/Shāfiʿī divide. However,
the discord was also personal, disrupting a community already under substan-
tial pressure from the colonial authorities. The most blatant example of inter-
nal conflict was the so-called “Jumaa Dispute” which came to a head in 1913. At
the core of the controversy was the question as to where Friday prayers should
be held, and who should act as imām.25
Hendricks resolved to put the jumʿa prayer dispute before his Meccan
teacher ʿUmar Bā Junayd. The latter responded by referring the matter to
Aḥmad b. Sumayṭ in Zanzibar, who in turn decided to send his disciple ʿAbd
Allāh Bā Kathīr to mediate in Cape Town. This process may be seen as another
example of a clear tendency towards social involvement of the ʿAlawī network,
involving themselves in matters far from their “home ground”. It is also yet
another, and earlier example of the role of the Sufi scholar as arbitrator, a func-
tion we have seen taken on by Abū ’l-Ḥasan Jamal al-Layl in 1949 and ʿUmar b.
Sumayṭ in 1972.
Bā Kathīr left Zanzibar some time in late 1913, accompanied by Rashīd b.
Sālim al-Mazrūʿī26 who was to act as an interpreter into English, as well as an
unnamed student of Bā Kathīr. Together they formed what in South African
literature is referred to as the “Ba Kathier delegation”.
The so-called Shāfiʿī Jumʿa agreement was signed on 27 Ṣafar 1332/24 January
1914,27 following a meeting attended by all but one of the Shāfiʿī imāms of the
Cape area. Here, the imāms agree to hold one jumʿa prayer in one (specified)
mosque, while the task of delivering the khuṭba (sermon) was to alternate
between the imāms who until now had led separate Friday prayers. However,
as it turned out, the compromise failed, the last imām refused to sign the docu-
ment, with the result that the community remained fragmented.
For his efforts, Bā Kathīr was offered a sum of money, which he, accord-
ing to Farsy, refused to accept. On his advice, and very much in the tradition
of the educationally oriented scholars of the ʿAlawī tradition, the money was

25 On the “Jumaa dispute”, see da Costa and Davids, Pages, 107–109.

26 Rashīd b. Sālim al-Mazrūʿī was liwālī of Takaungu just north of Mombasa and an influen-
tial member of the Mazrūʿī family. See Farsy/Pouwels, The Shafiʿi Ulama, 132.
27 The agreement is reproduced in da Costa and Davids, Pages, 109.
100 chapter 5

instead set aside to establish a madrasa in Cape Town, which was to be named
Madrasa Bā Kathīr.

The Azzawiya Mosque School of Cape Town

It is unclear whether this Bā Kathīr madrasa was ever established, or indeed
if Muhammad Salih Hendricks was even in charge of the funds. Instead,
family history remembers the years 1914–1919 as being times of further finan-
cial and personal struggle; Hendricks had neither money to set up his own
teaching institution, nor did his teachings necessarily win popularity in the
local community.
However, the educational zeal embodied in the reform-oriented milieu of
which Hendricks had become part, won out. By 1919, he completed the step
already taken by his Ḥaḍramī masters and by his fellow student ʿAbd Allāh Bā
Kathīr and established his own mosque-school in the Walmer Estate in Cape
Town. The institution was to be known as the Azzawiya (al-Zāwiya) Mosque.
Here, students were instructed in the various texts of ʿAlawī origin.28 The cur-
riculum included Rātib al-Ḥaddād which soon appeared in a translation into
Afrikaans (See below, Chapter 7), as well as Rātib al-ʿAṭṭās. Basic, Islamic edu-
cation for children was taught on the basis of al-Risāla al-Jāmiʿa, again a direct
derivate of the early Sufi reform of the eighteenth century ʿAlawiyya order and
of Zanzibar where the same text was being used in the Government Schools. A
translation into Afrikaans was completed, and since reprinted several times. In
addition, higher-level students were given thorough instruction in al-Ghazālī’s
Iḥyāʾ ʿulūm al-dīn, the main reference for ʿAlawī reformers’ emphasis on the
Quran and Sunna as the sole basis for the Sufi path.
Another element of Islamic practice introduced at the Azzawiya was the
mawlid celebration. Here, the model was clearly Zanzibari, reflecting the prac-
tice as it evolved there at the beginning of the twentieth century. The text
recited was the Mawlid al-Barzanjī by Jaʿfar b. Ḥasan al-Barzanjī (1690–1765).
This still is the most frequently recited poem both in Zanzibar and Cape Town
to celebrate the life and deeds of the Prophet Muḥammad.
Finally, the curricula of the Azzawiya also included specific teaching ses-
sions for women, including adult, married women. The issue of teaching
women was one that most clearly marked the Azzawiya as a first and foremost
a teaching institution, rather than a clear-cut mosque.29 The practice clearly

28 Interviews with Seraj Hendricks, Cape Town, March 2005, in Zanzibar, January 2006 and
in Cape Town, January 2009.
29 Interview, Seraj Hendricks, 22.08.2005.
the cape town muslim community 101

Image 3 The Azzawiya Mosque in Cape Town.

photo by the author.

points back to the eighteenth century reformers of the ʿAlawī ṭarīqa, includ-
ing Aḥmad b. ʿUmar b. Sumayṭ who used to teach women in special sessions
outside his mosque.

Other Travellers – More Daʿwa

Muhammad Salih Hendricks was a South-African born scholar who became

closely incorporated into the network of Indian Ocean reform. The reverse
traffic is also traceable in available sources: the daʿwa travellers who arrived
in Cape Town. A survey of the immigration records at the National Archives
of South Africa (Cape Town Repository) show a number of immigrants whose
names betray Ḥaḍramī-ʿAlawī or Meccan origin around the period 1910–1915.
These were a special breed of travellers, arriving in ports that until recently
had been off the circuit for Indian Ocean sailing but which became accessible
through the opening of regular steamer lines along the East African coast all
the way to Cape Town.
102 chapter 5

Sayyid Aḥmad b. ʿAbd Allāh al-Ḥibshī: Shibām-Cape Town

On the 22nd of February 1913, one Sayyid Aḥmad b. ʿAbd Allāh al-Ḥibshī stepped
ashore in Cape Town from the steamer S/S “General”.30 Aḥmad al-Ḥibshī had
embarked at Zanzibar, duly equipped with a passport issued by the British
Consul General of the Zanzibar Protectorate, the passport being dated
27 December 1912. Al-Ḥibshī travelled as a deck passenger, and he was 48 years
of age when he arrived in Cape Town. His stated profession was that of “busi-
nessman”, while also being referred to as a “priest”. According to the immigra-
tion paper, he was an “Arab” and he duly signed his name in Arabic script, being
(upon direct question) unable to sign in Latin script. According to his passport,
al-Ḥibshī was born in the city of Shibām in Ḥaḍramawt, Yemen. Interestingly,
his passport gives Al-Ḥibshī’s status to be that of “British Protected Subject”, i.e.
originally resident in one of the British colonies (possibly India, but in this case
most likely Aden). Al-Ḥibshī came to Cape Town for a reason, namely “to visit
co-religionists” – including two of his otherwise unidentified relatives, Ḥamza
and ʿAbd al-ʿAzīz, living in Rose Street in the BoKaap.
Upon arrival in Cape Town, al-Ḥibshī stated that he had spent one month
in Zanzibar, his destination all the while being Cape Town. Against a deposit
of 50 Pounds Sterling, the immigration department duly issued al-Ḥibshī with
a temporary permit to remain in the Cape Province for two months. When the
two months had passed, al-Ḥibshī applied for a one-month extension that was
promptly granted. On May 23rd 1913, he left Cape Town on board the steamer
“G. Woermann”, presumably returning to Ḥaḍramawt via Zanzibar.

Manṣūr b. ʿAbd Allāh al-ʿAlawī: Mecca-Cape Town

In April 1928, one Manṣūr b. ʿAbd Allāh al-ʿAlawī submitted an application
for temporary residence in the Cape Province.31 Like his predecessor above,
al-ʿAlawī’s stated purpose was to “visit co-religionists”, many of whom were
clearly well known to him through his role as pilgrimage guide in Mecca.
His application was supported by a letter from Mr. S. Dollie, Chemist and
Photographic Dealer of Cape Town, who stated that:

Sayyid Mansur was born in Mecca some 30 years ago. He is a direct

descendant of our holy Prophet Muhammad, is a Mohammedan bishop
of high standard, and is a guide for more than half of the total number

30 South African National Archives, Cape Town: KAB-IRC-1/1/347/987A: Immigration papers,

Sharif Ahmad b. Abd Allah al-Habashi.
31 South Africa National Archives, KAB-IRC-1/1/777–11220A: Immigration papers, Sayyid
Mansur Alavie.
the cape town muslim community 103

of pilgrims leaving South Africa annually. He has numerous personal

friends here, including myself.

Interestingly, Mr. Dollie adds that Sayyid Manṣūr’s father, too, had links to the
Cape Town Muslim community, dating back to the early years of the twenti-
eth century: “His father, Sayyid Abdallah al-Alawi, is a highly respected bishop
whose name still lives in the memory of the old Cape Malays. He had been to
South Africa some 25 years ago, and did much for the Cape Malays religiously.”
Sayyid Manṣūr had evidently prepared his visit in advance. After receiv-
ing a cable from him, Mr. Dollie had “called a meeting of Cape Malays”, all of
whom agreed that they were “anxious to have him out here.” One Mr. Gamiet
(Ḥamīd), most likely a participant in the same meeting, submitted a letter to
support the application, stating that: “Sayyid Mansur, whom I have personally
met in Mecca, is a direct descendant of our Prophet, and I can vouch as to his
being a person whose moral and intellectual qualities being of a high degree”.
In June 1928, Sayyid Manṣūr was granted a 6-month temporary residence in
Cape Town, arriving from Bombay on a passport issued in Jeddah. His address
was that of Mr. Dollie, in Castle Bridge. However, it appears that Sayyid Manṣūr
did not stay the extent of his visa. Already on the 31 of August 1928 he returned
to Bombay on the steamer S.S. Ellora.
Almost 20 years later, in 1947, Sayyid Manṣūr returned to Cape Town. Now,
he travelled on a passport issued by the Saudi state, but still with the chemist
Mr. Dollie as his guarantor. It also appears that Mr. Manṣūr had kept up his role
as pilgrimage guide for the Cape Muslims, while also working as an Islamic
teacher. Mr. Dollie was still vouching for Sayyid Manṣūr, stating that the lat-
ter was “an excellent type of person” who would conduct “religious services
among the Cape Malays.”

Sayyid Muḥsin b. Sālim al-ʿAydarūs (d. 1934): Mombasa – Cape Town

Born in Ḥaḍramawt, Sayyid Muḥsin had emigrated with his father to Mombasa
some time in the late nineteenth century.32 Coming of age in East Africa, he
established himself as a religious teacher. In 1909 he arrived in Cape Town,
where he settled for eleven years, returning to Mombasa in 1920. During this
period, he developed close contact with Muhammed Salih Hendricks.
Sayyid Muḥsin returned to Cape Town a second time in 1927, once again as a
religious teacher. During his seven-year absence from Cape Town, he is known

32 For what follows on Sayyid Muḥsin, see S. Hendricks, Taṣawwuf, 350–355 and A. Davids,
Slaves, Sheikhs, Sultans and Saints. The Kramats of the Western Cape, Unpublished MS,
unpaginated, in the possession of Seraj Hendricks.
104 chapter 5

to have founded a madrasa in Mauritius, in addition to one in Mombasa. Upon

his return to Cape Town, he also brought with him two of his fellow ʿAlawīs
(unknown if they came from Ḥaḍramawt, Mombasa or elsewhere: Sayyid
Muḥammad and Sayyid Ibrāhīm). The teaching consisted mostly of dhikri-
sessions taught and conducted in his home, or occasionally in the Azzawiya
mosque. The recitals were based on the ʿAlawī tradition, including Rātib
al-Ḥaddād recited every Thursday evening. However, Sayyid Muḥsin does not
appear to have initiated students into the ṭarīqa ʿAlawiyya. Rather, he seems to
have been an initiate of the Qādiriyya, and in retrospect the popularization of
the ṭarīqa Qādiriyya in Cape Town is attributed to him.
Like many of his contemporary ʿAlawīs traversing the Indian Ocean as teach-
ers and holy men, Sayyid Muḥsin was known in Cape Town as a performer
of karāmāt, extraordinary events or miracles. Amongst others, he reportedly
entered the cage of the lions in Rhodes Zoo, where the leader of the pride
advanced only to lay down meekly at the feet of Sayyid Muḥsin.
In addition to his teaching activities, Sayyid Muḥsin acted as a translator at
the Magistrates Court in Cape Town, translating from Arabic into English. As
has been pointed out by Seraj Hendricks, the apparent need for such transla-
tion indicate that quite a number of people of Arab origin sought residence in
Cape Town in the period ca. 1910–1935.33 Their travel patterns, as well as their
impact on Cape Islam has yet to be accounted for.
Sayyid Muḥsin died in Cape Town in 1934. His burial rites were conducted
by Muhammed Salih Hendricks in the Azzawiya Mosque. He was buried in
the Gamediyah Cemetary in the Observatory part of Cape Town. Shortly after
his death, a qubba (dome) was built over his grave, which is still standing. His
successors, the aforementioned Sayyid Muḥammad and Sayyid Ibrāhīm, con-
tinued his teaching, including the teaching of Qādirī practices.

Travellers in Reform and Their Audience

From their names and backgrounds is not difficult to identify teachers such as
Aḥmad al-Ḥisbhī, Manṣūr al-ʿAlawī and Muḥsin b. Sālim al-ʿAydarūs as repre-
sentatives of the sāda ʿAlawiyya. In a long tradition of ʿAlawī scholarship, and
by the 1920s also in a long history of Sufi-based reform, these travellers should
not be interpreted as mere “figureheads” whose perceived Sharīfian nobility
attracted the Muslims of Cape Town. Rather, they should be interpreted as
arriving for a specific reason, linked to specific disputes revolving around the
ever-present question: What is “correct” Islam and how is it best practiced in
our community? As such, they paralleled in South Africa a number of other

33 Seraj Hendricks, Taṣawwuf, 353.

the cape town muslim community 105

arrivals, notably from India, also proclaiming to hold answers to this question.
Travellers in what N. Green has called “Bombay Islam” is one example,34 bring-
ing to Natal (but also to Cape Town) their specific saints, practices and cultural
expressions.
The first question to be asked is why the travellers described in this chapter
chose to venture all the way to Cape Town, well beyond the long-established
routes of Indian Ocean scholarly Islam? As has been pointed out by N. Worden,
“Cape Town may not seem like an obvious place to be included in a study of the
Indian Ocean world.”35 The city seems, from a mere geographical point of view
to be too far south of the Arab/Swahili perceived cultural border and too far
west of any monsoon wind. However, as N. Worden also points out, and as this
chapter also has attempted to show, borders were not necessarily perceived as
obstacles, nor were trade winds the sole navigational aid for a traveling teacher.
Nonetheless, with reference to the spread of Sufism, Cape Town may indeed
be called an Indian Ocean port. Firstly, a feature of the late nineteenth and
early twentieth century was the increased tendency among students of Islam
to travel in order to obtain learning. This was also the case in the Muslim com-
munities in Cape Town, one prominent example being Muhammad Salih
Hendricks. Secondly, as mentioned above, Indian Ocean Sufi affiliations ran
deep in the Cape Islamic identity, with its origin in the Khalwatiyya and Shaykh
Yūsuf al-Tāj al-Khalwatī. Thirdly, in the late nineteenth/early twentieth cen-
tury, several steamship lines were established that trafficked the routes from
Aden, via Mombasa, Zanzibar and Beira to Durban and Cape Town, making
the trip to Cape Town accessible from several ports. Thus, the will, the back-
ground and the means were present.
We must nonetheless also ask how these links first came about. The exam-
ple of Sayyid Manṣūr above is poignant, as it indicates a pattern that very much
reflects the process in Madagascar whereby a son follows the travel pattern of
his father, and draws on the prestige of his father upon arrival. However, what
motivated the first traveller (of this particular family) i.e. Sayyid Sayyid ʿAbd
Allāh al-ʿAlawī, the “highly respected bishop”? For example, we do not know
who his host was, or how the journey came about. What we do see is that, like
in Mozambique, the need for a local “patron” was essential, not only for obtain-
ing a resident permit, but for the teaching system to become part of local
practice. This was precisely was Muhammad Salih Hendricks lacked. Finally,

34 N. Green, Bombay Islam, 211–234.

35 N. Worden, “VOC Cape Town as an Indian Ocean Port”, in H.P. Ray and E.A. Alpers, Cross
Currents and Community Networks. The History of the Indian Ocean World, Oxford
University Press, 2007.
106 chapter 5

the importance of Mecca must be emphasized again, both as a place where

intellectual friendships were forged (as between Muhammad Salih Hendricks
and ʿAbd Allāh Bā Kathīr) or more informal contacts made (as between Sayyid
Manṣūr the pilgrimage guide and the South African pilgrimage contingent).
Furthermore, we do not know if this was an organized effort nor how far
beyond Cape Town these travellers reached. Most likely, these teachers were
independent “operators” in a field where the daʿwa impetus played a crucial
role – possibly in collaboration with the Azzawiya school of Muhammad
Hendricks. Interviews with ʿAlawī descendants in Cape Town also indicate
that many of the early daʿwa travellers (early twentieth century) were “on
their own” and that they fanned out throughout the country.36 Their influence
on other Muslim communities (or previously non-Muslim communities) is a
topic for further research.
What impact ʿAlawī teachers in Cape Town had on the Islamic practices
of local Muslims, is more difficult to determine. As will be shown below
(Chapter 7), some practices they propagated are distinct to this day, notably
the widespread recital of Rātib al-Ḥaddād.
What does seem clear is that the “East African” network catered to a specific
segment of the South African Muslim community, different, for example from
that described by N. Green. As such, the development in Cape Town resembles
that in Mozambique and Madagascar, in the sense that divisions between eth-
nic and/or linguistic communities remained unaltered.
However, the Cape Malay community must also be considered a “special
case” as this was a community that already had a history of “ ʿAlawī Indian
Ocean Islam” and as such, the arrival of daʿwa travellers may be considered
a “reconnecting” rather than bringing to the field something new. It is also
evident that the Cape Muslim community was open and accommodating to
impulses from abroad, as is shows by the efforts of Mr. Dollie, the chemist.
We may interpret this willingness to “import” teachers in light what F. Becker
called “a response to a changing way of life” (See above, Chapter 1). The Cape
Muslim community was indeed facing changing times in the first decades of
the twentieth century, with increased facility of travel and increased opportu-
nity to connect to “global Islam” of various strands.

36 Interview, Nabil al-Saqqaf, Adil al-Attas and Uthman b. Abd Allah, Cape Town, August
2005. All three men name several names among their great-(or great-great) grandfathers
on both sides who came to South Africa “for daʿwa.” They settled in Kimberly, Cape Town,
Johannesburg and Port Elizabeth.
the cape town muslim community 107

This openness should not be overstated, as the hostility expressed towards

Muhammad Salih Hendricks indicate. What took place in Cape Town was
rather a process whereby an entire community reoriented itself towards differ-
ent versions of globalized Islamic discourse, the result being both a unification
(in the sense of being a Muslim community) and a fragmentation (in the sense
of being organized along different mosque/ṭarīqa lines).
 chapter 6

Travelling Texts
Arabic Literate Learning in Coastal East Africa, c. 1860–1930

This book,
What is in it, is in it,
And whosoever does not know what is in it
May the dog pee on him1

∵
While the previous chapters have focused mainly on the spread of Sufi orders to
different locations in the western Indian Ocean, this chapter addresses some-
thing more fleeting but also the most concrete aspect of reformist thought:
its words. These are manifested in manuscripts and printed books. In other
words, the object under study is the fabric underpinning the networks formed
by travel, trade and teacher-scholar relationships; the printed books, manu-
scripts, large and small, texts and commentaries that were part of the making
of Sufi reformism in the western Indian Ocean.
Surprisingly, this “parallel network” has often received less attention in stud-
ies of Islamic networks in nineteenth and twentieth century East Africa, in
favour of the network formed by individual teachers and scholars.2 Overall,
travelling bodies or spirits have been stressed, not texts. This is all the more
surprising, for the simple reason that to the historian, this is often the most
tangible and accessible of all the networks of the past. Indeed, it is often all
we have.
Today, what will be called here a “book network” can be traced in the form
of manuscript collections and mosque libraries and the succession of manu-
scripts, text editions, reprints, commentaries and additional textual material.
Furthermore, the words themselves – whether transmitted in manuscript or
print form – are our most explicit and lasting testimonies of ideas, and they
can be assumed to have had that function also in the past, albeit to the admit-
tedly limited group of people that B.G. Martin called “the learned class”.

1 Inscription in manuscript, Riyadha Mosque, Lamu. EAP466_RM18.

2 Notable exceptions when it comes to East Africa are the works by B.G. Martin, A. Ghazal,
S. Reese, V. Hoffmann and R. Loimeier.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�07

travelling texts 109

Textual Transmission and Religious Authority

As has been pointed out in earlier research, East Africa in the period c. 1860–
1930 saw an increased orientation towards textuality. I have argued elsewhere
that what took place was a “textualization of charisma”, as authority no longer
rested solely in a person or group, but rather in a person’s (or group’s) ability to
access texts.3 Islamic authority was, in other words, becoming “bookish”, con-
tingent on access to written words, be they produced locally or elsewhere, in
the present or in the past.
Here, the discussion will address text in the most concrete terms, as books
and manuscripts. What books were being circulated, and who owned them? In
many ways, these questions echo those of the German orientalist C.H. Becker,
who in 1911 asked: “What books come into the hands of local mwalimus and so
exercise a direct influence on the intellectual life of our colony? [Tanganyika]”4
A second question is how authority was gained and maintained through
the presence of texts. Explanations will be sought in the teaching, ownership,
exchange, circulation, copying, and commenting on manuscripts and printed
books. In total, the analysis will move towards a fuller understanding of the
textual basis for the spread of Sufism in nineteenth and early twentieth cen-
tury East Africa.
This chapter will also address a third and related topic: the transition from
manuscript to print in this period. As has been pointed out by, amongst others,
R. Schulze, the transition from manuscript to print implied a changing role
for the Islamic text in the public sphere. As the “printed sphere” emerged, a
new social discourse was created “in which the new reading public came to
distinguish between “tradition” (manuscripts without use in the market) and
what was considered “modern” (printed books).”5 Here, Schulze’s observa-
tions will be addressed from the point of view of authority: As textual trans-
mission moved into print, did the authority attached to a given text change?
Did reader­ship change? In other words: did print open for new interpreta-
tions or new interpreters? Or, to formulate the question slightly differently,
and more in line with M. Lambek’s emphasis on meaning as constructed by

3 A.K. Bang, “Authority and Piety”.

4 C.H. Becker, “Materials for the understanding of Islam in German East Africa” Edited and
translated by B.G. Martin, Tanzania Notes and Records, 68, 1968, 31–61.
5 R. Schulze, “The birth of tradition and modernity in eighteenth and nineteenth century
Islamic culture – the case of printing” in J. Skovgaard-Pedersen (ed.), Culture and History: The
introduction of the printing press in the Middle East, Culture and History, 16: 29–72, 29.
110 chapter 6

local ­hermeneutics of the text:6 What meanings were ascribed to the actual
object (the manuscript or the printed text) and to the textual content? How
did the meaning(s) attached to a text change as readers became more numer-
ous (however slightly) and more specialized?

Islamic Texts in the Western Indian Ocean: A Network of Global and

Local Authority
In the introduction to her study Islam Translated, R. Ricci proposes a literary
network, which, as her book illustrates for south and south-east Asia, “con-
nected Muslims across boundaries of space and culture”.7 Moreover, and more
importantly from the point of view of intellectual history, these networks of
texts facilitated the introduction and sustainability of “a complex web of prior
texts that were crucial to the establishment of both local and global Islamic
identities.”
As Ricci points out, a literary network is not a web where information only
flows one way. Rather, the diffusion of texts is a dynamic process which involves
many types of adaptations including editing, translating, commenting, read-
ing, reciting and performing through innumerable versions, variants (or
“tellings”) to the point where the notion of stable, original versions dissolve –
almost literally – in the sea.
This chapter analyses the diffusion of text in the Western Indian Ocean in a
related manner; along lines of integration versus diversification, or, in an alter-
native terminology; globalization and localization. It takes its cue from Ricci,
who takes as her starting point an “Arabic cosmopolis” where texts that were
originally in Arabic were disseminated into a translocal Islamic world “consti-
tuted and defined by language, literature and religion.”8 This cosmopolis can
be detected also on the western side of the Indian Ocean, and what will be
explored here is how – or more concretely: through what processes of incorpo-
ration, production, reproduction – the actors and activities within this literary
network contributed to the parallel processes of globalization and localization
of Islam in the region.

6 M. Lambek, “Certain knowledge, contestable authority: power and practice on the Islamic
periphery”, American Ethnologist, 17 (1), 23–40.
7 R. Ricci, Islam Translated. Literature, Conversion and the Arabic Cosmopolis of South and
Southeast Asia, University of Chicago Press, 2011, 1–2.
8 R. Ricci, Islam Translated, 4. It is important to note that Ricci by “Arabic cosmopolis” does not
imply a networks of texts in Arabic – but rather a literary network of several languages that
borrow and adapts from Arabic themes, tropes, genres, styles and narratives.
travelling texts 111

Finally, book networks may also be approached from the point of view
of language. Taking a related approach, but focusing on literacy rather than
explicitly on literature, P. Larson has analysed the Malagasy diaspora in the
Indian Ocean from the point of view of textual networks.9 His study traces the
relationship between writing and orality, vernacular renderings and creoliza-
tion, ultimately revealing the complexities of textual exchange. This study is
limited to Arabic, and will thus have little to say about the interplay of Arabic
and Swahili (let alone Emahkua, Kimwani, Malagasy or Afrikaans) in this
period, an approach that certainly would be fruitful and which undoubtedly
would yield interesting results. More concretely, a study of the texts in Swahili
that draw upon the textual corpus of reformist Sufism, would further deepen
our understanding of the complexities of religious authority in East African
society.10 On the assumption that the nineteenth century saw a “scriptural-
ization” of East African Islam, to what extent was authority also increasingly
being transmitted in writing in Swahili or other vernaculars? These questions
will only be raised here; they will not be answered.

Book Knowledge in the Age of Manuscripts: 1860s into the

Twentieth Century

The corpus of Islamic learning has numerous sub-disciplines, most of which

were present in East Africa by the mid/late nineteenth century. A discussion of
every discipline would be too long for this book, and I have therefore selected
Arabic grammar (naḥw), genealogy (nasab), and jurisprudence (fiqh) and Sufi/
devotional works (madḥ, duʿāʾ, qaṣāʾid) for a more in-depth discussion. The
reason for this selection is that these are the most representative for the book
collections consulted here:

– The Maalim Idris Collection in Zanzibar

– Checklist of the Arabic manuscripts in the Zanzibar National Archives, as
compiled by L. Declich, 2006.
– The Riyadha Mosque in Lamu, surveyed July 2010 and during digitalization
programme in 2011–2012.

9 P. Larson, Ocean of Letters. Language and Creolization in an Indian Ocean Diaspora,

Cambridge University Press, 2009.
10 R. Samsom, “Swahili Manuscript Culture”, Hamburg, Centre for the Study of Manuscript
Cultures, Newletter 4, 2011, 68–77.
112 chapter 6

– The manuscript accession list of the Centre National de Recherche in

Moroni, Grande Comore, photocopied in 1998. Unfortunately, I have not
been able to revisit Grande Comore and have therefore worked only from a
photocopy of the hand list.
– The Qādiriyya mosque in Mozambique Island

Arabic Grammatical Works: Arabic as a Foreign Language and as a

Learned Language
The increased emphasis on Arabic in the nineteenth century has been inter-
preted by R. Pouwels as a consequence of Bū Saʿīdī rule, with its corps of liwālis,
qāḍīs and administrators in every town on the coast.11 This was particularly
so during the reign of Sayyid Barghash, whose bureaucratic system provided
not only order, but also employment opportunities for the Arabic-literate. As
the authority of the traditional Swahili waungwana was waning, so the ques-
tion arose as to who should have religious authority over the stream of “new
Muslims” on the coast from the neighbouring Giriama, Pokomo, Oromo and
others.12
By the time of Sayyid Barghash, one of the main avenues for entering Swahili
society was some knowledge of the written corpus of Islam, notably the Quran
and the Sunna. By the late 1800s, this knowledge was increasingly being trans-
mitted through the expanding Sufi orders. For entry into the bureaucracy or
the learned class, knowledge of Arabic was an absolute prerequisite. Some
would know Arabic as a consequence of being either immigrants from Arabia
or first-generation immigrants. Others – especially individuals with a Ḥaḍramī
or Omani background – would travel to Arabia (Ḥaḍramawt, Mecca or Oman)
to learn Arabic there, while others in turn would follow instruction taking
place in mosques or with individual teachers (ḥalaqa), and later in the new
“mosque colleges”.
As has been pointed out by C.H. Becker, and also discussed by Loimeier,
the main introductory work of grammar was al-Muqaddima al-Ājurrūmiyya.13
Among the most widely used commentary was that by Aḥmad Zaynī Daḥlān,14

11 R.L. Pouwels, Horn and Crescent, 147.

12 J. McIntosh, The edge of Islam: power, personhood and ethnoreligious boundaries on the
Kenya coast, Durham (Duke Univ. Press), 2009.
13 Abū ʿAbd Allāh Muḥammad al-Ṣanhājī (d. 1323), Al-Muqaddima al-Ājurrūmiyya fī ʿilm
al-ʿarabiyya (GAL II: 237)
14 Aḥmad Zaynī Daḥlān, Sharḥ al-Ājurrūmiyya (GAL S II, 810). See also Loimeier, Between
Social Skills, 177. The earliest Egyptian printed version of this book was published in 1875
by Wahbiyya Publishers. It was then printed almost every other year until 1900 by Bulāq
travelling texts 113

who, as discussed in Chapter 2, was a c­ entral figure for Indian Ocean scholars
in Mecca. However, older works were also being reproduced on the coast. For
example, the Riyadha Mosque library holds a manuscript copy of the com-
mentary on al-Ājurrūmiyya authored by the f­ ifteenth-century Egyptian scholar
Khālid b. ʿAbd Allāh al-Azharī. 15
On this background it is not surprising to find that a substantial portion
of manuscripts from the period consists of Arabic grammatical works of dif-
ferent types and origins, a fact which substantiates Loimeier’s statement that
“language related disciplines as well as logic (manṭiq) were extremely impor-
tant [. . .]”.16 A clear indication is the fact that of the 80 manuscripts that were
brought to the Zanzibar National Archives from the former EACROTONAL col-
lection, one quarter (20 items) were concerned with Arabic language, grammar,
rhetoric or linguistics.17 In the Riyadha collection, the proportion is slightly
lower, as 12 out of 133 items are directly concerned with Arabic language (naḥw,
ṣarf, manṭiq).18

The Alfiyya and its Commentaries

A widely used text for Arabic grammar was the Alfiyya by Muḥammad b.
ʿAbd Allāh Ibn Mālik al-Tāʾī al-Jīlānī, known as Ibn Mālik (d. 672/1273). This
is a versified work of grammar that consists – as the name indicates – of one
thousand verses. The commentary by ʿAbd Allāh b. Abd al-Raḥmān Ibn ʿAqīl19
(d.769/1367), known as Sharḥ Ibn Aqīl, was a widely used commentary to this
work,20 and the standard companion to the study and memorization of the
Alfiyya. As such, it is not an unexpected title to find among the scholarly class
in East Africa by the late nineteenth century.

Publishers. See ʿAyīda Ibrāhīm Nuṣayr, Al-kutub al-ʿArabiyya allati nushirat fī Misr fī ’l-qarn
al-tāsiʿ ʿashar, Cairo, 1990, 145.
15 EAP466_RM008, Manuscript, Arabic, undated, 285 pages, Sharḥ laṭīf li-alfāẓ
al-Ājurrūmiyya.
16 Loimeier, Between Social Skills, 189.
17 I have tried repeatedly to consult the EAC collection at the ZNA, but unfortunately is has
been not available. This summary is therefore made from the checklist made by Lorenzo
Declich in the early 2000s. L. Declich, The Arabic Manuscripts.
18 The proportionally lower figure is partly explained by the fact that in the Riyadha
collection, one Quran with individually-bound juz’s is listed as separate items. Thus one
Quran is up to 26 items, eg. EAP466_RM153 – EAP466_RM178.
19 ʿAbd Allāh Ibn ʿAqīl, Sharḥ Ibn ʿAqīl ʿalā Alfiyyat Ibn Mālik, GAL I, 298–299.
20 S. Glazer, “The Alfiyya commentaries of Ibn ʿAqīl and Abū Ḥayyān”, The Muslim World,
31/4, 1941, 400–408. In 1330/1911, the Sharḥ Ibn ʿAqīl was required reading for 4th year
students at al-Azhar. http://sunniforum.net/showthread.php?t=6992
114 chapter 6

One of the early extant East African manuscript copies of the Alfiyya can
be found in the Riyadha library in Lamu.21 It was copied in Pate in 1866, which
by the late 1860s had already faded considerably from its eighteenth century
glory days. In terms of trade, military importance and as a centre of learning,
Pate had been eclipsed by Lamu as the leading power of the archipelago.22 In
this setting, where “lighted mansions echoed emptily”,23 on the 30th of Ṣafar
1283/16 of July 1866, ʿUthmān b. Ḥājj b. ʿUthmān b. Ḥājj b. Shayth (?) in Faza,
Pate Island finally completed his task. Having copied the entire Alfiyya and the
commentary by Ibn ʿAqīl in the margin, he noted briefly: “That was a moment
of joy” before duly listing his own credentials: “Shāfiʿī by madhhab, Fāzī (from
Faza, Pate Island) by residence and birth, and from the mosque of Shaykh Fātī
by employment.”
We do not know from what original ʿUthmān copied his version. Most likely,
he worked from a manuscript copy imported from Arabia or the Middle East, or
it could have been a locally or regionally produced manuscript. It is also conceiv-
able that the text that, despite its length, could have been copied from memory.
Interestingly, this is not the oldest rendering of Ibn ʿAqīl’s commentary to
be found in the Riyadha, as an earlier version is also in the collection. More
intriguingly, the earlier version is not a manuscript, but a lithograph printed
in Cairo in 1272/1855.24 This is a gloss of the commentary by Ibn ʿAqīl, by the
Egyptian scholar Muḥammad al-Khiḍr al-Dimyāṭī (1798–1870), completed by
the author in 1250/1834. The copy in the Riyadha library is thus one of the
­earliest printed versions of this book coming from Egypt, and hence one of the
earliest printed versions overall.25
The inscriptions in the front of the book show the history of this particular
book in Lamu. The oldest inscription says that the book was acquired by one

21 EAP466_RM28, Manuscript, Arabic, 1283/1866, 567 pages. Alfiyya with commentary by Ibn
ʿAqīl.
22 M. Tolmacheva, The Pate Chronicle, Michigan State Univ. Press, 1993 and P. Romero, Lamu,
passim.
23 The translation paraphrased here is by James de Vere Allen, Al-Inkishafi. Catechism of a
soul, Nairobi/Oxford University Press, 1972.
24 EAP466_RM27, Litograph, Arabic, 1272/1855, 719 pages. Hāshiyya Sharḥ Ibn ʿAqīl ʿalā
Alfiyya ibn Mālik by Muḥammad al-Khiḍrī al-Dimyāṭī.
25 ʿAyīda Ibrāhīm Nuṣayr, Al-kutub al-ʿArabiyya, 148. The overview by Nuṣayr shows only
three editions printed in 1272H, and none earlier. However, it also shows that the gloss
by al-Dimyāṭī was printed repeatedly in the years that followed, by several printers, and
no less than five times by the Būlāq Printing Press between 1865 and 1895. On the Bulāq
Printing Press, established by Muḥammad ʿAlī in 1821, see J. Pedersen, Den Arabiske Bog,
Copenhagen (Fischer), 1946.
travelling texts 115

Saʿīd Qāsim b. Saʿīd al-Maʿamrī in Rajab 1297/June 1880 and taken to Lamu by
him. The next thing we know, is that it was bought by one Saʿīd b. Rāshid from
the estate of Muḥammad b. Qāsim al-Maʿamrī, most likely the brother of the
first owner. Saʿid then made the book waqf for his son Nāsir to be used “as a
fount for knowledge”.
The fact that a printed version of this massive tome of Arabic grammar
ended up in Lamu 25 years after its imprint in Cairo, raises many questions.
Where was the book in the meantime? In private ownership? Although no
inscription indicates a previous owner, it cannot be ruled out. Sitting in the
storage house of the printing press? And how did Saʿīd Qāsim al- Maʿamrī
obtain it? Was it ordered from Cairo through middlemen and travellers? Was
it traded in Mecca and procured during ḥajj? Evidently, such questions can-
not be answered with reference to a singular manuscript or book, but will be
addressed below when it comes to the book network as a whole.
It is interesting to note that the two commentaries predate the Riyadha
college by several decades, even if we accept the earliest date for the estab-
lishment of the Riyadha, i.e. the teaching activities of Habib Saleh to have
started around 1875–1880 (see above, Chapter 2). This raises the question of
usage: How were these texts (the manuscript and the lithograph copy) to be
used before and after they became part of the Riyadha library? The most likely
answer is that they were copied (or, in the case of the lithograph, purchased)
for instructional purposes. The copy so laboriously produced by ʿUthmān b.
Ḥājj b. ʿUthmān b. Ḥājj b. Shayth would be used in education, probably in the
mosque of Shaykh Fātī where ʿUthmān was evidently employed as a khatīb.
Clearly, the “target audience” was students who were sufficiently advanced in
Arabic to understand its complexities.
For the Egyptian lithograph copy, we are left with fewer clues. However, it
seems clear that the book remained in the possession of the Maʿamrī family
from 1880 until at least some time in the early twentieth century. Most likely, it
was only deposited at the Riyadha some time after the 1903 foundation of the
mosque itself.
Another gloss on Ibn ʿAqīl that can be found in East Africa shortly after its
publication in the Middle East, is Sharḥ al-Shawāhid by ʿAbd al-Munʿim b.
ʿAwaḍ al-Jirjāwī.26 This is an edition from 1301/1882, printed in Cairo, and was
once part of the library of Burhān Mkelle.27

26 ʿAbd al-Munʿim b. ʿAwaḍ al-Jirjāwī, d. ca. 1776. He was one of the ʿulamāʾ of al-Azhar.
Ziriklī, iv, 168; GAL I, 299.
27 Maalim Idris Collection, printed books.
116 chapter 6

Muḥyī al-Dīn al-Qaḥṭānī: Sharḥ Tarbiyyat al-aṭfāl bi-taṣrīf al-afʿāl28

Contrary to the previous examples, this work originated in East Africa itself. Its
author, Muḥyī al-Dīn al-Qaḥṭānī (c. 1790–1869), has already been introduced
to the reader in Chapters 2 and 3 as a “first generation” Zanzibari ʿālim and
as an important shaykh of the Zanzibar Qādiriyya. He was also among those
most emphatically to emphasize written authority and the Arabic language,
as this work clearly demonstrates. The text is a commentary on al-Qaḥṭānī’s
own poem Tarbiyyat al-aṭfāl, an educational poem of grammar. The purpose
of this text, too, is clearly educational; to educate young people in the basics of
the Arabic morphology. Al-Qaḥṭanī even notes in his introduction that “some
of the brothers have asked me to produce a commentary on Tarbiyyat al-Aṭfāl”,
apologetically adding that “not all of my explanations [in the Tarbiyya] were
quite clear . . .” The image is that of a learned scholar who tries, and tries again,
to make young East Africans grasp the complexities of the Arabic language, a
task, as we shall see, was to pre-occupy also members of the next generation
of teachers.

Mirqāt al-iʿrāb (Mukhtaṣar fī ṣarf wa-naḥw)29

Mirqāt al-iʿrāb is a short treatise on morphology, grammar and pronunciation.
It is an introductory text to, and a condensed version of al-Iʿrāb ʿan qawāʿid
al-Iʿrāb by the 12-century scholar Jamāl al-Dīn Abū Muḥammad Ibn Hishām.30
The copy was made by Muṣabbiḥ bin Sālim b. Muṣabbiḥ al-Barwānī. The
annotation indicates that it was owned by Shaykh Sālim b. Makka al-Nawfalī,
inscribed by his student, in turn, ʿAlī b. Sālim al-Hināwī.
The first half of the manuscript deals with basic Arabic morphology (ṣarf ),
listing the Arabic sun- and moon letter. It then discusses, respectively, the
verbs, nouns and particles. The second half presents the basic rules of gram-
mar (naḥw), topic by topic, clearly marked and distinguished. This section
ends with exercises in the form of questions and answers. Again, the purpose
is educational, clearly intended for young people with limited knowledge of
Arabic.

