Shaikhdoms of Eastern Arabia Peter Lienhardt Edited by Ahmed Al-Shahi
St Antony’s Series General Editor: Richard Clogg...
324 downloads
1209 Views
3MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Shaikhdoms of Eastern Arabia Peter Lienhardt Edited by Ahmed Al-Shahi
St Antony’s Series General Editor: Richard Clogg (1999– ), Fellow of St Antony’s College, Oxford Recent titles include: Louise Haagh CITIZENSHIP, LABOUR MARKETS AND DEMOCRATIZATION Chile and the Modern Sequence Renato Colistete LABOUR RELATIONS AND INDUSTRIAL PERFORMANCE IN BRAZIL Greater São Paulo, 1945–1960 Peter Lienhardt (edited by Ahmed Al-Shahi) SHAIKHDOMS OF EASTERN ARABIA John Crabtree and Laurence Whitehead (editors) TOWARDS DEMOCRATIC VIABILITY The Bolivian Experience Steve Tsang (editor) JUDICIAL INDEPENDENCE AND THE RULE OF LAW IN HONG KONG Karen Jochelson THE COLOUR OF DISEASE Syphilis and Racism in South Africa, 1880–1950 Julio Crespo MacLennan SPAIN AND THE PROCESS OF EUROPEAN INTEGRATION, 1957–85 Enrique Cárdenas, José Antonio Ocampo and Rosemary Thorp (editors) AN ECONOMIC HISTORY OF TWENTIETH-CENTURY LATIN AMERICA Volume 1: The Export Age Volume 2: Latin America in the 1930s Volume 3: Industrialization and the State in Latin America Jennifer G. Mathers THE RUSSIAN NUCLEAR SHIELD FROM STALIN TO YELTSIN Marta Dyczok THE GRAND ALLIANCE AND UKRAINIAN REFUGEES Mark Brzezinski THE STRUGGLE FOR CONSTITUTIONALISM IN POLAND Suke Wolton LORD HAILEY, THE COLONIAL OFFICE AND THE POLITICS OF RACE AND EMPIRE IN THE SECOND WORLD WAR The Loss of White Prestige Junko Tomaru THE POSTWAR RAPPROCHEMENT OF MALAYA AND JAPAN, 1945–61 The Roles of Britain and Japan in South-East Asia
Eiichi Motono CONFLICT AND COOPERATION IN SINO-BRITISH BUSINESS, 1860–1911 The Impact of the Pro-British Commercial Network in Shanghai Nikolas K. Gvosdev IMPERIAL POLICIES AND PERSPECTIVES TOWARDS GEORGIA, 1760–1819 Bernardo Kosacoff CORPORATE STRATEGIES UNDER STRUCTURAL ADJUSTMENT IN ARGENTINA Responses by Industrial Firms to a New Set of Uncertainties Ray Takeyh THE ORIGINS OF THE EISENHOWER DOCTRINE The US, Britain and Nasser’s Egypt, 1953–57 Derek Hopwood (editor) ARAB NATION, ARAB NATIONALISM Judith Clifton THE POLITICS OF TELECOMMUNICATIONS IN MEXICO Privatization and State–Labour Relations, 1928–95 Cécile Laborde PLURALIST THOUGHT AND THE STATE IN BRITAIN AND FRANCE, 1900–25 Craig Brandist and Galin Tihanov (editors) MATERIALIZING BAKHTIN C. S. Nicholls THE HISTORY OF ST ANTONY’S COLLEGE, OXFORD, 1950–2000 Anthony Kirk-Greene BRITAIN’S IMPERIAL ADMINISTRATORS, 1858–1966 Laila Parsons THE DRUZE BETWEEN PALESTINE AND ISRAEL, 1947–49
St Antony’s Series Series Standing Order ISBN 0–333–71109–2 (outside North America only) You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England
Shaikhdoms of Eastern Arabia Peter Lienhardt
Edited by Ahmed Al-Shahi St Antony’s College Oxford
in association with St Antony’s College, Oxford
© Ahmed Al-Shahi 2001 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2001 by PALGRAVE Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N. Y. 10010 Companies and representatives throughout the world PALGRAVE is the new global academic imprint of St. Martin’s Press LLC Scholarly and Reference Division and Palgrave Publishers Ltd (formerly Macmillan Press Ltd). ISBN 0–333–94811–4 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Lienhardt, Peter. Shaikhdoms of eastern Arabia / Peter Lienhardt ; edited by Ahmed Al-Shahi. p. cm. — (St. Antony’s) Revision of the author’s thesis (Ph.D.)—Oxford University, 1957. Includes bibliographical references and index. ISBN 0–333–94811–4 1. United Arab Emirates—Social conditions—20th century. 2. United Arab Emirates—Politics and government—20th century. I. Al-Shahi, Ahmed. II. Title. III. St. Antony’s/Macmillan series (St. Martin’s Press) HN666.A8 L54 2000 306'.095357—dc21 00–066558 10 10
9 09
8 08
7 07
6 06
5 05
4 04
3 03
Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire
2 02
1 01
Contents List of Illustrations
vi
Acknowledgements
vii
Editor’s Preface
viii
1 The Political Complex Foreign protection and the extent of local autonomy Instability and structure The complex of the desert and the sea
1 1 16 24
2 Women and Men
33
3 The Bedouin
80
4 Towns and Maritime Activities Fishing Merchant seafaring Pearl fishing
114 125 136 150
5 The Shaikhly Families
165
6 The Shaikhs and Their People
194
Editor’s Epilogue
228
Appendix
231
Notes
233
Select Bibliography
241
Index
249
v
List of Illustrations Map
The Shaikhdoms of Eastern Arabia.
Fig. i
Peter Lienhardt and Shaikh Shakhbut bin Sultan of Abu Dhabi in 1961. Bedouin mother and her child. Bedouin. Bedouin at a well. Simplest (binary) model of segmentation. Model of segmentation with proliferation of groups among bedouin. A dhow. The growth of ruling families as illustrated in the descendants of Sultan bin Saqr. The genealogy of the ruling family of Abu Dhabi. Part of the Al Bu Falah genealogy showing the descendants of Khalifa and his two sons. Female links between two ruling families of Abu Dhabi and Dubai. A shaikh in his majlis with his guards. The formation of factions in the choice of a ruler. Rivalry within the ruling family of Kuwait.
Fig. Fig. Fig. Fig. Fig.
2.1 3.1 3.2 3.3 3.4
Fig. 4.1 Fig. 5.1 Fig. 5.2 Fig. 5.3 Fig. 5.4 Fig. 6.1 Fig. 6.2 Fig. 6.3
vi
xx
xii 36 82 92 101 101 126 166 176 178 179 198 214 220
Acknowledgements In the ‘Acknowledgements’ of his D.Phil thesis (‘Shaikhdoms of Eastern Arabia’) the late Peter Lienhardt (1928–86) thanked the Treasury Committee for Studentships in Foreign Languages and Cultures for the senior studentship which enabled him to undertake his anthropological research in the Trucial States. He also thanked ‘many members of HM Foreign Services and members of the British Communities in the Eastern Arabian Shaikhdoms both for their help and their hospitality’. Finally, he was ‘grateful . . . to the rulers of the various shaikhdoms in which [he] stayed, to members of the ruling families, and to many private individuals from among the people of the shaikhdoms for their kindness and co-operation without which [his] research could not have been carried out’. Peter Lienhardt received help from St Antony’s College, Oxford University, while he was a Fellow and latterly a College Lecturer. I am sure that Peter would want me to thank those friends and colleagues who have read and made comments on some earlier drafts of the chapters of this book. Their contributions are belatedly rewarded in the publication of this important study of the Arab shaikhdoms of the Gulf. It was customary for the brothers to exchange their writings for comments and refinements, and I am sure Peter would have liked his late brother, Godfrey, to be thanked for his comments. My thanks to Mrs Clare Brown for her help over deciphering some of Peter’s handwritten sentences, to Dr Eugene Rogan, the Director of the Middle East Centre, St Antony’s College, for his comments and suggestions and to the Middle East Centre for its contribution towards the publication of the book. Finally, I am grateful to Dr Alison H. Black for her meticulous editorial advice, to Dr Maike Bohn for her interest and suggestions and to my wife, Anne, for her help and support. The present volume is evidence of Peter Lienhardt’s intellectual ability, scholarly contributions and sensitivity to the workings of a different culture. He was kind, sympathetic and accessible to those who sought his help and advice. This book is a fitting tribute to his memory. AHMED AL-SHAHI Oxford
vii
Editor’s Preface
The present anthropological study of the political and social institutions of the shaikhdoms of the Trucial States has been long overdue. Peter Lienhardt1 intended to publish his anthropological study (conducted between November 1953 and February 1956) but he did not leave a complete manuscript in the form that appears in this book. Rather, he left chapters and notes, some typed and others handwritten, and no plan as to how they should finally appear in a finished form. As usual, Peter wrote his material in paragraphs, each beginning on a separate sheet of paper. There are numerous additions and reminders in the margin of the chapters which suggest that he was intending to revise the material for publication, but for various reasons he did not complete his final revision. As Peter’s Literary Executor, his former student and close friend, I have undertaken the task of assembling the available material and preparing it for publication with editorial notes. I have no intention to undertake any editorial changes either to the text of the manuscript or to the layout of each chapter. It is imperative that I should maintain the integrity of Peter’s text. The only necessary exceptions to this policy are to his ‘Introduction’ and his notes for a conclusion. Both are unfinished pieces, and while I have included in my Preface some extracts from the ‘Introduction’, I give in my Epilogue the full text of the points Peter earmarked for the conclusion. I do not think it would be fruitful to speculate on how he might have ended the ‘Introduction’ or how he might have written the conclusion or what further data he might have incorporated in both pieces. It is a wellknown practice with Peter that he used to revise his writings constantly, and hence the difficulty he found in completing any piece of his work. Originally the ‘Shaikhdoms of Eastern Arabia’ was successfully submitted as a D.Phil thesis at Oxford University in 1957. But since then and until his death in 1986, Peter reworked some chapters of the thesis, and the result of this revision is the present volume. He went to undertake research on these shaikhdoms not unprepared. Firstly, he studied Arabic and Persian at Cambridge University, and secondly, his anthropological interest in the Arabs began with earlier research for his B.Litt thesis (The Northern Arabs: an account of the social and political organization of some nomad and settled communities of Northern Arabia and greater viii
Editor’s Preface
ix
Syria), which was completed also at Oxford in 1953. The research was based on published sources, and some themes from this study were taken further in his fieldwork on the nomadic and settled people of the shaikhdoms of the Gulf. Equipped with language and research abilities, Peter went first to Kuwait and then to other shaikhdoms of the Gulf. His extensive travel and long stay helped in shaping his perceptive understanding of the people of the area and their complex culture. Moreover, he became acquainted with the ruling families of the Trucial States, a situation which enabled him to make further important observations on the politics of these families. Among his powerful acquaintances was Shaikh Shakhbut who, on the basis of Peter’s valuable experience, appointed him to be his adviser. Thus Peter went to Abu Dhabi in March 1961, with the hope of rendering useful advice to Shakhbut on future development. Of this episode in his career Peter wrote in his unfinished ‘Introduction’: Presumably on grounds of personal friendship, the then Ruler of Abu Dhabi, Shaikh Shakhbut bin Sultan [1904–66], invited me to come to act as his adviser or, as we agreed it was better to call me, Secretary to the Government. Unfortunately – perhaps for him as well as for me – the Ruler and I soon found ourselves in considerable disagreement over what should be done and I left after only three months. This experience is . . . one which has added a good deal to the general content of this book, since through it I came closer to the inner workings of a family of ruling shaikhs than an anthropologist engaged in field research can normally expect to do. Moreover, it brought me, however briefly, into contact with the problems of modern development and dealings with foreign employees and companies, some of the things that anthropologists used occasionally to blame themselves or each other for neglecting in favour of the traditional and the primitive. In an appraisal of the role of Shaikh Shakhbut in running his shaikhdom, Peter writes: Many readers will remember that Shaikh Shakhbut’s unwillingness to spend the increasing sums of money that he received from Abu Dhabi oil became one of the jokes of the world press in the later years of his rule. In Abu Dhabi itself, the fact that people were getting very little out of their oil caused increasing dissatisfaction among both the ruling family and the general public. It was clear
x
Editor’s Preface
that the situation could not go on like that indefinitely. Indeed, when I discussed the position in Abu Dhabi with a shaikh from another state in the early 1960’s, he asked me why I thought the British Government went on supporting Shakhbut. I said I did not think they were supporting him, but it would be difficult for them to depose him since gunboat diplomacy was now considered objectionable. He replied, ‘There would be no need for the British Government to depose him. If the Government simply let it be known that it was not supporting him, the people would depose him in no time’. And his own brother [the present Ruler of Abu Dhabi] had told him that unless he spent more money the people would come and burn the palace down. To clear the mystery and misunderstanding with regard to Peter’s departure from Abu Dhabi only three months after his arrival, it is important to look at this in the context of the relationship between Peter and Shaikh Shakhbut. Through his research Peter came to know the way Shakhbut conducted himself to his family and to his people in the shaikhdom, as well as in his relationships with other rulers of the shaikhdoms in the Gulf. Peter frequently sat in Shakhbut’s majlis and learnt a great deal about Shakhbut and the affairs of Abu Dhabi. Once Peter was appointed as adviser, then he assumed the role of suggesting to Shakhbut what was needed to improve conditions in his shaikhdom. The latter was reluctant to delegate some of his powers or let anybody share his authority, as he wanted total control over finance and policy. Within a short time, as Peter related to me, the relationship with Shakhbut changed from being friendly to being strained and frosty. Moreover, Shakhbut was, according to Peter, a very difficult person to deal with. He was suspicious, authoritarian and an unpopular ruler. Even official records hardly praise Shakhbut personally or appreciate his period of rule (1928–66). With disagreement over the nature and pace of development, Shakhbut would not take Peter’s advice and decided to terminate Peter’s contract. But the official reason given for Peter’s departure was absurd, to say the least. The British Political Agent reported in his ‘Trucial States Diary’ under June 1961 that Shaikh Shakhbut summoned Mr. Lienhardt and told him he was disappointed in the lack of progress since the latter’s arrival in Abu Dhabi. Mr. Lienhardt told him that if money were spent more freely, progress might follow. On 21st the Ruler accused him of disloyalty and (in veiled terms) of taking bribes from Halcrows. On June 22nd he gave him a month’s
Editor’s Preface
xi
notice. Mr. Lienhardt replied that he would rather be paid up and go; this was agreed and on June 25th he was paid and went.2 Shortly after his departure Peter was vindicated on two counts. Firstly, in his ‘Persian Gulf: Annual Review’ dated 28 December 1961, the British Resident in Bahrain reported to the Foreign Secretary: The Ruler [Shaikh Shakhbut] . . . did not embark on any major development scheme, and hopes that he might establish the necessary basic administrative structure were disappopinted when his British Secretary [i.e. Peter Lienhardt] left after three months as a result of unjust accusations by Shaikh Shakhbut.3 Secondly, after realizing his mistake, Shaikh Shakhbut sent an apologetic hand-written letter in Arabic to Peter in which he hoped that the latter would change his mind and resume his post. After the hurtful false accusation, Peter thought that there was no further trust between them and that it was pointless to return. Shakhbut’s action wounded Peter deeply and left him with a feeling of justifiable dislike towards a man who had questioned his integrity. However, this incident did not change Peter’s respect and regard to the people in the shaikhdom, and a year later he took an active role in trying, with the knowledge of the Foreign Office, to help Shaikh Shakhbut’s son, Sultan, while the latter was studying in England. In fact Sultan came to stay with Peter in Oxford in order to discuss his problems and affairs despite the fact that Peter refers to himself, in his correspondence with the Foreign Office, as ‘persona non grata’ with Sultan’s father, Shaikh Shakhbut. After this episode Peter went back once to Kuwait but never to the other shaikhdoms of the Gulf, even when Shaikh Shakhbut was deposed by his brother, Shaikh Zayid, the present Ruler. So what enticed Peter to undertake his anthropological research in the shaikhdoms of the Gulf in the first place? In his unfinished ‘Introduction’ he states: My first interest in anthropology was in politics – not so much politics in the day-to-day sense of current affairs or even the year-to-year sense of foreign relations, but rather politics in the sense of structures in terms of which political behaviour could be related to the way in which people lived, their physical circumstances and their values, though it was clear to me then and now that in order to understand the lasting [sic] one must, among other matters, inquire
xii
Editor’s Preface
Fig. i Peter Lienhardt and Shaikh Shakhbut bin Sultan of Abu Dhabi in 1961. From the author’s photographic collection.
into the things of day-to-day. My linguistic qualifications were in Arabic and Persian. These were the two reasons why I chose the shaikhdoms of Eastern Arabia – Kuwait, Bahrain, Qatar and the states of the Trucial Coast – as the area for my first anthropological fieldwork. . . . Politically these shaikhdoms were of special interest. They were not colonies or protectorates with foreign administrators. They had neither adopted representative forms of government from Europe nor military rule and . . . it was possible to enter them and move about freely. Moreover, the traditional circumstances of the various shaikhdoms had enough in common for some comparative study to be possible. Peter’s reflections on the circumstances and problems of conducting anthropological fieldwork in these shaikhdoms constitute an important contribution to the research methodology in Arab/Muslim communities. For example, he writes:
Editor’s Preface
xiii
As anyone knows who has done anthropological fieldwork in more than one area, not only the circumstances of the anthropologist’s life but the circumstances of gathering information vary very considerably with local conditions.4 For this reason alone, it is desirable for the anthropologist to say something of the circumstances in which he worked. In the case of the Gulf 5 there is an added reason, since the circumstances of life for the foreign visitor, whilst they have become much more comfortable physically have become much less so socially, assuming that is, that the foreign visitor is interested in knowing the local people, for as the people get richer they often get busier and less accessible. Moreover, as the number of foreign visitors increases they cease to have the interest of being curiosities, so that much company which was formally open and easy-going becomes remote and formal. Up to a point, this was the case in Kuwait even in 1953, but it was not so at all in the Trucial Coast until oil was discovered. I had become friendly with some Kuwaitis in England and they and their families remained kind and helpful in Kuwait. I spent a lot of time sitting with one group of friends outside a bicycle shop and in the mornings was able to sit in two or three offices of the civil service. These offices were places to which people, as well as coming for business, made purely social calls, just like going round the shops in the market. The civil servant’s desk was at one end and the rest of the walls were taken up with sofas and easy chairs packed in with their arms touching so that there should be ample room for visitors. It was possible to meet many people there, but not everyone was readily admitted or, when admitted, invited to sit down. Probably most Kuwaitis were, but some of the people to be seen waiting outside for hours on end were people who stayed or worked in the hotel. One frequent difficulty they had was that they were illegal immigrants wishing by some friendly influence to obtain residence permits. He continues: Helpful as my Kuwaiti friends were, I found it impossible to find a house, even a little mud house, that I could afford the rent of, because the rents were too high. It was out of the question to try to rent a room in a Kuwaiti house, since the women were strictly segregated from unrelated men. At that time I never met the wife or the mother of any Kuwaiti friend. Moreover, Kuwait proved to be a much more complicated place to begin a study than I had expected.
xiv
Editor’s Preface
The Kuwaiti people were already somewhat withdrawn from foreigners, be they Arabs from other countries or English people. It was much easier to meet them in semi-public circumstances than to have lengthy private conversations and so the questions one could ask had to be few and general. Had I wished to change the subject of the study, it would have been very easy indeed to study some group of Arab immigrants, for these were usually men who had left their families at home and had an endless desire for company and talk. There were already as many adult men who had immigrated to Kuwait as there were adult Kuwaiti men in the city. Their interests and outlook, and their privileges, were a good deal different from those of the Kuwaitis. They came to Kuwait to make a living and needed the employment which Kuwait provided, whilst Kuwait needed labour in its massive construction work and its growing commerce as well as the civil service, the health service and education. . . . Kuwaitis criticized the immigrants for ingratitude and sometimes for greed and dishonesty. It was not possible to mix easily in Kuwaiti and immigrant society since the two themselves did not really mix. As for the choice of place to undertake anthropological research, Peter writes: I decided that it would be better to go to a simpler place to start methodical research. I should have preferred the Jahrah Oasis, which is the main bedouin centre in Kuwait territory, but it was made clear to me that this would not be acceptable to the Kuwaiti Government, and so I went to the Island of Failakah which lies outside the Bay of Kuwait. The population was about three thousand and the main occupation was fishing. Even Failakah was a slightly tricky question because the late Ruler of Kuwait, Shaikh [Abdullah Al-Salim (1895– 1965), was very fond of the island, but I was given permission to go there on condition that I did not try to measure people’s heads or dig up their graves (neither of which, fortunately, I had the slightest wish or competence to do). . . . Many of the people living in Failakah village were immigrants, but apart from three or four Palestinian school-teachers they were all from the Gulf. Apart from the Palestinians, no one there spoke English and so it was much easier to practise colloquial Arabic than it had been in Kuwait where even little boys in shops had been eager to practise their English, however little they knew. And it was in Failakah that I first became aware of one of the basic traditional patterns of life in the Gulf, for I came across a group
Editor’s Preface
xv
of forty or fifty people, sailors and their families, who had moved as a corporate group from place to place in the Gulf with a family which owned three or four deep sea dhows. I also became aware at first hand how people could be forced to the hazards of the sea by debt, another theme in the traditional life of the people of the Gulf. He goes on: It was an excellent thing for me that none of the Failakah people knew any English. Living in these circumstances, apart from the interest of one’s work, the pressure of loneliness forces one to make the strongest possible effort to be understood and to understand. If one has no medium of communication excepting the language of a very foreign society, then one absorbs the representations of the society with the language. To take a simple example, the word ‘uncle’ in English has certain moral and emotional connotations (some of them quite complex if one thinks of the various jocular uses of the word). In Arabic, there is no word which would translate ‘uncle’ since the brothers of the father are distinguished from those of the mother. Hence there are two terms, [amm for ‘father’s brother’ and khal for ‘mother’s brother’. These terms have different connotations, since one’s relations with these two sorts of uncles are very different – as different, say, as the English relation with the father and with the grandfather. But the actual terms ‘father’s brother’ and ‘mother’s brother’ have very little moral or emotional connotation in English, since they are not normal terms for speaking of kin. It is only by thinking and reacting in terms of the very Arabic words [amm and khal that one can absorb their meaning. Such considerations as these cover a great deal of the anthropologist’s work. In classical texts, every time the name of the Prophet Muhammed or the names of other prophets are mentioned, they are followed by certain honorific expressions. When we read these as students in Cambridge,6 we read them purely mechanically. Hearing them spoken with piety in Failakah taught me that I had not been taking the religion of Islam as seriously as I should have. Though I had not realized it at the time, reading about Islam in circumstances totally divorced from the simple Islamic devotion, I had been thinking of the Prophet Muhammed far too much in terms of his old-fashioned Christian representation of the warrior and the authoritarian lawgiver, and far too little in terms of the most perfect human model of every virtue, including clemency and charity. And whilst sitting with the Governor of Failakah Island listening to the wireless one day, I asked
xvi
Editor’s Preface
a question which caused me acute embarrassment. I asked what was the singing that was being broadcast, and the Governor replied with surprise ‘It is not singing, it is the recitation of the Koran’. The Koran is intoned, but singing has no religious connotation in Islam, and my question was rather more inapposite than it would be if one confused the Agnus Dei7 with a music-hall song. But it seems to me far better to have to learn a language with all the frustration and embarrassment that this takes, than to have the assistance of an interpreter who, however helpful he tries to be, is going to try to break down local concepts and associations of thought and re-form them into what he supposes to be one’s own. After his stay in Failakah, Peter moved from Kuwait to Bahrain, Abu Dhabi, Dubai and other shaikhdoms of the Gulf. Out of these experiences he describes and analyses in the chapters of this book the complex political and social institutions of the people of the Trucial States. His marked sensitivity with regard to the portrayal of these institutions reflects his understanding of the language and culture of the area. It is not my intention to assess Peter’s work or to update some sections of the book as this would unnecessarily lengthen an already long manuscript. It is hoped that these topics will be pursued by future researchers. Suffice to say, however, that Peter left behind a unique anthropological portrait of the Trucial States. I say ‘unique’ in view of the quality of the ethnographic coverage at the time and the depth of understanding of Arab/Islamic culture. I think he is one of a very few scholars who had a real grasp of the traditional institutions of this culture. Though each shaikhdom forms a separate political entity, they all share some similar historical processes and their societies have some common social patterns. Thus, to study one shaikhdom, one has to take into consideration the others. Peter has tried to do this as can clearly be seen, for example, in his successful and illuminating abstractions on the position and role of men and women in the Trucial States. From his extensive anthropological fieldwork Peter published a few articles on the Trucial States (see Bibliography). However, it is appropriate to emphasize one crucial point: the present book and the articles he published must be seen in the context of the time in which the fieldwork was undertaken. The discovery of oil and consequent influx of wealth and people have transformed the economy and society of these states. The wealth of the Gulf states has become the envy of the Arab world and other developing countries. As a result of the high income of the people in this area, many are experiencing a comfortable style of living, international travel,
Editor’s Preface
xvii
the desire for luxury material goods, education, modern health facilities, administrative and political changes. Moreover, the influx of foreign workers (Arabs and non-Arabs) has shifted the population balance in favour of the immigrants. Substantial sums of money are spent each year on health, education, social security and defence. Affluence has led to the near extinction of pearl fishing and to the nomadic population becoming sedentary. But such drastic economic changes are not without problems. The emergence of modern states in the Gulf has, to some extent, weakened the power of the shaikhs, and ordinary people are increasingly seeking the democratization of these states in order to have more say in their running. In comparison with other shaikhdoms of the Gulf, Kuwait has been relatively liberal and progressive with regard to the need for political and social changes. On 1 December 1999, the Times newspaper reported that in Kuwait: The all-male parliament narrowly rejected a draft law granting women full political rights after seven months of debate in the conservative Gulf state. The vote was 32 against to 30 in favour. Two members abstained. Those opposed were mainly Sunni Muslim Islamist and tribal MPs; those in favour were mostly liberal and Shia Islamist elected MPs and government ministers. This is a radical change in the conservative attitudes of the people of the Gulf shaikhdom. The shaikhdoms of the Trucial Coast, however, are nowhere near accepting, let alone implementing, the policy of women’s participation in politics. These changes have widened the horizons of the local people. Certainly, members of the present generation have not experienced the kind of life that their fathers and grandfathers lived and that Peter described in this volume. On the other hand, recent changes are seen by some people in the area as the cause of the erosion of traditional moral values and changing attitudes of the younger generation. One major concern in the area nowadays is to strike a balance between the demands of social traditions and the desire for modernity. So far the societies of the Gulf shaikhdoms have not experienced the political and social upheavals which have been encountered by other Arab/Muslim societies in the Middle East and elsewhere. The need for political and economic co-operation prompted the shaikhdoms of the Trucial Coast to come together to form in 1952 the Trucial Council, the United Arab Union in 1968 and in 1972 the United Arab Emirates (UAE). This federation consists of Abu Dhabi, Dubai, Sharjah, Umm Al-Qaiwain,
xviii
Editor’s Preface
Fujairah, Ras Al-Khaimah and Ajman. But in 1981 Kuwait, Qatar, Saudi Arabia, Oman and Bahrain joined the UAE to form The Gulf Cooperation Council (GCC). This political development is seen as a means of narrowing the inequalities that exist between them and unifying their laws in the long run. The GCC has proved to be a workable institution and, so far, there has been no opposition to its continuity by the constituent states. In preparing the text for publication, names of individuals, tribes and places have been given as they appeared in Peter Lienhardt’s text. I give diacritical marks only to Arabic words, expressions and sayings which appear in italic. This is done in order to make the text less cumbersome in view of the many individuals, tribes and places cited in the text. However, in order to avoid confusion over the names of people, I have made use of two diacritical marks: the [ain ( [) and the hamza (]), which will be given for names of people and for words in italic. Anglicized Arabic words have been left out of the conventional system of transliteration. Gulf colloquial Arabic words have been maintained as Peter has given them. In the Notes at the end of the book, and occasionally elsewhere, double quotation marks have been used to identify his original footnotes. My additions within these are enclosed in square brackets. All other notes are my own. Peter incorporated in his thesis a small map of the shaikhdoms; hence my larger and more detailed map of the area. Apart from minor technical changes to the text and the incorporation of editor’s notes, the text of each chapter has been left as Peter wrote it. Anthropological monographs have a common layout: chapters are invariably divided into sections. However, Peter’s style of writing in this volume is in the form of an anthropological narrative. He was not writing a monograph on a specific community within the shaikhdoms of the Gulf. Rather, he was making abstractions on the political and social institutions of these shaikhdoms as a collectivity. Hence, to break up the coherence of each chapter would interfere with the continuity of ethnography and analysis. Peter makes frequent reference to the shaikhdoms as the Trucial States and to their geographical location as the Trucial Coast. Thus, in order to avoid confusion, it is appropriate to maintain the title of his thesis, ‘Shaikhdoms of Eastern Arabia’, as this term covers the Trucial States and other Arab shaikhdoms of the Gulf to which Peter also refers. The six chapters in this book have gone through Peter’s scrutiny many times as he was a perfectionist in his writing. I have read, as no doubt have others among his friends and students, some chapters of the manuscript during the 1960s and on re-reading them I have noticed changes and complete re-writing of some of the material. But no
Editor’s Preface
xix
manuscript of any chapter containing comments by anyone other than Peter has survived. He was careful as to who should read these chapters in view of the sensitivity of the material at the time. Finally, Peter did not include in his thesis any photograph from his collection. However, to give a contemporary visual perspective to the text I have included a photograph of Peter with Shaikh Shakhbut and a number of black and white photographs dating from the time of his fieldwork. The Trucial States constitute an important economic and strategic area for Arab and Western countries. This importance was very much emphasized during the Gulf War. The defence of the shaikhdoms of the Trucial Coast is seen as a factor vital to their stability and to the continuity of international trade and oil supplies. Many changes and issues have emerged since the writing of this book and these need to be given serious investigative consideration similar to that shown by Peter in his own research over forty years ago. I think he would have been pleased that, at long last, his major work on the Gulf shaikhdoms is in print and that anthropologists, Middle Eastern specialists and people in the region will be able to read and reflect on his work. After his research on the shaikhdoms of the Gulf, Peter undertook two further pieces of anthropological fieldwork. Firstly, among the coastal islands (Zanzibar, Lamu and Pemba) of East Africa (1957–59) and, secondly, in the Persian city of Isfahan (1965–6). He was well-qualified in the languages of these areas and he was able to switch from one language to another with ease and within a very short time. His fluency and understanding extended to reading modern as well as old literary and historical texts. In particular, his command of the Swahili and Arabic languages is manifest in his scholarly work of translating and editing The Medicine Man: Swifa Ya Nguvumali, by Hasani bin Ismail. His linguistic ability, his perceptive anthropological fieldwork and personal friendships have given Peter an entry and an insight into the complex and varied cultural patterns of the Arabs, Persians and the people of the East African Coast. Sadly, he did not sufficiently relate to us in print the fruits of his rich and varied experiences and his scholarship. But future researchers will, when the time comes, delve into the richness of Peter’s fieldwork observations. In the Arab World and Persia, Peter valued his time there and appreciated the friendships he formed with powerful as well as with ordinary people. But he was happiest in East Africa where he enjoyed the openness of people and the easy and relaxed social atmosphere.
xx
IRAQ Basrah
Failakah Is. KUWAIT
Kharg Is.
Bushire IRAN
Bander Abbas
The Gulf
Hormuz Strait of Hormuz. Qishm Is. Khasab Bukha Sha’m OMAN Rams RAS AL-KHAIMAH Diba UMM AL-QAIWAIN Lingeh
BAHRAIN QATAR Doha Waqra
SHARJAH
Dalmah ABU DHABI Yas
0
Miles
100
0
Km
160
Map
The Shaikhdoms of Eastern Arabia
AJMAN FUJAIRAH
Gulf of Oman
DUBAI
ED IT UN
AB AR Liwa
ES AT IR EM Buraimi
Kalba
MUSCAT
1 The Political Complex
This book is concerned with the traditional society of the Trucial Coast of Oman, more particularly in its political aspect, and with the government of shaikhs as it has existed there over the last two centuries. The society of the Trucial Coast is a small one and the style of government has been highly personal and, in that sense, simple. Excepting for the wider culture of Arabia and of Islam, the society has borrowed relatively little from the outside until very recent years. Foreign models and foreign ideologies1 have played no part in forming its dominant institutions. As the title of heads of state, shaikh, in these little principalities suggests, their form of government is connected with the institutions of the bedouin tribes of the Arabian Peninsula. And bedouin form an important and formidable part of the population. But something beyond tribal institutions has needed to develop, since a large part of the population has traditionally been made up of settled people in coastal towns and villages who depended for their living on the sea. The centres of government of the Trucial shaikhs have been the coastal towns.
Foreign protection and the extent of local autonomy The maritime activities of the people were not restricted to local waters. Apart from ordinary local fishing, they were engaged in pearl fishing in the open seas of the Gulf and active in trade carried by their own boats over the whole area of the Gulf and the Indian Ocean. These wider activities along a major trade route brought them into political contact and conflict with overseas powers. Similarly on the land side, not only the bedouin and oasis villagers but also the maritime people of the coast were engaged with other powers of the interior. On neither side has the Trucial Coast ever been a place of geographical isolation. It has, 1
2
Shaikhdoms of Eastern Arabia
however, until recent years, been too poor to offer any great rewards to conquerors. Moreover, the mobility of its population, both maritime and bedouin, and the complexity of its position between conflicting land and sea powers have made it difficult to govern from outside. Nevertheless, in the course of the nineteenth century, a situation developed in which the Trucial States were isolated, under British protection, from international politics whilst preserving their autonomy in internal affairs. The sequence of events which led up to this situation has elements of consistency, but it is far beyond the scope of this book to try to discuss it analytically. The facts derive from a far-flung and complex history of international as well as local relations. An outline of the events leading up to British involvement in the area will show something of the size of the subject. At the turn of the eighteenth century, the French were already in contact with people from Muscat through slave trading in Mauritius, and Napoleon sought to establish a position in Muscat itself in order to assist his projected conquest of India. Muscati trade, however, depended heavily on India, and the British were able to forestall Napoleon and constrain the Imam of Muscat and Oman to make a treaty with them excluding the French. Sir John Malcolm, the British negotiator, had pointed out to the Imam that his government had the power to close all the ports of India to Muscati shipping. But because of internal rivalries the Imam and his successor were insecure in their own territory. Moreover, Wahhabis from the interior of Arabia were pressing into Oman and also into what was subsequently to be called the Trucial Coast. The Jawasim of the coast provided a sea force which extended Wahhabi power beyond the land base of this politico-religious movement and threatened both Muscati and British shipping. In order to support the Imam and to protect their interests in the trade route, the Bombay Government mounted an attack on the Jawasim in 1809–10, followed by another and more thorough expedition in 1819–20. Upon this latter expedition there followed a peace treaty which was to form the foundation of British relations with the shaikhs of the Trucial Coast. 2 Thus, in combination with their susceptibility to pressure from the interior, it was the aggressive maritime activities of some of the people of the Trucial States, in the days when that name was yet to be invented and the area was called the Pirate Coast, which started a chain of events that was to preserve the states, in a more traditional political form than they might otherwise have had, up to the year of completing this book, 1971. The expeditions from India, aimed at suppressing piracy and protecting British shipping in the Gulf and the Indian Ocean, led to series
The Political Complex 3
of treaties, truces and agreements which prevented the states from being swallowed up and incorporated in any greater political unit in the area whilst at the same time preserving their independence in internal affairs and in their relations with each other. The British Government undertook to protect the Trucial States from all aggression except aggressions among themselves. In return, such foreign relations as the Trucial States had were conducted through the British Government, which excluded all other powers from direct dealings with any of them and in practice came close to excluding all other powers absolutely. Given the place of the British in India and Europe, their power concentrated in naval forces, it is clear enough how the outcome of maritime politics in the Trucial States since the early nineteenth century was related to the states’ geographical and ecological position. Living by the sea and uncontrolled by any greater political unit, they fell under the influence of the greatest naval power of the time. This influence was founded, in the first instance, on the use of force by the British expedition of 1819–20, and, though to a much lesser degree, force was used on various subsequent occasions. On the landward side, force was also involved. From this direction the Wahhabis were trying to reassert the beliefs of early Islam, one of which was that unbelievers, who included backsliders from the faith, must either submit to their government or be conquered. But in the missionary fervour of the Wahhabis appears another element in politics in addition to force – the element of ideology. Being Muslims who lived, for all their dependence on the sea, in an Arabian desert environment, the people of the Trucial Coast were close to the Wahhabis in their religious beliefs except in point of strictness. Their customs were not very different and their language was the same. Had it not been for the defeats suffered by the Wahhabis at the hands of other opponents, they might well have succeeded in controlling the Trucial Coast, as for short periods they did. As it is, their lasting influence there has been more cultural than political. What was the significance of British ideology in the influence which the British exerted in the Trucial States, and what followed from its consistency or lack of consistency with the ideology of the local people? It would obviously be impossible here to take into account all the materials of a long period of English history which a full discussion of these questions would demand. I discuss them, therefore, only in relation to certain major effects which the British connection had upon local affairs. To this subject and to the allied question of the extent to which local politics can be considered in local terms, as distinct from having
4
Shaikhdoms of Eastern Arabia
to be referred to an external influence amounting, some have thought, to domination, I devote the remainder of this section. During the course of the nineteenth century, agreements made by the Trucial shaikhs with the British or under their auspices produced security at sea, first by forcibly suppressing piracy and subsequently by restricting the politics of force to the land, except in so far as they were employed by the protecting power. Thus local political activity became largely incapsulated within the land area of the Trucial States and its desert hinterland. Movement between the Persian and the Arab shores of the Gulf had been going on since the earliest times, and in the eighteenth and nineteenth centuries there were a number of little Arab principalities similar to the Trucial shaikhdoms on the Persian coast of the Gulf. In the later nineteenth century, these were all more fully incorporated into Persia and their political affairs were finally severed from those of the Trucial States through the prohibition of all warlike activities at sea. In general, the confinement of warfare to the land probably increased the political importance of the bedouin, though this is a question difficult to assess. At the same time it undeniably contributed to the prosperity of the townsmen and villagers of the coast. Fighting at sea had damaged trade and often made the pearl fishery impossible. Thus in times of disturbance there had existed a vicious circle of political and economic insecurity, forcing people to try to seize what they could not earn. Intervention, in part forcible but largely diplomatic, by a foreign power concerned in the first place to protect its own shipping solved the problem, and there ensued a peace that no force based on local resources would have had the strength to impose. The correspondence of British officials of the time on sea warfare and related issues, even including piracy, reveals how relatively small was the part played by crude power politics in the political influence they exerted as representatives of a foreign power. The importance of moral considerations in the foreign political system which was influencing local politics is clearly demonstrated in letters of officers and officials sensitive not only to their own conceptions of political and social justice but also to local people’s ideas of right and reasonableness. Suppression of piracy had been the main object of the expedition of 1819–20, and for a long time afterwards the main British concern was to prevent its breaking out again. It was realized that the only economical way to achieve this end was through local co-operation. The expedition against piracy was conducted with the full vigour of war, and the town of Ras Al-Khaimah was destroyed. The people of the
The Political Complex 5
coast can have been left in no doubt of the power which the British could employ against them. But the terms of the General Treaty of Peace with the Arab Tribes concluded in 1820 were lenient, though partly by accident, since stricter instructions did not reach General Sir William Grant Keir, the British commander, from Bombay until after the treaty had been signed. The signatories to the treaty undertook to co-operate with the British in suppressing piracy. Piracy and the killing of prisoners, which the treaty proscribed, were still by no means ended, but from that time onwards a British naval force continued to seek out pirate boats. Instead of punishing captured pirates themselves or handing them over to their enemies, the British handed them over to the shaikhs from whose ports they had sailed. And they obliged the shaikhs to pay compensation both for property and for lives which had been lost. In the latter case particularly, the British conformed to local custom rather to their own. As I describe later, not only is execution for murder still very rare in the Trucial States but also, even after war, the bedouin custom is that each side should pay blood compensation for the enemy lives they have taken before peace is concluded. Warfare at sea continued, but it was apparent that it really led to acts of piracy against non-belligerents. The total abolition of sea warfare developed gradually through a succession of voluntary annual truces, starting in 1835, from which the name ‘Trucial’ Coast is derived. At first, the truces covered only the summer pearl fishing season, one of the causes of piracy being the poverty that followed from the interruption of this major occupation by fighting. Later they were made to cover periods of whole years, then a ten year period, and eventually, in 1835, there followed a Treaty of Maritime Peace in Perpetuity. The local rulers entered into truces and the final treaty voluntarily. The people of the coast found that they benefitted from peace at sea in respect both of pearl fishing and of ocean-going trade. The British themselves had been cautious about extending the truces, fearing that they might involve them in too heavy responsibilities and that they might prove unworkable by altering, for example, the balance of power between those states which were stronger at sea and those which were stronger on the land. It had been the shaikh of Ras Al-Khaimah, the former pirate stronghold, who had first suggested the total abolition of sea warfare in 1837, sixteen years before the British were prepared to undertake the responsibilities of the final treaty. British representatives demonstrated a similar practical but moral concern over taking action towards removing what was regarded as one
6
Shaikhdoms of Eastern Arabia
of the major causes of fighting at the pearl banks. The problem arose from breaches of the contract which was established between the owner of capital and the pearl fisher by the giving and acceptance of loans at the beginning of the pearl fishing-season. The pearl fishing industry provided employment for many casual or seasonal workers who came down to the coast from the interior as well as for local people who fished year after year from their own states. It was possible for such men as the former who had no local assets that might be sequestrated to accept loans whereby they bound themselves to go pearl fishing from a particular boat during the coming season, and then to leave for another town. There they took refuge with the ruler and went fishing in a boat from his town, defaulting on their debt and disregarding their contract. If their creditors from the first town tried to seize them at the pearl bank – the only place they could try to seize them – the fighting that broke out could then spread. Such calculated fraud contravened shari[a law, and the existence of the problem reveals the extent of competition and lack of co-operation between rulers, particularly since it seemed in the 1870s that some rulers were actually charging fees to absconders for granting them refuge. The British officials, however, appreciated that local customs of protection and refuge represented one very important defence that the weak had against the strong and were unwilling to do anything which might lead to undermining them. The problem was discussed as early as 1843, but a conclusion was only reached in 1879, in the form of an agreement whereby the rulers undertook to refuse to such ‘fraudulently absconding’ debtors and to pay a fine for infringement of the agreement. Disputed cases were to be decided by a council of rulers or their nominees presided over by the British representative in the Trucial Coast, and the council’s decisions needed to be confirmed by the British Resident in the Gulf. Thus in respect of acts of aggression at sea and of some situations that might provoke them, matters threatening the security of the sea lanes and the trade route, the British came to assume the position of arbitrators, undertaking responsibilities that went beyond diplomacy and good offices. The truces and agreements concerning peace at sea included clauses whereby rulers undertook to refrain from direct retaliation in case of infringements. They were to make their complaints to British representatives who would then themselves take action on behalf of injured parties. In all affairs of fighting at sea, the rulers gradually surrendered their autonomy, leaving the use of force to Britain, the dominant overseas power. And Britain acquired both the right and the
The Political Complex 7
obligation to intervene within a specific range of local affairs which only indirectly and potentially concerned British interests. In doing so, the British undertook to act impartially as between one ruler and another. The fact that the rulers’ centres of power, the towns, lay on the sea coast and depended economically on the sea enabled the British to exert what pressure they thought necessary by means of readily mobile naval force without running the risk of more complicated and equivocal operations on land. It followed from the incomparably greater power of the British at sea as compared with any other state in the area that they could insist on agreements being observed without fearing that each case might involve an expanding military commitment. The early historical records suggest that the consistent requirement that rulers should stand by their agreements was an innovation in local political practice. At the same time, the tactical position helped to preserve much of the integrity of land affairs from foreign infiltration and the snowballing of imperialistic interventions. In the later stages of British agreements with the Trucial States, the British interventions in land affairs increased as territorial jurisdiction became an increasingly significant issue. In the time of oil exploration, when the British Government had got the rulers to agree to a final settlement of territorial boundaries, the position of arbitrator was reserved to the political resident in cases where rulers could not reach agreement among themselves. And more generally the rulers were persuaded to meet together regularly to discuss their common concerns in a body called the Trucial Council in which the British political agent also participated. In spite of an earlier history of extreme disorder and insecurity at sea, the smoothness with which the Trucial System developed from its first beginnings in annual truces covering the pearl fishing-season to perpetual peace at sea suggests a measure of basic compatibility between the aims of British policy and the values and institutions of the local people. Historical records make it abundantly clear that the rulers were not coerced into accepting truces and that once the system was seen to work its continuation was welcomed. The practical need was for a sea force strong enough to guarantee that agreements would be kept and sufficiently impartial in carrying out its obligations to win local confidence. Once these two conditions were fulfilled, pearl fishing could go on without interruption and so make a greater contribution to local prosperity. At the same time, as a negative condition of political status, that prosperity was never so great as to tempt the British into steps which might have led to their taking control of administration on the land.
8
Shaikhdoms of Eastern Arabia
Behind these practical issues there lay the facts of attitudes and institutions. In the sphere of values, peacemaking and conciliation, which is what the British were often engaged in, are matters on which a very high moral value is set in Trucial Coast society, and one can safely assume that this was also true in the nineteenth century. The British, then, were doing something which was in the first place readily understood and in the second place acceptable. And the ground was prepared institutionally as well as morally. In bedouin custom in the Trucial Coast, when peace is made after a war it begins with a truce under the guarantee of a neutral mediator. This man needs to be a powerful shaikh because if the truce is treacherously broken by either side he must be able to gain compensation for the injured party and, if necessary, he must join the injured party in attacking the offenders. As guarantors of the truces at sea, the British stepped almost exactly into the institutional position of a local neutral. The last important political issue which arose from the involvement of the Trucial States in the trade routes between Basrah and the Indian Ocean was the slave trade. I will mention it only briefly since it had a limited significance in local politics. Here beliefs which became dominant in Britain as the nineteenth century advanced were in conflict with local belief. As far as the trade itself was concerned, British attitudes prevailed and the taking of slaves from Africa was suppressed, though at one time there were British officials who argued that once slaves had reached the Gulf they were extremely well treated and happy. The local status of slavery continued. When the decline of local slavery came, it was largely from economic causes rather than moral or political pressures. With the collapse of the pearl trade in the 1930s, it ceased to be economically profitable to own slaves, and in the general impoverishment that followed many people emancipated their slaves because they could no longer afford to keep them. Emancipated women slaves were left in a desperate condition without the means of subsistence. Some slaves remained, the property of the shaikhs and of private individuals, but the rulers were persuaded to agree to a regulation whereby any slave could go to the British political agency and apply for an automatic certificate of emancipation. Such applicants were quite a common sight performing a ceremony of taking off their sandals and stepping up to grasp the flag pole outside the Political Agency in Dubai in the 1950s. There were others who preferred to retain their status as slaves. Whilst protecting the Trucial States as political units, the British Government never made them into protectorates in the formal sense of
The Political Complex 9
that word. It took over neither the central power nor the machinery of government. The British did not make themselves responsible for the administration of justice or undertake any other kind of administration. There were no British district officers or even, until very late indeed, police officers or advisers. The shaikhs and the people of the Trucial States maintained control of internal affairs even to the point of being able to make war on each other without running the risk of British intervention. Similarly, in cases of internal struggles, civil war or coup d’état the outcome was left to the balance of local force. Only in political behaviour that might threaten either foreign nationals or the peace of the sea did local people lack full freedom in local affairs. Foreign subjects, whether British or otherwise, came under the jurisdiction of British consular courts rather than of the local rulers, at least until a few years ago when local powers in this direction were increased.3 And, in large scale politics, warfare by sea was forbidden in accordance with agreements reached in the nineteenth century. Infringements of these agreements in earlier days were punished by heavy fines which the rulers were obliged to collect from those responsible under threat of further punitive action by the British navy. Within these limits, however, local politics continued with remarkable lack of restriction. The unusual relationship between Britain and the Trucial States, besides preserving a great deal of an older political situation, had a conservative effect in cultural affairs. The British took no responsibility for, and officially had no right to interfere in, local affairs. Like administration, education was left in local hands. It was not until Kuwait, rich with oil, offered generous assistance that any modern school was opened in the whole area. At this point, the British Government decided to co-operate and make their own contribution. But even then, it was open to local rulers to refuse, as indeed the shaikhs of Dubai and Abu Dhabi did. They claimed to have adequate resources of their own to open modern schools without assistance. Both then delayed. By 1971 Dubai and Abu Dhabi had only had schools for about ten years whilst other and poorer states of the Trucial Coast had had them for almost twenty. The lateness with which any sort of modern education, even at its most elementary, reached the Trucial States meant that they remained largely closed to the influence of foreign thought, though some ideas percolated through from other parts of the Gulf. One can only speculate on the slowness of the shaikhs of Dubai and Abu Dhabi in doing anything themselves to improve education, and some of their private objections may have been religious ones, but they may well also have asked themselves whether the ideas connected with teaching in modern
10
Shaikhdoms of Eastern Arabia
schools were going to be compatible with the kind of government they themselves practised. The shaikh of Dubai had fought a civil war against a ‘progressive’ faction, inspired by a similar movement in Kuwait, in 1940. As late as 1955, there were no foreign newspapers circulating in the Trucial Coast, let alone any local publications. And at that time broadcasting from other Arab countries had little effect in provoking discussion among the general public. People often found it difficult to understand the type of Arabic in which news was broadcast from Egypt, and in as far as they referred to Trucial Coast affairs (it was the time of the Buraimi dispute) Egyptian broadcasts often provoked amusement by their ignorance of the simple details of life in the area. Some unusual names of local personalities were mispronounced, and the names of two of the more important states, Abu Dhabi and Dubai, were often conflated to produce a hybrid called ‘Abu Dubai’. In the Trucial Coast, there was a corresponding and widespread lack of knowledge of the life of the outside world. Few of the Arabs had even a smattering of a foreign language. Very few had travelled to the more advanced countries of the Arab world or to Europe and America, though India was more familiar. The shaikh of one of the villages at the Buraimi Oasis, a man who was to have a brief moment of international fame, could ask how big was the market in London and whether it was bigger than the market in Bahrain. And in the early 1950s a young merchant in Dubai, by far the most advanced town of the Trucial Coast, could find he had to remove a child’s toy globe from his shop because so many of the old men complained that it was against religion to pretend that the world was not flat. Though not everyone thought that way, those who did so were influential enough to sway public opinion. It was a bold move on the part of the Acting Ruler of Dubai to allow a cinema to be opened when many thought that films, by definition, were against public morals, though Dubai was not a puritanical place and the cinema there in fact preceded the opening of cinemas in Kuwait. In the 1950s, the Trucial States stood on the verge of profound changes – changes both economic and ideological. The main economic change followed, of course, from the discovery of oil. During most of the 1950s, however, it was not known whether oil in commercial quantities would eventually be found or not. Oil was found in Abu Dhabi both on the land and off shore, in quantities great in any terms but, in relation to the tiny population of the state, unprecedented in the world. In Kuwait, oil had offered a per capita income as high as that of the richest countries in the world. In Abu Dhabi the per capita income was more
The Political Complex 11
like that of the rich of the richest countries – it would have provided every man, woman and child with an income of something in the order of £5,000 to £10,000 a year. 4 The political implications of this economically freakish situation are, in 1971, not yet worked out. In one sense the situation of the people of Abu Dhabi could be described as the reductio ad absurdum of the presentday capitalist system, or at least of that system’s philosophy. Do rich countries have an absolute right to the independent enjoyment of their wealth, or does the right simply derive from their power to defend it? The answer that the right exists but needs the backing of power if it is to be asserted is scarcely satisfactory. A view commonly expressed in the larger Arab countries is that this is ‘Arab’ wealth and should not belong to any particular little country but to the Arab people as a whole: the few, with their wealth, should be incorporated in the many. Although widening the scale of things, such view is still, in a sense, national. The present-day wealth of the Trucial States has thrust out the question of the political status of small populations there into a worldwide context. In the local context, however, the issue has risen over and over again during the last two centuries. Apart from the early nineteenth century efforts of the Wahhabis to incorporate the people of the Trucial Coast into their Islamic state, both in earlier and more recent times the question of the independence of one village or another has often been disputed. The context was not usually religious but it was one, nevertheless, in which the moral criterion of rights was no less important than the criterion of power. Indeed, questions of the independence of small local groups, settled or bedouin, have been part of the whole tissue of the history of the Trucial Coast since its history has been known. But since the time when oil prospecting rights began to be sold the issues underlying the questions have substantially changed. In earlier times, land itself had very little value. Rights over people were much more important than rights over territory. Subsequently, for a time at least, the situation was reversed. It seemed natural to pay the fees for oil prospecting rights, and the subsequent royalties, to those who had ‘territorial jurisdiction’. Issues of sovereignty were raised by areas of desert which had, for practical purposes, been no man’s land in which influences overlapped. At sea, though the pearl banks had been equally open to everyone in the Gulf, offshore oil was judged, on a western model, to be the property of particular states. And on the land, conflicts arose in matters of hereditary right between shaikhs whom the British Government recognized as independent rulers by virtues of its treaties and other shaikhs who did not enjoy such recognition.
12
Shaikhdoms of Eastern Arabia
Assertions of sovereignty over territory, like the reception of oil payments by central authorities, implied a movement in the direction of absolutism which was at variance with the traditional political assumptions of the people of the Trucial States. It was, in different respects, also at variance with the values of that kind of modern education which was to be introduced and which the oil revenues paid for. Thus, even before its discovery, the prospect of oil was modifying the groundwork of politics. One significant innovation was the establishment of a small military force made up of local people controlled by British officers in 1950. It was at first called the Trucial Oman Levies, but the name was later changed to Trucial Oman Scouts, apparently because this name seemed to the Foreign Office politically less sensitive. The Scouts operated in the Trucial States with the permission of the States’ individual rulers. They were under purely British control. The force produced a remarkable change in conditions of security in the desert. Before and during the Second World War, the whole of the Trucial Coast had suffered severe depression. Poverty had driven many bedouin into raiding and then into indiscriminate robbery, whilst lack of funds had reduced the power of the rulers to keep the peace. Women and boys had been kidnapped from the outskirts of the coastal towns and taken into the interior to be sold into slavery. With much tact and remarkably little use of direct force, the Scouts brought to the desert a security probably greater than had ever before existed there. And this security was not to the advantage of the settled people alone: the proliferation of bands of robbers had disrupted the economic life of the bedouin themselves, thus forcing them into extending the violence which contributed in the aggregate to their own poverty. Hence the local view of the Scouts was commonly sympathetic on various moral grounds. Nevertheless, they represented an extension of British power, and, so long as the British were willing to support the rulers, a consolidation of the rulers’ authority. The founding of the Scouts certainly marked a new departure in British policy, hitherto based on security at sea and opposed to any steps which might imply responsibilities in the internal land affairs of the area. One can only guess at the motives which lay behind the change. It was quite commonly supposed in Dubai that the Ruler of Sharjah had first raised the issue by telling the British Government that unless they found some means of protecting him from the Bani Qitab,5 a bedouin tribe living in what he and the British Government regarded as Sharjah territory, he would have to apply for protection to the Saudi Arabians. But even if this were true – which it may very well not be – it
The Political Complex 13
would scarcely have been sufficient grounds for such a significant change. The British Government would have had other alternatives open to them since the act of applying to a foreign power would have been against the terms of the treaty. But the effects of the change of policy are indisputable. The Scouts who made the desert safer for local people and local economic life also made it safer for oil prospecting. The fruits of that prospecting are now being enjoyed by the people of the Trucial Coast, the oil companies and the British exchequer, and so financial success can be added to the list of improvements to which the Scouts have contributed. But the establishment of that force also had side effects of which the advantages were by no means so evenly shared. Though not under the control of the rulers of the Trucial States, the Scouts were at hand should it be decided to comply with any ruler’s request for assistance. They could reach almost any part of the Trucial States at very short notice and so became a highly significant potential force in local politics. In formal terms, the Scouts only acted in each of the Trucial States at the request, or at least with the permission, of the ruler of the state in question. In practice they made the British Government more of an arbiter in the question of who was to rule. In 1954 the Ruler of Dubai became alarmed at a threatening situation that had arisen between his son and the sons of one of his brothers and feared that the latter try to seize power. He arrived in the middle of the night at the British Political Agency to ask for help, and the Scouts (then Levies) co-operated with his own guards in expelling the brother and four of his sons from Dubai. They succeeded also in blocking the route to Sharjah, only seven miles away, where the brother might have hoped to find protection and so continue to be dangerous. Coups d’état have played such a repetitive role in the history of the Trucial States that it would not be going too far to describe them as one of the institutions of local politics. On this occasion, a straw in the wind of change, the people of the state were not left to resolve an internal conflict alone. The British authorities held the balance, and had British ideology been that of fifty years earlier one might have recognized such an action as one of the ‘inevitable’ steps towards the introduction of a protectorate or colonial system. The British authorities, whether or not this was their general intention (and I think it was probably not), had placed themselves in the position of being able to make sure that a particular coup d’état was not going to happen. This was a discouragement to others who might think of embarking on the same traditional course without considering what the British reaction might be. Such
14
Shaikhdoms of Eastern Arabia
intervention did not represent the limit of the Scouts’ powers: later, notably in Abu Dhabi in 1966, they were able to co-operate with a ruler’s opponents, a proceeding that had more to be said for it de facto than de jure, and make sure that a coup d’état was successful. If the Protecting Power took it upon itself to assist some rulers to maintain their positions, it automatically took on more responsibility for the style of government the rulers practised. All this was part of a change in the proportions of Trucial Coast affairs dating from the time when oil began to be considered as a serious possibility, not merely from the time when it was found. The discovery of oil confirmed changes which had begun already. Initially the prospecting agreements did not need to stipulate precisely the territorial boundaries in which prospecting was to be carried out. These boundaries for long remained in dispute between one ruler and another. Dubai and Abu Dhabi fought a war over a border area called Jabal [Ali, and at one period the shaikhs of Abu Dhabi suspected the shaikhs of Dubai of a plot to take over the Buraimi Oasis. Saudi Arabia claimed an enormous part of what the shaikhs of Abu Dhabi regarded as their territory, whilst the Ruler of Abu Dhabi complained bitterly that in his younger days he had taken fishing taxes from Umm Sa[id and that this oil port of Qatar was rightly his. Members of the British Foreign Service worked for years to sort out agreed local boundaries: it was no accident that they had not existed in many places before. The rulers were persuaded to agree that, in cases where they could not reach agreement among themselves about particular boundaries, the British Resident in the Gulf should act as arbitrator and have the power to make a final decision. In this and many other matters it is clear, however, that the British preferred that the rulers should reach agreement among themselves. To this end they had been persuaded in 1952 to hold regular meetings to discuss their problems in a body called the Trucial Council at which the British Political Agent was present. Though the prospecting agreements did not need in the first place to stipulate precise boundaries, it was at least necessary to decide with whom the oil companies were to sign their agreements and to whom the fees, and if oil was discovered the royalties, were to be paid. The course adopted, that of recognizing jurisdiction in those rulers who, with their predecessors, were in treaty with the British Government, meant disregarding many local hereditary rights to political positions. They were matters of ‘internal affairs’. Bedouin shaikhs such as the shaikh of the Bani Qitab were in no position to support a legal claim that their tribal territories were under their own or their tribe’s jurisdiction.
The Political Complex 15
Had they tried to establish a de facto claim by molesting oil company prospectors, they would have had the Trucial Oman Scouts to face as well as the ruler of the state. The same was the case with the hereditary shaikhs of villages, who were also often tribal leaders. In relation to the rest of the population, the Trucial Coast rulers were thus again accorded a status higher than the traditional way of life had allowed them. One complaint frequently heard was, why do the British give everything to the rulers and nothing to the people; does one feed a mother and leave her children to starve? The changes which came about before oil was discovered in the Trucial States and indeed for years afterwards took place piecemeal and on a whole imperceptibly. Excepting in a few instances, as they impinged in practice on most of the people, their immediate effects did not go beyond the limits of political behaviour which had been possible before. They merely approached the upper parameter of shaikhly power. The people of the Trucial Coast were accustomed to wide variations in the extent of their power. There had always been cases of village shaikhs asserting their independence and of town shaikhs taking villages under their control. As between one state and another there had been wide variations in wealth and strength. Methods of rule had also varied. Some rulers had tried to rule alone whereas in other cases authority had been widely distributed throughout a ruling family. The changes, therefore, could only be regarded as structural in a limited sense. They had not yet established themselves in people’s political assumptions and expectations or in their system of values. In Dubai, for example, in 1954 people were not apparently concerned that the Trucial Oman Scouts had helped Shaikh Sa[id bin Maktum to expel his brother Shaikh Jum[a. For them it was simply a matter of a ruler throwing out some members of his own family who might have seized power. Many rulers, including Shaikh Sa[id, had done that before. What was feared by many was that the British wanted to force an Adviser on Dubai on the pattern of Bahrain and thus, it was thought, increase British control over their lives. There was, however, one economic change that had occurred long enough ago and been serious enough to modify political relationships. This had been the disastrous decline of the pearl fishing industry in the 1930s, which had come about through a combination of the development of Japanese cultured pearls with the industrial depression in Europe and America. The industry had been a major source of wealth for the Trucial Coast as it had also been for Bahrain and Kuwait. Its decline, from which it never recovered, brought about serious hardship
16
Shaikhdoms of Eastern Arabia
and some political disturbance. After the Second World War, a few people went on fishing for pearls and made what they regarded as a reasonable profit. But by that time Kuwait and other places in the Gulf were developing fast and labouring work there offered as much profit as pearl fishing and much better working conditions. In political terms, the result of the decline of the pearl fishing industry had been to alter the balance of power in the towns as between shaikhs and merchants. To be a pearl merchant or a boat owner had been the only ‘noble’ occupation an Arab in the towns could take up. It had meant the leadership and control of men, often in large numbers. Labour in the industry had been organized on a basis of profit-sharing and loans rather than a system of wages, leading to a system of patronage and clientage between the owners of boats and those who worked in them. Like the patrons, many of the clients carried arms. Thus bodies of fighting men, sometimes numbering hundreds, existed in the towns under the leadership of families other than the shaikhs and were a significant factor in local politics. The decline of pearl fishing led to the dissolution of these political groups. They had disappeared before I first arrived in the Trucial States in 1954, and therefore anything said about this highly important factor within the whole traditional political complex of the Trucial States has to be a reconstruction of what had by that time ceased to be influential. Fortunately in 1954 it was only about twenty years since the collapse of pearl fishing, and what had happened in those times and earlier was still well remembered. And we do not need to rely on memory alone: there exists also an ample quantity of historical documentation. Being contained in political dispatches, it concerns itself more with crises than with day-to-day life, but of crises there were many.6
Instability and structure The extensive historical documentation which we owe to British dealings with the Trucial States from the early nineteenth century onwards contributes something more than simple depth to any study of Trucial Coast society. The sequence of events there recorded certainly reveals a process of development which makes it easier to understand how the Trucial States have come to be as they are. But a much greater part of the records is taken up with events in which people changed but situations remained the same. And no one who reads through the detailed history of the area 7 can fail to be aware of the striking number of repetitions of similar situations from time to time and from place to place.
The Political Complex 17
Comparable repetitions also appear in the history of other shaikhdoms of the Gulf such as Kuwait. In relation to any one state such repetitions would be noticeable, but the range of evidence is greatly widened when we have before us the history of numerous states in which ruling families have governed and been expected to govern in the same way. The number of similar governments and similar state communities in the Trucial Coast also makes it possible to study them comparatively in the kind of short term research on the spot represented by a couple of years of anthropological fieldwork. Here again, repetition of situations becomes conspicuous. Such repetitions, revealed both in field information and in history, suggest a pattern which is not simply one of sequence but which in an anthropological use of the term I shall call ‘structural’. But here some clarification becomes necessary. The types of repetition I shall consider are often repetitions of violence and political disturbance, and for many people the word ‘structure’ is associated with a more peaceful condition. Some detailed discussion of the anthropological concept of structure, and of the word ‘stability’ which often goes with it, in relation to the present study is therefore desirable, and I devote this section to it. In writings about society, both popular and technical, the word ‘structure’ has been used in various different ways and sometimes with disguised moral implications which produce confusion when the word is being used as an analytic concept. The crudest, and perhaps the most general, usage of the word ‘structure’ is as a mechanical metaphor derived from building – and it will be noticed that ‘building’ itself has a moral and emotional connotation when it is applied to societies or nations. The use of ‘structure’ in this sense implies that societies are like houses: if structural changes are rashly or inexpertly brought about in them, if a wall supporting the roof is mistaken for a mere partition wall, if their ‘foundations’ are disturbed, then they may ‘collapse’. (What might be identified as the social equivalent of the force of gravity is not usually investigated.) The wide currency of the word ‘structure’ in this sense suggests, however, that this is a significant, rather than a merely idle and illogical, usage. To take an example almost at random, when the Turkish army seizes control of the country a veteran politician is reported as saying that ‘the Turkish political structure’ is not ready for the big social and economic reforms which the army proposes. The implication is clear. Such reforms will disturb the political structure and hence promote instability and thereby, presumably, reduce national unity. ‘Structure’ here acquires a moral value, and one associated with the idea that
18
Shaikhdoms of Eastern Arabia
‘stability’, which is nothing more or less than internal peace and continuity, is a thing to which all moral people must aspire. These are propositions to which many people would in general assent, but most would probably be aware, in some different context, that in one part of the world or another there are conditions of stability maintained by a ‘political structure’ so tyrannous that they would happily see it disturbed. Here again, then, they are concerned with moral judgement. This is not to say that usage of the word ‘structure’ and its sister ‘stability’ with implied moral connotations and appeal is not analytic. But the place of such analysis lies ultimately within the sphere of political philosophy, as a part of a system of moral representations. There is another kind of analysis which can be applied to politics and society where the words are used in a sense closer to that which they hold in the natural sciences. Here the changing of one structure into another or the regarding of one compound as stable and another as unstable are matters which can be considered as morally indifferent. They are neither good nor bad. I do not suggest that anyone studying society can ultimately maintain the same moral indifference to it as does a chemist with regard to chemical processes. Indeed, to try to do so would require an attempt to separate the mind from the personality in a way inevitably naive. This, however, does not mean that the attempt to analyse social relations with a suspension of moral judgement and value judgement may not lead to conclusions which are both less confused than they might otherwise have been and also, eventually, of greater moral import. In relation to chemicals, the chemist is morally indifferent; in relation to society, the anthropologist simply tries to be morally dispassionate. In this book I use the word ‘structure’ in a sense now widely, though not universally, accepted among social anthropologists. Some write of structure as though it were inherent in the material and simply needed to be revealed once and for all. The sense I prefer, however, is one in which structure is created by the mind. Structure here is a construct of the writer’s imagination, suggesting what he regards to be the interrelationships of social and other facts within or relevant to a given society. The recognition of further facts as being relevant, or simply as existing, can be expected to lead to the subsequent modification or total abandonment of any particular construct, as happened, for example, at one time when historians began to apply economic theory to the study of history. This does not, of course, imply a merely arbitrary pattern: the construct is intended to embrace as many as possible of the available facts and represent them in appropriate proportions. Thus, as well as
The Political Complex 19
analysis, it should incorporate an accurate description of the society. Although any construct must in part strain and in part over-simplify the material it deals with, the less it does so the more satisfactory it must be regarded as being. And where it cannot be made to demonstrate casual relationships, as is so often the case in studies of society, it is at least intended to demonstrate the complex interrelationship of situations. The making of such a synthesis often helps to clear the mind of assumptions one has quite unconsciously imported into the material from one’s own society. Let me now give two examples, one of the relationship between an ecological and a political fact, and the other of the relationship between a social fact and a political one. I begin with the ecological example. In the Trucial Coast it is to be observed that relationships between the powerful and the weak are more egalitarian than those that have usually existed between, say, peasants and their masters in Middle Eastern countries. This relative independence of the common people can easily be related to mobility of population. The Trucial Coast has very few people whose livelihood depends on agriculture. The people of the interior are bedouin, whilst the settled people of the coast have traditionally been seafarers. Both the settled people and the bedouin have been able to move away from circumstances they found oppressive. If peasants were to move, they would have to leave their land and livelihood behind – a considerable disincentive. But just as the bedouin moved away with their herds, the settled people could sail off in their boats carrying many of their other possessions with them. Their livelihood was assured. Thus in this instance ecological circumstances, particularly the lack of water in the Trucial Coast which people there have so much regretted, have contributed to political independence, a fact first pointed out in relation to the Arabs who inhabited the Persian Coast of the Gulf by Sir John Malcolm in 1815. The second example is concerned with polygyny and divorce in relation to political fission. Both religion and social convention allow the people of the Trucial Coast to practise polygyny and divorce, and many do so. Hence many men have sons by more than one wife. This is so in the case of the shaikhs – all members of ruling families being called ‘shaikh’ – as with other people. But the shaikhs are something of a special case. To have many sons strengthens the position of a shaikh vis-à-vis other members of his family. Moreover marriages with the daughters of leading members of the community establishes him in a special relationship with these leaders which may be useful to him in politics, and useful also to them. People help and are helped by their
20
Shaikhdoms of Eastern Arabia
mothers’ families and their wives’ families, particularly when their wives have borne sons. And sons are, in general, subordinate to their fathers, so that a shaikh who has sons by more than one wife can hope to hold together the political connections he has made through his marriages. But at his death this unifying influence is removed. If a quarrel arises between half-brothers, each will probably have the support of his maternal kin. In this sense, the sons may be regarded as being united in terms of their father but divided in terms of their mothers, and since shaikhs often seize power from each other this situation has had its effect in politics. Not only the support but also the influence of their mothers’ families contributes to divisions between the sons of polygynous marriages once the father is dead. In isolation such examples may seem obvious. But when one tries to incorporate them in a more general construct, questions arise which oblige one to take a stricter view of the facts. If, for instance, one tries to work out the relationship between the first of these examples and the second, one finds that although both are true neither has been adequately stated. The earlier example is the sounder, but even here one has to recognize that there have been two kinds of population movement, both of which have contributed to political freedom but which have had different consequences in other respects. The second example, that of polygyny and political division, needs, however, to be formulated in the more general terms of the significance of relationships through women in a society where agnation is regarded as the dominant principle both in the family and in the unity of larger political groups. Polygyny thus becomes less of a self-contained phenomenon – which a person brought up in monogamous society might assume it to be – and more of a detail in a system of relationships between agnates and mothers’ and wives’ agnates. To take population movement first: the two kinds of movement have been of large groups and of individuals or little families. In the historical records – and we are fortunate here in having, preserved for historical use, the day-to-day reports addressed by its political agents to the Bombay Government – we find a great number of cases where people moved as large, agnatically-based groups in the form of whole tribal sections or substantial parts of them. In 1833 such a movement established independent Dubai as we now know it, and in numerous other cases large bodies of fellow tribesmen, displeased with one ruler or another, tried to set themselves up in a state of independence by setting off in their large ships and settling on some other part of the coast. On the other hand, out of dissatisfaction or in the search for better opportunities,
The Political Complex 21
many other people have moved as individuals or as little family groups. In either case, rulers whose style of government has displeased the people of the state have found the people melting away from under them. This has served to restrain some in their excesses and to lead to the overthrow of others by the people left behind. Both kinds of population movement have served to limit the powers both of rulers and of other great men. But in respect of kinship and marriage, the two kinds of movement have had different implications, particularly because of the two Islamic conventions whereby women are required to live secluded from men and can be divorced at will. In the society of the Gulf, the seclusion that women are required to practise makes it very difficult for them to maintain themselves either economically or socially outside the family group. And when a woman is divorced she must return home to be looked after by her own close paternal kin – her father, her paternal uncles or her brothers (unless she has a grown up son, in which case he should be the first person to look after her). Around this situation of seclusion and potential divorce is built a complex of reasons why women are usually unwilling to move away from their paternal families and settle in another place with their husbands. For a woman to marry a man who lives a long way away from her family is even more difficult. In the Gulf, as in other Arab societies, marriage with a close agnate is a preferred marriage, and another common way of marrying is where any sort of kinship connection exists already. Thus, when a large number of people have moved together from one place to another, the fathers and brothers of the women have moved with the husbands. The women have been able to move with the group. They have had their own kin there with them to turn to if unhappy or divorced. With regard to an individual man moving to a new place the state of affairs is different. He is likely to marry into the community of the new place, and even if his wife has come with him and stays with him there, his sons and daughters are very likely to marry into the local community they have joined. Thus through the ties of women they become incorporated into the local community, their community of adoption, and in the course of time the agnatic bond with the community they have left behind is weakened until it counts for very little indeed. This may lead us to look at the facts of polygyny and divorce in rather a different way. I said earlier that when men marry more than one wife their sons are united in the father but divided in terms of their mothers. I can see no alternative to using these words, ‘united’ and ‘divided’, short of trying to use some invented and indigestible vocabulary.
22
Shaikhdoms of Eastern Arabia
Nevertheless, ‘united’ does usually, like ‘stability’, have a good connotation and ‘divided’ a correspondingly bad one. We can correct this impression to some extent by putting the statement another way and saying that the sons are united with their mothers’ people, a fact of no small importance if we are thinking of social integration. This fact of social integration through women is not limited to the lives of the settled people alone. Although I will not go into detail at this point, it is significant that friendly relations between neighbouring bedouin tribes are spoken of in terms of a long tradition of intermarriage between them. Each stands collectively in the relationship of ‘mothers’ brothers’ to the other. This brings us to the last point which arises from the present example of structural investigation, the question of polygyny and political division. If we confine the inquiry simply to the ruling families, we find that it has been common among members of the ruling families in the Trucial States to have sons by more than one wife, but on the other hand many shaikhs have not done so. Shaikhs either marry within the family or marry girls from the families of leading local personalities. These marriages outwards extend the political connections of the family as a whole and of any member of it. They also mean that through wives’ and mothers’ sides of affairs various political interests of other groups are represented within the ruling families. Disagreements within the state are readily reflected by disagreements within the ruling family, in which these connections through women play their part: the system lays the ruling families open to popular pressure. Often rulers have been overthrown by other members of their families. They have never been overthrown purely from outside. Actual cases put the position of half-brothers in perspective. If we consider the numerous occasions when rulers have been overthrown we find instances where half-brother has overthrown half-brother, but also cases of the overthrow of full brothers by one another. Equally we find nephew acting against uncle and cousin against cousin. One very general uniformity runs through all the cases: those who are closer together in descent defend each other against those who are more distant unless one is the other’s principal opponent. When a shaikh overthrows his full brother he may be helped by a cousin, but he will not help his cousin to overthrow his brother. If the cousin tries to seize power, then both brothers are united against him. Now if we consider the position of links through women in these matters, we see that although they have the same connection with their mother’s family, full brothers will probably be connected with different
The Political Complex 23
families through their wives. Half-brothers have different connections through both mothers and wives, and first cousins likewise have different connections through mothers and wives, and of course are less united in their fathers. Looked at in this way, the relationship of halfbrothers appears simply as a stage in separateness lying between full brotherhood and cousinship, the stage at which people have neither father nor mother in common. At this point, any special analytic relationship between polygyny and the struggles that occur in ruling families disappears excepting inasmuch as it is part of a wider situation. This does not mean that the earlier perception of a relationship was wrong, but simply that it was inadequate. The different connections of the sons of a polygynous marriage through their mothers are better regarded in political terms as a step towards that division that comes when cousins have both different mothers and different fathers at the next generation. And this is the very point where the marriage of paternal first cousins is introduced in an effort to maintain family solidarity. The example I have just given may seem somewhat perversely chosen: had the initial formulation been in terms of what many present-day anthropologists would call the ‘alliance’ of agnatic groups through women, instead of being in terms of polygyny, then conclusions could have been reached a good deal more neatly and in much less space. But for this very reason the simpler formulation would have been unrepresentative. I think most anthropologists trying to arrange their material in a coherent pattern proceed in something of the way I have described, and the example chosen here is intended to demonstrate a method rather than to demonstrate, as a simpler example could have done, how conclusions can be supported. I have tried to show how evaluations of the significance of individual facts, plausible enough when presented in isolation, have to be modified in the course of their incorporation into a wider construct. Ideally it might be desirable for the writers of anthropological monographs to discuss formulations that they have subsequently modified or discarded and their reasons for doing so. In practice – and it is a practice I must follow in this book – they usually have to restrict themselves to incorporating an accurate description of a society in the structure they have formulated. To attempt more would be to make the work intolerably diffuse. But the demands of clarity have their disadvantages, as may be seen from the difficulties experienced by so many students of anthropology in the first few months they spend studying the subject. Firstly, it is difficult to grasp how conclusions are reached if one has no idea of what has been
24
Shaikhdoms of Eastern Arabia
discarded. Secondly, the very elegance with which some of the greatest anthropological monographs structure their material can suggest to the inexperienced that there are some societies in the world where none of the confusion and untidiness of mind and convention or the divided loyalties which are common human experience in our own society play any part. One may suppose that the same difficulties arise for other readers of anthropological works. It might seem better, therefore, to start with some general formulation which establishes a criterion of relevance and then to arrange the material accordingly. One might, for example, start with the plausible definition that politics is basically about power and proceed then to a discussion of cases arranged in terms of assertions of power, reactions to those assertions, sharing of power and so on. One would know of course that at the end there would be left a question no simpler than the one at the beginning: ‘Politics is basically about power’ – ‘And what is power basically about?’ But on the other hand, by adopting such a selective approach one might hope to avoid any suggestion of an allembracing whole – a structure of social phenomena so tightly knit that a substantial change of any part implies a radical changes of the whole. There is, however, a more serious objection to be made against introducing from the outset a system of classification according to logically prejudged categories, which is what such an approach would imply. Scarcely has one begun before one must start reinterpreting ideas, not to get them closer to the facts but to make them fit within the categories. As soon as even so simple a question as the ‘abuse’ of power is raised in such a context, difficulties arise, for the idea of abuse is not to be defined in terms of power but rather in terms of moral expectations and notions of right. More broadly, in religion, which has as we shall see a great deal to do with politics in the present context, ideas concerning power have their place within a body of mutually supporting assumptions and convictions which are to do much less with power than with whatever we may suppose religion to be basically about. Many difficulties of this sort can be avoided if the inquiry is formulated in structural terms – so long as we bear in mind from the beginning that the ‘structure’ is not the whole but only the best whole that we can propose.
The complex of the desert and the sea If the desert has made the tribes of the Trucial Coast, it is the sea that made the Trucial States, and the contrasting circumstances of bedouin
The Political Complex 25
and maritime life offer a key to the understanding of the political complex that has ensued. For here the two kinds of life have not remained hostile and apart, each repelling the other. Whilst not the same, bedouin society and the society of the settled people in the coastal towns and villages are not mutually exclusive and, just as the settled people have been active in the desert, the bedouin have played a part in urban life and the life of the sea. It is notable, however, that all the towns and most of the villages of the Trucial States lie on the coast, for in this area which one side joins with the Empty Quarter of Arabia, there is too little rain for many people to make a living by agriculture. In such arid circumstances, the little dusty oases that exist, such as that at Buraimi, gain a great local reputation with their supposedly enormous date gardens and sweet water ‘so cold that it cracks your teeth’. But in fact most settled life has depended on seafaring. And seafaring itself has been complex and not limited to one kind of activity. The waters of the Gulf are rich in fish and in pearl-bearing oysters, but except for an unsteady little contraption made of date stalks tied together with rope, there would be no boats from which to do the fishing were it not for a trade route through which timber can be imported from India. There is no local wood from which boats could be made, nor does it seem that there ever has been. And India used to provide the main market for pearls. Thus in more than one way the local maritime activities of the settled people of the Trucial Coast, and indeed of the whole of the Arabian coast of the Gulf, have always been connected with long-distance seafaring and the complexities of trade, unless we speculate on some primitive period of unknown remoteness but certainly before the Islamic, or even the Christian, era. Though circumstances of life in the desert are often very harsh, this life considered ecologically has a greater self-sufficiency. The bedouin move about in the desert with their camels and goats (circumstances here being too dry for sheep), spreading out to find pasture in the cooler months and gathering around permanent water in the most arid period of the summer. Together with their tents and the other belongings of the bedouin, their water is carried on the backs of their camels, and the skins in which to hold it are provided by their goats. And from their camels and goats they get milk, cheese and meat to eat and hair from which to make tents, ropes and clothing. Thus the desert of the Trucial Coast provides, without the need of outside links, the minimum necessities of life for a thinly scattered pastoral population. Like so much of the rest of the Arabian Peninsula, however, it would scarcely
26
Shaikhdoms of Eastern Arabia
do so were it not for the camel, since the camel serves the bedouin as their necessary beast of burden as well as providing them with food, clothing and shelter. Given the nature of the animals to be tended, an adequate minimal economic unit for the peaceful exploitation of the desert environment might be formed by a group of brothers with their wives and children. But if the political possibilities of quarrels and aggression are taken into account, the group of bedouin within which co-operation is practised needs to be of a very much wider scope. In the first place, there is the problem of reaching some agreement over who is to pasture his animals in any particular place when pasture is scarce, as it is at most times of the year. Secondly, there is the question of sources of water which do not dry up in the summer. Such permanent wells are limited in number, and therefore in the summer many bedouin families have to live close together in small areas where the water is and they must practise some co-operation over its use. Both of these matters, which demand wider scale co-operation than that of the group of brothers, wives and children that I have mentioned, can be thought of in economic terms as being matters of co-operation in the use of the changing resources of pasture and water. In a society under the control of an efficient central government they would be regarded as administrative matters rather than political ones. But in the absence of such government among the bedouin the necessary co-operation is formulated in terms of status relationships determined by shared descent in the male line. Groups of cousins regard themselves as united because of a shared ancestor, and the larger groups so formed regard themselves as being united into even larger groups by virtue of an ancestor further back in time. Thus, though we may regard the problem of co-operation as basically an economic one, we see that the formula which resolves it is expressed in terms of kinship, and kinship of a sort that shades off in politics. And there is a third question relevant to the formation of groups which is economic only in the sense that the maintenance of security contributes to economic welfare. Here we return to the camel, whose significance in desert society extends well beyond the simple provision of material means of survival. The range of movement of camels, which can be ridden, and also driven, over long distances at speed, implies a political problem, because camels provide both the means of raiding and the booty to be seized in raids. Times of great insecurity in the desert have also been times of great hardship, since in the course of indiscriminate endemic raiding everyone eventually loses as each little group in turn is suddenly deprived of its means of subsistence. The
The Political Complex 27
problem of avoiding such raiding can be related both to the bedouin idea of the tribe and to other institutions of bedouin politics. In bedouin society, agnatic descent groups of the sort I have mentioned build up into tribal sections, and the sections build up into a tribe. In the bedouin ideology, it is wrong to raid one’s fellow tribesman, let alone a member of one’s tribal section, though such aggressions are by no means impossible. There are, however, leaders who settle disputes and try to maintain peace between their fellow tribesmen. These leaders are the shaikhs. Each tribal section is usually headed by a hereditary shaikh, and the shaikh of one of these sections is accepted as the shaikh of the whole tribe. Apart from their concern with the maintenance of internal relations, any weakness in which has its repercussions in the tribe as a whole by weakening it in the face of enemies from the outside, the shaikhs also represent their tribal followers in relation to other tribes. Through their activities, co-operation can be extended and warfare avoided beyond the level of simple tribal unity, that unity which is conceived of in terms of common agnation. With the approbation and persuasion of the shaikhs as representatives and leaders of tribal opinion, it becomes possible for cases of inter-tribal homicide to be settled peacefully. And the shaikhs are able to extend protection to members of other tribes and enter into treaties with other neighbouring shaikhs in accordance with a whole range of bedouin political conventions which I shall discuss in another chapter. Thus, although in the ideology of the bedouin the tribe, united through descent in the male line from a common ancestor, is the largest group within which political unity should exist, bedouin institutions extend the possibility of political association over a wider range of people and a wider area of territory. Status extends into contract, and a response to demands of natural ecology develops into a body of institutions of increasing political complexity as human groups adapt themselves to each other. And here one last point of structural significance arises in relation to the camel, this time as a beast of burden. Bedouin institutions of protection render it possible, with the required agreement, for men to travel in relative safety through the territories of tribes who are strangers to them. Thus merchant caravans and pilgrims have been able to pass through Arabia from the earliest times, and even from remote places bedouin have been able to visit the towns to sell animals and buy goods and provisions. Consequently, although the bedouin economy is mainly one of subsistence, money as a medium of exchange is no new thing in the desert. And the idea of being paid for services equips bedouin to supplement their own and their families’ resources by going
28
Shaikhdoms of Eastern Arabia
to work in the towns, so long as that work does not involve humiliation before their fellows. In the Trucial Coast many of them used to find work in the pearling industry, and many still work as the guards of town shaikhs. Through money, the economic resources of the towns can be translated into political power. Thus the ancient institutions of the bedouin are associated with an economy that extends well beyond simple subsistence. Though it would be possible for the desert to support out of its own resources a smaller population living a life of even greater hardship, the actual economy that exists there is much less circumscribed. Much of the clothing of the bedouin, part of their food and some of their equipment are not even oasis-produced but are imported from overseas. Nevertheless, in the main this economy is one of subsistence, supplemented by dealing with and through the settled people. The coastal economy, on the other hand, is, not marginally but basically, an economy of markets and trade. Although the circumstances of settled life on the coast used to look rather simple in the absence of so many factory-products and the little interest shown in some of those that were available, nothing could be more remote either from subsistence or from communal selfsufficiency than the economic life of the coastal towns. The towns live through their external contacts, and politically as well as economically much depends on the relation of their people to people outside. Their political structure cannot simply be investigated in terms of their internal relationships. The widest traditional connections of the economy of the coastal towns are those of the trade route that extends between Basrah, the port of southern Iraq, and the coasts of India and East Africa. Over this extensive area of gulf and ocean, local boats based in the towns of the Trucial Coast and many more based on Muscat, Bahrain and Kuwait carried such local products as they had and other cargoes which they picked up on the way. The industry was financed by capital accumulated in the hands of the boat owners. The boat owners not only had to provide the boats and supply them with provisions. They also had to provide the sailors with loans to support their families whilst they were away on their lengthy voyages. Exchanges of money and labour between a boat owner and his crew or crews were long-term transactions. Trading voyages themselves lasted for many months, but in addition the system of loans for particular voyages readily developed into situations of long-term indebtedness. The same applied with the pearl fishing industry. Here, the main season lasted for four months in the summer. The boats, many of which
The Political Complex 29
had very large crews, had to be provisioned and the crews’ families had to be supported for this four-month period. But also there was little work available for many of the pearl fishers during the rest of the year, and the debt system extended itself from season to season and then, in many cases, for whole lifetimes. The boat owners’ capital was often insufficient to supply the demands made on it, and the great capitalists were the pearl merchants. Maritime enterprise was not, however, exclusively in the hands of those possessed of sufficient capital to maintain large boats and the loan arrangements associated with them. In the first place, ordinary fishermen who sold their catches fresh could go out for one night at a time in little boats manned by two or three men and owned by themselves, and sell their catches direct to the consumer. On a scale only slightly larger, it was possible to trade with little boats along the Trucial Coast and with the nearby Persian shore of the Gulf. And in pearl fishing, the records show ample evidence of there having been many small boats kept on the coast by people from as far away as the Liwa Oasis in Abu Dhabi territory and used by them, presumably in little family groups, during the season. Pearl fishers could not, of course, sell regularly to consumers since most of the latter were far away. They had to deal with pearl merchants, but these were many and were in competition with each other. In times of bad luck or financial stringency, small-scale pearl fishers certainly got themselves into debt with the merchants, but the good fortune of finding even one really large pearl could restore them to independence. Thus even in the case of the pearl fishing industry men with large assets were dominant but did not have entire control. Throughout the maritime industries, payment for labour was made by profit-sharing and not by wages. This is clearly to be attributed to the high degree of chance – or, in local terms, luck ordained by divine providence – involved in the operations. Even in the case of large pearling boats where catches were most likely to average themselves out, the finding of a few excellent pearls among little ones that could always be expected made all the difference between hardship and sudden, if brief, prosperity. The proportions in which profits were shared were customary ones, not subject to general alteration; but special individual services were rewarded by extra payments made, mostly by prior agreement, from the boat’s share of the profits. Members of the boat-owning families usually captained the boats, but other captains were sometimes employed and the boat owners also made extra payments for positions of lesser responsibility.
30
Shaikhdoms of Eastern Arabia
Some English accounts of the pearl fishing industry draw attention to the element of exploitation involved in relations between the owners of capital and the pearl fishers, particularly with regard to the tying of labour by lifelong debt. Local memories certainly recall commonly practised abuses such as the overcharging of supplies or the undervaluing of pearls through secret deals between merchants and captains. But there is conspicuous absence in both historical reports and local reminiscence of any evidence of generalized antipathy between ‘labour’ and ‘capital’ in terms of an idea of polarized interests. On the contrary, all the evidence shows that pearl fishers and seamen accepted rich men, so long as they were Arabs, as social and political leaders. In a general sense they were leaders of society and in a more particular and immediate way they were leaders of the men with whom they had economic ties. One would perhaps hardly speculate on the reasons for the absence of polarization between capital and labour in this instance were it not so widely assumed that conflict of interests between the one and the other was in the nature of the situation. But if the pearl fishing industry had continued to exist after some years of modern education it seems highly unlikely that the question would not have arisen. If this supposition is true, then it is significant that the concept of conflict of interest not only failed to emerge from local circumstances but also failed to enter local thought from some other part of the world. At this point of discussion, we are leaving considerations which can be regarded as following from ecology or economics and entering into the sphere of ideology. On the positive side, however, that of the integration of relationships between capital and labour, there is more to be said here. The sharing of profits – generous or disappointing according to the dispensation of divine providence – and the participation of capital in management meant that the owner of the capital and the suppliers of labour participated together in success or failure. Labour was not simply being ‘bought’ by the owner of capital: it had its own rewards out of the joint enterprise. Common interest did not come to an end with the end of each season. There was a great deal of continuity in employment, and a pearl fisher, his brothers and their sons would often work with the same captain. The pearl fisher who found himself in need during the winter months would expect to be able to borrow from the boat owner on the condition that he was going to work in the same boat during the next season. He was not applying to a market for a loan. Similarly the boat owner borrowed from the pearl merchant and so made sure that the boat’s pearls would be offered to him at the end of the next season.
The Political Complex 31
In a small society where there was ready social contact between rich and poor, the informal system of winter loans was elaborately adaptable to personal needs and sympathies. The rich man was thought to have particular responsiblities to those who worked under his auspices, and the respect in which he was held in society depended in no small measure on how he dealt with people through his wealth. And more than prestige was involved: where there was no strong and impartial government system to guarantee the security of the rich or the poor, a rich man had to rely on the loyalty of those he had gathered around him. They were not his employees but his followers – in the local term, jama[a – and just as he would sometimes need to act in the interests of one or another of them, at other times he might have to rely on their collective and forcible support or the threat of it. Hence the situation between the pearl merchant, the boat owner and the pearl fisher embraced more than the restricted roles of ‘capital’ and ‘labour’: it amounted to a system of patronage and clientage in circumstances of limited political security. Thus situations arising out of the economics of maritime organization affected the formation of political groups in the towns. And these groups were of a different kind from the entirely status-formulated groups of the bedouin. When we turn from the effect of maritime activities on local politics to the cultural and political influence (matters not always to be distinguished) of the foreign countries with which the people of the Coast came into contact along the trade route, we advance into issues of greater complexity. With immigration from overseas it became possible for communities of Indian and Persian Shi‘a traders, the former in particular under British protection, to establish themselves in the towns, where they played a considerable part in economic life. As unassimilated immigrants, they were not regarded by the Arabs as having any right to political self-assertion, and in practice they were almost entirely quiescent in politics whilst active in business. Hence the economic base of the towns which were more important in trade, particularly Dubai from the late nineteenth century onwards, became larger than the local political superstructure it supported. Added to this, as steamships came to carry more of the trade formerly carried by dhows, 8 overseas trade came to be concentrated in particular towns, partly in Sharjah at first but later overwhelmingly in Dubai, to the disadvantage of others. It was not only a matter of having relatively safe anchorages and convenient lighterage facilities. In order to make it commercially worthwhile for them to stop at any particular place, steamships needed cargo to load and unload in bulk, and bulk attracted
32
Shaikhdoms of Eastern Arabia
more bulk by increasing the frequency and regularity of steamship deliveries at Dubai through a cumulative process. During the first half of the twentieth century, before the discovery of oil, Dubai became stronger in economic terms than all the other Trucial States combined. On the political side, its territory was small and a very high proportion of its inhabitants were non-Arab.
2 Women and Men
A Bahrain proverb says of an inquisitive woman, ‘She’s blind but she still peeps out through the ventilation tower.’ One might ask, why not the window? But the answer is obvious: Arab houses simply present blind walls to the street, unless they have more than one storey, in which case the upper walls may be pierced by a few windows set at such a height that no one can possibly see in through them. Peeping out may be allowed; but it is peeping in that must be prevented. And so must the overhearing of private family affairs. Even the doorway of a house is so positioned that, when opened, it offers no view of the inner court on to which the windows and doors of the rooms are turned, for the doorway gives on to a stretch of curtain wall and in order to enter the court one must go round a corner. The rooms of the house are arranged around this inner courtyard, and if they are too few to surround the courtyard completely, walls or fences too high to see over complete the enclosure. The main door of the house is usually kept barred, and unexpected visitors are kept waiting for a few minutes before being invited inside. This allows the women of the house time to tidy up before they withdraw modestly out of the way. People who keep a regular sitting room1 for entertaining guests have it built just inside the main doorway of their house and usually have the passage that leads past it into the courtyard closed by a further door, which is kept fastened, again to ensure the privacy of women. It is possible for male guests who are not closely related to the family to stay in such a separate room, but most people consider it improper for any men other than close kin to stay right inside the house, where they are likely to meet the women all the time. If there is no guest room, then there is always the mosque to stay in. Another proverb says, ironically, ‘The house is yours, but the mosque is warmer for you.’2 33
34
Shaikhdoms of Eastern Arabia
For anyone who has travelled in the Middle East outside the modern parts of the great cities, this arrangement of houses falls within a familiar pattern characteristic not of the Gulf alone but of the traditional way of life throughout Islamic society. It expresses and puts into effect the principle of the modest seclusion of women, who should not mix socially with any men except their husbands and their very closest kin and affines – those with whom marriage would not be permitted. In considering relations between the sexes in the Trucial States and the effect that the seclusion of women has on the social order, we should bear in mind, therefore, that the seclusion of women is a part of local circumstances but cannot necessarily be accounted for in terms of these local circumstances alone. Its consequences are easier to define than its causes. Nevertheless, although seclusion is a widespread culture trait, it can still be looked at in terms of a smaller scale local structure of relations. Local circumstances do suggest various observations bearing upon the general sociological problem of why, how and with what results a society keeps its women in purdah. But in conformity with the general orientation of this book, I shall dwell mainly on the political aspects of the matter when seeking to reach conclusions. It is not necessary to dwell on the absurdity of popular representations of the Middle Eastern hareem where lascivious and jealous men keep collections of slave girls. Even for those sultans and shaikhs who did keep large numbers of concubines, there were still mothers, aunts, sisters, daughters and wives, all of whom grew old. When we speak of the society segregating its women we should not envisage some theatrical seraglio in which the facts of kinship and family life have no place. Rather we must think in terms of a segregation enforced not by men alone but by both sexes, in accordance with a common standard of morality and respectability. Though men, in general, have more authority than women in this segregated society, the standard is applied by women too. In the Trucial States, a woman is taught from before the age of puberty to ‘be ashamed’, tastahi, in the presence of men who are not closely related to her, and she is expected to be unwilling to come into contact with them. A girl whose curiosity is stronger than her shyness may perhaps get a caning from her father: she will certainly be rebuked by her mother and aunts. Girls begin to play at wearing veils at a very early age and wear them consistently from adolescence onwards. For most of the day, even inside the house, she wears a black cloth mask over her face and a veil folded over the back of her head and coming down below the waist, perhaps because there may be other buildings which overlook the courtyard or because men
Women and Men
35
other than the husband may come in. When she goes out of doors, she puts the veil down over her already masked face if there are men about. In the villages, when a stranger passes, women turn, thickly veiled as they are, to face the wall, though some are inquisitive enough to glance surreptitiously over their shoulders to see what they can make of the stranger. Of course it might be said that when a woman turns her face to the wall she is doing something quite different in a sense from ignoring the presence of a stranger – she is in a way acknowledging his presence. But the fact that women only turn to the wall when caught in the middle of the street and, given the opportunity, prefer to run down a side street shows clearly enough, if it needs to be shown, that turning to the wall is a gesture of avoidance. Such extreme reticence is no part of the manners of bedouin women – even of their adolescent girls. The women of the desert will readily come up to a total stranger, greet him and ask him what the news is. Though veiled, often masked like the women of the settled people, bedouin women are not expected to act shyly in the presence of stangers. An even more striking difference between the bedouin and the people of the towns is the behaviour bedouin permit their girls, who are allowed at celebrations to dance in public unmasked and sensously holding locks of their long hair in their teeth. Among the settled people, girls may only dance for their parents, for women of the family and for their husbands with no strangers present. For women of the town to dance in public is an occasion of high indecency, more so indeed than male dances of a clearly homosexual character. Bedouin girls and women dance in the open, whereas the dances performed by some women in the towns take place in secrecy, for fear of scandal and possible punishment. In these circumstances, one might expect townsmen and villagers, proud of the seclusion of their own womenfolk, to disparage the moral character of bedouin women and girls, or at least to criticise them for impropriety, but I have never known this to happen. Indeed, in general statements the contrary is more likely to be said. In the desert, there would be no point in travellers searching out the bedouin men whilst they are out with their herds. They make for the tents, where they can hope for refreshment and security, and here, if the husband is out in the desert, the wife must take his place as host. The honour of the tent demands that the traveller should be cared for, and the wife cannot care for him by remaining shy and secluded. Townsmen do say that bedouin ways are different from their own and that they could not permit their own women to copy bedouin ways even if the women wished to do so.