Genealogy (nasab): Identities in Writing

As mentioned above, genealogical works made up a sizable proportion of the
Arabic manuscripts that were produced, copied or imported to East Africa in

28 EAP466_RM38, Manuscript, Arabic, undated, 154 pages. Muḥyī al-Dīn al-Qaḥṭānī, Sharḥ
tarbiyyat al-aṭfāl.
29 Maalim Idris Collection, MS 6. Manuscript, Arabic, 1331/1913, 46 pages.
30 GAL II, 24, 27.
travelling texts 117

the period. The reason was that people of Arab (and especially of Ḥaḍramī
ʿAlawī) background were more likely than others to be literate in Arabic, hav-
ing either acquired Arabic from their family, or through studies in Arabia. They
were also more likely to take an interest in genealogy, as a proof of genea-
logical and spiritual identity, to establish themselves as figures of authority
vis-à-vis others – or both. Many Ḥaḍramī ʿAlawīs in the Indian Ocean – from
Indonesia to East Africa – were, as U. Freitag has pointed out “obsessed with
genealogies.”31 This “obsession” resulted in a whole literature of nasab works,
increasing in size and complexity as they were revised time and again. By 1989,
the Ḥaḍramī historian Saqqāf b. ʿAlī al-Saqqāf even saw the need to publish a
“simplified version” where the family lines are given in shorter form.32

An East African Genealogical Work: Ḥasan b. Muḥammad b. Ḥasan

Jamal al-Layl: Marsūma al-ʿAyniyya fī dhikrī baʿḍ al-sāda
al-ʿAlawiyya33
The quest for identity – and implicitly, authority – is the main theme of this
work, which was composed in 1896 or 1897 by Ḥasan b. Muḥammad b. Ḥasan
Jamal al-Layl, a Zanzibari-born ʿAlawī who died in 1904. The treatise was com-
posed as the result of a quarrel with his fellow scholar in Zanzibar, ʿAbd Allāh
b. Wazīr Mtsujini, who reputedly married a sharīfa without the presence of her
guardian, and is a harsh attack on Mtsujini’s motives, method and knowledge.34
In his survey of the Jamal al-Layl lineage, Ḥasan follows a straightforward
methodology, which he carefully lays out to the reader. First, he lists the gene-
alogies given in the Sharḥ al-ʿAyniyya by Aḥmad b. Zayn al-Ḥibsḥī (d. 1733).35

31 U. Freitag, “Hadhramaut: A religious centre”, 178.

32 Saqqāf b. ʿAlī al-Saqqāf, Darasa fī nasab al-sāda banī ʿAlawī dhurriyyat al-imām al-muhājir
Aḥmad b. ʿĪsā, M. al-Mukhtār al-Islāmī, 1989. This “user-friendly” version was composed
following a challenge from the well-known modernist and Islamic activist ʿAlī al-Tanṭāwī
(1908–1999).
33 Riyadha Mosque Library, EAP-RM60, Manuscript, Arabic, undated, but between 1896 and
1904 by textual evidence, 61 pages, Arabic, copied by Muḥammad b. Shaykh ʿAbd al-Qādir
al-Barāwī, 14 Rabīʿ I 1314/24 August 1896. Another, related text by Ḥasan b. Muḥammad
Jamal al-Layl can be found in the Zanzibar National Archives, ZA 8/58 (photocopy in
Bergen). For a discussion of this text, see V. Hoffmann, “The Role of the Masharifu on the
Swahili Coast in the Nineteenth and Twentieth Centuries”, in: K. Morimoto (Ed.), Sayyids
and Sharifs in Muslim Societies, London (Routledge), 2012, 185–197.
34 Farsy/Pouwels, The Shafiʿi Ulama, 80–82.
35 This is a commentary on Qaṣīdat ʿAyniyya by ʿAbd Allāh b. ʿAlawī al-Ḥaddād (1634–1719)
and contains much biographical detail which is implicit in al-Ḥaddād’s qaṣīda. This work
has been reprinted several times, in as diverse locations as Mecca, Egypt and Jakarta.
118 chapter 6

Then, he compares this with the genealogies that the Jamal al-Layl families
have with them in East Africa. Finally, he compares his finds with a genealogy
that was with Muḥammad Saʿīd Bāb Ṣayl and which evidently was copied for
him by one Sayyid Ḥusayn b. Muḥammad visiting Mecca. By this triangula-
tion, he arrives at his conclusion: “I found that the genealogy is the same as
what was written in Sharḥ al-ʿAyniyya and in the shajara in Mecca. Through
this endeavour, the truth of their nasab is clear.”
Ḥasan also gives the reader the motives for his efforts, and unsurprisingly, he
gives Prophetic traditions that emphasize knowledge of genealogy, whether of
Prophetic descent or not – such as: “Those who claim other fathers than their
own, will not enter Paradise.” He also explicitly says that “I did not do this to
be uplifted or to boast, because God said in his noble Quran: “The best people
in the eyes of God, are those who fear him.” In other words: Ḥasan is portray-
ing himself as a diligent and careful scholar who only does what is expected of
him as a good Muslim, with no ulterior motives. Whether or not we are to take
this at face value, the very fact that this text was circulated in manuscript form,
shows that Ḥasan did expect his work to produce some sort of authority – if
only within the field of nasab, but probably also more generally.
The element of authority becomes even more evident if we take into
account the foreword found in the manuscript copy of Riyadha mosque. It is
worth quoting at length, as it shows clearly the way in which nisba lines were
confirmed across the Indian Ocean, especially the type of authority that came
with certain genealogies. The background to the story is as follows: In 1896/97,
ʿAbd Allāh Bā Kathīr embarked on his journey to Ḥaḍramawt, on a quest for
knowledge – including, in his case too, genealogical knowledge.36 Ḥasan Jamal
al-Layl, who by then had worked on his genealogy for some time, took the
opportunity to send his work along with Bā Kathīr to have it verified by experts
in Ḥaḍramawt. After some time he received a letter from Bā Kathīr:

We arrived in Ḥaḍramawt on a blessed morning, and we visited it and saw

its cities and its ancient monuments. And we said duʿāʾs for you, as we
hope you will remember us in your duʿāʾs [ . . . ] We did not meet anyone
knowledgeable in nasab [until we met] Sayyid ʿAbd al-Raḥmān
al-Mashhūr, ṣāḥib of fatwas. He has with him a big shajara (lineage tree)
of all the sāda and all lines are mentioned in it. [ . . . ] So we compared
your nasab with the copy he has, and we found all your forefathers there.

36 On the journey of ʿAbd Allāh Bā Kathīr in Ḥaḍramawt, see A.K. Bang, Sufis and Scholars,
104–112. For the publication history of the account of this journey, see below.
travelling texts 119

[The letter then lists the lineage, name by name, confirming the migra-
tion of members of the Jamal al-Layl line to East Africa].
When Sayyid ʿAbd al-Raḥmān al-Mashhūr realized that your nisba is
the same as in the ʿAlawī family tree, he said: “I recognize you and your
children as ʿAlawī, ʿAlī and Ṣāliḥ. [ . . . ]. This is your recognized nisba in
the ʿAlawī shajara by ʿAbd al-Raḥmān al-Mashhūr. He greets you and
gives you a big gift, but I don’t want to send it by somebody else’s hand, as
it is very a precious book, but I will bring it myself after the Hajj. What he
composed is a book about all the sāda and the names they were known
by. May God grant us mercy through them.”

Presumably, Bā Kathīr is here referring to the main genealogical work of

ʿAbd al-Raḥmān al-Mashhūr (see above, Chapter 2), entitled Shams al-ẓahīra
al-ḍāḥiyya al-munīra fī nasab wa-silsila ahl al-bayt al-nabawī (The Luminescent,
encompassing mid-day sun on the lineage and genealogy of the people of the
house of the Prophet). This work was only completed by al-Mashhūr in manu-
script form by 1890, and it is interesting to note how relatively quickly a copy
made its way to East Africa. It does not transpire from the letter whether Bā
Kathīr refers to a manuscript copy or one of the early stone print versions,
but most likely it was the former. Shams al-ẓahīra was since copied repeat-
edly in East Africa, especially among ʿAlawīs and in the institutions linked to
the ʿAlawī reformist movement. One example is a copy held by the Riyadha
Mosque, copied in Ṣafar 1322/April 1904 by Aḥmad b. Ṣāliḥ b. ʿAlawī Jamal al-
Layl (Ahmad Badawi, the son of Habib Saleh, b. 1305/1887–88).37
In these exchanges of genealogical information, the element of authority is
brought out explicitly, as the genealogy is confirmed by the most prominent
authority available. Another important point is that the genealogy no longer is
simply retold orally, but has to be verified by written books, substantiated by
the Shams al-ẓahīra itself being brought by hand to Zanzibar.

37 This manuscript is not on the EAP website as it was insufficiently digitized. In the initial
listing by Shaykh Ahmad Nabhany, it was assigned the number RM30 (See below, sources
and bibliography). It is stored in the Riyadha library.
120 chapter 6

A Travelling Manuscript: Aḥmad b. Ḥasan b. ʿAbd Allāh b. ʿAlawī

al-Ḥaddād: Al-Fawāʾid al-Saniyya fī dhikr faḍāʾil man yantasibu ilā
al-silsila al-nabawiyya38
In addition to the genealogical component, this work includes descriptions
of Ḥaḍramawt, typically highlighting its many glorious mosques, scholars,
books, and its learned tradition. As such, it is a classic ʿAlawī work, describ-
ing at length the “sublime benefits” of the homeland, forming precisely what
E. Ho has called “travelling texts that formulate discourses of mobility.”39 It
was completed in 1203/1788–89, by Aḥmad al-Ḥaddād who was the grandson
of ʿAbd Allāh b. ʿAlawī al-Ḥaddād. Although the text reads like a “travel guide”
to the sacred homeland of Tarīm and its many learned men, mosques, qubbas
and holy sites, this text was actually composed in the Ḥaḍramawt itself and
not – as so many other texts of its kind –by a diaspora returnee overwhelmed
by the sanctity of the homeland.
The copy which is now in the Riyadha library was completed in 1253/1837–
38 (i.e. some 50 years after it was first written) by one Jaʿfar b. Muḥammad b.
Jaʿfar b. Muḥammad b. ʿAbd Allāh b. ʿAlawī al-Ḥaddād. It is not evident whether
the copying took place in Ḥaḍramawt itself or in some other location on the
Indian Ocean rim; either scenario is equally likely.
It has a fascinating set of inscriptions that establishes the diasporic life
of the book itself, moving from one end of the Ḥaḍramī ʿAlawī diaspora to
the other. The first inscription simply says that the copy was given away in
1309/1891–92 – in other words when the copy was almost 50 years old – by ʿAqīl
b. ʿAydarūs b. ʿAqīl “as a gift” to one ʿAbd Allāh b. ʿAlī al-Ḥaddād. A half-century
later, in 1366/1946–47, it was given away again – which indicates the long life
these manuscripts had, being re-circulated as gifts, donated as waqf or passed
on as inheritance. The inscription shows that the copy at this point was in
south-east Asia: “To Ṭāhā b. ʿAlī b. ʿAbd Allāh al-Ḥaddād Bā Faqīh from Shaykh
Bū Bakr b. Sālim in Bandar Batawī (Batavia/Jakarta).” Most likely, Ṭāhā b. ʿAlī
was the one who brought the copy to East Africa, most likely to Mombasa, from
where it passed on to Lamu.

38 EAP466_RM29. Manuscript, Arabic, undated, 244 pages. The full title of the manuscript
is: al-Fawāʾid al-saniyya fī dhikr nubdha min faḍl nisbat man yantasibu ilā al-silsila
al-nabawiyya wa-aʿnī bihim al-sāda al-ʿAlawiyya khuṣūṣan minhum al-qāṭinīn bi’l-jiha
al-Ḥaḍramiyya wa-dhikr shayʾ min manāqibihim al-ʿulyā wa-dhikr jihatihim wa-mā
ikhtaṣṣat bihim min al-faḍāʾil wa’l-khuṣūṣiyyāt al-marḍiyya khuṣūṣan baladuhum
al-maḥrūsa Tarīm. See E. Ho, The Graves of Tarim, bibliography.
39 E. Ho, The Graves of Tarim. 154.
travelling texts 121

What is clear is that this particular copy reached East Africa some time in the
latter half of the twentieth century, which shows that even at this late point –
when printed and audio Islamic material had long since arrived – m ­ anuscript
copies were still circulating. Here we can only guess that special value (cha-
risma or baraka) may have been attached to the copy due to its author, its
copyist and possibly also its previous owners. Finally, as R. Shulze pointed out,
by the late twentieth century, the manuscript may also have been circulated
due it its monetary value.40

Fiqh: The Legal Foundations of Islamic Society

As has been pointed out by several studies,41 the Shāfiʿī fiqh of East Africa was
very much influenced by the Ḥaḍramī ʿAlawī legal traditions, and this is dem-
onstrated by several manuscript copies of, most notably, al-Nawwawī’s Minhāj
al-Ṭālibīn. In the Riyadha library, for example, can be found Habib Saleh’s
own copy of the Minhāj.42 Especially Kitāb al-Nikāḥ (on marriage) and Kitāb
al-Farāʾiḍ (on inheritance) were widely copied, not unexpectedly, as these
were the sections that qāḍīs would have to consult most frequently in their
day-to-day work. Several manuscript copies of al-Nawwawī’s sections on mar-
riage and inheritance can be found both in the Zanzibar National Archives and
the CNDR, Moroni. Also widely circulated in manuscript form in the late nine-
teenth century was the work known as al-Muqaddima al-Ḥaḍramiyya.43 Its
accompanying text, the commentary by Ibn Ḥajar known as Sharḥ Ibn Ḥajar
ʿalā mukhtaṣar Bā Faḍl, seems also to have been widely known and circulated,
at least by the 1880s.44

40 R. Schulze, “The birth of tradition”.

41 C.H. Becker, “Materials for the understanding”, J. Schacht, Notes on Islam in East Africa”,
Studia Islamica, 23:1965, 91–136; B.G Martin, B.G. Martin, “Notes on some members of the
learned class of Zanzibar and East Africa in the Nineteenth Century”, African Historical
Studies, 4:3, 1971, 525–545; Loimeier, Between Social Skills.
42 EAP466_RM18, manuscript, Arabic, undated, 413 pages.
43 Al-Mukhtaṣar al-ṣaghīr li’l-muqaddima al-Ḥaḍramiyya, by ʿAbd Allāh b. ʿAbd al-Raḥmān
Bā Faḍl al-Ḥaḍramī (d. 918/1512–13). On its circulation, see Becker, “Materials for the
understanding”.
44 EAP466_RM57, manuscript, Arabic, 1279/182–63, 357 pages, Sharḥ Ibn Ḥajar ʿalā mukhtaṣar
Bā Faḍl. The copy was completed in 1279/1862–63 by one Muḥammad b. Masʿūd b.
Muḥammad al-Wardī. It is impossible to determine where the manuscript was copied,
but judging solely from the name of the copyist, the copy was not produced in East Africa.
There are no notes of ownership that could indicate how or when it arrived in East Africa.
122 chapter 6

Made waqf in Pate: Ibn Ḥajar al-Haytamī: Minhāj al-Makkī/Tuḥfat

al-muḥtāj li-sharḥ al-Minhāj45
Minhāj al-Makkī by Ibn Ḥajar was completed in 1550s and is an accompany-
ing text to Tuḥfat al-Muḥtāj – which in turn is a commentary on al-Nawwawī’s
Minhāj al-Ṭālibīn.46 This manuscript is presently in the Siu Friday Mosque, and
has neither a date of copy, nor the name of the copyist. Given the prevalence
of Minhāj al-Ṭālibīn as the main legal text in East Africa, it is not surprising that
one of its most widely used commentaries could be found in Pate by the mid/
late nineteenth century. More surprising is an interesting waqfiyya note, by one
Mwana Barka of Somali origin who made this book waqf, unfortunately at an
unspecified date.47 The inscription does not make any particular references
to how exactly the book was to be used, but it nonetheless shows that manu-
scripts were highly valued objects and that those who were in a position to do
so would and could make them waqf. Of course, the fact that this particular
text –a complicated and massive legal tome; certainly not for beginners – was
made waqf by a woman, is interesting in itself. We can only assume that the
manuscript was kept in the mosque, or with the family of Mwana Barka herself.

Purchased in Mecca for the Benefit of the Community: Muṣṭafā

Khayr al-Dīn al-Ḥanīfī, Tartīb qawāʾid al-ashbāh wa’l-naẓāʾir48
This is a work of uṣūl al-fiqh, completed in 1064/1653–54, by the Ḥanafī scholar
Muṣṭafā Khayr al-Dīn (d. 1613).49 As a Ḥanafī work, it is unlike most fiqh texts
to be found on the coast. It relevance in this chapter has more to do with the
inscriptions of ownership than with content, as they demonstrate how manu-
scripts of this type could be acquired. What we find is that the two first owners
were evidently Meccan, which indicate that the manuscripts itself was most
likely produced in Mecca. The last owner given is Abū Bakr b. ʿAbd Allāh b.
ʿUmar al-Shāṭirī, who was then residing in Mombasa. He adds: “I bought this
book in Mecca for half a piaster (niṣf qirsh)”, thus providing us with an indi-
cation of what a manuscript of several hundred pages would fetch in Mecca

45 Manuscript, bound, Arabic, unknown number of pages, partly photographed in Siu, Pate,
July, 2010.
46 Most likely, the manuscript contains both texts by Ibn Ḥajar. Unfortunately, the state of
the manuscript prevented it from being fully opened and photographed.
47 Judging from the paper and ink, the manuscript copy seems to be late nineteenth century,
while the inscription by Mwana seems to be somewhat later, possibly early twentieth
century.
48 The poor condition of this manuscript did not allow for its full digitalization. Its original
reference in the Riyadha catalogue is RM6.
49 GAL II, 310; GAL SII, 425.
travelling texts 123

in the late nineteenth century. The most likely conclusion is that Abū Bakr
al-Shāṭirī bought the book for himself, but there are also indications that the
Shāṭirī family in East Africa (who for some reason was known to be better off
than the other ʿAlawī families) acted as philanthropists on behalf of the com-
munity. They were known for buying books during their travels which were
in turn donated to mosques, individual teachers, or, like in this case, to the
Riyadha mosque.50 Another possibility is of course that al-Shāṭirī ran a book
import business, and that the book was sold onwards to the Riyadha at a later
stage, no note being made of this transaction.

An Example of Local Copying: ʿAbd Allāh b. ʿUmar al-Hijrani

al-ʿAdanī: Mishkāt al-Miṣbāḥ li-sharḥ al-ʿUudda waʾ-’l-silāh51
This is a text concerning marriage law.52 The manuscript was copied in Lamu
some time between 1256 and 1266/1840–1849 by one Sharo ʿUthmān Abū Bakr
ʿAlī al-Sūmālīi. As such, it is an example of local copying of textual material. We
do not know from which original Sharo made his copy, but most likely an ear-
lier, pre-existing copy was to be found in the scholarly community and dupli-
cated. Like in the case of Mwana Barka (above), we see the important role of
individuals of Somali origin, this time as a copyist. While not surprising in a
region near to the Benadir coast, it is worth noting the way this “book network”
cut across ethnic and linguistic boundaries.

Devotional Works (Sufi Litanies and Prayers) and Sufism

Superogatory prayers, litanies and various types of religious poetry have
most likely been recited on the East African coast since the arrival of Islam.
Furthermore, a corpus of poems and prayers in Swahili developed over cen-
turies and were learnt by heart by successive generations. However, from
the nineteenth century, a new set of prayers and devotional litanies in Arabic
seem to have arrived on the coast, this time in the form of written texts. The
“seem to” is important here, as this assumption must remain tentative. It is
very difficult – if not impossible – to exclude the possibility that the same texts
were already circulating in writing by the eighteenth century. Contrary to long
prose texts of fiqh, grammar or taṣawwuf, the devotional texts lend themselves
to oral transmission, and their presence in manuscript form by the mid/late

50 Interview, Aydaroos and Ahmad Jamal al-Layl, Lamu, 5 December 2011.

51 EAP466_RM30, Manuscript, Arabic, dated between 1256 and 1266/1840–1849, 354 pages.
The manuscript contains seven different texts by several authors, both legal, devotional
and genealogical. Only the legal section is referred to here.
52 294; GAL SII, 972.
124 chapter 6

nineteenth century should not lead us to conclude that they were not known
before. It may simply mean that their transmission before this point was oral,
or that manuscript copies that were produced before, were lost. Here, in turn,
it is worth noting the role of institutions such as the Riyadha or the Madrasa
Bā Kathīr in acting precisely as agents of book preservation. The institutions,
then, in addition to their educational function, became facilities where all
kinds of manuscript copies could be stored, thus accumulating the history not
only of Islamic scholarship, but also of expressions of religious devotion.
Among the most widely copied texts was Mawlid al-Barzanjī. The CDRS in
Moroni holds two manuscript copies, unfortunately both undated. The Maalim
Idris collection also holds one copy, dated 1873.53 The Zanzibar National
Archives has two copies, both dated 1875.54 The Riyadha mosque has several
copies, both nineteenth and twentieth century, the latter often in school note-
books together with other devotional texts.55
The distribution of the Mawlid al-Barzanjī in vocalized, Arabic writing had
the specific effect of formalizing proper Arabic vocalization, as Abdallah Saleh
Farsy noted: “I have heard that when the Barzanji maulidi first was introduced
to East Africa, the Arabic was unvocalized and everyone was reciting it as he
pleased. Finally, when some books arrived with proper vocalization, it was
vocalized the way Sayyid Abdu’r-Rahman was reciting it!”56 In other words:
The oral authority of Sayyid Maṣab b. ʿAbd al-Raḥmān was supported by “some
books”, and the question of vocalization settled.
Also widely copied were madāʾiḥ (praises) for the Prophet, to be recited dur-
ing the month of Ramadan. These can be found throughout the ZNA collec-
tion, the Maalim Idris collection and in the Riyadha library, dating from the
1860s to the 1890s.
One specific text that seems to have been often written down is Wird al-Laṭīf
by ʿAbd Allāh b. ʿAlawī al-Ḥaddād (often copied alongside his Rātib, as will be
discussed below, Chapter 7). Another text that was regularly transmitted in
writing was Dalāʾil al-Khayrāt, by the Moroccan scholar Muḥammad al-Jazūlī
(d. 1465). Copies of this text were registered by C.H. Becker and can also be

53 Maalim Idris Collection, MS 16, Arabic, 31290/1873, 313 pages. The volume includes a
collection of devotional texts.
54 ZA8/8 and ZA8/35. See Declich, The Arabic Manuscripts.
55 EAP466_RM15, Manuscript, Arabic, undated, 79 pages; EAP466_RM25, Manuscript, Arabic,
undated, 241 pages, containing various devotional texts; EAP466_RM79, Manuscript,
Arabic, 1349/1928, 314 pages containing various devotional texts.
56 Farsy/Pouwels, The Shafiʿi Ulama, 168. The ʿAbd al-Raḥmān referred to here is ʿAbd
al-Raḥmān b. Aḥmad Jamal al-Layl (d. 1897).
travelling texts 125

found in the Riyadha library dating from the 1910s and 20s. Its recital has been
linked particularly with the Shādhilī order, but its diffusion in writing shows
that it was widely recited in East Africa by the late nineteenth and early twen-
tieth century (as indeed it is today).
As was shown above in the case of the “travelling manuscript”, hand-written
texts could have extraordinarily long lives, for use in educational settings, and
later possibly also as exceptional textual specimen to be used on special occa-
sions.57 Related texts, such as various versions of the witriyya58 prayers are also
among those found in every collection.
Another genre of devotional text is the poetic rendering of the asmāʾ
al-ḥusnā, the 99 names of God. Several such copies can be found in the col-
lections. One example is a beautifully ornamented copy made by Sulaymān
b. Khalfān b. ʿUbayd b. Muḥammad al-ʿAlawī, dated 20 Shawwāl 1302/2 August
1882.59 The text contains two different poems that elaborate on the names. The
first presents all the names in the form of a duʿāʾ (prayer) seeking God’s bless-
ings and seeking the meanings attached to each of His names. The last poem
is only partial, but follows the same theme. Unfortunately, no author’s name is
given to any of the poems.
Among the most widely hand-copied documents overall are the Sufi “text-
books” or manuals containing the adhkār, prayers and poetry specific to the
different orders. Both the Qādiriyya and the Shādhiliyya, as they diffused,
brought with them a set of adhkār and prayers that were performed orally but
that were also written down as text. Manuscripts versions of Qādirī prayers and
litanies can be found in Zanzibar dating from 1900,60 and near-identical texts
from Mozambique Island in the 1930s (i.e. after the alignment with Zanzibar,
known as the Qādiriyya Sadat).61 The compilations include the standard Qādirī
prayers and the prescriptions on how to initiate seekers into the Qādiriyya.
They also include poetry by Shaykh al-Uways and a brief biography of his life.
The Sufi poetry and adhkār (litanies) of the ʿAlawiyya, not unexpectedly,
make up a significant part of the Riyadha library and we can assume that they

57 See also A.K. Bang, “Authority and Piety”. The discussion concerns a madīḥ text that was
re-circulated in Zanzibar between the 1850s and the 1930s.
58 Takhmīs al-Witriyya by Muḥammad b. ʿAbd al-ʿAzīz al-Iskandarī seems to be the most
widespread. See for example MS 7 in CNDR Moroni Collection.
59 Maalim Idris Collection, MS 15, Manuscript, Arabic, undated, 50 pages. The latter poem
gives the name of the writer, but it does not indicate if both are by the same author. Most
likely they are, but the question mark remains nonetheless.
60 Maalim Idris Collection, MS 21, Manuscript, Arabic, undated, 182 pages. By internal
evidence, this manuscript is written before the death of Shaykh Uways in 1909.
61 QADIRI QASIDA/ILHA 1 and QADIRI QASIDA/ILHA 2. See also above, Chapter 3.
126 chapter 6

Image 4 Sample of manuscript, devotional writing. From the collection of Maalim Idris
Muhammad Saleh.
Photo by the author.
travelling texts 127

were widely read and taught among teachers and students. In fact, texts that
can be classified as either devotional or explicitly Sufi, make up the majority of
the Riyadha manuscript collection.

Identified Poetry/Devotional/Sufi texts in the Riyadha manuscript collection

EAP466_RM4, Manuscript, Arabic, EAP466_RM7, Manuscript, Arabic,

1316/1898, 22 pages. undated, 150 pages.
Title: Al-bakūra al-Janniya fī ʿamal al-āla Title: Takhmīs al-Witriyya, for recitation
al-Jaybiyya. Poetry of praise. during Ramaḍān.
Author: Muḥammad b. Yūsuf al-Khiyāṭ Author: Abū ʿAbd Allāh Muḥammad b. Abī
al-Makkī, d. c. 1915. Bakr al-Baghdādī al-Shāfiʿī known as
al-Witrī (d. 1264).
EAP466_RM10, Manuscript, Arabic, EAP466_RM11, Manuscript, Arabic,
1324/1906–07, 46 pages. undated, 79 pages.
Title: Mawlid Dibāʾī. Poetic text for Title: Mawlid al-Barzanjī
recitation on the occasion of Mawlid. Recited throughout East Africa on Mawlid.
Author: ʿAbd al-Raḥmān al-Dibāʾī. Author: Jaʿfar al-Barzanjī, d. 1765, Medina.
EAP466_RM19, Manuscript, Arabic, EAP466_RM24, Manuscript, Arabic,
1346/1927, 244 pages. 1326/1908, 455 pages.
Title: Diwān al-ʿAdanī. Poetry. Title: Fatḥ al-Mubīn. Sharḥ anfās ʿAydarūs.
Author: Abū Bakr b. ʿAbd Allāh Auhor: ʿAbd al-Raḥmān b. Muṣṭafā
al-ʿAydarūs, known as al-ʿAdani al-ʿAydarūs
(1447–1508), Aden.
EAP466_RM25, Manuscript, Arabic, EAP466_RM28, Manuscript, Arabic,
undated, 309 pages. 1336/1917–18, 223 pages.
Title: Various devotional texts. Title: Īḍāḥ Asrār ʿulūm al-muqarribīn.
Author: Several authors. Author: Muḥammad b. ʿAbd Allāh b.
Shaykh al-ʿAydarūs
EAP466_RM34, Manuscript, Arabic, EAP466_RM36, Manuscript, Arabic,
1340/1921–22, 42 pages. undated, 130 pages.
Ahl Badr. Duʿāʾ for intercession. Title: Sharḥ matn al-Burda (commentary
on Qaṣīdat al-Burda by Muḥammad b.
Saʿīd al-Būṣīrī, d. 1269)
Author: Shaykh Khālid b. ʿAbd Allāh
al-Azharī, d. 1499.
128 chapter 6

EAP466_RM41, Manuscript, Arabic, EAP466_RM42, Manuscript, Arabic,

undated, 254 pages. 1314/1896–97, 604 pages.
Title: Kitāb al-barqa al-mashīqa fī dhikr Title: Kitāb uns al-sālikīn fī manāqib baʿd
libās al-khirqa al-anīqa. al-ṣāliḥīn. Biography of Sufi scholars and
Author: ʿAlī b. Abī Bakr al-Sakrān b. ʿAbd saints.
al-Raḥmān al-Saqqāf, d. Ḥaḍramawt, Author: ʿAbd Allāh b. ʿAbd al-Rahmān b.
1490. ʿAlī Bā Harūn.
EAP466_RM48, Manuscript, Arabic, EAP466_RM49, Manuscript, Arabic,
undated, 121 pages. 1323/1905–06, 171 pages.
Title: Wiṣayāt wa-jamʿiyya al-ijāzāt Title: Mukātibāt ʿAlī b. Muḥammad
ʿAydarūs b. ʿUmar al-Ḥibshī. al-Ḥibshī.
Author: ʿAydarūs b. ʿUmar al-Ḥibshī, d. Author: ʿAlī b. Muḥammad al-Ḥibshī, d.
Ḥaḍramawt, 1896. Ḥaḍramawt, 1915.
EAP466_RM52, Manuscript, Arabic, EAP466_RM53, Manuscript, Arabic,
undated, 34 pages. undated, 326 pages.
Title: Simṭ al-durar. Text introduced to Dalāʾil al-Khayrāt wa-mashāriq al-anwār fī
East Africa in the late 19th century for dhikr al-ṣalāt ʿalā al-nabī al-mukhtār.
recitation during mawlid. Author: Muḥammad b. Sulaymān al-Jazūlī,
Author: ʿAlī b. Muḥammad al-Ḥibshī, d. d. 1465.
Ḥaḍramawt, 1915.
EAP466_RM58, Manuscript, Arabic, EAP466_RM59, Manuscript, Arabic,
1347/1928, 314 pages. 1353/1934, 129 pages.
Title: Miscellaneous devotional texts, Prayers and praises of the Prophet
incluing Rātib al-Ḥaddād. Muḥammad (Majmūʿ al-aṭāʾif al-ʿarshiyya)
Author: Several authors.
EAP466_RM64, Manuscript, Arabic, EAP466_RM67, Manuscript, Arabic,
1341/1923, 13 pages. undated, 34 pages.
Title: Iḥdā ʿashariyya. Poetic devotional Title: Mawlid miʿrāj (“qiṣṣat al-miʿrāj”)
text recited on the 11 of every month. Author: Jaʿfar b. Zayn al-ʿĀbidīn
Author: Aḥmad b. Muḥammad al-Barzanjī, d. 1765, Medina.
al-Miḥḍār, d. 1304/1886–87,
Ḥaḍramawt.
EAP466_RM83, Manuscript, Arabic, EAP466_RM88, Manuscript, Arabic,
20th century, 16 pages. undated, 9 pages.
Unknown sharḥ on a qaṣīda on tawḥīd Title: Iḥdā ʿashariyya. First section of
by ʿAbd al-Ghanī al-Nābulusī (d. 1731). poetic devotional text recited on the 11 of
Possibly by ʿAbd Allāh b. ʿAlawī every month.
al-Haddād, d. 1719, Ḥaḍramawt. Author: Aḥmad b. Muḥammad al-Miḥḍār,
d. 1304/1886–87, Ḥaḍramawt.
travelling texts 129

EAP466_RM89, Manuscript, Arabic, EAP466_RM91, Manuscript, Arabic,

undated, 42 pages. 1334/1915–16. Manuscript, Arabic,
Title: Iḥdā ʿashariyya. Poetic devotional 34 pages.
text recited on the 11 of every month. Title: Ṭayyib al-asmāʾ al-mubāraka.
Author: Aḥmad b. Muḥammad Author: Unknown.
al-Miḥḍār, d. 1304/1886–87,
Ḥaḍramawt.
EAP466_RM99, Manuscript, Arabic, EAP466_RM105, Manuscript, Arabic,
1337/1918–19, 11 pages. 1334/1915–16, 8 pages.
Title: Wiṣayāt Aḥmad b. Abī Bakr b. Title: Duʿāʾ birr wālidayn. A prayer for
Sumayṭ ilā ʿAbd Allāh Bā Kathīr honouring one’s parents.
Author: Aḥmad b. Abī Bakr b. Sumayṭ, d. Author: Muḥammad b. Aḥmad b. Abī
1925, Zanzibar. Text in the author’s own ’l-Ḥubb al-Ḥaḍramī, d. 611/1214–15,
hand. Ḥadramawt.

EAP466_RM106, Manuscript, Arabic EAP466_RM108, Manuscript, Arabic,

and Swahili, 1353/1934, 143 pages. undated, 29 pages.
Title: Kitāb jāmiʿ al-adhkār. Collection of Title: Kitāb khilāṣa fī al-ism al-ʿaẓam.
dhikr and prayers for the prophet. Author: ʿAlī b. Ḥasan al-ʿAṭṭās,
Author: Several authors. 1172/1758–59.
EAP466_RM116, Manuscript, Arabic,
1346/1927, 265 pages.
Title: Diwān al-ʿAdanī.
Author: Abū Bakr b. ʿAbd Allāh
al-ʿAydarūs, known as al-ʿAdani
(1447–1508), Aden.

Figure 10 Identified devotional and Sufi texts in the Riyadha manuscript collection

The devotional and Sufi texts used in the Riyadha show a clear Ḥaḍramī ʿAlawī
influence, particularly in terms of authorship (such as for example the poetry
of Abū Bakr al-ʿAdanī62 and the Sufi manuals of ʿAlī b. Muḥammad al-Ḥibshī).

62 For example, Habib Saleh owned an illuminated copy of the diwān of the fifteenth
century ʿAlawī-Adeni Sufi and poet Abū Bakr b. ʿAbd Allāh al-ʿAydarūs, known as al-ʿAdeni
(1447–1508) EAP466_RM19, Manuscript, Arabic, 1347/1928, 244 pages. Another copy of
the same diwān, and by the same scribe, was produced one year earlier, in 1346/1927:
EAP466_RM76, Manuscript, 1346/1927, 266 pages. On al-ʿAdeni and his importance in the
formation of the “ʿAlawī way”, see E. Ho, The Graves of Tarim.
130 chapter 6

However, the Riyadha also held multiple copies of texts that were common
to all of East Africa, and indeed beyond to the wider Muslim world (such as
Qaṣīdat al-Burda and Mawlid al-Barzanjī). From their presence in an educa-
tional institution, we can deduce that knowledge of these texts – literally, the
ability to read them – was essential for being an educated member of con-
temporary Muslim society. This, in turn, confirms Loimeier’s findings from
Zanzibar of a “bildungskanon”, an educational common ground that extended
beyond Lamu; to Ḥaḍramawt, Zanzibar and the Comoro Islands.

From Manuscript to Print: Parallel or Converging Authorities?

As has been pointed out in several studies, the first Arabic printing press in
East Africa was imported by Sayyid Barghash in 1879, complete with experi-
enced Lebanese printers. In its early period, it published primarily Ibāḍī legal
works (its main output was Kitāb Qāmūs al-Sharʿīyya), and can thus not be
said to have played a part in the spread of Sufi reformist Islam on the coast.63
Nonetheless, the Bū Saʿidī sultans played an important role also when it came
to Shāfiʿī textual material, maintaining close contact not only with their co-
religionists, but also with the wider nahḍa in the Middle East and beyond.
A. Ghazal has described how Ibāḍī scholars of Zanzibar incorporated nahḍa
ideas into the works, and how they participated in the wider re-orientation
(or reform) of Ibāḍism in Oman and Algeria.64 Her study emphasizes the
emergence of journals, the flow of printed matter (particularly from Algeria,
but also from the Middle East overall) and the rise of associations that in turn
became the seed of nationalist sentiments.
By the end of the nineteenth century, journals and newspapers from the
Middle East were accessible – at least in Zanzibar, and most likely also in
Mombasa. However, the question must be raised if the Shāfiʿīs (and specifi-
cally, the representatives of Sufi orders), too can be said to be participants in
the ongoing nahḍa and thus in the discourse of global Islam? This, in turn, can
be linked to the question of authority; to what extent were the new, printed
books and publications such as journals overriding earlier authorities, i.e.
manuscripts, or even oral renderings of text?

63 Only one work by an East African was published before 1900, and this too was by an
Ibāḍī; Abū Muslim al-Rawwāḥī’s account of the son of Sultan ʿAlī to the east. On this, see
Sadgrove, P. 2004. “From Wādī Mīzāb to Unguja: Zanzibar’s scholarly links”. In: Reese, S.
(ed). The transmission of learning in Islamic Africa. Leiden (Brill). 184–211.
64 A. Ghazal, Islamic Reform.
travelling texts 131

Arabic Books Imported from the Middle East

It is impossible within the framework of this study to establish an overview
of the number and type of printed Arabic books imported to East Africa dur-
ing the nineteenth and early twentieth century. Clearly, by the late nineteenth
century, Arabic printed books were being brought from Arabia (primarily
Ḥaḍramawt and Mecca), from the Levant, and above all, from Egypt. The recip-
ients ranged from the Bū Saʿīdī sultans themselves, to mosques and teaching
facilities to individual scholars.65 The overall impression, after a survey of book
collections in Zanzibar and Lamu is that the import of printed books started
in earnest around the turn of the century, and really picked up between 1910
and 1925. This could happen through purchase, or when books were given as
part of a student/teacher relationship. A case in point is the brief remark made
by Abdallah Saleh Farsy in his biography of the Zanzibari-Comorian scholar
Aḥmad b. Muḥammad Mlomry, who was known for his wide travelling and
for seeking out famous shaykhs abroad: “He also attended the classes of Sh.
Salym al-Bishry,66 the Shaykh al-Islam of Egypt, and he went to Sayyid Ahmad
Bey al-Husayni, the first person to publish Imām Shafi’i’s Al-Umm. He gave Sh.
Ahmad a copy of al-Umm and other books.”67 Farsy here probably refers to
the 1904 edition of Kitāb al-Umm, published by Būlāq Printing Press. In other
words; East African scholars were aware of, and in possession of the Islamic
books that were being printed in Egypt.
An interesting collection that reflects the “in-between” period between
manuscript and print can be found in the Azzawiya mosque library. In bound
volumes, several of the works and letters of ʿAbd Allāh b. ʿAlawī al-Ḥaddād
are collected, some printed, some photocopies of manuscripts that reside in
Mecca. Among them is a 1998 print of Aḥmad b. Zayn al-Ḥibshī’s commentary
on al-Ḥaddād’s poem al-ʿAyniyya, which, as shown above, was used by Ḥasan
Jamal al-Layl a basis for his claim to lineage-based authority.68 This early print
was produced by Maṭbaʿat al-Mīriyya in Mecca and we can only guess that this
version also was circulated in Zanzibar.