36
Shaikhdoms of Eastern Arabia
Fig. 2.1 Bedouin mother and her child. From the author’s photographic collection.
But at the same time it is common to hear the women of the bedouin praised in the towns for the strictness of their sexual morals. Far from considering them to be laxer than their own women, townsmen tend to hold up the women of the bedouin as models of propriety and to speak with admiration of the strict code of morality which the bedouin are said, vigorously and remorselessly, to enforce. But bedouin themselves joke about sexual matters much more readily than townsmen. The bedouin version of the puzzle about the wolf, the sheep and the bundle of hay is of a husband trying to flee from danger with his wife and two other men. He has to work out how to ferry them across a creek in a boat that will hold only three people, knowing what will happen to his wife if she is left alone with either of the men.
Women and Men
37
Although conservative-minded townsmen may commend the strict morality of bedouin women, they by no means regard their social manners as a model for their own women’s conduct. Among the qualifications for a family’s prestige in the towns are putative desert origins and the strict seclusion of women, two conditions that might superficially seem incongruous. But in this case it is a remote desert background that enhances status rather than a recent, and therefore more certain one. The former is a qualification that guarantees pure Arabness without detracting from the rights of established wealth and long settlement in the town community. So it could happen in Kuwait, that when Shaikh Mubarak rashly insulted the great pearl merchant Hilal Al-Mutairi he chose to call him a rough and lowly Mutairi tribesman who had come to Kuwait with nothing and made a fortune. For all the prestige attached to desert origins, townspeople would not enhance their status by copying conventions which are purely those of the desert. In the case of the town family, a putative desert origin is important as a token which helps to indicate a social category rather than a substantiated fact. As time makes it more remote from the fact, the value of the token increases, for it can then be more unambiguously applied to the present social reality in which its true significance resides. At the same time, however, the token could hardly retain its social force if it were not regarded as being the fact that it seems to be, rather than the symbol of something else, which it is. It is this which renders it adaptable when there is pressure towards a change in conventions. Thus, in the present context, those men of the modern Trucial Coast who regard the seclusion of their own women as something less satisfactory than the more emancipated circumstances they see abroad need not quote the foreign cases which provoke the comparison. They can quote instead the bedouin example as proving that seclusion of women is not a ‘truly’ Arab custom. In the arrangement of accommodation for social use, the bedouin have a good deal in common with townsmen, even if protocol concerning women is less strict. When men gather together to talk, they do so in a special part of the tent where women do not join them. The bedouin women gather together in the part of the tent where cooking is done. Out of ‘shame’ they eat apart from men, even their husbands and their sons, just as women in the towns usually do. Only young children are unaffected by this separation of the sexes at meal times. But when men are gathered together to talk or eat in the tent, then what divides them from the women in the kitchen quarters is only a tent strip and not a wall. The women can peep through this and can call out greetings
38
Shaikhdoms of Eastern Arabia
and, in less formal circumstances, remarks and comments, so that the division of social space is not so much a division of social contact as it is in the houses of richer people in the towns. On the other hand, many of the town houses of the poor are only huts made of date sticks, little more substantial than the tents of the desert. Overcrowding makes it too inconvenient for women to keep utterly apart from men visitors except on formal occasions. The strictest form of seclusion of women in traditional society goes therefore with prosperity and high status. This is why the convention can break down with such surprising speed as it has done in Kuwait under the impact of great wealth, because it is the rich who are most exposed to the influence of education, travel and foreign models. And it should not be supposed that the rather greater freedom of bedouin women than that enjoyed by women in towns and villages masks any approach to social equality with men. Most bedouin are poor and undernourished, whether they be men or women, but the best of the food goes to the men, and after them to the boys. At formal bedouin meals, the whole slaughtered animal set on a great dish of rice is first served to the men. When the men have got up from the meal, the boys take their places, and when the boys have finished, the messy remains of rice, gravy and bones go to the women. Since the animal is cooked whole, there is no opportunity of keeping any of it back, excepting the offal, for the women and small children. Within the bedouin system of values, their ideology, as one may call it, the men are fighting men and must be strong and hardy as a part of the aggresive masculinity that they expect of themselves and the ideology expects of them. And it is their ‘strength’ that produces strong and numerous children. In the division of labour between the sexes, the work that requires the greatest physical strength and perhaps courage, such as dealing with angry camels or lifting the heaviest weights, is the province of men. But short of these, regular and continuous physical effort of the sort that produces extreme fatigue is the province of the women. In the desert life, women not only cook, weave and look after the children, but they also help to water the goats and camels, and it is quite normal for a woman to get out looking for a camel that has strayed. Bedouin women who come to the doctors for medical treatment are found much more commonly to be suffering, apart from other complaints, from exhaustion than are bedouin men. It is well known amongst the settled people that the standards of veiling and seclusion practised by their own women are more extreme than the Koran and the religious law demand. The Koran instructs
Women and Men
39
women to hide their ornaments and behave modestly in the presence of men who are not close members of the family, and, in addition, the religious law requires them to conceal their hair and their necks and their bodies down to the forearms and ankles. It says nothing of concealing the face, let alone of the hiding away of the whole person inside the house. This is acknowledged to be custom, not binding on religious grounds. Even in the towns, the custom varies to some extent according to practical possibilities. In the tents of the desert it would, of course, be quite impossible practically for women to live in total seclusion even were it thought desirable that they should. The practical circumstances of women’s life in the desert are, in other ways too, different from the circumstances of women in settled places. In this connection the question of going in search of a lost camel illustrates the practical fact bearing on social life and moral possibilities. The camel leaves a record of its movement in the tracks it makes in the sand. For purely practical reasons, anyone living in the desert must develop a precise and analytical eye to the varieties of tracks which he or she finds. The tracks of snakes and scorpions tell one where not to tread or sit. From the way in which the wind has blurred the tracks it is possible to assess how recently they were made, which is a matter of equal practical value. This perception of the age of tracks is also indispensable when people are hunting. With camels, the families who owns them can tell their individual tracks from all others, and this they need to do, because the tracks of a camel that has strayed may pass over stretches of sand where many other camels have been wandering about, producing a situation utterly confusing to any inexperienced eye but usually plainly read by people of experience. Thus, for the bedouin, the desert sand preserves an easily interpreted record of what has passed across it during a period of two or three days, and the record is naturally as good for human movements as for the movements of other creatures. A woman may wander about, but in case of suspicion it is not difficult to tell where she has been and whether she spent all her time alone. More important than this, however, is the effect of the social groupings of people in the desert on the circumstances of women’s social life. Throughout most of the year, groups of kin are moving about together, usually at some distance from other such groups. In the summer when the time comes for people to concentrate round the permanent wells, those groups of kin gather with other groups from their own tribal section. Thus most bedouin women for most of the time are living among people related either to them or to their husbands, and often to both, since marriage with cousins is held in high favour. Their social
40
Shaikhdoms of Eastern Arabia
milieu is that of kinship, and tracts of territory divide them from regular contact with strangers, whether the strangers be men or women – and the latter, as will be seen, are highly significant in questions of sexual morality – more effectively than do the walls of houses among the settled people. Apart from exceptional occasions of the arrival of guests or of visiting the towns, the regular social circle of a bedouin woman is made up of people who are likely to be closely concerned with her honour. Her honour and theirs are one, because bedouin women and the men with whom they are in regular contact partake in the same whole of honour and reputation. Hence, should there be any individual inclination to deviate from strict mores, there is a very strong pressure of social control to counteract it. It may also be true that any such deviation that occurs is more easily hushed up, but this is a matter which, in the nature of the case, one can do little more than guess at. Even the possibility is not readily admitted amongst bedouin or townsmen, though some bedouin have told me that if a girl bears an illegitimate child, though she may be killed, it is more likely that only the child will be killed and she will get off with a beating. The difference in practical circumstances that makes a contrast between the social condition of bedouin women and the life of women who live in villages and towns is illustrated by the problem of fetching water from the wells for use in the houses of settled people. Perhaps because of the association of water with cooking and washing, it is thought to be undignified and indeed shameful for men to fetch the water the family needs. In some towns piped water can now be provided. Otherwise, poor immigrant men, usually Baluchis,3 bring round water in paraffin tins carried on the backs of donkeys. Water at the wells is invariably free, and poorer people in the towns who cannot afford to pay for the transport collect their own, as do most families in villages. But it is very rare for a girl to be allowed to go by herself to fetch water. Usually women and girls of the family go out together, carrying their pots or tins on their heads, and chaperone each other. People say that there is no knowing whom a woman may meet when she goes to the well. The trouble is not merely that she might come into contact with men on the way – men who are not in this case, as with the bedouin, kinsmen or at the very least fellow tribesmen, but strangers. All the diverse families of the village or the town have free and equal access to the wells. Some of the other women collecting water may well be women of laxer morals or even prostitutes or procuresses on the lookout for girls whom they can tempt or to whom they can pass on messages entrusted to them by admirers. And should this not, in fact,
Women and Men
41
have happened when one woman who is either little known or little respected speaks to another at the well, gossip can invent it. The reputation of the women of a family is one important index of family standing, and so we may observe here, even in small villages, an element of class differentiation entering the situation. The wells are invariably focuses of scandal. They are the only free or neutral places where women regularly meet together and no one can be excluded. The contrast of practical social circumstances goes a good way to explaining the differences of behaviour between bedouin women and the women of the settled people. Bedouin women spend almost all their time in the company of their nearer or more remote kin and, when they marry, the kin of their husbands, whereas in spite of their domestic seclusion the women of the settled people much more commonly come into contact with strangers – if not strange men, at least strange women. It is clear that, whatever might seem superficially to be the case, the greater reticence of most settled women does not express a total contrast in social attitude from that of the women of the desert. When bedouin women get married to men in the towns, then they readily adopt the behaviour thought appropriate to women of the settled people. Even when visiting the towns, bedouin women speak to strangers less readily than they do in the desert, and usually go around in the company of their husbands. Their relative freedom of manners belongs to the social environment of the desert, in which they are secure among kin. Townsmen who have divorced bedouin wives do not complain that they are incorrigibly gregarious. They are much more likely to complain that bedouin women are always giving away their husbands’ property to their fathers, brothers, cousins and fellow tribesmen. It is here that the difference in values resides, suggesting a more strategic point of departure for comparative comment. At the root of the difference is the way that members of the community conceive of, and need to conceive of, their relations with each other. The members of any bedouin group would be an easy prey to human enemies and a harsh environment were they not bound together by very strong ties of social and political solidarity. The bedouin ethos sets a high value on the redistribution of resources, conceived of in terms of generosity and the right to expect and demand help from each other. A bedouin who is generous to her cousins does no more than the bedouin ethos requires of her. Private resources are only private in a very limited way: they are subject to many and strong communal demands. The solidarity which produces these demands and is reinforced by them is expressed in terms of kinship and shared agnatic
42
Shaikhdoms of Eastern Arabia
descent. What may be interpreted from the outside as an adaptation to environment conducive to the more secure exploitation of limited resources is expressed within bedouin society in terms not relative or arbitrary but absolute – terms of value, virtue and status. All members of the group are, in a sense, ‘cousins’, and as such identified with each other. The dishonouring of any member of the group, man or woman, affects everyone’s honour, and so the sexual honour of the woman concerns the group as a whole. For settled people (other than those who live in small oasis villages and whose ethos remains close to that of the bedouin with whom they are closely connected) the situation is very different. Most settled communities contain bodies of people who suppose themselves to come from different tribal groups. In addition, settled communities readily incorporate individuals and little families, Arab, Persian and Indian, who for their own private reasons settle in one place or another and form accretions to the population of long-settled groups. There is no sense whatsoever in which the members of a settled community need to regard themselves as ‘cousins’ together. In spite of common interests that they share, they see themselves as being basically divided by their origins, a principle expressed in the words al-asl wa-]l-fasl, ‘origin and separation’. Cousinship is an expression of, and a justification for, a sort of moral unity, and the lack of it means that members of a settled community are not fully identified with each other as are the members of a group of bedouin. The segments of the community are internally united in terms of cousinship, but no such terms of status unite these segments with each other. Hence, the members of different segments do not share a common involvement in honour and reputation, except in as far as a tradition of intermarriage between one segment and another has converted terms of contract into terms of status. Thus, in relation to the present issue of the honour of women, it is possible in a settled community for a woman to throw away her honour to the extreme point of becoming a prostitute without bringing shame to most of the community’s other members. The shame only involves the segment or segments of the community with whom she shares a status relationship. One woman may be a member of an agnatic group who will kill her for a much less extreme breach of the code of sexual morals. Another may come from another place and have no local agnates at all, or she may be a member of some disreputable little group who do not care what she does and are doing the same thing themselves. In the latter case, no one is likely to prevent or punish her unless people decide
Women and Men
43
that her behaviour is a scandal and expel her from the community. The scandal is to their religious conscience: their honour is not impunged. And in religious terms the matter is equivocal because, on the one hand, there exists the doctrine that a good Muslim must prevent evil as far as he possibly can, whilst, on the other, is the principle that he should not pry into the houses of his neighbours. The moral reputation of a woman is the concern of her father’s family throughout her life. This jealous regard for the reputation of the women of the family is closely connected with prestige. If a woman is unchaste, the main social burden of shame does not lie upon her husband. The reproach of cuckoldry, as it appears in so many Shakespearean jokes, was part of a system within which divorce was impossible and a husband took full responsibility for his wife from the time of marriage onward. In the Trucial Coast, if women are not chaste the main scandal descends not on their husbands, who can divorce them at will and without any public process which requires the giving of reasons or the proof of justification, but on the women’s own agnatic kin, to whom when divorced they must return. Expressions of abuse give some indication of the pattern of values in which they are current. There is no word of abuse current in the Trucial States, nor I believe in the Arab world in general, which implies that a man’s wife is a whore. On the other hand, three of the most common insulting expressions, if they are taken literally (which, like most such expressions, they do not usually have to be), imply respectively that a man’s mother is a whore, that his sister is a whore and he does not mind, and that he is a pimp for his sister. And if we turn from the profane aspect of the question to the sphere of religion, we find a conservative opinion about women’s literacy which, oddly unpractical as it is, speaks for itself on the matter. This opinion, which one used quite commonly to hear expressed by otherwise sensible people, was to the effect that it was a good thing to teach women to read but a bad thing to teach them to write. I do not know whether anyone ever actually tried to separate the two skills, but the justification given for this conservative opinion was that if women could read, they would be able to read the Koran and books that were morally good for them, but writing might tempt them to carry on correspondence with lovers. In practice, it was very uncommon indeed for women to be taught to read or write until very recently, which meant that even if a husband wrote to his wife from abroad, the wife had to get this private letter read to her by her father or brother. When girls’ schools started, more parents than people had ever expected proved ready to let their daughters attend them. It seemed that the
44
Shaikhdoms of Eastern Arabia
reason for this was practical in terms of marriage rather than in more general eductional terms. Once some girls went to school, their marriage prospects were improved because educated young men did not want illiterate wives who lacked their own knowledge of cosmopolitan values and sophistication. Other parents did not want their own daughters to be left behind and downgraded in such an important matter. In this way an important motive again was family prestige expressed again, though in a different guise, in terms of the marriage prospects of girls, since it is important to the status of the whole family that a girl should marry either her social equal or her social superior. Marriages are arranged between families rather than between individuals. They affect the families as wholes, and girls have little and sometimes no choice in who their husband is to be. Educated girls nowadays, being among other things older, are a little freer – but not much. Young men have rather more power than girls to express their own will, but even they cannot choose very easily, since if they have any idea of their own about the girl suggested it is probably no more than a memory of what she was like as a child. It is an irony that, in a society in which the honour of its women is of such great significance to the family, the status of women in general should be low in comparison with that of men. Of course what men, or some men, say about women in general would horrify and anger them if it were said to them specifically about the women of their own families or the families of their friends. General statements therefore should not be taken as being more serious than their context warrants. Another result of segregation is the dichotomy between the general and the particular, a social representation which affects the position of the young wife who comes into the family from the outside. One hears men characterize women as being easily tempted and eager to tempt. When speaking generally, some men will say that women have a devil of temptation at their nose ends and that only one in a thousand goes to paradise. A common piece of doggerel, ascribed, as it happened, to a religious teacher, invokes blessings on the Prophet and then goes on to say that women have neither sense nor religion and are the ropes whereby devils drag men down to hell. There is a story, not so much canonical as told as a joke, of how the Prophet is supposed to have been asked by the women of his followers to give them a leader of their own. He told them to choose the most trustworthy woman amongst them and gave her a box to keep for him. When she brought it back the Prophet opened it and showed the people that it was empty. He explained that he had put a pigeon in the box, knowing that no woman
Women and Men
45
could be trusted to conquer her inquisitiveness and that it would fly away when she looked inside. Only very rarely have I heard jokes about particular women in the towns, but with their usual ribaldry some young bedouin guards of the shaikhs compose indecent satires about particular girls and recite them to their friends. The subjects of such satires are not likely to have been respectable girls in the first place, otherwise the incidents would not have occurred and the girls would not have been known. As far as I could discover, there is very little counterpart among women to such masculine disparagements of themselves. There are one or two proverbs: ‘For lack of men, she took little Willy for a husband’ and, on divorce and remarriage, ‘One donkey takes the place of another’. Such details have to be thought of in the context of jokes and chatter in social groups of men. It would be wrong to set much store by such details. On the other hand, a positive religious belief which cannot be set aside is that a women is of half the value of a man. The statement that a woman is equal to half the value of a man, or, as it is more commonly put, that a man is doubly ‘better than’ or ‘preferred to a woman’, 4 is advanced as a fact of revealed religion, and hence as a general, and indeed absolute, principle which is not open even to empirical or logical questioning. Indeed, when I have tried to argue that there must, at least, be exceptions to this statement, mentioning some insignificant men and some remarkable and famous women of the Trucial Coast to whom men applied for guidance, help and even leadership, I have been rebuked. I was firmly told that the people of the Trucial Coast hold this belief because it is a thing laid down by Allah and the Prophet Muhammed, and, moreover, they hold a further belief, namely, that what is laid down by Allah and the Prophet Muhammed is not to be questioned. The doctrine that a man has twice the value of a woman has perhaps the same logical status as the Christian (or political) belief that all human beings are equal; and to argue about it brings into discussion so many questions of society and of the self that people may find it either confusing or boring. I am by no means certain that the principle has been laid down either by the Koran or by what is reputably attributed to the Prophet in the hadith5 literature as an absolute pronouncement on human affairs. The evidence adduced to prove the principle is taken from a relatively small number of facts of the religious law of inheritance and the law of evidence. The Koran tells men to ask the advice of women but not necessarily to take it, a phrase which one Arab feminist in the Gulf said was interpreted by men to mean, ‘Ask their advice and do the opposite.’
46
Shaikhdoms of Eastern Arabia
But most people regard ironical interpretations of the Koran to be in bad taste. The facts seriously adduced in relation to the law are purely practical. In general principle (established on a strong Koranic base) when a man and a woman are in equivalent relations of kinship to a dead person, the man inherits more than the woman does in the proportion of two to one. The religious law is not, in fact, entirely uniform in this respect, and with some cases, particularly where the categories of kin upon whom the major part of the inheritance normally devolves are not represented, can become very complicated indeed. But even in relatively simple circumstances where a person who has children and whose parents are still alive dies and leaves an estate to inherit, then the father and the mother inherit equally, each taking a sixth of the estate. Nevertheless the principle that a man’s share of an estate in inheritance is double the woman’s is overwhelmingly established on the Koranic statement ‘God charges, concerning your children: to the male the like of the portion of two females’,6 and the Koran itself gives some consistent details of this. When his father or his mother dies, their son inherits twice as much of the estate as does their daughter, and if there are several sons and daughters the estate is so divided that each of the former inherits double the share of each of the latter. When, in the more complicated – but far from uncommon – cases of the law a sister inherits from her brother (within the technical category of fara]id), then the fact that she inherits draws in her brother automatically as an inheritor, and he has a right to twice the share that she receives. It might be supposed that this principle arises from the greater responsibility of the men, who have to look after the women of the family. But the matter is not so simple, for the basic principle and authoritative moral justification of that category of inheritance called fara ]id is held to be the charitable care of women. Moreover, a widower inherits from his dead wife twice the proportion of the estate that a widow inherits from her dead husband. When there are no children, then the shares of widower and widow are increased, but in both circumstances they are doubled. Here it could be argued that in general a man, particularly a helpless old man, has qua man responsibilities that increase his needs, whilst a woman, however helpless, merely has needs. But where society makes it very difficult for a woman to earn a living for and by herself, there are many cases of the working of the law where women come out at a great disadvantage. Thus, the implications of the law of inheritance could be argued in more ways than one. The law of evidence is quite plain in its implications and raises none of these questions of need, responsibility and interpretation. The law of evidence draws a clear distinction between
Women and Men
47
men and women. If witnesses are disagreed, then the testimony of two women is equivalent to the testimony of one man. When we turn to the general moral position of women within their families, though we see this formal principle implied in action, and more particularly in the exercise of authority, a situation of a reverse sort also appears. This for women is a situation of much greater strength; a strength accentuated, perhaps, by the formal weakness of their position in relation to financial or social independence. Though the religious law places women on a lower scale than men in principle, the women in a family are usually said to be more strict and tenacious of religious observances and attitudes than are the men, if any contrast is to be made at all. Women’s religious observances take place in the house, but they may dominate the house. At most times, modesty denies to women the practice of religious observances in public. When the time for the five daily prayers comes, men may pray in any clean place where they happen to be, whether it is on a seashore where they are mending fishing nets or on a patch of waste ground in a town, excepting, that is, for the upper classes who tend to wait for a relative privacy and defer the hour of their prayer until they reach an appropriate place. Women pray the five daily prayers at home. On Fridays, for the noon prayer when attendance in the congregation is obligatory for them, men go to the principal mosque. Women do not usually attend the mosque at all, though in Hanbali communities they may go there for the special evening devotions of the penitential month of Ramadan, when a special screen of date sticks is put up in the mosques to divide them off from the men. Men are expected to arrive at the mosque early so as not to be trying to pass through the door at the same time as the women. It is only on the two major religious feasts of the year, the feast of the Breaking of the Fast which ends Ramadan and the Feast of the Pilgrimage, the time when pilgrims who have gone to Mecca perform the sacrifice there, that the women of the Trucial Coast pray in public. No mosque is big enough to hold the total congregation, so the community gather to pray together and listen to the formal sermon of the occasion in the open. The men form up in long ranks at the front and at some distance behind them there form the ranks of the women. At most times, however, women pray at home. In some senses the house is a sacred place, to be respected as much as is the mosque.7 The phrase which is a good, and a normal thing, to say when entering upon any activity is, ‘In the name of Allah.’8 This is formally said aloud by people entering a house and in less formal circumstances it is assumed
48
Shaikhdoms of Eastern Arabia
to be said in secret or at least intended. ‘In the name of Allah’ is invariably said aloud when people begin eating, and it makes the beginning of legal contracts and of letters. The phrase should also be spoken at the beginning of sexual congress. In case of extreme danger, the house is a better place of sanctuary than the mosque, for it affords personal protection to fugitives even at a risk to the owner of the house. Since it is usually thought impolite to refer to a man’s wife directly as ‘your wife’ when speaking to him, people commonly use a periphrasis. The expression used, which can be extended to all the women of the house, is ahl al-bait, which means literally ‘the person (or people) particularly connected with the house (or home)’. As this implies, the house is not merely the sphere in which women happen to be; it is their particular sphere. Though women lack authority as compared with men, the separation of the domestic duties endows them with a good deal of control over what happens in the house. For example, if they are strict about the observance of fasting in Ramadan, as they are usually said to be, it is almost impossible for one of the men of the family to obtain water to drink or food at home should he find the fast too hard for him. The situation would involve too many embarrassments and reproaches. Nor is it possible for men to entertain anyone privately at home, any more than it is in societies when there is no segregation of women, for their men are either sitting in public or else they are in the women’s quarters, from which, in all propriety, strangers are excluded. As with many social questions, there is no possible or conceivable way of finding out or learning whether women in general are truly stricter in religious observance than men, but this is commonly said to be the case by men who are disinclined to be strict themselves. I think these men probably speak the truth, more particularly because women are deprived of so many other outlets for their energies, and cut off so much from most public concerns in which they might be active, that religious devotion may indeed be the only general activity to which they can devote themselves. A passage in that chapter of the Koran entitled ‘The Light’ begins by instructing the Prophet: And say to the believing women, that they cast down their eyes and guard their private parts, and reveal not their adornment save such as is outward; and let them cast their veils over their bosoms, and not reveal their adornment save to their husbands or their fathers, or their husbands’ fathers, or their sons, or their husbands’ sons, or
Women and Men
49
their brothers, or their brothers’ sons, or their sisters’ sons, or their women, or what their right hands own, or such men as attend them, not having sexual desire, or children who have not attained knowledge of women’s private parts; nor let them stamp their feet, so that their hidden ornament may be known. And turn all together to God, O you believers; haply so you will prosper. 9 As this passage makes clear, the modest behaviour of women should spring from their own virtue and moral responsibility. Moreover, the Koran establishes the superiority of men over women in the following passage in the chapter entitled ‘Women’: Men are the managers of the affairs of women for that God has preferred in bounty one of them over another, and for that they have expended of their property. Righteous women are therefore obedient, guarding the secret for God’s guarding. And those you fear may be rebellious admonish; banish them to their couches, and beat them. If they then obey you, look not for any way against them; God is All-high, All-great. 10 Consistent with the teaching of the Koran and sunna, there is also a contrary representation, equally general in reference but more widely distributed. Side by side with general ideas of the faults and inadequacy of women there exists the idea of a model of feminine virtue, ‘the good woman’, al-hurmat al-saliha, which even those willing to speak lightly of women readily describe. The characteristics of this ideal figure are enumerated consistently, with great respect and usually some emotion, by men of widely different characters. They express a collective representation of virtue in women, interpreting what may be experienced at home and what is generally hoped for in marriage. Among her most important characteristics is the way she looks after her husband and children and their interests, and it is interesting to see that the model is not one of total submissiveness. They care for the moral welfare of their menfolk as their menfolk care for them, though by slightly different methods. The ‘good woman’ looks after her husband so well at home and makes herself so attractive to him that he is not tempted to spend his spare time outside the house wasting his money and mixing with doubtful company in the cafés. She herself prefers not to spend too much time visiting other women in their houses. If her husband is foolish, she is clever enough to find a way of reforming him: hence, the simple story of a clever woman who married a foolish and quarrelsome
50
Shaikhdoms of Eastern Arabia
husband. After they had set up house together, she foresaw each day what he would want for dinner and was able to please him by asking what he could fancy and then showing him it already prepared. When she noticed that he could no longer be pleased and wanted to pick a quarrel with her, she reformed him: she had foreseen that he would demand a dish of faeces and had that ready prepared for him too. Though women as a class are regarded as inferior to men, as individuals they can be conceived of, whether the imagination is serious or jocular, as superior to individual men in respect to religion, character or intelligence; and on the Last Day the good and the bad of each sex will be separated and men and women will be joined together according to their virtue: Upon that day God will pay them in full their just due, and they shall know that God is the manifest of Truth. Corrupt women for corrupt men, and corrupt men for corrupt women; good women for good men, and good men for good women – these are declared quit of what they say; theirs shall be forgiveness and generous provision.11 The Islamic conception of paradise thus includes the continuance of sexual relationships. If we bring together the Christian and the Islamic ideas of this world and the next and contrast them, we find among Christians that the dogma that there are no marriages in heaven is associated with monogamy, the prohibition of divorce in the secular life and a widely held ideal of holy celibacy in the religious life. In Islamic belief sexual relationships continue in heaven, whilst in this world polygyny, concubinage and divorce are permitted and celibacy, whether the motive for it be secular or religious, is disapproved. This is not to say that polygyny, concubinage or divorce are recommended in Islam: they are merely tolerated. Both old-fashioned people in the Trucial States and those who consider themselves more progressive readily quote the doctrinal statement that, of all things which are not positively forbidden by Allah, divorce is the one most disapproved. And amongst progressives, though not much in other circles, it is argued that men should not practise polygyny because the religious condition that husbands who have more than one wife must treat them equally is an impossible one to fulfil – an example, no doubt, of the reorientation of dogma to accord with social aspirations. The possibility of legal concubinage disappears when slavery ceases to exist. But it is very strongly believed that everyone should marry; that both men and women should be married young to keep them out of temptation; and that
Women and Men
51
widowers and widows should marry again, though some widows may justifiably refuse to do so for the sake of their children. In current usage, a man begins to be called a man, rayyal, rather than a boy, walad or subay, and a women begins to be called a woman, hurma, when they are married. There is, therefore, a sense in which men, like women, need marriage and parenthood before they can achieve full adult status. But the degree of parallelism between men’s and women’s positions is limited. The word just quoted that means ‘a man’ stands absolutely, whereas the word for ‘a woman’, hurma, only has meaning in relation to men – an instance which defies classification according to any jejune dualistic system. The word rayyal, ‘a man’, is connected with the idea of manliness, and if one says huwa mu bi-rayyal (lit. ‘he is not a man’) it means ‘he is a rotter’, relating to some context in which the subject lacks a virtue such as honesty, generosity, integrity or, less commonly, courage. Hurma, ‘a woman’ is, both philologically and in the popular mind, one of a group of words connected with the idea of what is forbidden in the divine dispensation. It connects with haram, which is the word defining something forbidden by Allah, be it fornication, theft, murder or the drinking of wine, and with harim, a collective meaning ‘women’, used also in some places for the women’s private apartments and naturalized into English in the form ‘hareem’. Hence hurma is haram in the sense of being forbidden to men, and so, far from being absolute, the word takes its meaning from a relativity.12 The prohibition applies even in death. This is not to say that men and women are buried in different parts of cemeteries. They are not: there is no question of segregating the dead from the dead. But the funeral of a woman is markedly different from that of a man. Whereas it is held meritorious for any man to assist in carrying the body of a dead man to the grave, a woman’s bier should only be carried by those men whom she did not have to avoid in life. Her body is wrapped in an extra shroud, and it is further concealed from public view by a cloth stretched over the bier on its way to the cemetery. At the grave itself the body is committed to the ground under the cover of a canopy. The reason given for this difference between the treatment of the dead bodies of women and men is specifically that the body of a woman is haram, in the sense here described. We have now reached a stage in the discussion where we can try to sum up some of the differences and the extent of complementarity between what is represented as appropriate to women and to men in Trucial Coast society, taking the representations at their widest and
52
Shaikhdoms of Eastern Arabia
their least personal. We have seen how women are represented as subordinate to men and of half their value according to the divine dispensation – according, that is, to what people regard, within the same representation, as the very nature of rightly-lived human life and rightly-organized society. The modesty with which virtuous women are endowed is consistent with the role which rightly organized society accords them: modesty causes them to avoid all but the very limited circle of men closely related to them and hence related to each other. Men’s life may be bound up in home and family, but women’s life is almost entirely bounded by them. Men have their life both within the family and outside it, whereas there is no public sphere for women, even among themselves. Hence the life of men and the life of women do not form two equal and complementary spheres, any more than do members of the two sexes have equal value. As exemplified in the congregation of the faithful at the Friday prayer, when the men of a community must not only attend a mosque but attend the same mosque, and as appears in secular affairs in the markets or in the majlis, public meeting, of the bedouin, there exists a general, public and inclusive society of men which lacks any feminine counterpart. The mosque and the market are inclusive, whereas it is in the nature of the home to exclude. Whether in Trucial Coast society or in other societies across the world, the home provides not only physical but also social shelter. In the Trucial Coast, we see such social shelter taking an extreme form in association with women’s seclusion. As well as excluding men, the home excludes most women. Let us consider a few more details in the separate social lives of men and women in the Trucial Coast. For men, the market plays a greater part in social entertainment than it does even for women in Western society. Men spend a good part of their daylight leisure there. In shops and offices, well supplied with seats, business is mixed with conversation and social calls, and visitors are served with tea, coffee or soft drinks. In addition there are the cafés. Whether in shop or café, people are sitting in places open to the street, being greeted and joined by passersby. The arrangement results in ready accessibility not only between equals but also between men of widely separated social status, though the introduction of air conditioning is now changing the situation and threatens to increase social divisions. But within the prevailing social accessibility, social divisions are a part of traditional town circumstances, as becomes apparent when one compares the men one sees sitting in shops with those one finds in cafés. For most young men it makes little difference whether they sit in a shop or in a café, except that
Women and Men
53
in a shop their conversation is likely to be more constrained because of the presence of older men. But to find older men of high status in cafés is most unusual. The older men one finds in cafés are people of fairly low social standing. The café is, so to speak, neutral ground where conversation is unrestrained and, for some, over familiar. Middle-aged and elderly men who do not wish to mix on the café’s terms with all and sundry usually choose more personal places in which to sit – shops which, though beside the public thoroughfare, offer the relationship of host and guest. Visitors who sit down in shops are guests whose company the proprietor welcomes or is, at least, prepared to tolerate. Others come to conduct their business, whether to buy or to beg, without being invited to sit and join in the conversation. If some inquisitive person stands about outside the shop listening, he is asked, politely but without any nervous hesitation, to move on. Thus the society of the market is public but at the same time differentiated. Women have little social life outside the privacy of houses. Only rarely does one see a woman sitting in a shop, and then she will be sitting apart with her husband making special purchases, shopping in general being men’s business. One would never see a woman sitting in a Trucial Coast café – not even if she were the most brazen of prostitutes. In contrast with the element of public display in the wearing of jewellery in western society, women here wear their ornaments for displaying in private to their husbands, and part of their more valuable jewellery, if they are rich enough, is a gold belt worn around the waist underneath the clothing. In practice, however, the Koranic precept on concealing ornaments is not observed with absolute strictness. Gold necklaces and bracelets may be displayed, or, at least, negligently hidden. Nevertheless, out of doors and in the presence of men who are not close kin, women show a sombre exterior. Their masks and veils and their outer robes are all black, though the robe itself has an edge adorned with a patterning of gold thread, as do the robes of men. Below the robes, one sees the lower parts of trousers made from patterned cloth, finished off at the ankles with silver embroidery, and similar bright patterns can be seen underneath the veiling when it is disarranged a little in movement or blown by the wind. From such glimpses and from the types of cloth on sale in the shops, it is clear that the clothes women wear underneath their black covering are far from sombre, but the attractions of the face and figure and of clothing and jewellery are not for all men to look at. The Arab husband does not display his wife before other men or use her as a vehicle for displaying his wealth to them. A woman’s pretty clothes and her finery may be
54
Shaikhdoms of Eastern Arabia
seen by other women, but beyond that they are for the private pleasure of her husband. One woman, usually the wife (a senior wife, if he has more than one) of the master of the house is the ‘mistress’ or ‘guardian’ of the house.13 She is in charge of household expenditure and of the preparation of the food. Among poorer people it is very common for the mistress of the house to be in charge of the family money. Her husband and her sons bring most of the money they earn to her, and she keeps it in a special box, which is a part of her personal property, and dispenses it to them when they need it. On the whole, women do not do the shopping for the family: their modesty keeps them out of the market, and shopping is done by the men. But, as the guardianship of the family money and the supervision, if not control, of family expenditure suggests, senior women are not in such a humble position of tutelage as might at first be supposed. Living so much in the company of other women, they learn a great deal more than their husbands usually do about family situations, particularly since the social proprieties of what can be discussed are less of a restraint on them than on men. It is, in general, improper for men to speak about women, whether those of their own families or of other peoples’, and it would be difficult to discuss family affairs without mentioning the women. Hence men tend to learn more about private affairs from their womenfolk than they do from other men. The total income of a family is not kept as an undifferentiated whole. Rich or poor, it is normal for women who earn or possess money of their own to keep it separate from the money of the household. It is considered to be the duty of a husband to support his wife without any financial assistance from her, even if she has the means to help him. A woman with private income of her own is expected to keep it or spend it on her children and not to use it merely as a supplement to the income of her husband. This does not mean, of course, that no woman ever helps her husband financially. To suppose this would be absurd, but the assumption of the separation of assets is a significant social fact, particularly when considered in the context of men’s behaviour in the family. So long as sons, whether before or after marriage, continue to live in their father’s house they do not separate their income from his, and when brothers live together after their father’s death they share the household expenses. The normal expectation that women do not need to mix up their own money with their husbands’ is not so much connected with any idea of the independence of women – for in fact they have very little independence – but with the possibility of divorce.
Women and Men
55
Though, in practice, many men remain faithfully married to the same women all their lives, the power to divorce their wives at will, without proving or even offering any justification, is one of their basic rights. All a man needs to do is to announce to his wife that he divorces her and to pay the remainder of the bridewealth if there is any owing, and then, whatever she may have to suffer, a woman has no appeal and no right to any alimony or further support. If the divorced wife is keeping the husband’s money in her box, then she returns the money to him and takes only the empty box away. Just as a wife’s property is separate from her husband’s, so is the property of the husband quite strictly defined, even including some of the things which are in his wife’s hands. Jewellery is a case in point revealing the social distinction between a wife and her husband. Some of a woman’s jewellery is her own absolutely. With the rest of it – and these things do not need to be family heirlooms – she is the bailee, wearing the jewellery for her own and her husband’s pleasure so long as the marriage lasts. Unless the jewellery a woman has from her husband is either part of her bridewealth or has been specifically given to her as a formal gift, then it remains her husband’s property. If he divorces her she must return it to him. Unless he has divorced his wife three times, a man is able to ask her to resume the marriage after he has divorced her. In this case, he must try to persuade her to return by giving her a present, often a substantial one. Since it is possible for a man to speak the words of divorce in a rage without having thought well about the matter, such resumptions of marriage are by no means uncommon. A wife cannot divorce her husband, though a divorce may be obtained for her by her family on a very limited range of grounds which have to be proved in a legal action. The grounds are as follows: that the husband is failing to support her; [and that he has been cruel to her or he has deserted her]. There are no other grounds for actually forcing the divorce upon the husband, though he may be persuaded to divorce the wife according to a form whereby he himself receives compensation or part of his bridewealth back should the wife and her family be anxious for the separation to occur. It is also possible here for the rulers to apply pressure. I was present one day when the Ruler of Ras Al-Khaimah was returning from the mosque to his majlis.14 He was pursued by an angry woman who wanted to divorce her husband because he had married a second wife. The husband, who was present, did not want to do so, and to add to the complication, the woman was pregnant. When he learnt that the woman’s family were in favour of the divorce and were willing to return
56
Shaikhdoms of Eastern Arabia
the bridewealth, the Ruler simply told the husband that there was no point in remaining unhappily married and that he should accept the family’s offer and divorce his wife there and then in front of witnesses. And so the husband divorced his wife on the spot. He would have had to be very stubborn not to do so in the circumstances. As the whole sum of his bridewealth was not readily available, the husband was given a date garden as a pledge, on condition that he should tend it and receive half the produce until such a time as the money was paid in full. If a man marries a second wife, this is also an occasion when a husband has to give presents to his first wife to ‘please’ her or ‘make her content’, if that is possible. There is another way of procedure, not indeed legal in the strict sense but usually practical, which a woman can adopt if she wants a divorce. This is simply to leave her husband and return to her family. In such circumstances, it lies within the legal rights of the husband to insist on her family’s sending his wife back to him, and if necessary to appeal to a court, but unless the woman’s family are so displeased with her that they make it clear that they will not have her back, then she can run away again. For the husband, the result is embarrassment and public scorn, since people think it absurd that he should insist on keeping his wife if she does not want him. In this case, a social rather than a legal sanction forces the divorce. When a wife leaves her husband, this is another occasion when he may try to ‘make her content’ and persuade her to return to him by making her a present of some of his property. One curious situation which arose in Ras Al-Khaimah was of an elderly and on the whole a ferocious man, a former brigand, who had quarrelled frequently with his wife, to whom he had been married for over twenty years. She had left him several times, but on each occasion she had relented when he had appealed to her and offered her a present. In the course of their frequent quarrels, the wife had been given most of his property – his four cows, a few date palms and even the house they lived in. He felt that his wife was growing old and less attractive, and he also thought he would like to have some more children, so he proposed to his wife that she might allow him to marry a young girl and they could all live happily together. The old wife flatly refused and told him that if he married a young girl she would leave him again, and this time permanently. This he was apparently prepared to suffer. But her final ultimatum was to point out to him that if she left him on this occasion it would effectively amount to his leaving her, since she owned the cows and the palm trees and the house, and so if he wanted a new wife he had better find a house for her first.
Women and Men
57
A man is legally allowed to have up to four wives at the same time, but most men have only one, and few are able to maintain four. Divorce and remarriage are, however, quite common and there is no social stigma attached, except perhaps in the case of very rich old men who go on marrying young women, not much caring what families they come from or whether their young wives are simply hoping to get rich in a month or two and then publish their memoirs after they have been divorced. Religion, however, lays down one condition for men who keep more than one wife at once, and this is that they should treat them equally. This condition is taken as being an impossible one to fulfil by modern-minded Muslims who regard polygyny as backward, and so they argue that men should be restricted to one wife. Some men who travel a good deal keep wives and families in more than one place and visit each from time to time, but others who do not may have different houses for their wives so as to avoid strife. One old man gave a vivid description of how intolerable he had once found it when he tried to live with two wives in the same house, saying that he had spent alternate nights (following the principle of equal treatment) with the two wives but he naturally had not always felt the same. One night he would be in good spirits and spend most of the night awake enjoying the first wife’s company. The next night he would be very tired and want to go to sleep. But the second wife would keep him awake with her reproaches and complaints, telling him that she had heard him and the first wife enjoying themselves the night before and why could he not enjoy himself with her? In the society of the Gulf there is no idea that women do not or should not enjoy sexual congress with their husbands, highly as the idea of feminine modesty is valued. The precept of treating co-wives equally is well known in the society of Trucial Oman to be laid down by Islam. As a counsel of prudence, many people extend the principle to include women living together in the same household. The master of the house who has his son and daughter-in-law living with him takes care, when he buys a present for his wife, that he buys the same or an equivalent present for his daughter-in-law at the same time so as to avoid trouble and jealousy between the two women. This practice is one consciously determined on and commonly made explicit, a fact which indicates the importance of the problem as much as it indicates the importance of the remedy. The social importance attached to marriage is, in principle, the same for both sexes. Marriage is both a religious duty and a social necessity. There follows from this the need for children, a matter in which the social pressures upon men and upon women are rather different. ‘One
58
Shaikhdoms of Eastern Arabia
hand cannot clap’, is a local proverb applied to both situations, lack of a wife and lack of children, in relation to men. (I do not know whether it is used by women in relation to their own affairs.) For both men and women, the practical need of children is considerable, and this is reinforced by social pressure, so that it becomes a matter of social embarrassment to have none or even to have only daughters. Both situations may be ‘excused’, in explanations of a man or his friends, in terms of love of the wife, but they are situations where excuse or explanation is thought necessary. Gruffer people will advise a man who has had no children after being married for some years to marry again. It is of course known perfectly well that just as some women are barren so are some men impotent, but perhaps because the choice of remarriage or further marriage lies with the man, the blame for lack of children, or for the children of a marriage being only daughters, tends to be put on the woman. And, in ways which I shall discuss, the security of a woman depends more on having children than does the security of a man. In the case of women, this practical and social importance attached to the bearing of children clearly intensifies the natural devotion they feel, thereby increasing rivalries which exist already or often producing new rivalries between women whose children are anyway, whether potentially or in reality, in competition. Between co-wives, this rivalry is, as far as one can gather, very common, although there are some who get on well together. Less commonly, but by no means infrequently, the rivalry impinges on the relations of brothers. These differences apply less often in relation to the children of two men by the same wife, because not only is the formal stress on the paternal line but also, given the social circumstances, it is more likely that a woman will stay with the children of only one husband than that a husband will stay with the children of only one wife. If a wife has children, she is less likely to be divorced, and if divorced she is less likely than a man to marry again. Should she marry again, then she produces a counter pressure to that the prevailing one and compels her sons to modify their formal agnatic duty in society in favour of their mothers. Hence, differences within families are, like the lack of children, frequently, and no doubt often rightly, blamed on the character or nature of women, without taking into account the pressures through which the society so strongly influences the orientation of these moral circumstances. If a woman is divorced by her husband, she veils herself from him, packs her belongings, and returns to whoever is her legal guardian. This is her father, if he is still alive, or otherwise her brother or her father’s brother. She cannot marry again for at least forty days, during which it
Women and Men
59
will become clear from her menses whether she is pregnant by the husband who has divorced her.15 The same period must elapse before a widow can remarry, but unlike the divorced woman, the widow spends the time in seclusion and mourning, and even after this period it is not thought entirely proper for her to marry again too quickly. The position of the widower is different. For him there is no period of formal mourning, and the speed with which he marries again is said to be entirely his own business. His position is thought to be different from that of a widow in that he needs someone to run the house; but in any case it is held to be better for both the widow and the widower to remarry. Such, at least, is the general doctrine, based on religious principle. In practice, some old men and many widows who have children prefer not to marry again, and remarriage is socially rather less acceptable for widows with children than for men. There was an old fisherman whose wife died, to his great grief. He was about seventy and she had been his first, and only, wife, and so the couple must have been married for something like fifty years. He said she had been a perfect wife, looking after the house and the money, and that he had never struck her once, even with his hand, during all the years they were married. Nevertheless, within a fortnight of the funeral he was talking of marrying again and thought he would be able to marry the daughter of a shopkeeper in the village. The reason he first gave for wanting to remarry so urgently was that he could not bear having no outlet for his sexual appetites, but this was not the whole of the story. His daughter and his son-in-law had been sharing the house with him. Once the old wife died, his daughter became the mistress of the house, and the daughter was more interested in pleasing her husband than her elderly father. The daughter certainly did not want her father to marry again and bring in another woman in her mother’s place to be mistress of the house. But instead of trying to please her father who was not easy to please, she had quarrelled with him. He had then struck both her and her husband, whereupon the daughter had refused to cook his food, and so he was having to cook for himself. Had the old fisherman been rich, then the situation would have been different. He would have had a household full of servants who would have obeyed him rather than his daughter, and his social connections with other families of rich men would have been strong enough to ensure that his daughter and son-in-law treated him with adequate respect and subservience. But, as it was, when his wife died and his daughter became mistress of the house, the old man found himself being superseded as master.
60
Shaikhdoms of Eastern Arabia
The marriage with the shopkeeper’s daughter fell through. The old fisherman said he had found that the shopkeeper had no marriageable daughters. One was a young girl, and he was too old to marry a young girl, whilst he had been told that the other was an unpleasant woman. He had inquired about a third woman, but he found she was beyond childbearing, and he wanted children. I fear the fact was simply that he had been refused. Then he had a letter written to kinsmen who lived some distance away saying that they must provide him with a wife. At this time he talked a good deal about the impossibility of marrying certain women who would, he thought, have been available in the village, since they came from low families without honour, sharaf, and said, ‘If I took a girl from a family like those, what should I be doing? It would be like buying a slave to do the housework and sleep with.’ The kinsmen proved unobliging: a poor old fisherman is no great catch for a girl. The last stage of the negotiations I heard of was equally hopeless. There was a respectable girl living with her widowed mother, but she was blind. The old man proposed that he should marry the girl and the old woman should come along to do the housework and look after her daughter and him. The old woman was anxious to get her daughter married, but this was because she was tired of looking after her and the two were having quarrels already. If the daughter were to get married, she had no intention of looking after her any more, let alone her husband. None of the marriage proposals succeeded, and the old fisherman became ill and died a few months later. People unable to look after themselves must turn to their families rather than anyone else for help, and hence the position of those who have had no children or who have had them and lost them is an unhappy one. I never came across people who wanted only sons and looked on the birth of a daughter with dismay, as the Koran blames the pagan Arabs for doing,16 and as some modern books report of other countries in the Middle East. Nor did people want to have very large number of children. But sons on a whole have more power and resources to help their parents than do daughters. Since, in this society, a woman’s place is in the home to such a point that it is enormously difficult for most women to support or look after themselves independently, a woman’s need of sons is intense, whilst a barren woman may find her life disastrous. A girl is looked after by her father until she marries – traditionally at an early age, not very long after reaching puberty. If she is divorced or becomes a widow, then she must return to her father until she can marry again. So long as her father is alive, returning to her own home usually represents no difficulty. But if her
Women and Men
61
father is dead, then she must return to the house of some other man of the family, usually a brother, and for a middle-aged woman the situation then may be far from simple. Her brothers, however kindly they may feel towards her and however strongly they may regard their family duty, have wives of their own. It is usually the most senior of the brothers who takes the place of the father, and so the middle-aged widow or divorced woman is likely to find herself something of an intruder in a family in which the relationships are already well established. She is busy in another woman’s kitchen and with another woman’s children, and it must be remembered that poor people in the towns and villages live in very cramped circumstances. With sons a woman has a much stronger right than with brothers, and a daughterin-law is expected to give away to her husband’s mother and accommodate herself to her. Of course, the characters of the individual people involved in such situations have a great deal to do with what actually happens in particular cases. Life does not proceed by mechanical regularities, but what is being described here is the general pattern within which such situations tend roughly to fall. One also hears of women living perfectly happy with their brothers and sisters-in-law, and, though more rarely, of women who are not well looked after by their sons. The fact remains that within the general system of values, whilst it is thought to be a bad thing for a brother to fail to look after his sister happily, for a son to fail to look after his mother is thought very shocking indeed and despicable. Caring for the parents, and particularly the mother, is a social imperative implanted very strongly in the conscience of children from their early youth, and should children fail in their affections and their duty, then society applies the sanction of steady disapproval. To have sons is also some safeguard for a wife against divorce itself. In the first place, when a man finds he can beget no sons by his wife he may begin to think of marrying again. In the second place, if he has sons by his wife and then divorces her, the helplessness of the woman’s position draws her sons towards her and away from him. Children should, according to religious law, remain with their mothers until they reach seven years of age. But after that, when a son is old enough to earn money for himself, his first duty is to care for his mother if she is in need. Particularly among poorer people, it is expected that a young man whose mother has been divorced by his father will leave his father’s house. There is more than one motive involved. The son is likely to be indignant that his father should divorce his mother, and also to be intolerant of another wife of his father who takes his
62
Shaikhdoms of Eastern Arabia
mother’s place in the house. But the main motive lies in his mother’s helplessness. The son leaves his father’s house and goes to make a new home for his mother. This, however, is only so long as the mother does not marry again. If she does, then it is not usual for a grown-up son to stay with her and her new husband. His natural place again becomes his father’s house. But certainly fathers consider, if tempted to divorce the mother of their sons, that in doing so they may be losing their sons as well. In this sense, though one could scarcely call it more than a consolation prize, the very helplessness of women in a highly segregated society can provide them with a certain measure of security, since their inability to care for themselves alone strengthens the bond of moral duty in which their sons are bound to them. As might be supposed, it produces an extremely strong emotional dependence of mothers upon their sons, for, even in marriage, they represent a great part of the mother’s security, whilst in bereavement or divorce they represent almost the whole of it. Segregation also removes from women most potential interests outside their families and those of their friends and neighbours. The disabilities which segregation and the ideal of modesty lay upon women who might need or want (though in traditional society the latter would scarcely be conceivable) to be independent are numerous, and in relation to poverty they increase. Given intelligence and more strength of character, a woman who has some financial assets can increase them. She can buy cloth and other supplies and sell them to other women from her house, and she can speculate in land and shops or in trading voyages or the profitable, if hazardous, smuggling enterprises of the Trucial Coast. The greatest and most remarkable woman of the Trucial States in the present century was the mother of the present Ruler of Dubai, Shaikha Hussa bint Al-Murr. 17 As wife of the late Ruler of Dubai, she had an assured position, and being gifted with extraordinary abilities and strength of character she made a great fortune as well as asserting a dominating position in politics. Shaikha Hussa was unusual in that she regularly held a public majlis of her own which was attended by men – and indeed it is said to have been more crowded than her husband’s, though he was the ruler. The wives of shaikhs are often important in public affairs. Women, and sometimes men, come to them in private and ask their advice or try to persuade them to use their influence on their behalf, but for a shaikhly lady to hold a majlis ‘like a man’ is a very rare thing. 18 In addition, however, Shaikha Hussa bought and hired out many of the little ferry boats that transport people
Women and Men
63
from side to side of the Dubai creek between the two halves of the town, and she invested money in buying quite a large proportion of the shops in the Dubai market, property which became more valuable with the development of trade as Dubai grew and its neighbour Sharjah declined with the silting up of its creek and the maladministration of some of its rulers. And she made further wealth by entrusting money to merchants and boat captains to use in trade on her behalf. The use of money on the model of Shaikha Hussa is open to any prosperous woman, though a woman of shaikhly status has some advantages. A mature woman does not, if she is veiled and not alone, need to feel shame at seeing unrelated men in her own house, so long as it is on business. She may choose to speak to them through the doorway from the far side of an adjoining room, and even to have a curtain drawn across the doorway, but so long as the decencies are observed there is no possibility of scandal. In Shaikha Hussa’s time (she died about twenty years ago)19 a rich woman would have men and women slaves in the house and probably dependants from her own family. For the prosperous, domestic circumstances are such that, without her needing to leave the house, she can engage effectively in financial dealings, with men as well as women, without the slightest detriment to her reputation. Such a woman can patronize and pay men who will do work in public circumstances under her direction, and they do not even have to be men of her family. But, for the woman whose only asset is her ability to do manual work, the situation is entirely different, for, however willing she is, there is little work she can do excepting work within the family, unpaid, as a wife, mother, or dependant female relative. One of the things that a ‘good woman’ is said to do if she has some money from her bridewealth is to buy a sewing machine20 with which she can save money by sewing clothes for the family and earn a little by sewing for other people. But without a sewing machine there is very little profit in sewing. However, there is no manual work where the products are sold that is not normally done by men. It is extremely difficult for a woman to preserve her respectability if she goes to work as a servant in the house of another family. This is no matter simply of ‘bourgeois’ respectability derived from the nature of the work, but rather a matter of respectability associated with the nature of the situation between the sexes. A slave woman – there are few left now – can remain highly moral and religious living in the house of the man who owns her, but the religious law permits her master to sleep with her should he wish to do so. If she bore him a child whom he acknowledged, then that would be the legitimate child of her master.
64
Shaikhdoms of Eastern Arabia
But, low as is the status of slaves, a woman in service in the house of another family has a status even lower, for she has not even got the religious guarantee for her situation with the men of the family. The implication for the employer’s family is almost as difficult, and the result is that a poor woman is deprived of the possibility of such work both by her own feelings and the feelings of those who might employ her. When poverty descended on the Trucial States as the pearl fishing industry collapsed, and hardship was increased for most people by the Second World War, many people who found they could no longer keep their slaves granted them – more particularly the women among them – emancipation, whether they wished it or not. People in the Trucial States who were critical of the selfishness of such owners of slaves said that many of the slave women had no possible means of supporting themselves except prostitution, unless they were to starve to death. To emancipate a slave is held in Islam to be one of the greatest acts of charity, but emancipation for this motive requires of the master that he should ensure that the slave has enough means to live on. The highest respectability is not so much expected of slave women as of the free, but there were two emancipated slave women living in Ras Al-Khaimah who were said never to have been into the market in their lives. It is said that one of the ways Shaikha Hussa ensured the well-being of her slaves without financial loss during this period of general poverty was to send them as presents to princes in Saudi Arabia, receiving from that already rich country other gifts in return. The position of Shaikha Hussa was a phenomenon within the traditional system. The emancipation of this great lady was a personal matter, not directed in the slightest towards increasing the freedom of women in general or anything that could be described as feminism. Like her husband, the late Shaikh Sa[id bin Maktum, Shaikha Hussa appears to have been something of a conservative, suspicious of modern education and administrative reform. Her efforts, like his, were directed towards commercial prosperity, military security, political stability and charitable good deeds after the traditional Arab fashion. (And to these one should no doubt add looking after the interests of her husband and her children within the state and ensuring that they retained the primacy within the ruling family). Shaikha Hussa died some years before her husband, and it was only after his death and the accession of their son Shaikh Rashid bin Sa[id that modern schools, a regular court and various civil service departments were established in Dubai. Although poorer women are much less able to enjoy financial independence than richer ones (and even with the latter, the actual
Women and Men
65
situation is possible rather than usual), the domestic constriction of the poor forces upon the women a greater familiarity with unrelated male company. Several people living together in a tiny house leave little possiblity for the woman to manœuvre out of the way of visitors. Considering the strictness of the value system, poverty and the desire to save money produce anomalies such as suggest many unspoken complications in family life. Indeed, further research might establish a pattern much laxer in some ways than I am aware of, so long as that research was done by a woman, for it is scarcely possible for a man to make any methodical investigation of the circumstances of housing and who sleeps where, unless he is prepared to seem disreputable. (The situation might now be different, since people can be employed on official work in social welfare.) But two examples, neither of them scandalous, suggest to me the possible extent of the domestic problems arising from cramped circumstances. I think now that I should have tried to inquire into these circumstances more thoroughly, but they are very difficult things to ask about – more difficult than a reader might suppose. The first was a legal complaint, in at least that it was a complaint made to a ruler by the family of a wife and he had to make a decision on it and try to settle the matter. There were two brothers living in the same house. They were both recently married and had no children or dependants living with them. But they were poor and the house had only one room, so that the brothers and their respective wives all had to sleep together. The family of one of the wives complained that the situation was improper. The brothers said they would build another room in the course of time. The family that was raising objections said that the brothers owned a hut not far from their house and that it would be better if their daughter and her husband lived there. The brothers said this would be far more expensive for them because their cooking would have to be done separately. The ruler said they must use the hut, because he agreed with the complainants that the circumstances of the women at night were improper. The second case was not one which was thought to involve any impropriety. A middle-aged man and his wife had a married son and a house that comprised a sitting room and a kitchen. The father found a rather helpless youth who proved to be a nephew, and decided that he must take him to the house to stay. The whole family, apart perhaps from the mother of it, were very simple-minded, otherwise possibly there would have been trouble, but the son’s wife was no more than eighteen and moreover the whole family were strangers to the village, not even the young daughter-in-law being a local girl. The arrangement
66
Shaikhdoms of Eastern Arabia
of the rooms at night was that the father and the son took it in turns to sleep with their wives on alternate nights. The nephew always slept in the other room. But neither of the women, his uncle’s wife or his cousin’s wife, could sleep in the room where he slept, so that the father slept with his wife in the presence of his daughter-in-law and the son slept with his wife in the presence of his mother. Here again, there was talk of building a third room, but this revolved round the idea that the nephew ought to get married, a thing which would have rendered even the relative propriety of the existing arrangements impossible. I have no representative body of fact upon which to generalize, but neither of the situations mentioned was considered extraordinary in local circumstances – what can the poor do? A third case was from Bahrain. A widow had two sons, each of them married and each having a house with only one room, which contained a bed. The widow had to stay with her sons, for there was nowhere else she could go. She visited one son, and being his mother accepted the bed to sleep in. Since the bed had room for two, the son slept in it with her, whilst his wife slept on the floor. (Many people sleep on the floor and it is not thought particularly uncomfortable). But after a few months, the wife became irritated with the arrangement and quarrelled with her mother-in-law, with the result that the mother-in-law indignantly went off to her other son, where exactly the same arrangement was repeated, ending up with another quarrel and a return to the first son. This had been going on for four years when I was told of it by one of the wives. So do general standards have to be adapted to poverty? It is generally thought in the Gulf that honour and respectability are every woman’s right and duty and that wealth simply makes greater problems for people. There is no idea, as there is in the western world, that poverty itself is shameful. But still, with poverty such embarrassing difficulties cannot help but arise. In normal conversation they are not mentioned. The poor are thought to have as good a right to their privacy as the rich, and there is as yet no fashion among the latter of supposing that poor people need moral, as distinct from financial, help or that they can be shown how to maintain their poverty in a more respectable way. In the cramped circumstances of a poor man’s house, a wife can scarcely avoid seeing and being seen by people who come to see her husband and becoming acquainted with those who come to the house regularly. With such people, the shyness may wear off and the woman then starts to bring refreshments into the room herself for guests she has seen before and join in the conversation on the way in and out. Inquisitive young women may have to be told not to hang about the
Women and Men
67
room when men are visiting. In bedouin circumstances, a wife may act as host to travellers to fulfil the duties of hospitality which fall upon the tent when her husband is away and there is no grown men to act as host in his place. This is much less the case in the towns and villages where there is always somewhere else for people to go, but nevertheless women do sometimes invite familiar male guests into the house when their husbands are out and serve them with refreshments whilst they wait for the husband to come back. There is nothing shameful in this as long as the husband does not disapprove, since the duties of hospitality and politeness of the house can be said to outweigh the requirement that the woman should remain withdrawn. But some conflict of values does exist, and this is expressed in the local formulation of one sin which Allah will not forgive even if the man who commits it repents. The sin is that of the sabbab,21 the trouble maker, who makes friction between other people, particularly between men and their wives. The trouble maker may act by gossiping and telling false stories, or, Iagolike, by feeding a husband’s suspicions. But the worst act of the trouble maker is to accept entertainment from a woman whilst her husband is away and then to seduce her. One of the domestic problems created by seafaring life is the frequent absence of the men from home. Even fishermen, who are away from home for less lengthy periods than pearl divers or trading voyagers were, have to go out to their work at night and may be absent from home for several days on end. There are circumstances where, if suspicions are to arise at all, there is ample scope and opportunity for them. It is better, therefore, for women not to be alone when their husbands are away but to have around them other women of the family, for this reduces the possibility of a woman’s being tempted, unjustly compromised, or suspected. It might seem that, in any analytical sense, the segregation of women from men not related to them could equally well be regarded as the segregation of men from women. In some ways, this is true, so long as it is borne in mind that there exists a difference in the quality of the segregation. There is much stronger obligation requiring women to practise the avoidance of men outside the narrow degrees of kinship within which marriage is prohibited than any obligation which requires men to avoid women. When men avoid women, it is for the sake of the women’s modesty that they do so, unless the women are patently immoral, in which case they may be avoided for the sake of a man’s own respectability. Men who do not conform to the code risk some loss of reputation and family reproaches, and in case of a complaint being made to the shaikhs they may suffer a beating. But women who do not
68
Shaikhdoms of Eastern Arabia
conform risk not only a total loss of reputation but also death at the hands of their families, unless the men of the family are prepared to be despised. In matters of sexual misbehaviour, the honour of the whole family depends on the women incomparably more than on the men, and should one of the women of the family compromise it then it is a socially approved thing for her father, brother or cousin to kill her – rather than the man supposed to have seduced her. Such a murder is not strictly legal, but it is tolerated and socially approved. The effects of segregation on the social life of men and women are also different. Men, going about and sitting in the open, live in a wider and more gregarious society than women. At the widest scope of women’s social life, when they are entertaining or being entertained by other women (a time when the men of the family cannot come in to them), the place where women see each other is either their own house or someone else’s. They are never on, so to speak, neutral ground to which everyone has equal and unrestricted access. Women choose whom they will entertain and whom they will exclude, and they also choose whom they will visit, considering both their hostesses and the sort of people their hostesses may admit to their company. Hence the social life of women with other women is more narrow than the social life of men with other men, and is limited effectively to the women of their own and their husband’s family, the women of the families these two families visit, and the women who visit those families. It is thought that both girls and boys should marry young in order to avoid temptation and become accustomed to appropriate habits and responsibilities. In traditional society, people have no record of their birthdays and do not count their ages, but girls usually marry in their teens from about twelve years of age onwards, and boys from about fifteen but more usually nearer eighteen or nineteen. Since marriages are arranged within and between families, the timing much depends on the possibilities. (It is not usual for any strict understanding to be reached on behalf of very young children, as may happen elsewhere in the Middle East.) At this age, the boys are not old enough to set themselves up in houses of their own. They do not have the resources, nor is it considered appropriate that they should leave the family home. And girls are far too young to manage a house independently. After the first few nights of the wedding which she spends with her husband in her father’s house so as to become accustomed a little more gradually to married life, the bride and her husband take up residence in the house of the husband’s father. Here she exchanges the authority of her own parents for the authority of her husband and his family.
Women and Men
69
Unless a girl’s parents are very indulgent, she has no say in whom she is to be married to. In her position as a young girl (and among the merchant classes, before the days of going to school, girls ready for marriage were never allowed outside the house at all) she can have no knowledge of men or the whole question of marriage, and therefore her opinion of what would make her happy does not count. At the betrothal ceremony, when the marriage is legally entered upon, the father must take a witness to his daughter, who is modestly concealed behind the door of a room, and ask her formally whether she empowers him to contract the marriage on her behalf. A woman who has been married before must reply to the question, but a virgin is required to be too modest to speak, and her silence is formally interpreted as implying consent. Within the law, it would be possible for a girl to answer and say no, but one never hears of any case in which this has happened. When one asks hypothetically what would happen if a girl were to refuse in this way, the answer is that she would have been made to agree in any case before the betrothal ceremony was embarked upon, and, if she were suddenly to refuse, then her father could not only maltreat her at home but also refuse to let her marry at all so long as he lived. This, however, is not to say that fathers are callously interested in disposing of their daughters in marriage for their own financial, or even political, advantage. The impression one forms in talking to people about this question is that fathers are interested in assuring their daughters’ well-being and a happy relationship between families, and this is certainly what is socially expected of them. In the context of marriage, two local customs would seem to express something of the psychological situations which may follow from the way in which social realities are formulated. The first of these affects only the husband and is to do with the betrothal ceremony. The local custom is that this ceremony should be absolutely private, the only person present, apart from close male kin of the bridegroom’s and bride’s families, being the religious functionary who makes the contract. Those present sit cross-legged on the ground and they must keep their hands flat down on their knees throughout the ceremony.22 The popular explanation of the exclusion of strangers and of the manner of sitting is that it is to avoid any possibility that sorcery might damage the marriage. Were it not for having to keep his hands flat on his knees, it is said, an ill-intentioned person might interfere with the marriage by doing such things as breaking a matchstick in his fingers or turning a date stone round and round. The symbolic representation need not be elaborated on (though I should add that people are well aware of
70
Shaikhdoms of Eastern Arabia
what the matchstick and the date stone can be thought to resemble). The result would be that the bridegroom would find himself unable to consummate his marriage.23 The second custom is that the consummation of the marriage should be a formalized rape. The girl is expected to defend her virginity and only to submit to her husband after a fight. Some girls are said to fight in earnest, and one woman is spoken of in Dubai whose first two or three husbands had to divorce her because they could not persuade or force her to submit to them. I do not know whether this is a tall story: the husband who was at last successful is said to have tricked her by fastening her in a mat when she had fallen asleep. This formalized rape as the consummation of a marriage would seem to express two things: the ideal of chastity in women and the ideal of masculine dominance. I cannot say whether it ever has disastrous emotional results. One does not hear of them, but the facts would in any case not be public knowledge, and would only be known, if at all, among women. In legalistic terms, the control a man exercises over his wife is very considerable. At an extreme, he can even forbid her to visit her parents excepting when they are dying. Hence some practical conflict may arise between the law and the local norm of family solidarity, from which a woman should never be divided. In one such case, a man had quarrelled with one of his wife’s brothers and forbade his wife to see him, though he allowed her to visit her other brothers. She went to the village where her brother lived in order to deliver a medicine, over which a religious man had read verses of the Koran, to the sick child of her brother-in-law who also lived in this nearby village. Her brother heard that she was there in her brother-in-law’s house and sent a message round telling her to come and visit him. This she refused to do. When she was leaving, her brother-in-law accompanied her, and on the way she was waylaid by her angry brother and had to be defended from him. It was the brother who complained to the Ruler of the state about the matter. Another man had to take his wife’s mother and her unmarried sister to live in the house with him, but he found that he could not control the sister, who was always out of the house visiting people. He thought this improper and feared it might start gossip that would damage his reputation, so he warned his wife that he would divorce her unless she and her mother could make her sister stay at home. The sister could not be restrained, and so the husband divorced his wife and got all the three women out of his house. At first, he said, he regretted the divorce but later he was glad of it because he heard that the sister had borne an illegitimate child. As a result of the divorce, this did not
Women and Men
71
affect him and his reputation did not suffer. These were poor people living in a village and not leading members of the community. What happened to the sister-in-law is an illustration of what can happen among the less strict families. The child is said to have been killed by its grandmother at birth (though bastard children are not invariably killed). The child’s mother married one of the Shaikh’s young guards, who was very poorly paid, for only a token sum of bridewealth. The guard’s home was a long way, and he had every intention of returning there. There was very little question of his taking his wife with him. Had he been at home, his family would scarcely have allowed him to make the marriage, which was disapproved of by fellow kinsmen of his who were working locally and was regarded as little better than co-habiting with a prostitute, but, according to his friends, he wanted a woman to look after him and to live with for the time being until he should marry properly. Brought up in seclusion and then removed from the domestic circumstances and personalities which represent her whole experience of life, a newly married girl is in an unhappy position unless not only her husband but her mother-in-law and the other women of her husband’s family are actively kind to her. The girl comes particularly under the control of her mother-in-law both in matters of domestic work and in social situations. It is not uncommon for the intense emotional attachment women have for their sons to have a deleterious effect on their relations with their daughter-in-law, producing quarrels and unhappiness. There are many examples of mothers who have quarrels with successive daughters-in-law to a point where the son divorced them in turn, but in such cases the likely result is that sooner or later a girl of stronger character turns up and persuades the son to leave home and set up a separate household with her. This is much easier in these present times of plentiful opportunities of employment for wages. Safeguarding the happiness of the girl is an important motive that people have for arranging to marry their daughters into families to which they are already related. The girl’s parents know that these families of kin and affines are likely to treat their daughter well and make her happy. The social connections of women’s visiting already exist, and so the daughter is not going away to a life of isolation but merely taking up another position within what is largely the same social sphere. Strictly speaking, the preferred marriage for a man is with the daughter of his father’s brother, which is the nearest possible marriage outside his prohibited degrees of kinship. But practically speaking, any marriage with a girl closely related to a man already, the daughter of an aunt or
72
Shaikhdoms of Eastern Arabia
uncle either on the father’s or on the mother’s side of the family, is also preferred, as is a marriage to a girl related to the family by some tie of affinity. It is said that such marriages are not preferred particularly by young men, and men commonly say that there is no guarantee that a man will be happier married to a close kinswoman than married to a woman with whom he had no previous relationship of any kind: it is a matter of luck and character. But such close marriages are generally preferred by parents, partly for the sake of the girl and partly with the object of renewing and strengthening family ties and holding existing relationships together. And, of course, these relationships cannot remain static by being left as they are, because they are altered by the coming into the family of a new wife. In a society in which the family is overwhelmingly the most important unit of economic co-operation and the group within which, much more than the individual, social status is invested, this is far more than a matter of sentiment, however important that may be. It is a matter of great practical significance. 24 There is, however, another practical side to the matter which prevents such marriages from being the dominating pattern throughout society. This is provided by the prevalence of divorce and the fact that girls of suitable age are often not available in closely related families at suitable times. Otherwise, within the value system, inward marriage would be the prevalent pattern of the society. As it is, the tendency towards it is stronger in the greater families, who tend to be greater in numbers as well as in importance because there is more economic reason for them to remember their relationships. A corollary of the attachment of women to their own families and their insecurity when they are taken away from their kin is that married women are usually most unwilling to go and live at any distance from their parents and their parents are usually unwilling that they should go. Hence women are not only conservative in relation to the home but they are also conservative in relation to small geographical areas. Here, therefore, is a society in which men have always been, as they still are, able because of the type of work they do, and also willing, to move and go to take up residence in other places, whereas women tend to represent local stability. They do so in more than one sense. Men who wish to move may find they cannot because of their wives. If the men move far, then their wives may not be willing to go with them. The men marry in the new places they go to and thereby establish ties to a new place through their wives and their wives’ families. Up to this point, I have spoken of ‘the family’ in a vague way without defining. In fact, it is not possible to define the family as a strict exclusive
Women and Men
73
group in the society of the Gulf – anymore, perhaps, than in any society. On the other hand it is necessary to point to distinctions within the family. These distinctions rest upon general differences in the quality of relationships, which in their turn are closely associated with the status of women as compared with men. Just as there is a contradiction between the way in which men sometimes speak of women in general and the way they speak of the women to whom they are related, so there are two somewhat contradictory ways of speaking of women in relation to the children they bear to men. One way of speaking suggests that it does not matter who the child’s mother is; it is his father who is important. Hence appears the proverbial expression, ‘Descent is from the foot,’ an expression which is explained by another one, ‘I have kicked off my sandals and will get some more,’ which is supposed to express a man’s attitude to divorce. In harmony with these expressions, if a man sleeps with a slave woman of his and has a child by her whom he acknowledges, then that child is legally the equal of his other children. It was not uncommon in the days of slavery, which are now ending, for fathers to acknowledge their children to slave women in this way. Socially, the children are at a some disadvantage, as compared with the children of free women whose fathers are of equal rank, but the disadvantage is not overwhelming. But the other aspect of the situation is expressed in a proverb of a different import – ‘If you want a son, look for a mother’s brother for him.’ In other words, it is important to marry a woman whose family her son can be proud of. The possible uses of a third expression (the counterpart of ‘He is the son of (i.e. takes after) his father’) which says, ‘Two-thirds of the boy are from his mother’s brother’, in praise and in irony are many, and if the father’s marriage has been unsuitable can make a very harsh comment. As I have said, marriages involve families as wholes. It is easy, then, to imagine the complications which arise if a man chooses to marry a woman whom his mother and his sisters would prefer not to meet, for faults either of her own or of her family connections. The question regularly asked if a man reveals such an intention is, ‘If you marry her, who will marry your sister?’ The inconsistency of the two attitudes to father right (to use an old term, without wishing to rake up old controversies) is not merely a phantom result of translation and partial description. The inconsistency remains when all the local circumstances have been understood in their own terms. It lies within the tensions inherent in the society. The basic distinction drawn within family relationships, both in vocabulary and in fact, is the distinction between the father’s side of
74
Shaikhdoms of Eastern Arabia
the family and the mother’s. The linguistic distinction does not start with the grandparents, since the same two words mean respectively ‘grandfather’ and ‘grandmother’, whether they refer to grandparents on the father’s or the mother’s side. The two can only be distinguished by the use of an expression which literally translates ‘on the father’s/ mother’s side’. The absence of linguistic distinction in this case probably implies no more than the simple fact that by the time men and women are fully adult and concerned in affairs, their grandparents are likely to be either failing or dead. The word jidd, ‘grandfather’, has a further meaning of ‘ancestor’ and, in this second meaning, it would be uncommon to use it for referring to an ancestor on the mother’s side without expanding the term with the appropriate qualifying phrase. 25 The distinction drawn between the two sides of the family starts with – and here it is necessary to introduce Arabic vocabulary – the father’s brother, [amm, and the mother’s brother, khal. The feminine forms for aunts on the two sides of the family correspond to them – [amma and khala – and the terms for their children follow the pattern as ‘son of’ and ‘daughter of’ added to the four preceding terms, so that we have wuld al-[amm (the al being a prefix meaning ‘the’), ‘the son of the father’s brother’, bint al-[amm, ‘the daughter of the father’s brother’, wuld al-[amma, ‘the son of the father’s sister’, and so on. The social representation of the character of the relationship of a man or woman with the mother’s and the father’s sides of the family corresponds to a degree to the social representation of a son’s relationship with his mother and his father. Putting the contrast admittedly too crudely, the father represents authority, whereas the mother represents affectionate indulgence; the father stands for power and force, whereas the mother stands for peacemaking and influence. These harder and softer relationships (both, it should be borne in mind, being relationships of affection and duty) characterize the quality of the social ties between a man or woman and the father’s and the mother’s side of his, or her, family. Both in law and in social practice, the group of kin connected with one through one’s father is a much more formal body than the kin connected with one through one’s mother. The formal relationships of the latter, or the side of the authority, are amongst themselves. They have neither responsibility towards, nor authority over, the children of their sisters as they have with the children of their brothers. Even the women on the father’s side share, in so far as they are capable as women, in the side of authority, as is appropriate to the fact that they are much more likely to be living in the same house when children are being brought up than are the sisters of the mother.
Women and Men
75
As far as I know, this distinction would apply equally truthfully to all parts of the Arab world, including those with ancient urban communities. But I will not try to speculate here on the survival and revivification of social forms throughout all the varied lands and circumstances in which Arab peoples live. Even were one to wish to do this, adequate analytic material is too scarce and thin to allow one to do so. But even the towns of the Trucial Coast are on the fringe of the Arabian desert – and not merely in a geographical sense but in a social sense also. As I shall describe later, inasmuch as they are formulated, the values of the towns are inspired by the desert. The desert, for all its poverty and squalor, which townsmen do not forget, represents in some of the aspects of its life an ideal of honourable social relationships to which townsmen, in company with their shaikhs, aspire.26 The desert represents, among other things, independence; and independence, among other things, means having no superior to protect one. Among the bedouin, the minimal political group founded on status is made up of the men who are descended agnatically from a common great or greatgreat grandfather, and these are the people who are responsible and answerable to each other’s lives in the feud. The mother’s side of a person’s family has no such place in the feud. They may be involved, but this is only in exceptional circumstances or in the case of exceptional people. Thus, in the life of the desert, the distinction made between father’s and mother’s kin is a clear, formal and practical distinction of overwhelming importance in the circumstances of harsh political existence. Here are the responsibilities of freedom, and freedom is one of the local ideals. The general and simple situation which the institution of feud reveals illustrates a crucial difference between the bond of agnation as compared with the bond of kinship through women. Agnation is the basis upon which people conceive of tribal unity in that the customary way of picturing a tribe is as a group of people all of whom are descended in the male line from a common ancestor. Herein lies the principle whereby fellow tribesmen co-operate with each other in peace, fight on the same side in war, and assume responsibility for each other’s wellbeing. But this political principle should, I think, be viewed as being produced as an analogy with kinship rather than as itself representing kinship. The group of four or five generations of agnatic kin which is involved in the feud is, however, certainly conceived of as a kin group. They are the descendants in the male line of the man who was literally the great grandfather or great-great grandfather of the person killed or of the person who killed him, and they hold a common responsibility,
76
Shaikhdoms of Eastern Arabia
a common ‘answerableness’ as it is put, for the act which produced the feud or for the conduct of the feud, depending whether they are the kin of the killer or of the killed. Here, in the face of death, is an ultimate common interest and common responsibility among those who share kinship through male links for which there is no counterpart for those who are kin through women. Qualitatively, the side of the family represented by the paternal uncle is characterized by authority and responsibility. A man’s subordination to his father’s brother is similar to his subordination to his father. A criticism made of Christianity in comparison with Islam illustrates the associations of thought in this matter, when a local Muslim recognizing that Christians worship the same God as Muslims says, ‘How can Christians obey Allah without obeying his Prophet? It is as if one were to obey one’s father without obeying one’s father’s brother.’ On the other hand, the side of the family represented by the mother’s brother is characterized by lack of authority, and a corresponding lack of responsibility. The positive side of the relationship lies in affection and influence and the generous, as distinct from dutiful, giving of help. This is not, of course, to say that relationships between paternal kin are commonly lacking in affection, influence or generosity, but merely that duty intervenes in this case to modify the quality of the relationship. But in the towns and villages one becomes aware of a difference of emphasis in representations of kinship from that which applies in the desert. The terms remain the same, and so does the explicit ideology. People still speak in general terms of paternal cousins all being ‘one’ like brothers and of the primacy of the male line, but in conversations about specific situations the trend of speech implies a tendency to attach in practice rather more importance to kinship through women – and also to relationships of affinity – than applies in bedouin life. This tendency corresponds to a difference of situation in the relative importance of the political and the economic aspects of day-to-day life. In the desert, the agnatic kin group is a unit of co-operation in both economic and political life. At times of instability in the desert – and this was prevalent until twenty years ago – economic and political life were scarcely distinguishable, since keeping camels also meant defending them constantly from raiders, and also raiding camels from other people. The elementary family of a man with only his wife and unmarried children around him was too small a group to be effective in tending camels, let alone in defending them from parties of raiders. By virtue of their concentrations of population and their system of rule, the towns and even the exposed villages are in a different position.
Women and Men
77
Warlike and defensive actions are co-ordinated by the shaikhs, and minor violence is controlled by their guards. In trade, as compared with pastoral life, small families can be effective units of organization. On the other hand, in the towns a man’s connections become much more significant, whether it be in the pursuit of politics in peaceful situations within the hierarchical scheme of political organization, or in the pursuit of financial profit. These two activities are, of course, like the pastoralism and fighting of the bedouin, very closely intertwined. In the towns and villages one quite often finds men, who within their social sphere are prosperous, drawing their sisters’ sons or daughters’ sons towards them, especially when they themselves have few or no sons of their own whilst the families that their sisters and daughters have married into are economically weak. ‘Family’ remains the basic economic grouping, but the personnel of this grouping changes with the change from what, with bedouin, is largely subsistence pastoralism to the more capitalistic organization of trade and work among the settled people. A man who is rich in himself but weak in male kin needs his sisters’ children or his sons-in-law, and his wealth may attract them to him. This situation is not expressed in the formal ideology of kinship among settled people, any more than it is expressed in the society of the desert. It is, however, expressed in speech, and sometimes in complaints. The tension between the ideal of what ought to be – of what is ‘natural’ and ‘honourable’ and exists among the bedouin – and the actual, comes into the open quite conspicuously with talk about relationships of affinity. The mother’s brother is not much raised in such conflicts of ideal and actual values. He is, after all, kin by blood relationship. The nasib, affine, is another matter, for here are only the ties expressed in terms of who will, or can, marry whom. If the marriage has again joined together people already related by blood, then it is more common to refer to those among whom it produces ties of affinity by their titles in kinship. Hence, in practical use, the word nasib means generally an affine who is not already a kinsman on either the maternal or the paternal side, unless perhaps by some remote previous connection. People commonly complain of how, in these degenerate days, the values of kinship are giving way to the opportunism of marrying one’s girls to those who are more eligible on grounds of wealth, ability or connections. A rather coarse expression draws attention to this, saying, ‘Better an in-law, not a paternal cousin; the in-law is (as close as) part of one’s own bottom’. This matter must be looked at in two ways, the synchronic and the diachronic. When a man gives his daughter in marriage to a man of
78
Shaikhdoms of Eastern Arabia
a family hitherto not related to his own, then his sons, of course, become the mother’s brothers of her children. The same occurs when a mother’s brother gives his daughter in marriage to that sister’s son: the sons of the givers are mother’s brothers to the children. But the agnatic connection is naturally and by definition only established by the marriage of a son27 – in other words it is established in relation to the pre-existent strength in numbers of men of the agnatic family. In the circumstances of the settled people, that weakness in numbers of men which is a disaster for a bedouin kin group can be overcome, given the wealth appropriate to a man’s status, by the bringing in of affines and potential children through women. It is by no means out of the question for these daughters’ children to be assumed into the family name of their mothers’ kin. Naturally all agnatic families, and most individual men, stand in relation to some other families as mother’s kin as well as being father’s kin in relation to their own, but the use they can make of this relationship depends on their circumstances within the economic and political conditions of sedentary life. That they can make such use of privileges in circumstances where wealth and numbers are differentiated is a cause of conflict between the circumstances and the ideology of kinship which is taken from the practical bedouin standard of the primacy of the male line. It should be noted that the conflict seems to apply more to poor people within the gradation of their poverty than to the rich, unless it is that the richer people are more guarded when speaking of it. People in the Trucial Coast do not usually consider their own social positions and relationships in terms of long term diachronic comparisons – ‘long term’ even in the sense of covering the changes of status which occur in the course of the four or five generations covered by the feuding kin group – any more than they do in England. As far as I have come across comparisons of status and wealth, they extend in personal circumstances no further back than the grandfather. But within this range of awareness there certainly exists a distinction in status between ‘pure’ birth and wealth. In practice, it is not common to arrange marriages in terms of either of these criteria separate from the other. The frequently used idiom ‘origin and separation’ (al-asl wa- ]l-fasl, which is a rhyme like most such expressions) implies that people’s basic separation in social circumstances lies in their origin. But though these origins may have a semblance of being a historical matter of the most ancient kind, there are good reasons for supposing that the time scale in which they are known and considered is a shallow one. Nevertheless, the tension between the two standards of birth and wealth is a real
Women and Men
79
one – and socially, as distinct from the analytic terms of an outsider’s account of social history, the real one. What is significant is that the gap cannot totally be bridged by one rich man in his own lifetime. Hence the telling importance of such a precept as, ‘If you want a son, find him a mother’s brother first’, and the uncomfortable reminder of family status, ‘If you marry such-and-such a girl, who will marry your sister?’ The importance people attach to superiority of birth slows down the rising of rich families and the decline of poor ones, and the tensions of adjustment in individual cases represent adaptation within circumstances of relative stability within the circumstances of any one or two lifetimes. Changes are gradual rather than sudden and catastrophic. And when they are catastrophic (as I shall discuss in relation to exiled shaikhs and leading families) the catastrophe is often concealed by the removal of the children of the great from the scene of their predecessors’ glory, since either they flee to save their lives or they emigrate to save their embarrassment. The scale of social history is determined by social memory. And one might add that in western society where it is a matter of scholarship, the scholarship has had no relation to determining status excepting as that very one-sided scholarship of genealogy which is naturally invoked only by those who wish to improve their status or preserve it. In other words, this would seem to be a common part of the human predicament in hierarchically organized societies.
3 The Bedouin
The contrast – and the antipathy – between the desert and the town makes one of the ancient themes of the Middle East. What it means in terms of attitudes to life, evolved in both cases in experience and vicissitude, can be seen if we juxtapose two verses, one of them widely known in Arabia, the other equally famous in Persia. The Arabic verse says, Pour me not out the cup of life in humiliation, Pour me rather the bitter cup of the Bitter Apple.1 The Persian verse recommends a different philosophy: Learn to be humble if you would be blessed, Ground that stands high up is never watered. The Arabic verse draws its metaphor from the desert; from place to place among the general barrenness, this cucumber-like plant the Bitter Apple spreads out over the sand and produces a fruit of extreme bitterness. Bedouin say it acts as a laxative if merely rubbed on the sole of the foot, but has no other use. The Persian metaphor refers to the ancient system of irrigation by underground channels coming down from the high land into the valleys and plains. These channels are in use at the Buraimi Oasis and in other parts of the Trucial Coast and Oman. In Persia they are called qanat and in the Trucial Coast falaj or shari[a.2 When constructing them, the Persians sink deep wells in the high ground and join them together at the bottom by means of an underground channel. The channel is brought down to the plain at a slope slightly less than that of the water table, so that it eventually joins the surface of the low ground and delivers the water for agriculture. Being 80
The Bedouin
81
irrigated in this way, the low ground in Persia can produce rich crops, whereas the high ground, watered only by the meagre rain, produces poor crops or none at all. The Persian verse expresses the experience of a history of conquest, tyranny and exploitation, repeated over and over again through the centuries. It reflects a pattern of life in which both the rich and poor had to practise prudence, caution and deviousness in order to survive, and where, for those who were lucky, self-abasement opened up paths of advantage and favour which blunt straightforwardness could never have negotiated. In Persia even the idioms of address, with their marked reference to relative status, would seem to correspond to this attitude of mind. When speaking to a superior, one does not say, ‘You say’ or ‘I say’. One says instead, ‘You command’ or ‘I submit’. The Arabic verse, with its metaphor of the desert, its arrogant independence and its proud, even bombastic tone, invokes the values of the bedouin, and a code in which politeness is usually devoid of any attempt to flatter or even, often enough, of any particular wish to please. In bedouin life, perhaps the highest values are personal reputation, independence and self-assertiveness, combined with an unswerving self-interest coextensive with self-reliance. In the physical hardship of bedouin life, the high value set on generosity and mutual help is a condition of survival. Behaviour which, seen from the standpoint of any other scale of values, presents a confusing mixture of altruism and shameless greed, is in bedouin terms entirely self-consistent. In the Trucial Coast, one finds no such strongly marked division between the life of the bedouin and the life of the settled people as exists in the more fertile parts of the Middle East. These are desert states, and there has been a constant coming and going between the desert and the town. The Trucial Coast has no large population of peasants for the bedouin to maraud, and the settled places have had no system of government essentially different from that which is applied in the desert. Many Trucial Coast bedouin own date gardens and some own solid houses, whilst until recently many townsmen used to own herds of camels. Even now, villagers hand over some of their flocks for the bedouin to look after. Most of the settled Arabs have spent some of their time in the desert. In particular, the shaikhs of the towns still receive a good deal of their upbringing there, where they can combine the sports of hunting and hawking with political training. Ruling families are shaikhs in the desert as well as in the towns and villages, and desert politics concerns them just as much as the politics of the settled people. Among the Arabs of the towns – though it must be remembered
82
Shaikhdoms of Eastern Arabia
Fig. 3.1
Bedouin. From the author’s photographic collection.
that much of town commerce is in the hands of non-Arabs – tribal standards still prevail to a marked degree. In some towns, even the settlement pattern is on tribal lines, the quarters of the town bearing the names of the tribal sections from which most of their inhabitants come, and the word meaning a quarter in a town is the same as the word for a bedouin encampment, farij. Arab tribal values are founded principally upon status relationships. The tribe into which a man is born compels his loyalty and duty, and that of his sons and all his descendants in the male line, theoretically for as long as the line continues to exist. Outside the tribe, there is no group of people to whom he owes this loyalty in an equal way. Within the tribe, should there be conflict, then his duty is owed more strongly to those who are nearer to him in descent, also in the male line only, than to those who are more distant. This is the basic morality, or at least a large part of it. If a man finds fellow tribesmen involved in a fight, whether in the desert or in a town, then it is not for him to consider what might be the rights and wrongs of the matter, but to join them.
The Bedouin
83
The basic pattern, then, is a segmentary one, the tribe being conceived of as a group all of whose members are descended in the male line from a common ancestor, divided internally into sections and sub-sections by descent through more proximate ancestors, with the whole tribe acting as a united body in relation to other tribes, and the members of a section uniting when necessary in opposition to another section of their own tribe by virtue of their greater closeness of common descent, and similarly the members of a sub-section uniting in opposition to another sub-section within their own section of the tribe. But, though this tribal segmentation is the rough basic pattern of the bedouin idea of tribal duties, what bedouin think they should do, let alone what they in fact do, in varied political circumstances is a much more complicated matter. One has only to think of the antiquity of trade in Arabia and of the custom of the pilgrimage, requiring the crossing of successive tribal territories, or of the bedouin institutions of protection, hospitality and companionship in travel, to be reminded that there are other principles at work to make the bedouin system what it is. Broadly speaking, we may describe the relationships which are part of the bedouin system but outside the simple lines of segmentary political relations as being relationships of a contractual order, leading to the possibility of treaties and to other contractual relationships in which the part played by tribal shaikhs is of great importance. In respect both of status relationships and the relationships of contract, bedouin are usually quite explicit about what is the right and normal thing to do. There exists a greater need for consensus in the desert than in the towns, where the shaikhs occupy a more dominating, though often a less secure, position. Hence, in the politics of the towns, we find the desert principles somewhat eroded, the gaps being filled in partly by shar i[a law and, substantially, by a precarious absolutism. The account given by the bedouin of their tribal loyalty and other associated institutions is the most explicit and consistent formulation of political values that we find in the society of the Trucial Coast. It is obvious that the structure of relationships to which tribal values correspond could not survive entirely unchanged in the very different circumstances of settled life, and particularly in the circumstances of town life, where differences of social class become conspicuous and immigration for purposes of trade and labour makes the populations increasingly heterogeneous. Nevertheless, the towns are only the centres of states, not the whole of them, and the bedouin play an important part in the politics of these states and, indeed, even in the internal politics of the towns. The values of tribal loyalty, therefore, are not a matter
84
Shaikhdoms of Eastern Arabia
restricted merely to the desert and peripheral to other political values. In the Trucial States, the loyalty which binds the members of a tribe together is called shaff.3 A description of what is meant by this word, and of the groups within which shaff exists, will demonstrate some of the formal principles of tribal structure. Shaff is a principle of action whereby people related to each other in the male line (and others, as I shall mention later) are expected to tolerate each other more than they would tolerate strangers, help each other in misfortune and support each other in war and diplomacy against the rest of society. In accordance with the same principle, agnates who are more closely related by descent support each other when necessary against other agnates of theirs who are related to them more distantly. We may, then, begin to investigate this principle by describing the formation of the groups within which it is regarded as existing. The tribes of the Trucial Coast are small, the biggest of them numbering perhaps twelve thousand people and most consisting of no more than two or three thousand. The members of a tribe describe themselves as all being agnatic cousins, their term for which is bani ]l-[amm, ‘sons of the paternal uncle’. In southern Iraq, the term for tribal loyalty is bin [amma, a term obviously derived from bani ]l-[amm, though the second m of the word [amm is elided and the whole spoken as one word with the accent on the first syllable. Considering a tribe from the outside, as a whole political unit, one thinks of it as being internally segmented to form smaller groups and groups within those groups, each smaller group having greater political cohesion than the larger groups of which it forms a part. There is, however, much to be said for looking at the tribe as a building up of smaller units, by stages, into a larger whole, since there is no reason for supposing that the slow division of the tribe through genealogical stages is historically a true one, whilst there is every reason for supposing that the more units are built together the less is the political or social cohesion of the whole. Hence, the idea of ‘whole’ and ‘part’ can be subtly, or even crudely, misleading. The agnatic political groups to which a bedouin regards himself as belonging, working outwards from the smallest to the greatest, are usually no more than the groups which I will call the kin, who are the agnatic descendants of his great-grandfather or great-great-grandfather, then the secondary section, which is a number of kins united in a more distant agnatic ancestor common to all of them, then the primary section, similarly composed of secondary sections united in a more distant common ancestor, and last the tribe, made up of the primary sections which are united in the ancestor of the tribe as a whole.