65 For an account of the titles used in early twentieth century education, see Loimeier,
Between Social Skills. See also Bang, “Authority and Piety”.
66 Sālim al-Bishrī was a contemporary of Muḥammad ʿAbduh and Shaykh al-Azhar from
1900–1904 and again from 1909–1916.
67 Farsy/Pouwels, The Shafiʿi Ulama, 106.
68 Aḥmad b. Zayn al-Ḥibshī, Sharḥ al-ʿAyniyya, Maṭbaʿat al-Mīriyya, 27 Dhū ’l-Ḥijja 1315/18
May 1898. In Cairo, only two of al-Ḥaddād’s works were printed before 1900; al-Daʿwa
al-Tāmma (1887, Maṭbaʿat ʿAbd al-Razzāk) and Nasāʾik al-Dīniyya wa’l-wiṣāya al-imāniyya
(1876, Bulaq Printing Press and again privately printed in 1888 and 1891).
132 chapter 6

A survey of the book collection of Burhān Mkelle (1884–1949) can serve as

an indicator of what types of books a “public intellectual” would have in his
collection.69 The collection consists of about 250 books (probably there were
even more, but the collection was partially dispersed after his death). Some
have inscriptions that indicate that they were originally owned by others. For
example, a part of the collection was first originally in the possession of the
Zanzibari qāḍī and scholar Saʿīd b. Dahmān (d. 1926). Others have inscription
that indicate how they came into Burhān’s possession – either through pur-
chase or – in at least one case – as a gift. Finally, several of the books also have
waqfiyya inscriptions where Burhān passed the books on as waqf for his chil-
dren and then for the benefit of all Muslims.70
Firstly, the collection contains the general works of fiqh, tafsīr and ḥadīth
that one would expect from a scholar with Burhān’s educational background.
Given Burhān’s interest in Arabic language and literature, there are also several
works of grammar. As a member of the Qādiriyya and the Shādhiliyya, Burhān
also owned a good collection of Sufi literature, including a four-volume anno-
tated edition of al-Ghazālī’s Ihyāʾ ʿUlūm al-Dīn from 1939.71 However, he also
owned several items that can be called “general literature”. For example, Burhān
owned a copy of Jazīrat al-ʿArab fī ’l-qarn al-ʿishrīn (The Arabian Pensinsula in
the twentieth century) by the Egyptian-born scholar and journalist, and later
Saudi diplomat Ḥāfiẓ Wahba (d. 1969).72
Furthermore, although the collection is clearly incomplete, it is evident that
Burhān subscribed to the journals that were being published in the Middle
East. In the collection can be found for example copies of al-Hilāl dating from
1913 to 1925, demonstrating an awareness of the nahḍa. Clearly, Burhān was not
the sole subscriber in Zanzibar or on the coast, and there is also every reason
to believe that the copies were circulated among other readers. Also present in
the collection are copies of the al-Azhar-published journal Nūr al-Islāmī, dat-
ing through the 1930s.
Finally, Burhān’s collection was not limited to works of Islamic law, history
or interpretation in Arabic. Among his books was the French translation of
Minhāj al-Ṭālibīn by L.W.C. van den Berg, which by the late nineteenth century

69 The collection was in the possession of the late Muhammad Idris Muhammad Salih,
Zanzibar. For the purposes of this survey, I have looked at books printed before 1360/1941 –
i.e. eight years before Mkelle’s death.
70 See discussion in Bang, “Authority and Piety”.
71 Iḥyāʾ ʿUlūm al-Dīn, with commentaries by ʿAbd al-Qādir b. Shaykh ʿAbd Allāh al-ʿAydarūs
and Shihāb al-Dīn al-Shuhrawardī, Maktabat Muṣṭafā al-Imbābī, Cairo, 1358/1939, Vols 1–4.
72 This work was first published in the 1940s and was republished in the 1950s and 60s.
travelling texts 133

was the most widely used reference work of Islamic law by colonial adminis-
trators in Dutch East India and in British India.73

Publication by East African Authors with Publishers in the Middle

East or Arabian Peninsula
We know little about how East African Islamic scholars came to have their
works published in printing presses in the Middle East. Most likely, con-
tacts with publishing houses such as al-Ḥalabī printers in Cairo, came about
through personal contacts, go-betweens and intellectual networks in general.
It also seems clear that there existed, at least by the 1920s, a group of individu-
als willing to sponsor the printing of locally-produced works. A glimpse of the
negotiating over printing can be gained from a letter found among the papers
of Burhān Mkelle, from “The Great Commercial Library” in Cairo (Maktabat
al-Tujjāriyya al-Kubrā), owned by one Muṣṭafā Muḥammad.74 In the letter,
dated 28 September 1922, Muṣṭafā Muḥammad explains that printing 1000 cop-
ies will be more profitable than printing only 500, as the extra is “only the cost
of paper”. Unfortunately, we do not know which text Burhān wanted to have
printed; it could have been one of his own works, or another text to be used in
the schools.

Written in Zanzibar, Published in Mecca: The Works of Aḥmad b.

Sumayṭ
Aḥmad b. Sumayṭ’s Manhal al-Wurrād75 was one of the earliest books of Shāfiʿī
scholarship by an East African scholar to be printed. It was published in Mecca
by Maṭbaʿat al-Mīriyya in 1315/1897–98. According to the final note, the writing
was completed in 1895, which would mean that it went more or less directly
to print in Mecca without first being circulated in manuscript form. No man-
uscript copies have been found in the Riyadha library nor in the collection
of Maalim Muhammad Idris deriving from the Madrasa Bā Kathīr – both of
which are institutions where this text could be expected to have been used.
Unfortunately, the edition says nothing of how exactly the work was brought
into print, such as who paid for it, or how contact with the Meccan publisher
was established. In this case, we can only guess that the contact came through

73 L.W.C. van den Berg, Le guide des zélés croyants. Manuel de jurisprudence musulmane selon
le rite de Châfi’î. Vol I&II, Batavia Government Print, 1882, 1884.
74 Maalim Idris Collection, Mkelle Papers, Letter from Muṣṭafā Muḥammad to Burhān
Mkelle, 28.09.1922.
75 Aḥmad b. Abī Bakr b. Sumayṭ, Manhal- al-Wurrād min fayḍ al-amdād bi-sharḥ abyāt
al-Quṭb ʿAbd Allāh b. ʿAlawī al-Ḥaddād. See A.K. Bang, Sufis and Scholars, 205.
134 chapter 6

Aḥmad b. Sumayṭ himself, through his long study periods in Ḥaḍramawt and
Mecca in the 1880s.
Furthermore, very little is known about the Maṭbaʿat al-Mīriyya. Most likely,
this was the name of the government printing press set up in Mecca under the
auspices of the Ottoman Sultan ʿAbd al-Ḥamīd. This press was also closely con-
nected to Aḥmad Zaynī Daḥlān, who (as outlined in Chapter 2) was a leader
for ʿAlawī scholars in Mecca, and whose close associates Aḥmad b. Sumayṭ met
during his sojourn in Mecca.76 As has been pointed out by U. Freitag, this press
also came to be very influential in the teaching of Islamic sciences in Mecca, as
it provided students with printed copies of the main books.77
Another early print by Aḥmad b. Sumayṭ was Tuḥfat al-Labīb.78 This work
was completed in 1911, and appeared in print by 1322/1913–1914–14. This time,
the publisher was Dār al-Kutub al-ʿArabiyya al-Kubrā, established in Cairo in
1276/1859–60 by Muṣṭafā al-Bābī al-Ḥalabī and his brothers. By the early twen-
tieth century, it was a well-established press, as evident from their post-script
blurb:

Anyone who is concerned with the East and with the Arab countries, will
know that Cairo (Miṣr) is the most important when it comes to the pub-
lishing of Arabic books, and that the most reputed publishers [in Cairo]
now are the Dār al-Kutub al-ʿArabiyya al-Kubrā.79

The “Dār al-Kutub” label specialized in books of ʿilm, and published an annual
catalogue, distributed free of charge. Most likely, Tuḥfat al-Labīb, like its prede-
cessor, went more or less directly to print and had only very limited circulation
in manuscript form.

76 In his portrait of Aḥmad Zaynī Daḥlān, C. Snouck-Hurgronje mentions different printing

presses in Mecca during the 1880s. One was apparently a “government print” opened in
1883 and comparable to the one in Zanzibar. Another one, (but quite possibly referring
to the same one) is simply referred to as “de jonge Mekkaansche drukkerij” (the young
Meccan printing press). C. Snouck-Hurgronje, “Een rector der Mekkaansche universiteit”,
Bijdragen tot de Taal-, Land- en Volkenkunde van Nederlandsch-Indië, 36:3, 1887, 344–404.
77 U. Freitag, “Der Orientalist und der Mufti: Kulturkontakt im Mekka des 19 Jahrhunderts”,
Die Welt des Islams, 43:1, 37–60, 43.
78 Tuḥfat al-labīb sharḥ ʿalā lāmiyāt al-Ḥabīb, Cairo, M. Dār al-Kutub al-ʿArabiyya al-Kubrā,
1332/1913–14. This, too is a commentary on a poem by al-Ḥaddād and deals with the
genealogy and spiritual leaders of the ṭarīqa ʿAlawiyya. See A.K. Bang, Sufis and Scholars,
206.
79 Publisher’s note on the Tuḥfat al-labīb.
travelling texts 135

This was not the case with some of Aḥmad b. Sumayṭ’s other works, which
were printed only after his death. For example, his commentary on Minhāj
al-Ṭālibīn entitled al-Ibtihāj fī bayān iṣṭilāḥ al-Minhāj80 was only published
in 1935, 10 years after the death of author. Again, it was printed in Cairo, by
the al-Ḥalabī press. The treatise seems to have been left unfinished by the
author’s death, and some of the gaps are filled by Ibn Sumayṭ’s son ʿUmar. The
posthumous publication was financed (ʿalā nafaqa) by one Aḥmad BāShaykh
al-Qaḥṭānī – most likely a relative of the abovementioned Muḥyī al-Dīn
al-Qaḥṭānī (see above, Chapters 2 and 3).
Another of Aḥmad b. Sumayṭ’s works took even longer to see print. His
Kawkab al-Zāhir was only printed in 1960, some 60 years after its completion
by the author – again in Cairo, but this time by al-Madanī printing press.81 This
was published together with a prayer for the Prophet, apparently written upon
the encouragement of Sayyid Manṣab b. ʿAbd al-Raḥmān and the Riyadha
Mosque. The belated printing would be typical of works that were composed
by a well-known and revered author, and widely circulated in hand-written
form, and possibly also orally.
Another example is that of a work by Aḥmad b. Sumayṭ’s father Abū Bakr
(d. 1873). He wrote a commentary on Irshād al-Muslimīn, a widely used intro-
ductory text in Quranic schools all over East Africa.82 The commentary, enti-
tled al-Tiryāq al-nāfiʿ min al-ʿammā, was spread in manuscript form throughout
the coast, but was first published some time around 1930–1940 (judging here
by the look of the book and print; the publication itself is undated) – again
by al-Ḥalabī publishers in Cairo. This time, the cost of publication was spon-
sored by one “Molla Karīmjee Molla Muḥammad Bihāʾī al-Baḥrī and his sons
in Zanzibar”.83 New prints of the same text followed in the 1950s, ensuring a
steady circulation for teachers to use in the Quranic schools. Publications of
this text has continued to the present.84
Finally, it should be noted that there are several well-known works that
never made it into print, despite being written by high-profile scholars and

80 Al-Ibtihāj fī bayān iṣṭilāḥ al-Minhāj, Cairo, Maṭbaʿat Muṣṭafā al-Bābī al-Ḥalabī wa-awlādihi,
1357/1937. In the collection of Maalim Muhamad Idris in Zanzibar.
81 Al-Kawkab al-zāhir ʿalā nasīm ḥāhir, Cairo (M. al-Madani), 1381/1961. In the collection of
Maalim Muhamad Idris in Zanzibar.
82 This text is known as the “Bāb mā jāʾ” or “Babu Majā” and its author is unknown but
probably from Shela, Lamu. See R. Loimeier, Between Social Skills, 175. The “Banu Majā” is
among the most frequently reprinted texts on the coast. Particularly widespread were the
publication by the Maktaba wa-Maṭbaʿat al-Hājji Muḥammad wa-awlādihi in Mombasa.
83 This edition is in the Maalim Idris collection.
84 For example, it was re-published in the 1990s by the Maṭbaʿat al-Khayriyya in Zanzibar.
136 chapter 6

being well-known – at least within the scholarly community. Perhaps the most
prominent examples are the works of Burhān Mkelle, especially Taʾrīkh Jazīrat
al-Qamar al-Kubrā (A history of Grande Comore) and his collected poetry
(known as the Burhāniyāt).85

The Publication History of Riḥlat al-Ashwāq al-Qāwiyya

As mentioned above (and in Chapter 2), ʿAbd Allāh Bā Kathīr set out for his
journey in Ḥaḍramawt and the Ḥijāz in March 1897, and returned to Zanzibar
in May/June 1898 (Muḥarram 1316).86 Fairly soon thereafter, he must have com-
pleted his account of the journey in manuscript form, most likely based on a
diary (the account is very specific on dates, times and locations). By 1899, man-
uscript copies of Riḥlat al-Ashwāq were already being produced in scholarly
centres on the coast. One example can be found in the Riyadha library, which
holds a manuscript copy of the Riḥla dated 27 Jumāda II 1317/1 November 1899.87
The relative speed of diffusion must have meant that this was a popular text
among certain members of the learned class, especially those with a Ḥaḍramī-
ʿAlawī background or training. Being a riḥla, it is also a relatively accessible
text, straightforward in language and content, and without convoluted argu-
ments, a point that may have added to its popularity.
If the manuscript moved quickly, this did not necessarily mean an equally
quick transition into print. The travelogue was first printed by Maṭbaʿat
al-ʿUlūm in Cairo in 1358/1939, a full 14 years after the death of the author.88
This was not merely a publication, but rather an edited and updated version as
is carefully explained on the frontispiece of the book:

Riḥlat al-Ashwāq [gives the full title] was in need of annotating and edit-
ing in print, and of some information on all those [people] mentioned
in it, or at least some of them, and a registration of some of those who
have died, insofar as it is possible. So, ʿAbd Allāh b. Ḥamad b. Ḥāmid b.
ʿUmar al-Saqqāf, master of the history of Ḥaḍramī poetry and of many
other works, proceeded to write those annotations aiming for brevity and
conciseness. [The annotations] were placed below the [text of] the Riḥla,
under a dividing line.

85 Manuscript copies in the Maalim Idris collection. Efforts are being made today to have
these works edited and published.
86 For an account of the journey, see A.K. Bang, Sufis and Scholars, 104–113.
87 EAP466_RM154, Manuscript, Arabic, 1317/1899, 120 pages.
88 Edition in the Riyadha Library, Lamu.
travelling texts 137

The publication was sponsored by [ʿalā nafaqa] Shaykh Saʿīd b. Muḥammad

al-Khishtī and Muḥammad b. ʿAbd Allāh Mūsā in Zanzibar. The edition does
not say which manuscript al-Saqqāf worked from, but he would certainly have
more than one to choose from as the Riḥla was duplicated in several copies.
What we see here is that the fact that a text was fairly widely read did not
necessarily mean that it was quickly moved into print. In other words: The cir-
culation of manuscript copies – especially of a relatively short text like the
Riḥla – was sufficient to saturate the “market” of potential readers in schol-
arly centres like Zanzibar, Mombasa and Lamu. From the introduction in the
printed version, the editor seems to have been mostly concerned with updating
the text – such as providing information on scholars who had died since the
events took place, detailing further information on their careers etc. However,
the printed version, like the manuscript, seems to have been circulated quite
widely and rapidly. We find the printed edition both in the Riyadha library,
the collection of Burhān Mkelle and – obviously – in the library of the former
Madrasa Bā Kathīr.
The Riḥla was reprinted again in 1984, based on the 1939 edition, but again
with “updates” on the various scholars and with a slightly different layout.

The Zanzibar Government Print and Educational Efforts

From c. 1905, education in Zanzibar was increasingly viewed as not solely the
responsibility of individual teachers in mosques, but rather as an effort to be
undertaken by institutions. Two main lines can be identified here; one which
aimed towards a systematic, curriculum-based Islamic education modelled
mainly on the Ḥaḍramī ribāts (exemplified by the Madrasa Bā Kathīr), and one
which accorded that Islamic education could also take place within the frame-
work of the emerging, colonial government schools (most clearly exemplified
through the career of Burhān Mkelle, but also Abū ’l-Ḥasan Jamal al-Layl).89 In
either case, what we see in the period is an institutionalization of the field of
the ʿulamāʾ.90 The gradual incorporation of Islamic subjects into the govern-
ment schools led to a small flurry of printing activity by the Zanzibar govern-
ment printed to produce educational material for instruction.

89 A.K. Bang, Sufis and Scholars; R. Loimeier, Between Social Skills.

90 A.K. Bang, “Hadrami teachers, scholars and educationalists. The role of Hadramis in
Islamic education in Zanzibar, c. 1870–1930,” Asian Journal of Social Science, 2/3, 2007,
435–456.
138 chapter 6

Title/Content Author Year of print

Murshid al-Fityān ilā ʿulūm al-bayān. Burhān Mkelle 1916

Intruction to the style and grammar of Note: Copies can not
Arabic poetry as well as recitation. For be found, only
advanced students. manuscript version
has been consulted
here.
Alfiyya al-Wādiha (Alfiyya Burhān Burhān Mkelle 1917
Mkelle) Note: Copies can not
be found, see ALA
IIIB/Loimeier, 206
Al-Tamrīn. Book of Primary Lessons on Burhān Mkelle 1918
Grammar,Part I Containing parts of Saleh b. Ali91
Speech.
Arabic primer for children.
Al-risālat al-jāmiʿa fī bayān uṣūl al-dīn Aḥmad b. Zayn 1926
wa’l-fiqh wa’l-taṣawwuf li-madāris al-Ḥibshī (d. 1733) (2nd ed. 1927)
Zinjibār/Al Risalat al Jamya. Kitabu cha Edited and trans-
kufundisha ibada za kiislamu katika lated by Aḥmad b.
vyuo vya Unguja. Abī Bakr b. Sumayṭ
Basic introduction to Islamic religious
duties, with the Ibāḍī rules for prayer
added to the Shāfiʿī.
Kitābul Tahara Abū Zakariyā 1927
(Extract from the chapter on ritual al-Nawwawī
purity from the Minhāj al-Ṭālibīn) Translated into
Note that edition gives the translation Swahili by Abū
into Swahili in both the latin and the ’l-Ḥasan b. Aḥmad
Arabic script. Jamal al-Layl and
Saleh b. Ali
Kitāb intishār al-khaṭṭ al-’Arabiyya. Burhān Mkelle 1939
Introduction to the Arabic script.

Figure 11 The Islamic educational publications by the Zanzibar Government Print

1900–1940s91

91 Saleh b. Ali was an employee of the British Residency. We can assume that he
was the “government arm” in the production of these booklets.
travelling texts 139

It should be noted that the Zanzibar Government Print also printed works that
were not directly meant for the educational sector. In fact, it seems that the
press could be used for any type of print, as long as there were ­benefactors
prepared to pay for the printing. For example, when the Ḥaḍramī ʿAlawī
teacher ʿAlawī b. Raḥmān al-Mashhūr (See above, Chapter 2) visited Zanzibar
in 1911–1912, he arranged to have his own work printed there and later sent
to him in Ḥaḍramawt. Interestingly, the work – a mawlid92 – was printed at
the Government Printer at the cost of (ʿalā-nafaqa) the Ikhwan al-Safa music
ensemble. We can only assume that the Ikhwan al-Safa had performed this
mawlid during ʿAlawī’s sojourn in Zanzibar and that they had appreciated it
sufficiently to sponsor the publication of the text. Given that the Ikhwan al-
Safa were themselves sponsored by the Sultan, we can assume that the spon-
sorship of this print indirectly came from Sayyid ʿAlī b. Ḥamūd.

Manuscripts, Printed Books and Religious Authority

The above samples go some way to answer Becker’s question about what books
came into the hands of local religious teachers and thus exercised a “direct
influence on the intellectual life of our colony”. The sample is evidently not
fully representative, as it is taken mainly from collections that were closely
connected to the Ḥaḍramī ʿAlawī tradition. However, it is still representative
in the sense that this tradition, in combination with the works of Shādhilī and
Qādirī scholars was the most influential on the coast in this period when it
comes to higher learning.
The first observation to be made is that manuscripts and printed books are
best understood as part of a broader pattern of reading, writing and collect-
ing of Islamic texts. Two main centres can be identified: Ḥaḍramawt, for the
manuscript period, and Cairo for the printed books. Mecca, too, was a cen-
tral marketplace for books, most likely in connection with the ḥajj or study
periods. In addition, we can also detect the impact of the flourishing printing
activities taking place in south-east Asia. This is especially so for the smaller,
educational works (such as al-Risālat al-Jāmiʿa) but – as we have seen in the
case of Fawāʾid al-Saniyya – manuscripts too could make their way from one
end of the Indian Ocean to another.

92 Al-Mashhūr, Lawāmiʿ al-Nūr, I, 91.

140 chapter 6

From Manuscript to Print: Same Words, New Authorities?

The transition from manuscript to print was slow, gradual and, by 1940, still
incomplete. Manuscript copies continued to play a role well into the twen-
tieth century. As we have seen, manuscripts copies were acquired from over-
seas as late as the last decades of the twentieth century (Fawāʾid al-Saniyya).
We have also seen that shorter devotional texts were produced and circulated
in manuscript form well into the 1930s and 40s. As has been shown elsewhere,
manuscript copies that were made waqf in the nineteenth century, came to
new uses in the 1930s when they were re-donated from private to institutional
ownership.93
Several explanations are likely to be at play here. First, East Africa (and espe-
cially coastal East Africa) did not see a proliferation of printing presses com-
parable to that in south- and south-east Asia. Thus, manuscript transmission –
especially in the educational context – was a question of necessity well into
the twentieth century. This is the period when we find manuscripts written in
modern, mass produced exercise books intended for secular (colonial) schools.
Secondly, not all titles were easily obtainable, even if they did exist in printed
versions. For example, despite being a well known text which was, as we have
seen, carefully transmitted in manuscript form, I have yet to see a copy of the
earliest printed version of Shams al-ẓahīra in East Africa, a lithograph printed
as early as 1911 in Hyderabad. This is all the more surprising as this first edition
was distributed free of charge.94 In other words: Despite the good networks of
individual scholars, not all books were obtainable. Nonetheless, in the case of
Fawāʾid al-Saniyya, one may equally well assume that the manuscript, by the
latter half of the twentieth century, was obtained due to its value as a heritage
object documenting the history of the ʿAlawī sāda or even as a family heir-
loom (in the widest ʿAlawī sense). Finally, it may also have been acquired for
its baraka. In these cases, manuscripts seems to have had (and still have) a lon-
gevity that surpassed their immediate “use value”, representing a way of com-
munication that did not vanish entirely with the age of print.
When it comes to the actual circulation of both manuscripts and printed
books, we have seen that manuscripts were bought overseas, copied locally,
given away as presents and donated as waqf. There are many indications, but
no concrete evidence of a genuine manuscript/book-market within East Africa
itself; none of the books described here or surveyed for this work show any sign
of having been exchanged for a price in Zanzibar, Lamu, Mombasa or else-

93 A.K. Bang, “Authority and Piety”.

94 The first edition was printed at the expense of one Sayyid ʿAbd Allāh Muḥammad
al-Ḥāshimī. E. Ho, Graves of Tarim, 346.
travelling texts 141

where on the coast. A few intriguing suggestions can be found, however, giving
us a few hints about a “book exchange” taking place within East Africa itself. For
example, on the front of EAC-075 of the Zanzibar National Archives, another
copy of a commentary on yet another of Ibn Malik’s grammatical works, the
second owner noted in 1864 that he bought the book at the “Zanzibar market”.95
Another suggestive note can be found on the Ibāḍī manuscript ZA 13/15. In 1912,
the unknown owner has noted: “I bought this book in a public sale at the Bayt
al-Ajāʾib.”96 However intriguing, these indications are too few and far between
to be conclusive, beyond stating that manuscripts had the potential to be, and
probably also were, commodities.

The Usage of Manuscripts and Printed Books

Turning to the usages of the textual material, we find a difference between the
manuscript and print period. Usage is closely connected to the value attached
to the manuscript, which by the mid/late nineteenth century was purely utili-
tarian and mainly educational. Moving towards the twentieth century, we see
the first examples of text moving more or less directly from manuscript to print
(such as Ibn Sumayṭ’s Manhal al-Wurrād), indicating that textual transmission
was aiming beyond local circulation and thereby beyond local concerns. The
text of Manhal al-Wurrād is clearly meant as a work that speaks to the wider
tradition of ʿAlawī literature, and to the Arabic Islamic textual tradition overall.
This may have been a function of the stature of Ibn Sumayṭ as a scholar, but the
examples also show that some of his works remained unpublished until well
after his death. In these cases, we can imagine that manuscript circulation was
considered sufficient for educational and devotional purposes. A final possibil-
ity, and certainly one that is very likely, is that funding for publication simply
could not be found.

Texts as a Basis for Religious Authority

Can we conclude from the books themselves how and to what extent they
functioned as a basis for religious authority? In some cases, it is evident from
the text itself, such as in the case of Marsūma al-ʿAyniyya which promotes
genealogy as a basis for authority, and which does so in a highly scriptural-
ized manner. However, on a more general level, the presence of quite substan-
tial amounts of text from the mid nineteenth century – combined with the
emergence of Islamic teaching institutions towards the end of the century –
show that authority was increasingly becoming text-based. However, the

95 L. Declich, The Arabic Manuscripts, 35.

96 L. Declich, The Arabic Manuscripts, 88.
142 chapter 6

examples given above also show that the Sufi orders (notably the Qādiriyya
and the Shādhiliyya) also promoted a text-based type of authority, insofar as
the writing down of dhikr and prayers promoted an authority based on Arabic
literacy. Other devotional texts seem to have had the same effect, moving from
previous existence as orally transmitted recitations to manuscripts to note-
books to pamphlets. The same can be said for the activities of Burhān Mkelle
and his compatriots within the framework of the Government Schools. Here,
too, emphasis was on reading and understanding the textual foundations
of faith.
 chapter 7

Ritual of Reform – Reform of a Ritual

Rātib al-Ḥaddād in the Southwestern Indian Ocean, c. 1880–1940

This chapter picks up the discussion from the previous chapter, where increased
textual transmission of Islamic knowledge was shown to form alternate types
of authority in East African Muslim communities in the late nineteenth and
early twentieth century. However, as was pointed out with reference to devo-
tional literature, oral transmission continued alongside the written, and par-
ticularly so in settings such as the Sufi dhikr or prayers. This chapter focuses
entirely on one such text and ritual, the text known as Rātib al-Ḥaddād com-
posed by the Ḥaḍramī poet and Sufi ʿAbd Allāh b. ʿAlawī al-Ḥaddād (1634–1719).
Rātib al-Ḥaddād is a dhikr – a “remembrance of God” performed by Muslims
in large parts of the Islamic world, but it is particularly widespread in the
Indian Ocean region. It is recited by men and women, in mosques or private
homes, to express devotion and to seek blessing. It is thus a ritual that in the
context of a group, a local community or a segment of a population makes up
part of what it means to be a Muslim.
In this chapter, Rātib al-Ḥaddād will be viewed as both text and ritual, focus-
ing on the relationship between textual transmission and performance, and
how this changed in the late nineteenth and early twentieth century Lamu,
Zanzibar and Cape Town.

Rātib al-Ḥaddād as Sufi Reform

The use of devotional litanies has, at least since the eighteenth century, been
one of the main hallmarks of the ṭarīqa ʿAlawiyya, in East Africa as elsewhere
around the Indian Ocean. As a result, the text of Rātib al-Ḥaddād conveys and
expresses spiritual, social and cultural value for people in locations as diverse
as Tarim (Yemen), Singapore, Jakarta (Indonesia), Diego Suarez (Madagascar)
and Maputo (Mozambique). It is, in other words, one of the widely practiced
Indian Ocean expressions of Muslim devotion, and one of the better known
texts.
Taking as a starting point that the Rātib, by the mid/late nineteenth century,
was an established part of ritual practice (at least among ʿAlawīs) in East Africa,
this chapter traces changes in transmission, practice and perceived meaning

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�08

144 chapter 7

in light of the wider reformist wave that started at the time of al-Ḥaddād him-
self. What changes can be traced in educational and textual transmission as
well as recited performance in Lamu, Zanzibar and Cape Town in the period
c. 1880–1930?
There are two major problems with this approach. First, while we can trace
the Rātib in writing, it is very difficult (if not impossible) to establish how a
recital was performed more than a hundred years ago. However, oral history
and present-day practices go some way to establish the history of recital in spe-
cific locations. Insofar as possible, written sources have also been consulted.
Secondly, given that the ritual practice, transmission and meaning may vary
over time, one must also take into account that it may vary from one location
to the next, and that the researcher may interpret as change what in fact is
local variation. These variations, in turn, may determine social as well as per-
sonal interpretations. For example, in Cape Town, singer Wahied Kannemeyer
remembered “visiting a different house every Thursday evening, to participate
in Rātib al-Ḥaddād”.1 For Kannemeyer, the recitals are remembered with over-
tones of anti-apartheid resistance, sentiments that obviously would be absent
outside that particular South African context. Another example: Several studies
have demonstrated how the Islamic text – but more especially its accompany-
ing recital, be that of the Quran itself or of a dhikr or other types of prayer –
played a vital part in the integration of newly converted African Muslims into
Swahili society in East Africa. Their expression of devotion was also an expres-
sion of a new identity – a perceived meaning that would not be present for
example in Tarim, Ḥaḍramawt.2 This problem will be sought overcome by his-
toricising the ritual in each location.

The Author of the Rātib: ʿAbd Allāh b. ʿAlawī al-Ḥaddād

ʿAbd Allāh b. ʿAlawī al-Ḥaddād was born in 1044/1634 in al-Sabīr in the out-
skirts of Tarim in Ḥaḍramawt. He went blind around the age of four and died
in Tarīm in 1719, at the age of 85.3 His legacy has been of vital importance to the
development and spread of the ṭarīqa ʿAlawiyya, and he is generally referred to

1 http://www.astrolabe.com/product/2529/Divine_Blessings.html.
2 See for example J. Glassman, Feasts and Riot; F. Becker, Becoming Muslim.
3 On his life and works, see al-Mashhūr, Shams al-Ẓahīra, 568–571 and passim. See also Mostafa
al-Badawi, Sufi Sage of Arabia. For an analysis of the importance of al-Ḥaddāḍ as a reformist
figure, see U. Freitag, Indian Ocean Migrants, 92–96. For an overview of his writings see GAL
II, 407–408, SII, 566; Kaḥḥāla, vi, 85 and Ziriklī, iv, 104.
ritual of reform – reform of a ritual 145

within the ʿAlawī tradition as the quṭb (spiritual pole or axis) and the mujaddid
(renewer) of the 12th Islamic century.4
The work of al-Ḥaddād and his students has also, at least in retrospect, been
interpreted as a genuine revival of the ʿAlawiyya and his works have also been
interpreted as bringing forth the first wave of reformist ideas within the ṭarīqa.
The fame of al-Ḥaddād stems from his personal piety, his karāmāt (supernatu-
ral events linked to his actions) and his role as a teacher, but above all from
his poetry. The recitation of his awrād (devotional litanies) quickly became
integral elements of the ʿAlawī Sufi way. His best-known work is the long
poem known as Qaṣīdat al-ʿAyniyya, which recounts the history of the ʿAlawī
sāda, their homeland and teachings, while incorporating the essence of their
religious tenets.5 As we have seen above (Chapter 6), the poem was read and
commented upon also by East African ʿAlawī scholars. Ḥasan b. Muḥammad
b. Ḥasan Jamal al-Layl used it as a basis for establishing his own genealogy, in
a direct bid for religious authority. The commentaries of Aḥmad b. Sumayṭ
on the poetry of al-Ḥaddād offers more general interpretations, but these, too,
have a underlying subtext of authority based on Prophetic decent.
The works of al-Ḥaddād are also generally acknowledged for being not
overly long and convoluted, but rather brief and clear in order to instil in the
reader the “firmest possible foundations for faith.”6 To this day, al-Ḥaddād
is understood as the “ḥaddād al-qulūb”, the smith of hearts, alluding to the
name Ḥaddād also meaning blacksmith.7 His texts are not read as intellectual
works, but as words of devotion that speak directly to emotions, to form and
strengthen the faith of all believers. They are understood as texts that instil
in the faithful the bond between himself and the creator, and thus (at least in
theory) accessible to all, regardless of class, education and ethnic divisions.
As was outlined in Chapter 2, al-Ḥaddād also emphasized missionary
(daʿwa) activity, and particularly “inner mission”, such as preaching to people
who were long since Muslims. His activities also included social work, which
was later taken up by his students.8 This is what in turn has been interpreted
as the first wave of reform within the ṭarīqa ʿAlawiyya, a gradual shift towards

4 Mostafa al-Badawi, introduction to English translation of ʿAbd Allāh b. ʿAlawī al-Ḥaddād, The
Book of Assistance, Fons Vitae Spiritual Masters Series, 2005, vii–viii.
5 A commentary on Qaṣīdat al-ʿAyniyya quickly followed, composed by al-Ḥaddād’s student
Aḥmad b. Zayn al-Ḥibshī (d. 1733), known as the Sharḥ al-ʿAyniyya.
6 Mostafa al-Badawi, Sufi Sage, vii.
7 Interview, Muhdhar Khitamy, Mombasa, May 3rd, 2010. As for the reason why his family line
came to carry this name, see Mostafa al-Badawi, Sufi Sage of Arabia, 6–7.
8 U. Freitag, Indian Ocean Migrants, passim.
146 chapter 7

social activism. The Rātib al-Ḥaddād was part and parcel of this development,
reaching beyond the practices of the ʿAlawiyya to the Muslim world in general.

Rātib al-Ḥaddād, the Text

Rātib al-Ḥaddād is a typical example of the devotional and accessible nature of
al-Ḥaddād’s orientation, and by far his most widely known and recited work.
The Rātib, as the word implies (Ar: rātib: ordering, structuring) combines verses
of the Quran with poetic prayers (duʿāt) and praise of God. It starts with the
Fātiḥa (the opening verse). It then moves on with the Āyat al-Kursī (“the verse
of the throne”; Quran 2:255), the verse considered to be the most beneficial for
recitation, bringing the most blessing upon the believer because it contains
the name of God more frequently than any other verse.
Then follows the last two verses of the Sūrat al-Baqara. These are verses
of a penitential nature, focused on seeking forgiveness. They are also among
the most widely used Quranic passages in all Islamic devotional text,9 and as
such, Rātib al-Ḥaddād is similar in nature (and in some cases also in composi-
tion) to other devotional texts within and beyond the ṭarīqa ʿAlawiyya. The
similarities to the use of Quranic passages in the adhkār of the Shādhiliyya and
the Qādiriyya are very evident. This is not surprising given the close lineage
connection between the ʿAlawiyya and the Shādhiliyya. It is also reported that
ʿAbd al-Qādir al-Jīlānī was an important figure in the life of ʿAbd Allāh b. ʿAlawī
al-Ḥaddād. In fact, al-Ḥaddād considered al-Jīlānī among his Sufi masters.10
Next follows the affirmation of the unity of God through the Kalimat
al-Tawḥid; the statement that God is one and has no partner, no equal and that
He has the power of everything, to be repeated three times.
The text then moves on to various invocations (tasbīḥ; Ar: proclaiming the
glory of God through variation of the phrase “Subḥān Allāh”) and prayers to be
recited a specific number of times, usually three or four repetitions, but some-
times more, glorifying the name of God, praising, taking refuge from evil, etc.
Most of these invocations originate in ḥadīth quotations.
However, the distinct ʿAlawī Sufi nature of the text also comes through as
the text contains supplications for blessings not only on behalf of the Prophet,
but also for his forefathers and descendants among the Bā ʿAlawī, as in verse 27:

9 C. Padwick, Muslim Devotions. A Study of Prayer Manuals in Common Use, Oxford

(OneWorld) 1996 (1st. ed. 1961), 116–119. The other two most widely used Quranic passages
are the Sūrat al-Ikhlāṣ and the Sūrat YāSīn.
10 Mostafa al-Badawi, Sufi Sage of Arabia, passim.
ritual of reform – reform of a ritual 147

(Recite the) Fātiḥa on the noble soul of our leader, the jurist and head,
Muḥammad son of Ali Bā ‘Alawī, and on the soul of all his forefathers and
offspring and all those who they owe any right; may the almighty Allāh
forgive and grant them mercy and elevate their position in the Jannah;
make us benefit from their blessings, secrets, luminescences and their
knowledge, in our religious and worldly affairs and in the Hereafter.11

It is worth noting that the text particularly mentions Muḥammad b. ʿAlī Bā

ʿAlawī, known as within the ʿAlawī tradition as al-Faqīh al-Muqaddam or “the
first jurist” (d. 1255). Al-Faqīh al-Muqaddam is considered to have been the
initiator of the ʿAlawī Sufi path and he is also at a “crossroads” of the ʿAlawī
descent lines, so blessings supplicated in his name invokes the entire ʿAlawī
descent line. The implication is that spiritual authority and mystical insight is
transmitted in the entirety of the Prophetic bloodline.12 The following verses
reinforce the emphasis on bloodline transmission, reciting fātiḥas for the “Sufi
masters” and, of course, al-Ḥaddād himself.
In line with al-Ḥaddād’s general style, the Rātib – on the face of it – is a fairly
simple text that can be learnt by without too much effort. Its apparent simplic-
ity is part of its appeal; being understood as personal, something close to the
individual believer. In the words of Shaykh Seraj Hendricks, “It embraces you
almost more than you embrace it. It does not demand a lot from you, it is like a
calm pool, and you walk calmly into it because you see that it is safe. And you
emerge refreshed.”13
The text of Rātib al-Ḥaddād appears, in other words, to have been written
for the general public, as an expression of devotion and with sensitivity to their
needs and aspirations. This is understood as a reflection of the reformist orien-
tation of al-Ḥaddād; the will to incorporate new groups into existing practices
and to add new practices to strengthen faith.
However, it is important to note that to educated Sufi shaykhs, the Rātib is
also understood as a gateway to metaphysical insights. It is perceived as a text
that must be understood at different levels, and the understanding of the text
is dependent on what stage on the Sufi path (maqām) the individual seeker
has reached. Thus, the educated (and very Sufi) view is that the Rātib carries

11 Translation from G. Muhamed (Ed.), A translation and transliteration of Athkaar, Duaas

and Salawaat, including the Ratibul Haddad, Cape Town (Digital Action), 2010, 97. See also
numerous translations on the web, into English, Bahasa Indonesia and others.
12 On the position of al-Faqīh al-Muqaddam, see E. Ho, The Graves of Tarim, passim and A.K.
Bang, Sufis and Scholars, passim.
13 Interview, Seraj Hendricks, 07.12.2010.
148 chapter 7

meaning on two levels; the immediate one where devotion is expressed (for
the general public, the ʿawāmm), and the mystical, where an esoteric truth is
conveyed to those able to plummet it (the advanced seekers, the select few, the
khawāṣṣ). The Rātib is thus understood as a combination of something acces-
sible, easy and – by implication – communal; and something very exclusive,
accessible only to an elite and – by implication – individual, or even personal.
This combination of something widely accessible and a truth that needs pro-
tection has caused tensions in the diffusion and the teaching of the Rātib, as
will be discussed below.

Rātib al-Ḥaddād in East and South Africa

It is not known exactly when Rātib al-Ḥaddād first was recited in East Africa,
but a likely guess is that it came with migrants shortly after it was composed by
al-Ḥaddād himself, some time between c. 1655 and 1710 (when al-Haddād was
between 20 and 65 years of age). This coincides approximately with the first
migrations of the Āl Shaykh Abī Bakr b. Sālim to Pate in the mid-seventeenth
century, but, as discussed in Chapter 2, these migrations are shrouded in leg-
end and dates should be treated with caution. However, given the closeness of
the Āl Shaykh Abī Bakr b. Sālim with al-Ḥaddād and the scholars surrounding
him, it is not unlikely that the migrants from this family took the prayer to their
first settlement in Pate.
Other possible early transmitters of Rātib al-Ḥaddād include the Jamal al-
Layl family, who (as described in Chapter 2) settled in Pate in the mid-16th
century. Their back-and-forth migrations between Ḥaḍramawt, Pate/Lamu,
Zanzibar, the Comoro Islands and Madagascar over the next centuries is likely
to have contributed to the diffusion of the ritual.
Turning to the late nineteenth and early twentieth century, we find that C.H.
Becker does not mention the Rātib among the textual material he examined.14
As mentioned above (Chapter 6), he does, however, note two copies of the
Dalāʾil al-Khayrāt which has later tended to be instructed “hand-in-hand” with
Rātib al-Ḥaddād in the higher classes of East African madrasas. It’s presence
among the material examined by Becker most likely means that the Rātib is
absent simply by chance.
When it comes to the recital of the Rātib, it is of course impossible to know
how, by whom and at what times the text was recited in the seventeenth and
eighteenth centuries. A likely guess is that Rātib al-Ḥaddād was recited both

14 C.H. Becker, “Materials for the Understanding”.

ritual of reform – reform of a ritual 149

in mosques and in private houses. It is impossible to know if the recitals fol-

lowed a systematic pattern (at regular hours, on regular days), but we may
assume that it was most commonly recited by evening time, probably after
ʿishāʾ prayers, as per the instructions of al-Ḥaddād himself.
However, the institutionalized form of recitation, which also included ele-
ments of education to incorporate young people or new Muslims, seems to have
developed much later, most likely by the mid-nineteenth century. This explicit
“scheduling” of a devotional litany must be understood in connection with the
institutionalization of Islamic learning in the Islamic world in general and in
Ḥaḍramawt and Mecca in particular, as was outlined in Chapter 2. The com-
munal, and organized system of recitation that defines Rātib al-Ḥaddād today
was only fully formulated in Ḥaḍramawt and Mecca by the mid-­nineteenth
century, more than a century after the death of its author.
A central figure in this process was Aḥmad Zaynī Daḥlān (1817–1886) in
Mecca, who, as shown in Chapter 2, was not only the teacher of an entire gen-
eration of Ḥaḍramī ʿAlawīs settling in Mecca, but also of East African scholars
who spent study periods in the Holy City. The notion that the organized pat-
tern of reciting, and implicitly also teaching, the Rātib should have spread to
East Africa via Aḥmad Zaynī Daḥlān in Mecca may seem somewhat surprising,
given the high frequency of migration directly from Ḥaḍramawt to the urban
settlements of East Africa in the nineteenth century. However, the educational
changes that took place in Ḥaḍramawt were mirrored by, or even modelled on,
those undertaken by Aḥmad Zaynī Daḥlān and his associates in Mecca. As was
shown in Chapter 2, the frequent intellectual exchange between East Africa
and Mecca makes it likely that the recitation- and prayer schedules practiced
in Mecca were repeated by East African returnees.
The recital of Rātib al-Ḥaddād was further popularized in twentieth century
East Africa through the efforts of the Egyptian ʿālim Ḥasanayn b. Muḥammad
al-Makhlūf (1890–1990). Al-Makhlūf, like his father and fellow Muḥammad
Ḥasanayn al-Makhlūf (1861–1936), was an al-Azhar graduate.15 He was also
the grand Mufti of Egypt in the period 1946–50 and again in 1952–54 and in
1962, he was one of the founding members of the Muslim World League. While
his father took a clear anti-Wahhābī stand (expressed through writing pub-
lished in the 1920s), Ḥasanayn al-Makhlūf only went as far as sanctioning prac-
tices such as grave visitation and prayer for intercession. He was a member

15 On the life and career of al-Makhlūf, see J. Skovgaard-Petersen, Defining Islam for
the Egyptian State. Muftis and Fatwas of the Dār al-Iftā, Leiden (Brill), 1997, 170–180;
R. Schulze, Islamischer Interntionalismus im 20. Jahrhundert: Untersuchungen zur
Geschichte der Islamischen Weltliga, Leiden (Brill), 1990, 123–125 and passim.
150 chapter 7

(although not an initiate) of the Egyptian Khalwatiyya, although, as noted by

J. Skovgaard-Petersen, “he supported the reformed and moderate version of
Sufism.”16 Interestingly, he also maintained a good and long-lasting relation-
ship with the Saudi monarchy. He was a conservative scholar and seems to
have been particularly inspired by the works of al-Ḥaddād. In his foreword to
a new, Arabic edition of al-Ḥaddād’s The Lives of Man, al-Makhlūf wrote: One
of the best things I have ever come across and been guided to, among the trea-
sures left behind by [. . .] al-Haddad, is [. . .] The Lives of Man.” In this foreword,
as well as in several of his other works, al-Makhlūf essentially endorses that
spiritual benefits may be reaped from litanies such as Rātib al-Ḥaddād.
Again, the involvement of an Egyptian scholar may seem surprising given
the rich fabric of direct exchange between East Africa and Ḥaḍramawt. The
importance of al-Makhlūf, however, must be linked directly with the twen-
tieth century emergence of the Egyptian printing presses and the diffusion
of printed material, as was described above (Chapter 6). Al-Makhlūf’s father
has been a part of this movement and one of the central figures behind the
efforts to provide reprints of important works at the Būlāq Printing Press.
Ḥasanayn al-Makhlūf’s importance must also be understood in the framework
of the tradition following from Muḥammad ʿAbduh and the impact of Dār
al-Iftāʾ al-Miṣriyya – in other words the “global Islam” as outlined in Chapter 1.
As we saw in Chapter 2, East African scholars like Aḥmad b. Sumayṭ would
turn to ʿAbduh and the Dār al-Iftāʾ to clarify questions of fiqh. In other words:
Al-Makhlūf first and foremost made the Rātib “acceptable”, well into circles
that by the 1940s and 50s had developed clear modernist- and/or Salafi lean-
ings. What we see, in other words, is a process whereby a practice that was long
since part of the East African tradition became “re-packaged” and sanctioned
by the Ḥaḍramī, Meccan or Egyptian ʿulamāʾ, first in the 1860s and 1870s, and
then again in the twentieth century, and thus sanctioned as part of what it
meant to be an East African Muslim.

Lamu and the Riyadha Mosque

In Lamu, the recitation of Rātib al-Ḥaddād became strongly (but far from
exclusively) associated with the Riyadha Mosque. Given the local impact of
the Riyadha in the early twentieth century, Rātib al-Ḥaddād emerged as a
text that, in the local context, came to define “being Muslim”. The practice
in Lamu seems to have started before the actual inauguration of the Riyadha
mosque. The way its history is remembered in the Riyadha mosque, communal
recitation took place already in the 1880s when Habib Saleh settled in Lamu.

16 J. Skovgaard-Petersen, Defining Islam, 180.

ritual of reform – reform of a ritual 151

The Rātib also has a long history in several other mosques in Lamu, among
them Mskiti Rawda and the mosque in Shela.
In the Riyadha Mosque, Rātib al-Ḥaddād probably was, already from the
start, recited after ʿishāʾ prayers, except for Fridays and during Ramaḍān.17 The
system is likely to have been quite similar to today: upon finishing the regular
prayers, those present could either take their leave directly, or choose to stay
for the recitation. As several informants indicated, “most stay”, leading to the
conclusion that the Rātib in the Riyadha context is considered as an almost
integral part of the obligatory evening prayer. This pattern seem to have been
unchanged for a long period, as several informants of at least 60 to 70 years
of age could not remember there having been any change in this pattern dur-
ing their lifetime. This stands in contrast to for example the Mandhry mosque
in Mombasa (to which the Riyadha is strongly connected through family ties)
where the timing of the ritual actually has changed. There, too, it used to be
recited after ʿishāʾ but is now recited immediately after maghrib prayers.18
As for teaching Rātib al-Ḥaddād, all teachers interviewed agreed that the
Rātib, for as long as it has been practiced, has been most simply learnt through
listening, through internalizing the sounds, intonations and rhythm repeat-
edly. However, as will be discussed below, the text of the Rātib was also copied
in manuscript form, either for the process of memorizing, to ensure proper
vocalization, or for further educational use.

Rātib al-Ḥaddād in Zanzibar

It is impossible to know when Rātib al-Ḥaddād first became widely recited
in Zanzibar. What is known is that by c. 1900, it was recited regularly in the
Madrasa Bā Kathīr and in the Gofu mosque.
The main problem with obtaining an oral history of the Rātib in Zanzibar
is the 1964 revolution. The post-revolutionary climate created an entirely
new setting for religious discourse, and for a period even outlawed it entirely.
Moreover, the revolution led to the expulsion of many of the Rātib leaders, such
as for example ʿUmar b. Sumayṭ.19 By 1970, many of those who had led prayers

17 In Ramadan, the tarāwīḥ Quranic prayers are recited instead.

18 Interview, Muhdar Khitamy, Mombasa, May 3rd, 2010.
19 Fact and figures about the Zanzibar revolution are notoriously hard to find and verify and
I will not even try to give an estimate here of how many mosque/prayer/religious leaders
who left Stone Town in the period 1964–1970. It has been indicated that as many as half of
the approximately 2000 houses in Stone Town were vacated following the revolution. As
many of these were inhabited precisely by people who would be considered “leaders”, one
can imagine a broken chain in the form of a series of empty buildings in the 1960s and 70s.
152 chapter 7

in the Stone Town mosques had left (some to Kenya, others to Ḥaḍramawt,
to Europe, Canada or to Oman/the Gulf, yet others to the hereafter – either
as direct casualties of the revolution or by natural causes in the years that
followed).20

Rātib al-Ḥaddād in Cape Town

In Cape Town, Rātib al-Ḥaddād has long traditions. In fact, Seraj Hendricks, in
his thesis on Cape Sufism, indicates that the text may have been introduced to
the Cape as early as with Tuan Sayyid ʿAlawī in the mid-1700s.21 Tuan Sayyid
ʿAlawī was deported by the Dutch from Indonesia and arrived in the Cape
in 1744. There, he spent 10 years imprisoned on Robben Island before being
released and taking up the role as preacher to the Muslim community from
the 1750s and until his death in 1803. Being of Ḥaḍramī ʿAlawī extraction, Tuan
Sayyid ʿAlawī is likely to have been introduced to Rātib al-Ḥaddād, either in
Indonesia or during a sojourn in Yemen.
As S. Hendricks has pointed out, litanies such as Rātib al-Ḥaddād would –
due to their simplicity – be exactly the type of knowledge Tuan Sayyid ʿAlawī
would have passed on. This trajectory is certainly possible, given the migra-
tion pattern of Ḥaḍramī ʿAlawīs to Indonesia during this period, although it
presupposes that the Rātib spread to Indonesia, became part of practice, and
then was brought onwards to South Africa within the space of approximately
40–50 years.
Like in East Africa, the Rātib, during its early years did not have the orga-
nized and communal form by which it came to be known in the Indian Ocean
regions in the nineteenth century. It is not clear when this form of the practice
reached Cape Town, but we may assume that travellers such as those encoun-
tered in Chapter 5 played a role in establishing a new, scheduled pattern. In
twentieth century Cape Town, the Rātib was widely practiced, both in private
homes and in mosques, as indeed it is today. Private recitations are usually
held to mark private events, such as the commemoration of a deceased p ­ erson,

For one interpretation, see R. Loimeier, “Memories of Revolution. Zur Deutungsgeschichte

einer Revolution (Sansibar 1964)”, Africa Spectrum, 41:2, 2006, 175–197.
20 Much research remains to be done on the Zanzibari diaspora. For reference here, see f.
ex. E. Gilbert, “Oman and Zanzibar. Historical roots of a Global Community”, in: Ray, H.P.
& Alpers, E.A. (eds.) Cross Currents and Community Networks. The History of the Indian
Ocean, Oxford (Oxford Univ. Press), 2006. For the Zanzibaris in Oman, see also A.K. Bang,
“Zanzibar-omanere og Oman-zanzibarere. Bindestreker over det Indiske Havet – og
videre”, Babylon, 5:2,2007, 98–104.
21 S. Hendricks, Taṣawwuf (Ṣufism): Its role and impact, 45 and 215–230.
ritual of reform – reform of a ritual 153

or on more irregular occasions, such as the inauguration of a new house.22 In

terms of schedule, it is organized along the same patterns as in all other loca-
tions, with recitations on Thursdays, 3–7 days after a death, and on the anni-
versary of a death.23 The Thursday pattern may have been implemented by
Muhammad Salih Hendricks, as we know that recitations were being held on
Thursdays after ʿishāʾ shortly after the foundation of the Azzawiyya.

The Rātib in Writing: Textualization of Charisma

From the mid-nineteenth century until the present, Rātib al-Ḥaddād is one of
the most widely copied and distributed devotional texts in coastal East Africa.
Before the “age of writing” (see above, Chapter 6) one must assume that Rātib
al-Ḥaddād in the southwestern Indian Ocean was first and foremost trans-
mitted orally, although manuscript copies are also likely to have circulated in
private possession.
The earliest extant manuscript version of Rātib al-Ḥaddād on the East
African coast is (as far as can be ascertained at present) a copy in the Zanzibar
National Archives dating from 1860. Interestingly, this copy was made by a
member of the Ibāḍī Mundhirī family, who were close to the Bū Saʿīdī rulers,
indicating that either the Ibāḍīs, too, were reciting the Rātib, or that the con-
version of prominent Ibāḍīs to Sunnism started shortly after the transfer of the
Bū Saʿīdī capital to Zanzibar.24
The Riyadha Mosque library also contains several manuscript copies of
Rātib al-Ḥaddād, but these are later, dating from 1920s to the 1930s (typically
copied together with the Dalāʾil al-khayrāt or other devotional texts). The
books are daftars (students notebooks), more or less carefully written up, but
clearly written for the purpose having the text of the Rātib accessible, either
for educational or devotional purposes, or simply for the process of memoriza-
tion. Their very presence indicates two things. Firstly, that the Rātib was being
taught through writing, i.e. that students (and teachers for that matter) would
choose to write the text down, in order to memorize or for later rehearsal. It is
also possible that the act of writing in itself was understood as an act of piety
and devotion. The act of writing, in this case, was certainly not necessary, in
the sense that there were many, many people around who knew the Rātib by

22 Interview, N.B. al-Saqqaf, Cape Town, February 2005 and August 2005.
23 Interview, Nabiel al-Saqqaf, 21.08.2005.
24 ZA8/41: L. Declich, The Arabic Manuscripts, 80. I have tried to access this manuscript, but
it was unavailable.
154 chapter 7

heart. Secondly, the act of writing also indicates that the printed leaflet-ver-
sions that are ubiquitous today, probably were not so during the first decades
of the twentieth century.
In the twentieth century, Rātib al-Haddād has been printed repeatedly
throughout the Indian Ocean, and there are publications with both interlin-
ear and additional translation into English, French, Malay, Swahili, Afrikaans –
to mention a few. Rātib al-Haddād also has a number of commentaries, most
of them printed. The most widely known is that by another Ḥaḍramī ʿAlawī,
ʿAbd Allāh b. Aḥmad Bā Sudān (d. 1266/1849). The only known commentary
by an East African author was produced in manuscript form by the renowned
reformer of Swahili Islam, Muhammad al-Amin al-Mazrui (d. 1947).25 A third
widely used commentary today is a more recent sharḥ by ʿAlawī b. Aḥmad
al-Ḥaddād.26
It is not entirely clear when the first printed versions of Rātib al-Ḥaddād
were available in East Africa, nor when the first ones were produced by East
African printers. In the latter half of the twentieth century, a widely circulated
edition was printed by the Madrasat al-Nūr al-Islāmiyya in Malindi.27 This ver-
sion had an introduction in Swahili in the Arabic script, while rendering the
Rātib itself without translation.
The East African printed editions were relatively late compared to Cape
Town, where printing of Islamic teaching material started in the early twenti-
eth century. In 1910, a translation of the Rātib into Afrikaans in the Arabic script
was produced in Cape Town by one Shaykh Taha Gamieledien.28 As mentioned
above, the Azzawiya too, produced transliterated copies in the 1920s, where
transliteration was rendered as Ratiep AlGaddat or Raatieboel-Gaddad.
This, too was relatively late compared to the situation in south-east Asia,
where this type of “leaflet-print” can be found from the 1870s and onwards.
It is not clear when the first printed version of Rātib al-Ḥaddād appeared in
south-east Asia, but as early as 1875, printing presses in Batavia/Jakarta were

25 As late as 1947, the summary was reproduced again in manuscript form, copied by ʿAlawī
b. ʿAbd al-Wahhāb Abī ’l-Ḥasan Jamal al-Layl. R.S. O’Fahey et. al. ALA IIIB, typescript,
Chapter 2, 6. Unfortunately, I have been unable to consult this manuscript.
26 ʿAlawī b. Aḥmad b. al-Ḥasan b. ʿAbd Allāh al-Ḥaddād BāʿAlawī, Sharḥ Rātib al-Ḥaddād,
Tarim/Singapore, 1425/2004.
27 The latest version of the Madrasat al-Nur print dates from 1408/1987–88 and is the fifth
edition. I am grateful to Gerhard Bruinhorst, Leiden University, for providing me with a
copy of this publication.
28 Leaflets, dated 1910 and 1935. Another widely used printed version was produced in 1968
by Abbad Cloete. This was reprinted in the 1980s. I am grateful to Dr. Shamil Jeppie for this
information and for a copy of the editions.
ritual of reform – reform of a ritual 155

Image 5 An early printed version of Rātib al-Ḥaddād, Cape Town.

156 chapter 7

producing lithographs of closely related textual material, such as for example

al-Risālat al-Jāmiʿa by al-Ḥaddād’s student Aḥmad b. Zayn al-Ḥibshī..29 For
Lamu and Zanzibar, it is fair to conclude that textual transmission of the
Rātib took place in handwritten form well into the twentieth century, before it
was eclipsed by booklets printed in India, south-east Asia or East Africa.
In Cape Town, locally printed copies were supplanting manuscript versions by
the 1920s.

The Rātib Performed: Reform of a Ritual?

As outlined above, we see in the late nineteenth and early twentieth century a
process whereby the Rātib became “institutionalized” in new teaching institu-
tions and mosques. The question can be asked how, of if at all, this altered the
perceived meanings attached to recitation? And: If so, can these changes too
be understood as part of a broader reformist agenda?
Before continuing, some general observations are in order about the Rātib
recital. First, it should be noted that when recited, Rātib al-Ḥaddād is not
necessarily filled with any esoteric significance. The main meaning and ulti-
mate goal of the recitation of the Rātib is to bring blessing to the individual or
group reciting. As such, and as a performed text, it hovers somewhere between
what C. Padwick as far back as 1961 called “sheer magic” and “utter devotion
most humbling to our cowardly souls.”30 For the purposes of this discussion,
it should be added that the personal perception of the text on the continuum
between “magic” or “devotional” depends largely on the intent of the reciter,
his/her level knowledge of the text and of the Arabic language, the context for
the individual, and finally the wider socio-political context in which recitation
takes place.
A second observation is needed when it comes to the discussion of timing
and schedules. Rātib al-Ḥaddād has, for as long as it can be established with
any certainty, been considered an evening prayer – to be recited after sunset,
contrary to for example another text by al-Ḥaddād, Wird al-Laṭīf, which is usu-
ally recited after the fajr (dawn) prayers. When asked why Rātib al-Haddād is
suited for evening prayers (rather than for example mid-day), many present-
day shaykhs and prayer leaders had no particular explanation, beyond the
fact that “it is better after sunset, because then one is relaxed; done with the

29 L.W.C. van den Berg, Le Hadhramout et les Colonies Arabes, 87.

30 C. Padwick, Muslim Devotions, xiii.
ritual of reform – reform of a ritual 157

days’ work.”31 There is, in other words, no doctrinal reason, nor anything in the
text itself that determines at what time the Rātib should be recited, except for
tradition.32 However, it seems clear that the timing as such (after sunset) has
roots in the Sunna, whereby the Prophet is said to have prescribed an after-
noon siesta – understood for the believer to be refreshed for his/her evening
devotions (and implicitly saying that evening is the ideal time for devotional
prayers).33
Finally, Rātib al-Ḥaddad is supposed to always have the same rhythm, while
the speed of recital may vary. In the madrasa, for example, it may be recited
very slowly, in order for the children to learn. At a “normal” speed, reciting
Rātib al-Ḥaddad takes about 15–20 minutes, but it can take considerably longer
for pedagogical reasons, in order for the young students to hear the full enun-
ciation of each syllable and sound. Thus, in the madrasa setting, the age and
needs of the students overrules the “set” timing of the recital.

Teaching the Rātib: Text and Inherent Blessings

In East Africa, students today learn the Rātib in madrasa from an early age, and
it is likely that this was the case also a hundred years ago. Contrary to the teach-
ing of for example longer Sufi texts, there was no particular teaching system
for texts like Rātib al-Ḥaddād. It was taught simply through recitation, without
necessarily considering the actual meaning of the words. Furthermore, there
is not, and never seems to have been, such thing as an ijāza (certificate) for the
Rātib. A young person, student or (in the case of East Africa) a new Muslim
would simply learn it through hearing it repeatedly. There are, however, exam-
ples in literature of individual students seeking an ijāza after having recited the
Rātib with a particularly venerated shaykh. In these cases, the ijāza was solely
for the inherent baraka (referred to as an ijāza tabarruk) in the act of reciting
together and not for the actual learning of the text (referred to as an ijāzat
al-irshād). Typically, in his travelogue from Ḥaḍramawt in 1897, ʿAbd Allāh Bā
Kathīr related that upon spending time in Tarīm with the Ḥaddād family, he
received an ijāza in “all the adhkār and awrād [. . .] and especially in the Rātib
which is to be read at the time of ʿishāʾ.”34 In this case, the esoteric significance

31 Interview, Mwalimu Husayn Soud al-Maawy, Lamu 29.04.10.

32 An exception would be inclusion in the Rātib of the last two verses of Sūrat al-Baqara,
for which there exists strong ḥadīth traditions that they should be the last words recited
before sleep.
33 Interview, Seraj Hendricks, Cape Town, 07.12.2010.
34 ʿAbd Allāh Bā Kathīr, Riḥlat al-Ashwāq al-Qāwiyya, 76.
158 chapter 7

(the inherent baraka of Prophetic descent and that of the family of al-Ḥaddād)
of the text is what is certified in the form of an ijāza.
Here we see that the learning of the actual text and the effect of reciting it
are considered two different processes and two different “products”, or what
M. Lambek has called “alternate forms of legitimacy” representing two differ-
ent form of authority.35 This, again, points back to the perceived dual nature of
the Rātib; its immediate accessibility for devotional purposes, and its depth as
a conveyor of metaphysical insight.
The formal ʿAlawī view has, at least since the nineteenth century, been
that the baraka of Rātib al-Ḥaddād rests in the tartīb of the Quranic/ḥadīth
quotes, i.e. the way in which the lines are ordered in sequence.36 This means
that although the words themselves were not originally written by al-Haddād,
his way of linking known quotes together was inspired and has the potential to
convey baraka when recited precisely in that order. This belief has amounted
to orthodoxy, and is where the Sufi reformist understanding parts way with
more explicitly modernist and Salafi views.

Communal Recitation and the Process of Institutionalization

Most scholars consider Rātib al-Haddād to be more beneficial when recited
communally, although there is nothing that prevents an individual from recit-
ing it alone. The rationale often refers to the hadīth saying that when a group
praises God together, they are surrounded by angels. The emphasis is first and
foremost on baraka, and reciting the Rātib gives blessing to those reciting, all
the more so when it is done communally. Moreover, the story is often repeated
that a group reciting Rātib al-Ḥaddād brings blessing not only to the house in
which the recitation takes place, but also to the forty houses surrounding it.37
This is where the demarcation line is most explicit between the Sufi orienta-
tion and later modernist/Salafi orientation that came to expand in East Africa.
The individual aspect of worship is today often downplayed by Sufi leaders, in
favour of the implicit communal or social benefits of a devotional ritual such
as Rātib al-Ḥaddād. The implicit understanding, as it was conveyed also in
the late nineteenth and early twentieth century, is that an individual Muslim
actually can pray on behalf of others, and call for blessings upon their behalf.

35 M. Lambek, “Choking on the Quran. And other consuming parables from the western
Indian Ocean Front”, in: W. James (ed.), The Pursuit of Certainty. Religious and Cultural
Formulations, London (Routledge), 1995.
36 Interview, Seraj Hendricks, Cape Town, 07.12.2010.
37 Interview, Mwalimu Husayn Soud al-Maawy, Lamu 29.04.10 and with Saleh Muhammad
Badawi Jamal al-Layl, College of Islamic Studies, Lamu, 01.05.10.
ritual of reform – reform of a ritual 159

The modernist approach, on the other hand, would emphasize the explicit
(and by implication; personal) statement of belief.
Several studies, among them Dale Eickelmann’s from Oman, have pointed
out this shift from ritual to a more personal statement of belief and linked it
to levels of secular education. In the case of Oman, Eickelmann links the shift
to mass higher education in the Qaboos era, whereby young Omanis became
equipped to explain and understand their beliefs in a way not previously
possible.38
While this interpretation may have been valid at the time and place of
Eickelmann’s study, we see a slightly different trajectory unfolding in East
Africa. An explicit example can be found in mid- and late twentieth century
Kenya, where the “personalization” of ritual performance came from within
the Sufi tradition as much as from the outside (be that secular education or
Wahhābī daʿwa). Salafi ideas are not “new” to the Kenyan coast, nor are they
wholesale imports from Arabia. On the contrary, the educational silsilas of
Salafi reformist scholars of the 1930s, 40s and 50s, such as Shaykh al-Amin al-
Mazrui (d. 1947) and Abdallah Saleh Farsy (d. 1982) were exactly the same as
those scholars whose orientation remained more distinctly Sufi in outlook.39
In other words, while they certainly were exposed to, and influenced by ideas
from outside – notably Egypt – their reformist orientation was that of a reform
from within, with a strong link between reformist programs and local practices.
Present-day modernist scholars view devotional recitals such as Rātib
al-Ḥaddād as being among the less harmful practices among Sufi practices.
However, the doctrinal ground is straightforward and goes to the belief in
baraka: the Salafi view holds that there rests no particular baraka in reciting
the lines of Rātib al-Haddād in this particular order.40 There is nothing harm-
ful in reciting them, on the other hand they could potentially be recited in any
other order. The bidʿa, to the Salafis, lies in the belief that this particular order –
over any other – is a conveyor of baraka.
Thus in the setting of modern Kenya, reciting Rātib al-Ḥaddād is a choice –
a personal expression of belief in the way outlined by Eickelmann. However,
recitation is also a statement of belonging. It is not as clear an expression as
the participation in the mawlid celebrations, but it nonetheless identifies
the person as a non-Salafi. This ties in with the concept of ʿibāda (worship).

38 D.F. Eickelmann, “Mass Higher Education and the Religius Imagination in Contemporary
Arab Societies”, American Ethnologist, 19:4, 1992, 643–655.
39 K. Kresse, Philosophising in Mombasa, passim; Farsy/Pouwels, The Shafiʿi Ulama;
A.K. Bang, “My generation”.
40 Interview, Seraj Hendricks, Cape Town, 07.12.2010.
160 chapter 7

What are the correct ways of worshipping God? While a text like the Rātib
celebrates first and foremost God and the Prophet, it also, as has been shown
above, praises named members of the Bā ʿAlawī clan, and the descendants of
the Prophet in general. The leaders of the Riyadha, for example, could make
the argument that this recitation is simply a praise of great Muslims before
them, an established practice (mila) – thus not in conflict with the demands of
ʿibadāt. However, as Kresse has pointed out, this argument can easily backfire
and undermine the very significance of the ritual itself, by reducing it to mere
performance and its propagators and leaders to mere performer – or, slightly
better, “cultural personalities”.
A similar, but somewhat later development can be traced in Cape Town,
where the arrival of Salafi/Wahhābī groups in the 1970s and 1980s challenged
the practice of the Rātib. The then-leader, Shaykh Mahadi Hendricks, initially
opposed all types of popularization – based on the notion that the Rātib should
be taught alongside “proper” ʿilm. However, he changed his view some time in
the 1980s and called for a less elitist, more outgoing approach to the Rātib reci-
tations. The difficult life of Muslims in the suburbs and townships must also
have played a role, as many were unable (for financial or political reasons) to
attend Rātib classes. Instead, Shaykh Mahadi called for recordings to be made
and distributed of several adhkār, including Rātib al-Ḥaddād, for basic educa-
tional purposes. In these recordings, the recital drew on long-­established Cape
Malay singing traditions, including its rhythm and tone, thus placing the reci-
tation nearer to what is generally perceived of as “performance”. However, even
on these early cassettes, the recitation would be prefaced by a lesson on its
purpose. The recordings succeeded in their goal of teaching people the actual
texts to be recited, but it is less certain that they actually succeeded in instilling
its purpose of baraka.
By the 2000s, the “protective layer” of knowledge offered at Azzawiya, how-
ever, was not enough – despite the best efforts of the leaders. According to
Shaykh Seraj Hendricks, they decided to simply stop reciting Rātib al-Ḥaddād
altogether. The rationale given is that recitals were stopped to prevent the
Rātib from becoming a mere performance, “put in a G-minor”, to quote Seraj
Hendricks.41 In other words: When the recital becomes more concerned with
its tone than its content, it no longer has any meaning and should be stopped,
according to the teaching of the Azzawiya. It was, in other words, in the pro-
cess of becoming a social event rather than a quest for understanding. The hia-
tus is also partially explained by the need to protect the meaning of the Rātib,

41 Interview, Seraj Hendricks, Cape Town, 07.12.2010.

ritual of reform – reform of a ritual 161

i.e. to stop when people start misunderstanding or misperceiving the baraka

and spiritual states the text is perceived to induce.
Finally, it should be noted that the continued recitation of the Rātib also
can be understood from the point of view of language. In East Africa, Rātib
al-Ḥaddād is recited in Arabic, and few attempts are made at translation as
the majority of the text consists of Quranic quotes that are considered imme-
diately recognizable (if not immediately understandable) to all. This may in
itself contribute both to the continuity and popularity of the Rātib, as the mag-
ical essentialist function of Arabic has been strong among many non-Arabic
speaking Muslims. For example, as J. McIntosh has pointed out, the Giriama
ascribe “intrinsic, underlying potencies” not only to the speech act, but the
entire Arabic language.42

The Rātib al-Ḥaddād: New Reform of a Reformist Ritual?

In the latter half of the nineteenth century, the recitation of the Rātib (like
other devotional rituals, notably the mawlid celebrations) moved from the
family/small group to recitations that in principle were open to all. Gradually,
the recitation became, in institutions such as the Riyadha, an almost integral
part of evening prayers.
At the same time, the text itself became the object of writing, and this text
was used alongside “learning by ear”. As in Ḥaḍramawt, the role of teaching
institutions, (Riyadha, the Madrasa Bā Kathīr and the Azzawiya), were vital
in transforming the recitations to a scheduled event, meaning that students
and mosque attendees would learn the Rātib as a litany to be performed at
certain times, in a certain speed and with a certain pronunciation and into-
nation. At the same time, the same institutions were transmitting (indeed, as
in the case of the Azzawiya, printing and distributing) the ritual as text. To
higher-level students, the same institutions would offer interpretations of the
Rātib, by including its commentaries in their curriculum. Finally, as discussed
in Chapter 6, they also became the custodians of the manuscripts themselves,
the written words of the Rātib as noted down by parents and grandparents.
In other words: the Rātib, at the time of al-Ḥaddād himself, was mainly an
exercise performed by individuals and/or groups as a method to achieve unity
with God, but from the mid/late nineteenth century it moved with the flow of
general Sufi reform towards institutionalization, and thus towards increased
social relevance. This relevance was consciously transmitted and taught by

42 J. McIntosh, The Edge of Islam, 223.

162 chapter 7

teachers within the institutions. Teaching was conducted both orally (learning
by listening) but also on the basis of textual reproductions of the Rātib itself.
Finally, teaching was based on an established set of commentaries as well
other Sufi works within the ʿAlawī tradition that emphasized daʿwa and social
involvement. As we have seen, the endorsement and ritual “recipes” by schol-
ars considered authoritative in East Africa – such as Daḥlān and al-Makhlūf –
also came to have an impact on the way the Rātib was practiced.
The element of social relevance has remained strong also in the twentieth
century. The emergence of locally produced modernist thought, and not least,
the emergence of Salafism/Wahhābism in the twentieth century, has only led
to partial changes, most likely because Rātib al-Ḥaddād is among the least
offensive practices from the Salafi point of view. Political upheavals – ­notably
the Zanzibar revolution – has caused more radical breaks in the tradition than
decolonization, mass education and intellectual shifts within the Muslim
world. This indicates the continued strong emphasis on oral transmission, and
implicitly on leaders who “know” the Rātib, in the sense that they are able to
access its perceived esoteric content.
 chapter 8

Consolidating the Network

Waqf Distribution and New Organizations in Zanzibar, c. 1900–1930

We are told that it is your wish to send your brother and son to us in
Mecca.
We are delighted to hear this.
Ṣāliḥ b. Abī Bakr Shaṭṭā in Mecca to Ṭāhir al-Amawī in Zanzibar, 1929

∵
By the late nineteenth century, the Indian Ocean was becoming an increas-
ingly crowded place. European steamers trafficked the routes to their respec-
tive colonies, while sailing ships of all types and nationalities journeyed from
one coastal city to the next. At the same time, dhows kept plying the waves
between Arabian and East African shores and between the different ports of
East Africa. As we have seen, all these vessels occasionally carried travellers
who journeyed for a very specific purpose, i.e. to spread ideas of organized,
reformist-oriented, Sufi-based Islam. We have seen travellers heading from
Zanzibar to Mozambique, from Mombasa to Cape Town, from the Comoros to
Madagascar and repeated visits to Mecca, Ḥaḍramawt and Egypt.
A final set of questions remains to be raised in this study, concerning the
funding and organizational form of the Sufi networks. The first is the most
difficult to answer as the schools, ṭarīqas and scholarly families upon whose
libraries this book is based, has tended to preserve their religious heritage in
the form of books and manuscripts, while discarding mundane sources such
as receipts, account books, tickets and travel documents. This chapter will thus
only focus on one particular link in what was probably a dense grid of financial
transactions, namely the long-standing transfer of waqf funds from Zanzibar to
the holy cities of Mecca and Medina.
The second question, too, is somewhat difficult to answer, as the organi-
zational form of the network was precisely that; a network of individuals in
varying relationships to one another, and with varying obligations towards
each other. However, we find by the 1920s that these networks were gradually
being “translated” into new organizational forms, and here at least some writ-
ten sources can be found.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�09

164 chapter 8

By the early twentieth century, connections between East Africa (Zanzibar

in particular) and the Ḥaramayn were well established, and s­ teamships from
Zanzibar called on Jeddah at frequent intervals. As we have seen in the previous
chapters, letters passed to and fro with greetings to families at home, spiritual
instructions (such when and how to recite texts such as Rātib al-Ḥaddād), or deal-
ings with publishing houses on the print of a new book (such as Burhān Mkelle’s
correspondence with “The Great Commercial Library” in Cairo). Last but not
least there were more formal letters from qāḍīs and civil servants, on matters of
inheritance and distribution of funds deriving from pious foundations (waqfs).
The latter type of letter will be the starting point of this chapter. Dating from the
1920s and 1930s, there exists in Muscat, Oman,1 a correspondence from the then
Shāfiʿī qāḍī of Zanzibar, Ṭāhir b. Abī Bakr al-Amawī (1877–1938).2 A correspond-
ing file is kept in the Zanzibar National Archives, deriving from the colonial
administrative body known as the Wakf Commission, of which Ṭāhir al-Amawī
was one of the appointed Commissioners. Both correspondences focus on old
waqfs; that is the distribution of revenues from waqfs established well before the
British Protectorate, in some cases as far back as the 1850s. They deal with waqfs
where the beneficiaries were “the poor of Mecca and Medina.”
What this chapter will show is that connections between East Africa and
Mecca were long standing and binding, and that the religious and intellectual
network that underpinned much of the reformist drive also had a function as a
“money transfer” network where funds could flow within a circle of trust. It will
also show that teachers and students in Mecca, in turn, were partly financed by
money derived from waqfs in Zanzibar, distributed by a handful of qāḍīs who
themselves were very much part of the network and who used it precisely in
order to ensure smooth and consistent transfer of money.

Scholarly Networks and the Zanzibari “Meccan Waqfs”, c. 1880–1940

The Zanzibar Wakf Commission was established by the British Colonial

authorities in 1905, as one of the many efforts of the British to reform and con-
trol Islamic law and institutions inherited from Bū Saʿīdi period.3 Its p ­ owers

1 This correspondence is currently kept in the private library of Muḥammad b. Aḥmad Al Bū

Saʿīdī in Muscat, Oman.
2 On Ṭāhir al-Amawī, see A.K. Bang, “Another scholar for all seasons?”.
3 Abdulkadir, Abdulkadir Hashim, Reforming and Retreating: British Policies on transforming
the Administration of Islamic Law and its Institutions in the Busaʿidi Sultanate, 1890–1963, PhD
Thesis, University of the Western Cape, 2010.
consolidating the network 165

were formalized in the so-called Wakf Property Decree of 1907, which was
­formulated to grant the colonial authorities maximum control of waqf proper-
ties and revenues. Failure to notify the Commission of a new or old waqf – as
well as any transaction concerning a waqf property – was deemed an offence.4
The Wakf Commission consisted of four British officers as well as one Sunni
and one Ibāḍī qāḍī, to be appointed by the British Resident, subject to the
approval of the Sultan. For the period from 1905 to his death in 1925, Aḥmad b.
Sumayṭ served as the Sunni representative. He was replaced by Ṭāhir al-Amawī,
who held the same position until 1936. He, in turn, was replaced by Aḥmad
b. Sumayṭ’s son ʿUmar b. Sumayṭ, whom we encountered in Chapter 4 as a
trader-scholar in Diego Suarez in the 1920s. He served as Wakf Commissioner
in Zanzibar well into the 1950s. In other words, the Sunni representatives
were identical with the Chief Sunni Qāḍīs. They were also closely connected
with each other, either by family (as the father-son Sumayṭs), or by religious
adherence, as Shaykh Ṭāhir who was closely connected with the ʿAlawī ṭarīqa,
although his primary Sufi affiliation was the Qādiriyya, through his relative
ʿAbd al-ʿAzīz al-Amawī and the network of Brawanese scholars.
The records of the Wakf Commission in Zanzibar show that a number of
town houses and shambas (farms) in the countryside had been dedicated at
least partially for the benefit of “the poor in Mecca and Medina”.5 By the time
the Wakf Commission started to take over administration of Zanzibari waqfs,
most of the endowments for Mecca were already old. Only in very few cases
was the original waqfiyya document at hand, while the rest of the cases were
“traditional”, as the British called it. The lack of a central waqf administration
during the Bū Saʿīdī period, meant that the later colonial Wakf Commission
often had to work without the original documents and purely from “traditional
knowledge”, concerning which properties were made waqf for what purpose.
In cases were the waqf was disputed, the original purpose of the waqf was fre-
quently reconstructed in court by witnesses who gave statements from mem-
ory or hearsay.
Around 1915, the British members of the Wakf Commission were becoming
increasingly concerned about the “Meccan waqfs” in Zanzibar. As was the case

4 For a full account of the waqf administration during the Zanzibar protectorate,
see N. Oberauer, Waqf im kolonialen Sansibar. Der Wandel einer islamischen Stiftungspraxis
unter britischer Protektoratherrschaft, Würzburg (Ergon-Verlag), 2012.
5 Four of the Zanzibari waqfs dedicated for Mecca and Medina are gathered in the file
ZA-HD10/5. This file records the history of properties dedicated as Ḥaramayn waqf in the
period 1877–1904. Others are scattered in individual files, depending on when – and if – they
came to be vested in the Wakf Commission. See below, Appendix 1.
166 chapter 8

for most waqfs, they were worried about the lack of revenue deriving from the
property, but in this case they also had another, more practical problem: How
were they actually going to forward the money to the poor in Mecca, and how
could one make sure that the money actually reached its destination? Although
the Wakf Commission had been aware of the money for more than a decade,
the First World War and the political upheavals in the Ḥijāz had prevented
them from setting up any formal procedure for the transfer. Many options had
been tried, and in 1913, the later British resident John Sinclair voiced his exas-
peration to the British Consul in Jeddah:

At one time I understand the money was sent by one of the Kathis here
to the care of the Mufti of Mecca, but in view of some doubt arising as to
how this gentleman disposed of it, this practice was discontinued.6

The British Consul in Jeddah, in turn suggested various solutions, but none
were put into effect, probably due to the political unrest in the Ḥijāz. By 1915,
the secretary of the Wakf Commission had reached a frustrated conclusion:
“I would suggest that we send nothing to the poor of Mecca and Medina. It
would certainly never reach the poor.”7

The Religious Networks Used for Money Transfer

By 1926, the revenue from the waqf properties had accumulated to the con-
siderable sum of 18,069 Rupees waiting to reach Mecca and Medina. With the
Wahhābīs and the embryonic Saudi state now in possession of the Holy Cities,
the problems of actually getting the money there was as unresolved as ever.
The Wakf Commission in Zanzibar decided to turn to Sayyid Khalīfa for sug-
gestions. The Sultan could only reply that he considered “the present state of
Arabia so disturbed and the whole question such a difficult one” that he pre-
ferred to let the matter rest indefinitely.8 Nevertheless, three months later, he
was able to suggest a solution – quite possibly on the advice of qāḍīs on the
Wakf Commission.9
The Ibāḍī money was to be sent with an honourable Ibāḍī going on pilgrim-
age, and receipts were to be returned to the Wakf Commission. The Sunni

6 Letter from Sinclair to the British Consul at Jeddah, 18 February 1913. ZA-HD10/5.
7 Note from the Secretary to the Wakf Commission, 25 August 1915. ZA-HD10/5.
8 Letter from Sayyid Khalīfa (by his private secretary Battiscombe) to the Secretary of the Wakf
Commission, 10 August 1926. ZA-HD10/5.
9 Letter from Sayyid Khalīfa (by his private secretary Battiscombe) to the Secretary of the Wakf
Commission, 2 November 1926. ZA-HD10/5.
consolidating the network 167

money should be entrusted to a “prospective Sunni pilgrim of good ­character”.