The Bedouin
85
These ancestors4 are not all the same kind of imaginative object. A tribesman will usually know, or able to find out, the names of his ancestors back to his great-grandfather or great-great-grandfather. These are totally real people who, if they are dead now, did exist as living people in the memory of men whom their descendant knows or knew when he was younger. The descent of one from another of these ancestors is known to be the literal historical truth, with no link in the chain omitted. After these ancestors of the kin, a man knows the names of the ancestors of the secondary section, his primary section, his tribe, and often the name of the ancestor of his tribe. No man, not even the most simple-minded, will say that these seven or eight ancestors through whom he is joined to his tribe’s founder are all his ancestors. Whereas he speaks of the ancestors of his kin, that is to say those as far as back as his great- or great-great-grandfather, as a genealogy going back through single steps of generations, of ancestors more distant he will say that they have an unknown number of generations between them. Distant ancestors are spoken of as men who actually lived but about whom little or nothing is known. A man has no connection with them as personalities. The names of tribes and their sections usually connect verbally with the name of the tribe of the relevant ancestor. Members of the [Awamir tribe say that their tribal ancestor was called [Amir, and members of the Manasir tribe say their ancestor was called Mansur. [Amir and Mansur are personal names, and the names of the tribes supposed to be descended from these ancestors are possible plural forms taken from the names. The Bani Yas say that they are descended from the Prophet Al-Yas.5 The battle cry of a tribe usually incorporates the name of the tribal ancestor, so that the Manasir cry, ‘Sons of Mansur’ whilst the [Awamir cry, ‘Sons of [Amir’.6 On the other hand, the Bani Yas of Abu Dhabi cry, ‘Sons of Zayid’. The shaikhs of Abu Dhabi are also the shaikhs of Bani Yas, and the Zayid whose name is called in battle is identified with the great shaikh Zayid who ruled Abu Dhabi from 1855 to 1908. Not all the names of tribes and tribal sections bear any resemblance to ordinary personal names, however. In the case of those that do not, one may find stories connecting the ancestor with some other word of vocabulary. The stories which connect the Chiyabil section of the Al Bu Shamis and the Hawamil section of the Bani Yas with their respective ancestors reflect something of the seafaring background of the society of the Gulf. The first ancestor of the Chiyabil, one is told, was a man called Chaibali who came from the Persian province of Fars. Chaibali
86
Shaikhdoms of Eastern Arabia
was a man who had long hair all over his body. Whilst he was swimming, he got caught in a fishing net and was carried to the shore on a float (a fishing float made of three palm branch stumps fastened together is called chaibal, and chiyabil is a plural of this word). Chaibali was caught by the fishermen, and he married one of their women and she bore to him the ancestors of the Chiyabil. A similar sea story is told by the Hawamil section of the Bani Yas when they are talking about how they got their name. The ancestor of the Hawamil is said to have floated in from the sea on a spar carrying a woman with him in a box – an idea that the Hawamil find amusing. When people asked him what he was doing, he said, ‘I am just wandering about,’ and this is how the Hawamil got their name, because hamil means ‘wandering about’ and Hawamil is a plural of this word. The man had children by the woman he had brought with him and married them with people from the various sections of the Bani Yas. It is worthy of note that, in spite of what this story might seem to suggest, the Hawamil are camel-owning bedouin and not fishermen of the coast. The [Awamir give the full name of their tribal ancestor as [Amir bin Sa[sa[a, the latter name being a famous one in classical Arabic genealogies. According to the [Awamir, Sa[sa[a was forced to become a Muslim in the early days of Islam and to get his own back he stole the Prophet’s camel. So, say the [Awamir, the predatory habits of the tribe are based on Sa[sa[a example. Other people joking with or about the [Awamir say that they are not very strict Muslims and take it in turns to fast separate days of Ramadan. According to the Bani Yas, the full name of their own ancestor was Al-Yas bin [Amir, and both they and the [Awamir say that this [Amir was the same man as was ancestor of the [Awamir tribe. Of course, it is not possible to fit together logically the idea that Al-Yas was a prophet before Muhammed with the other idea that Sa[sa[a was a contemporary of the Prophet Muhammed in the terms of this idea of genealogy. The bedouin themselves are perfectly capable of seeing this. It seems to concern them no more than the personality of King Canute concerns most Englishmen. A matter of greater interest here is the linking of tribes together, in a range wider than that covered by any tribal name, through such further stages of putative ancestry. This is one of the reasons why I have preferred to speak of kins building up into tribes rather than of tribes dividing into kins, but further discussion of this matter must wait until a later stage. Besides finding linkages in genealogy beyond the scope of existing tribal names, one also finds the same tribal name shared by people who recognize no important political connection between themselves. The attitude of bedouin towards their ancestry is ambivalent. Those who are
The Bedouin
87
united by a common ancestor and are also normally united in political action will attribute their unity to this common descent. On the other hand, tribesmen who share the same tribal name and who, obviously enough to the outside observer and also to the more intelligent among themselves, were originally of the same tribe but now share no common tribal loyalty, will deny that common ancestry necessarily imposes any duty of loyalty. When one questions bedouin about such matters, one rapidly becomes aware that the explanations given are justifications of existing political facts and of the regular and specific status relationships in which people were brought up, whilst general descriptions ignore many matters which, in the terms the descriptions use, would be anomalous. When speaking of other bedouin whose tribal name they share but with whom they do not regard the loyalty relationship of shaff as existing, some people will say of the other tribe that they know nothing of its ancestry and are not interested in it. Others, in my experience the more intelligent, will say that when a tribe has divided of its own accord the relationship of loyalty between its parts ceases to exist. There are, for example, groups of a section of the Bani Yas called the Mazari[, which is the biggest of the Bani Yas sections, all over the Trucial States, and there is no political unity among them now they have divided, though a certain amount of intermarriage takes place between the different fragments, and it would no doubt be somewhat easier for a Mazru[i (this is the singular of Mazari]) from one of the groups to settle in another if he wished to do so than it would be for him to settle with a completely different tribe. Where the separation of tribesmen has occurred within living memory, some part of the original loyalty tends to remain. Bedouin speak of tribal loyalty largely in political terms, but the duties and sentiments of it are not limited purely to political action. The sort of expressions that people use in connection with tribal loyalty are, ‘We are one tentpost,’ nahn [amud wahid, or, ‘We are one bundle, there is no difference between us,’ nahn rabta wahida, ma shai ] farq bainna, and, ‘Our war is one war and our peace is one peace,’ harabna wahida wa-sulhna wahid, meaning that they do not make war or peace separately from each other, and in the same meaning they speak of going out on raids together and returning from raids together, yughazzun wa-yanchif un raba[a. And a further expression suggests that men will fight on behalf of those with whom they share loyalty even more fiercely than the latter will fight for themselves: ‘If they strike at the body, we strike at the head,’ idha hum yadrubun f i ]l-yiththa, nahn nadrub f i ]l-ras. Should a
88
Shaikhdoms of Eastern Arabia
member of one tribe kill a member of another, this is a cause of war, unless the shaikh of the offending tribe takes immediate steps to avoid it, and hence the expression used of those with whom people share tribal loyalty: ‘Our blood and their blood are one, our debt and their debt are one,’ damna wa-damhum wahid, dainna wa-dainhum wahid, this referring to retaliation and suffering retaliation for bloodshed. And, in more peaceful circumstances, people bound by a common tribal loyalty ought to protect each other’s reputations by objecting to any scandal they may hear spoken of any of them. Should members of two different tribes become involved in a fight, loyalty requires any fellow-tribesman present on either side to join in and give immediate assistance to his own man. Should a man be killed, then any fellow tribesman of his should retaliate immediately by killing a member of the other tribe, preferably, but not necessarily the man who committed the deed. When a fight happens between members of different sections of the same tribe, any member of either section who happens to be present will also assist his own man, though he will probably try to separate the two opponents if possible and to restrain the member of his own section if he feels justified. Thus the loyalty between the sections modifies the behaviour of their members towards each other. If anyone is killed in such a fight between fellow-tribesmen, then it is said to be likely that a member of the same section as the dead man will retaliate immediately if he is present. After this first crisis situation, however, the matter is different. In the case of such a killing within the tribe, tribal loyalty restricts the feud that develops to the respective kins of the killer and the killed. The idea of tribal loyalty does involve, therefore, some positive idea of tribal unity, not merely in opposition to outsiders but also in the internal relations of the tribe. This loyalty of tribesmen towards each other is a moral and emotional condition, and one or two examples will show how it lingers on even after members of a tribe have intentionally divided from each other over political boundaries and severed their common interests and political environment. In 1945, war broke out between the states of Abu Dhabi and Dubai and lasted for two years. Shortly before the war began, a group of the Manasir tribe, who live in Abu Dhabi territory, went and asked the Ruler of Abu Dhabi to give them some money for a subsidy, but he refused. Being dissatisfied with their treatment, they went to Dubai, where the ruler was very ready to treat bedouin generously, and when the war broke out they fought on the side of Dubai against Abu Dhabi whilst the rest of their tribesmen were fighting on the side of Abu Dhabi against Dubai. This group of Manasir, who had no grounds
The Bedouin
89
for hostility against their fellow tribesmen and would probably have stayed at home had they not been dissatisfied with their failure to get a subsidy, never fought against the other Manasir during the course of the war, and on one occasion saved the life of some of them. The Dubai Manasir had joined an ambush composed of a strong party consisting mainly of men of Dubai. When the enemy from Abu Dhabi, greatly outnumbered, was about to walk into the ambush, the Dubai Manasir recognized some other Manasir in the party they were just about to attack. They then called out to each other by name. The other Manasir recognized who they were, and so were able to flee, taking the rest of the Abu Dhabi party. This behaviour was looked on indulgently even in Dubai, whose people were the losers by it. Another case from the same war is of a different sort. During the war itself, also apparently because they wanted money, a section of the Mazari[ left Abu Dhabi and joined the forces of Dubai. It will be remembered that the Mazari[ are themselves a section of the Bani Yas, whose principal shaikhs are the shaikhs of Abu Dhabi. Hence, on various grounds, their behaviour was much more open to criticism than that of the Manasir who had gone to Abu Dhabi before the war and are not themselves Bani Yas. This large group of Mazari[ fought against other Mazari[ who had stayed on the side of Abu Dhabi. Moreover, in company with other bedouin following Dubai, they took part in a raid against the Manasir in which a large group of Manasir ran out of ammunition and many of them were shot down by the raiders without any opportunity of defending themselves. The numbers killed in these local wars have not been great, but then the numbers of people involved have been very small. In proportion to the total membership of the tribe, the number of Manasir killed in this raid was comparable with the number of British soldiers killed in the First World War in proportion to the British population. Ten years after this raid occurred, the Manasir of Abu Dhabi still bore a grudge for the massacre against Dubai and those of the Mazari[ who had fought for Dubai, not surprisingly, but their resentment, though in a milder form, extended also to the Abu Dhabi Mazari[. This was in spite of the fact that they had fought on the same side during the war in which the massacre of Manasir occurred and that the Mazari[ of Abu Dhabi had actually fought the Mazari[ who had joined Dubai. Here we have the idea of tribal loyalty under two different aspects. In the first case, fellow tribesmen fighting on opposite sides in a war were nevertheless unwilling to fight each other, and one group even saved a group of the others when it could. In the other case, in spite of the fact that they had fought against their fellow tribesmen, the members of a tribal section were still
90
Shaikhdoms of Eastern Arabia
blamed for losses that their allies had suffered at the hands of those fellow tribesmen they themselves had been fighting against. Just as it is commonly known in practice that not all people who are supposedly descended from the ancestor of a tribe are any longer members of it or share any part of the tribal loyalty, it is also well known to everyone that not every man who is a member of a tribe and has his loyalty in it can, in fact, be descended from the tribal ancestor. All tribes have within them groups of people who are known to have joined them at some time in the known past and now act at one with them in tribal loyalty. To return again to a name familiar already, there is a section called Manasir incorporated in the Bani Qitab, who are bedouin of Sharjah territory (though it is said that they once lived in the territory of Abu Dhabi). They are said to have fled to the Bani Qitab, who are the traditional enemies of their own original tribe the Manasir, because of a blood feud some fifty or sixty years ago, and never returned. Up to now, they have married mainly among themselves and little with their tribe of adoption, so that by descent they are for the most part Manasir on both paternal and maternal sides. Within the Bani Qitab, they are distinguished from other sections by the same name Manasir, but as members of a tribe they are called Bani Jitab like the rest of their tribe of adoption, and they share in the loyalty of this tribe and not of their tribe of origin. Such attaching of members of one tribe to another often happens because of internal dissentions within the tribe, but it can also come about in other ways, as when a tribal section becomes reduced in numbers to a point where it can no longer muster the force to maintain its separate identity. Curiously enough, such is the case with the bedouin part of what ought to be the most distinguished section of the Bani Yas. This is the bedouin part of the Al Bu Falah. The more important part of the Al Bu Falah are the Shaikhs of the Bani Yas and of Abu Dhabi as a whole, but they seem to have favoured the bedouin of their section – ‘cousins’, it will be remembered, only in a tribal and not in a family sense – very little. The Al Bu Falah bedouin declined in numbers and then, instead of attaching themselves to any other section of their own tribe, they attached themselves to a section of the Manasir who, though friendly with the Bani Yas, are a separate tribe. They are now united in loyalty with the Manasir. They have also adopted their camel brand. Camel brands, however, are largely a practical matter, since the people a man lives among will recognize his camels individually, whilst people further afield will recognize only the brand. Thus the brand identifies a camel with a particular group of people who are living together, and
The Bedouin
91
once they have the camel they can recognize which member of their own group it belongs to. One commonly comes across such cases of tribes attached to and incorporated in tribes other than their tribes of origin. In these circumstances, people make no attempt to manufacture spurious genealogies, perhaps because beyond the ancestors of the kin, whose position is important should a feud occur, genealogy is of the most skeletal variety. When, because of feud or other reasons, a small group attaches itself to another tribe, that tribe defends it, and the section to which it attaches itself defends it if necessary against other sections of the tribe, through customs I mention later. These duties of defence are reciprocal. They would seem to grow in a relatively short time into duties of loyalty. Nevertheless, in all the tribes in which I asked about the matter, people insisted that if not all, then most of their fellow tribesmen were descended from the tribal ancestor. Only the [Awamir, mentioned earlier, had a different opinion of themselves. The [Awamir are a tribe which goes out into the former no man’s land of the Empty Quarter, and there are other [Awamir beyond the Empty Quarter in what is now South Yemen. In the Trucial Coast, the [Awamir are not considered to have any tribal territory of their own or any strong traditional connection with any of the states. Neither are they a very strongly united tribe. When asked specifically whether they thought that all the members of their tribe were descended from the original [Amir, a number of members of the tribe said they were not. They said that only the shaikhs of the tribe were descended from the tribal ancestor, whilst the rest were descended from men who were his followers from the beginning. No member of the tribe whom I asked denied this. On the other hand, when I mentioned this statement to members of other tribes, no one would agree that this was also the case with his own tribe. On the contrary, people critical of the [Awamir, who, in the times of insecurity in the desert before 1950, and even after that time, have been very disunited and also dangerous to others, said that no doubt this was the reason why their tribal loyalty was so weak and they were always fighting each other. So far we have considered only the duties and relationships which are supposed to follow, though with exceptions, from shared descent in the male line. Numerous and traditional connections through women are, however, regarded as a further basis of common interest and co-operation, though a weaker one. The most absolute territorial divisions in the desert are those that follow from enmity, for with enmity goes danger. Bedouin tribes between whom traditional enmity exists do not dare,
92
Shaikhdoms of Eastern Arabia
Fig. 3.2
Bedouin at a well. From the author’s photographic collection.
even when formal peace obtains between them, to meet in a common boundary area. There lies between them a neutral zone in which no one dares to live for fear of being raided or killed. These areas may be places which have some potential value in that they have wells in them, but the wells are only used by raiding parties. Tribesmen living on separate sides of such divisions consequently remain quite separate. On the other hand, tribesmen who are, on the whole, friendly towards each other are in quite a different position. Tribes that are on friendly relations with each other, though not supposed to share any common agnatic ancestry, mix freely in the desert in the winter and may even gather around the same wells in the summer, or at least gather round wells that are not far separated from each other. Coming together at the wells is a sign of much closer political association than is mixing in the open desert, for at the wells quarrels easily arise. One man waters his camels before another and the other becomes irritated. If water is short, he may try to force his way in. Or camels may
The Bedouin
93
start to disturb each other and to fight, provoking their owners to blame and insult each other. Old injuries are remembered, and fights easily break out, drawing in other people on the pattern described earlier in this chapter. When speaking of the friendship of members of two different groups, and how great it is, bedouin will often say that they camp together at one well. Such is the imaginative significance of the act that a peacemaking ceremony of the interior, which I was told of but have never seen, includes the opponents’ drinking together out of the same bucket to symbolize their complete reconciliation. One of the chief objections to the giving of women in marriage outside the cognatic kin or the tribe is that the woman will have to live a long way apart from her own family. If she is maltreated she has no one to turn to and her father’s family loses touch with her children. If two tribes are living on friendly relations with each other and mixing freely in the desert, this barrier to intermarriage becomes much less serious. Such tribes have often intermarried over a number of years, further cementing their relationship of friendship, to a point where the maternal relation is not considered merely in terms of particular links of marriage between individual families, but each tribe is regarded as being in the general relationship of mother’s brothers to the other. In such a situation, the tribes are said to have ‘entered into each other’, tadakhalu, or they have said to be, again in general, ‘related’, taqarabu, which goes with the word for proximity. In relation to such situations, though it remains true that men’s ultimate responsibilities lie with their agnates, the members of their own tribe, nevertheless they should consult their mother’s kin before they take important action. The relationship of consultation produced by a tradition of intermarriage between tribes is sometimes referred to as a connection ‘by advice’, bi-]sh-shur. In some circumstances, people speak of loyalty, shaff, as existing between tribes so related by proximity and intermarriage. Such loyalty, however, is not of quite the same sort as loyalty between agnates. It is defined in the narrow terms of unity against enemies in war, so that it is said: idha harabathum qabilatum yanb, shaff bainhum, wa-idha bain anfushum, maku shaff, ‘If a foreign tribe makes war on them, they have loyalty to each other, but considered merely between themselves there is no loyalty.’ It is not such a far-reaching loyalty as that between fellow tribesmen, and is only given the word shaff in war situations, when a further tribe is fighting one of the two and these two are expected to make common cause against the enemy of either. Outside war situations, the relationship is described by the common word sadaqa, friendship, a word which in other contexts has only a very weak significance, as when
94
Shaikhdoms of Eastern Arabia
making a treaty of peace after a war; it is comparable with the political use of the word in English. We see, then, that the regular and long-standing use of territory by groups of bedouin, requiring, as it does, co-operation or at least tolerance between those groups that are in regular contact with each other, is expressed in terms of the common descent of groups of bedouin through men and also through women, producing a general statement of territorial situations in terms analogous with those of kinship. The more compulsive side of the formulation is that which is made in terms of descent through men alone, but of course we must not forget the fact that when the most strongly approved form of marriage is to marry back as closely as possible into the family, then common descent through men implies also common descent through women. The result of this preference for marriage with close kin is a concentration and a constant reconsolidation of the bonds which unite and preserve the continuity of smaller groups. The tradition of taking and giving women in marriage also affects the relations of the sections of a tribe. Even though all members of a tribe are classed as the descendants of a common agnatic ancestor, nevertheless those groups between which there is a tradition of shared maternal ancestry are considered to be more closely related than those between whom relationship can be counted on an agnatic model alone. This is so with the larger as well as the smaller groupings of the members of a tribe. The type of generalized relationship between primary sections of a tribe which results from the tradition of intermarriage is comparable with the generalized relationship that results from the traditional intermarriage of two different tribes, excepting that in the case of intermarriage between sections of the same tribe the relationship through women reinforces the more important agnatic tie instead of acting as weaker substitute for it. To take an example from the Manasir, the Manasir have five primary sections, each of which divides up again into secondary sections. Considered simply in terms of supposed descent through men, then all of these primary sections are at an equal structural distance from each other. Nevertheless, it is explained by Manasir that some of these primary sections are ‘closer’ to each other than they are to other sections of the same order of segmentation. The names of the five primary sections of the Manasir tribe are, Al Bu Rahama, Al bi-]l-Khail, Al Bu Mindhir, Al bi-]sh-Sha[r and Al Bu Hamir. Of these, the Al Bu Rahama and the Al bi-]l-Khail are considered to be ‘closer’ to each other than either of them is to any of the other three primary sections. Similarly, the Al bi-]sh-Sha[r and the Al Bu Hamir are closer to each
The Bedouin
95
other than they are to the others. The same differentiation of ‘closeness’ is made between various of the secondary sections which make up each primary section. For example, the Al Bu Rahama primary section has six secondary sections, namely, the Al Salimin, Al Al-Tu[aib, Al Wabran, Al Al-Taraif, Al-Taraiarifa, Al Makhazima. The Al Salimin are closer to the Al Wabran than they are to the rest. The Al Wabran, however, enjoy this greater ‘closeness’ with three of the secondary sections, namely, the Al Salimin, the Al Al-Tu[aib and the Al Makhazima. And the Al Al-Taraif and the Al-Taraiarifa are closer to each other than to the rest. This ‘closeness’ of tribal sections to each other does not necessarily correspond to physical propinquity at the present time. As I have said, the Al Bu Rahama and the Al bi-]l-Khail primary sections of the Manasir are considered to be closer to each other than they are to the other three primary sections of the Manasir tribe. On the other hand their centres are quite far apart. The Al bi-]l-Khail now live in the region between Abu Dhabi and Buraimi, whilst the Al Bu Rahama are centred on the Liwa Oasis, so that the centres of these two sections are [well over a] hundred miles apart. The Al bi-]l-Khail say that they lived near Liwa in the past but their palm trees were engulfed by the sand (a thing that easily happens in the Liwa area) and they had to move and make a living out of the carrying trade between the coast and Buraimi. There may, however, be a particular reason for preserving this special relationship. The Al bi-]l-Khail are near to Buraimi, which is the inland centre that the ruling family of Abu Dhabi most frequently stay and live at, whilst the shaikhs of the Al Bu Rahama are also shaikhs of the Manasir tribe as a whole. It is politically necessary for the shaikhs of the Manasir to visit the Buraimi Oasis every so often and maintain contact with the shaikhs of Abu Dhabi there, and the Al bi-]l-Khail, as a section of the Manasir tribe, have no shaikhs of their own who might stand as rivals to the shaikhs of the tribe. To sum up at this point, we see that bedouin describe the pattern of tribal politics in terms of one dominating principle, the principle of agnatic descent. All the members of a tribe are supposed, in principle, to be descended in the male line from the founder of the tribe. Descent from the tribe’s founder has branched out in the past into divisions through descendants of his, who are the ancestors of the tribe’s primary sections. In similar fashion, the lines of descent from these ancestors of the primary sections have branched out into subdivisions through the ancestors of the secondary sections, and, branching again from these ancestors of the secondary sections, come the ancestors of the various kins. Being the smallest political units, these kins are the most compact.
96
Shaikhdoms of Eastern Arabia
According to the terms in which the system is conceived of, looking backwards in time we find the nearer the ancestor is who unites any two groups, the more strongly are they united. The kins that make up a secondary section have closer ties with each other than they have with any of the other kins of the tribe. Similarly, the secondary sections that make up a primary section have closer ties with each other than they have with the tribe’s other secondary sections. The primary sections are joined together into the tribe, a major group separate from other tribes and ready to unite against these outsiders because all the members of the tribe share descent from the tribe’s common ancestor. This paradigm, ordering groups within a tribe according to analogy with a family tree, represents the political behaviour of tribal groups as generalized in conformity with one basic principle. But, important though it is, the agnatic principle is not absolute. It has to be modified by others. Firstly, when the parts of a group divide and separate from each other, then physical distance produces, in the course of time, political distance. Secondly, in spite of having supposedly separate agnatic origins, tribes which are neighbours and are on friendly terms with each other are able to develop another kind of relationship, again analogous with an aspect of kinship relations, in terms of descent through women. They become, in a political sense, each other’s ‘mothers’ brothers’ kin’. And, moreover, this special position of mothers’ brothers’ kin can serve to express a special closeness, over and above the agnatic system, of agnatic groups within any particular tribe. The agnatic paradigm, therefore, is a good deal more elastic than it looks on first sight. When we find it described, in uncompromising terms, as being the system of a tribal community, this happens partly because agnation is indeed the most important single principle in the light of which people can give a generalized account of their political relationships, but also the description only takes this form because various other things are taken for granted. Apart from the division which follows upon the separation of agnates, and the drawing together which relationship through women can affect, there is a third matter, which is the way in which the whole affects the part. Where the whole is the tribe and the part is one of its sections, this principle is consciously associated with the working of the agnatic model. But when the whole is tribal society and the part is one of the tribes, the association is not made so readily, for the matter depends on the general values of the society which are part of a different context of consideration. In this latter case, the context is defined by the two related criteria of honour among men and duty towards Allah.
The Bedouin
97
The question of solidarity between the kins that ultimately make up a tribe, and the cognate issue of solidarity between a number of tribes that are neighbours and have a tradition of intermarriage, need to be considered in the context of the practical dangers and difficulties of life in the desert. In desert life, were every man’s hand truly against his neighbour, neither the man nor the neighbour could live, for neither would retain the means to continue that struggle against the harshness of the environment which is the main condition of survival. Here we need to consider some of the characteristics of the camel, as both milk giver and mount. The mobility of the camel, considered in its two aspects as an animal that carries raiders and also forms their booty, extends the area of ground over which co-operation needs to be practised. Bedouin depend for their nourishment and livelihood on camels and flocks. The harsher – and hence the more precarious – are the desert conditions in which they live, the greater is their dependence on the camel. They drink its milk, eat its flesh and weave its hair. Furthermore they need the camel to carry their families and possessions about as they move from place to place in the desert. The fact that the camel is both a hardy and a highly mobile animal enables bedouin to live in places which would not otherwise support human life. But since the camel needs only occasional water and can move at a considerable speed, it is both an excellent mount for raiders and an attractive booty. Riding on camels, raiders can make a sudden attack on the herds of their victims from a great distance. Travelling at speed, they can approach by unexpected routes where there are no wells. [Awamir who raid in the Empty Quarter say that a raiding party sometimes takes a special camel, gives it a huge quantity of water to drink, cuts out its tongue, and loads it with water skins. The cutting out of the tongue is supposed to make the animal expend less water, since it prevents it from eating. When the water skins are exhausted, the raiders get more water to drink by forcing a stick down the camel’s throat to make it vomit. The final stage is to kill the camel and drink the water that remains inside it. With camels for their mounts and for their booty, desert raiders can arrive from great distances and drive back with them at considerable speed the herds they have captured. Profitable as such raiding might be for some people in the short term, it is obvious enough that were warfare and raiding endemic and totally unchecked, the cumulative result would be not only that herds would be constantly changing ownership, but that the herd’s productive capacity would be destroyed and the survival of those dependent on their milk be put in jeopardy. Hence, were it not for the existence of some
98
Shaikhdoms of Eastern Arabia
social and political institutions capable of restricting raiding and guaranteeing some measure of security to groups already engaged in a struggle with a harsh environment, then the environment would speedily conquer and swallow up the raiders with the raided. That war and disorder in the desert lead to poverty is a fact of life that the bedouin themselves well understand out of experience. Factors of various kinds have to be reckoned with when considering the solidarity of the tribes. One is the number of units into which each group divides. Here those numbers which are nearest arithmetically to unity would seem, for reasons which can be explained later, to imply the greatest disunity. The [Awamir of the Trucial Coast have only two primary sections and are the most disunited of the tribes. The more common pattern is for each tribe to be made up of several primary sections, and for each of these primary sections to subdivide into several secondary sections. The implications of the pattern of division and subdivision are clear. When a tribe has a number of primary sections, and a conflict occurs between two of them, then the remaining primary sections are, in a sense, neutral to the dispute. Their agnatic segmentary loyalties do not draw them in to assist either side. On the other hand, the remaining primary sections are as closely connected with both of these sections that are in conflict as either of them is with the other. Apart from any matters of sentiment, which may be important, it is against the political interest of the other sections that the conflict should continue. Internal conflict weakens the tribe as a whole in relation to other tribes. The interest of the other sections is to bring what good offices and pressure they can bear in order to pacify the opponents. The same situation applies, and indeed applies even more strongly, should a conflict arise between two of the numberous secondary sections that make up a primary section of the tribe. Such a conflict weakens the primary section as a whole and hence all members of the primary section are concerned to bring the opponents to terms. The proliferation of groups of the same order of segmentation – that is to say, that a tribe should have not just two but several primary sections, and so on – would seem therefore to be one of the factors that contribute to tribal solidarity. The second factor to be taken into account whilst considering tribal solidarity is tribal leadership and the form it takes. So far I have discussed tribal structure only in terms of the balance and counterbalance of numbers. But the full political pattern of a bedouin tribe is more complex than this implies. Leadership has also to be taken into account, and amongst the bedouin leadership is associated not only with special,
The Bedouin
99
recognized talent but also with hereditary status. For all their egalitarian tendencies, bedouin are not all equal in their inherited power to lead. All tribes, and most primary sections of tribes, are led by families of shaikhs. From each of these families, the member most acceptable to the family and the tribe or primary section is recognized as being the leader – the shaikh. In both primary section and tribe, the office of shaikh is hereditary in one particular family. The hereditary shaikhs of one of the primary sections are also hereditary shaikhs of the whole tribe. Hence we meet among the shaikhs not only hereditary leadership but a hierarchy of hereditary leadership. Whereas the pattern of bedouin sectional groupings is roughly symmetrical, with numbers balancing each other, hierarchical groupings cannot be symmetrically distributed. The distribution of shaikhs on an asymmetrical pattern through the segmentary sectional groupings of a bedouin tribe is a form of distribution of leadership corresponding to the way in which the shaikhs stand partly within and partly above the segmentary opposition of their fellow tribesmen. By so doing, the shaikhs exert a significant additional influence towards tribal solidarity and cohesion. In considering tribal structure, therefore, we have to take into account two complementary bases, the terms of which are different. One is the segmentary group, a group built up of smaller groups, which are in potential opposition to each other, and each of which is itself built up of smaller groups in the same way. The other is the hereditary leadership of particular families in its relation to the various levels of tribal segmentation. When we take into account symmetry in the counterbalancing of equal groups and a hierarchy asymmetrically distributed through these groups, then we reach a pattern revealing enough of the relationship between its own parts to be called a structure. It is usual for a tribe of bedouin to have within it several families of hereditary shaikhs. These kins, as we may call them more precisely, are the leaders of the various primary sections of the tribe. One of these kins, as well as being shaikhs of one of the tribe’s primary sections, also occupies on a hereditary basis the position of being shaikhs of the whole tribe. The members of this kin are therefore shaikhs in two senses, leaders of one of the sections of the tribe in relation to the other sections and leaders of the tribe as a whole in relation to other tribes. Secondary sections in general do not have shaikhs, but the shaikhs of primary sections are, naturally, also members of secondary sections. Hence, in any dispute within a tribe, there are shaikhs involved who are not opposed to other members of the tribe simply as equals. They are
100
Shaikhdoms of Eastern Arabia
endowed with more influence and authority than the remaining groups which are otherwise, according to the simple pattern of tribal segmentation, in comparable positions. If two secondary sections within one primary section of a tribe are involved in a dispute, and the shaikhs of their primary section belong to one of the other secondary sections that are not directly involved in the quarrel, then they can claim to represent all the secondary sections not involved in the dispute in order to bring influence to bear towards a settlement. If they belong to one of the secondary sections involved, then they can either, as shaikhs of the whole primary section, claim the support of the secondary sections which are directly involved, or else, if they find themselves generally thought to be in the wrong, they can relinquish their claims in the dispute in the name of unity. When a dispute arises between primary sections of a tribe, each of which is likely to have its own shaikhs, then the principle of asymmetry relates to the position of the family who are shaikhs of the tribe as a whole. Whether they are directly concerned or not, the settlement of the dispute concerns them in relation to their position in the tribe as a whole. Hence, in either case, whether in a small scale or a large scale dispute within the tribe, there exists a family of shaikhs whose status transcends simple leadership of one segment of a tribe in opposition to another. Whether directly involved in the dispute or not, this family has something more than the dispute itself to consider, because its position relates to a stage one degree higher, in the building up of smaller units into a tribe, than the stage of segmentation within which the dispute occurs. Thus, the hierarchy of tribal leadership is something quite different from, for example, a military chain of command in which, at each stage of greater co-ordination, there exists a superior whose position is equidistant from all the leaders one stage below him – the leaders of the groups which his command unites. The position of a shaikh is a little more like that of a prime minister who also holds the portfolio of a particular ministry within the cabinet, except that all the cabinet positions would have to be hereditary. But this would still be a very remote comparison. If we consider the bedouin tribe in terms of its internal structure, we can see two variations from the simplest logical model of segmentation in which the division is into pairs. The first is that groups proliferate at each order of segmentation. The second is that the bedouin have leaders whose position is asymmetrical in the way just described. The first difference can be illustrated in the contrast between Figure 3.3 and Figure 3.4. The pattern of segmentation in a bedouin tribe and the
The Bedouin
Tribe
101
Opposed Tribe
Segments
Fig. 3.3 Simplest (binary) model of segmentation. Opposition between any two smaller segments results in (a) the internal uniting of each of the biggest segments that includes one and excludes the other, and (b) the opposition of the biggest segments so formed to each other.
Tribe
Opposed Tribe
Segments
Fig. 3.4 Model of segmentation with proliferation of groups among bedouin. Opposition between any two smaller segments results in (a) (as in Fig. 3.3), (b) (as in Fig. 3.3), (c) the smallest segment that includes both parties to a dispute being weakened by the dispute in relation to segments that include neither party to it, and (d) attempts to settle the dispute by those weakened by it and not directly involved in it.
102
Shaikhdoms of Eastern Arabia
asymmetrical position of the tribal and sectional shaikhs can therefore be seen to be complementary. Within the group, be it tribe or section, that he heads, the shaikh is the channel through which the pressure of third parties, injured by dissension but not directly involved in it, is brought to bear in order to reassert solidarity. But his authority is largely moral: he can persuade much more easily than he can command. In this sense, he has been spoken of as primus inter pares. His personal interests are, formally at least, no more important than those of any other member of the group he heads. Apart from the members of his own family, no smaller group which he leads is formally any more important than others within this larger group – for example, a primary section headed by the shaikh of the whole tribe is not superior to the other primary sections of the tribe. Thus, a bedouin shaikh does not simply head his tribe: his position is integrated with the positions of his fellow tribesmen at all levels. The system of leadership is election within a hereditary system. The fact that heredity determines candidature restricts competition for positions of leadership (i.e. of pre-eminence) to one family in each case. If the hereditary principle did not exist and any member of a section might be a candidate for the position of shaikh of the section, the result would be that segmentary groups were always vying with each other on segmentary lines, resulting then in the opposite of solidarity – disunity. But when heredity restricts competition, most of those qualified to support a candidate are unqualified to be candidates themselves. Where the shaikhs become a ruling family in the towns, competition is for something more than pre-eminence; it is for positions in which there is more power and authority and less leadership than among bedouin shaikhs. Competition for positions of power can be conducted by violent means more easily than competition for positions of leadership, since ‘consent’ is less important. The way in which the ideology of tribal society counteracts the fissile tendency inherent in segmentary tribal structure is illustrated with some precision by the conventions which control the blood feud. Feud is, in the first place, a way in which the hostility automatically brought about by the death of a tribesman at the hands of a member of his own tribe is restricted. Moreover, it is possible to extend the feud to cases where homicide of a member of one tribe is committed by a member of a different tribe. Here again, the effect of the feud is to restrict hostilities, which would otherwise set the two tribes against each other as wholes. Following this limitation of the social area of conflict, it often becomes possible to restore, or at least to patch up, relations by the
The Bedouin
103
payment of blood compensation, which the kin of the dead man accept from the kin of the man who has killed him. When a bedouin kills a member of his own tribe, the resultant hostilities are restricted to those four or five generational groups that I have called the kin. These are the groups of agnates whose members know their precise genealogical connection with each other. The two groups are made up of all the descendants in the male line from the great-grandfather or great-great-grandfather of the dead man, and all the descendants in the male line of the great-grandfather or greatgreat-grandfather of the man who killed him. As I have mentioned already, bedouin often do not know who their great-great-grandfather was. Where he is the nearest genealogical link joining his descendants together, some of those descendants may have separated from each other, forgotten their precise historical relationship, and no longer retain the ties of special common interest that kinship implies, having passed out into being no more than fellow tribesmen. The fact, therefore, that when bedouin are asked to define the groups involved in the feud they give these alternatives of those descended from the same great-grandfather or the same great-great-grandfather is of some significance. To this extent, the definition is pliable and able to cover actual social relations, excluding where necessary those which no longer apply in practice. When the social link that joins together as kin the descendants of a common great-great-grandfather ceases to exist, then also the ancestral link which records and signifies this common interest disappears. This is why it is of some theoretical importance to distinguish those personally remembered ancestors who mark the internal relations of kin from the tribal ancestors, with their unknown numbers of generations separating them, who mark the segmentary relations of tribal sections. Among the bedouin, the kin are not, in any strict sense, a segmentary group. This is shown by tribal convention governing what should happen if a man kills his own kinsman. Here feud, in the normal sense of the word, is impossible and the payment of blood compensation, both in theory and practice, is out of the question. Should a man happen to be killed by his agnatic first cousin, for example, then his son may revenge him on the man who killed him. But there is no question of the total group of the first cousin, his brothers and all their sons being suitable objects of revenge. In principle, the bedouin speak of such a killing as being such a horrible disaster that nothing can be done about it institutionally. The kin as a whole, not as an internally segmented group, are answerable for each other in feud, and so, if a case of homicide
104
Shaikhdoms of Eastern Arabia
occurs within the kin itself, then the situation is not one to which feud applies. When a man is killed by his own fellow tribesman, then all the members of his kin and all the members of the kin of the man who killed him are involved as principals. Any man of the kin of the dead man has a right, and a duty, to kill any man of the kin of the man who killed him. Women, and boys below the age of puberty, are not considered suitable objects of retaliation, the boys because of their ‘innocence’ and the women because they are generally classed as being defenceless and fighting is not normally part of their business. I did, however, come across one man who had recently killed a man and a woman (his sister), who were together, in feud, but he claimed that he killed the woman by mistake. But it was generally thought to have been intentional since two men from their kin were being avenged. I suspected that he had hoped to remove the evidence by preventing her from being a witness. When one man kills another, it is not in the first place considered important whether he did it by accident or by design. Practically speaking, among the bedouin, there is no institutionalized way of deciding formally whether a man was killed by pure accident, by carelessness or rashness or the various categories of responsibility up to premeditated murder. On the other hand, sometimes the fact is self-evident even to the kin of the dead man. Such a case was one where, during a war, a band made up of tribesmen were sleeping in the desert, keeping their loaded rifles at their sides because they feared a surprise attack. [Peter Lienhardt here leaves a gap in his manuscript to ‘add detail’. This was not completed, and one can only infer that one of the tribesmen was killed unintentionally.] In such a case, the kin of the dead man are, at least, less eager to retaliate than they would be were the killing intentional, and they have much stronger justification for settling the implied feud by accepting blood compensation, since in doing so they are unlikely to suffer much shame to their courage and to gain honour in another way for their humanity. The killer and his kin are collectively responsible for an act of homicide. To avenge the dead man, any of them may be killed, so long as it is a youth or a man, and not a woman or child, and bedouin strongly deny that there is any preference for whom to kill. They make one exception, usually that it is better, if possible, to retaliate by killing the man who killed their own kinsman. But apart from this no distinction is made amongst the offending kin in any qualitative sense, whether it be according to closeness of relationship to the killer or according to personal importance. In retaliation, there is no preference for killing, say,
The Bedouin
105
the offender’s son rather than his brother’s son. Nor, if it is a leading man who has been killed, is there any preference for killing in retaliation a leading man from the offending kin. The unity and interchangeability of the members of the kin, vis-à-vis the outside world, is expressed in a proverbial expression that says, ‘Preferring one to another only applies to the offspring of goats,’ at-takhayyur f i [ayal al-ghanam. This means that, when choosing a goat to kill and eat, people select the one they think will do best, but human standards are different. In retaliating for a human death, men are exacting a human life, not calculating and choosing something they want to eat. The question of who, among the kin of the dead man, should retaliate for his death is somewhat different. Here the closeness of relationship to the dead man is regarded as important. Though the dead man’s kin have all suffered the loss, those closest to him are regarded as having a stronger right and a stronger responsibility to perform the deed of retaliation. If they fail to do this, then the responsibility for reasserting the honour of the kin falls on the dead man’s more distant kinsmen, but it would be wrong for the latter to act prematurely, since it would by implication be denying the prior right and, more important, the manliness of the close kin. The most telling and suggestive fact, in relation to the question of segmentary division and tribal unity among bedouin, is that the convention of the feud forbids the tribal section of the offending kin to unite in defence of their member against the tribal section of the dead man, and, similarly, the tribal section of the dead man do not need to and are not expected to unite in support of his kin against the dead man’s kin. Within the conditions I have described, the members of each kin are united and regarded as being interchangeable. But, apart from this, the convention of the feud obliges the other members of the tribe, whether more closely or less closely related to either of these two kins, to remain neutral. Moreover, tribal convention forbids the offending kin to stay and fight the matter out with the kin of the dead man. The offence is theirs, and it is an offence against the tribe as a whole as well as against the dead man and his kinsmen. The offending kin must, therefore, flee and seek protection outside the tribe (the protected man is called in the Trucial Coast zabin and in other Arab countries he is called dakhil). The convention leads to the removal of the dispute outside the spatial area in which the members of the tribe normally move. Hence the convention of the feud works in two stages: it limits the social area of hostility within the tribe to the two kins, so that the segmentary divisions of the tribe are not called into action, and then it
106
Shaikhdoms of Eastern Arabia
removes one of the kins socially and spatially from the tribe, thus again working towards the re-establishment of tribal unity. Though it is never possible for a kin exposed to retaliation in a feud to seek lasting protection within its own tribe, in some tribes it is possible for them to seek temporary protection, under the guarantee of which they can escape to a protection of a more lasting sort. It is interesting to observe the way in which the convention works here, for what might seem to imply a reassertion of segmentary involvement proves in fact to be quite the opposite. Where it is possible to give protection within the tribe, those who are able to do so are not those who are close to the offending kin but, on the contrary, they are those fellow tribesmen, and those only, who are remoter from the offending kin as the latter are from the kin of the dead man. One tribe in which temporary protection is possible in this way is the [Awamir, who, as I have said, are one of the less united tribes of the Trucial Coast. They have – and this may indicate their disunity – only two primary sections. Each of these divides up into a number of secondary sections, which are themselves made up of numerous kins. Among the [Awamir, if a man kills a member of his own secondary section, then he and his kin can seek temporary refuge with any of the other secondary sections within his own primary section of the tribe, or with members of the opposite primary section. If, however, a man kills a member of one of the other secondary sections within his own primary section, he and his kin cannot seek refuge with any of the other secondary sections which make up his own primary section. They can only seek temporary refuge with members of the opposite primary section of the tribe. And if he kills a member of the opposite primary section, then he cannot ask for temporary protection anywhere within the tribe. Among the Manasir, members of a primary section do not give protection in feud to kins from among themselves. If the two kins involved in the feud are from the same primary section, then the offending kin can seek temporary protection in one of the other primary sections. If the two kins are from different primary sections, then the offending kin cannot be protected at all within the tribe. Thus within the Manasir the principle is the same as with the [Awamir, but its implication is less wide. The convention of the feud therefore specifically bars from involvement those groups within the tribe that might be drawn into hostile feelings by their closer segmentary ties. Thus the feud is prevented from developing into segmentary fighting and is limited to the two kins, and where any group can give even temporary protection to the offending kin it can only be a group whose position in the
The Bedouin
107
segmentary order of relationships is patently neutral. Neutrality is not even guaranteed by an equal distance from the two kins involved in the feud. As we have seen, some groups which are regarded as being at the same distance from each other in agnatic segmentation are nevertheless said to be closer together than they are to others of the same order of segmentation because of links through women. The conventions of the feud decree that if there is no one in the tribe who is positively further from both of the kins that are principals in the feud than the two kins are from each other, then no one can offer even temporary refuge to the offending kin. The conventions of the bedouin feud thus restrict the area of hostility that follows from a case of homicide and prevents it from extending itself throughout the segmentary groups, some of whose members are involved, and drawing them into enmity and bloodshed. Within the kins concerned, however, a feud can in theory continue indefinitely. The achievement of retaliation by killing a member of the offending kin does not end the matter but merely reverses the roles of the two kins. The offending kin, by virtue of the death of one of their members, become the offended kin and vice versa. The only way in which the feud can be settled, if it is to be settled at all, is when the kin of the dead man can be persuaded to waive their right to retaliation by accepting compensation, diya.7 In the 1950s, the compensation for homicide was calculated in rupees, the normal currency, and was relatively low, amounting for a man to about sixty pounds sterling. If the kin who have to pay it are unable to do so, they can go begging assistance anywhere in the tribe. To give such aid is a mark of generosity and a good work. The difficulty of accumulating the sum is never, therefore, insuperable. The problem lies in the other direction, that of persuading the offended kin to accept it. This is a matter of lengthy social pressure and is sometimes not successful at all. Accepting money in compensation for the death of a kinsman always involves some measure of shame, though as we have seen the shame is reduced when the death is well known to have been the result of an accident. Moreover, the fact that compensation has been accepted does not necessarily mean that the feud is at an end, but only that there is a possibility of avoiding further violence. The Trucial Coast bedouin say that even when compensation has been accepted there may still, years afterwards, be a member of the kin of the dead man who is not satisfied. He may be a young man who was a child when the kin agreed to accept the compensation. He may be excitable and eager to make a reputation for himself as a hero who cannot tolerate the shaming of his kin, and so
108
Shaikhdoms of Eastern Arabia
may go of his own accord and retaliate by killing a member of the other kin. Thus, either because compensation has been persistently refused or because its acceptance is no final guarantee of security, kins who flee in feud often flee for good and never return to the tribe. Even after the acceptance of compensation and the passage of many years, members of the two kins cannot meet without danger. But sometimes, when the generations who were alive at the time of the original killing have passed, the offending kin is able to settle the matter finally and with full assurance by taking in marriage a woman from the kin of the dead man or a woman in some way related to them. We may now add a third category to the status situations within which homicide can occur. The closest status situation, as we have seen, is where the dead man and the man who kills him are related to each other within the status of agnatic kinship. If a man kills another member of his own kin, then there is no convention to govern what should happen subsequently. It will be remembered that there is one situation in which bedouin (and also settled people) think it honourable to kill a member of one’s own kin, namely, when it is a woman who has injured the collective honour of the kin by not conforming to the strict convention of chastity. The kin are too closely identified together for there to be any regular way in which they could subdivide into groups over a case of homicide. The second category is that of the tribe, and, as we have seen, homicide within the tribe, bound together as all its members are in tribal loyalty, automatically results in feud. This feud automatically restricts hostilities to two kins, one of which must flee, and the conventions of the feud are such as to avoid as far as possible the extension of hostility into the segmentary divisions of the tribe. There is no doubt that homicide puts a strain on tribal solidarity. If relations within the tribe are very strained already, then homicide can be one of the causes of the tribe’s dividing and abandoning its common loyalty, but the convention of the feud is certainly a way of reducing strain by limiting the area of conflict – so long as tribal solidarity is strong enough in the first place for it to be possible to preserve the tribe as a united group. The third category within which homicide can occur is between members of separate tribes. When homicide occurs between two different tribes, then there is no automatic way in which the conflict is restricted to groups within the tribe. The incident automatically sets the respective tribes of the killer and the dead man against each other as wholes. The whole tribe, bound by tribal loyalty, on each side is the status group involved. The automatic implication of homicide in such a situation is war. But even here
The Bedouin
109
the conflict can be limited and war avoided. I have already described how a close relationship can exist between tribes outside the terms of shared agnatic descent. Such a relationship of traditional friendship and intermarriage affects the conventions of protection when a case of homicide has occurred in one of the tribes. In such circumstances, the friendly tribe acts as though it were an extension of the other, one step further away in the agnatic tribal system. A kin fleeing for refuge in the feud from another kin of its own tribe can only be granted temporary, not permanent, protection from the tribe related to its own by traditional friendship. Hence, in this situation, the position of the friendly tribe is parallel to the position of one primary section of a tribe affording temporary protection to a kin from another primary section, one of whose members has committed homicide in his own section. Here, then, is another conventional means for preserving solidarity: in this instance, the solidarity is that which obtains in circumstances of traditional friendship and intermarriage between tribes. The way in which warfare is avoided between friendly tribes following on a case of homicide where a member of one of the tribes kills a member of the other is by the immediate payment of compensation at a tribal level. This gesture dissociates the offending tribe as a whole from the offence committed by one of its kins. By accepting it, the tribe one of whose members has been killed compounds the homicide as a tribe, with the result that hostility is restricted, as in the case of a killing within a tribe, to the kins of the killer and the killed. This inter-tribal compensation is a different matter from the compensation that may eventually be accepted by the kin of the dead man. Inter-tribal compensation is paid to the shaikh of the offended tribe by the shaikh of the tribe that has committed the offence. The sum is equal to half the sum that has to be paid for blood compensation between kins. It is called muhtasham, ‘respect money’. When able, the shaikh of the offending tribe may recover the sum from the kin actually responsible for the offence. Respect money, unlike blood compensation, is an immediate payment, and if offered it has to be accepted. In this sense, therefore, there is a great difference between blood compensation and respect money. There is always some measure of shame involved in accepting the former, whereas it would be disgraceful to refuse the latter. The acceptance of respect money removes the ‘blame and answerableness’, lawm wa-luzum, for the killing from the tribe that has paid it. The convention of respect money extends beyond situations where traditional friendship exists between tribes. It can be paid between any two tribes (unless, of course, they are formally at war already when it
110
Shaikhdoms of Eastern Arabia
would be meaningless), though it is said to be less likely to be offered where relations between tribes are already seriously hostile. The payment does not automatically prevent war: it simply prevents automatic war. But after respect money has been paid, the behaviour of the man who committed homicide and of his kin depends on the closeness of relations between the two tribes. When the tribes are on such friendly relations that they are regularly mixing together in the same areas, the offending kin have no security from retaliation if they stay with their own tribe. They must go, therefore, and seek protection at a distance. Where the tribes concerned are separate and not friendly, then the offending kin are able to stay with their own tribe. The arrival of a group from the other tribe in the territory is an unusual and suspicious thing, and if they prove to be a party of the dead man’s kin seeking to retaliate, they can be turned back. Moreover, if the shaikh of the tribe who has paid respect money wishes to avoid further troubles, and he learns that a party of the dead man’s kin is going to try to retaliate, then he can warn the shaikhs of the other tribe of the danger and enable them to take precautions against it. If we consider the conventions of the bedouin feud, as I have described them so far, in a structural sense, one general principle is immediately clear. The remoter are the tribal connections of the two individual men at the root of the incident, namely the dead man and the man who killed him, the more important is the position of the shaikhs if warfare is to be avoided and hostility is to be restricted to the respective kins who are most immediately involved. From the closest to the most remote relationships within which homicide can occur, there are various stages which can be distinguished analytically. The closest group is the agnatic kin, and, as we have seen, this relationship is thought to be so close that there is no formal conventional way of behaving in such a case. The next stage, analytically, is the tribal section, within which the members are united by an agnatic relationship analogous with kinship. Here, temporary protection can be given by another section of the same tribe. At this stage, the shaikhs become involved in their character as leaders of tribal sections, since people seeking protection can be most secure if they go to the most powerful protectors they can find, and if they do not manage to reach them first they are likely to be taken to them. Hence, the tendency is for this temporary protection to be granted by the shaikhs of the other section. The third stage is the tribe, again bound by ties analogous with those of agnatic kinship. In a case of homicide between two primary sections of a tribe, no group within the tribe can give temporary protection,
The Bedouin
111
but another group more remotely related can. This is a tribe related to the tribe of the offending kin by traditional friendship, the relationship with which is spoken of in terms extended and generalized from those of common descent through women. The members of each tribe are the political ‘mothers’ brothers’ kin’, each to the other. Here again, it is the shaikhs who are likely to be called upon to give temporary protection. The next stage outwards is where a homicide occurs between these two friendly tribes. This is the stage which combines the end of status relationships and the beginning of contractual relationships. In formal terms, the relationship through women is a much weaker one than the agnatic tie in kinship. A man does not retaliate for the killing of his mother’s brother or his mother’s brother’s son unless the dead man is isolated and has no agnatic kin to avenge him. In a formal, agnatic sense – this being the dominant criterion in terms of which political relationships are formulated – the two friendly tribes are separate from each other. This formal separation means that the containment of hostilities within the kins of the dead man and the man who killed him is not automatic. The shaikhs’ gesture of paying respect money here becomes necessary. And there is no formal need for the offending kin to flee: only a practical one. The final stage is where the tribes are fully separate. Here, respect money is a condition of preserving peace between the tribes when a member of one tribe kills a member of the other, and the offending kin does not need to flee. Here also, one of the tribes can give permanent protection to a kin fleeing from a feud in the other. Within this final stage of separation, there are degrees of good and bad relations. The worse the relations are between the two tribes, the less likely it is that respect money will be offered, and, conversely, the more secure is the protection that can be given by either of the tribes to a group fleeing from a feud in the other. Bedouin in the Trucial Coast explain the difference between agnatic tribal loyalty and friendship in terms of quality. They say that friendship involves the quality of ‘face’ whereas tribal loyalty does not. An explanation of the difference between ‘face’ and tribal loyalty is that ‘face’ comes from the human face, whereas loyalty is from people’s hearts. I translate the well-known Arabic tribal expression wajh (locally pronounced waih) as ‘face’, which is its literal meaning, with misgivings because it brings to mind such English expressions as ‘saving one’s face’ and the foreign concept naturalized now into English in the expression ‘loss of face’. Though the Arab idea of ‘face’ has some affinity with these usages, in the sense of being incorporated in expressions concerned
112
Shaikhdoms of Eastern Arabia
with honour and reputation, there is nevertheless, as might be expected, a good deal of difference. The most important Arabic expressions connected with the figurative use of the word are, in literal translation, ‘May Allah whiten your face,’ used in praising or thanking a man for his honourable behaviour; the converse of this, ‘May Allah blacken your face,’ used in reviling a man for dishonourable behaviour; ‘There is no face between you and me,’ a declaration of serious hostility expressing the severance of any friendly or even neutral relations; ‘The face between you and me is broken (or, untwisted),’ a more formal version of the last expression, used in a declaration of war (‘untwisted’ referring to the untwisting of a rope); and, ‘in my face’, which means, ‘under my protection’. Where the agnatic tribal loyalties, which are ‘of the heart’, end, ‘face’ begins. A man who, whether because of feud or for other reasons, has been granted protection is ‘in the face of’ his protector. On leaving the state of protection, it is customary for him to acknowledge that it has been fully and honourably given by saying to his protector, ‘May Allah whiten your face.’ As this implies, to grant protection to a stranger is an honourable deed. The protector maintains or enhances his own reputation by demonstrating his power, his willingness to commit himself to an obligation which may be arduous and dangerous, and hence his capacity for acts of moral and physical courage. Protection depends mainly on force and power. It is unlike sanctuary, for there is no question of an enemy’s ‘respecting’ the state of protection, or of his being restrained by piety or the fear of supernatural punishment, let alone by shame. What there is to restrain him is the knowledge that if he succeeds in getting through the protector’s guard and taking vengeance he is calling down on his own head and the heads of his family the vengeance of a powerful adversary and possibly an implacable one. In the Trucial Coast, protection is no automatic right. But protection involves honour, and so once it has been asked it is difficult, and to some extent shameful, to refuse to grant it. Since it depends mainly on force and power, a weaker man cannot effectively grant lasting protection against a strong one. All he can do is to protect the fugitive as best he can for a short period and try to smuggle him away to a place where he will be more secure. Custom also sets a limit to the period of protection which a powerful man needs to grant in order for there to be no reproach against his honour. The formal state of protection lasts only for one year, though at the end of this period the man who has been protected may still be allowed to stay. Protection has a high place in the bedouin scale of values. The conventions relating to the granting of protection make no difference
The Bedouin
113
in kind between the powerful and the weak. To some extent any man, weak or strong, devalues himself if he fails or refuses to grant it, in that others may reproach him for cowardice. Allah will not be asked to whiten his face on this occasion. Society, after all, is made up of people who might, any one of them, some time need such protection, even as a result of some accident over which they had no control and which they would dearly have loved to avoid. But although the conventions of protection make no difference in kind between man and man, practical standards of behaviour and expectation take into account differences of quality. The more important and powerful a man is, the stronger is his protection. In this sense, the working of the system is co-ordinated with ideas and values of hierarchy. And at the head of that hierarchy are the greater shaikhs. In the society of the bedouin, as in other societies where feud exists, the institution of feud can be interpreted analytically as limiting the range of political fission implied in any case of homicide. It is not homicide between total strangers that leads to feud. Feud results from cases of homicide which occur between members of groups which are, in the first place, bound by significant political ties. As we have seen, the conventions of bedouin feud limit the area of open hostility to the kins of the dead man and the man who killed him, and thus provide a corrective to the segmentary principle of tribal structure. Here, feud within a tribe is a way of reacting to homicide whilst preserving the solidarity of the tribe as a whole whereas, according to the standards of ‘closer loyalty’ within tribal sections, the sections concerned would otherwise be united in enmity against each other as wholes. According to the bedouin conventions of feud, those united by closer loyalty with either of the two kins who are principals in the feud must stand neutral and outside the dispute, and are prevented from granting any protection whatsoever. Through the institution of respect money in the Trucial Coast, the institution of feud is capable of extension from cases of homicide that occur within a tribe to cases that occur between two different tribes. Here, the alternative of respect money and feud preserves inter-tribal solidarity and provides an alternative to war. If we accept that feud has the function8 of restricting hostility and thereby preserving political solidarity, this means that we can by no means define the limits of bedouin political solidarity, as expressed in bedouin institutions, as being coterminous with the limits of tribal loyalty. The limits of institutionalized relationships extend over a much wider range. The institutionalized customs of protection suggest how broadly they extend.