He was to hand it over to Shaykh ʿUmar b. Abū Bakr Bā Junayd in Mecca, who
would conduct the distribution and return receipts. Bā Junayd, as we have
seen in Chapters 2 and 5, was an associate of Aḥmad Zaynī Daḥlān, and an
important teacher to several East African scholars who spent time in Mecca.
He was also one of the main teachers of Muhammad Salih Hendricks during
his Meccan sojourn, and the person who appointed ʿAbd Allāh Bā Kathīr to
journey to Cape Town to resolve the Jumʿa dispute there.
Nonetheless, Bā Junayd’s credentials as a Shāfiʿī-Sunni scholar was not
enough to convince the Ibāḍī sultan of Zanzibar of his trustworthiness,
let alone the British Wakf Commissioners. Sayyid Khalīfa advised that one
should check on Bā Junayd: Was he really a man of good reputation? The Wakf
Commission promptly instructed the British Consulate in Jeddah compile a
report. The reply was that Bā Junayd was a man of the best reputation; “a mem-
ber of the Majlis al-Shuyūkh in Mecca”.
In 1927, the Wakf Commission thus decided to entrust ʿĪsā b. ʿAlī al-Barwanī
(on him, see below) with 1000 RS. He would guarantee for the money and ensure
its safe forwarding to Bā Junayd in Mecca.10 The Wakf Commission made it
perfectly clear that this was to be regarded as “an experiment.” Receipts were
to be produced before any further funds would be made available. The result of
the “experiment” makes for fascinating reading from the point of view of the
scholarly networks between Zanzibar and the Ḥaramayn. The actual money
transfer is interesting in itself and lists the names of the central figures of the
reformist Sufi network of the western Indian Ocean in the 1920s:
The 1000 RS money first passed from ʿĪsā al-Barwānī to ʿUmar b. Sumayṭ and
from him to Abū Bakr b. ʿAbd Allāh Bā Kathīr, the son of the founder of the
Madrasa Bā Kathīr, who had died in 1925. They, in turn remitted it by bank to
Sayyid Muḥammad b. ʿUmar Bā Zurʿa in Aden who finally sent it onward to
Bā Junayd.11 What the money transfer in fact shows, is that the same network
that since the early twentieth century had led various Islamic educational and
daʿwa functions, also was reliable and efficient in overseas money transfers.
This system was so successful – and apparently also acceptable to the British
Waqf Commissioners – that the method was repeated every year without fail
until the Second World War.

10 Letter from the Wakf Commission to ʿĪsā b. ʿAlī al-Barwanī, 10 January 1927. ZA-HD10/5.
11 Letter from ʿUmar b. Abī Bakr Bā Junayd to ʿĪsā b. ʿAlī al-Barwanī, 17 Shaʿbān 1345/
20 February 1927. ZA-HD10/5.
168 chapter 8

The Distribution of Waqf Funds in Mecca and Medina: Scholar to

Scholar Aid
Bā Junayd lived up to his reputation. By February 1927 he had produced a
detailed set of receipts for the money.12 What the list of receipts show, is that far
from being randomly distributed among “the poor”, there was a clear pattern.
They were, in fact, distributed almost exclusively among the sāda ʿAlawīyya
and the Ḥusayni ashraf, for the benefit of themselves and those dependent on
them.
A closer study of distribution in the period 1927–1936 reveals that the most
substantial amounts were allotted to a handful of scholarly families, either
ʿAlawī or closely intellectually associated with them.

November May May May September November

1927 1929 1930 193113 1933 1936

Āl al-Ḥibshī 25 RS 25 RS 40 RS 85 RS 105 RS 40 RS
Āl al-Bār 25 RS 20 RS 40 RS 100 RS 110 RS 20 RS
Āl Bāb Ṣayl 20 RS 20 RS 40 RS 40 RS 45 RS 24 RS
Āl Shaṭṭā 12 RS 25 RS 40 RS 360 RS 150 RS 496 RS
Āl Daḥlān 100 RS 25 RS 40 RS 185 RS 185 RS 65 RS
Students 40 RS 30 RS 60 RS 70 RS 70 RS Not
at the Sāda registered
Ribāṭ

Figure 12 Distribution of the “Meccan Waqfs” of Zanzibar, 1927–1936. 13

Clearly, the distribution favoured families who were extremely influential

in the period when these waqfs were established (late nineteenth century).
Daḥlān and Bāb Ṣayl have already been mentioned repeatedly as central teach-
ers to a generation of East African and Ḥaḍramī ʿulamāʾ. Equally important
were the teachers of the Ḥibshī family, whether located in Ḥaḍramawt or in
Mecca. Other ʿAlawī families, such as the Āl al-Bār held the same importance
as keepers of extraordinary baraka. The individuals in this network were tied
by strong bonds, both in terms of family (being of common Sharīfian ­ancestry),

12 ZA-HD10/5.
13 This year the amount to be transferred increased from 1000 RS to 2050 RS – hence
the increase in the figures.
consolidating the network 169

and in shaykh-disciple relationships that were considered binding both in this

world and in the more metaphysical sense.
Also very important, as has been pointed out repeatedly in this book, were
the ribāṭs, the teaching institutions. It is worth noting that throughout the
period, the ʿAlawī (Sāda) ribāṭ in Mecca received substantial funding from
the Zanzibari Wakf Commission.
As the above table shows, only a small amount was originally granted to the
Shaṭṭā family, namely the sons and relatives of Abū Bakr Shaṭṭā. However, by
1929, they are on the list, as recipients of 25 RS. In 1930, too, they received a sim-
ilar amount as their companions of the Daḥlān, Ḥibshī and Bāb Ṣayl families.
By 1933, the qāḍīs in Zanzibar had decided that no less than 1/4 of the remitted
amount was to be earmarked for the head of the Shaṭṭā family, Ṣāliḥ b. Abī
Bakr Shaṭṭā, “for they are numerous and their students cannot be under one
hundred according to what we have heard from persons who come from that
place.”14 Shaykh Ṭāhir al-Amawī could thus write to Ṣāliḥ b. Abī Bakr Shaṭṭā:

After the best of greetings, and peace be upon Your Honour [. . .] I turn to
the letter which Sayyid Muḥammad Saʿīd ʿUthmān [Shaṭṭā] sent to our
paternal cousin, the Qāḍī Shaykh Burhān [al-Amawī], including the note
on the funds (khayrāt), dedicated for the awlād Shaṭṭā. This surplus has
been maintained since days long gone, but these days the revenue has
been disrupted, and tears of sorrow flooded because of its ceasing, since
the waqf was meant to be continued for years to come. I should inform
you that the instructions (ʿamr) originally were passed by the hands of
the late Sayyid Aḥmad bin Abī Bakr bin Sumayṭ, and Shaykh ʿAbd Allāh
Bā Kathīr, then it was conducted by Shaykh Abū Bakr b. ʿAbd Allāh Bā
Kathīr, and it was always sent to Shaykh ʿUmar Bā Junayd, according to
what Shaykh Abū Bakr says. Nothing came to our mind of that which
pervades the heart of suspicion and lies. [Then] Sayyid Muḥammad
Saʿīd ʿUthmān was cut from the revenue, and our hearts were filled with
doubts. However, now it will resume every year as before or even with an
increase – God willing.15

According to the letter and notes by Ṭāhir al-Amawī, the Shaṭṭā family ­simply
“inherited” the revenue which until then had gone to three other, clearly iden-
tified persons. These names are interesting as they show the impact of the

14 Minutes of Meeting of the Wakf Commission, 25 May 1933. ZA-HD/10–5.

15 Letter from Shaykh Ṭāhir al-Amawī to Ṣāliḥ b. Abū Bakr Shaṭṭā, 21 Shawwal 1348/22 April
1930. Amawī File I, Doc. No. 34, Maktaba Aḥmad Āl Bū Saʿīdī, Muscat.
170 chapter 8

Qādirī network of Brawanese scholars (notably ʿAbd al-ʿAzīz al-Amawī and

his son Burhān) on the distribution of waqf funds from Zanzibar. The three
beneficiaries were Abū Bakr b. ʿAbd al-Raḥmān al-Amawī (90 RS), “the virtu-
ous Sayyida Sharīfa bt. ʿAbd al-Azīz al-Amawiyya” (90 RS) and Sayyid Aḥmad
b. ʿUmar Qullatayn (90 RS), the son of the central Qādirī scholar in Zanzibar
(see above, chapter 3). What we see here is a distribution of waqf funds entirely
within the circle of the Amawī-Qullatayn families, all closely tied together by
their Brawanese background, the Qādirī ṭarīqa, and probably also by marriage
and kinship ties reaching back to Brawa.
Now, in 1930, these three recipients were all dead and Shaykh Ṭāhir, for
reasons unknown, redirected the entire amount to the Shaṭṭā family. Ṣāliḥ b.
Abī Bakr Shaṭṭā was to distribute the funds according to a system laid out by
al-Amawī, and he was to return receipts in the same way as Bā Junayd.
In return, Shaykh Ṭāhir got a letter of thanks, some issues of the journal
“Umm al-Qurrāʾ” (which Ṭāhir decided to “keep in a safe box so it will not be
harmed”) and select literature from Mecca.16 The change to the Shaṭṭā family
was far from random. As we have seen in the previous chapters, the core of this
network was its intellectual connection over two generations, and that this was
an outlook that emphasized books, and a strong orientation towards reformist
ideas and a willingness to stay updated on developments elsewhere.
However, we may speculate that Shaykh Ṭāhir also had more personal
motives for the sudden increase in the Shaṭṭā funds. In 1929, Ṣāliḥ b. Abī Bakr
wrote to Shaykh Ṭāhir al-Amawī: “We are told that it is your wish to send your
brother and son to us in Mecca. We are delighted to hear this . . .”17 Here, we
also get a glimpse of how the sending of students came about; through per-
sonal, long-standing networks that were cemented (in this case), by money.
Smaller funds (amounts of 2–5 rupees) were also meticulously distributed,
and receipts returned to Zanzibar. The list of receipts from 1928 may serve as
an example of the ways in which these funds were used to support families,
institutions and individuals who had historical affiliations (whether through
family or Sufi association) to the Zanzibari scholarly community as repre-
sented by its chief qādīs:

16 Letter from Ṭāhir al-Amawī to Ṣāliḥ b. Abī Bakr Shaṭṭā, 21 Shawwāl 1349/11 March 1931.
Amawī File I, Doc. No. 35, Maktabat Aḥmad Āl Bū Saʿīdī, Muscat.
17 Letter from Ṣāliḥ b. Abī Bakr Shaṭṭā to Ṭāhir al-Amawī, 6 Jumādā I 1348/9 October 1929,
Amawī File I, Doc. No. 28, Maktabat Aḥmad Āl Bū Saʿīdī, Muscat.
 consolidating the network 171

54 RS Muḥammad b. Muḥsin al-ʿAṭṭās

50 RS Muḥammad b. Ṣādiq Daḥlān; ʿAbd al-Raḥmān b. Ṣādiq Daḥlān
40 RS Inhabitants of the Sāda Ribāṭ
25 RS Āl al-Ḥibshī; Āl al-Bar
20 RS Āl Bāb Ṣayl
15 RS Āl BalKhaywīr
12 RS
Sayyid Muḥammad b. ʿAbd Allāh Āl al-Miḥḍār ʿUmar Ba Faqīh and family
Yarūm and family
Sayyid Aḥmad Ba Rūm The inhabitants of the Ribāṭ The descendants of Sayyid ʿAbd
of Shaykh Tāj Allāh Miḥḍār
Sayyid Muḥammad Shaṭṭā
10 RS Family of Sayyid ʿAbd Allāh BāRūm
9 RS Sayyid Ḥasan Fadʿaq
8 RS
Sayyid ʿAbd Allāh b. Muḥammad Sayyid ʿUmar b. Ḥasan Sayyid Aydarus b. ʿAlawī al-Jifrī
al-Jifrī al-Yatimī (?) and his family and his family
Ḥasan ʿAbbās and his family Sharīfa ʿAyisha ʿAṭṭāsiyya Ibn Shaʿīb ʿAbd al-Raḥmān and
his family
ʿAbd Allāh al-Malik
6 RS
Family of Shaykh Asʿad Dahan Family of ʿAbd al-Raḥmān The daughters of ʿAlawī al-Jifrī
Dahan
The daughters of Ṣāliḥ b. al- Sharīfa Mazīna Jifriyya and Sharīfa ʿAlawiyya bt. Sayyid
Shaykh Abī Bakr her daughters Ḥusayn Jamal al-Layl and her
children
Family of Shaykh ʿAbd Allāh Āl Family of ʿUthmān Dahrān
Bū al-Khayr
5 RS
The family of ʿAlī Bā Dhanī The family of ʿAbd al- Sharīfa Ruqayya bt. Sayyid Salīm
Wahhāb al-Aṭṭās b. Aḥmad al-ʿAṭṭās
The daughters of ʿAlawī al-Saqqāf The daughters of ʿAlawī b. Ḥasan b. ʿAlawī al-Jifrī
and their brothers Naṣr
Sharīf ʿAbd Allāh al-Kalfūt The family of Sayyid Hāshim Shaykh ʿAbd Allāh Ba Rāhim
b. ʿAlawī al-Jifrī
Shaykh Muḥammad Saʿīd Āl Bū Shaykh Muḥammad Bā Shaykh ʿUmar Hamdān
al-Khayr Qayl (?)
Shaykh Zaydān Family of Aḥmad Bā Shaykh Saʿad al-Yamanī,
Sulaymān Madrasa in the Ḥaram Mosque
Family of Shaykh Ṣāliḥ Kamal Aḥmad b. ʿAbd Allāh al-ʿAṭṭās
 172 chapter 8

(cont.)

4 RS
Sayyid Ḥasan b. ʿAbd Allāh al-Jifrī Sayyid Aḥmad b. ʿAbd Sayyid ʿAlī b. Ḥasan al-Yatimī
al-Raḥmān al-Jifrī
Sayyid ʿAbd al-Qādir Muqbīl (?) Sayyid Ḥusayn b. Muḥsin Sayyid ʿAlawī b. Aḥmad
Muqībil al-Saqqāf
Family of ʿAbd Allāh b. Ḥasan SinʿAlīha Madaha (?) Sharīfa Fāṭima bt. Ḥusayn Bā
al-ʿAṭṭās Aqīl
Sayyid ʿAlawī Atrajī (?) Sayyyid ʿAlī al-Saqqāf and Muḥammad Kāmil al-Sandrī
his sister
Shaykh Ḥasab Allāh (?) Sayyid Aḥmad b. ʿAbd Sayyid Ḥusayn b. Dāwūd
al-Raḥmān al-Saqqāf
ʿAbd al-Mālik Murdād Sharq b. Yaḥyā Shaykh Aḥmad Nadirīn
Sharīfa ʿAlawiyya Jifriyya Sayyid Muḥammad b. Nāṣir Shaykh ʿAbbās Mālikī
Shaʿīb
Shaykh ʿAlī Mālikī Shaykh Darwīsh Amīmī Sālim Amimī
ʿUmar b. Khalīl Amimī Sayyid ʿAbd al-Qādir b.
Muḥammad Dahrān
Shaykh Jamal Mālikī Shaykh Muḥammad
al-Bandarī
3 RS
Sayyid Ḥasan Barqah Shaykh Bakr Rāfīʿ Family of Shaykh b. Shaykh Abī
Bakr
Fāṭima bt. ? (unreadable) Ṣāliḥ al-Mubārak and family
ʿAyisha Muslim The house of Shaykh ʿĪsā Amīna bt. ʿUmar Bā Satrūd (?)
Muḥammad b. ʿAbd
al-Raḥmān b.ʿUmar Bā Sudān
Abū Bakr Bashash
2 RS
Sayyid Ḥasan b. Muḥsin al-Yatimī Sayyid ʿAbd al-Raḥmān b. Sayyid ʿAydarūs al-Saqqāf
Ḥasan al-Yatimī
Layla bt. Muḥammad al-Saqqāf Fāṭima bt. Muḥammad Fāṭima al-Ḥaddād
al-ʿAṭṭās
Ḥasan b. ʿUmar al-Jifrī ʿUbayd Mubārak ʿAlī Sardajī
Maryam bt. ʿAbd Allāh Khadīja bt. Aḥmad ʿAbd al-Fatḥ Bashash
al-Yamanī
Salmā bt. Ḥasan al-Yaltamī Ḥasan b. ʿUbayd Allāh Ḥasan b. ʿAbūd BaJābir
al-ʿAydalī
Ḥusayn Bā Jābir Sharīfa Zaynab bt. ʿUmar Sharīfa Jamīla bt. Muḥammad
al-Saqqāf al-Saqqāf
 consolidating the network 173

(cont.)

Sayyid Hāshim ʿAqīl Family of Muḥammad Muḥammad b. Mazmī

BalBayd
Nūr bt. Muḥammad b. ʿUmar Sharīfa Maryam bt ʿAbd Family of ʿAbd Allāh b.
al-Saqqāf Allāh al-Kutī Muḥammad Bā Jābir
Sayyid Aḥmad b. Shaykh ʿAlī Bā Sālīm Sharīfa Yamīna
al-Yaltamī (?)
Fāṭima al-Yamaniyya ʿAyisha Bā Qūsih (?) Ruqayya Juwiyānī
Fāṭima bt. Bā Wazīr Sayyid Ḥasan b. Ṣāliḥ b. Saʿīd Shaykh ʿAbd al-Ḥamīd al-Miṣrī
ʿAtiyya bt. ʿAbd Allāh al-Sawwālī Ruqayya bt. Muḥammad Bā ʿAbd al-Qādir Nūrī
Jābir
Students at al-Shaḥūmī Ribāṭ Fāṭima bt. Zagha Sayyid ʿAbd Allāh b. ʿAlawī
al-Saqqāf
Sharīf Sharq b. Ṭurāb and his Sharīfa Khadīja bt. ʿAbd Sharīf Aḥmad b. Hizaʿ
family Allāh al-Saqqāf
Sharīf Ḥammūd Sharīf Ḥamza al-Yarkatī Sharīfa Kamīla Jifriyya
Sharīfa Maryam bt. Aḥmad b. Zayn Sharīfa Zaynab bt. Sayyid ʿAbd al-Raḥmān and his
al-Jifrī Muḥammad al-Saqqāf brothers
Sharīfa Maryam bt. ʿAbd Allāh Ismāʿīl Zaframi (?) Zaynab bt. Ṣāliḥ Bā Nawīr
al-ʿAṭṭās
Fāṭima bt. Muḥammad Bā Hadīq Sharīfa Zaynab bt. ʿAwn Sharīfa ʿIzza bt. Abī Bakr
al-ʿAydalī al-ʿAydalī
Khadīja bt. Abūd Bā Jābir Khadīja bt. Naṣr Bā Jābir Fāṭima bt. ʿUbayd Bā Ḥadīla
Sharīfa bt. ʿAbd al-ʿĀl Salma bt. Muḥammad Ṣāliḥa bt. Muḥammad Bā Jābir
Bā Jābir
Saʿda bt. Muḥammad Bā ʿAbad ʿAyisha bt. Abūd Bā Jābir Fāṭima bt. Muḥammad Bā Jābir
Shaykh ʿAbbūd Bā Jābir Sharīfa Ṣāliḥa bt. Sayyid Sharīfa Safiyya bt. Sayyid Sharf
Zaynī Jamal al-Layl Jamal al-Layl
Sharīfa Fāṭima bt. Sayyid ʿAlawī Sharīfa Fāṭima Muqībilah
BāʿAqīl
1 RUPEE
Shaykh Muḥammad al-Yamanī Khadīja bt. ʿAbd Allāh
al-Saqqāf

Figure 13 Receipts signed and returned to Zanzibar 12 Jumada I 1347/26 October 1928.

What we see is that among the smaller donations, a significant portion was
granted to women. In fact, among the 62 recipients of 2 rupees, 40 were
women. Named, individual women were also the recipients of more significant
174 chapter 8

amounts, such as Sharīfa ʿAlawiyya bt. Ḥusayn Jamal al-Layl who, along with
her children, received 6 rupees. Here, we catch for the first time a glimpse of
the Meccan teaching circles run by women. Keeping in mind the observation
by Snouck Hurgronje that the “almost dwarfish” Bāb Ṣayl provided ladies of
the better classes with “useful sentences” quoted from fiqh, ḥadīth or theology,18
it is not unlikely that some of the ladies on this list – or their mothers – had
been participants in these classes. We may also guess that some of the women
on this list also acted as teachers for other women. However, it is also very
likely that many of these women were simply widows who were supported
by the network. What is clear is that these donations, too tended to follow fam-
ily lines.
We also see smaller donations being made to students (“inhabitants”) of
specific ribāṭs. The Haḍramī ʿAlawī impact on the list is very strong, as well-
known ʿAlawī and non-ʿAlawī Ḥaḍramī names keep appearing as recipients.
Overall, the distribution pattern seems to indicate first and foremost an
over-representation of specific families (as shown in Figure 11) as recipients
of larger amounts. Next, the funds were spread thin among individuals and/or
families (as shown in Figure 12), and here, too, the over-representation of cer-
tain families is clear. However, individuals with no known connection to nei-
ther the ʿAlawiyya nor the Qādiriyya can also be found on the list of recipients.
The question should be raised why the Zanzibar qāḍīs and scholars chose
this carefully detailed distribution and not – say – simply entrust the money
to the Mufti of Mecca or any official qāḍī in the Ḥaramayn? As mentioned at
the outset, most of the original documents of these waqfs are not available, nor
were they available to Aḥmad b. Sumayṭ or Ṭāhir al-Amawī who distributed
the funds on behalf of the Wakf Commission. They were, in other words, free
to distribute the funds as they saw fit, although partly tied by local knowledge
about the original intentions of the waqf. Another limitation was the need of
the colonially administered Wakf Commission to maintain good book-keeping,
i.e. produce receipts that the money was actually distributed. In other words:
The importance of the intellectual and family networks came to decide also
the more temporal distribution of pious endowments.
When it fell to Aḥmad b. Sumayṭ to oversee the distribution of the funds “for
the poor of Mecca and Medina”, it is not surprising that he chose to channel the
funds through the networks which he knew and trusted. This is especially so if –
as the aforementioned British comments indicate – remittances through the
Mufti of Mecca gave rise to doubts. Ibn Sumayṭ’s successor, Shaykh Ṭāhir, sim-
ply followed suit in the established pattern. His successor, in turn, Ibn Sumayṭs

18 C. Snouck Hurgronje, Mecca, 200.

consolidating the network 175

son ʿUmar b. Sumayṭ, was himself trained in Ḥaḍramawt and the Ḥaramayn,
and he proceeded to ensure the same distribution well into the 1950s.

The Waqf of Muḥammad b. Aḥmad al-Ḥātimī

The discussion so far has focused on waqf funds which were transferred to the
Ḥaramayn as collective funds, i.e. deriving from several different waqfs but
sent to the Ḥaramayn as a lump sum.
However, not all waqfs were taken over by the Waqf Commission. Some
remained under the administration of a traditionally appointed mutawallī,
despite the efforts of the British to obtain control of all waqf funds in Zanzibar.
Furthermore, as mentioned above, not all waqfs were administered on the
basis of an original waqfiyya, but rather on qāḍī’s knowledge of the original
intention of the founder. It is worth discussing in closer detail one such waqf
for which the original document can be found (albeit only in a much later
translation), and which for a very long period was administered (and its rev-
enues distributed) outside the reach of the Wakf Commission.
In February 1853 the Brawanese-Zanzibari19 Muḥammad b. Aḥmad al-Ḥātimī
had his fellow Brawanese, the qāḍī Muḥyī al-Dīn b. Shaykh al-Qaḥṭānī (see
above, Chapters 2 and 3), draw up a waqfiyya for him. Although little is known
about his background, Muḥammad b. Aḥmad must have been a successful
man as he owned his own stone house in Zanzibar Town, as well as at least
3 shambas on the island, which he now made waqf:

Muḥammad b. Aḥmad b. ʿUmar al-Ḥātimī declares that he has dedicated

as waqf his share in the house where he lives at Kiponda in the town of
Zanzibar and which is built of stones and lime. It is eternal waqf in the
way of Allah, and it is neither to be sold, gifted nor inherited until Allah
inherits the earth and whatever is on it and He is the best of inheritors.
The income derived from the said house is to be divided into 4 parts: the
first part for the upkeep of the house according to the discretion of the
trustees, the second part for his daughter and he has allowed her to live in
the said house i.e. to occupy the portion which he was living in without
any rent, the third part for the poor of Mecca, the fourth part for the
Sunni mosques which are in the town of Zanzibar.
The trustees of this waqf are Sayyid Sulayman b. Hamid, Muhyi al-Din
al-Qahtani, Omar b. Abd al-Rahim and Nur b. Omar b. Abi Bakr.

19 In other known documents drawn up on his behalf, Muḥammad al-Ḥātimī uses the nisba
“al-Barāwi”. Amawī File, II, doc. 16 and Amawī File III, Doc. 30, Maktaba Aḥmad Āl Bū
Saʿīdī, Muscat.
176 chapter 8

He also declares that he has dedicated as waqf his 3 shambas at Kianga

at the island of Zanzibar, together with their contents such as clove trees
and other kinds of trees, including his slaves who are in the shambas. It is
eternal waqf in the way of Allah, and it is not to be sold, gifted nor inher-
ited until Allah inherits the earth and whatever is on it and He is the best
of inheritors.
The income derived from the said shambas is to be divided into four
parts as that of the house mentioned herein. One part goes to the upkeep
of the shambas and purchase of a slave to replace one who is dead,
according to the discretion of the trustees, the second part for his daugh-
ter, the third for the poor of Mecca, and the fourth for Sunni mosques in
Zanzibar Town ( . . . )
4 Jumādā I 1269/13 Feb. 1853.
Muhyi al-Din b. Shaykh al-Qahtani in his own hand.

What concerns us here are the parts allotted for “the poor of Mecca”. If this
document is to be trusted as authentic,20 it gives absolutely no specifications as
to how the money is to be distributed among the poor of Mecca, but conforms
to the generally accepted norms and language of Ḥaramayn waqfs throughout
the Islamic world.
Let us now move 77 years forwards in time, to 1930. By then, Muḥammad b.
Aḥmad al-Ḥātimī was long dead, as was Muḥyī al-Dīn al-Qaḥṭānī who drew up
the waqfiyya. Still alive was Ṭāhir b. Abī Bakr al-Amawī, by now chief Shāfiʿī
Qāḍī of Zanzibar.
Ṭāhir now administered the waqf, but did so by his own authority as
appointed mutawallī, and not in capacity as a member of the Wakf Commission.
It appears that the Ḥātimī waqf came to be vested in the Wakf Commission as
late as 1939, after the death of Shaykh Ṭāhir al-Amawī. According to the Minutes
of Meeting, the Wakf Commission was dissatisfied with the lease agreements
executed by Shaykh Ṭāhir on behalf of al-Ḥātimī’s granddaughter Nana. It tran-
spires from the context that Shaykh Ṭāhir had wilfully kept this waqf away from
the Wakf Commission, despite being a long-term member of that body. Why

20 Waqfiyya translation in English, probably produced in 1938, when the administration of

this waqf was disputed by the Wakf Commission; ZA-HD6/108. From the context of the
dispute, it seems likely that the translation was produced from a document provided
to the Wakf Commission by al-Ḥātimīs descendant Muḥammad b. ʿAbd al-Raḥmān
al-Ḥātimī who wanted his sister removed from the trusteeship of the waqf. Whether he
was in possession of the original document or a later copy is not stated. Neither can we be
certain that the translation actually reproduces the full text.
consolidating the network 177

he should have done so, is unclear, but it may be that the “Brawanese network”
wished to keep this particular waqf under their own control, and that Shaykh
Ṭāhir fit the job description perfectly, being Brawanese, Qādirī and a capable
mutawallī.21
On the 21 Shawwāl 1348/ 22 April 1930, Shaykh Ṭāhir wrote the following to
his colleague Ṣāliḥ b. Abī Bakr Shaṭṭā in Mecca concerning the Ḥātimī waqf:

Of these (i.e. of an original amount of 500 RS) are 200 RS revenue from
the waqf of the late Muḥammad b. Aḥmad al-Ḥātimī, administered by
Mwana Nāna bt. Nūr al-Ḥātimīyya for the poor of Mecca and Medina.
Of that, 50 RS is to be distributed among the poor of Mecca by Sayyid
Ḥamza b. ʿAbd Allāh who shall keep 15 RS for himself and return receipts
to your honour from those who received it. Twenty-five RS goes to the
house of Sayyid Aḥmad Zaynī Daḥlān, and receipts shall be returned to
us. Furthermore, 25 RS goes to the house of Muḥammad Saʿīd Bāb Ṣayl,
to be distributed among them according to instructions and receipts
to be returned to us. Furthermore, 50 RS goes to the offspring of Sayyid
Aḥmad Shaṭṭā to be distributed among them, and receipts returned to
us. Furthermore, 50 RS shall be sent to Medina, to Aḥmad Khalīlī, Shāfiʿī
Imām of that blessed realm. Of that, 15 RS are to be kept by him and
the rest is to be distributed among the poor of Medina. He is to return
receipts to your honour from those who receive it.22

What we find is that Shaykh Ṭāhir now gives very clear and concise details on
how the surplus from the Ḥātimī waqf is to be distributed. The question that
must be raised is how this actual distribution system came about, given that
it is not specified in the waqfiyya? Again, the most likely answer is that it was
initiated in the time of Muḥyī al-Dīn al-Qaḥṭānī and then by Aḥmad b. Sumayṭ
who served as mutawallī for a number of waqfs before the Wakf Commission
came into existence, and thereupon, as mentioned, as Wakf Commissioner.

Avoiding the Wahhābīs: The Network into the 1940s

The transfer of waqf funds from Zanzibar to Mecca seems to have functioned
well until it was brought to a halt by World War II. During the war years, ʿUmar
b. Sumayṭ made some attempts to transfer the funds by hand through “reliable
travellers”, and it seems that some transfers of this type did take place.

21 ZA-HD10/7.
22 Letter from Shaykh Ṭāhir al-Amawī to Ṣāliḥ b. Abī Bakr Shaṭṭā, 21 Shawwāl 1348/22 April
1930. Amawī File I, Doc. No. 34, Maktaba Aḥmad Āl Bū Saʿīdī, Muscat.
178 chapter 8

However, in 1946, the Saudi Government intervened. Saudi authorities had

also tried to intervene in 1927, when Ibn Saud claimed the right to all waqf
revenues destined for the Ḥaramayn. At that time the qāḍīs of the Wakf
Commission in Zanzibar flatly refused to change their well-working system for
transferring money to Mecca. In this, they had the support of the British, who
above all were concerned about the effects any concession to the Wahhābīs
might have in India:

. . . it would almost certainly give rise to pain in some quarters in India

and to the dangerous misunderstanding that His Majesty’s Government
were prepared to interfere with the courts in a matter of such delicacy on
behalf of iconoclasts.23

According to the 1946 regulations, Saudi authorities would no longer permit

private persons to distribute funds deriving from foreign waqfs. Rather, this
should be left to the appropriate Government department, together with a
detailed list of recipients.
It seems that the ʿulamāʾ in Zanzibar, and especially ʿUmar b. Sumayṭ, had
little faith in the Saudi Government. In a meeting in June 1946, the Zanzibar
Wakf Commission decided to check the receipts they received from the Saudis.
Did the people on the list actually receive their money or not?24 ʿUmar was to
write to his correspondents and ask. The answer, according to ʿUmar, was that
the money never was distributed.
ʿUmar apparently decided to follow the matter up more closely. In 1956,
when on his riḥla (journey) to Ḥaḍramawt and Ḥijāz, he “made inquires” con-
cerning the possibility of bypassing the Saudis and remitting money directly,
“through other channels”.25 In Zanzibar, ʿUmar clearly reported that the money
never reached the intended beneficiaries, as defined by two generations of
Zanzibari practice. Although his attitude may have derived from a qāḍī ’s need
for orderly book-keeping, the Sufi distrust and dislike of the Wahhābī Saudi
government should also be taken into account. Personal relations is another
explanation; most likely, the families of Daḥlān, al-Ḥibshī etc. did not rank on
top of the Saudi distribution list and ʿUmar b. Sumayṭ and his colleagues in
Zanzibar may have felt obliged to continue to remit funding to people they

23 Letter from Downing Street to the British Resident at Zanzibar, 16 May 1927, ZA-AB34/38.
24 Minutes of Meeting of the Wakf Commission, 24 June 1946. ZA-HD10/5.
25 Letter from the Secretary of the Wakf Commission to the British Embassy of Saudi Arabia,
18 December 1956. ZA-BA10/5.
consolidating the network 179

considered family, important teacher and students, as well as fellow scholars

within their respective ṭarīqas.
In 1958, ʿUmar returned to Mecca for a second time, to discuss the matter
further. Again, it seems that ʿUmar b. Sumayṭ preferred to keep the money,
which now had amounted to an annual 1000 Pounds Sterling, well away from
the Saudi Government. Unfortunately, this is where the file ends. It is unclear
if the Zanzibari “Meccan waqfs” finally became integrated into the Saudi Dār
al-Awqāf, or if it continued to be remitted through the long-standing, parallel
network of Sufi channels.

Waqf Distribution within Intellectual Networks: Consolidating

Reform through Waqf Funds

The preoccupation with teaching is evident when we look at the distribution

of the Zanzibar waqfs in Mecca. It is clear that the contribution was carefully
distributed among a selection of teachers whose intellectual orientation had
been central to East African reformist Sufi circles since the mid/late nine-
teenth century.
We must also assume that these contributions (as well as corresponding
ones from elsewhere) enabled Shaṭṭā-Junayd-Daḥlān scholars to continue
their work in Mecca, especially after the Wahhābī-Saudī takeover of the
Ḥaramayn. For more than two generations, decade after decade, the scholarly
class of Zanzibar – in their respective capacities as qāḍīs and members of the
Wakf Commission – remitted funds to their colleagues in Mecca.
Besides their official roles, these men could activate a network of fellow
scholars and relatives. This was not the British Colonial authorities and their
banking facilities, but the scholarly network of qāḍīs, teachers, and students,
all of whom knew each other personally, had studied with each other or each
other’s fathers and had married off sons and daughters to each other. The
same description fits well on the recipients of the waqf funds: the money was
in effect distributed among a group of people who shared common shaykhs,
common teaching and a dedication reformist Sufism, be that ʿAlawī or Qādirī.
The private correspondence between al-Amawī and Shaṭṭā leave little doubt
that the actual “Meccan” distribution was entirely beyond the scope and under-
standing of the British members of the Wakf Commission. From the British
point of view, the commissioners were quite satisfied as long as the commis-
sioners saw receipts, accompanied with the seals and signatures of recipients.
180 chapter 8

Image 6 Receipts from Mecca with signatures and seals.

Zanzibar National Archives.

The Ḥātimī waqf also leaves little doubt that the actual selection of recipients
was formulated not by the original waqfiyya, but rather by the long-standing
relationship between shaykhs and scholars in both Zanzibar and Mecca. As
was shown above, the pattern of distribution may date back to the time of
Shaykh Muḥyī al-Dīn al-Qaḥṭānī, and through him to the “Brawanese” network
of Qādirī scholars, such as ʿAbd al-ʿAzīz al-Amawī and ʿUmar Qullatayn – and
finally to Ṭāhir al-Amawī. One may also guess that the generation of Aḥmad
b. Sumayṭ and ʿAbd Allāh Bā Kathīr (i.e. ca. 1890–1920) – with their close links
to Ḥaḍramawt, Mecca and Medina – were instrumental in establishing a pat-
tern that favoured the Shaṭṭā, BabṢayl and Daḥlān families, as well the students
and staff of the Sāda Ribāṭ. This was the pattern that Shaykh Ṭāhir inherited,
and which he chose to continue, and which ʿUmar b. Sumayṭ maintained into
the 1950s. In other words: Intellectual links that were formed in the 1880s and
1890s were continued well over eighty years later, in the form of continued
scholarly contact and as financial support. From the correspondence of ʿUmar,
the distrust of the Saudi waqf distribution system is spelled out clearly, and
the underlying theme is that of continued support of a “reform from within”,
within the parameters of Sufism. Most likely, this pattern was only broken by
the Zanzibar revolution in January 1964.
consolidating the network 181

From Networks to Organizations: The Rise of the Jamʿiyya,

c. 1900–1930

A hallmark of the modernist/Salafi reformist movements of the 20th century

has been the formation of organizations and associations, which often had
programs that went beyond the immediate teaching of religion, and typically
offered health-care, secular education, support for women and children etc.
As the above section has showed, the organization of the “Sufi reform” taking
place from the late 19th century was primarily organized as networks around
family and religious background, where educational and social efforts were
financed through traditional means, notably waqf.
In the 1920s, a new organizational form appeared; in the form of jamʿiyyāt or
associations founded for specific purposes or on the basis of specific religious/
ideological positions, while still retaining many long-established ethnic- or
family networks. These changes will be discussed both in the wider context of
Islamic reformist developments, as well as the existing networks from which
the new associations emerged.

The Network as Organizational Model: The Qādiriyya and the

ʿAlawiyya
In addition to an allotment for “the poor of Mecca and Medina”, the Ḥātimī
waqf discussed above also stated that one quarter was to go to the upkeep of
Sunni mosques in Zanzibar Town. In reality, the funds seem to have been spent
largely on the upkeep of the Ḥātimī mosque in Kiponda, (presently known as the
Halwa Mosque).26 The Ḥātimī family was one of several influential Brawanese
clans in Zanzibar, alongside the Amawī, Qullatayn and al-Qāḥṭānī families,
whose members we encountered in Chapter 3 as khalīfas of the Qādiriyya and
in Chapter 6 as authors of manuscript texts, teachers and transmitters of writ-
ten Islamic knowledge. Although less is known about their role in the Zanzibari
context, the Āl al-Ḥātimī too was a scholarly family, and included for example
Imām Aḥmad b. al-Rāfiʿī al-Ḥātimī al-Barāwī who was a copyist and a reciter
in the Ḥātimī mosque.27 Another mosque closely ­connected to the Brawanese
community was the Mnara mosque (Mskiti Mnara). It was reportedly built by
Brawanese immigrants, and was particularly connected to the Amawī family.