4 Towns and Maritime Activities
All the towns of the Trucial Coast lie on the seashore and have depended for their existence on the sea rather than the desert. Nevertheless, even now, in spite of all the new building that is going on, they retain a desert flavour, different from those Middle Eastern towns and cities where a large peasant population intervenes between bedouin and urban life and provides the strongest element in urban recruitment. In Eastern Arabia the desert sand runs straight down into the sand of the beaches: the coastal towns are desert towns and ports at the same time. There are towns where bedouin move with ease and confidence – bedouin who in the past played a part in the life of the sea, just as townsmen made sorties with their camels into the desert. Towns have grown up in places where there have been anchorages and some fresh water of however poor a quality. Some of them are sited in positions which afford natural defence against land attack. The first requirement of permanent settlement has, of course, been fresh water, but it has clearly not been essential that this should be either good in quality or ample in quantity. According to an early report, perhaps near enough to the event to constitute reliable history, the town of Abu Dhabi came into being with the discovery of drinking water there about 1760.1 There is only enough of this water for domestic uses and in flavour it is as bad as some of the worse bedouin wells. In the 1950s, before any water distillation plant had been set up, sweet water to drink was being sold in the cafés in Abu Dhabi at an anna (roughly, then, a penny) a glass. This drinking water for the cafés and some of the houses was imported from the Persian Coast or from Qatar by launch in oil drums and retained some of the flavour of the containers, but it was still much preferable to the local water. Such elaborate efforts to obtain good water were not unusual in the towns of the Gulf. 114
Towns and Maritime Activities 115
In Bahrain, some of the drinking water used to be produced by men diving with water skins to springs on the seabed. And in Kuwait, where richer people, like those in the Trucial Coast, collected rainwater in the wintertime and stored it in underground cisterns for drinking as a luxury in the nights of Ramadan, drinking water was also brought by boat from the higher reaches of the Shatt Al-Arab. As in the case of Abu Dhabi, this meant that the water had to be carried for about a hundred miles. J. H. Stocqueler, a traveller who visited Kuwait in 1831, remarked that Kuwait within the walls yielded nothing and that the water was ‘far from sweet’. 2 He found it difficult to conjecture how such a site came to be chosen for the establishment of four hundred families, which is what the population then amounted to. Nevertheless, Kuwait’s population was to increase more than tenfold during the course of the nineteenth century. No town in the Trucial Coast is quite as barren as was Kuwait, where in the 1950s I saw only one date palm and a few eucalyptus trees. But the date palms one sees at Abu Dhabi, and even Dubai, give a deceptive impression since they suffer from the salinity of the water underground and therefore do not produce well. Substantial water supplies, then, were no condition of urban settlement. Natural defence against land attack was clearly of importance in the siting of towns, as may be seen from examining their geographical positions. There is no high ground for defence, for the coastal land is almost entirely flat until the mountains come down to the sea beyond Sha[m, but some natural defence was provided by water. Little peninsulas of sand, and islands so close in to the shore that the intervening stretch of water could be forded or easily swum, nevertheless provided some measure of protection. Shandaqah, the quarter of Dubai where the old houses of the shaikhs are, is at the tip of a sandy peninsula at one side of the Dubai creek and used to be cut off completely by high tides. Dairah, the quarter of Dubai at the other side of the creek, is also a peninsula, as was the now eroded site of the main part of Ras Al-Khaimah. Abu Dhabi is an island so close in to the shore that it was possible, with the simplest techniques, to join it to the mainland by means of a causeway. But even very shallow water, or simple mud, is a defence against men mounted on camels, for the form of the camel’s foot that makes it so sure-footed in dry sand makes the animal slip or stick in wet sand and especially in mud. When approaching slippery places, camel riders have to dismount and lead their animals, and in the salt marshes which are common in the coastal area one sees the bones of many camels that have died there after slipping and breaking their legs.
116
Shaikhdoms of Eastern Arabia
Man-made defences, additional or supplementary to those provided by water, were produced by building towers and forts, many of which not only survive but are still manned. As recently as the 1930s, when an airstrip was established at Sharjah, it was thought necessary to build the little rest house there in the form of a fort. In Abu Dhabi, the fort protecting the causeway I have just mentioned is still jealously guarded. The simplest form of defence tower, like those protecting Rams from the Shihuh mountaineers, is only about twenty-five feet high and three or four yards square. The flat roof, from which guards can keep a look out and fire their rifles from behind a parapet, connects with a room below it where the guards keep their food and water and can shelter from the sun. The only means of entry to this room is by climbing up a rope let down by those within. Such towers would have provided no impressive defence on a European terrain even before the invention of gunpowder, but their weakness corresponds to the difficulty of producing heavy weapons of attack or, even if such existed, of transporting them across the sand. To judge from the positions of the towns, the most telling physical factors which helped them to flourish were that they should have good anchorages and that access to them should not be through sandbanks between which it was difficult to navigate. It should be remembered, however, that none of the local boats was too big to pull up on to a beach from time to time, and so even good anchorages were not allimportant for home-based seafaring. Much depended on the kind of seafaring that was undertaken. For foreign dhows which were merely calling in at a port, safe anchorages in creeks protected from the sudden and violent storms of the Gulf were a clear advantage. The same was true for dhows plying the short distance between the Trucial Coast and the coast of southern Persia, and trade with the Persian coast was, and remains, an important part of the local economy. Creeks, the safest of local anchorages, are not, however, permanent; they change and silt up in the course of time. During the present century, such silting up of the creek has destroyed the trade facilities of Sharjah, at one time the most important of the ports of the coast. Trade from Sharjah moved to Dubai, but Dubai too was subsequently threatened by the same disaster, and the silting up of the mouth of the Dubai creek was only prevented by expensive modern dredging which could not have been accomplished without foreign technical expertize. By that time, the great bulk of trade goods was carried by steamship and landed by lighter, rather than being carried in dhows, but the same factors applied nevertheless for the lighters were of no deeper draft than the
Towns and Maritime Activities 117
dhows. Even more recently, on this shifting coastline, a part of the town of Ras Al-Khaimah has disappeared into the sea. The shifting of the sands of the coast, blocking some ports and eating others away, is perhaps a reason why the coastal towns as we know them have had such a short history. Archaeologists are now seeking for the remains of earlier towns in places no longer inhabited, and one of these, apparently of considerable antiquity, has been discovered in Abu Dhabi territory. The ancient port of Ommana, referred to in the second century Greek travel document The Periplus of the Erythraean Sea,3 which may have been somewhere within this area, has yet to be found and may by now be well under the sea or far inland. The kind of anchorage available played some part in determining the types of seafaring occupations practised in various places. Seafaring provided three general sources of livelihood, namely fishing, pearl fishing and deep sea trading. Fishing boats are easily drawn up on the shore from day to day, and even the bigger pearling boats could be beached when they returned from this seasonal occupation. Indeed, they had to be, in order to be caulked, repaired and preserved. It was the carrying trade which particularly needed good anchorages. Ras Al-Khaimah, Sharjah and Dubai, all of them with safe anchorages, prospered one by one in long-distance trading, whereas Abu Dhabi, a state which was never less in power than any of the others, had little income from trade as compared with the profits it received from pearl fishing. On the other hand, pearl fishing suffered no disadvantage from good anchorages at home, and pearl fishing was carried on from all the coastal towns and villages, whether they had anchorages or not. Hence, it was long-distance trading which determined the balance of power at sea during the early nineteenth century, before the shaikhdoms were prevented from engaging in sea warfare, and here Abu Dhabi was at a disadvantage as compared with her rivals. It will probably not be long before none of the buildings or even the streets of the coastal towns and villages preserve any sign of traditional life. Dubai and Abu Dhabi, with their high modern buildings and their streets widened and cleared for motor traffic, have already changed a great deal, and one can judge from modern Kuwait, where scarcely a single old Kuwait house has been left standing, what further developments the future holds in store. If we consider some of these coastal places as they were a few years before oil was discovered in the Trucial States, more continuity with the past appears. Still, we must bear in mind the earlier crisis, in this case a crisis of poverty, which had already occurred, changing the traditional situation and bringing much of the
118
Shaikhdoms of Eastern Arabia
coast into decay. Pearl fishing had by then dwindled to almost nothing, the dhow trade had greatly declined and steamer traffic had helped to concentrate almost all major trade at Dubai, whilst many of the people of the coast had moved to places further up the Gulf which already had oil and so offered greater opportunities of employment and business. Nevertheless, thirty years’ poverty produces fewer changes than ten years of extreme wealth. The latter produces a vigorous, positive reconstruction in which survivals from the past are rapidly swept away, whereas the former only brings about a slow decline and decay. So long as we bear in mind that this decline had already occurred, a description of two or three of the coastal towns and villages in the years just preceding the discovery of oil will show something of what traditional life was like. Abu Dhabi had by that time become so small that one hesitates to call it a town. The population was probably about three or four thousand. Here, the effects of the decline of pearl fishing were most conspicuous in the ruinous state of the more solid buildings, buildings which had been the houses of the few pearl merchants who had lived in the town. Most of the people of Abu Dhabi lived in houses made of date sticks, the cheapest and also the most perishable sort of house. Such flimsy houses, generally called barasti, have various local names according to type such as loga, sebli, khaima and [ar ish, this last being a sort of house occupied only for the summer because they are cooler. Recently some effort has been made in various towns of the Gulf to get rid of palm houses because, with the use of paraffin stoves for heating, the fire risk of great compact areas of them is very considerable. A very small palm house cost, at that time, about a hundred rupees to build and then about forty rupees a year in repairs. These houses were the nearest thing to a bedouin tent among the settled people: they could if necessary be dismantled and removed to another place. Their occupants did not usually own the ground they were set up on, but had a right of occupancy deriving from the shaikhs. The right, once given, was an established one which could not arbitrarily be taken away, not, at least, so long as the shaikhs conformed to local canons of justice. It was thought wrong to make people move without their consent unless they had misbehaved themselves, and when it was necessary to persuade them to move they were often given presents to induce them to do so. But once people did move from this land their right in it lapsed. This is a matter of some significance in relation to the old customary law which applied throughout all the Trucial States, for the custom was that if a family left the state and went to live elsewhere, their house, or at least the land it was constructed on, was sequestrated. In the case of
Towns and Maritime Activities 119
palm houses, so long as the family left peacefully, they had a perfect right to remove or sell the materials. Solid houses, if they had not been prudently sold in advance, were simply confiscated, though they might be handed over to the family again if they returned. Conversely, a person newly arrived in one of the Trucial States and wishing to settle there received a plot of land from the shaikhs and some financial or material assistance towards erecting a house on it. The custom of sequestration was no longer followed in the 1950s except in political cases. Some people had bought land in the towns and villages from the shaikhs or from each other and had documentary evidence of their title to it in the form of tiny scraps of paper which they carefully preserved and sometimes pretended to have mislaid lest they should be made away with by more powerful or unscrupulous claimants. But the lack of any general registration of titles to land was, comparatively speaking, only one of the minor deficiencies of the administrative system, where very little record was kept of anything. Abu Dhabi preserved evidence of its tribal nature in the names of its own quarters. The local Trucial Coast word meaning ‘a quarter’ in a town, far ij, is in itself the same word as that for a bedouin encampment, and the quarters of Abu Dhabi were called ‘the farij of the Rawashid’, ‘the farij of the Rumaithat’, ‘the farij of the Qubaisat’, ‘the far ij of the Gumzan’, ‘the far ij of the Al Bu Muhair’ and ‘the farij of the Sudan’, all of these names being the names of tribal sections of the Bani Yas tribe of Abu Dhabi. Between some of the different quarters of the town were patches of open sand, and so there was room for quarters to expand on tribal lines. On the other hand, people were under no obligation to live in their own tribal quarter and some of them did not, so that a map of the town was by no means a simple map of tribal population. Conspicuous as the Bani Yas tribal sections were in the names of the town quarters, Bani Yas tribesmen were not at all conspicuous in trade. In the market, which was in the Qubaisat quarter and had about forty shops all told, not a single shopkeeper was a member of an Abu Dhabi, or even a Trucial Coast tribe. They were all Persians or Perso-Arabs with the exception of one long-established Indian. There was, however, one local Arab family of contractors who were very successful, it being the policy of British Petroleum, who were already prospecting in the area, to use local contractors when possible. Although there were some local Arabs engaged in retail trade in other parts of the Trucial Coast, and even a few Arab shops in Abu Dhabi’s inland villages, shopkeepers who came originally from Persia, India and Pakistan far outnumbered them. In traditional life, retail trade had not been thought to be a ‘noble’
120
Shaikhdoms of Eastern Arabia
occupation for an Arab. Wholesale trade was considered more respectable, but it was difficult to become a wholesaler without starting from small beginnings in retail business. And the only really noble trade to be engaged in had been the sale of pearls and the management of pearling boats. Thus, when economic decline had struck the Trucial Coast it had struck the Arabs of high status hardest of all. At that time, Abu Dhabi was dominated by the fortified palace of the Ruler. This was the centre of such little administration as there was. No administrative building of any sort existed in the town, excepting one room for a customs house. Nor, apart from a couple of cafés and the mosque, were there any places for meeting or entertainment excepting in private houses or out in the open. There was no school except for learning the Koran, and there was no court-house. Nor were there any public services, apart from the shaikhs’ guards acting as a rudimentary police force. Families fetched their own water or had it fetched for them and stored it mostly in earthenware jars, in which mosquitoes bred. There were no medical services or facilities nearer than Dubai, which had a small hospital with an English doctor and his assistant, and an American medical mission at Sharjah. As a latrine, people used the seashore, the women trying to go there at night so as to avoid embarrasment, though some people did have fenced off pits in their yards which they used as little as possible because they merely had to be covered over when full. As in the whole of the Trucial Coast at that time, there was no refuse collection whatsoever. Domestic animals that died of sickness were left on the shore and eventually removed by the tide. Some people tried to keep clean the narrow passageways around their houses. Others were more negligent. In those days, none of the roads in Abu Dhabi or any other town of the coast was metalled or surfaced in any way. Camels that people had been riding or that had been carrying firewood or charcoal squatted at the roadside with a rope tied round one of their front knee joints to stop them wandering away. There were a few motor cars, jeeps and lorries, mostly belonging to the shaikhs or the oil companies. At that time, the only remnants of seafaring activity were fishing and a very little trade by launch. A large number of small fishing boats were drawn up on the beach and one or two launches lay at anchor off the shore. One might see a group of carpenters with their adzes making a new launch. In the mornings and late afternoons, fishermen sat out of doors mending their nets or making new ones. This was the full extent of manufactures in Abu Dhabi, apart from Persians making rounds of unleavened bread in bakery ovens and a few tailors using sewing
Towns and Maritime Activities 121
machines in their shops. It seemed ironical that the word karkhanah, which in its original Persian meaning is a workshop or a factory, had been adopted into Trucial Coast Arabic to mean no more than a sewing machine. The village of Rams, at the opposite end of the Trucial Coast from Abu Dhabi, presented a similar picture of decay. Rams had always been a small place, but its population by that time was down to about four hundred people. More of the houses in Rams were of stone than in Abu Dhabi, because the village lay very near to the foot of the mountains, where loose stone was readily available. But many of the stone houses of Rams were falling down. At the foot of the mountains were some date gardens, this area, which includes Ras Al-Khaimah, near the mountains being one of the more fertile of the coastal areas. Rams had no market in the strict sense, but four shops which were not all together. One saw the same fishing boats and fishermen, and here there was one medium-sized dhow drawn up on the beach. The brother of one of the shopkeepers made a single journey in it each year carrying salt fish to the north of the Gulf. Goats and a few cows wandered about the streets. Being so small, Rams had only three quarters. The eastern one was called the Persian quarter (again, farij), the central one was called by two alternative names, the Tanaig quarter and the central quarter, and similarly the western quarter was called either the Bani Hamud quarter or simply the western quarter. As in Abu Dhabi, the quarters of Rams were not exclusive. For example, some members of the [Awamir tribe and one family from the Sudan tribe lived in the Persian quarter. The central quarter had Sawalim and Manani as well as Tanaig, who in fact were very few, the former shaikhs of the village having belonged to them and having recently been driven out by the Ruler of Ras Al-Khaimah. And there were Marazig living in the Bani Hamud quarter. The bedouin connections of Rams were very much slighter than those of Abu Dhabi, and this being so it is worthy of note that some of the people settled there were members of a tribe that of all the Trucial Coast bedouin goes furthest out into the Empty Quarter in its transhumant life, namely the [Awamir. Such is the continuity between desert and coastal village or town. Of all the towns and villages in the Trucial Coast, Dubai alone was in any way flourishing at that time. Dubai had extensive quarters of palm houses, but also had wide areas of more solid buildings, many of them two and a few three storeys high and most of them in good repair, surmounted by handsome wind towers in the Persian style. Heaps of coral for further building stood along the shore drying out. A group of
122
Shaikhdoms of Eastern Arabia
men were constantly at work cleaning a sandbank in the creek and carrying the sand in baskets on their heads to build up one of the banks. This was a minor part of dredging operations intended to safeguard the trade of Dubai, but the new area of building land which was thus incidentally produced in the very centre of the town eventually made a profit for the present Ruler, 4 in his private capacity, of several hundred thousand pounds. Men in little rowing boats, many of which had been owned by the Ruler of Dubai’s late wife, ferried people from side to side of the creek, which runs right through the middle of Dubai and then several miles inland. Many launches – dhows, that is to say, with engines as well as sails – were tied up on one side of the creek, and there was usually at least one steamship unloading off the coast. On the sea beaches, one saw again the fishing boats and great piles of tiny fish drying in the sun. A few people fished in the creek itself, but most people regarded such fish with distaste because the creek was the town’s general sewer, being used like the seashore at Abu Dhabi. Dubai had become the main and almost the only centre for import and export for the whole of the coast, taking the place of its near neighbour Sharjah. But by this time Dubai was probably more successful in trade than Sharjah had ever been. Steamship traffic had helped in concentrating trade in one place because of the technical arrangements and organization required for dealing with the ships and their bulk cargoes. Dubai was an entrepôt, re-exporting goods to Persia as well as to the other Trucial States and the interior. From a Persian point of view, a good part of Dubai’s trade with Persia was smuggling, since the goods exported from Dubai in local or Persian launches could often be landed in Persia secretly without payment of customs duties. This was even more the case in India, a considerable part of Dubai’s trade with which lay in traffic in gold. Arab seafarers carried the gold from Dubai to India hidden in the cargoes or the hollowed-out timbers of launches and sold it there privately at high rates. Dubai being technically an independent country in the sterling area, British-protected but not a protectorate, merchants there were able to buy as much gold as they needed. They then smuggled it into India at enormous profit, with occasional heavy losses when it was found by Indian customs officials. At present, something in the region of forty million pounds’ worth of gold is sent to Dubai annually from the London bullion market, which is the main exporter, and no doubt smaller quantities of gold are imported from other sources. The main destination of the gold is still India. This enterprise – which can be represented as parasitism on controlled economies or as free trade, depending on one’s point of view – was one
Towns and Maritime Activities 123
in which Kuwait too was highly successful at that time. Secret hiding places were part of the original design of some new launches at the time when they were built. And there were other complicated ways in Dubai of making profit out of the Indian trade, in that at the time when bad political relations stood in the way of direct trade between India and Pakistan, goods from each were imported into Dubai and then re-exported to the other. Apart from oil prospecting, what little there was of European commercial activity, which was to say British commercial activity because of the Treaty, was mainly in Dubai at that time, though a little remained at Sharjah. Dubai had one good-sized modern bank, and there would have been others, as there are now, had not the British Bank of the Middle East then held a monopoly. Also in Dubai was the main branch of the British shipping agents. On the other hand, the airstrip serving Dubai was still at Sharjah, which had an Royal Air Force base. The British Political Agency for the Trucial States had recently moved to Dubai from Sharjah. There, every so often, one saw slaves taking off their sandals, stepping up on to the slab of concrete which supported the Agency flag pole with its Union Jack, and grasping the pole with both hands in a ceremony by which they made it known that they wanted certificates of emancipation. I do not know whether the ceremony was spontaneous or was one thought up by some earlier British Political Agent. In both main parts of the town of Dubai, on the two sides of the creek, were long and intricate markets roofed over with palm matting for shade. Only two shops in the whole town had glass windows. All the rest were open-fronted in the traditional way with the shopkeeper sitting with or behind his goods. At night, the shops were boarded up and padlocked, and the markets were dimly lit with oil lamps and guarded by night watchmen with daggers at their waists and rifles in their hands. Dubai had numerous simple cafés in and around the market, but the more respectable people preferred to sit talking and drinking coffee or tea outside shops where not everyone had the right to sit down and make himself at ease. There was also an open air cinema owned by a merchant from Kuwait who had been living in Dubai for most of his life. This was before Kuwait itself had any cinemas at all, the shaikhs there having refused permission to build them on the grounds that they would corrupt morals. The Dubai cinema showed as many Indian films as Arab ones. Apart from Indian clerks working for the merchants, a very large number of the shopkeepers in Dubai were Indians, particularly in the cloth market. There were also many Persian
124
Shaikhdoms of Eastern Arabia
shopkeepers, as well as Arabs whose original homes were in other parts of the Trucial Coast. Indeed, the number of people living in Dubai who came originally from elsewhere probably outnumbered the indigenous inhabitants. In the administrative circumstances of Dubai at that time, there was no way I could conceive of whereby any accurate estimate of the population could be produced apart from a full census. Recently, when a full census of the Trucial Coast was carried out, the rulers only permitted it on condition that the figures should not be made public. Their reasons for making this condition may have been that there were far fewer people in most states than their rulers had claimed. For example, in 1961 the then Ruler of Abu Dhabi 5 thought his country had ninety thousand people when my own most generous guess could not put it at more than twenty thousand. In 1967, with development under way, the population seemed to be in the order of twenty-five thousand. For all its size, with a population then supposed to amount to about forty thousand, and its relative prosperity, Dubai suffered the same lack of public services as the other Trucial Coast towns. A few people had private electric generators, but most used pressure lamps or ordinary oil lamps, the latter being both cheaper and cooler. Some poor people had lamps made of old polish tins with wicks stuck into them. There was no cleaning of the streets or public refuse collection. Dirt and rubbish were thrown into the street unless people took the trouble to take their refuse down to the creek. Baluchi water carriers, poor immigrants who could not understand Arabic and so were treated more or less like imbeciles by their customers, sold water from door to door, carrying it in paraffin tins loaded in panniers on the backs of donkeys. Dubai did have, however, a post office and a small hospital run mainly on British funds. Education was as backward in Dubai as in Abu Dhabi. When the Kuwait Government had offered schools to the Trucial States, some of them had accepted.6 But the Rulers of Dubai and Abu Dhabi having refused on the grounds that they could afford to provide schools and teachers themselves, had taken no steps towards doing so. And, like the rest of the Trucial States, Dubai had no official court-house open to the general public. The question of a court-house had been a subject of contention in the Dubai civil war of 1939–40, and it was one not very far removed from education. In that war, the few educated people in Dubai had been opposed to the Ruler and they, for the time, had lost. On the other hand the bedouin, to provide whose pay during that war the Ruler of Dubai was said to have sold even his coffee pots, were still generously treated, as was any poor petitioner. Whilst the people of Dubai lacked public services, they also paid no taxes apart from customs duties of
Towns and Maritime Activities 125
four per cent. The customs house was the only administrative building. Like their subjects, the shaikhs of Dubai lived off trade and the sea though their political concerns extended far into the desert.
Fishing Whatever their political orientation, the towns of the Gulf as a whole always owed their economic existence to the sea. In considering the three types of maritime activity, fishing, deep sea trading and pearl fishing, in which the Arabs of the Gulf engaged, I will begin with the occupation which was economically the simplest in the past and is the only one to maintain its importance at the present, namely fishing. Even here, the economic circumstances were by no means primitive. Not merely was fishing part of a money economy but, more complicatedly, fishing depended on long-distance imports even for making the fishing boats, and it produced long-distance exports. There used to be one very primitive and inefficient form of boat made of local materials in the Gulf. It was in the form of a deep raft made of date sticks tied together with cord, with a hollow in the middle in which one man could sit and float along precariously and half submerged. But for more efficient boats, timber had to be imported from Malabar, for there was no local material from which to construct them. And, correspondingly, salt fish from the Trucial Coast was carried by dhow and sold in Basrah or even in East Africa. The long-distance trade in salt fish has been going on for a long time. In 1828, the Ruler of Ras Al-Khaimah, the famous and in some ways notorious Shaikh Sultan bin Saqr, sent a letter to the British Resident in the Gulf complaining that for a considerable time past all the boats belonging to his subjects arriving from Basrah had informed him that the Kuwaitis had taken fish from each of them. (He accused the Kuwaitis of trying to destroy his credit, without mentioning that they were retaliating for attacks on their shipping.) But though fishing has more than local implications in terms of finance or of organized co-operation, it is relatively simple as compared with the trading voyage or the pearl industry. Fish is the one food that is plentiful in the Gulf and it forms the main protein of the coastal diet. Most people eat fish every day. Usually they eat it fresh, but when the sea is rough or at times in the summer when fish is less plentiful in the south of the Gulf, they eat salt fish, and salt fish is taken inland and eaten as something of a delicacy, or at least a change, at the oases of the interior and even in the desert. In addition to fish for human consumption, the Gulf also produces huge quantities
126
Shaikhdoms of Eastern Arabia
Fig. 4.1
A dhow. From the author’s photographic collection.
of little fish like sardines, which the fishermen catch and dry in the sun. People buy this dried fish to feed to their goats and the few cows they have, since there is usually so little else for them to eat. (One can also buy date stones for feeding to goats.) In recent years, a great deal of dried fish has been sold to Europe as fish manure. For the coastal towns and villages, fish also provides one further product, in that people make oil from the livers of some of the bigger fish, particularly sharks, and use it for painting on boats to preserve the wood. The oil has a particularly unpleasant smell, which pervades some parts of the Trucial Coast towns. Fish is the only food eaten in the Trucial Coast none of which needs to be imported. By comparison, meat forms only a minor part of most people’s diet. Only during Ramadan or when entertaining guests do people tend to have meat when they can afford it. Until recently, the
Towns and Maritime Activities 127
only town in the whole area where there was a meat market and where it was possible regularly to buy meat ready butchered was Dubai. Elsewhere, the danger to the butcher of losing money from lack of customers was so great that no one would regularly undertake the enterprise: if one wanted meat one had to buy a whole live animal. The exception was Ramadan, when goats and sheep were imported from Persia and the demand was great enough to make the provision of readybutchered meat practicable in all the towns and larger villages. Fishing, like other maritime occupations, is entirely men’s work even to the making and repairing of the nets, excepting that women sometimes help the men of their families by sewing ropes onto the nets, and that in the smaller places some of them sell the catch, not, of course, taking shops in the market but squatting outside with the fish on the ground before them. Fishing is a poor and, on the whole, humble occupation, and prouder townsmen and villagers, including some fishermen, would not have their womenfolk selling goods in public in this way. The only explanation men give as to why women never make or mend fishing nets is that they do not understand the work. This explanation not being plausible, one must assume that the real reason is one of the imaginative association between fishing nets and the work of fishing. In a similar way it is thought shameful for men to fetch water, except on expeditions in the desert, or to sell eggs, both of these being connected with the house. The same applies to washing clothes for other people, and laundry work is done by Adenese or Somalis. Men usually make their own fishing nets, though the very big nets are often made to order at piecework rates or bought ready-made. The pay for net making is poor, though the work is tedious rather than arduous. In the early 1950s, a man could earn one or one-and-a-half rupees a day making nets, which was about a third of the daily wage of a building labourer. A new sixty-yard-long net of the small-meshed kind used for catching fish to dry cost about thirty rupees, of which the labour cost was between five and eight rupees. The only local materials used in making fishing equipment are from date palm, in that the simplest form of float is made from the stump of a date frond. But good wood is more lasting and effective and the type used comes from East Africa. The twine for making nets is imported, as is sail canvas. Sometimes the twine is twisted into a suitable number of strands by the net maker and sometimes it is bought ready twisted. All fisherman know how to make nets but the boats themselves have, of course, to be made by specialist carpenters. Some small boats were made by the carpenters working on ocean-going dhows during the long voyages. In addition to
128
Shaikhdoms of Eastern Arabia
nets, and sometimes hook and line, for fishing, fishermen in the Gulf use two kinds of fish trap. One kind, hazra, which is twenty or more yards across, is made of date sticks or other stakes planted and fastened vertically in a curve in that part of the seashore where the water is shallow at low tide. The trap is so shaped that the fish can enter at high tide but are held in the shallow water when the tide recedes. Another version of this kind of trap is made of net supported on stakes and has a door to shut when the tide starts to go out. For allowing fishermen to construct such traps on suitable parts of the seashore, the shaikhs had a traditional right to charge a rent. The other kind of trap, gargur, is made of wire and is in the shape of a hemisphere. These traps vary in size from about two to about eight feet high. The local way of measuring them is, however, not by height but in handspans going from the base over the top and down to the base on the opposite side, a way of measuring apparently derived from the length of the wires used in constructing the trap. The entry to this kind of trap is through a sort of wire funnel which can be removed to allow the fish to be collected. These traps are taken out to sea by boat, baited with seaweed, and let down in fairly deep water. Many fishermen own their own boat and nets and go out fishing with a son who is living in the same house with them. In this simple situation, no division of the proceeds of the work is necessary. But as soon as there arises the question of different household expenses, or of different ownership of boats, nets or other fishing equipment, we come to a method of division of profits characteristic of all the maritime occupations of the Gulf. The division of profits gets more and more complicated as the enterprises grow bigger and bigger. In all these maritime affairs, there are very few wages or set payments. Almost everything is profit sharing. The sharing of profits, taking into account capital and labour, and often also expertize and responsibility, has always been the method of paying for work at sea, though wages and fees are common enough on the land. One reason for its prevalence, not only in the Trucial Coast but throughout all the shaikhdoms of the Gulf, may lie in the extreme variability of the profits themselves in relation to an equal amount of effort expended in obtaining them. The simplest form of profit sharing is that practised when men are fishing from small boats for fish to be sold fresh. This is a day-to-day occupation requiring little financing and little equipment, apart from the boat itself, and concerning very few men in each individual trip. When two men fish together in a boat belonging to one of them, using the boat owner’s nets, which is the arrangement that usually applies in
Towns and Maritime Activities 129
the Trucial Coast when fishing for fish to sell fresh, the two fishermen divide their takings into three parts: one part is for each of the men and one for the boat and nets. If the owner is not working in the boat, he is sometimes able to take a half of the fish by persuading some other man to go out fishing in the boat with only a boy to assist him. The man takes a half of the proceeds and then gives the boy a quarter of his half (i.e. one eighth of the total takings). The other half of the proceeds is for the boat and nets. Arrangements for the division of profits from fishing in small boats are capable of variation. For example, when fishing is being done with wire fish traps from a boat, the owner may take other men with him and then simply arrange with them that he should take more of his own traps with him than they do of theirs, say in the ratio of five of his and two for each of them. The various fishermen do not pool their catches but keep them separate. It can also be agreed beforehand that when men are fishing with nets or with lines from a small boat they should keep their fish in separate baskets, and in this case the owner receives from each man a set quantity of fish, irrespective of the size of the catch. This is the only example I know of in the fishing industry or in the other seafaring industries of the Gulf where there is a set return for capital equipment and no profit sharing. All these sharing arrangements are contractual. In small enterprises of the sort just mentioned, kinsmen, such as maternal or paternal cousins, will usually see no need for anything more than a verbal agreement. But for people outside the family, a written agreement is preferred. The manner of dividing profits between capital equipment and labour, which is practised between even a couple of fishermen, the nets and the boat, becomes more elaborate as the fishing enterprises become more complicated. When the catch is to be dried or salted, more men work together in a bigger boat using bigger nets, and they no longer sell their catch from day to day. From one part of the Trucial Coast to another, local variations in the division of the proceeds correspond to special local circumstances of fishing. In the Ras Al-Khaimah area, when a group of men has been out catching fish for salting, the division of the proceeds from the sale of their catch is as follows. The leader of the enterprise, who usually owns the boat and all or most of the nets, first deducts the cost of food and tea consumed on the trip. He then subtracts one third of the remainder as the share for the nets and ropes. Lastly, he divides the remaining two thirds into as many shares as there are men in the boat, including himself, plus one extra share, the extra share being for the boat itself. Thus, in his capacity of the capitalist, he takes one third of the profit plus one sailor’s share, and in his capacity
130
Shaikhdoms of Eastern Arabia
of labourer he takes a further man’s share. In his capacity of captain, he takes nothing, since he is running the enterprise himself. If he cannot join the expedition and another man has to take command, then the owner must pay him, according to prior agreement, an extra sum calculated in terms of an individual crew member’s share and fractions of it. This he pays out of the sum he receives from the capital equipment and it is no charge on the earnings of the crew. Thus leadership or control is a charge on capital and not on labour. The nets required for catching little fish to dry are considerably more expensive than those used for catching larger fish to salt for human consumption. They have to be of a very narrow mesh and between sixty and a hundred feet long. The nets are supported at one side with floats and weighted at the other side with small stones, and the method of using them is to net round almost the whole creek. When this type of fishing is done at Ras Al-Khaimah, the division of the proceeds is different from the division just mentioned, in that the nets and ropes take not a third but a half of the remainder after the cost of food has been deducted. The other half is divided between the fishermen with a fisherman’s share for the boat as before. But in Abu Dhabi, the circumstances of catching fish to dry are rather different from those at Ras Al-Khaimah. The sharing procedure varies accordingly. At Abu Dhabi, a longer trip is required because the fishing is done in creeks a good distance down the coast. Hence a bigger boat is necessary. The fishing party consists of ten or a dozen men who stay away from home for about a fortnight. Here, as at Ras Al-Khaimah, the division of proceeds begins with the subtraction of the cost of food. (On such a trip, the food represents quite a considerable expense. It will be noticed that the form of the division of proceeds is such that the cost of the food is in effect divided between capital and labour.) After subtracting the cost of food, the leader of the enterprise in Abu Dhabi proceeds in a different way from his counterpart in Ras Al-Khaimah. He first subtracts one fifth of the remainder for the boat. Of the sum then left, he subtracts a half for the shares of the nets and ropes, then he divides the other half into equal shares for everyone in the boat including himself, but with no extra share for the boat since that has already been paid for. Throughout the Trucial Coast, the leader of the enterprise has one extra cost, which has not yet been mentioned, to pay out of his own money. This is the rent for the particular creek in which he has a right to fish, and it is paid annually to the ruler of the state. In the 1950s, creeks were rented out for fishing at between one and two hundred rupees a year (a hundred rupees being about seven pounds ten shillings at that time)
Towns and Maritime Activities 131
according to their profitability for fishermen. Access to particular creeks, therefore, is not a matter of traditional right for particular families of fishermen, nor is it a matter of favour or a claim in use. It is a right obtained by payment, and the payments were competitive. If the nets used on any particular expedition have various owners, the part of the proceeds due to the nets as a whole is divided up between the various nets, particularly big nets taking a double share. In considering the large share of the profits paid to the owner of the nets on such expeditions, it should be remembered that the nets are not only big but also highly perishable as compared with a boat. Even with repairs, a net is not expected to last more than two years. It is a duty of the crew to repair the nets without extra payment. This is a service which labour gives to capital. The same system of profit sharing as is followed in fishing operations in the Trucial States is applied also to the division of profits between capital and labour in other maritime activities there. In the case of fishing, as we have seen, one finds within the Trucial Coast itself minor variations in the actual shares which accrue to different parties in an operation. With a further range of local variations, this same practice of profit sharing applies with a great deal of basic consistency to all maritime occupations in the shaikhdoms of all the Gulf, from the Trucial States to Kuwait. There are no wages or set fees except in certain particular cases which I shall discuss later. Profits are always shared between the owners of capital equipment and those who do the work of the operation. The actual form of sharing is adapted to the circumstances of particular operations so that, for example, there are differences between the sharing of the profits which accrue from fishing and the sharing of profits that accrue from trading voyages, and there are further differences when we come to pearl fishing. And just as the way in which the work is done can result in a particular way of sharing profits in the fishing industry, so too it can in pearl fishing. The broad system is adapted to the relatives values of the equipment and the labour done with it. As I shall describe when giving an account of trading and pearl fishing, there can also be distinctions drawn between different sorts of labour such as do not apply to fishing operations as I have so far described them. They do, however, apply to fishing from launches in the places where this is practised, but since the method of dividing the profits made from fishing by launch is more similar to that practised in dividing the profits of a trading voyage, it will be more convenient to mention the exception when the trading voyage comes to be described. One general principle holds, namely that the more complicated an
132
Shaikhdoms of Eastern Arabia
operation is in types of work and of equipment used, the more complicated are the sharing arrangements for the profit accruing from it. I have begun this account of profit-sharing with a description of the system as it is applied in the fishing industry for two reasons. One is that, of the three types of maritime occupation which have been important in the traditional life of the Trucial Coast, fishing is the one that vigorously survives. And the second reason is that fishing shows us profit-sharing at its simplest. But I think it would be wrong to assume that, just because it is the one that produces the simplest form of profitsharing, fishing therefore gives us the key to the whole system. We have no way of telling how the custom of sharing profits originated or whether it began in one sector of maritime affairs before the others. Nor can we build any suppositions on the assumption that fishing, being technically a simpler operation than diving for pearls and marketing them or making a long-distance sea voyage, must have been practised first. As we have seen, fishing from boats as it is now practised could not exist without the trading voyage, because without the latter there would be no wood available to build the boats. There is a further reason why I think it would be wrong to assume that the sharing system practised in relation to fishing is the basis of the whole system, and this is that the very simplest type of fishing operation is in two ways rather different from other maritime occupations. Firstly, in this type of fishing, no arrangement needs to be made for food or other supplies; the significance of this will appear more clearly later. Secondly, and here perhaps is the more telling difference, each individual operation is so small that it usually involves co-operation only between two members of a family, probably in practice two of the very closest kin such as a father and his son. If this son is living in his father’s household, as many sons still do now and even more did in the past, then there is no need for sharing arrangements. The father is in charge of household expenditure, even to the point of buying clothes for his son’s wife. He keeps, or else entrusts to his wife, the little money that he and his son earn between them, and the son asks him for spending money when he needs it. And even when a fisherman and his assistant are not kin, relations between the two are extremely personal. When boats that hold only two men are in question, there is no class of boat owners separate from another class of fishermen who do not own boats. But once larger scale operations are embarked on, a distinction begins to arise. Thus, so long as neither of them sees cause for complaint, what happens between individual pairs of fishermen is more of their own business and less a matter of public concern or public policy than what happens between
Towns and Maritime Activities 133
the owner of a pearl-fishing boat and his crew. The pearling crew are on the whole neither relations nor friends of the boat owner. The arrangements made between capital and labour in respect of a pearling boat have a bearing on relations between all capital and all labour in the pearl fishing industry. This being so, custom is stricter and is more strictly enforced by those in authority, since uniformity is a matter of public importance. So far, I have discussed only formal financial arrangements, but one further financial matter needs to be mentioned before we go on to discuss the trading voyage: the question of debt. In the fishing industry, debt is informal in the sense that it is not a regular part of all or most transactions. In sea trading and pearl fishing, the opposite applies; here debt is a regular part of the relations between capital and labour. In general, any man may, of course, borrow money from such people as are prepared to lend it to him and according to such terms as he may make with them. Though it is forbidden by Islam and most people in the Trucial States express disapproval of it, moneylending does exist there, and there are also perfectly legal ways of making a profit out of loans. But it is not with such matters that I am concerned here. I speak of debt, rather, when it occurs as part of a business relationship, as, for example, between a boat owner and a member of his crew, or between a boat owner and a merchant to whom he sells his goods. Debts of this kind have less chance of arising in the fishing industry than in other maritime occupations, but they can exist even in fishing, and when they do so they bind particular seamen to particular boat owners and similarly bind particular boat owners to particular merchants, just as they do in other maritime industries. Whether he is a fisherman who owes money to a boat owner, or a boat owner who owes money to a merchant, a man who wishes to break the commercial relationship must, in either case, first free himself from the relationship of debt by paying what he owes. In the fishing industry, the relationship of debt only arises significantly, as a bond in economic organization, when people are engaged in dealing with fish in bulk. Although he takes a part of the nightly catch for the use of his boat and nets, in practice the owner of a little fishing boat that needs only two men to man it is no capitalist. Logically he may seem to be so, inasmuch as he receives a return for his capital equipment. But to regard him as a minor capitalist in theory would tend to conceal more significant distinctions. People with money to invest do not find it profitable to invest it in buying a number of little fishing boats and getting other people to work them: the arrangement
134
Shaikhdoms of Eastern Arabia
would be too complicated and the margin of profit too small. Ordinarily, the fisherman who owns a little boat lives near the margin of subsistence. It is perfectly possible for an assistant who works with him to have more money to spare than he has himself, since he may have elderly parents and a number of small children to look after, whereas his assistant may have no special calls on his earnings. Moreover, the assistant may well have a father and brothers who own boats and may be saving up, with their help, to buy a boat of his own, whereas the sons of the boat owner he works with may, when they grow up, have to go and work in other people’s boats, excepting for the one who can help his father in the assistant’s place. Because of the very small amount of capital represented by a little fishing boat and its nets, there is no social inequality between boat owner and assistant. And, as we have seen, economic organization in this sort of fishing is mainly co-operation within the family. For producing fish to sell fresh, both the working units and the amount of capital required for each unit are small, and, in addition, the time-scale of work in relation to receipt of payment is very short-term. Fishermen who sell their fish fresh receive the income from their work daily – or almost daily, for there are some days when they do not put to sea. Unlike the owners, or the crews, of trading dhows, they do not have to spend several months without payment between the start of the enterprise and the time when its profit can be shared out. And unlike those who sell goods in bulk, whether they be dried fish or pearls, they sell fresh fish direct to the general public, acting as their own retailers. Their customers buy as their immediate household needs warrant and need not make long-term arrangements to ensure regular supplies. Hence, the context of such a fisherman’s work does not oblige him to seek loans until he can receive the payment for his work, nor does it provide him with anyone to whom he can automatically turn for a loan. If he is in debt, as he may well be, the reason is that he has overspent his income. The creditor to whom he is most likely to owe money is a shopkeeper whose bill he has not paid. Here indeed is a relationship made binding by debt, for if he cannot pay the bill he will have to go on patronizing the same shop. But his relationship with the shopkeeper has no direct connection with his work. People who own bigger boats and catch fish to dry or to salt are in a different position. Those who catch fish to dry, more particularly, are selling to merchants rather than to the general public. Often these merchants deal in great quantities of the product, accumulating it and then exporting it in bulk. Unlike daily shoppers, the merchants need to
Towns and Maritime Activities 135
ensure their supplies and can take steps to do so. Also, the boat owners, in their turn, are dealing with capital and labour on a rather larger scale than the small fishermen I have been speaking of. They need bigger boats and more expensive nets; and their expeditions last up to a fortnight so that neither they nor their crews are earning their livelihood from day to day. In this part of the fishing industry, circumstances begin to arise where debt becomes part of economic organization. The dried-fish merchant needs to ensure that he receives supplies from a number of boat owners. Each of the boat owners, in his turn, needs to ensure his labour supply and keep as good a crew as he can together. The boat owner may need money for a new boat or to replace his nets, or even to buy supplies for the dozen men who form his crew during the fortnight they spend at sea, or like anyone else he may need to borrow money for personal reasons unconnected with his work. Since he sells his fish regularly to one merchant, he expects the merchant to lend him money when he needs it. The merchant, for his part, ensures, by providing a loan, that until the loan is repaid the boat owner will continue to give him the first refusal for his catch. Similarly, by lending money to a member of his crew, a boat owner can ensure that the man stays with him. On the crew member’s side of the matter, such a loan can only be one with no direct, positive connection with his work. The crew member has no equipment to replace, and since accounts at shops are normally paid only at the month end, and with some elasticity, the crew member has no need of borrowed money to buy food that will tide his family over until he returns from a fishing trip. But from the boat owner’s point of view, the loan may have another significance. If the crew member is a good worker who might easily go to work for another boat owner, then a loan ensures that he stays in the same crew until he repays what he owes. Thus loans can bind labour to capital and can bind smaller amounts of capital to greater ones. They reduce freedom of choice and freedom of action. For this, they themselves are the return. Loans such as these that merchants give to boat owners and boat owners give to crew members in the Gulf have a moral as well as an economic aspect. The merchant or the boat owner who refuses a loan – and sometimes a gift – without adequate reason, is making a personal moral decision which lays him open to disapproval on the part of the unsuccessful applicant and of other people. He is refusing ‘help’, musa[ada, and the refusal is likely to provoke resentment. He is thought ‘mean’. Conversely, if he complies with the request he is thought ‘generous’. To some extent the recipient is expected to be grateful, and often is so.
136
Shaikhdoms of Eastern Arabia
A person who asks for a loan and is refused may take his work or his supplies elsewhere, not so much out of calculation of his own advantage as out of resentment. Such loans as these are free of open interest, or even of the ‘concealed interest’ that Islamic law allows. Suppose that a man in need of cash buys goods at more than the current market price to be paid for in arrears. To realize cash, the purchaser then has to sell the goods, probably at less than the market price, since he is not a regular trader in them. On the other hand, they establish a relationship in which the debtor can no longer take the fullest advantage of the market whether in selling his goods or his labour. This does not mean that he has to accept less than most other people are receiving, but until he repays the loan he is tied to the general market price of goods or labour and cannot take advantage of special offers. At the time when I was engaged in research in the Gulf, the implications of the situation were clearer in the already prosperous circumstances of Kuwait than they were in the Trucial States. In Kuwait, the owner of a fishing launch preferred to have some money owing to him from the better members of his crew. So long as he owed his money, a good fishermen would work for the launch owner for an ordinary crew member’s share of the profits. But if the man owed him no money, the launch owner might have to offer him an extra quarter of a crew share to persuade him to stay with the boat. This extra money was not included in the share of the profits divided among the crew as a whole but was paid from the share due to the launch owner. Similarly, a merchant could buy fish from a boat owner who owed him money at the price for which most merchants were buying fish. He could not be expected to make a specially high offer, nor could the boat owner take advantage of a high offer from some other merchant without being pressed for repayment of the debt. In this sense, debt is an advantage to creditors, and the more general is the situation of debt, the less free is the market. On the other hand, the advantage of interest-free loans to the debtors is obvious, and the restriction of the debtors’ freedom of action can, in some sense, be regarded as a return for the credit he receives.
Merchant seafaring Of the three interdependent maritime activities of the Trucial States, fishing is the simplest and smallest in terms of the organization of any individual operation, and pearl fishing is the largest and most complicated. In these terms, merchant seafaring comes between the two, and
Towns and Maritime Activities 137
that is why I consider it at this point. But the placing of it second in the terms of the present discussion does not imply a similar position in any scale of importance based upon other criteria. That supplies should be delivered and means provided for the export of pearls was essential to the Trucial States. Whether or not home-based shipping provided the means was far less important. Fish and pearls, the former mainly for home consumption and the latter for export, were the major traditional assets of the coast. So long as there were ships from anywhere calling in at the ports to bring food and raw materials from overseas and carry away pearls to foreign markets, it was not of overwhelming significance whether the ships were home-based or foreign. Home-based shipping was more important to the economy of some states than of others, but there were always other towns in and near the Gulf such as Muscat, Sur and Kuwait, whose shipping had greater share in the business of the trade route extending from Basrah along the Gulf and out into the countries of the Indian Ocean. I discuss, therefore, fishing, merchant seafaring and pearl fishing in the present order in relation to a scale of complexity in social and economic organization and not in relation to a scale of importance in traditional life or to any hypothetical sequence in the evolution of the coastal towns. The most conspicuous thing stimulating people of the Trucial Coast to undertake long and arduous trading voyages was not the goods they had to export but their need for imported goods. The barrenness of the countries along the coast was such that not only did they produce little that they could export in bulk, but they also lacked a substantial part of the necessities of settled life both in the way of raw materials and of food. I have mentioned already that the timber people needed for building boats had to be brought from India: naturally enough, so too did timber for the doors and window frames of houses. And, though date palm trunks provided roof beams for houses at the oases of the interior, on the coast mangrove poles from Kenya and Tanzania were always used. Most of the cotton stuff for clothing, even the traditional veils of the women, comes from India. And from India and Persia come rice, which in the Arabic of the Gulf is called [aish, or ‘life’, a term that the corn-growing culture of Egypt applies to bread. As well as rice, India and Persia were the traditional suppliers of flour for the coast, which also imports coffee, tea and sugar and numerous other foodstuffs. The trading voyage usually began with one foodstuff the coast could export: salt fish. A market for salt fish existed in Iraq. And at the port of Basrah in southern Iraq the dhows could take on a major cargo supplied by the enormous area of date gardens planted along the Shatt Al-Arab,
138
Shaikhdoms of Eastern Arabia
where the Tigris and Euphrates, the former already joined by the Karun river of Persia, meet to flow into the sea. This is the only area in the whole of the Gulf where there is an abundance of fresh water. From Basrah the dhows returned down the Gulf and, if they were making the full trading voyage, they sailed out to the ports of the Indian coast, loading and delivering cargoes and passengers on the way. From India they crossed to the coast of East Africa, and sailed back up this coast, along the southern coast of Arabia and back into the Gulf. The whole voyage could take a year or more. Some [dhows] only went to India. A further cargo that was for a long time taken on in East Africa was slaves, bought from slave traders or sometimes kidnapped. Within living memory, mothers in Zanzibar used to keep their children indoors when the dhows of the Gulf were in port. Boats are built by specialist carpenters working for a daily wage. There are no special ceremonies for launching or naming the boat, but when a boat is nearing completion and ready to be caulked with cotton waste and fish oil, all the local boat builders come and join in the work of caulking without payment. The head of the firm of shipwrights that is building the boat entertains them all to a meal on the deck when they have finished. The local carpenters thus combine in a gesture of making the boat seaworthy, for it has many perils to face. Even sailing within the Gulf itself can be dangerous because of sudden and violent storms, and the journey to India or East Africa, or both in succession, is one that some crew members embark on unwillingly, forced into the work by the need to earn a living or by debt from an earlier voyage. The idea exists in the Trucial Coast, as in Kuwait, 7 that if a barren woman steps over the keel of a boat whilst it is being built she will bear a child and subsequently one of the sailors manning the boat will lose his life as a substitute for the life she has been able to produce. More important, however, is the idea that no one should go up on to a boat wearing his sandals, even when it is in the course of being built. Sometimes a visitor breaks this rule, and then the carpenters say that one of his sandals is forfeit. They cut the sandal in two and nail one half to the boat. Whilst aboard a boat at sea, no one wears sandals, for it is said that people have to pray regularly on the deck, including the congregational Friday prayer, and hence the boat is like a mosque. It comes, therefore, into a high category of sanctity, though this is not overtly associated with the dangers those who sail in it have to face. Many people, in politeness, used to comply with the practice of removing their sandals even in the little rowing boats that ferried them across the creek at Dubai.
Towns and Maritime Activities 139
When talking to boat captains, one may at first suppose that they practice an extraordinary amount of astro-navigation, but one soon finds that the reason why the stars are mentioned so often in their conversation is that all the points of the compass have the names of the risings and settings of stars and constellations. Asked about navigation by the stars, one launch captain who was sailing his boat to India said, ‘How can one sail by the stars when they move in the sky?’ Captains do, however, make use of navigational instruments besides the compass, or else they have on board a mu[allim, literally ‘teacher’ or ‘person who conveys knowledge’, who does so. It seems that more navigation was done by the stars in the past, for there are books of astronomical tables still extant. But it is clear that a great deal of the expertize of navigation depends upon long experience, during which the captain comes to know the appearance and condition of the coast around which he sails, together with the reefs, sandbanks, currents and shallows. Much of the sailing is done close in to the coast, which can provide both directions and shelter from storms. An old manuscript sailing book from the Gulf, written in colloquial Arabic, gives advice about all the landmarks over a wide area. It provides little sketches of the shapes of mountains and hills seen from particular angles, mentioning any distinguishing patches of colour or groves of trees, and it gives the bearing according to which one must then proceed. It instructs the reader as to the bearings he can take on buildings and natural features for making his way safely into anchorages through the reefs, and tells him of the white water he will see if he has passed by the channel. The author has not, however, confined himself to giving purely navigational assistance. He tells, for example, of how there are plenty of fish to be caught in one anchorage, and of how in another place there is plenty of water and firewood, adding the warning that the sailors should fetch it for themselves because the people thereabouts exploit travellers and overcharge them.8 On his journey in a Kuwaiti boat from Bombay to Muscat in 1831, Stocqueler noted that the commander took no share in the navigation of the vessel while it was crossing to Muscat, this duty being entrusted to an old Arab who understood the use of the sextant and who was so correct in his observations that we made the Ras-el-Had within an hour of the time he had predicted we should.9 The navigational expert owns his own instrument and charts and is paid for the use of them. But the captain may take on this extra duty, in
140
Shaikhdoms of Eastern Arabia
which case, if he suddenly falls sick or dies at sea, it is quite possible that the boat and its crew be left in a desperate condition, no one else having an adequate idea of how to navigate. It seems that there has never been any custom of training in navigation at sea excepting for a member of the family or by special favour. What Stocqueler said of the observance of discipline on a trading boat in his day applies equally now. ‘The sailors’, he says, ‘acknowledged a kind of paternal authority on the part of the commander, and mixed with their ready obedience to his mandates a familiarity quite foreign to English notions of respect, and the due maintenance of authority’.10 And it was no small number that the commander of Stocqueler’s boat was dealing with, for the crew, he tells us, numbered forty or fifty men. The captain of a boat has no special disciplinary powers, but if a member of the crew behaves particularly badly he can complain of him to the shaikhs on his return home and expect him to be punished by them, probably severely. The organization of a trading voyage is more complex than that of a large-scale fishing expedition, both in terms of skill and specialization and in terms of finance. The boat needed for the voyage is bigger and, with the distances to be covered, more complicated to manage. Since most of its time is spent in actual sailing, a trading boat needs a mate and helmsmen as well as a captain. It also needs a cook and, if it is a launch, an engineer. The time spent at sea produces a need not only for the provisioning of the boat but also for the support of the sailors’ dependants who stay at home: the crew must be provided with advances of money to keep their families until they return. Trading boats carry passengers for fares, but in the case of goods two different terms of carriage may be adopted. As in European ships, goods may be carried in return for freight charges. But also a boat can be working in co-operation with people who provide capital with which to buy a cargo, which is then sold on the way and other goods bought with the proceeds, perhaps to be sold and replaced again in the course of the voyage home. Now, as in the past, trading boats are often owned not by their captains but by merchants, in which case the captains for them have to be provided by the owners. Nevertheless, captains and crew are not paid wages. Like fishing, the trading voyages is a co-operative enterprise between the owner of a boat and the men who sail it, but here a provider of capital may have to be taken into the enterprise in addition. The division of profits received from passengers’ fares and freight charges on a launch manned by a captain, mate, engineer, three helmsmen, a cook, ordinary crew members and a boy is as follows. The engineer and the boy do not receive any set share in the proceeds of the
Towns and Maritime Activities 141
voyage and are not included in the calculations. The engineer is paid a monthly wage by the owner of the boat. The boy is paid nothing: he is probably travelling with his father or brother to give him experience and accustom him to the sea. He receives his food, and the captain and crew give him presents of money. From the gross profits of the voyage, the cost of food and petrol are first deducted, giving the net profit, which is then divided into halves, one half for the sailors, including the captain, and the other half for the boat. The sailors’ half is divided equally between them, including the captain as one of them. This establishes a man’s share. The sailors with no special responsibilities receive this share and no more. Then, according to what has been agreed with the owner of the boat before the start of the voyage, those with special responsibilities receive extra sums from the boat’s half of the profit, calculated on a man’s share and fractions of it. The captain may then receive three extra shares, the mate half an extra share, and the three helmsmen and the cook may receive a quarter of an extra share each. All these extra shares and parts of shares are charges against the owner of the boat, and in addition the boat owner pays the captain a sum previously agreed on for his use of charts and navigational instruments. The captain and crew are allowed to make extra money by carrying some personal merchandise of their own in the boat free of charge. The merchandise they carry in this way is a private matter and each deals with his own and retains his own profit without any pooling of resources. A simple enterprise of exporting salt fish to Kuwait from the village of Rams near Ras Al-Khaimah provides an example of the share taken by capital in a trading voyage. In earlier days, Kuwait itself used to export salt fish, but when the town suddenly became wealthy and grew in size, all the fish caught locally could be sold fresh and salt fish became an import. In this case, the owner of the boat, who was himself the captain, had some capital of his own, but not enough to buy a full cargo. Hence various people produced shares of the capital required to buy the cargo, including the owner himself. The journey to Kuwait, with a following wind, took five days, sailing across to Qishm Island and up the Persian coast and crossing to Kuwait due west from Kharg Island. If the wind was against the boat, the journey took up to a fortnight. The boat would stay in Kuwait for several days to allow the captain to do the business he had to conduct, and so the round trip would usually take three or four weeks. Having sold his cargo in Kuwait, the captain would buy such goods as he could there that were cheaper than they were in Dubai and bring them back to sell in the Trucial Coast. After returning
142
Shaikhdoms of Eastern Arabia
home and selling the goods, he took the total sum of money realized by the voyage and deducted from it the cost of food for the trip, the sum of capital which had been spent on buying the salt fish, and the customs duties and other expenses of the voyage. The captain then divided the net profit into thirds, of which one was for the capital, one for the sailors working in the boat, including himself, and one for the boat. He divided the profit from the capital amongst those who had provided the initial capital in proportion to the sum they had supplied. He divided the third which was for the sailors into equal shares including one share for himself, and then, out of the boat’s share, he paid the extra fractions of shares due to the helmsmen and the cook. Such a voyage could make as much as 200 per cent gross, but the crew took a risk in that they had no assured return, even of a minimal variety, for their work. If, for example, the salt fish was condemned by the new health department in Kuwait as being unfit for human consumption, which sometimes happened, then the journey might make no profit at all, and would make a substantial loss for those who had invested capital in it. In the best conditions, crew, captain and investors of capital all had excellent returns. Those who provided the capital, including the captain, made a net profit of about 60 per cent on their investment. This was not, however, a return for just one month as it might appear, because the capital had to be available some months in advance in order to accumulate the cargo. The return, therefore, was not quite so high as might seem at first sight, and moreover such opportunities for investment were only occasional. Cash lay unused for long periods. The captain did not think it worth his while to collect the cargo, take his sizeable boat up the Gulf and do business in Kuwait, more than once a year. Unless the profits were spectacular, the journey was not worth taking. In the case of longer sea voyages, the profit might be very low, especially if the boat had been unsuccessful in smuggling and had had to drop gold in the sea, as happened more than once, to avoid arrest by the Indian customs authorities. The crew might then have too little profit out of the voyage even to repay the money they had borrowed from the boat owner to keep their families while they were away at sea. Until they could repay their debts, they were under an obligation to go on a subsequent voyage in the boat whether they did so willingly or not. When part of the profits of the voyage came from fares and freight charges which had to be halved between the crew and the boat, and another part came from this kind of merchant adventuring and had to be divided into three parts in order to include those who had provided the capital, then the cost of food consumed on the
Towns and Maritime Activities 143
voyage had to be divided proportionately between the two kinds of sum received. It was quite common, especially amongst Arabs whose families had been of some distinction in the past, to prefer trying to make large sums of money by occasional ventures rather than persevering in work that produced steadier, and cumulatively greater profits. But others were prepared to persevere with work that paid badly. Another man in Rams had a small boat which took goods to Dubai and usually found little to bring back on the return journey. The rates he could charge were low: one rupee for a sack of goods and ten rupees for a bull, bulls being used for drawing water for the date gardens and then later sold to Dubai for slaughter. Four men were needed to man the boat, which often took well over a week to get to Dubai and back and so could not usually make more than two trips a month. The total profit for a round trip was between a hundred and a hundred and fifty rupees. But the owner had other work and did not usually make the trip himself, whilst two of the men who manned the boat were members of the owner’s family and lived in the house with him. These two had little else they could do for a living unless they went to work away from home, which the family did not want them to do, and the boat and their work between them contributed an additional sum of a hundred and sixty pounds or more a year to the family income. In the village of Rams, as in most parts of the Trucial States during the 1950s, there was very little work indeed, and many people went to earn a living in the oil-rich states further up the Gulf, leaving wives, children and parents at home. Some neglected their families after they had been away for a time. Others continued to support them and authorized their families to run up bills in their names at particular local shops. One shopkeeper had partners who did not work in the shop itself but collected the debts that were owing and at the same time bought new stock, travelling up and down the Gulf as passengers on launches and carrying goods with them. When one of the partners left for Kuwait, he would take bills due to his own and other shops. I do not know what he charged his competitors for collecting their debts for them, but in other circumstances it is usual to charge ten per cent for collecting a debt on someone else’s behalf. He also delivered letters from families in Rams and Ras Al-Khaimah to their kinsmen who were working in Kuwait, charging one rupee per letter and bring back a reply free. One rupee was more than the postage for one letter each way, but the difference for each person was only a few pence, and the partner might have to write the letters down himself, most people being illiterate. The advantage of
144
Shaikhdoms of Eastern Arabia
sending letters by hand was that it ensured that they reached those they were addressed to. It was not easy to send a letter by post to one of the thousands of immigrant workers in Kuwait unless the writer had exercised more than usual prudence in arranging an address. Personal delivery was simpler, because the people from each place abroad tended to see a lot of each other in Kuwait and knew each other’s whereabouts. As soon as he reached Dubai on the way to Kuwait, the partner invested as much money as he could afford, including what he had been given for delivering the letters, in goods to sell in Kuwait. One thing he could buy was animal ghee, most of the ghee available in Kuwait at that time being vegetable cooking fat made in Europe. For this he had a ready market amongst the people from home, more particularly because ghee can easily be adulterated and in selling it personal recommendation counts for a good deal. And the delivery of letters, which he often had to read to their recipients, including thanks for money received or pleas for help, no doubt helped the partner to collect the debts, though it was also a part of his social life, not to be assessed merely on a basis of financial calculation. Having collected his and other shopkeepers’ debts in Kuwait, he then spent most of the money on buying goods to sell in Dubai on his way back home and stock for his partner’s shop in Rams. No one could make himself rich on these terms, but this was one of the ways of making a slender living, in the spirit of the Gulf proverb, ‘Better a busy jackal than a lion asleep.’ Village shopkeepers, who make what living they can out of a very small turnover, always charge prices substantially higher than those charged in the towns, even when selling to customers who owe them no debts. In the 1950s, emigrant labourers provided an important part of the village shopkeepers’ income, for, impoverished as the villages were, they would have been far more so had it not been for money earned abroad and used to support dependants at home. Shopkeepers not only supplied goods on credit for the families of emigrant labourers, but also provided the initial money needed to enable emigrants to leave home, charging for their help in a way permitted by the shari[a law. A man who proposed to travel abroad needed his fare, the cost of a passport (which amounted to a tax when the passport was provided by some of the shaikhs), and some money to live on until he found work. Furthermore, he needed to leave some supplies in the house to help feed his family. The shopkeeper provided him with supplies to be paid for considerably in arrears and charged him 20 per cent above his normal price. The emigrant left part of these with his family for their use, and part he sold in order to raise cash for his trip. The shopkeeper
Towns and Maritime Activities 145
would refuse to buy the goods back himself, on the grounds that to do so would be tantamount to taking interest on a loan, a thing which was not only contrary to the shari[a in a technical sense but also a particularly serious sin anathematized by the Koran. The emigrant had, then, to sell the goods to another local shopkeeper, unless he could sell them to private people at a slightly better price than the shopkeepers would give him. In any case, he stood to lose doubly on the deal, buying at more than a simple retail price and probably selling at less than a wholesale one. The refusal of a shopkeeper to buy back his own goods was no ruse: he was conforming to religious principle, and failure to do so would have hurt both his conscience and his social reputation. But in keeping his conscience clear he saved himself some trouble in not having to argue what was a just return price for the goods. The bargaining was done for him, in effect, by other shopkeepers on simple terms of the market, free of any social obligation to buy. Thus the other shopkeepers were at once his competitors and his colleague in the workings of this detail of the economic system. Terms of 20 per cent extra on one side, coupled perhaps with a 20 per cent loss on the other, may seem a high price to pay for a short-term loan, and when such considerations arise the step from socio-economic to moral assessments is all too easy. In favour of the shopkeepers, however, it should be noted that the trouble of collecting debts is considerable, and that if one equates their procedure to the taking of interest, as they do not, the interest is not compound. Nor, in a village society, do shopkeepers live in those impersonal circumstances where it is easy to refuse credit in doubtful cases. Village shopkeepers do not become rich men, but in a money economy, where the margin of subsistence is narrow the cost of credit within that margin is high. In such a case as this, a shopkeeper is not unlike the owner or captain of a trading dhow: travel is part of his business and he trades on his way, and also he facilitates other men’s travel abroad for work in that he provides credit for their dependants to live on while they are away. But there is one great difference: unlike the boat owner, the shopkeeper does not himself provide the work. The boat owner provided, as a few still do, both credit and work – one cannot easily call it employment since it was not wage-based – and by so doing he acquired a position of patronage and power such as is not accessible to people simply dealing in credit. The bond between a boat owner and his sailors was often lifelong, and even more than lifelong, for it continued in the families on both sides. That this was so in the case of the owners of merchant boats and pearling boats explains the high prestige of those two occupations
146
Shaikhdoms of Eastern Arabia
in the past. It is what made them the truly ‘noble’ occupations, based on economic position but from there extending, because of the control of men as well as money, into much wider social and political spheres. Throughout the shaikhdoms of the Gulf, it is said that the profits of trading voyages never compared with those of pearl fishing, and certainly in the Trucial Coast the pearl merchants and captains greatly predominated over the owners of trading boats. But both sorts of occupation can be classed within the same broad social category as providing work and patronage. When the economic position of the boat owners declined, a decline accentuated in the case of the pearl merchants by their crews’ inability to repay outstanding loans, then their social and political position began to decline correspondingly. But prestige passes less quickly than wealth, and it has certainly not died utterly even yet. Traces of the old social dominance remain, and some of the former socio-economic practices survive. The most remarkable case that I came across of this survival of the old order of things into the 1950s related to a family of trading boat owners settled at that time on Failakah, the fishing island of Kuwait, with their followers and making up a total community of two hundred and seventy people. With these followers, the family had moved its home several times, ranging both across and from end to end of the Gulf. They had started on the Persian coast, which, it will be remembered, has a considerable Arab population, and then had moved to Khasab, on the Musandam Peninsula, which is officially Muscat territory lying between the mouth of the Gulf and the Trucial Coast. From Khasab they had gone to live in Qatar, before moving to Failakah and coming under the government of the shaikhs of Kuwait. This move of theirs by sea from Qatar to Kuwait accompanied by their followers reminds one of the earlier move which the shaikhs of Kuwait made, according to traditional history. They too, with their followers, had sailed from Qatar, which they are never supposed to have ruled, to the region of Kuwait and subsequently to Kuwait itself. The Kuwait shaikhs had travelled up the Gulf at a more disturbed period, and one when it was still possible for such a group to establish an independent territory. Nevertheless, though the days of the proliferation of tiny autonomous bands of sailors had passed in the Gulf, it was possible to see from the more modest example of these owners of trading boats what a family with its followers is like. The principles remain the same even when not pushed to quite the same lengths, and one of the things the case reveals is how intangible are the criteria of independence and even of the authority of the state if we leave out the standards of an international law that has
Towns and Maritime Activities 147
grown from sources as far removed culturally from these little ‘nations’ as from the jus gentium of the Romans. 11 The principal members of the family which, with its followers, made up a community of two hundred and seventy people among the three thousand villagers of Failakah Island were three brothers, who between them owned four dhows. The whole group had been allowed to come from Qatar in the late 1940s by the Kuwait shaikhs, on condition, however, that they settled on the Island of Failakah, eighteen miles out to sea, rather than in the town of Kuwait itself. At that time, the members of the group were not regarded as subjects of either Kuwait or Qatar, though some of them had left kin behind in the latter. They were in Kuwait territory on sufferance, and it was probably because they were such a large, unified group rather than because Kuwait was overcrowded that they were settled on the island instead of in the town, because by that time Kuwait was readily opening her gates to thousands of immigrant workers from all over the Middle East. Most of the adult men of the group had worked as sailors in the dhows of the family around whom they congregated, though some had also worked as pearl fishers. The dhow trade was, however, declining, and many members of the group began to make their living independently from the central family, some as fishermen and others as builders, whilst some were allowed to go across individually and work in Kuwait. The community as a whole, however, stayed on the island and retained a considerable degree of unity. If there were quarrels, members of the community turned to the family at the centre for adjudication, not to the Kuwait shaikhs or their governor on the island. It was the governor’s majlis, not the religious teacher’s majlis, that the older-established villagers of Failakah attended. When members of the teacher’s community had to have dealings with the Kuwait authorities they did so through the central family. Had any of them committed serious criminal acts, then it would have been necessary for the Kuwait government to deal with the matter, for the members of the central family could not resort to force; but had such a thing happened it is likely that the whole community would have found itself expelled from Kuwait. In the matter of pronouncing on what was right and wrong, the eldest brother and effective head of the family was as well qualified as most people, since not only was he elderly, experienced and judicious, but he was also a religious scholar. Every morning he would sit in his public room reading out long extracts from religious books to those who visited him. Whilst refreshments were served he would expound his texts, answer questions on these and any other matters anyone wished to raise, and discuss issues
148
Shaikhdoms of Eastern Arabia
of the moment. The morning gathering consisted almost entirely of his own followers, but they attended faithfully. Often as many as thirty people attended the meeting, and for those who gathered there it was the main social occasion of the day. For the other villagers of Failakah, it was a different matter: he was a stranger and so his opinions had no special weight. The questions discussed in public often had to do with the actual behaviour and problems of members of the community. For example, after a long and moving reading of the Koranic story of Joseph and his brethren during which some of the audience wept, a man asked whether it was permissible in religion not to reply when someone gave one the greeting of peace. It transpired that the man was making a complaint against his own brother, with whom he was having a quarrel, the reason for the quarrel being that the brother had asked him to give his daughter in marriage to the other’s son at a reduced bridewealth and he had been unwilling to do so. The brother had no strictly legal cause for complaint but he certainly had a moral one, and ventilated his objections openly, with the religious teacher on the whole taking his side when he came to sum the whole matter up. This was no conversation or argument such as one might find between people in a group of friends. The complaints were being brought before the leader of the group for a ruling. His ruling had no judicial force – nor, indeed, in that particular question would the ruling of one of the shaikhs, since the matter was a personal one in which there was no place for compulsion – but there certainly was a place for pressure to be brought to bear on the side of what was thought to be the appropriate fraternal behaviour. The procedure in this matter would have been the same in the majlis of a shaikh. To refuse advice and peacekeeping laid the brothers open to public shame, and moreover if they refused help on an occasion such as this they might find it slow in coming on some other occasion when they needed it. I have said that there is a difference between such a meeting and the majlis of a shaikh in that in criminal or political matters the shaikh has the ultimate sanction of force at his disposal. This is constantly made manifest by the presence in a shaikh’s majlis of guards sitting ready to carry out his commands. But most town shaikhs are only willing to use force cautiously and as a last resort, unless they are dealing with people they think to be of no account. And with bedouin shaikhs there is much less of compulsion and of the sanction of force, for bedouin shaikhs only have recourse to these in circumstances of crisis. The sort of community gathered around the leading family, such as I have been describing, is called, in the Arabic of the Gulf, jama [a, a word
Towns and Maritime Activities 149
which carries the underlying sense of being ‘gathered together’. The central family around which the other families were gathered was of some significance to the local world in general, partly because of its financial resources and partly because of the number of its followers. Any other family of the group was of little significance to the local world except by virtue of its membership of the group. So long as it remained in the group, however, it could expect the family at the centre to take advantage of its higher status in smoothing out any difficulties it might have with people outside the group. The self-respect of the central family and its prestige both within and outside its own depended on its being ready and able to help in appropriate cases. On the other hand, there was no formal obligation to require members of the central family to make these efforts on behalf of their followers. No other authority had placed the family in a position of leadership. It had not been imposed on the group from the outside nor had it been elected, or even accepted, from the inside. It was simply a family that had gathered other families around it. The bond tying the other families to it was initially economic but had developed into something more embracing through the patronage the central family could provide. This does not mean that adherence to this or some other leading family was essential to the well-being of any member of the group. It is simply that, on many occasions, he would find it an advantage to have some influential family ready to take his part when he had to deal with other influential or powerful people. To the group as a group, however, the central family was essential. But for its continued existence, the group could have no corporate identity. The community which remained, still gathered as a cohesive group, around this religious teacher and his family is not a representative social fact of the present day. It should be regarded, rather, as a survival from a period of a generation earlier when pearl fishing and the dhow trade still flourished. With the decline of the first of these maritime activities and the collapse of the second, the economic basis of such groups was removed, the bonds which held them together became tenuous and the groups themselves began to melt away. But so long as these maritime industries continued to provide the livelihood of townsmen and coastal villagers in the shaikhdoms of the Gulf, the social groupings formed by the economic organization of the industries played a telling part in political organization. The largest scale of economic organization existed in pearl fishing, the dominant industry of the Trucial Coast, and the economic form of the industry extended itself into the social order. The pearl merchants and boat owners exercised control not only over money
150
Shaikhdoms of Eastern Arabia
but also over men. This is the reason for the superior honour attached to their business as compared with all others – particularly since their followers bore arms.
Pearl fishing Although pearl fishing still exists in the Gulf, it is on a very much reduced scale compared with the past and has lost all its former economic and social importance. There are few important pearl merchants left, and some of them have sources of income outside the pearl trade. For the men who go out in the pearling boats, pearl fishing is no longer their lifetime’s work but simply an occasional occupation. No one depends upon the industry any longer, whereas up to thirty years ago it was the mainstay of coastal society. Some of the old customs of loans and sharing survive in the present reduced state of the industry. But the total system which applied in the past, when pearl fishing was central to the life of the coastal people and important to many even in the interior, has gone. The way pearl fishers manage their affairs is now of very little public importance, whereas in the past it was of such moment that certain breaches of uniformity incurred punishment. For these reasons, I shall speak of pearl fishing in the past rather than the present tense. Some pearl fishing is possible in the Gulf all the year round, because pearl-bearing oysters are to be found near the shore as well as in the deeper waters. But it was deep water that used to provide the most profitable fishing, and here divers could only work when even the depths had been warmed by the oppressively hot summer of the Gulf. The work of diving was in any case hard, but in the full heat of the summer, gaid, it was possible for a diver to continue in the sea from about an hour after sunrise to nearly sunset with a break in the middle of the day. This was the procedure which was followed in the main pearl fishing season,12 which started in the second half of May and lasted for about four months. In the earlier part of the summer, saif, there was a period of up to a month during which it was warm enough to dive so long as the divers were able to spend half of their time relaxing and warming up on the boat. Thus the earlier part of the summer provided an additional short season, 13 and there was a further possible season14 in conditions similar to those of the early summer after the main season had ended. Whilst cold affected the conditions of the divers’ work, it made no difference to the other main part of the labour force, the ‘pullers’ whose job was to let the divers down to the seabed and to
Towns and Maritime Activities 151
help pull them up again when they had completed a dive. Hence the proportions of the labour force in a pearling boat were determined by the season of the year. During the main pearl fishing season, the work would be done with one puller to each diver, but when diving was done in rather colder circumstances one puller would look after two divers, who took it in turns to rest on the boat. Whereas diving was a relatively skilled occupation, the job of the puller was simply hard work. To enable the diver to get to the seabed quickly and not waste time plunging, the puller let him down on a stone attached to a rope and had to pull the stone up again after the diver had finished his descent. In addition to this rope, the diver had a thinner rope which he held between his toes and to which his basket for the oysters he collected was attached. When he had stayed under water as long as he could bear, he jerked on the rope as a signal to the puller to help him back to the surface and the side of the boat. The puller took the oysters from the diver’s basket whilst the diver got his breath back and prepared to go down again. With labour comprising both the skilled work of the divers and the unskilled work of the pullers, the pearl fishing industry provided opportunities for casual as well as for regular labour. For many people of the coastal towns and villages pearl fishing was the only source of livelihood, there being little for them to do during the rest of the year, though others fished, manned trading boats or attended to what little cultivation there was. For those not capable of the skill and stamina required of the divers, the less rewarding work of a puller was available. But in addition bedouin came from the interior to the coastal towns, as they did in all the shaikhdoms of the Gulf, to work on the pearling boats, and other immigrant labourers came by dhow from the Persian coast and from as far as Aden and Somalia. Quite a number of pearling boats were owed by people who lived at oases in the interior such as Buraimi and Liwa and who came down to the coast in the summer to join in the work of the main season. Pearl fishing, then, had a substantially larger labour force than that provided by the coastal towns and villages alone. The coastal towns profited from the extra labour in more than one way. Apart from the profit of the pearl trade itself, there was profit from provisioning the boats and further profit to be made by selling goods to the departing migrant labourers at the end of the season when they had received the pay for their work. People from the interior would take back household supplies, which were cheaper to buy on the coast than at the oases. People who did not need to take supplies home would at least take back
152
Shaikhdoms of Eastern Arabia
presents for their families. Thus the coastal towns profited indirectly as well as directly from the pearl fishing industry, taking a profit twice over from much of the money that immigrant labourers earned. The dhow trade profited from providing transport, but much more important was the profit of suppliers of cloth and of foodstuffs such as rice, dates, tea, coffee and sugar. One Kuwaiti merchant, complaining of the depression of trade in Dubai, said of the pearl fishing days, ‘Those were times when the poor had money; and prosperity comes from the poor, not from the rich.’ Now that the shaikhdoms of the Gulf have other sources of employment and wealth, no one who worked at sea in the pearl fisheries regrets the passing of the industry. What is remembered most is the squalor of life on the boats, and particularly the nasty food, the dirty sodden rice and the maggotty dates that the pearl fishers had to eat for months on end. The cost of food eaten on the boats was subtracted from the profits and so the food provided was at the lowest level anyone could tolerate. It was often overcharged nevertheless, either because it was bought on deferred terms when the captain could not afford to pay cash for it, or else because it was made to provide an extra, hidden profit for the captain himself. In addition to bad food, there was also the physical exhaustion that the work entailed, the lack of help if anyone fell sick, and, for the divers, the dangers of the sea, not so much from sharks, which seem to have caused little trouble, as from sting rays and poisonous jellyfish. People complain much more of the bad physical conditions of pearl fishing than of the chronic affliction of debt, though debt is said to have been lifelong for most of those who worked regularly in the industry and often, moreover, handed on from father to son. Debt restricted people’s freedom of action, tied them to their work, and laid them open to exploitation. But, being endemic, it involved no disgrace. Nor could a creditor practise the level of extortion and emotional persecution possible to English moneylenders in the last century 15 (and perhaps even to-day). Speaking of the pearling days, people say with great seriousness that what sustained the pearl fisher was the hope that Allah would favour him and provide him with sudden good fortune. When discussing the state of mind with which pearl fishers went to their work, people speak much of divine providence: Allah provided sustenance, not man, and reliance on Allah was part of the faith. A number of pious stories circulated, such as that of the ant with which the Prophet Solomon conducted an experiment. Solomon had asked the ant how much it ate in a year, and the ant had told him it ate one grain of rice. And so
Towns and Maritime Activities 153
Solomon shut the ant in a box with a grain of rice and kept it in his treasury. When a year had passed, he opened the box and found the ant with half of the grain of rice still left, and so he asked the ant why it had told him it ate a whole grain in a year whereas it had only eaten half of one. The ant replied: When I was living in the open, my sustenance, rizq, came from Allah, and He provided me with a grain of rice every year, but when you, Solomon, shut me up I knew that I now depended on man, and feared that after a year had passed man might be negligent and forget me. This is why I tried to eat less than I normally do. Like some other pious stories of the Gulf, the story of Solomon and the ant will not perhaps bear much logical attention. But to question it would have been improper. It was not regarded as an edifying fantasy but as being literally true. And to doubt it would have been to undermine the reassurance it conveyed. Hope, rija], in a theological sense, the hope of divine power, was supremely important to the pearl fishers, for their work had much uncertainty. Profit depended on the quality of the catch, not on the bulk. A few big and perfect pearls among the mass of small pearls always found by the fishers in any boat made all the difference between a failure and a glorious success. Talk of the pearls themselves acquired a religious overtone: when comparing natural with cultured pearls, pearl merchants would say that cultured pearls are man-made and lose their lustre after a few years, but the natural pearl is from Allah and stays the same for ever. Pearl fishing, then, was something of a laborious game of chance played by believers. Nowadays some people have invented a gambling game with unopened pearl oysters, selling them off stalls five at a time to passers by who open them on the spot in the hope of finding pearls. The profits must be high. As in merchant seafaring, much of a pearling captain’s expertize lay in experience. It was a job for the middle-aged and even for the elderly, not for young men. The pearling captain needed an accurate memory of the pearl banks and the conditions of the sea bed. The fact that it is still related as a sign of the remarkable progressive-mindedness of the late Shaikh Mani[ of Dubai, himself a pearl merchant, that he made a map of the pearling banks of the Gulf, shows how captains in general relied on memory rather than on any written record in order to return to what they judged to be the most promising spots for fishing. The same story is current throughout the length of the Gulf, referred to different captains
154
Shaikhdoms of Eastern Arabia
in different places, suggesting how accurate their knowledge was. It relates how one year a pearling captain dropped a saucer (at other times it is a tea glass or a coffee cup) over the side of the boat out at sea, and how when he came back the next year he was able to tell a diver exactly where to dive and recover it. Whether true or not, this story was represented as being true, and generally accepted. It expressed the pride of the pearl fishing people in the skill and wisdom of some members of their community, whatever they may have lacked in education – a pride, in short, in their own intelligence. Knowledge of the sea bed was not the only thing required of a captain. He needed some knowledge of the qualities and values of pearls, since it was he who had to sell them to a merchant on behalf of his crew. Moreover, he needed to know how to manage his crew, particularly at the pearl banks, where fatigue might make them quarrelsome, but also year in and year out when they came asking for a special subvention. The ant’s example of trusting in Allah but treating man with greater caution was certainly followed by the pearl fishers, both in their dealings with each other and with their captains. Whilst the oysters were being opened, four ‘guards’ or ‘watchers’, haras, from among the crew would sit inspecting the work to make sure that no one who found a large pearl could hide it away to keep for himself. I never heard, however, of anyone who had tried to steal a pearl on a pearling boat, and was told that a thief would have been frightened of being beaten to death by the rest of the crew. There was no wage economy. It was their profit he was stealing. The crew also had witnesses present when their captain was selling the pearls to a merchant, to make certain what price he had been paid. Nevertheless, they were aware that the captain sold to the pearl merchant at a price considerably lower than that at which the merchant would later dispose of the pearls, and they knew too that some captains secretly accepted presents from the merchants in order further to lower the price. A diver who was particularly good at his work might, again secretly so as to avoid jealousy, be given a present by the captain in order to induce him to join his crew for the season. In this case, if the diver was wise, he would make sure that the sum was recorded as a present, or else at the end of the season he might find the captain claiming that it was simply a loan that the diver had received, and taking the money back again. As I have said already, there was also in very many cases outstanding debt to keep people working in particular boats over the years. But even a debtor could transfer to another boat so long as he could persuade its captain to loan him money with which to free himself from the captain of the boat he wished to leave.
Towns and Maritime Activities 155
The division of proceeds in the pearl fishing industry was somewhat different from that applied in fishing or the trading voyage. In pearl fishing, the share of labour in the operation was higher in proportion to that of capital equipment, and labour took a higher share of the proceeds. But also, in the Trucial Coast (the method of division in Bahrain being slightly different), labour paid the cost of damage to equipment and also some of the price of special responsibility. Whereas in the other maritime industries the first calculation was to reduce gross profit to net profit, the first thing the pearling captain did when he began to share out the proceeds of the season’s work was to deduct a fifth from the gross profit, that is, from the selling price of the pearls. This was later to be divided between the boat and the divers. He then subtracted the cost of food and damage from the remaining four fifths and also, from this four fifths, he extracted the equivalent of one per cent (expressed, however, in terms of ten in a thousand) of the gross profit. This one per cent, calculated on the total selling price of the pearls but deducted from the four fifths and not from the whole, was an extra payment 16 for the mate. The sharing out of what was left of the four fifths after these deductions had been made depended on the manner of diving. If the manner of diving was two pullers to one diver, ghaisain ghaisain, then the sum was divided into equal shares to include everyone on the boat, excepting that the young assistants17 counted as only half a share each and that one additional share was calculated, which was a tax, taraz, for the shaikhs. If the manner of diving was one puller to one diver, ghais ghais (an arrangement which became rare in later years), then each diver took twice the share of each puller. The captain took a diver’s share, whilst the mate and cook each took a puller’s share, and the assistants took half a puller’s share each. It will be seen that in either case, whether there were two pullers to one diver or one puller to one diver, the basic group for division was three, divided simply into thirds when there were two divers to one puller in the working group and in the ratio of two thirds to one third when there was one diver to one puller. There remained to be dealt with the initial fifth that had been subtracted. This fifth was divided into halves, and one half was divided equally amongst all the divers, the captain, but not the mate, being counted as a diver. The other ‘half of the fifth’, nuss al-khums, as it was called, was for the boat. If the captain was not the boat owner, then it was for the boat owner to pay him out of this part of the proceeds, according to previous agreement, the sum usually being about a tenth. The captain himself usually gave a present to the cook.