26 Abdul Sheriff, “Mosques, Merchants and Landowners in Zanzibar Stone Town”, Azania 27,
British Institute in Eastern Africa, Nairobi, 1992, pp. 1–19.
27 A.K. Bang, “Authority and piety.” Although we cannot be certain, it is likely that Imām
Aḥmad was either the father or the uncle of Aḥmad al-Ḥātimī, the founder of the waqf for
the Ḥātimī mosque.
182 chapter 8

Marriage connections were close within the community. The marriage

between ʿAbd al-ʿAzīz al-Amawī and Shaykh Uways’ daughter has already been
mentioned (Chapter 2). Another example is Ṭāhir al-Amawī, who was married
to Fāṭima, the daughter of ʿAbd al-ʿAzīz’ son Burhān al-Amawī.28
This Zanzibari Qādirī network was not only highly educated and high-­
profile; it was also well-connected. Moreover, it was cemented by family- and
ethnic ties, as well as by a series of ijāzas, whereby the scholars appointed their
selected khalīfas to positions within the order. Organizationally, however, the
“Brawanese Qādiriyya” before the 1920s did not move beyond the organiza-
tional structure of the Sufi order, i.e. khalīfa-appointments serving as religious
authorization. Funding for ṭarīqa activities (such as teaching, and the orga-
nization of the mawlid, dhikr and ziyāras) seems to have come mainly from
waqfs, such as the one established for the Ḥātimī mosque. It is also possible
that some funding came directly from the Bū Saʿīdī sultanate, although no
sources have been found that can confirm this directly.
The picture is very similar if we turn to the Riyadha mosque and the ʿAlawī
network surrounding it. A significant waqf for the Riyadha mosque in Lamu
was signed on the 20th of Shawwāl 1320/20 January 1903.29 The waqfiyya was
registered with the Lamu Registrary of the East Africa Protectorate on the
same day and one month later, it was returned, duly stamped by the colonial
authorities. The waqf was established by Sayyid Abū Bakr b. ʿAbd al-Raḥmān,
who, as the waqfiyya states, was “of the Shaykh Abū Bakr b. Sālim”, but is bet-
ter known in East African historiography as Sayyid Mansab b. Abd al-Rahman.
As outlined in Chapter 2, he studied in Ḥaḍramawt and Mecca, with ʿAlī b.
Muḥammad al-Ḥibshī, ʿAbd al-Raḥmān al-Mashhūr and Aḥmad Zaynī Daḥlān,
and he was also encountered in Chapter 6 as the scholar who provided the
definitive vocalization of Mawlid al-Barzanjī.
According to the waqfiyya, Sayyid Manṣab b. ʿAbd al-Raḥmān donated as
waqf the land “on which stands the Riyadha Mosque”, plus additional sur-
rounding land for the upkeep and support of the Riyadha. The waqf states
explicitly that funding shall be made available for the reading of the Sūrat Yā
Sīn, the Sūrat al-Ikhlāṣ, blessings upon the Prophet and other prayers, as well
as for the Imam of the mosque, “Sayyid Ṣāliḥ b. ʿAlawī Jamal al-Layl or whom-
soever replaces him”. Habib Saleh himself also established several waqfs for the
Riyadha, as did many of his family members and other inhabitants of Lamu

28 ZA-HC15/4. The file concerns the estate of Shaykh Ṭāhir.

29 Waqfiyya, signed by Sayyid Manṣab b. ʿAbd al-Raḥmān and registered by the Lamu
Registrary, in the possession of the Riyadha Mosque.
consolidating the network 183

Town.30 Over the years, several shambas on Lamu and Pate were also made
waqf for the Riyadha.
The organizational form was traditional and had the form of a network.
Among the foremost intellectuals in the Riyadha network was Aḥmad b.
Sumayṭ. In the years 1880–1881, he spent a period of study in Ḥaḍramawt, with
the same teachers who had taught Sayyid Manṣab b. ʿAbd al-Rahmān. Some
years later, in 1885–86, he studied in Mecca, with the associates of Aḥmad
Zaynī Daḥlān, who by that time had died. Not only was Aḥmad b. Sumayṭ, like
Habib Saleh and Sayyid Manṣab, of Ḥaḍramī ʿAlawī descent, he had also grown
up with Habib Saleh in Grande Comore. The two had the same teacher in their
early years, Abū ’l-Ḥasan b. Aḥmad Jamal al-Layl, known as Mwinyi Bahasani,
who was introduced in Chapter 2 as an important figure in developing Islamic
education in Grande Comore. Abū ’l-Ḥasan was also an elder cousin of Habib
Saleh. Marrying into the same Comorian female descent line, Habib Saleh
and Aḥmad b. Sumayṭ also shared another family connection, which in the
Comoros is considered equally, if not more, important than the patrilinear
descent line.31
As outlined above, another important person in the Riyadha network was
the Lamu-born scholar ʿAbd Allāh Bā Kathīr, whom we have encountered
repeatedly in this book. He founded the Madrasa Bā Kathīr in Zanzibar, which
was an exponent of Sufi reform within an educational framework. It is worth
noting that the Madrasa Bā Kathīr, like the Riyadha, was funded partly by sup-
port from the al-Ḥusaynī (Shaykh Abū Bakr b. Sālim) family.32 What should
also be noted for this discussion is that Bā Kathīr married a daughter of Habib
Saleh as well as the daughter of another member of the Jamal al-Layl clan,
the Zanzibar-resident scholar Ḥasan b. Muḥammad b. Ḥasan Jamal al-Layl
(d. 1904), author of the Marsūmat al-ʿAyniyya that was described in Chapter 6.

30 By 1969, the Riyadha was endowed with 44 waqfs in and around Lamu town, and the
names of their founded show a dominance of ʿAlawīs, but also non-ʿAlawīs names. Ten
of these were endowed by women, including the daughter of Habib Saleh. Document in
possession of the Riyadha mosque.
31 S. Blanchy, Maisons des femmes, cités des hommes : Filiation, âge et pouvoir á Ngazidja
(Comores), Nanterre (Société d’ethnologie), 2010.
32 R. Loimeier, Between Social Skills, 105. ʿAbd Allāh Bā Kathīr built his house and school in
Zanzibar upon land owned by the Ḥusaynī family, decendants of “Mwinyi Mkuu” Sultan
Aḥmad (1792–1875), the Sultan of Moroni, Grande Comore. This is another branch than
the one personified by Sayyid Manṣab b. ʿAbd al-Raḥmān in Lamu, but confusingly also
with a “Sayyid Manṣab” as a famous religious scholar; Sayyid Manṣab b. ʿAlī. On him, see
below.
184 chapter 8

What we see is a network of people of Ḥaḍramī, mainly ʿAlawīs ʿAlawī

descent that were connected through family, teacher-student relations, spiri-
tual connections and a common cause: An approach to Islamic practice that
acknowledged and supported elements such as grave visitation, dhikr, the
use of instruments in mawlid celebrations, but also the emphasis on the lim-
its imposed by the Sharīʿia. The network was reinforced by visits, the sending
back-and-forth of ijāzas, books and students. During the last decade of the 19th
century, and the first decade of the 20th, it also became consolidated in the
physical form of the Riyadha mosque itself, in the waqf for its upkeep and for
the religious leadership of the Jamal al-Layl family.

The Age of Associations: Into the 1920s and 30s

The rise of European colonialism in East Africa and the implementation of
British social, legal, educational and bureaucratic reforms disrupted coastal
communities in the early years of the 20th century. So far-reaching was the
impact of British colonial rule that it, according to R. Loimeier, “in fact set the
framework for the emergence of a different group of Muslim reformers who
were no longer associated with the esoteric episteme.”33
By the 1930s, a new generation of modernist-oriented scholars was emerg-
ing from within East African scholarly circles, including important figures like
Shaykh al-Amin al-Mazrui and Abdallah Saleh Farsy. What they had in common
was the rejection of some – none of them rejected all – features of established
Sufi traditions, such as for example grave visitations. They saw the traditional
Islamic schools as “unable to confront the challenges of modern times.”34 As
has been pointed out in several studies, neither of these early modernists/
Salafist were “outsiders”. On the contrary they came from within established
East African families with strong connections to the Qādiriyya, Shādhiliyya and
ʿAlawiyya. Al-Mazrui for example, had studied with both Aḥmad b. Sumayṭ and
ʿAbd Allāh Bā Kathīr, but became, from the 1920s, increasingly influenced by
the works of Egyptian reformers, notably Muḥammad ʿAbduh and Rashīd Riḍā.
Al-Mazrui founded the first reformist journal in East Africa; al-Shahīfa in
1930 – and then, in 1932, al-Islāḥ.35 Still, he remained close with the ʿAlawiyya
and taught in the Ghazzali Muslim School in Mombasa. He also composed the
educational work Hidāyat al-Aṭfāl, which was approved by the ʿAlawīs on the

33 R. Loimeier, Between Social Skills, 71.

34 R. Loimeier, Between Social Skills, 109.
35 On the publication history of al-Islāḥ and its ideological outlook, see N. Matthews,
“Imagining Arab Communities: Colonialism, Islamic Reform and Arab Identity in
Mombasa Kenya, 1897–1933”, Islamic Africa, 2, 2013.
consolidating the network 185

educational board for use in government schools in Zanzibar. Al-Mazrui’s criti-

cism of both social and religious traditions was explicit: Parts of Muslim social
life on the coast was, in fact, bidʿa; wrongful innovations.
One of the hallmarks of the emerging Salafi movement was the forming
of associations for the purpose of “guidance” or “enlightenment”. The classic
example is Jamʿiyyat al-Daʿwa wa’l-Irshād, founded by Rashīd Riḍā in Cairo in
1912. This jamʿiyya soon grew into a college that drew students from all over the
Muslim world.
However, there are also several examples associations that remained within
the esoteric episteme. One example is Jamʿiyyat al-Khayr, founded in Batavia
(Indonesia) in 1901, and registered 1905.36 This organization was established
mainly by ʿAlawī reformist-oriented merchants, with a purpose of providing
“modern education” for young Arabs in the colony. It engaged in teaching
Arabic and religion, and recruited teachers from the Middle East. The Jamʿiyyat
al-Khayr also held weekly sessions for adults, men and women. The orienta-
tion was pan-Islamic (its leaders expressed support for the Ottoman Empire),
but its outlook was clearly Sufi based. Here, its stood in contrast to the some-
what later Jamʿiyyat al-Iṣlāḥ wa’l-Irshād al-Islāmiyya (Association for Reform
and Guidance), founded in 1915 in Batavia under the tutelage of the Sudanese
reformist teacher Ahmad Surkatti and like-minded Ḥaḍramis. Its main activ-
ity was the running of schools, but it also supported journalism, hospitals and
primary health care.37 This organization was clearly modernist, Salafi in orien-
tation, and relied heavily on Muḥammad ʿAbduh and Rashīd Riḍā in its views
on education and daʿwa.
In Zanzibar itself, the emerging religious associations could draw on the
example of ethnically based associations that were formed in the first decades
of the 20th century. The first was the Arab Association, formed in 1901, but only
formally registered in 1922. According to A. Ghazal,38 the constitution of the
Arab Association was drawn up in 1920. Its purposes were many: Originally, it
was founded to claim compensation from the British after the 1897 abolition
of slavery, but later developed into an organization to protect Arab economic
interest overall, vis-à-vis the Indian and the Swahili population. The Arab

36 On the Jamʿiyyat al-Khayr and the Jamʿiyyat al-Iṣlāḥ, see U. Freitag, Indian Ocean Migrants,
230–250 and passim.
37 N. Mobini-Kesheh, The Hadrami Awakening. Community and Identity in the Netherlands
East Indies, 1900–1942, Cornell Univ. Press, 1999: U. Freitag, Indian Ocean Migrants, 248–
253 and passim; M.F. Laffan, Islamic Nationhood and Colonial Indonesia. The Umma below
the winds, London (Routledge), 2003, 189–210.
38 A. Ghazal, Islamic Reform and Arab Nationalism, 97.
186 chapter 8

Association was for example consulted during the struggle to include reli-
gious education in the government schools,39 and was also active in support-
ing Arabic language education. The main mouthpiece of the Arab Association
was the journal al-Falaq, first appearing in 1929 (its predecessor was al-Najāḥ,
edited by Abū Muslim al-Rawaḥī), edited by Hāshil b. Rashīd al-Maskarī.40 It
had a wider reach than Zanzibar alone and spoke to the entire Omani intel-
ligentsia, but also the Arab nahḍa overall.
Another ethnic organization formed in the early 20th century was the
Comorian Association,41 which dates back to 1911, when a waqf was established
under this name. The formal association of the same name was founded in
June 1924, explicitly to unify Comorians who had previously belonged to vari-
ous “county” (midji) groups, depending on their place of origin in the Comoros.
The purpose of the Comorian association was to abolish the costly midji fees
for marriages and funerals (this was explicitly formulated in the rule book of
the association). Its leadership consisted mainly of young, modernist-minded
men; Ibuni Saleh, Mgeni Ali and Saleh Yahya. These were known locally as the
“Yaminis” (“rightists”).
A rival association formed shortly after, “The Comorian Association Liberal
Party” (Hizb al-Ahrar) in 1925. This organization did not prohibit members
from also being a part of the midji groups – in fact it functioned as an umbrella
organization for them. Its members were known as “shimalis” (“leftists”) and
saw themselves as defenders of custom.
The ethnically based associations that emerged in the twentieth century
must be understood as a function of the British urge to organize the hetero-
geneous Zanzibari society into comprehensive “segments”, typically based on
what they defined as “ethnic” categories. As L. Fair has shown, this was not only
a matter of racial definition, but also a matter of access to political represen-
tation, employment, education and food.42 The “pigeonholing” of Zanzibari
residents even went beyond regulating access to social and political influ-
ence, but, as W.C. Bissell has shown, ordered the regulation of the city itself,

39 A. Bang, Sufis and Scholars, 173–177; R. Loimeier, Between Social Skills, 306–308.
40 A. Ghazal, Islamic Reform, 99.
41 I. Walker, “Identity and citizenship among the Comorians of Zanzibar”, in: A. Sheriff and
E. Ho, The Largest Continuum. Oceanic Movements and formation of new societies in the
Indian Ocean, forthcoming, London (Hurst), 2013.
42 L. Fair, Pastimes and Politics. Culture, community and identity in post-abolition urban
Zanzibar 1890–1945, Oxford (J. Currey), 2001. See also A. Purpura, Knowledge and Agency:
The Social Relations of Islamic Expertise in Zanzibar Town, PhD Dissertation, City
University of New York, 1997.
consolidating the network 187

built on a British vision of “racial containment” in separate quarters.43 Here,

the ethnically based organizations can be viewed as entities organized for the
colonial powers to have representatives to talk to, and which they viewed as
the sole representatives of organizational life. However, as the example of the
Comorian associations demonstrate, internal, and opposing interests, also led
to the formation of fraction-groups, irrespective of the British ideal of “one
ethnic group, one organization”.
While the ethnic associations certainly can be viewed as creations of colo-
nial rule in Zanzibar, the connection is less obvious for the new type of reli-
gious organization that emerged in the 1920s; the jamʿiyya.

The Religious Jamʿiyyāt in Zanzibar: The Hallmark of the Modernists –

and the Sufis?
As outlined above, the most explicit change in last decades of the nineteenth
century was the emergence of the ṭarīqas in East Africa. Towards the end of
the century, and in the first years of the twentieth century, the most important
organizational innovation was the rise of mosque-colleges; a phenomenon
which, as we have seen, can be dated back to the 19th century in East Africa
and even earlier in the Ḥaḍramawt and Mecca. These were based in scholarly,
Sufi-oriented networks and funded by traditional means (waqf). In the 1920s,
a new form of religious organization emerged, both within the Sufi-oriented
groups and the more modernist-oriented ones. In other words, what we see is
a move from the previous, ṭarīqa- and family based network to formal associa-
tions with a president, a written statement of purpose etc.
In Zanzibar, the first (known) association with an explicit religious
agenda was al-Jamʿiyyat al-Islāmiyya, founded in 1922.44 This organization
was founded by the Zanzibari scholar Sālim b. Saʿīd b. Sayf al-Shuhaybī, an
Ibāḍī who changed his madhab to Shāfiʿism and studied under the tutelage of
Aḥmad b. Sumayṭ and ʿAbd Allāh Bā Kathīr. Loimeier describes this as “prob-
ably the first Salafi organization in Zanzibar”,45 primarily on account of the
activities of al-Shuhaybī’s main student, Rashīd b. Ḥamīd al-Ḥaḍramī (d. 1963).
Another organization – probably with a similar aim, but about which little is

43 W.C. Bissell, Urban Design and Colonial Power in Zanzibar, Bloomington (Indiana
University Press), 2011. As Bissell also demonstrates, the British never truly succeeded in
implementing their vision of a city divided into “racially” organized quarters.
44 Loimeier, Between Social Skills, 106, 111; Farsy/Pouwels, 114
45 Loimeier, Between Social Skills, 106.
188 chapter 8

known – was al-Jamʿiyyat al-Islāḥiyya, also founded in Zanzibar 1922 by one

Naqīb ʿAbd al-Karīm Ṣalāḥ.46
The emergence of a corresponding organization with a Sufi outlook was not
long in the making. In 1927, Jamʿiyyat al-Sunna was founded in Zanzibar, evi-
dently by a group of sayyids and scholars closely connected with the Madrasa
Bā Kathīr. Its first president was Sayyid Manṣab b. ʿAlī (1863–1927).47 He was
a well-known scholar in Zanzibar, and has been portrayed by Abdallah Saleh
Farsy as somewhat of a modernist, reading al-Manār and painting a portrait
of the Sultan Sayyid Khalīfa b. Saʿīd. However, in this context, it is clear that
Sayyid Manṣab was part of a circle of scholars whose outlook did not reject
the Sufi paradigm, but rather aimed social reform from within. Another found-
ing member and the second president of the organization was ʿĪsā b. ʿAlī
al-Barwānī, who by family was closely connected to the Madrasa Bā Kathīr.48
ʿĪsā b. ʿAlī was encountered above as part of the network of trust within which
money was transmitted from Zanzibar to Mecca. Finally, the Secretary of the
newly formed organization was Abū Bakr Bā Kathīr, the son of ʿAbd Allāh Bā
Kathīr.
Also a prominent member was Sayyid Abū ’l-Ḥasan Jamal al-Layl, whose role
as a Qādirī scholar and transmitter of the order to Mozambique was described
in Chapter 3 and whose efforts as a public educator was outlined in Chapter 6.
Here, he emerges as a fund-raiser for the Jamʿiyyat al-Sunna, as outlined in his
obituary from 1959:

When the Jamiyyat al-Sunna (“The Sunna Association”) was founded in

1345 (1927), Sayyid Mansab became its first president. Then, Shaykh Isa
al-Barwani became president, and Abu Bakr BaKathir became Secretary.
Abu al-Hasan Jamal al-Layl became the leader of a committee that was
appointed to raise funds for the Jamiyya.
He did much work in order to bring the Jamiyya forward and to recruit
people and promote the name of this Jamiyya. He was sent to the coun-
tryside, in Pemba and in the mainland, to raise money, which he did.

46 Farsy/Pouwels, The Shāfiʿī Ulama, 114.

47 On him, see Farsy/Pouwels, The Shāfiʿī Ulama, 22–24. Sayyid Manṣab b. ʿAlī was the
grandson of Mwinyi Mkuu Sultan Aḥmad of Moroni and the Madrasa Bā Kathīr was built
upon part of his family land in Ukutani, Zanzibar Stone Town.
48 ʿĪsā b. ʿAlī al-Barwānī was the maternal uncle of Ali Muhsin. Ali Muhsin al-Barwani,
Conflicts and Harmony in Zanzibar, Dubai, 1997, 26–27. Note also Loimeier, Between Social
Skills, 104, on the relation between Bā Kathīr and the Barwanis: The mother of Bā Kathīr
was a Barwani (Binti Abdallah).
consolidating the network 189

Then, he was sent to Pemba to build a new Friday mosque in Wete for
some of the money that had been collected and he built this mosque
Kitutia where Friday prayers were held, as well as other prayers and edu-
cation, year after year, until some time ago a very big Friday mosque was
built in Wete and this one was indeed demolished, and the new was built
partly upon the old one.
He was given this task of traveling everywhere and to raise funds
because he was well-known and a respected person whom all spoke well
about [. . .].49

Little is known about Jamʿiyyat al-Sunna in Zanzibar, as no sources have yet

appeared about its inner life. What is certain is that it clearly resembles simi-
lar organizations for example in Indonesia and elsewhere, with its purpose of
“beneficial” (“khayrī”) work, such as building mosques and schools and sending
teachers to remote areas.
It is possible that Jamʿiyyat al-Sunna was founded as a response to the
two organizations above, which seems to have had a more Salafi approach.
Another likely assumption is that Jamʿiyyat al-Sunna was modelled on (or at
least mutually inspired by) a corresponding organization in Grande Comore
named Jamʿiyyat Ikhwān al-Hudā (The Society of the Brothers of Guidance).
The central figures behind this organization were ʿUmar b. Sumayṭ, Sālim b.
ʿAbd Allāh Wadʿān (d. 1942) and Aḥmad Qamar al-Dīn (d. 1974).50 Especially
the two former had very strong connections with Zanzibar and spent much
of their lives there as scholars in various capacities. It is thus very likely that
the men behind the two organizations were well aware of each other’s activi-
ties and reformist objectives. In fact, ʿUmar b. Sumayṭ mentions one occa-
sion where Abū ’l-Ḥasan, Burhān Mkelle and Sālim b. ʿAbd Allāh Wadʿān were
together, clearly counted among ʿUmars “friends” (aṣdiqāʾ) from the circle
of the Madrasa Bā Kathīr. They all attended a celebration upon the return
of ʿUmar from Ḥaḍramawt in 1923, and on this occasion, both Abū ’l-Ḥasan,

49 Obituary, Mwongozi. I am grateful to Kjersti Larsen for assistance in the translation of this
text from Swahili.
50 T.A. Mohamed, Ahmad Qamardine, 323–328. Shaykh Sālim (d. 1942) was born on Grande
Comore to a father of Ḥaḍramī origin. At a young age, he was sent to Zanzibar to study.
Later, he became editor of the journal al-Falaq. He was also a central figure in two other
organizations in Zanzibar: The Young Hadrami Association and the Young Muslim
Association.
190 chapter 8

Burhān and Salim b. ʿAbd ʿAllāh composed poetry (although only that of Sālim
b. ʿAbd ʿAllāh is quoted).51

From Networks to Organizations: Responses to Colonial Rule or

Salafi Modernity?
The shift from looser networks, built on Sufi connections (teacher-student),
family or ethnic origin to organizational forms that has an explicit, socially
beneficial purpose, can be understood in different ways. The jamʿiyya was
clearly a new organizational form for the Sufi scholars on the coast, at least
when it comes to religiously based activities. However, contrary to the “eth-
nic organizations”, which were responses to the British authorities, societ-
ies such as Jamʿiyyat al-Sunna clearly had an older genealogy, and were built
upon other connections than intellectual concordance alone. Rather, from
the names of members and what we know of their activities, it is clear that
Jamʿiyyat al-Sunna was rooted squarely in the same Sufi network that we have
seen conducting money transfers, building mosque-schools, sending out
daʿwa scholars, writing, reading and importing books since the 1860s. The new
organizational form is thus best understood as yet another way of consolidat-
ing existing network, in addition to the money flow that was described above.
In fact, one may even view the jamʿiyya as another way to obtain funding for
the same goal; Abū ’l-Ḥasan’s efforts to raise money for the mosque in Wete is
different from the fundraising for the Riyadha only in method; not in scope.
As has been pointed out by R. Loimeier, the jamʿiyya may be the hallmark
of twentieth century Salafism, but in terms of programme, the Sufi and Salafi
societies were, in fact, remarkably similar; “doing good”, building mosques,
providing education, alleviating poverty etc.52 As has been demonstrated by
S. Reese, the genealogies of these organizations, too, were deeper than mere
intellectual orientation, as Salafi-oriented organizations grew out of older Sufi
networks, but took on the hallmarks of Salafism in terms of social activism.53
Although there are many similarities between the Salafi and Sufi-based organi-
zations in East Africa, there is one thing we don’t see among the Sufis: journals
and newspapers. Instead, as was outlined in Chapter 6, we see an increased
activity of reprinting (i.e. funding of reprinting) of “classics”, especially devo-
tional literature. The main difference then seems to be their different usage of
the print media, at least in the first decades of the 20th century.

51 ʿUmar b. Aḥmad b. Sumayṭ, Al-Nafḥa al-Shadhdhiyya ilā al-Diyār al-Ḥaḍramiyya

wa-ṭalbiyyat al-ṣawt min al-Ḥijāz wa-Ḥaḍramawt, 110.
52 R. Loimeier, “Is there something like ‘Protestant Islam’?”, 233.
53 S. Reese, “Salafi transformations.”
 chapter 9

Conclusions

On Ripples and Reefs: Agency in a Translocal World

Returning to the imagery of the ripple and the reef, the introduction to this
book outlined a dual focus, whereby both the translocal Sufi-teacher and
the local community were assigned equal agency. What we have seen over
the past chapters is a repeated pattern, whereby translocal agents – all part
of a network that had more than one centre – arrive in locations that have
a specific ethno-linguistic composition and specific hierarchies of their own.
By the early twentieth century, the daʿwa scholar is seemingly everywhere; in
Mozambique, in northern Madagascar, and in Cape Town – and surely also in
several other locations that are beyond the scope of this study. He (the net-
work is male, although we do catch occasional glimpses of the education also
of females) preaches, he converts people, he organizes dhikr, introduces new
rituals or transforms old recitals to scheduled rituals, such as was the case with
Rātib al-Ḥaddād. He refers to text for authority, be that genealogical (as was
the case with Ḥasan Jamal al-Layl in Zanzibar) or theological (such as with
the wide diffusion of legal texts). Above all, he establishes schools, with educa-
tional programmes that go beyond the traditional Sufi teacher-disciple pattern
(The Riyadha, the Madrasa Bā Kathīr and the Azzawiya are all examples of
this). In all these activities, he is drawing upon pre-existing family networks,
but also forging new connections, often beyond his own ethnic or linguistic
background and into new communities. To do this, he may even have made
a conscious effort to understand the customs and habits of his new place of
residence, “to settle into its ways, to know its people”, as Burhān Mkelle wrote
about ʿĀmūr b. Jimba in Mozambique.
At the same time we see that the new organizations (i.e. ṭarīqas) intro-
duced by these translocal agents do not produce an “output” that overrides
local stratification and/or social classification structures, be they ethnic, lin-
guistic or descent-based. Neither in Mozambique, Madagascar or Cape Town
do the arrival of these agents of “Indian Ocean Islam” result in a construction
of Islamic identity that overrides earlier categories. Instead, we see a ten-
dency to assign authority (sometimes retrospectively, as the example of the
Qādiriyya in Ilha de Mozambique shows) to the source that best fits the “facts
on the ground”. In Madagascar, the gradual formation of a Malagasy branch
of the Shādhiliyya, with a parallel branch mainly dominated by Comorians,

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�10

192 chapter 9

show a similar ­development. In Cape Town, the steady “import” of teachers

show a diversification of thought and practice, and – at times – a contesta-
tion of authority. However, all the cases show a re-orientation towards glo-
balized Islamic discourse, while maintaining aspects of pre-exisiting social
organization.
This in itself is hardly new knowledge. L.J.K. Bonate in 2007 showed the
importance of local kinship ties in the incorporation of ṭarīqa Islam, and
N. Guenier has illustrated the emergence of a “Malagasy Islam” as being ṭarīqa
based. In the case of Cape Town, Yusuf da Costa showed already in 1989 the
variety of “receptions” of Islam to the local community. In fact, when posing a
question that assigns dual agency to translocal agents and local communities,
the resulting answer is almost bound to be “yes, please, a little bit of both”.
In very few cases in global intellectual history will the response of a local com-
munity or a people to new ideas be either a total acceptance or a unanimous,
wholesale rejection.

Sufi Reform on the Move

The picture becomes more complicated if we nuance both sides of the ques-
tion. As argued in the introduction, the emergence of ṭarīqa-based Islam in
the late nineteenth century produced “something new” in a diverse set of loca-
tions. It was also argued that we, as historians, may observe this in the set of
texts circulated both within the translocal network, as well as in the locations
where they were used. Furthermore, historians have access to the introduction
of new rituals forms and (at least to some extent) to the ways in which the
drive towards change was organized. Overall, its was argued that this “change
with a programme and a willingness to travel” constituted a reform movement
which remained within the esoteric (Sufi) episteme, and which thus may be
viewed as a forerunner to the modernist/Salafi reformist movements of the
twentieth century. The question then becomes how “change”, as agitated by
translocal agents and local leaders alike, came to influence both local society
and Islamic thought overall.

Textual Dissemination and Usage

Chapter 6 discussed the widespread production and circulation of Islamic
texts in the latter half of the nineteenth century, mainly in the scholarly
­centres of East Africa (Lamu, Zanzibar, Comoro Islands), but also in Cape
Town. The ­argument was made that we see an increase in production and cir-
culation in the period, and that this, in turn, resulted in new forms of authority.
conclusions 193

The amount of Islamic material in Arabic dating from this period is simply
too large to be explained as a continued tradition, implying that whatever was
there before has just been lost. This said, it must also be emphasized that the
emergence of the mosque-colleges themselves played a vital part in preserving
the collections, and that there thus may have existed more Islamic scriptural
material also before the nineteenth century but that no institutions were in
place in Lamu or Zanzibar act as long-term repositories.
If we look at the composition of the scriptural material, we find three main
themes, all of which can be related to authority. First, the prevalence of texts
related to Arabic language and grammar indicates that access to the sources of
Islam was becoming paramount for authority. Secondly, the emphasis on fiqh
indicates again that reference was increasingly being made to sources external
to the local context in matters of law. Thirdly, the wide prevalence of devotional
texts indicate the “standardization” of ritual practice with reference to text, not
solely to the ritual leader. The way the vocalization of Mawlid al-Barzanjī was
standardized is a case in point. However, as S. Reese has pointed out, “Discursive
interconnectedness [. . .] should not be confused with intellectual uniformity.”1
This strong element of interconnectedness, combined with diverse paths of
dissemination and usage is one of the primary findings of this study: While
the emphasis on Arabic (and thereby on access to the “global Islam – how-
ever defined) was universal, the modalities of how text was employed and used
were not the same in all locations. In Lamu, Zanzibar and Cape Town, the texts
surveyed here were used in schools, and the purpose of their very existence
was educational. In Ilha de Mozambique, where no comparable school was
established in the period, we see more of the devotional texts transmitted by
Sufi orders; typically the Qādiriyya dhikrs and the poetry of Shaykh Uways. The
picture can thus be nuanced: Not only does the ripple and reef jointly produce
the impact of the wave, but so does the form and shape of both. The presence
or absence of educational institutions in one such variable, where there is still
room for more research. Another is in the network itself: Details are still miss-
ing on how exactly books and written material came to be produced, bought
and sold, seen into print and distributed.

1 S. Reese, Renewers of the Age, 5. Reese’s agenda here is to nuance the picture of the “accom-
modation” versus rejection of Europe and colonial rule, but the argument can also be made
for the various types of dissemination and usage of Islamic textual material.
194 chapter 9

Rituals, Texts and the Sufi Orders in the Southwestern

Indian Ocean
The various dhikrs and recitals advocated by the late nineteenth and early
twentieth Sufi orders is what came under attack from the later generations
of reformer, variously clothed as modernists, Salafis or Wahhābīs. The notion
of intercession, as well as the idea of praying on behalf of others, was thought
to violate the principle of tawḥid and assign to others (saints, ritual leaders)
powers that belong to God alone. The recital of the Qādirī dhikr, for example,
was certainly a novelty when new ṭarīqa recruits gathered to recite it in Ilha de
Mozambique in the late nineteenth century. The rituals of the Shādhiliyya, too,
were new in Zanzibar, Ilha de Mozambique and Diego Suarez.
The history of Rātib al-Ḥaddād, on the other hand, is an example of previ-
ously existing ritual (or at least one that we may safely assume was already
known, at least in some family circles) that became increasingly “regular-
ized” as new teaching institutions were founded in the late nineteenth cen-
tury. There are aspects of the recital that are clearly “global” and which refer
to authority beyond the local. The text is to be recited at night, and this is
known on the authority of al-Ḥaddād himself, as well as on textual elements
in the Rātib itself. However, it is important also here too, to emphasize the
effect of the reef, as the Rātib was regularized in ways specific to East African
and Capetonian teaching institutions to the extent that its recital became an
ingrained part of Islamic practice. In other words; the “proper” way of perform-
ing the Rātib became specific locations (Every day/every Thursday; after ʿishāʾ/
after maghrib; on what types of occasion etc.). As such, it became an expres-
sion not only of devotion, but also of identity, specific to place, rather than to
a diffuse network.

Organizing the Wave

We have seen in the previous chapters that Sufi daʿwa travellers came and
went, and we have seen scholars from Lamu, Zanzibar, Cape Town and else-
where travel to other centres in search of knowledge. On their journeys, they
brought along books, took locally produced manuscripts off for publication in
Cairo or Mecca, or even took notes for their own riḥla (travel) accounts to be
produced. We have seen them bring ideas on how the Rātib was to be recited,
and textbooks of Qādirī dhikrs to be recited on specified occasions. Finally, we
have also seen that this was not a wave that maintained momentum entirely at
random. Rather, a set of like-minded scholars channelled money within their
own circle to finance for example education (such as the funding for students
at the Sāda Ribāṭ in Mecca) and functions within their mosques (such as the
money assigned for Habib Saleh or “whomsoever replaces him” and al-Ḥātimī
conclusions 195

mosque in Zanzibar). Until the 1920s, the drive towards change – and the readi-
ness to travel! – was channelled through the traditional framework of Islamic
institutions, mainly the mosque, the madrasa and the waqf. The individuals
in the network knew each other after two generations of translocal scholarly
interaction, were to various degrees related to each other, and could operate
on the basis of trust – a trust that entirely eluded the British efforts to regu-
late the waqf “market”. This model differs substantially from its successor, the
Jamʿiyya, which was to become the main organizational form of twentieth cen-
tury reformism. Here, the “change with a programme” was formulated explic-
itly and from the outset, be that irshād (guidance) or iṣlāḥ (reform) or similar.
The individuals were here assigned clearly defined roles, and were expected to
act towards a defined goal (such as raising money for a new mosque in Pemba).
It is worth noting here that this study has shown several, and partly overlap-
ping networks of Sufi reformist groups that we may assume all operated in sim-
ilar ways, although varying according to access to political and legal power, as
well as funding. The Qādiriyya, Shādhiliyya and ʿAlawiyya dividing line is one,
and within each there were also sub-divisions that often corresponded to eth-
nic or socio-economic class, or with access precisely to overseas networks. The
Brawanese Qādirī network in Zanzibar Town and the network around Shaykh
Mjana Kheri is one example. As we have seen, they were all important trans-
mitters of the Qādiriyya, although with slightly different approaches and “tar-
get groups”. The jamʿiyya, on the other hand, could conceivably be the model
that could do what the traditional ṭarīqa organizational form could not; create
structures that could truly cut across existing social divisions and create a basis
for an Islamic identity beyond the local. It seems that in Zanzibar, at least, the
jamʿiyya was to become the vehicle preferred by the “twin waves” of Salafism
and nationalism, here understood in the same vein as Sufism is understood in
this book. Reformers within the Sufi episteme, as we have seen, continued to
organize its activities within the traditional format well into the 1950s, funding
schools by waqf and having books printed “ʿalā nafaqa.” Instead, through the
colonial experience and the process of decolonization, nationalism emerged
as the “new wave” which had the potential to supersede social, cultural and
political lines, with Salafism as a strong cross-current (or vice-versa, depend-
ing on point of view). Here, despite excellent studies by several researchers in
the past few years, details are still missing. A strong hypothesis is that the new
organizations (be they Salafi, Sufi or even political parties) built at least partly
on earlier networks.
196 chapter 9

The Ecumene that wasn’t – yet?

Summing up the above, we may say that the efforts of reformist Sufi leaders
and commoners in the southwestern Indian Ocean did change practice on the
ground, and that ideas of this change travelled. However, we have also seen
that changes in practice came to be understood as specific to place and/or
population group.
What we have seen is that the spread of Sufi orders in the southwestern
Indian Ocean and their implicit reform of Islamic practice took place based
on pre-existing networks and utilizing new modes of transmission. However,
as this study also has shown, they created new networks of authority that
relied on access to written textual sources, first and foremost the Quran and
the Sunna. They also relied on rituals that catered to the spiritual needs of new
populations. Finally, they also relied on a system of money transfer that oper-
ated within a circle of trust, but also with a budding set of new organizations
structured along different lines. This was transmitted through personal inter-
actions, sealed with authority issued in writing (“I appoint him as my khalīfa”)
and with rituals that expressed the same authority.
Before the full emergence of Islamic modernism/Salafism and – later –
nationalism, and before the transmission modes that D. Eickelmann called
“Islām al-Ṣawt” (let alone Islām al-Internet), these were powerful trans-
missions that carried a potential that frightened observers like C. Snouck-
Hurgronje in ways very comparable to the ways “internet jihad” today frightens
politicians and intelligence agencies (as well as certain media houses) in the
Western world. If the twin discourses of nationalism and Islamic modern-
ism/Salafism were to be the main levellers of the twentieth century Islamic
world, one should ask here if Sufi reformist thought had some of the same
effects in the late nineteenth and early twentieth century? The answer, as this
book has shown, is no. The processes of localization did not, in fact, realize the
potential inherent in Sufi reformist thought for an enlargement of scale in the
sense feared by C. Snouck-Hurgronje. The travellers may have come back from
Mecca, Ḥaḍramawt or elsewhere with notions of a global Islam, embedded in a
wider structure that was framed by European colonialism. Daʿwa scholars may
have set out with similar ideas. However, upon settling in locally, they were
inevitably implanted into hierarchies of power, social structures and cultural
practices that could not be ignored. For a fondy of the Shādhiliyya in northern
Madagascar to drink rhum is but one example; for the ʿAlawī travellers in the
Comoros to marry into matrilineal clans is another.
On the face of it, this book has shown how specific locations in the south-
western Indian Ocean became incorporated into a wider, globalized Islam.
conclusions 197

We have seen the Shādhiliyya-Yashruṭiyya spreading from Palestine, via a

meeting between al-Darwīsh and al-Maʿrūf in Mecca, to Zanzibar, the Comoro
Islands and onwards to Ilha de Mozambique and to northern Madagascar.
Although the trajectory of the ʿAlawiyya is slightly different, being already
widely diffused in the Indian Ocean, the way in which the educational activi-
ties of the ʿAlawīs operated in the late nineteenth and early twentieth century,
can be interpreted as an example of the same: Reformist ideas emanating from
a perceived centre (be that Mecca/Medina or Ḥaḍramawt) and then replicated
in new locations with reference to a textual authority. This “wave”, then, theo-
retically had the capacity (and in some cases, clearly also the intention) to flat-
ten the landscape, erase or alter previously existing hierarchies and ethnic or
linguistic categories, forming one Shāfiʿī-Sunni ecumene relying on the same
texts, performing the same rituals and organized along the same lines of reli-
gious affiliation.
As will be clear from the above chapters, this not what happened. Instead,
we see that the Sufi orders came to be organized along pre-existing kinship,
ethnic or linguistic lines. It is as if the wave is ecumenical as long as it remains
at sea, but takes on different capacities at landfall. This process is especially
clear in Ilha de Mozambique and in northern Madagascar, where the daʿwa
agents – the Comorians, and beyond them, the ʿulamāʾ of the scholarly centres
of East Africa, Ḥaḍramawt and the Ḥaramayn – were outsiders whose leader-
ship was not at all taken for granted. Instead, we see two parallel processes tak-
ing place: One which emphasizes local leadership and authority, and another
which draws on a translocal notions of authority. In northern Madagascar,
this translated into Malagasy (Antankarana) and Comorian silsilas. In Ilha de
Mozambique, it translated into the Qādiriyya and the Qādiriyya Sadat. The
Qādiriyya as spread by the students of Mjana Kheri to Palma present us a
slightly different picture, with no parallel chain of authority but rather a pro-
cess of localization that went through the chain of transmission among what
F. Becker called “commoners”. In Cape Town we see another parallel develop-
ment, as the Muslim community was both unified (qua Muslim community)
and fragmented (into different sets of beliefs which could or could not corre-
spond to ethnic lines).
On the most general level, the diffusion of a specific type of Islamic knowl-
edge in the Southwestern Indian ocean in the late nineteenth and early twen-
tieth century, and the ways in which it came to be appropriated in different
locations, addresses the question of the social and political consequences of
trans-societal interrelations. As J. Prestholdt has pointed out, the presence of
long-term migrational culture challenge the notion of discrete sociocultural
spaces and speaks directly to the issues of globalization which concerns social
198 chapter 9

scientists and publics today.2 The question then is to what extent the next gen-
eration of reformists, the modernists/Salafis/Wahhābīs actually “flattened”
the twentieth century landscape, or to formulate the final question of this
book more explicitly: to what extent did twentieth-century reformers in the
southwestern Indian Ocean build upon the lines established by the generation
before them, the local and translocal agents decribed in this book?

2 J. Prestholdt, Domesticating the World, 1.

Appendix 1

The Zanzibari “Meccan Waqfs” Contained in ZA-HD10

The compilation of this file was completed in 1926, when the Wakf Commission made
an effort to get an overview of the “Meccan Waqfs” in Zanzibar. As shown above, by
the example of the Ḥātimī waqf, it was by no means complete. A number of waqfs
containing clauses benefitting “the poor of Mecca and Medina” remained outside the
administration of the Wakf Commission for several years.

1. The Shamba “Magogani”, dedicated by Muḥammad b. ʿAlī b. ʿAmīr al-Barwānī

(Sunni).
2/3 of the revenue to go to the poor of Mecca.
In this case, the Wakf Commission knew that there had existed an original waq-
fiyya, dated 24 Shaaban 1308/ 4 April 1891. However, by 1926, only indirect refer-
ences existed to the original document.
2. Town Houses 105, 106, 107. Waqf for the poor of Mecca.
No knowledge existed of the original dedicator or document, only what the
British called “traditional” knowledge.
3. Town Houses no 86 and 87, Sokomohogo. For the poor of Medina and Mecca,
respectively.
No knowledge existed of the original dedicator or document, only what the
British called “traditional” knowledge.
4. Houses at Hurumzi. Dedicated by Sayyid Ḥammūd b. Aḥmad Āl Bū Saʿīdī.
Ibāḍī. Registered in 1904.

The dedication of this town house for Ibāḍīs in Mecca was part of a very substantial
waqf settled by Sayyid Ḥammūd b. Aḥmad Āl Bū Saʿīdī in 1877 (however, it only came
to be registered in 1904). Sayyid Ḥammūd was a member of the Bū Saʿīdī family, a close
companion of Sayyid Barghash, and a known benefactor for Ibāḍīs both in Zanzibar
and overseas.1 Until 1917, the administration of the houses at Hurumzi was supervised
by the Ibāḍī qāḍī, but it was then taken over by the Wakf Commission.2 At that point,
the houses yielded approximately 300 Rupees per month, a sum which the British
Commissioners considered to be well below its potential. Interestingly, the original

1 For an overview of the charities and educational efforts of Sayyid Ḥammūd b. Aḥmad, see
al-Mughayrī, Saʿīd b. ʿAlī, Juhaynat al-Akhbār, 362-363.
2 A detailed history of all the waqfs established by Sayyid ḤHammūd is contained in ZA-HD6/55
and ZA-HD3/12.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�11

200 appendix 1

document of this waqf explicitly states that the surplus is to favour deserving Ibāḍīs in
Mecca, and secondly deserving Muslims in general:

[. . .] The above Sayyid Ḥammūd b. Aḥmad has also made waqf his house, which
is situated at Hurumzi, [. . .] and which is built of lime and stone, and which he
acquired by purchase from Ladda Demji, the Banya [. . .] as perpetual waqf until
resurrection day, not to be sold and not to be given away as a gift.
That the said house be given on rent and that its income should first be uti-
lized towards its upkeep, whenever required and its caretaking, and towards the
upkeep of the adjoining passages surrounding the house on all four sides, and
that the remainder from the income be divided into two equal parts.
The first half is to go to the poor of Mecca and Medina – may God honour
these two cities for ever. And if there be found in these two cities poor who
belong to the Ibāḍī sect, or other Muslim poor who seek religious knowledge or
other knowledge pertaining to divine service, they should be given from such
income according to the extent of the income, and likewise if it is required for
the purpose of digging a well for passers-by or for burying poor Muslims or sup-
plying water to a pilgrim or other person, or for helping a needy worthy of help
or to provide clothing for the unclad or to feed a poor or an orphan or for any
other good and pious act – this is allowed and made lawful for the mutawallī to
do so in a suitable way, surely God does not waste the rewards of the doers of
good.3

However, after its inclusion in the joint administration of the “Meccan Waqfs” by the
Wakf Commission, the funds were channelled through the same, exclusively Shāfiʿī
networks. From what can be gained from the sources, no attempts were made to seek
out Ibāḍīs in particular.

3 Waqfiyya of Sayyid Ḥammūd b. Aḥmad Āl Bū Saʿīdī, dated 18 Ṣafar 1294/4 March 1877.
The existing document is a copy from 1904, and with an English translation dated 1951.
ZA-HD3/12.
Sources and Bibliography

This study is based on a variety of sources, oral and written, located in state archives
and private collections, English, French, Swahili and Arabic. What follows is an over-
view of the sources, as well as a discussion of how selections were made.

Arabic Textual Material

Given the nature of the inquiries made in this book, the majority of the sources are
textual material in Arabic (manuscript and print), today located in various collections
in East Africa and Cape Town. I have been privileged to have access to two major man-
uscript and book collections in East Africa; The Riyadha Mosque College library in
Lamu and the collection of Muhammad Idris Muhammad Saleh in Zanzibar. In addi-
tion, manuscript silsilas and dhikris were photocopied in Mozambique Island, courtesy
of Hafiz Jammu. In Cape Town, I had access to the material in the Azzawiya Mosque,
courtesy of Shaykh Seraj Hendricks. I have also used material held in the Zanzibar
national archives. These derive from earlier holdings and from the EACROTONAL1 col-
lection that was brought from Dar-es-Salaam to Zanzibar in 2004 and catalogued by
Lorenzo Declich.2
Furthermore, I have drawn on the surveys of Arabic textual material conducted in
the 1990s and 2000s. The first was the World Survey of Islamic Manuscripts, published
in 1993 by al-Furqān foundation. The entries for Kenya and Tanzania gives an overview
of what manuscripts are there, and indicate category according to the classical Arabic
categorization of fiqh, naḥw, taṣawwuf, ṣarf, ḥadīth, falak etc. However, the al-Furqān
publication gives only very limited information concerning dates, titles and authors.
Unfortunately, the World Survey does not even have an entry for Mozambique. The
entries from Madagascar lists exclusively the so-called Sora Be (Great Writing), which –
although certainly interesting from the point of view of writing, and writing traditions –
cannot be said to convey the type of “book-Islam” under scrutiny here. For South Africa,
the focus is on the so-called Jāwī manuscripts and the ones in Arabic-Afrikaans, but
there are also entries of regular Arabic works of theology, fiqh and Sufism among them.

1 Eastern African Centre for Research on Oral Traditions and African National Languages. The
manuscripts collected by the EACROTONAL initiative were listed by Muhammad Burhan
Mkelle, and the handlists have also been used in this study.
2 L. Declich, The Arabic Manuscripts of the Zanzibar National Archives: A Checklist,
Supplemento No 2 Alla rivista degli Studi Orientali, Nouva Serie, Vol. LXXVIII, Pisa/Roma
(Instituti Editoriali e Poligrafica Internazionali), 2006.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004276543_�12

202 sources and bibliography

Another baseline for this work has been the information gathered by R. Seán O’Fahey
for the forthcoming Volume IIIB of the Arabic Literature of Africa series (here referred
to as ALA IIIB), intended to cover Kenya and Tanzania. The surveys entered so far in ALA
IIIB were mainly conducted in the late 1990s and in the period 2000–2005. Although
much of the material referred to in this work will expand the ALA IIIB volume, its
entries nonetheless have provided a basis for the understanding of the diffusion of
Arabic textual material in the region.

The Riyadha Mosque, Lamu, Kenya

The manuscript collection of the Riyadha consists of approximately 150 manuscripts,
presently housed in the library of the educational facility of the mosque. The manu-
scripts date from the 1830s to the 1930s, and is the largest to known to exist in Kenya.
The manuscripts range from 400-page tomes of Islamic law to smaller leaflets of 40–50
pages meant for use in an educational setting. The older ones are bound in leather,
whereas the younger ones are written in lined schoolbooks. The manuscript collec-
tion of the Riyadha mosque is unique from several perspectives. Firstly, it provides an
important overview of the historical orientation of Islamic education in East Africa.
The collection also contains works that so far has been found nowhere else. Several
of the manuscripts have inscriptions that name owners over decades, indicating the
economy of books and reading. Finally, the presence in some of the manuscripts of
inter-linear Swahili translations in the Arabic script opens for research on the use of
the Arabic script before colonial education.
My first visit to the Riyadha was in April 2010. I then returned in July 2010, with col-
leagues from the University of Bergen and from the University of Cape Town. Together
with Aydaroos and Ahmad Badawi of the Riyadha, we made a preliminary inventory of
the manuscript collection, simply listing the manuscript in the order for 1 upwards, list-
ing title, author, copyist and date of copy only. In total, we listed at that time 40 manu-
scripts. Some of them are referred to in this book as “Riyadha MS XX/2010 Survey). At
the same time, we decided, together with Aydaroos and Ahmad Badawi, to apply for
funding from the British Library Endangered Archive Program to digitalize the entire
manuscript collection. This funding was obtained in August 2011, and digitalization
was completed in December 2012.
At this point, it was decided to give the manuscripts numbers that accorded with a
previous listing, completed by Shaykh Ahmad Nabhany for the World Survey of Islamic
Manuscripts in the early 1990s.3 This list was not complete, but listed 75 of the manu-
scripts, and categorized them according to the Islamic system as fiqh, naḥw, taṣawwuf,
ṣarf, ḥadīth, etc. Stickers were made at that time and placed on the manuscripts giving

3 See entry for the Riyadha in: G. Roper (Ed.), World Survey of Islamic Manuscripts, London
(Al-Furqān Islamic Heritage Foundation), Vol II, 1993, 158.
sources and bibliography 203

the number (RM1, RM2 etc.) and the category (fiqh, taṣawwuf etc.) only. Thus, as digita-
lization progressed, the manuscripts that already had an accession number, were given
this number, whereas those that did not were given numbers beyond 75. Consequently,
the majority of the manuscripts have the prefix of the Endangered Archives Program
number, plus their accession number as assigned by Shaykh Nabhany, thus for exam-
ple: EAP466_RM12.
It should be noted that the Riyadha Mosque also has a substantial collection of
printed books dating from the entire twentieth century. Although this collection
would have been an obvious resource especially for Chapter 6 for this study, time and
money did not allow a systematic survey at this point.

The digitized collection is accessible at the EAP website:

http://eap.bl.uk/database/results.a4d?projID=EAP466

The Manuscript and Book Collection of Maalim Muhammad Idris

Muhammad Saleh, Zanzibar
Due to the revolution in 1964, and the exodus of many of the scholarly families from
the islands, much less textual material can be found than one could expect given the
history of Zanzibar as a scholarly centre. For example, the Sumayṭ family left in 1965,
taking with them the entire library of he father and son Aḥmad and ʿUmar b. Sumayṭ,
both of them former chief qāḍīs and prominent Muslim scholars. In other cases, man-
uscripts were simply burnt or done away with by other means.
In terms of Arabic textual material used for this book, the collection consists of
about 40 manuscripts and a large number of printed books that until were 2012 held
in the home of the late Maalim Muhammad Idris in Zanzibar. The manuscript and
books range in age from the 1860s to the present. In the period 2007–2009, 33 manu-
scripts were digitalized under the research programme “Linking Global Cities” at Uni
Global/Chr. Michelsen Institute. These are both Ibāḍī and Shāfiʿī and contain a vari-
ety of material: poetry, mawlids, legal t­ reatises, grammatical works, medicine and one
Quran. The manuscripts come from a mixture of collections, but some of them must
(by internal evidence) have belonged to the twentieth century scholar Burhān Mkelle
(see below). A list was made of the manuscripts that were digitalized, ranging from
“Maalim MS1” to “Maalim MS33” and that is how they are referred to here.
The collection of printed books is too large to present an overview of. Instead, I have
singled out collections that had a known owner, in order to get a sense of what types
of books a scholar would have in his collection. I decided to focus on the collection of
the twentieth century scholar Burhān Mkelle (1884–1949), whose work will be referred
to throughout this book. The reason for this selection is that although the career of
Burhān Mkelle falls within the period of colonial Zanzibar, it also bridges the period
from manuscript to print transmission, and from the clearly Sufi-based reformist
204 sources and bibliography

o­ utlook towards a more modernist-oriented reformist view. His book collection is also
interesting, in that it incorporates books from earlier scholars, thus showing a “recy-
cling” of reading material. For example, several books in the collection had previously
belonged to Saʿīd b. Daḥmān (d. 1925), and then passed on to Burhān, presumably after
the death of Daḥmān.
I have also relied on several unpublished manuscripts that exist in photocopy in the
collection of Maalim Muhammad Idris, as well as newspaper cuts, letters and notes,
journal extracts etc. in Arabic, Swahili and English. Among the more substantial such
works are the works of Burhān Mkelle.
The bulk of the longer manuscripts (texts, treatises) are made available at a sepa-
rate website (kindly contact the author for login):
www.vipidocu.com

Shorter texts and the majority of the textual material deriving from Burhan Mkelle is
available at:
www.swahiliweb.net

The Collection of the Azzawiya Mosque, Cape Town

Through the generosity of Shaykh Seraj Hendricks of the Azzawiya Mosque in
Cape Town, I was allowed to go through the material pertaining to his grandfather,
Muhammad Saleh Hendricks who founded the mosque in 1919. The library collection
consists mostly of printed books, some dating back to the early twentieth century.

The Private Library of Muḥammad b. Aḥmad Āl Bū Saʿīdī, Muscat,

Oman
This collection was the property of the late Muḥammad b. Aḥmad Āl Bū Saʿīdī and is
a private collection accessible to the public. It contains a rich manuscript collection,
mainly of Omani origin but also some manuscripts deriving from Zanzibar. For the
purposes of this work, I consulted mainly the correspondence of the former qāḍī of
Zanzibar, Ṭāhir b. Abī Bakr al-Amawī (d. 1938). He played a central role in the transfer
of money within the scholarly network, discussed in Chapter 9.

Arabic Manuscript Sources (Listed)

BURHĀNIYĀT
Unpublished manuscript in the author’s own hand containing the collected
poetry of Burhān Mkelle. In the collection of the late Muhammad Idris
Muhammad Saleh, Zanzibar. The manuscript is available online at:
www.swahiliweb.net
sources and bibliography 205

MKELLE/ZNZ-1 and MKELLE/ZNZ-2: Burhān Mkelle

The Taʾrīkh by Burhān Mkelle is a central source for the intermediate role of the
Comoros in the spreading of the Shādhiliyya, ʿAlawiyya and the Qādiriyya to
Madagascar and Mozambique. It is also an important biographical source on
individual travelling scholars. Completed in the early 1930s (dated on internal
evidence in the text, such as for example events taking place in the 1920s),4 it
deals at some length with Comorian migration to other lands in the region,
principally Madagascar, Mozambique and – above all – Zanzibar.
This text, written in Arabic, exists in several versions. Two of them have been con-
sulted for this book:

1.
MKELLE/ZNZ-1. A photocopy of an Arabic MS in the collection of Muhammad Idris
Muhammad Saleh in Zanzibar. This copy runs to 67 paginated pages, but with pages
33–42 missing. Most likely, this in Burhān Mkelle’s own hand.5
It is possible and even likely that this photocopy was made from the version known as
EAC-003, based on the index of the EACROTONAL6 collection made by Burhān Mkelle’s
son Muḥammad Burhān.7 The entry in the EACROTONAL list says that the manuscript
runs to 78 pages, while MKELLE/ZNZ-1 runs only to page 67. However, the text ends
abruptly, and – as mentioned – is missing pages, and there is reason to believe that the
version contained additional text that either has been lost in later photocopying or
was not photocopied by whoever did the original copying.8
MKELLE/ZNZ-1 is available online at www.swahiliweb.net

4 For example, Burhān Mkelle gives the death of Shaykh ʿĪsā b. Aḥmad al-Qādirī who died in
Zanzibar in 1925. MKELLE/ZNZ-1, 65.
5 This is based on a comparison with another original MS in the Maalim Idris collection, which
is the collected poems of Burhān Mkelle, known as the Burhāniyāt. This is clearly in the
author’s own hand, and the handwriting is very similar. The judgement is also based on the
comments in the English typescript translation – see below.
6 Eastern Africa Centre for Research on Oral Traditions and African National Languages. This
institution was founded in Zanzibar in 1979 and closed in 1987. Muḥammad Burhān Mkelle
was employed by the EACROTONAL as an expert of Arabic manuscripts.
7 Résumé of old Arabic manuscripts collected in Zanzibar Island by M.B. Mkelle, Volume 1,
EACROTONAL, Zanzibar, 1981. Note that Muḥammad Burhān dates the version to “around
1920” whereas in fact it must be completed at least after 1925.
8 Unfortunately, this cannot be checked against the original, as the EAC-003 listed by
Muḥammad Burhān in 1981 was not part of the EACROTONAL collection when it was listed
by Lorenzo Declich in 2004–2005. The most likely scenario is that it was in fact part of the
EACROTONAL collection but was lost either in the process of transfer to Zanzibar or mis-
placed by the ZNA.
206 sources and bibliography

2.
MKELLE/ZNZ-2. An original Arabic MS in the collection of Muhammad Idris
Muhammad Saleh in Zanzibar. The manuscript is part of a bundle of documents deriv-
ing from Burhān Mkelle, including some of his correspondence, an account book, mis-
cellaneous poetry and general notes to lectures and speeches. The text is in black ink
in a lined exercises book rebound with string clips threaded through two holes. The
paper is broken is some places and has been repaired with tape, which renders the text
partially unreadable.
In this version, Comorian and Malagasy place names are spelled out in Latin let-
ters in brackets, which is not the case in MKELLE/ZNZ-1.9 This copy runs to 56 pagi-
nated pages, and is most likely in the same hand as MKELLE/ZNZ-1, i.e. thought to be
by Burhān Mkelle himself. The introduction found in MKELLE/ZNZ-1 is lacking; i.e. the
invocation of blessings on the work, as well as the motivation for writing the text (“so
our young people can know the works of their forefathers”).
The manuscript bears signs of having been used for another edition. Small notes on
a different type of paper is inserted between the pages, some in Arabic and some in
English. Most likely, this was done by Burhān Mkelle’s son Muḥammad for a partially
completed English version (see below).
On the question of dating, there are several indications that MKELLE/ZNZ-2 is a later,
edited version of MKELLE/ZNZ-1, completed by the author himself. Firstly, MKELLE/
ZNZ-2 contains more information on specific people and events, which it would not
make sense to edit out should the editing process have been the reverse (i.e. from
MKELLE/ZNZ-2 to MKELLE/ZNZ-1). For example, in a biography of the well-known
religious scholar Muhammad al-Moroni, MKELLE/ZNZ-2 adds an anecdote on the
relationship between al-Moroni and ʿAbd Allāh Bā Kathīr, the time they spent together
in the Ḥijāz and the high esteem Bā Kathīr gave to al-Moroni.10 This is not mentioned
at all in MKELLE/ZNZ-1. Given the prestige of Bā Kathīr in the Zanzibar scholarly
environment, it would make little sense, within the format of Islamic hagiography, to
edit this section out. In addition, MKELLE/ZNZ-2 includes textual references to events
taking place later than what can be found in MKELLE/ZNZ-1. For example, MKELLE/
ZNZ-2 notes the death of Shaykh ʿAbd al-Raḥīm b. Maḥmūd in Pemba in March 1936.11

9 Note that this is not consistent throughout the text. In some cases, the “difficult” names
are spelled out in Arabic letters in brackets and with vocalization. Either, Mkelle could
not make up his mind which system was the most clarifying, or he simply (consciously or
unconsciously) abandoned consistency.
10 MKELLE/ZNZ-2, 51.
11 MKELLE/ZNZ-2, 22. The sentence noting the death of this shaykh is added in a different
pen, but what appears to be the same handwriting. It is the latest date to be found in any
of the two manuscript versions consulted here.
sources and bibliography 207

The conclusion is thus that this manuscript copy was completed by the author himself
some time after March 1936.
A section of MKELLE/ZNZ-2 (the section on Comorian language) was published by
A. Aboubakar in 1983.12
Reference should also be made to a photocopy of an Arabic manuscript origi-
nally in the possession of B.G. Martin, bearing the author’s autograph, enti-
tled Taʾrīkh Jazīratinā Qamar al-Kubrā.13 In his work Muslim Brotherhoods in
19th century Africa,14 B.G. Martin referred to the manuscript with page reference, for
example when it comes to the spread of the Shādhiliyya to Mozambique at the hands
of Shaykh Mrūzī. It is clear that these references correspond exactly to MKELLE/ZNZ-
2. Without having seen the actual photocopy, is therefore very likely that the manu-
script referred to by B.G. Martin is (or is identical to) MKELLE/ZNZ-2.
Muḥammad Burhān Mkelle, in addition to his work on the EACROTONAL cata-
logue, also made a list of books that he had in his possession some time in the 1980s.15
On this list features we also find the Taʾrīkh Jazīrat al-Qamar al-Kubrā, this in a copy
running to 56 pages.16 There is every reason to believe that this copy is identical to
MKELLE/ZNZ-2.
It is therefore reasonable to conclude that there were two copies of the History
of Grande Comore: One that was listed by Muḥammed Burhān Mkelle as part of the
EACROTONAL collection and one that was listed by Muḥammad Burhān Mkelle in
his own list, and which at some point in time passed to Maalim Muhammad Idris in
Zanzibar.
MKELLE/ZNZ-2 is available online at www.swahiliweb.net

12 Aziza Aboubakar, “Le manuscript de Burhan Mkelle et la langue comorienne”, Études

Océan Indien, II, 1983, 117–122. In the article, a section of the Arabic text is reproduced,
and although the reproduction is not very clear, it seems to be identical to MKELLE/
ZNZ-2. Unfortunately, Aboubakar does not say where she viewed the copy.
13 B.G. Martin, Muslim Brotherhoods in 19th century Africa, London (Cambridge University
Press), 1976. This title is given in the bibliography (page 239), and noted as “xerox of
author’s manuscript in my possession”, but without any further information. Note the
wording “jazīratinā” instead of “jazīra”. It may be that MKELLE/ZNZ-2, by the time Martin
viewed it, actually had a front page giving this title.
14 B.G. Martin, Muslim Brotherhoods, 156 and notes.
15 I am grateful to Ridder S. Samsom for proving me with a copy of this handlist, and to his
friend, in turn, who provided the first copy.
16 Entry 15 in “List of some manuscripts written by some of the ʿulamāʾ of Zanzibar”
compiled by Muḥammad Burhān Mkelle. It is possible that this list was compiled before
the EACROTONAL list, as a introductory work.
208 sources and bibliography

In addition, at least three other versions of this text are known to exist:

a)
A fragment manuscript version (4 pages) of the same text photographed by Drs. Ridder
Samsom, Hamburg University, in the collection of Muḥammad Burhān Mkelle. This
is clearly another copy, and is most likely in the same hand as the two copies outlined
above. It is found among other fragments of text that clearly derive from Burhan
Mkelle. The fragment is dated, in hijra and CE date (17 Ṣafar 1349/14 July 1930) and starts
out from the same point as MKELLE/ZNZ-2, with the headline “The name of the island
and its geographical location”. It bears marks of annotations and corrections. The most
likely interpretation is that this is one of the earlier and incomplete drafts of the text.

b)
An Arabic manuscript used by Djemal Eddine Gaba for French a translation published
in 1981.17 This MS was, according to Gaba, obtained in Grande Comore in 1977. As stated
by Gaba, the MS runs to 33 pages. However, it is clear both from the translation, the
various sections of the document, as well as the amount of detail provided under each
topic, that the version consulted by Gaba is different from both MKELLE/ZNZ-1 and
MKELLE/ZNZ-2.

c)
There exists also in the collection of Maalim Muhammad Idris in Zanzibar a partial18
English translation, in the form of an unpublished typescript. This was, in all likeli-
hood, done by Burhān’s son Muḥammad. The typescript has some notes, which makes
it possible to deduce that it was based on two Arabic texts, one referred to as “ms”
and one as “writer’s MS”. However, it is clear that the actual text corresponds exactly
to MKELLE/ZNZ-2, in terms of word-by-word content. This, in turn, strengthens the
hypothesis that MKELLE/ZNZ-2 remained with the family and eventually passed to
Maalim Muhammad Idris while MKELLE/ZNZ-1 became part of the EACROTONAL
Collection.

Considering MKELLE/ZNZ-2 as the latest, and most conclusive version, it is the preferred
reference throughout this book. However, references are made to both MKELLE/ZNZ-1
and MKELLE/ZNZ-2, depending on which version has the information referred to.

17 Djemal Eddine Gaba, “Le Manuscrit de Burhan Mkelle”.

18 The translation is complete up to the section entitled “The (racial) origins of the
Comorians”.
sources and bibliography 209

ZANZIBAR/QADIRI 1
Photocopy of first page of silsila and ijāza given to ʿUmar Qullatayn. Arabic, 17 lines.
In the collection of the late Maalim Muhammad Idris. Photocopied by A.K. Bang,
February 2008. Copy in Bergen.

ZANZIBAR/QADIRI 2
Scroll, 77 lines, Qādirī silsila, Arabic, undated, signed by Shaykh Uways. In the col-
lection of the late Maalim Muhammad Idris Muhammad Saleh, Zanzibar. The scroll
is badly torn, and parts of the text impossible to read. Photographed by A.K. Bang,
October 2011. Copy in Bergen.

QADIRI SILSILA/PALMA
Scroll, 232 lines, Qadiri silsila and text, Arabic,. Photographed 1988 in Palma by Ryszard
Czajkowski, and copy given to me by Prof. E. Rzewuski, University of Warsaw. Copy in
Bergen.

QADIRI SILSILA/ILHA.
Scroll, 74 lines, Qādirī ijāza and silsila issued by Shaykh ʿĪsā b. Aḥmad. Photographed
by Dr. Elke Stockreiter in January 2007. Copy in Bergen.

QADIRI QASIDA/ILHA 1
Notebook, 83 pages handwritten (more than one hand), Arabic, some entries dates
1940s. Contains a collection of Qādirī poetry and prayers, some of it composed by
Shaykh Uways.
Photographed by A.K. Bang in Mozambique Island, March 2011, courtesy of Hafiz
Jammu. Copy in Bergen.

QADIRI QASIDA/ILHA 2
Notebook, 135 pages handwritten (more than one hand), Arabic, dated 1940s. Contains
a collection of Qādirī poetry, songs (prefixed by the instruction: inshād) and prayers, in
praise of Qādirī shaykhs, and particularly ʿAbd al-Qādir al-Jīlānī.
Photographed by A.K. Bang in Mozambique Island, March 2011, courtesy of Hafiz
Jammu. Copy in Bergen.

SHADHILI/ILHA1.
Scroll, 43 lines, dated 12 Dhū ’l-Ḥijja or Dhū ‘l-Qaʿda 1324/Dec 1906/Jan 1907, Arabic.
Photographed Mozambique Island, 2007 by Dr. Elke Stockreiter. Copy in Bergen.
210 sources and bibliography

SHADHILI/ILHA 2.
Scroll, 59 lines, undated, Arabic. Photographed Mozambique Island, 2007 by Dr. Elke
Stockreiter. Copy in Bergen.
By internal evidence (the person being initiated, Ṣāliḥ b. ʿUbūd, is 3 links removed from
Muḥammad Ghulām), this document is more recent than SHADHILI/ILHA 1. This is
also indicated by the appearance of the paper and ink.

SHADHILI/ILHA3.
Scroll, undated, 48 lines, Arabic. Photographed by Chapane Mutiua, Eduardo Mondlane
University. Copy in Bergen.

SHADHILI/ILHA4.
Scroll, undated, 46 lines, Arabic; Photographed by Chapane Mutiua, Eduardo Mondlane
University. Copy in Bergen.

ABU AL-HASAN/ZANZIBAR1.
Manuscript, 3 pages, Arabic, undated. By internal evidence by Abū ’l-Ḥasan Jamal al-
Layl. In the collection of the late Maalim Idris Muhammad Saleh. Photographed by
A.K. Bang, Zanzibar, July 2010. Copy in Bergen

QADIRI DHIKR/ZANZIBAR1
Collection of Qādirī dhikr (hādhā tartīb al-dhikr). Manuscript, Arabic, bound in
leather, 182 pages. Some pages are missing as the pages have loosened from the leather
binding. The text is in one (unknown) hand, except for the last 13 pages. The first pages
contain the instruction on passing on the Qādirī dhikr to the murīdūn. Then follows
instructions on dhikr and when to perform them. The volume also contains the sayings
of al-Jīlānī and Shaykh Uways, Qādirī poetry and dhikr, as well as poetry by Shaykh
Uways.
In the collection of the late Maalim Idris Muhammad Saleh.
Photographed by A.K. Bang, Zanzibar, October 2011. Copy in Bergen.

MORONI/1.
Notebook, handwritten in several hands, 51 pages, Arabic. Containing notes and poetry
by Aḥmad and ʿUmar b. Sumayṭ, in family possession, Moroni, Grande Comore. Copied
by A.K. Bang, Moroni, June 1997. Copy in Bergen.

QADHIS DIARY
This is the diary of the Brawanese-Zanzibari scholar ʿAbd al-ʿAzīz b. ʿAbd al-Ghanī
al-Amawī (1838–1896). The original is in the collection of the late Muhammad
Idris Muhammad Saleh in Zanzibar, and is unpaginated and in no particular order.
sources and bibliography 211

A ­paginated photocopy exists, prepared by Muḥammad Burhān Mkelle, probably some

time in the 1980s. This photocopy is not complete, nor consecutive, as pages are clearly
missing. For two readings of this very rich source, see Mkelle (1992)19 and Hoffmann
(2006).20 Copy in Bergen.

Archival Sources

South Africa National Archives, Cape Town

South Africa News 1903–1905
South African National Archives, Cape Town: KAB-IRC series: Immigration papers

Zanzibar National Archives

ZA-HD10: Waqf files

Oral Sources/Interviews

Chapter 6 (on the ṭarīqa Shādhiliyya in Northern Madagascar) draws on a series of

interviews with Shaykh Yusuf Abud, presently Imām of the Domoni mosque in Diego
Suarez and Shaykh Abdu Samash, presently shaykh of the Shādhilī mosque in Diego
Suarez in the period 20 July 2008 to 30 July 2008. Furthermore, it is based on one inter-
view conducted with Shaykh Abduh Samad b. Ahmad Khalid, resident of Ambilobe,
for the past decade the khālifa of the ṭarīqa Shādhiliyya in Northern Madagascar, on
27.07.08. It is also based on one interview with Mr. Cassam Aly, representative of the
Shādhiliyya in Diego Suarez and on conversations with Mr. Isa Abduh Husayn, leader
of the Islamic Association of Diego Suarez. Furthermore, the chapter is based on one
interview and a series of conversations with members of the Ismail family in Diego
Suarez, presently proprietors of the Jacob Mosque (also known as the Sunni-Hanafi
mosque). Finally, it is based on conversations with Mr. Abd al-Rasul, former student of
ʿAbd al-Raḥmān al-Saqqāf, and presently a member of the ṭarīqa Shādhiliyya in Diego
Suarez. All interviews were conducted in French, but in some cases in a combination
of French, Swahili and Arabic.
Chapter 5 draws on repeated interviews and conversations in the period 2006–
2010 with Shaykh Seraj Hendricks of the Azzawiya Mosque in Cape Town. Shaykh
Hendricks is the grandson of Muhammad Salih Hendricks, and presently the co-leader

19 Mkelle, Muhammad Burhan, “A Scholar for all Seasons. Shaykh Abd al-Aziz al-Amani”
(sic: for al-Amawi), Journal of the Institute of Muslim Minority Affairs, 13:1, 1992, 116–121.
20 V. Hoffmann, In His (Arab) Majesty’s Service.
212 sources and bibliography

of the Azzawiya mosque, in addition to teaching in the Religious Studies Department

in the University of Johannesburg, a member of the Muslim Judicial Council of South
Africa and a public intellectual. When quoted specifically, dates of formal interviews
are given as a reference.

Interviews in chronological order

– Maalim Muhammad Idris Muhammad Saleh. Repeated conversions in the period

2005–2012.
– N.B. al-Saqqaf, Cape Town, South Africa, February 2005 and August 2005.
– Adil al-Attas and Uthman b. Abd Allah, Cape Town, South Africa, 14.08.05.
– Ahmad b. Abi Bakr b. Umar Qullatayn, grandson of Umar Qullatayn, Zanzibar,
21.07.2006.
– Ismail Hasan Jacob, 21.07.2008. Also: Conversations with members of the Jacob
family, present day proprietors of the Jacob mosque in Diego Suarez, Madagascar,
20.07.2008 and 01.08.2008.
– Bakar Momad, Diego Suarez, 26.07.2008
– Isa Aboudou Hussen, Islamic Association of Diego Suarez, Diego Suarez,
26.07.2008.
– Abduh Samad , Khalīfa of the ṭarīqa Shādhiliyya, Ambilobe, Madagascar,
27.07.2008.
– Abd al-Rasul, leader of the ṭarīqa Shādhiliyya, Antsirananana, Madagascar,
27.07.2008.
– Shaykh Yusuf Abud, Imām of the Juma mosque, Antsiranana, Madagascar,
01.08.2008.
– Mwalimu Husayn Soud al-Maawy, Lamu, Kenya, 29.04.10.
– Saleh Muhammad Badawi Jamal al-Layl, College of Islamic Studies, Lamu, Kenya,
01.05.10.
– Muhdhar Khitamy, Mombasa, Kenya, 03.05, 2010.
– Seraj Hendricks, Azzawia Mosque, Cape Town, South Africa, 10.03.05 and
07.12.2010, and repeated conversations between 2005 and 2011.
– Aydaroos and Ahmad Badawi Jamal al-Layl, Lamu, Kenya, December 2011 and
December 2012.

Bibliography

Unpublished Papers
ʿAlawī, Ṭāhir Muḥammad, Tarjama ʿUmar b. Sumayṭ, Typescript, Arabic, 6 pages, ND, in
family possession, Moroni, Grande Comore.
sources and bibliography 213

Aly, Cassam, “Islam dans le Nord de Madagascar”. Unpublished paper, ND.

Davids, A., Slaves, Sheikhs, Sultans and Saints. The Kramats of the Western Cape,
Unpublished MS, unpaginated, in the possession of Shaykh Seraj Hendricks.

Published Works
Abdulkadir, Abdulkadir Hashim, Reforming and Retreating: British Policies on trans-
forming the Administration of Islamic Law and its Institutions in the Busaʿidi
Sultanate, 1890–1963, PhD Thesis, University of the Western Cape, 2010.
Ahmed, C., Religion et Politique aux Comores. Évolution de l’autorite spirituelle depuis le
protectorat français (1886) jusqu’à nos jours, PhD Thesis, École des Hautes Études en
Sciences Sociales, Paris, 1996.
———, Islam et Politique aux Comores. Évolution de l’autorité spirituelle depuis le pro-
tectorat français (1886) jusqu’à nos jours, Paris (L’Harmattan), 2000.
———, Ngoma et Mission Islamique (Da’wa) aux Comores et en Afrique Orientale, Paris
(L’Harmattan), 2002.
———, “Networks of the Shādhiliyya Yashruṭiyya Sufi Order in East Africa”, in:
R. Loimeier and R. Seesemann, The Global Worlds of the Swahili. Interfaces of Islam,
Identity and Space in nineteenth and twentieth Century East Africa, Hamburg (Lit
Verlag), 2006.
Ahmed, C.C., God, Anti-Colonialism and Dance. Sheekh Uways and the Uwaysiyya, in :
Gregory Maddox (ed.), Conquest and Resistance to Colonialism in Africa. New York
(Garland Publishing), 1993, 145–167.
Alpers, E.A., “A complex relationship. Mozambique and the Comoro Islands in the
nineteenth and twentieth century”, Cahiers d’études africaines, 161, 2001.
———, “East Central Africa”, in: N. Levtzion and R. Pouwels, The History of Islam in
Africa, Oxford (J. Currey), 2000, 303–326.
———, “Indian Ocean Africa: The Island Factor”, in: E.A. Alpers, East Africa and the
Indian Ocean, Princeton (Marcus Wiener), 2009, 39–54.
Ameir Issa, A., “The legacy of Qadiri scholars in Zanzibar”, in: R. Loimeier and
R. Seesemann (eds.), The Global Worlds of the Swahili, 343–62.
Anderson, J.N.D., Islamic Law in Africa, London (Frank Cass), 1970 (2nd ed. 1st. ed., 1955)
Asad, T., “The Idea of an Anthropology of Islam”, Occasional Paper Series, Georgetown
University (Center for Contemporary Arab Studies), 1986.
Babavatan, H., Understanding Afrikā-yi Osmani in the late Ottoman period. The case of
Zanzibar, BA thesis, Bogazici University, Turkey, 2000.
Bāb Ṣayl, Muḥammad Said, Al-Qawl al-majdī fī ’l-radd ʿalā ʿAbd Allāh b. ʿAbd al-Raḥmān
al-Sindī, litograph, Batavia/Jakarta, 1309/1891–92.
al-Badawi, M., Sufi Sage of Arabia. Imām Abdallah ibn Alawi al-Haddad, Louisville
(Fons Vitae Spiritual Masters Series), 2005.
214 sources and bibliography

al-Badawi, M., Introduction to English translation of ʿAbd Allāh b. ʿAlawī al-Ḥaddād,

The Book of Assistance, Fons Vitae Spiritual Masters Series, 2005.
Badawī, Ṣāliḥ Muḥammad ʿAlī (Shaykh Bā Ḥasan), Al-Riyāḍ bayna māḍīhi wa-ḥādirihi,
Typescript, NP, 1410/1989.
Bakar, Abdourahim Saïd, “Small Island Systems: A case study of the Comoro Islands”,
Comparative Education, 24,2, 1988, 181–191.
Bā Kathīr al-Kindī, ʿAbd Allāh b. Muḥammad, Riḥlat al-Ashwāq al-Qawiyya ilā Mawāṭin
al-Sāda al-ʿAlawiyya, Reprint, Cairo, 1405/1984 (first ed. Cairo, 1934).
Bang, A.K., Sufis and Scholars of the Sea. Family networks in East Africa, c. 1860–1925,
London (RoutledgeCurzon), 2003.
———, “My generation. Umar b. Ahmad b. Sumayt (1886–1973): Inter-generational
Network transmission in a trans-oceanic Hadrami Alawi family, ca. 1925–1973”, in
L. Manger and M. Assal, Diasporas within and without Africa – Dynamism, hetereoge-
neity, variation, Nordiska Afrika Instituttet, Uppsala, 2006, 87–103.
———, “Hadrami teachers, scholars and educationalists. The role of Hadramis in
Islamic education in Zanzibar, c. 1870–1930,” Asian Journal of Social Science, 2/3,
2007, 435–456.
———, “Zanzibar-omanere og Oman-zanzibarere. Bindestreker over det Indiske
Havet – og videre”, Babylon, 5:2,2007, 98–104.
———, “When there are no foreign lands and all lands are foreign”, in: N. Naguib and
B. de Vries (eds.), Heureux qui comme Ulysses a fait un beau voyage. Movements of
People in Time and Space, Bergen (Bric), 2010, 151–165.
———, “Authority and Piety, Writing and Print: A preliminary Study of the Circulation
of Islamic Texts in late nineteenth and early twentieth Century Zanzibar”, Africa,
81:1, 2011, 89–107.
———, “Zanzibari Islamic knowledge transmission revisited: loss, lament, legacy –
and transformation”, Social Dynamics, 38:3, 2012, 419–434.
Bangstad, S., “Diasporic Consciousness as a Strategic Resource: A Case Study from a
Cape Muslim Community,” in: L. Manger and M. Azzal, Diasporas within and with-
out Africa – Dynamism, hetereogeneity, variation, Uppsala, 2006, 32–60.
F. Barth, Sohar. Culture and Society in an Omani town, Baltimore (Johns Hopkins
University), 1983.
Becker, C.H., “Materials for the Understanding of Islam in German East Africa,” edited
and translated by B.G. Martin, Tanzania Notes and Records, 68, 1968, 31–61.
Becker, F., “Commoners in the process of Islamization: reassessing their role in the light
of evidence from southeastern Tanzania”, Journal of Global History, 3, 2008, 227–249.
———, Becoming Muslim in Mainland Tanzania, 1890–2000, Oxford (Oxford University
Press), 2008.
Bennet, N., “Zanzibar, Portugal and Mozambique: Relations from the late eighteenth
century to 1890,” Working Papers in African Studies, Boston University, 1987.
sources and bibliography 215

van den Berg, L.W.C., Le guide des zélés croyants. Manuel de jurisprudence musulmane
selon le rite de Châfi’î. Vol I and II, Batavia Government Print, 1882, 1884.
Bissell, W.C., Urban Design and Colonial Power in Zanzibar, Bloomington (Indiana
University Press), 2011.
Blanchy, S., “Images Coloniale de la Société Comorienne. Les Raisons d’une mécon-
naisance durable”, in: Images coloniales de sociétés de l’Océan indien, Paris (Presses
Universitaires Paris-Sorbonne), 2009.
Blanchy, S., Maisons des femmes, cités des hommes : Filiation, âge et pouvoir à Ngazidja
(Comores), Nanterre (Société d’ethnologie), 2010.
Bonate, L.J.K., “The Ascendance of Angoche. The Politics of Kinship and Territory in
Nineteenth Century Northern Mozambique”, Lusotopie, 2003, 115–140.
———, “Matriliny, Islam and Gender in Northern Mozambique,” Journal of Religion in
Africa, 36:2, 2006, 139–166.
———, Traditions and Transitions. Islam and Chiefship in Northern Mozambique, c.
1850–1974. PhD Thesis, University of Cape Town, 2007.
———, “Roots of diversity in Mozambican Islam,” Lusotopie, XIV (1), 2007, 129–149.
———, “Muslim Religious Leadership in Post-Colonial Mozambique”, South African
Historical Journal, 60 (4) 2008, 637–654.
———, “The use of Arabic script in Northern Mozambique”, Tydskrift vir Letterkunde,
45:1, 2008, 120–129.
———, “Islam in Northern Mozambique: A Historical Overview” History Compass, 8,
2010.
———, “Documents in Arabic Script at the Mozambique Historical Archives”, Islamic
Africa, 1:2, 2010.
Bonate, L.J.K. and Mutiua, C., “Duas Cartas de Farallahi”, Estudos Mocambicanos, 22:1,
2011, 91–106.
Bose, S., A Hundred Horizons. The Indian Ocean in the age of Global Empire, Cambridge,
Mass. (Harvard University Press), 2009.
Boulinier, G., “Le Rôle de Said Mohamed el-Maarouf dans le Developpement de la
Confrerie Shadhili”, ND (?), Bulletin CENDERES, Moroni.
Bourde, A., “The Comoro Islands: Problems of a Microcosm”, Journal of Modern African
Studies, 3,1, 1965, 91–102.
Bourgoin, F., “From Network to Class? Towards a More Complex Conception of
Connection and Sociability”, Working Paper of Danish Institute of International
Studies. Copenhagen: Danish Institute of International Studies, 31:2009.
Brennan, J., Taifa. Making Nation and Race in Urban Tanzania, Athens, Ohio (Ohio
University Press), 2012.
van Bruinessen, M., “The Origins and Development of the Naqshbandi Order in
Indonesia”, Der Islam, 67, 1990, 150–179.
216 sources and bibliography

de Brye, E., “L’enfant comorien et l’apprentissage de l’Islam. Quelques observations”,

Études Océan Indien, 6, 1985, 41–51.
Caplan, A.P., Choice and Constraint in a Swahili Community. Property, Hierarchy and
Cognatic Descent on the East African Coast, London (Oxford University Press), 1975.
Chouzour, S., “Histoire et Sociologie de Ngazidja. Le Manuscrit de Said Hussein.
Présentation, traduction et notes”, Études Océan Indien, 1:1982, 15–54.
Cohen, R., Global Diasporas: An Introduction, London (UCL Press), 1997.
———, Migration and its enemies. Global Capital, migrant labour and the nation state,
Aldershot (Ashgate), 2006.
Cornell, V.J., The Way of Abū Madyan. Doctrinal and Poetic Works of Abū Madyan Shuʿayb
ibn al-Ḥusayn al-Anṣārī (c. 509/1115–594–1198), Cambridge (Islamic Texts Society),
1996.
Daḥlān, Aḥmad Zaynī, Al-Durr al-Saniyya fī ’l-radd al-Wahhābiyya, 4th ed., Cairo, 1980.
Dallal, A., “The Origins and Objectives of Islamic Revivalist Throught, 1750–1850”,
Journal of the American Oriental Society, 113:3, 1993, 341–359.
Davids, A., The Mosques of Bo-Kaap, Cape Town (The South African Institute of Arabic
and Islamic Research), 1980.
da Costa, Y., Islam in Greater Cape Town. A Study in the Geography of Religion, PhD
Thesis, University of South Africa, 1989.
——— and Davids, A., Pages from Cape Muslim History, Pietermaritzburg (Shooter and
Shuter), 1994.
Decary, R., L’Ile de Nosy Be de Madagascar. Histoire d’une colonization, Paris (Edition
Maritimes et d’outre-mer), 1960.
Dechlich, L., The Arabic Manuscripts of the Zanzibar National Archives: A Checklist,
Supplemento No. 2 Alla rivista degli Studi Orientali, Nouva Serie, Vol. LXXVIII, Pisa/
Roma (Instituti Editoriali e Poligrafica Internazionali), 2006.
Eickelmann, D. F. and Piscatori, J., (eds.), Muslim Travellers. Pilgrimage, migration, and
the religious imagination, London, 1990.
Fair, L., Pastimes and Politics. Culture, community and identity in post-abolition urban
Zanzibar 1890–1945, Oxford (J. Currey), 2001.
Farsy, A.S., Baadhi ya Wanavyoni wa Kishafii wa Mashariki ya Afrika/The Shafiʿi Ulama
of East Africa, ca. 1830–1970. A hagiographical Account. Translated, edited and anno-
tated by R.L. Pouwels, University of Wisconsin, African Primary Text Series, III, 1989.
Feener, M., “Hybridity and the “Hadrami Diaspora” in the Indian Ocean Muslim
Networks,” Asian Journal of Social Science, 32:3, 2004, 353–372.
Ferrand, G., Les Musulmans a Madagascar et aux Iles Comores. Vol 3: Antankarana,
Sakalava, Migration Arabes, Paris, 1902.
Freitag, U., “Hadramaut: A religious Centre for the Indian Ocean in the late nineteenth
and Early twentieth centuries?”, Studia Islamica, 89, 1999, 165–183.
sources and bibliography 217

———, “Der Orientalist und der Mufti: Kulturkontakt im Mekka des 19. Jahrhunderts”,
Die Welt des Islams, 43, 1, 2003, 37–60.
———, Indian Ocean Migrants and State Formation in Hadhramaut, Leiden (Brill),
2003.
——— and van Oppen, A., Translocality. The Study of Globalising Processes from a
Southern Perspective, Leiden (Brill), 2010.
Fuji, C., ‘ “Tarīqas” without Silsilas: The Case of Zanzibar’, Kyoto Bulletin of Islamic Area
Studies, 2–1 (2008), 23–34.
Gaba, Djemal Eddine, “Le Manuscrit de Burhan Mkelle sur la Grande Comore”, Asie du
Sud-Est et Monde Insulindien: Islam et Littératures dans l’Archipel Comores, Vol. XII:
3–4, 1981, 43–80.
Gamieldien, F., The History of the Claremont Main Road Mosque. Its People and their
Contribution to Islam in South Africa, Published by the Claremont Main Road
Mosque, 2004.
Geertz, C., The Religion of Java, Chicago (Chicago Univ. Press), 1960.
Ghazal, A., Islamic Reform and Arab Nationalism. Expanding the Crescent from the
Mediterranean to the Indian Ocean (1880s–1930s), London (Routledge), 2010.
Gilbert, E., “Oman and Zanzibar. Historical roots of a Global Community”, in: Ray, H.P.
and Alpers, E.A. (eds.) Cross Currents and Community Networks. The History of the
Indian Ocean, Oxford (Oxford Univ. Press), 2006.
Glassmann, J., Feasts and Riots, Revelry, Rebellion and Popular Consciousness on the
Swahili Coast, 1856–1888, Portsmouth (Heinemann), 1995.
Glassman, J., War of Words, War of Stones. Racial Thought and Violence in Colonial
Zanzibar, Bloomington (Indiana University Press), 2011.
Glazer, S., “The Alfiyya commentaries of Ibn ʿAqīl and Abū Ḥayyān”, The Muslim World,
31/4, 1941, 400–408.
Gonia, J.W., The Umma Ideal and Muslims in Madagascar. Movement from Community
to Community, Master of Theology Thesis, Luther Theological Seminary, St.Paul,
Minnesota, 1997.
Green, N., Bombay Islam. The Religious Economy of the Western Indian Ocean, 1840–1915,
Cambridge (Cambridge University Press), 2011.
———, Sufism. A Global History, Wiley-Blackwell, 2012.
Le Guennec-Coppens, F., “Social and Cultural Integration: A Case Study of the East
African Hadramis”, Africa, 59 (2), 1989, 185–195.
Gueunier, N.J., Les Chemins de l’Islam à Madgascar, Paris (L’Harmattan), 1994.
Gupta and Ferguson, Anthropological locations: Boundaries and Grounds of a Field
Science, Univ. California Press, 1997.
Guy, P. and Abdourahmane bin Cheih Amin, La Vie et l’Oeuvre du Grand Marabout des
Comores, Said Mohamed bin Ahmed Al-Ma’arouf, Tananarive, 1949.
218 sources and bibliography

al-Ḥaddād, ʿAbd Allāh b. ʿAlawī, al-Daʿwa al-tāmma wa’l-tadhkira al-ʿāmma, word press
online edition, 1461/2000: http://seekerofthesacred.wordpress.com/arabic-pdf-
books/
Hafkin, N.J., Trade, Society and Politics in Northern Mozambique c. 1753–1913, PhD
Dissertation, Boston University, 1973.
Haj, S., Reconfiguring Islamic Tradition. Reform, Rationality and Modernity, San
Francisco (Stanford University Press), 2011.
Haron, M., “Islamic Dynamism in South Africa’s Western Cape”, Journal of Muslim
Minority Affairs, 9:2, 1998, 366–372.
———, “Da’wah movements and Sufi tarīqahs: Competing for spiritual spaces in con-
temporary south(ern) Africa”, Journal of Muslim Minority Affairs, 25:2, 261–285.
Harries, L., (ed.), The Swahili Chronicle of Ngazija by Said Bakari bin Sultani Ahmed,
Bloomington (African Studies Program), 1977.
Hefner, R.W., “The Political Economy of Islamic Conversion in Moderen East Java”, in:
William R. Roff (ed.), Islam and the Political Economy of Meaning, London/Los
Angeles, 1987, 53–78.
Hendricks, S., Taṣawwuf (Ṣūfism): Its role and impact on the Culture of Cape Islam, MA
Thesis, University of South Africa, 2005.
Ho, E., The Graves of Tarim. Genealogy and Mobility across the Indian Ocean, Berkeley
(University of California Press), 2006.
Hoffman, V., “In his (Arab) Majesty’s Service: The Career of a Somali Scholar and
Diplomat in Nineteenth-Century Zanzibar”, in Loimeier, R. and Seesemann (eds.),
The Global Worlds, 251–273.
Hoffmann, V., “Annihilation in the Messenger of God: The Development of a Sufi
Practice”, Int. J. of Middle East Studies, 31:3, 1999, 351–369.
———, “The Role of the Masharifu on the Swahili Coast in the Nineteenth and
Twentieth Centuries”, in: K. Morimoto (ed.), Sayyids and Sharifs in Muslim Societies,
London (Routledge), 2012, 185–197.
Jamal al-Layl, Al-Ḥabīb Abū ’l-Ḥasan b. Aḥmad, Diwān al-Fatḥ wa’l-Imdād, printed in
Dubai, ND.
Jeppie, S., “Leadership and Loyalties: The Imāms of Nineteenth Century Colonial Cape
Town, South Africa”, J. of Religion in Africa, XXVI, 2, 1996, 139–162.
———, “Re-classifications: Coloured, Malay, Muslim”, in: Z. Erasmus (ed.), Coloured by
History, Shaped by Place. New Perspectives on Coloured Identities in Cape Town, Cape
Town (Kwela), 2001.
Kaḥḥāla, Umar Riḍā, Muʿjam al-muʾallifīn. Tarājim muṣannifī ’l-kutub al-ʿarabiyya,
Beirut, 1376/1957.
al-Junayd, ʿAbd al-Qādir b. ʿAbd al-Raḥmān b. ʿUmar, al-ʿUqūd al-Jāhiza wa’l-wuʿūd
al-nājiza, Printed in India, ND (2009?).
sources and bibliography 219

Kent, R., Early Kingdoms of Madagascar, 1500–1700, New York, 1970.

Kitamy, BinSumeit, “The role of the Riyadha mosque college in enhancing Islamic
identity in Kenya”, in: M. Bakari and S.S. Yahya (eds.), Islam in Kenya, Nairobi
(MEWA), 1995, 269–276.
Knysh, A., Ibn ‘Arabi in the later Islamic Tradition. Making of a Polemical Image in
Medieval Islam, New York (SUNY Series in Islam), 1999.
Kresse, K., Philosophising in Mombasa. Knowledge, Islam and Intellectual Practice on the
Swahili Coast, Edinburgh (Edinburgh University Press), 2007.
Lambek, M., “Choking on the Quran. And other consuming parables from the western
Indian Ocean Front”, in: W. James (ed.), The Pursuit of Certainty. Religious and
Cultural Formulations, London (Routledge), 1995.
——— and Walsh, A., “The Imagined Community of the Antankarana: Identity,
History and Ritual in Northern Madagascar”, J. of Religion in Africa, 27, 3 (Aug. 1997),
308–333.
———, The Weight of the Past. Living with History in Mahjanga, Madagascar, Palgrave,
NY, 2002.
Larson, P., Ocean of Letters. Language and Creolization in an Indian Ocean Diaspora,
Cambridge (Cambridge University Press), 2009.
Lienhardt, P., “The Mosque College of Lamu and its Social Background”, Tanganyika
Notes and Records, 1959, 228–242.
Loimeier R., and Reichmut, S., “Zur Dynamik religiös-politischer Netzwerke in muslim-
ischen Gesellschaften”, Die Welt des Islams, 36/2, 1996, 145–185.
——— and Seesemann, R., The Global Worlds of the Swahili. Interfaces of Islam, Identity
and Space in nineteenth and twentieth Century East Africa, Hamburg (Lit Verlag),
2003.
———, “Patterns and Peculiarities of Islamic Reform in Africa”, Journal of Religion in
Africa, 33:3, 2003, 237–262.
———, “Is there something like “Protestant Islam”?”, Die Welt des Islams, 45:2, 2005,
216–254.
———, “Zum sozialen Kontext eines religiösen Rituals”, Der Islam, 83, 2006, 170–186.
———, “Memories of Revolution. Zur Deutungsgeschichte einer Revolution (Sansibar
1964)”, Africa Spectrum, 41:2, 2006, 175–197.
———, Between Social Skills and Marketable Skills. The Politics of Islamic Education in
twentieth century Zanzibar, Leiden (Brill), 2009.
Lupi, E.C., Angoche. Breve Memoria Sobre uma das capitanias-móres no Districto de
Mozambique, Lisboa, 1907. http://ia600404.us.archive.org/3/items/angochebreve
mem00lupigoog/angochebrevemem00lupigoog.pdf
Mandaville, P., Transnational Muslim Politics. Re-imagining the Umma, London
(Routledge), 2001.
220 sources and bibliography

Martin, B.G., “Muslim Politics and Resistance to Colonial Rule: Shaykh Uways b.
Muhammad al-Barawi and the Qadiri Brotherhood in East Africa”, Journal of African
History, Vol 10:3, 1969, 471–486.
———, “Notes on some members of the learned class of Zanzibar and East Africa in
the Nineteenth Century”, African Historical Studies, 4:3, 1971.
———, “Migrations from Hadramawt to East Africa and Indonesia, c. 1200–1900”,
Research Bulletin, Centre for Arabic Documentation, Ibadan, 7, III, 1971, 1–21.
———, “Arab Migrations to East Africa in Medieval Times,” Int. Journal of African
Historical Studies, VII, 3, 1975, 367–390.
———, Muslim Brotherhoods in nineteenth Century Africa, Cambridge (Cambridge
University Press), 1976.
———, “A Palestinian Arab and Writer on the Comoros, 1903”, Études Océan Indien, 6,
1985, 71–124, 102.
Al-Mashhūr, ʿAbd al-Raḥmān b. Muḥammad b. Ḥusayn, Shams al-ẓahīra fī nasab ahl
al-bayt min banī ʿAlawī. Furūʿ Fāṭima al-Zahrāʾ wa-Amīr al-Muʾminīn ʿAlī, 2 vols.,
2nd ed., edited by Muḥammad Ḍiyāʾ Shihāb, Jiddah (ʿĀlam al-Maʿrifa), 1984.
al-Mashhūr, Abū Bakr al-ʿAdanī, Lawāmiʿ al-nūr. Nubdha min aʿlām Ḥaḍramawt, 2 vols.,
Ṣanʿāʾ (Dār al-Muhājir), 1412/1991–92.
Matthews, N., “Imagining Arab Communities: Colonialism, Islamic Reform and Arab
Identity in Mombasa Kenya, 1897–1933”, Islamic Africa, 2, 2013.
Mitchell, J.C., “The concept and use of social networks,” in: Mitchell, J.C. (ed.) Social
Networks in Urban Situations, Manchester, 1989.
McIntosh, J., The edge of Islam: power, personhood and ethnoreligious boundaries on the
Kenya coast, Durham (Duke Univ. Press), 2009.
Mkelle, Muhammad Burhan, “A Scholar for all Seasons. Shaykh Abd al-Aziz al-Amani”,
Journal of the Institute of Muslim Minority Affairs, 13:1, 1992, 116–121.
Mohamed, G. (ed.), A Translation and transliteration of Athkaar, Duaas and Salawaat,
including the Ratibul Haddad, Cape Town (Digital Action), 2010.
al-Mughayrī, Saʿīd b. ʿAlī, Juhaynat al-Akhbār fī taʾrīkh Zinjibār, 4th ed., Ministry of
National Heritage and Culture, Oman, 1994.
Muḥammad b. ʿAbd al-Raḥman b. Shihāb, “Al-Sādat al-ʿAlawiyya fī jazāʾir al-Qamar wa-
Madaghaskar”, al-Naḥda al-Ḥaḍramiyya, Singapore, 10/1933, 24–25.
Munthe, L., La tradition Arabico-Malgache vue à travers le manuscript A-6 de Oslo et
d’autres manuscript disponibles, Antananarivo, 1982.
Naude, J.A., “A Historical Survey of Opposition to Sufism in South Africa”, in: F. de Jong
and B. Radtke, Islamic Mysticism Contested. Thirteen Centuries of Controversies and
Polemics, Leiden (Brill), 1999.
Newitt, M., “The Early History of the Sultanate of Angoche,” Journal of African History,
XIII:3, 1972, 397–406.
sources and bibliography 221

———, A History of Mozambique, London (Hurst), 1995.

Nimtz, A.H., Islam and Politics in East Africa. The Sufi Order in Tanzania, Minneapolis
(University of Minnesota Press), 1980.
Nuṣayr, ʿAyīda Ibrāhīm, Al-kutub al-ʿArabiyya allatī nushirat fī Misr fī ’l-Qarn al-tāsiʿ
ʿashar, Cairo, 1990.
Oberauer, N., Waqf im kolonialen Sansibar. Der Wandel eier islamischen Stiftungspraxis
under britischer Protektoratherrschaft, Würzburg (Ergon-Verlag), 2012.
O’Fahey, R.S., and Radtke, B., “Neo-sufism reconsidered”, Der Islam, 1, 1993, 52–87.
——— Et. Al., The Arabic Literature of Africa, Vol. IIIB (forthcoming).
Padwick, C.E., Muslim Devotions. A Study of Prayer Manuals in Common Use, Oxford
(OneWorld) 1996 (1st. ed. 1961).
Pedersen, J., Den Arabiske Bog, Copenhagen (Fischer), 1946.
Penrad, J-C., “La Shâdhiliyya-Yashrûtiyya en Afrique orientale et dans l’océan Indien
occidentale,” in: Éric Geoffroy, Une voie soufie dans le monde. La Shāadhiliyya, Paris
(Maisonneuve and Larose), 2005.
Pouwels, R.L., Horn and Crescent: Cultural Change and Traditional Islam on the East
African Coast 800–1900, New York (Cambridge University Press), 1987.
Prestholdt, J., Domesticating the World. African Consumerism and Genealogies of
Globalization, Berkely (University of California Press), 2008.
Purpura, A., Knowledge and Agency: The Social Relations of Islamic Expertise in Zanzibar
Town, PhD Dissertation, City University of New York, 1997.
Radtke, B., “Sufism in the 18th century: An attempt at a Provisional Appraisal”, Die Welt
des Islams, 36:3, 1996, 326–364.
Reese, S. (ed.). The transmission of learning in Islamic Africa. Leiden (Brill), 2004.
———, Renewers of the Age. Holy Men and Social Discourse in Colonial Benaadir,
Leiden (Brill), 2008.
———, “The death of Shaykh Uways of Somalia”, in: J. Renard (ed.), Tales of God’s
Friends. Islamic Hagiography in translation, Berkely (Univ. California Press), 2009.
———, “Salafi transformations. Aden and the changing voices of religious reform in
the interwar Indian Ocean”, Int. Journal of Middle East Studies, 44, 2012, 71–92.
Renard, J., (ed.), Tales of God’s Friends. Islamic Hagiography in Translation, Berkely
(University of California Press, 2009).
Robinson, D., Paths of Accommodation. Muslim Societies and French Colonial Authorities
in Senegal and Mauretania, 1880–1920, Oxford (James Currey), 2000.
Roff, W.R. (ed.), Islam and the Political Economy of Meaning, London (Croom Helm),
1987.
Romero, P., Lamu. History, Society and Family in an East African Port City, Princeton
(Marcus Wiener), 1997.
Roper, G. (ed.), World Survey of Islamic Manuscripts, London (Al-Furqān Islamic
Heritage Foundation), Vol. II, 1993.
222 sources and bibliography

Rzewuski, E., “Origins of the Tungi Sultanate (Northern Mozambique) in the light of
local traditions”, in: S. Pilaszewicz and E. Rzewuski (eds.), Unwritten Testimonies of
the African Past. Proceedings of the International Symposium held in Ojzan, Warsaw
7–8 November 1989, University of Warsaw (Orientalia Varsovensia), 1991.
Sadgrove, P., “From Wādī Mīzāb to Unguja: Zanzibar’s scholarly links. In: Reese, S. (ed.).
The transmission of learning in Islamic Africa, 184–211.
Said, I., Chamanga, M.A., Gueunier, N.J., “Un Qasida Arabe en l’honneur de Sayyid
Ahmad al-Kabir, missionaire de l’Islam a Madagascar de 1896 a 1919”, Études Océan
Indien, Vol.1, 1982, 137–148.
Said, M., “Le Périple D’un Comorien de Zanzibar a Mascate et Retour,” Études Océan
Indien, 10, 1988, 103–122.
Saleh, I., A Short History of the Comorians in Zanzibar, Dar es Salaam (Tanganyika
Standard), 1936.
Schacht, J. “Notes on Islam in East Africa”, Studia Islamica, 23:1965, 91–136.
———, “Sayyid Aḥmad Zaynī Daḥlān”, Encyclopedia of Islam; 2, 1960.
O. Scharbrodt, O., “The Salafiyya and Sufism: Muhammad ʿAbduh and his Risalat al-
Waridat (Treatise on Mystical Inspirations)”, Bulletin of the School of Oriental and
African Studies, 70:1, 2007, 89–115.
Schulze, R., “The birth of tradition and modernity in eighteenth and nineteenth cen-
tury Islamic culture: the case of printing” in J. Skovgaard-Pedersen (ed.), Culture and
History: The introduction of the printing press in the Middle East, Culture and History,
16, 1997, 29–72.
———, Islamischer Interntionalismus im 20. Jahrhundert: Untersuchungen zur
Geschichte der Islamischen Weltlige, Leiden (Brill), 1990.
Sedgwick, M., Saints and Sons. The making and remaking of the Rashīdī Sufi order, 1799–
2000, Leiden (Brill), 2005.
———, Muhammad Abduh, Oxford (OneWorld), 2010.
Sharkey, H., “Aḥmad Zaynī Daḥlān’s al-Futūḥāt al-Islāmiyya: A Contemporary view of
the Sudanese Mahdī”, Sudanic Africa, 5, 1994, 67–75.
Sheriff, A., Dhow Cultures of the Indian Ocean. Cosmpolitanism, Commerce and Islam,
London (Hurst), 2009.
bin Shihāb, Muḥammad b. ʿAbd al-Raḥman, “Al-Sādat al-ʿAlawiyya fī jazāʾir al-Qamar
wa-Madaghaskar”, al-Naḥda al-Ḥaḍramiyya, Singapore, 10/1933, 24–25.
Skovgaard-Pedersen, J., Defining Islam for the Egyptian State. Muftis and Fatwas of the
Dār al-Iftā, Leiden (Brill), 1997.
Snouck-Hurgronje, C., “Een rector der Mekkaansche universiteit”, Bijdragen tot de
Taal-, Land- en Volkenkunde van Nederlandsch-Indië, 36:3, 1887, 344–404.
Snouck Hurgronje, C., Mekka in the latter half of the nineteenth century, Leiden (Brill),
1931.
sources and bibliography 223

ʿUmar b. Aḥmad b. Sumayṭ, Al-Nafḥat al-Shadhdhiyya min al-Diyār al-Ḥaḍramiyya

wa-ṭalbiyyat al-ṣawt min al-Ḥijāz wa-Ḥaḍramawt, Privately printed, Tarim/Aden (?)
1955. Second, revised and annotated edition: Al-Nafḥat al-Shadhdhiyya ilā al-Diyār
al-Ḥaḍramiyya wa-ṭalbiyyat al-ṣawt min al-Ḥijāz wa-Ḥaḍramawt, Privately printed,
Jiddah, 1988. Note the change in the title from min to ilā al-Diyār al-Ḥaḍramiyya.
Reference in this work is made to the 1988 version unless otherwise noted.
Tayob, A.K., Islamic Resurgence in South Africa, Cape Town (University of Cape Town
Press), 1995.
———, Islam in South Africa. Mosques, Imams and Sermons, University Press of
Florida, 1999.
———, “Muslims and Globalization in post-Apartheid South Africa”, paper presented
to the AEGIS European Conference on African Studies conference, London, 2005.
Toibibou Ali Mohamed, La Transmission de l’Islam aux Comores (1933–2000), Paris
(L’Harmattan), 2008.
———, Ahmad Qamardine (1895–1974). Un intellectuel Comorien et ses réseaux, PhD
Thesis, University of Paris Diderot, 2010.
Trimingham, J.S., Islam in East Africa, Oxford (Oxford University Press), 1964.
ibn ʿUmar, ʿAbd al-Raḥmān, al-Jawhar al-Nafīs fī khawāṣṣ al-Shaykh Uways, Cairo
(Maṭbaʿat al-Mashhad al-Ḥusaynī), 1964.
de Vere Allen, J., Al-Inkishafi. Catechism of a Soul, Oxford (Oxford University Press), 1972.
Verin, P., The History of Civilization in North Madagascar, Rotterdam (A.A. Balemka)
1986.
———, “Le Rapport Pechmarty sur les “ecoles koraniques” a la Grande Comore (1917)”,
Études Océan Indien, 16, 1993, 105–123.
Versteegh, K., “Arabic in Madagascar”, Bulletin of the School of Oriental and African
Studies, Vol. 64, 2, 2001, 177–187.
Vertovec, S., and Cohen, R., (eds.), Migration, Diasporas and Transnationalism,
Cheltenham (Elgar), 1999.
Vikør, K.S., Sufi and Scholar on the Desert Edge. Muḥammad b. ʿAlī al-Sanūsī and his
Brotherhood, London (Hurst), 1995.
Voll, J., “Neo-Sufism reconsidered again”, Canadian Journal of African Studies, 2/3, 2008,
314–330.
Walker, I., “From Moroni to Mukalla: Hadhramis on the Island of Ngazidja (Comoros)
and in the Hadhramout”, Journal des Africanistes, 72:2, 2002, 111–121.
———, “Identity and citizenship among the Comorians of Zanzibar”, in: A. Sheriff and
E. Ho, The Largest Continuum. Oceanic Movements and formation of new societies in
the Indian Ocean, forthcoming, London (Hurst), 2013.
Walsh, A., Constructing “Antankarana”. History, Ritual and Identity in Northern
Madagascar, PhD Dissertation, University of Toronto, 1998.
224 sources and bibliography

———, “What Makes (the) Antankarana Antankarana? Reckoning Group Indentity in

Madagascar”, Ethnos, 66:1, 2001, 1–22.
El-Zein, A.H.M., The Sacred Meadows. A Structural Analysis of Religious Symbolism in
an East African Town, Evanston, 1974.
Al-Ziriklī, Khayr al-Dīn, al-Aʿlām: Qāmūs tarājim li-ashhar al-rijāl wa ’l-nisāʾ min al-ʿarab
wa ’l-mustaʿaribīn wa ’l-mustashriqīn, 6th. ed., Beirut, 1984.
Index

ʿAbd Allāh Qāḍī ʿAbd al-Salām (Tuan Guru) as a place of publication 114–115,
91 131 n. 68, 133–136, 139
ʿAbduh, Muḥammad 12 n. 25, 13, 15, 131, 150, Comoro Islands 2, 16, 22, 25, 34, 37–42, 44,
184, 185 46, 54, 67, 70, 74, 76, 82, 88, 130, 148, 197
Aḥmad al-Kabīr (Aḥmad b. Mwinyi Mkuu
Aḥmad al-Saqqāf) 43, 79–81, 83–85 Daḥlān, Aḥmad Zaynī 27–28, 29, 32, 33, 34,
Ahmad Qamar al-Dīn 189 36, 112, 134, 149, 162, 167, 168, 177, 182, 183
al-Ājurrūmiyya 112, 113 n. 15 Diego Suarez (Antsiranana) 1, 2, 11, 16, 39,
al-Alfiyya 113–114 73–74, 75 n. 13, 76, 80, 83, 85–89, 143, 165,
al-Amawī, ʿAbd al-ʿAzīz 32, 34–35, 50, 165, 194, 211–212
170, 180
al-Amawī, Ṭāhir b. Abī Bakr 163, 164, 165, Farsy, Abdallah Saleh 35, 37, 57, 124, 131,
169–170, 174, 176, 177, 180, 182, 204 159, 188
Amur b. Jimba 67–69, 191 al-Faqīh al-Muqaddam, Muḥammad b. ʿAlī Bā
Antankarana (people) 72–74, 76–84, 88, 197 ʿAlawī 94–95, 147
al-ʿAydarūs, Muḥsin b. Sālim 103–104
Azzawiya Mosque, Cape Town (founding) Grande Comore (Ingazija) 20, 22, 36, 38–41,
100–101, 204 42, 43, 44–45, 52, 53, 55, 56 n. 36, 57, 59, 64,
67, 75–77, 85 nn. 46–47, 86, 92, 112, 183, 189,
Bāb Ṣayl, Muḥammad Saʿīd 28, 32, 33, 34, 208, 210, 212
36, 118, 168, 174, 177
Bā Junayd, ʿUmar b. Abī Bakr 32, 94, 96, 99, al-Ḥaddād, ʿAbd Allāh b. al-ʿAlawi 10, 24,
167–170, 179 30–31, 95, 117 n. 35, 120, 124, 128, 131, 133, 134
Bā Kathīr, ʿAbd Allāh 33–34, 57, 63, 66 n. 55, n. 78, 143, 144–146, 147, 194
94, 96, 99–100, 106, 118, 119, 129, 136, 157, Ḥaḍramawt 1, 2, 16, 21, 22, 25, 33, 36, 40,
167, 169, 180, 183–184, 187–188, 206 42, 46, 50, 56, 76, 77, 85 n. 46, 88, 94, 96,
Bā Kathīr, Abū Bakr b. ʿAbd Allāh 167, 169, 102, 103, 104, 112, 118, 120, 128, 129, 130, 131,
188 134, 136, 139, 144, 148, 149, 150, 152, 157,
Barghash b. Saʿīd, Sultan of Zanzibar 32 n. 161, 163, 168, 175, 178, 180, 183, 187, 189,
34, 33–35, 39, 50, 52, 112, 130, 199 196, 197
al-Barwānī, ʿAlī b. Khamīs 32 as a scholarly centre 29–32
al-Barwānī, ʿĪsā b. ʿAlī 167, 188 al-Ḥalabī, Muṣṭafā Bābī, Publishers 133, 134,
Brawa 34, 44 n. 44, 50, 170 135
“Brawanese network” 48–50, 165, 170, Ḥaramayn 2, 16, 25, 26–29, 32, 33, 46, 88, 95,
177, 180 164, 167, 174–176, 178, 179, 197
al-Barāwī, Uways b. Muḥammad 34–35, 48, al-Ḥātimī, Muḥammad b. Aḥmad 175–177,
49, 50, 51, 53, 56, 58, 59, 60, 62, 64, 71 n. 65, 180
125, 182, 193, 209, 210 al-Ḥātimī mosque, Zanzibar 181–182
Hendricks, Muhammad Salih 93–101,
Cape Town 1, 2, 4, 11, 16, 96, 98–107, 143, 144, 103–107, 153, 167, 211
163, 167, 197, 201, 204 al-Ḥibshī, Aḥmad b. Zayn 30, 31, 117, 131, 138,
Muslim community in 90–93 145 n. 5, 156
Ratib al-Haddad in 152–153, 154–156, 160 al-Ḥibshī, ʿAlī b. Muḥammad 31, 36, 42, 95,
Cairo 13, 15, 25, 164, 185, 194 128, 129, 182
226 index

Ihyāʾ ʿUlūm al-Dīn 100, 132 Mkelle, Burhān 33, 38, 39, 43, 51, 53, 55, 56
Ilha de Mozambique 16, 47, 54–57, 64, 65, n. 35, 57, 63, 67, 71, 75, 115, 132, 133, 136, 137,
67–69, 81, 83, 88, 191, 193, 194, 197 138, 142, 164, 189, 191
History of Grande Comore by 205–208
Jamal al-Layl, ʿAbd Allāh b. Aḥmad (“Master Mmakonde, Said b. Abdallah Lindy 57
of birds”) 20, 21, 22, 35, 41, 45, 76 al-Moroni, Muḥammad b. Aḥmad b. Ḥasan
Jamal al-Layl, Abū ’l-Ḥasan b. Aḥmad 32–33, 93, 206
(Mwinyi Bahasani) 45, 46, 55, 183 Mtsujini, ʿĪsā b. Aḥmad (“al-Injazījī”) 56–59,
Jamal al-Layl, Abū ’l-Ḥasan b. Aḥmad 64, 66, 205 n. 4
63–64, 65, 70, 76, 87 n. 55, 99, 137, 139, al-Muhājir, Aḥmad b. ʿĪsā 30
188–190, 210 Muḥammad “Mkelle” b. Adam 51–53, 58,
Jamal al-Layl, Ḥasan b. Muḥammad b. 59
Ḥasan 117–119, 145, 183 Muhammad b. Thābit 60–62
Jamal al-Layl, Ṣāliḥ b. ʿAlawī (Habib Saleh) Muḥammadan light 9, 23–24
35–36, 45, 115, 119, 121, 129 n. 62, 150, 182, Musa Muhammad Sahib Quanto 54, 56
183, 194 Mwinyi Mkuu Sulṭān Aḥmad 41, 43, 79, 183
Jamʿiyyat al-Sunna 63, 65, 188–190 n. 32, 188 n. 47

Khalīfa b. Saʿīd, Sultan of Zanzibar 188 Nationalism 14–16, 195–196

al-Khishtī, Saʿīd b. Muḥammad 137 Neo-Sufism 8–10
Nosy Be 20, 39, 73–74, 75 n. 13, 76–78, 80
Madrasa Bā Kathīr 34, 85 n. 46, 100, 124, 133,
137, 151, 161, 167, 183, 188–189, 191 Pan-Islamism 13, 26
al-Makhlūf, Ḥasanayn b. Muḥammad
149–150, 162 Qādiriyya 1, 8, 17, 34–35, 47–53, 56, 58–61,
Manṣūr b. ʿAbd Allāh al-ʿAlawī 102–103 63–66, 70–71, 112, 116, 125, 132, 142, 146, 165,
al-Mashhūr, ʿAlawī b. ʿAbd al-Raḥmān 32, 174, 195, 197, 205
139 as organization 181–182
al-Mashhūr, ʿAbd al-Raḥmān b. Muḥammad branches of the Qādiriyya 48–53, 56–63,
31, 118–119, 182 64–66
al-Maʿrūf, Muḥammad 42–43, 67–69, 78–79, in Zanzibar 34–35, 48–53, 116
197 in Ilha de Mozambique 54–56, 64–66,
Maṭbaʿat al-Mīriyya 131, 133–134 70–71, 191, 194, 197
Mawlid al-Ḥibshī 24, 36 al-Qaḥṭanī, Muhyī al-Dīn 32, 50, 116, 135,
Mawlid al-Barzanjī 24, 53, 58 n. 40, 100, 175–177, 180
182, 193 Qullatayn, ʿUmar b. Qullatayn al-Naẓīrī 34,
textual diffusion of 125, 127, 130 35, 48, 49–50, 59, 62, 64–65, 180, 181, 209
al-Mazrui al-Amin b. Ali 154, 159, 184–185
al-Mazrūʿī, ʿAlī b. ʿAbd Allāh b. Nāfiʿ 32 Rātib al-Ḥaddād 2, 17, 94, 100, 104, 106, 128,
Mecca 2, 3, 12, 25, 31–33, 36, 43, 50, 91, 92–94, 156, 157–162, 164, 192, 194
96, 98, 99, 102–103, 106, 112, 113, 115, 118, 122, as text 146–148, 153–156
149, 182, 183, 187, 195, 196, 197, 199–200 transmission of 148–150
as a place of learning 26–29 in Lamu 150–151
printing presses 131–134, 139 in Zanzibar 151–152
waqf distribution to 164–175, 176, In Cape Town 152–153
177–181, 188 Riḍāʾ, Rashīd 13, 24
Minhāj al-Ṭālibīn 45, 121–122, 132, 135, 138 Al-Risālat al-Jāmiʿa 31, 94, 100, 138, 139, 156
Mjana Kheri, Abd Allah 50–51, 59–61, 66, Riyadha Mosque (Lamu) 35–37, 108 n. 1,
70, 195, 197 150–151, 160, 161, 190, 191
index 227

as book collection 111, 113–116, 118–121, Sumayṭ, ʿUmar b. Aḥmad b. 9 n. 16, 15, 44,
123, 130, 134–137, 153, 201, 202–203 85–88, 99, 151, 165, 167, 175, 178, 179, 180,
as organization 182–184 189, 203
Sumayṭ, Aḥmad b. ʿUmar 30–31, 95, 101
Saidina b. Umar 77, 78
Salafism 13–14, 162, 190, 196 ʿUthmān b. ʿAbd al-Laṭīf 72, 77–79, 82
Sayyid Manṣab b. ʿAbd al-Raḥmān (Abū Bakr
b. ʿAbd al-Raḥmān) 33, 34, 36, 135, 182, Wadʿān, ʿAbd Allāh 189
183 Wird al-Laṭīf 124, 156
Sayyid Manṣab b. ʿAlī (Aḥmad b. ʿAlī Ittibārī)
79 n. 26, 183 n. 32, 188 Zanzibar 1, 2, 4, 5, 11, 13, 14, 16, 25, 37, 38, 39,
al-Saqqāf, ʿAbd al-Raḥmān 80–82, 83, 85, 42, 46, 57, 59, 70, 71, 73, 75, 77, 86, 87, 92,
211 99, 100, 102, 105, 116, 117, 119, 125, 130, 139,
Shādhiliyya 1, 8, 27, 34, 42–44, 46 n. 87, 48, 144, 148, 151–152, 153, 162, 164, 181, 183, 185,
85, 86, 88, 89, 94, 125, 132, 142, 146, 184, 191, 191, 192, 193, 195, 197
194, 195, 196, 197, 206 as printing centre 31, 137–139
in Mozambique 67–71 as scholarly centre 32–35, 43, 44–45,
in Madagascar 72, 77–84 48–56, 58, 60–63, 66, 67, 74, 79, 85 n. 46,
Shams al-Ẓahīra 31, 119, 140 88, 96
Sharḥ al-ʿAyniyya 117, 118, 131 n. 68, 145 n. 5 organizations in 65, 185–190, 194–195
Sumayṭ, Aḥmad b. Abī Bakr b. 9 n. 16, 15, 31, text circulation in 130–133, 134, 135,
33, 34, 45, 63, 66 n. 55, 94, 95, 96, 99, 129, 136–137, 156, 193–194, 203–204
133–136, 138, 141, 145, 150, 165, 169, 174, 177, waqf and waqf distribution 164, 167,
180, 183, 184, 187, 203 168–175, 176, 177–179, 180, 199–200