156
Shaikhdoms of Eastern Arabia
Just before the pearling season began, every man who was going to work in a particular boat received a loan from the captain to enable him support his family while he was away. The loan, guwad, which was a set sum decided upon by the ruler of the state in consultation with more important captains, also served a second purpose, in that its acceptance by the pearl fisher constituted a contract between him and the captain whereby the pearl fisher committed himself to working in the boat for that season. Once having accepted the loan before witnesses, the pearl fisher was not at liberty to withdraw from the contract even if he was willing to repay the loan. A slip was written on his behalf stating the terms of the contract in the following words: I state that I am N son of N and that I have contracted for such and such a sum of money from the aforementioned [the captain whose name was at the top of the paper], my obligation only being ended by fulfilment and completion [of my work].18 He signed this or put his thumb print on it and it was retained by the captain. The sum was subtracted from his share of the profits at the end of the season. Though the form of loan paid at the beginning of the season was the only one which constituted an irrevocable contract, a pearl fisher who borrowed from a captain between the end of one season and the beginning of the next, as many found they had to do, committed himself to join the captain’s boat for the next season unless he could repay the debt or persuade someone else to do so. Like the loan for the season, these loans were free of interest, and the consideration lay in the obligation to work. Debt being widespread, the captains operated a system of passes to prevent men who owed money to one captain from defaulting and going to work for another. Having received a between-season loan, 19 a man would be given a slip certifying how much he owed the captain, worded in the following way: To any captain who should see this: we are owed the sum of [so much money] by N son of N. Let anyone who wishes to sign him on repay the sum. 20 Any other captain could, therefore, accept the man’s services on condition that he advanced him the money to repay his original debt. When a man left his captain, owing him nothing, he was given another slip to certify that he was free. The slip was worded:
Towns and Maritime Activities 157
To any captain who should see this: N son of N owes us nothing. Anyone who wishes to sign him on may do so. 21 In respect of loans, casual labourers who came to the Trucial States from a distance and normally lived outside local jurisdiction were naturally in a different position from the pearl fishers who lived locally and worked regularly in the fisheries. Only the latter could raise loans from the captains other than the official loan provided before the season began. Winter loans were at the captains’ discretion, according to what they were prepared and able to lend particular families. Such loans might continue from year to year. The loans were usually in the form of goods, but it was not necessarily the captain who made a profit on the goods, since he often had to take them over as a loan from a merchant or another captain. In much the same way as the pearl fisher was bound by debt to a captain, the captain himself was bound to a pearl merchant. He commonly had to borrow from the merchant in order to provide the loans that were expected of him by his crew and to obtain provisions for his boat during the season. By virtue of his debt, the captain was obliged to offer first refusal of his pearls to his creditors. If they were offered to any other pearl merchant, customary law forbade him to buy them. This does not mean, however, that the merchant could underprice the pearls at his whim. He was expected to pay what, in comparison with what other merchants were paying, was considered a reasonable price for the goods. A captain who was seriously dissatisfied could take the matter to the shaikhs, who would then ask some leading pearl merchants to say what they considered to be reasonable minimum price for the pearls. Moreover, a merchant who was particularly mean damaged himself in another way. When a boat had had a good season, both captain and crew might be free of debt and have enough money to support them until the next season. If they were dissatisfied with the pearl merchant with whom they had been dealing, they could then transfer themselves to another who had the reputation of treating his captains and crews better. Or, failing this, the captain might be able to transfer his loan, and with it his services, like a good diver who was dissatisfied with his captain. In either case, the mean merchant stood to deprive himself of his sources of supply, and the fact that captains were generally tied to particular merchants would then make it difficult for him to buy and would militate against his recovering his position. Many pearl merchants themselves were not totally independent. Lacking capital, the poorer pearl merchants in small places would have,
158
Shaikhdoms of Eastern Arabia
besides their own business, run on a capital of a couple of hundred pounds, the position of agents for rich merchants in the bigger towns. A poor merchant would buy pearls with cash borrowed from a rich one and then divide the profit with him, each party taking half. In hard times, even important pearl merchants might have to borrow money, mortgaging their houses and other property. The fact that, throughout the whole scale of the pearl fishing industry and the pearl trade, indebtedness restricted men’s freedom to seek work or do business where and as they chose, might lead one to surmise that financial independence, with its concomitant freedom of action, must have held a high place in the aspirations of everybody. Surely freedom from debt vouchsafed freedom to manoeuvre and the chance of standing out for the most favourable price whether in selling or buying? At the same time, the very fact that indebtedness extended so widely through every occupation connected with pearls and even into the formal organization of the whole industry might lead one to wonder whether it did not serve some purpose beyond the exploitation of labour by capital and even hold some advantage for the debtor. Otherwise, how could such a system survive on the edge of the desert, bordering on a life so noted for its insistent freedom and impatience of restraint as that of the bedouin, and including in its regular labour force people from the near-bedouin oases and settled members of bedouin tribes, open to the scorn and the instigations of their nomadic cousins, some of whose daughters were their wives at home? What is said of the independent operator throws some light on both of these questions. The independent operator, khaluwi, was in fact no highly respected or envied character. He was a minor figure, significant to no one, a man in a small way of business who went offering his pearls all the time to every possible buyer. The general attitude towards him is reflected in a jocular phrase used of a loose-living wife: ‘So-and-so’s wife is an independent operator, khaluwiya, she’s selling in the market.’ It was the lesser people in the pearling industry who sold their pearls to anyone, without having a standing arrangement with a particular merchant. Hawking pearls round the market was a sign of the small man, who was regarded with a certain contempt. His freedom of action meant isolation; when he found himself in difficulties, he had no one he could turn to. What he significantly lacked was not so much debt as credit. A regular arrangement to work in a particular boat or to sell to a particular merchant provided an insurance against financial misfortune in any particular year. As I have mentioned, the proceeds of a season’s pearl fishing in any one boat were uncertain, because everything depended
Towns and Maritime Activities 159
on finding pearls of high quality. Hence, however hard-working and prudent a pearl fisher was, he could never be sure of husbanding enough of his own resources to survive not just one but perhaps two or three poor seasons in a row. The ordinary vicissitudes of human life might also face him with financial difficulties. By providing winter loans, in a relationship which extended readily from the economic to the personal, the boat owner evened out financial hardship between his pearl fishers. If he did not have more than one boat himself, thus reducing the extremes of good and bad years, the boat owner was likely to come of a boat-owning family who would help each other in times of need. He could also, on behalf of his fishermen or himself, turn to his pearl merchant for help. Living from season to season meant living from hand to mouth, which is what the independent operator did. The system of debt, on the other hand, was a system of insurance spreading liabilities over from year to year. The pearl fisher or the boat owner had at least this advantage out of a system that constricted his sphere of action and possibly reduced the profit he could make at any particular time. As well as its importance for individuals, this fact clearly had its overall significance in the organization of the industry. First of all, the fact that most people did not need, in order to survive bad years, to keep substantial sums of money hoarded implies that the system allowed a greater proportion of local capital to be productively invested in the pearling industry than would otherwise have been possible, thereby increasing the scale at which pearl fishing could be carried out. Secondly, without any provision for mutual aid – even though the aid had to be paid for – the risk to individual families would have been correspondingly greater, and one can therefore presume that the number of people prepared, or even able, to live by the industry would have been smaller. There can, therefore, be little doubt that for these two related reasons the credit system increased the scale of exploitation of one amongst a very limited range of natural resources along much of the coast of eastern Arabia. In the Trucial States, the spirit of mutual aid was never so strong, however, as it was in Kuwait where it is said that the merchants never allowed each other to go bankrupt until in recent years. There, if an owner of trading boats, for example, lost heavily by shipwreck, then his creditors would cancel part of his debts and other well-wishers might give him such contributions as they could afford to allow him to stay in business. This was common custom rather than any exceptional charity, and it is easy to see how, in the chances and dangers of maritime life, the custom served the prosperity of the whole Kuwaiti community. People in the Trucial Coast knew that this custom
160
Shaikhdoms of Eastern Arabia
obtained in Kuwait but never adopted it themselves. The most that was usual was to give contributions when fires broke out in the towns and left people destitute. When one considers the question of debt in the traditional economic system of the Trucial States, and compares the resultant situation with traditional landlord and peasant economies in the Middle East, one difference that stands out is the status of basic resources. In peasant societies, debt was one of the means whereby the land fell into the hands of landlords. Once the landlord had taken his share of the harvest not only for the land but for the water, the seed and the animals that drew the plough, the peasants were often left with only a tiny share of what their labour produced. Even though pearl fishers were often over-charged for their food and under-paid for their pearls, nevertheless their labour started with a great advantage over Middle Eastern peasants in general in that it received almost nine-tenths of the price of the catch. Moreover, the pearl banks themselves, which were the basic resources, could never fall into private hands. Perhaps this is one reason why the formal share of labour in the catch continued to stand so high. Indeed not only could the pearl banks never become private property; they never even fell into the control of particular states. The sea bed was common property – or, more strictly, no man’s property – until it was divided up at European instigation for the sake of prospecting for oil. And the same freedom of the basic resources which applied in the pearling industry also protected the interests of those fishermen who made their living by ordinary fishing. Because the pearl banks were not susceptible of individual ownership, neither wealth nor credit could gain control of basic resources. Anyone could go fishing for pearls so long as he had the minimal practical requirements of a boat, a little equipment and some supplies. He needed no concession and no licence. He lived, indeed, in a collecting economy, though it was one that was organized in terms of markets and not of subsistence and the things collected had to be translated into terms of cash. Still, there was a limit to what capital could gain possession of. Similarly, the profit-sharing system whereby earnings were divided meant that in the pearl fishing industry there was no ‘labour market’ in any simple sense of the term. As far as the yearby-year relations between pearl merchants, boat owners and crews were concerned, the fact that the rewards of labour were calculated in terms of customary sharing arrangements meant that the system was less open to private and piecemeal manipulation than an unregulated wage system could be. At times of depression, no individual capitalist could
Towns and Maritime Activities 161
contract directly for a greater share of the profits, and indeed, even when there came the great depression of the 1930s which was to destroy the industry, the customary shares held unchanged. Casual labour, usually the most susceptible to the fluctuations of a market, retained the same formal shares as regular workers. The immigrants who came irregularly, with no local person to interest himself particularly in their rights, had exactly the same shares in the profits of a pearling expedition as the local people who worked with the same captain from year to year. This is not to say that the persistence of customary proportions in the shares distributed should be taken to imply that custom was ‘sacrosanct’ in any superstitious sense. But the shares of divers and pullers were their customary rights, and these rights were the widespread interest of a high proportion of the coastal population. Moreover, they were quite simple and specific. Hence they were not easily susceptible of direct and individual erosion. But, as facts given already have implied, the system was neither totally inflexible nor was it incapable of private manipulation. Both flexibility and manipulation lay in the system of credit, a system which lasted the year round but had its particular power, the stronger for its lack of clear formality, in the unproductive period of the year. Most people depended on credit, and when money was in short supply, either in general or for individuals, the price charged for loans, that is to say the price charged for supplies that were to be paid for in arrears, could rise. In providing winter loans, the owners of capital and those whose credit was good invested in future labour. The charge put on the supplies provided on deferred terms, whether for families during the winter or for provisioning boats in the summer, effectively altered the gains which any indebted pearl fisher had to take home at the end of the season. The debtor was obliged to go on working for his creditor. This being the case, simple logic might suggest that, whilst holding to formal sharing arrangements, the owner of capital still suffered no restraint as to the real proportion of profits that he could take when all the terms of work were taken into account. But like most simple logic this argument has a flaw, the flaw in this case lying in the question of obligation. Who was there to oblige the debtor to work, and what was the price of compulsion to the creditor? In this question, the economic order of Trucial Coast society meets up with the political order and a situation where the possibilities of control were slight and the exercise of compulsion was arbitrary. There existed the possibility of enforcing rules, but it could be costly, chancy and humiliating to try to do so.
162
Shaikhdoms of Eastern Arabia
The following example may illustrate how pliable a formal situation could be. When the pearl trade declined, perhaps those who suffered most were the slaves who worked in the industry. These slaves were, in the first place, investments, assets owned and physically saleable for profit. Their earnings belonged to their masters, and it was for their masters to feed and clothe them. As slaves, they lacked all the rights that the most oppressively exploited debtor could lose, and more. During the times of depression, there was one group of slaves belonging to a boat owner in the Trucial Coast who became dissatisfied with the poor food they were given to eat and the tattered clothes they had to wear. They told their owner that it had become obvious that he could not afford to keep them any longer and that they found their life with him intolerable. Then they left him and went and sold themselves to a rich man in another state of the Gulf. One of them came back to his former master to give him the money and tell him what his slaves had done, and then he returned to his new master. Later, when emancipated, some of these ex-slaves continued to visit the man who had been their first owner whenever they returned to the town. The question of the practical enforcement of the law is, however, only a part of the broader issue of social and moral expectations. There being no separation between judicial and political responsibility, a shaikh prepared to enforce a law once might not find it expedient to enforce it repeatedly; and a man who was constantly appealing for the enforcement of his rights could easily be regarded as a nuisance and ignored, on the grounds that he should have managed his interests with more prudence or less cunning. A creditor had to consider whether he ought, in decency, to insist on the repayment of a loan, and what would be the effect on his personal reputation, bearing in mind that everyone would hear of all the circumstances. He also had to consider whether the ruler would be prepared to enforce the law at his behest, or would say instead that he was too busy. Sometimes he might even have to wonder whether the ruler might not find some excuse for appropriating the money after obtaining it on his behalf, thereby depriving him of his social advantage as a generous creditor and at the same time making him appear as a fool and a person of no consequence. The problem of the enforcement of laws in respect of the formal rights of creditors and owners of property thus takes us beyond economics into politics, and here the number of matters that have a bearing on politics is very wide. Beyond matters of practical power, organization and expediency, they include, for the public in general, standards of morality and social conformity, and, for the individual, the half-thought and half-sensed
Towns and Maritime Activities 163
concerns lying along that complex scale that runs from prestige, through self-confidence and into the conscience. Here it is important to recall the higher measure of prestige attached to pearl merchants and boat owners. These two related callings were thought by the Arabs of the Trucial Coast to be the only truly honourable ways in which a townsman of high status could earn his living. As far as the Arabs were concerned, shame attached to some occupations absolutely, and it still does, whereas with regard to other occupations it was relative to inherited status. Thus a man might prefer to go hungry rather than earn money by working as a porter or water-carrier. Some find it embarrassing even to carry their own possessions, let alone possessions belonging to anyone else; talking about the days when poverty came upon the Coast, people often used to speak of men who had lost everything and would yet, when they could only afford one fish for the household lunch, pay a porter to carry it home for them. I never saw this happen, though I have seen an elderly shaikh carrying home a fish dangaling from his finger on the piece of raffia that is threaded through the head – whether out of eccentricity or from some private principle I do not know. But the example of the porter had become almost proverbial. On the other hand, to be a manual worker as a member of the crew of a boat, whether in pearl fishing or other work at sea, carried no shame for ordinary people. It was decent, respectable work, so long as the man who had to do it had not been born of a family accustomed to higher occupations. Proceeding now to work that required capital, there was not much prestige in being a simple retail shopkeeper, though it was perhaps better than being a pearl diver in that it provided more opportunity of becoming prosperous. To be a merchant, however, was much superior, for it implied wealth – in Arabic of the Gulf, the word ‘merchant’, tajir, is commonly used to mean ‘rich man’ whether the rich man is in fact a merchant or not. But the great prestige attached to the ownership of the larger boats and even more to the trade in pearls was related not only to the control of wealth but also to the control of people. It was implied by the existence of the jama[a, the group of people collected around a particular family, which I referred to in an earlier section of this chapter. The prestige of the boat owners and the pearl merchants corresponds to the status in social leadership, patronage and political power which this group of working dependents, the jama [a, conferred and supported. So, by virtue of the groups their wealth and business gathered around them, the richer pearl merchants and boat owners were within the highest ranges of prestige. Only the shaikhs were above them. And even
164
Shaikhdoms of Eastern Arabia
this comparative ranking was equivocal. To be a pearl merchant was no shame for a member of a ruling family, whereas those members of ruling families who had been relegated from the leading positions of political power and wealth were people of much less significance than the greater pearl merchants. Thus creditors who perhaps paid their men too little for the pearls they produced, perhaps charged them too much for the food they were in need of, and ultimately offered them the bondage of debt, occupied some of the main positions of honour. But of course this is not the whole story. The general characteristic of the merchant class of society was to accumulate and centralize wealth – had that not been the case, then the pearl industry would have had to be organized differently if it could have been run at all. To families that lacked wealth, the dominant positions in society were not accessible. But within the wealthy class, the prestige of particular individuals and families depended on their manipulation of finance in relation to people rather than their manipulation of people in relation to finance. As in many places more isolated than the Trucial Coast, the society was too small to regard withdrawn exclusiveness as other than a mark of insanity. It was also too insecure to permit of impersonal management conducted through any apppointed hierarchy of intermediaries. Personal acquaintance with the great and powerful, extensive enough in its range of families of shaikhs and of rich men to be comparative, was common; at no more than one remove it was almost universal. There was neither need nor opening for fantasy in the minds of the poor, and so little was available for elevating the rich excepting their own behaviour, given the financial resources they had at their disposal. The local terms in which men were judged were generosity, which wealth could make superior; wisdom – in the sense of wordly wisdom and intelligence as well as the philosophic variety; truthfulness, unless overriden by such wisdom; and courage. The lack of any of these could mean the loss of both prestige and wealth.
5 The Shaikhly Families
A family tree (Fig. 5.1) illustrates the growth of ruling families. It gives the descendants of two of the sons of Sultan bin Saqr, who became shaikh of the Jawasim in 1803. In these two lines are included the shaikhs who are now associated with Ras Al-Khaimah. Other descendants of Sultan bin Saqr are descended through other sons and are associated with Sharjah. These have been omitted. The third generation included in this chart, that is, the generation of shaikhs Muhammed bin Salim (bin Sultan bin Saqr) and Sultan bin Salim (bin Sultan bin Saqr) is still alive. Shaikh Muhammed bin Salim, the father of the present Ruler of Ras Al-Khaimah, 1 is probably over seventy years old. His brother Shaikh Sultan bin Salim, himself Ruler until he was driven out a few years ago2 by his nephew Saqr bin Muhammed (bin Salim) who took his place, is considerably younger. The last generation mentioned in any line is made up for the most part of children and young men who have not yet reached an age to be politically important. This example of the growth of a ruling family sets one of the main political problems of the Eastern Arabian shaikhdoms. This is the problem of how such a rapidly-increasing number of shaikhs is to be maintained on state revenues and inherited property, when neither the state revenues nor the property increase at a proportionate rate. Here we have one of the main causes of strife within the ruling families. In the example given above, Shaikh Muhammed bin Salim was Ruler of Ras Al-Khaimah. He resigned under pressure and criticism from his brother, Shaikh Sultan bin Salim, who replaced him as Ruler but left him and his sons to continue to live in Ras Al-Khaimah. The sons of Muhammed bin Salim resented the inadequacy of their allowances from the new Ruler, and some years later, Shaikh Saqr bin Muhammed seized his opportunity whilst his uncle was away in Dubai, took the fortified house of the 165
166
Shaikhdoms of Eastern Arabia
Saqr
Humaid
Kaid Muhammed Salim
[Abdullah
Salim
Khalid Sultan Muhammed Khalid [Abd Al-[Aziz [Abdullah Ahmed Majid Sa[id [Abd Al-Malik Sa[ud Khalid Muhammed Majid Faisal Sultan Humaid [Umar Faisal Sultan
Hamad
Sultan
Sultan
Majid Nasir
Salim Saqr Khalid [Umar Fahim Faisal [Abd Al-Rahman Nasir Sultan Hamad Sultan
Muhammed
Salim Ibrahim
Hamad Nasir [Abdullah
Fig. 5.1 The growth of ruling families as illustrated in the descendants of Sultan bin Saqr. Compiled by Peter Lienhardt.
Ruler by surprise attack and set himself up as a ruler, refusing to allow his uncle to return. Thus, among men still living, two rulers have been replaced by rivals. This again, as we shall see, is characteristic. In Bahrain and, very recently, in Qatar and Kuwait steps have been taken to make a distinction between the Ruler’s private estate and state property. This has not yet occurred in any of the Trucial States.
The Shaikhly Families 167
Traditionally there was no distinction. The late Ruler of Kuwait, Shaikh Ahmed Al-Jabir Al Sabah,3 had received payments for the Kuwait Oil Concession though he died before oil production reached its present level. When he died he left an estate said to be in the region of four million pounds.4 This was inherited, according to the ordinary Muslim law of inheritance, by his issue and not by his successor as Ruler. A large sum was, it is true, set aside to be distributed as alms and was recently 5 shared among poor men whose houses had been severely damaged in the heavy rain of 1944–5, but to leave large sums to be distributed as alms is a normal act of charity of rich men in the shaikhdoms as in other Muslim countries, and it therefore remains true to say that this fortune which was accumulated for the most part out of state revenues was treated as any other private estate of a dead man. The development of oil and the spreading of Western ideas have now led to some questioning of the traditional attitude and to a desire to distinguish between public and private income, instead of, as formerly, regarding the income of the ruler as being for him to spend and distribute as he thinks fit with little account taken of whether he received it from taxation or from private property. There is now, to the best of my knowledge, no ruler in the shaikhdoms who does not receive some income from oil companies, though the amounts, of course, vary enormously between the sums received in Kuwait and Qatar and the sums received by most of the shaikhs of the Trucial Coast, whose income from oil companies is usually no more than a very few thousand pounds a year each (except in the case of Abu Dhabi where serious oil drilling is being done and where the income is considerable). 6 Nevertheless, with the decline of the pearling industry, even these small sums have become extremely important to the poorer of the Trucial States as at Ras Al-Khaimah where, by both the Ruler’s assertions and by what private calculations I was able to make, more than half of the income of the Ruler of Ras Al-Khaimah comes from his oil concession. Without these oil payments it would be almost impossible for some of the poorer of the Trucial States to be maintained as separate governments. See the estimate, made at the beginning of the present century by a Persian merchant, of the total annual income of Shaikh Mubarak Al-Salim Al Subah (ruled 1896–1915), Ruler of Kuwait at the time (Table 5.1). 7 It is added that the shaikh denied receiving anything from the pearling boats but that it was generally said at the time (1904) that he took one diver’s share per boat. He denied also taking interest on money, but was said to have lent a large sum to an Arab merchant at 20 per cent interest. 8 Thus, approximately one-third of Mubarak’s income came
168
Shaikhdoms of Eastern Arabia
Table 5.1
Annual income of Shaikh Mubarak in dollars (i.e. riyals)
Date gardens at Fao (Iraq) Rent of shops in Kuwait bazzar Zak at on bedouin flocks Tax on sheep brought into the town Tax payable by butchers Tax on sale of camels Sea customs Duties on exports to the interior Tax on pearl divers Fines Share of fish brought in Interest on money lent to merchants
108,000 9,000 25,000 2,000 2,500 1,500 150,000 20,000 60,000 6,000 7,000 8,000
Total
399,000
Source: J. G. Lorimer’s Gazetteer (1915 edition)
from family property whilst the rest came from taxation. It should be added that Mubarak was a ruler of remarkable strength of character and succeeded in increasing old taxes and imposing new ones, which were abolished after his death. The most important source of Mubarak’s income apart from taxes was his date gardens at Fao. These date gardens were an important part of the investments of the ruling family of Kuwait. When Mubarak was already having trouble with his brother, the then Ruler whom he later murdered and replaced, the brother offered his date gardens at Fao on condition that he should go and live outside Kuwait territory. The Shatt Al-Arab area is, of course, famous for the great quantities of dates produced there, and because of the lack of water in the shaikhdoms dates have little importance in the total economy of the states we are considering. Such date gardens as there are, however, are for the most part in the hands of the shaikhs and the merchants and for many of the shaikhly families they are of some importance. When shaikhs are driven out, it is usual for their successors to take possession of their date gardens as of the rest of the property which they leave behind. The income of Mubarak quoted above also includes a relatively small sum received from letting out shops to rent, and another sum apparently obtained by usury. It is quite possible that the money in the latter case was lent in a legal way: I do not know of the particular terms on which it was lent, but the case is still remembered by the family which borrowed the money particularly because when that family was unable
The Shaikhly Families 169
to repay the debt Mubarak heirs seized houses belonging to their debtors. It is common for members of ruling families to invest money in building and also in boats. Some of the members of the ruling families in the past were pearl merchants. At the present time, many of them own shops and launches, giving them some private income extra to the income from the date gardens which many of them own. The rulers themselves cannot engage directly in trade, but are able to invest money in the cargoes of ships and take the profit from this (when money is invested in this way, the division of profit is usually one-third for the boat owner, one-third for the crew and one-third for the ras al-mal, capital). I have met with examples in which rulers who owned launches were prepared to use their prerogative to bring about the using of their launches in preference to those of private individuals, though such behaviour naturally is resented and damages the popularity of a ruler. There are other means too whereby shaikhs, both those ruling and the other members of their families, can gain money more or less illicitly. In Muslim law and in the custom of these societies of Eastern Arabia, women can and do own and inherit property. Indeed, strictly speaking, the bridewealth paid by the bridegroom’s family to the bride’s family at a wedding is the property of the bride: this is the only payment which is strictly required for a valid marriage. In practice, most of the payment which is made from one family to the other and which we must include in the term ‘bridewealth’ is not received by the bride herself. I give the terminology of the various parts of the bridewealth as they are used in the Trucial Coast: 1. Al-mal ( also called al-mahr and al-sadaq). This is the property of the bride herself. It is always calculated in tumans, a result, presumably, of the earlier use of Persian currency before the more modern use of the Indian rupee. It is paid to the woman if she is divorced, though she may receive it earlier in the shape of gold ornaments. 2. Al-sabaha. This is an additional sum equal to one quarter of the mal. It is only paid in the case of a woman’s being married for the first time and is paid to her father or whoever brought up the girl as compensation for the cost of her upbringing. 3. Al-talaba. This is the largest sum and is made up of presents for the bride’s family, especially her father, but also her mother and brothers, to show the good will of the bridegroom’s family. The bridegroom’s family also pays for the cost of the marriage feast and for musicians.9 It is not at all unusual for the bridegroom to hand over gold ornaments to the bride which are regarded as being, strictly
170
Shaikhdoms of Eastern Arabia
speaking, on loan, so that if the girl is divorced she surrenders to her husband the gold ornaments which she has from him in excess of the value of the mal which is her right. The richer are the families involved in the marriage, the less importance is attached to the presents which are given to the other members of the bride’s family and the greater are the presents given by the bride’s father to the bride herself. Thus, whereas among the poorer people a man may make a certain amount of financial profit at the wedding of his daughter, among the richer people it is almost as expensive to marry a daughter as to marry a son. If a man angers his wife to such a point that she leaves his house and returns to her father and he wishes to persuade her to return to him, or if he divorces her and wishes to marry her again, or if he wishes her to remain with him but wants also to persuade her to allow him to marry an additional wife, it is usual for him to try to ‘please’ her, yurd iha, by giving her a substantial present. This again is her private property. A woman also inherits from her father or mother a share equal to half of that of her brother, if she has a brother, half of the estate if she is the only child and an equal share of two-thirds of the estate if there is more than one daughter. A woman inherits a share of her husband’s estate if he dies before her and a share of her son’s if he dies before her. In all these cases, the property is the private property of the woman herself and is not under the control of her husband. It is unusual for a woman to allow her own property to be confused with her husband’s. She may of course give him money if he is in need of it, but we must remember that a woman is always liable to be divorced and that the more money she has the better is she insured against this. A man’s inability to maintain his wife is in practice the most important ground for allowing her, through the intervention of her father or other matrimonial guardian, to divorce her husband. There are many women who earn a certain amount of money by their own efforts. The poorer women sell the eggs from their domestic poultry and sell their chickens also (the former trade, particularly, being considered to be women’s business and below the dignity of men). Some of them sew or do embroidery for other women. Some even have small shops, for women only, inside their houses. These remarks about the financial status of women in general have been necessary in order to provide a background for the consideration of the women who are either born or married into the ruling families. The women of the shaikhly families have their own private property as other women do. Most of the ruling families of the States refuse to give their daughters outside their own families, even if this means that they
The Shaikhly Families 171
are not allowed to marry at all. The present Ruler of Abu Dhabi10 had eight aunts on his father’s side, and of these six never married. Some people attribute this refusal to give daughters in marriage outside the ruling families to the desire of the ruling families to keep their property inside the family and not to lose it by having it inherited through women by the children of other agnatic lines. It is more common, however, and probably more correct to attribute the refusal to family pride, a pride which, in the case of the ruling families, is not without its practical purpose as will be discussed later. In the Trucial Coast, where women and particularly those of the shaikhly families have a rather higher status than have their counterparts in Kuwait, 11 the shaikhas quite often engage in public affairs and in trade and investment with considerable success. A most notable case was that of the Shaikha Hussa bint Al-Murr, the late wife of the Ruler of Dubai and mother of all his children. Her case is extreme and she is locally spoken of in the Trucial Coast as having been unlike other women, but some account of her activities will show how an outstanding woman of a shaikhly family can act. This remarkable lady was a daughter of Al-Murr bin Huraiz, a member of the Al Bu Falasa section as the Bani Yas, that is, of the same tribal section as that of the ruling family of Dubai. Her mother was the daughter of a shaikh of the village of Bukha, a Hinawi village of the Shihuh which lies in the inaccessible region of mountains near to the Strait of Hormuz. She married Sa[id bin Maktum,12 who became Ruler of Dubai in 1912, and remained his only wife until she died, bearing him two sons and two daughters. At the beginning, as the wives of rulers often are, she was available to be seen privately by the women of Dubai, to intervene in their disputes and possibly report on them to her husband. Such a possibility saves women the embarrassment of having to approach a ruler personally. Then, she made herself available to the men at arms when necessary. This too is not uncommon, for the wife is often the most reliable person to leave in secret control of affairs whilst her husband is away. But later, Shaikha Hussa began to hold a public majlis into which men were admitted, so that there would be two majlises proceeding at the same time, that of her husband and her own. She acquired a great reputation both for prudence and for generosity, and took an active part in planning strategy during the Dubai civil war of 1939. Though she was illiterate, as almost all women of the Trucial Coast still are,13 she engaged successfully in investment. The town of Dubai lies on both sides of a long creek and there is a constant passage of small rowing boats, [abara, ferrying people from one side to the
172
Shaikhdoms of Eastern Arabia
other. She bought a number of these and rented them out to the rowers. She also bought launches for trading with the Persian Coast. Out of her profits she began to build shops in the market to be let out for rent. When the Second World War came, she became even more successful. Dubai was supplied with rations of sugar, rice, tea and other things by the British and was the main centre of food distribution for the Trucial Coast. It was a time of serious shortages of food and clothing in the shaikhdoms and also in Persia. Many people in Dubai and in other parts of the Trucial Coast made fortunes out of smuggling sugar, and also rifles, to Persia, and among these it appears that Shaikha Hussa was one of the more competent. When she died about 1950, she left an estate which included several hundred Arab shops in Dubai, the smaller ones rented at Rs.500 per annum and the bigger ones at Rs.1,000 or Rs.1,200. It is said in Dubai that her sons allowed more than the proportion of her estate stipulated by the shari[a to be inherited by her daughters, since the sons could expect to share in the customs receipts of Dubai and were therefore assured of a substantial income. These two sisters, Shaikha Moza and Shaikha Shaikha, now own much of the suq in the Dairah quarter of Dubai, and they too engage in a little trade. They have also been given gardens at Ras Al-Khaimah by their father the Ruler of Dubai. The late Shaikha Hussa is still spoken of with respect and a certain amount of wonder in the Trucial Coast. Her extraordinary behaviour is considered to have suited her extraordinary character and ability. Other women are not, like her, able to hold their own in public affairs with men as equals and even to make themselves superiors. The only woman I heard of who is said to have resembled her is Shaikha [A]isha bint Muhammed who was the father’s brother’s daughter of Shaikh [Isa Al Khalifa, a former Ruler of Bahrain and grandfather of the present Ruler. Shaikha [A]isha was married to her cousin Shaikh [Isa. This shaikha too is said to have held her own majlis to which men were freely admitted and also to have conducted her own business affairs. The wives, sisters, and mothers of shaikhs, however, even though they do not measure up to Hussa bint Al-Murr and [A]isha bint Muhammed, have greater advantages than other women in being able to give full scope to their abilities. Though it is said that a woman per se is the equal of only half of a man per se in Islam and the law of inheritance is adduced to prove it (together with such other Koranic stipulations as those to witnesses), and men will say as generalization: [andana ar-rayyal afdal min al-hurma bi-darajatain, ‘With us [the Muslims] a man is twice as good as a woman,’ rebutting objections that Hussa bint Al-Murr appeared to be superior to most men by saying that Muslims do not question the Koran and the
The Shaikhly Families 173
Prophet; still, in all social affairs save those prescribed differently by the law, the women of the ruling families are treated and spoken of by most men as superiors. There is one difference between them and men and that is that their status depends as much on the men they are married to or whose mothers they are as upon their birth: if a woman is the mother or the wife of an important shaikh, it makes little difference whether she is by descent a member of the ruling family or not. A great deal of importance is attached to the beauty of a girl if a shaikh is to seek her in marriage, and it is not uncommon for shaikhs to marry the daughters of poor men so long as their families are considered to be of pure Arab descent. Moreover, in Kuwait particularly but also occasionally in other places, the shaikhs take slave women as concubines and the sons of these concubines are brought up, as the law requires, as sons equal to the other children of their fathers. Both the sons of the present Ruler of Kuwait14 are the sons of negresses, whilst of the two most powerful shaikhs in Kuwait other than the Ruler, one, the Ruler’s brother, is the son of a negress and the other, the Ruler’s uncle, is the son of a jariya, that is, a white concubine. Political needs and social ambitions do not, in these countries, override the personal and sexual part of marriage. Nor is the woman’s attitude to the marriage completely disregarded. Many fathers deny that they would consult their daughters at all about whom they wish to be married to, in the case of a first marriage of the girl,15 but on the other hand it is recognized that if a girl is married to a man quite against her will she is likely either to run away from him or else to quarrel with him so much that the marriage is quickly brought to an end, and it is said that no marriage can come to anything ‘unless the wife loves her husband and wants him’: illa idha chan muhibba f i galbha, turiduh. Still, the local proverb, idha baghait walad dawwar lah khal, ‘If you want a son, search out a mother’s brother for him’, well reflects the importance which is attached in the shaikhdoms to the social aspect of marriage. The proverb stresses the desirability of seeking in marriage a girl of respectable family. The particular form in which the proverb is stated is of some interest, for in it the marriage is looked at not from the point of view of the contracting of affinal relationships but, prospectively, from the point of view of the son of the marriage. If we take divorce into account, and divorce is quite frequent, the difference is of some importance. A man can, at will, divorce his wife and have nothing to do with her, whereas he is inescapably connected with his maternal kin, even though his paternal kin should be preferred to them. In the ruling families, it is quite common to attempt to repair rifts by giving girls in marriage. A ruler will often give his daughters in marriage
174
Shaikhdoms of Eastern Arabia
to his principal rivals in his own family. The attempt is often unsuccessful. At the time when Shaikh Sultan bin Salim was Ruler of Ras Al-Khaimah and his nephew the present Ruler, Shaikh Saqr bin Muhammed, was living in the town in narrow circumstances, Sultan bin Salim is said to have called Saqr into his presence and to have offered him his daughter in marriage. He accepted her, but when he went out again he is reported to have said ironically to his followers: a[tani bintu, yaftakir ana bi-anam ba[d, ‘He has given me his daughter: does he think that I shall sleep now?’ Saqr was not pacified and later he expelled his uncle and took his place as ruler. The daughter of Sultan bin Salim remains with him as his senior wife and is the mother of two of his sons. Another daughter of Sultan bin Salim was given in marriage to Saqr’s brother Humaid and is the mother of his eldest son. In Dubai, the daughter of the late Shaikh Mani[ bin Rashid, who was leader of the party opposed to the present ruler Shaikh Sa[id bin Maktum in the Dubai civil war, married the younger son of Shaikh Sa[id. One of the results of this was that a son of Mani[ was allowed to return to Dubai and live there with the Ruler’s family until that it was discovered that he was plotting the murder of Shaikh Sa[id’s elder son, the acting Ruler. This young man was expelled again and now lives at the Buraimi Oasis under the protection of the Abu Dhabi ruling family. His sister remains married to Khalifa bin Sa[id and she is in communication with her brother. In Dubai, the Shaikha Shaikha bint Sa[id, the Ruler’s daughter, married Maktum, one of the sons of the Ruler’s brother Jum[a. This did not prevent the deterioration of relations between the sons of the Ruler and their uncle Jum[a and his sons, and Jum[a and four of his six sons were eventually expelled from Dubai. Maktum, the son of Jum[a who was married to Shaikha bint Sa[id, was allowed to remain, but was shorn of all authority. Marriages between different ruling families, either for political or other purposes, are not common. The wife of Shaikh Rashid bin Sa[id, acting ruler of Dubai, is Latifa bint Hamdan of the Al Bu Falah. Shaikha Latifa’s father was Ruler of Abu Dhabi, and was murdered in 1922 by his brother Sultan, father of the present Ruler of Abu Dhabi.16 His wife went to Dubai with her two daughters and there bore a child to her dead husband. Her daughter Latifa married Rashid bin Sa[id, as we have said. Her other daughter, Miryam, married Shaikh Hazza[ bin Sultan of Abu Dhabi, the son of her father’s murderer (Sultan himself was murdered a few years after he killed his brother). The posthumous son, who was called Hamdan after his dead father, married his mother’s sister’s daughter in Dubai. The eldest daughter of Hamdan bin Zayid, the murdered
The Shaikhly Families 175
shaikh, remained with her husband Rashid whilst Rashid was leading the war of Dubai against Abu Dhabi, where the shaikhs were the maternal kin of his children. We have seen how Shaikh Hamdan of Abu Dhabi was murdered by his brother Sultan. These two brothers were min batn wahid, ‘sons of the same mother’, their mother being Miryam bint Muhammed bin Saif, a distant paternal cousin of their father. In 1927, Shaikh Sultan was murdered by a half-brother, Saqr, whose wife was Saliha the daughter of Ahmed bin Dalmuch, one of the most important pearl merchants of Dubai, a powerful political figure and a member of the Al Bu Falasa, the same tribal section as that of the ruling family of Dubai, into which one of the daughters of Hamdan, murdered by Saqr, married. Saqr himself was murdered one year after he had killed his brother by the son of his eldest brother. This man did not come to rule but brought back the eldest son of Sultan bin Zayid, and he has ruled from that date. Shaikh Muhammed bin Khalifa (bin Zayid) married, among others, Shaikha a daughter of his uncle Saqr whom he murdered and also married Miryam a daughter of his uncle Sultan whom Saqr murdered. The daughter of Saqr bore to Muhammed bin Khalifa a daughter called Hussa, who was married to Zayid bin Sultan, another son of the Sultan murdered by Saqr. Thus Hussa is the wife of her mother’s father’s brother’s son, and mother of his children, and these children are paternal grandchildren of a grandfather murdered thirty years ago by their mother’s mother’s father, who himself was murdered by their mother’s father. When Muhammed bin Khalifa killed his uncle, Saqr, he called in to help him a party of Manasir, and it was they with the assistance of a slave who killed the Ruler. In doing this, Shaikh Muhammed was calling in the assistance of his mother’s tribe and his father’s mother’s tribe. His father was the son of Maitha bint Salimin, a daughter of a shaikh of the Manasir, and his own mother was Shaikha bint Muhammed bin Salimin, which is to say that he was married to his mother’s brother’s daughter, bint khal. The family tree in Figure 5.2 displays the genealogy of the ruling family of Abu Dhabi as it is recorded in Lorimer.17 It is based upon diplomatic records and is therefore to be accepted as reliable, at least since about 1800. I have brought it up to date as far as possible, though some of the men named in it have been, over a long period, expelled from Abu Dhabi and it is difficult to obtain information about them. The ruling family of Abu Dhabi gives us the most extreme example of internal strife among the shaikhs which is to be found in the history of the shaikhdoms, perhaps, indeed, more extreme than the notorious
176
Shaikhdoms of Eastern Arabia
Harib
Dhiyab
Ja[fur
Shakhbut Ahmed Muhammed Sultan Muhammed Khalid Sultan (1922–7)
Khalifa (1833–45)
Muhammed
Shakhbut (1928–66)
Sa[id Sultan
Hazza[ Zayid (1966–) Hamdan (1912–22)
Dhiyab bin [Isa
Saqr (1927–8)
Shakhbut (1793–1816)
Khalifa
Hamdan Rashid Zayid Dhiyab
Sa[id
Sultan
Tahnun(1909–12) Sa[id
Zayid (1855–1909)
Khalifa
Muhammed
Mubarak Tahnun Hamdan Khalifa
Hazza[ Sultan
Sa[dun Hamdan Sa[id
Hilal
Dhiyab
Hilal
Dhiyab
Hilal
Tahnun (1818–33)
Saqr Sa[id (1845–55) Hamdan
Muhammed (1816–18)
Fig. 5.2 The genealogy of the ruling family of Abu Dhabi, as recorded in J. G. Lorimer’s Gazetteer (1915 edition).
history of the Al Rashid dynasty of Ha]il.18 The list of the Al Bu Falah rulers of Abu Dhabi is as follows: Dhiyab bin [Isa was killed by Hazza[ bin Zayid in 1793. Hazza[ bin Zayid ruled 1793–5 and escaped during a revolt. Shakhbut bin Dhiyab then came to power and ruled 1795–1816, when he was deposed by his son Muhammed bin Shakhbut, who ruled 1816–18.
The Shaikhly Families 177
Muhammed bin Shakhbut was expelled by his brother Tahnun bin Shakhbut, who ruled 1818–33, when he was killed by his two brothers Khalifa and Sultan. Khalifa bin Shakhbut ruled 1833–45 with the assistance of his brother Sultan. He was killed by [Isa bin Khalid for an unknown reason after a feast. [Isa bin Khalid was killed by Dhiyab bin [Isa, who himself was killed by Khalid the son of [Isa bin Khalid whom he had killed. This all occurred in 1845. Sa[id bin Tahnun ruled 1845–55, when he was overthrown and succeeded by Zayid bin Khalifa. Zayid, after a remarkably long reign 1855–1908, died peacefully and was succeeded by his son Tahnun bin Zayid, who ruled for four years and then died in 1912. He was followed by his brother Hamdan bin Zayid, who ruled 1912–22 and then was killed by his brother Sultan bin Zayid, who ruled 1922–7 and then was killed by another brother Saqr bin Zayid. Saqr ruled for a year and was killed by the arrangement of his nephew Muhammed bin Khalifa in 1928, and Shakhbut bin Sultan was brought back from Sharjah and has ruled since that date.19 The family tree in Figure 5.3 shows a part of the Al Bu Falah genealogy including the mothers of the descendants of Zayid bin Khalifa (ruled 1855–1908) as far as I was able to ascertain them. The shaikhs of the Al Bu Falah living at present in Abu Dhabi are as follows: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Shakhbut bin Sultan, the Ruler. Sa[id bin Shakhbut ⎫ elder and younger sons of the Ruler. ⎬ Sultan bin Shakhbut ⎭ Hazza[ bin Sultan, brother of the Ruler. Khalid bin Sultan, brother of the Ruler. Muhammed bin Khalid, son of the above. Zayid bin Sultan, brother of the Ruler (Zayid’s son is still a child). Muhammed bin Khalifa, paternal first cousin of the ruler. Hamdan bin Muhammed Mubarak bin Muhammed sons of the above. Tahnun bin Muhammed Khalifa bin Muhammed
⎫ ⎬ ⎭
Other surviving shaikhs of the Al Bu Falah are living in exile. Taking into account the general background of strife and murder in this family, the marriages of the descendants of Zayid bin Khalifa which we have listed, and the shaikhs listed above as being now present in Abu Dhabi, we may now proceed to enquire more closely into the history of the
178
Khalifa
Maitha bint = Latifa bint Sa[id = Zayid* = Woman of Al Bu Falasa = Miryam bint Muhammed = Hussa bint Rashid (Ras Al-Khaimah) (Dubai) bin Saif Salimin bin Tahnum (Al Bu Falah) (Manasir) (Al Bu Falah)
Khalifa = Shaikha bint Muhammed bin Salimin (Manasir)
Tahnun
Saqr = Saliha bint Hazza[ Sa[id Ahmed bin Dalmuch (Dubai)
Dhiyab
Hamdan = Shamsa bint = Shamsa bint Sultan = Salama Muhammed 8 Girls Sultan bint (6 unmarried) Ahmed bin Buti (2 married sons of bin Mujrin Khalifa Dhiyab bin Khalifa) (Qubaisat) (Dubai) (Sudan Tribe)
Muhammed = Daughter of Shakhbut = Daughter of Zayid Zayid
Fakhira
Muhammed = Moza bint Ahmed = Shaikha bint = Miryam bint Sultan Shaikha Zayid Dhiyab Rashid Miryam Latifa San[a = Hamdan Miryam Hazza[ = Miryam bint Shakhbut = Fakhira bint Khalid = Moza bint Zayid = Hussa Miryam Hamdan Dhiyab Muhammed bint bint Saqr bin Khalaf Muhammed Roda (Al (Al [Utaiba) bin Bu Falasa) Khalifa (Dubai)
Hamdan = [A] isha bint Mubarak = Gout bint Tahnun Maitha Hussa Khalifa Hussa Sultan** = Girl of Sudan Shakhbut Shakhbut Tribe
Moza
Tahnun
Sa[id = Maitha bint = Malna bint Khalid Muhammed bin Khalifa
4 Girls
Muhammed = Bint Shakhbut
Wudima
Fig. 5.3 Part of the Al Bu Falah genealogy showing the descendants of Khalifa and his two sons. *“In Lorimer’s genealogy, the mother of Zayid bin Khalifa was of the Sudan Tribe.” **“Sultan’s mother was Hazza[’s daughter, says Shakhbut (i.e. presumably daughter of Hazza[ bin Zayid).” The broken lines represent the movement of women in marriage. Compiled by Peter Lienhardt.
1Girl
Khalifa
Sultan
1Girl
The Shaikhly Families 179
Zayid of Abu Dhabi
Mujrin of Dubai
Maktum of Dubai
Sa[id
Sultan
Hamdan = Shamsa
Latifa = Rashid
Roda
Mudiya = Jum[a
San[a = Hamdan
Mujrin
Thani = Bint Mujrin
Fig. 5.4 Female links between two ruling families of Abu Dhabi and Dubai. Compiled by Peter Lienhardt.
Abu Dhabi shaikhs in recent times. Our account will show the personal nature of the conflicts involved. The murders which occurred in the last generation of rulers are still remembered with pain by the shaikhs of the present. To kill a brother is considered as only slightly less shameful and disastrous than to kill a father. Hamdan bin Zayid who ruled 1912–22 had two wives. One was Shamsa bint Sultan bin Mujrin of Dubai. She was of the same tribal section as the shaikhs of Dubai and had two sisters and a brother. Her family is closely connected with the ruling family of Dubai through several female links, as the family tree in Figure 5.4 shows. Only those of the Al Maktum of Dubai and Al Zayid of Abu Dhabi, who are involved directly in the marriages, are included. Thani bin Jum[a is the son of Jum[a by another wife than the one here mentioned. Shaikh Hamdan of Abu Dhabi had another wife who was also called Shamsa.
180
Shaikhdoms of Eastern Arabia
She was the daughter of Ahmed bin Khalifa of the Sudan tribe. Hamdan had no son until after his death. Sultan bin Zayid, brother of Hamdan, was married to Salama bint Buti. She was the daughter of the shaikh of the Qubaisat section of the Bani Yas. She is still alive and living in Abu Dhabi, and her brother Hamid bin Buti is shaikh of the Qubaisat. The trouble between the brothers Hamdan and Sultan is locally attributed to jealousy between their wives. Hamdan had no sons, and the sons of his brother Sultan are said to have been constantly with him, more indeed than with their own father. Hamdan was a man of extreme generosity, giving away all he had to those who asked for it and therefore extremely poor. Thus he was not able to give his brother the sums of money which the brother and his wife thought he should. They appear to have imagined that he was much richer than he was, because he was generous, and that he was intentionally depriving them of money. When Sultan killed Hamdan quite suddenly and unexpectedly with his own hand, he is said to have immediately opened the treasury where he found nothing, and then to have returned and wept. As we have said, Hamdan’s wife returned with her daughter and her unborn son to her family in Dubai. Her daughter Latifa married Shaikh Rashid bin Sa[id, the Ruler of Dubai’s elder son. One of their children is called Hamdan in memory of his grandfather (it is usual to name children after dead kin). The son of the murdered Hamdan later married his mother’s sister’s daughter and still lives with her in Dubai. The new Ruler of Abu Dhabi, Shaikh Sultan bin Zayid, was murdered in 1927 by his brother Saqr. Saqr was closely connected with Dubai. Both his mother and his wife came of the Al Bu Falasa and were related to the Dubai ruling family, and also to the wife of the dead Hamdan. Saqr is said to have murdered Sultan out of simple ambition. Three of Sultan’s sons (one, Zayid, a child) were away with their mother at Buraimi, whilst Khalid bin Sultan was with his father in Abu Dhabi. Saqr invited Sultan to dinner together with his son Khalid. When Sultan arrived at the door of the house, Saqr went down to meet him and shot him with a revolver as he came up the stairs. When the shot was heard, the sons of Saqr who had been warned beforehand stabbed Sultan’s son Khalid from both sides as he was sitting in the upper room waiting for his father to arrive. Khalid, however, was not killed and after falling down the stairs and being left for dead he was taken away and protected by his mother’s brother, Hamid bin Buti, Shaikh of Qubaisat. Hamid bin Buti also sent a message to Buraimi to warn his sister and her sons there of what had happened and that Saqr had sent out a party of men to intercept and kill the remaining sons of Sultan. The wife and
The Shaikhly Families 181
sons were able to escape and take refuge with the Shaikh of Hamasa village at the Buraimi Oasis and then to make their way to Sharjah where they were in greater safety, and there joined them Khalid, their brother who had been wounded in Abu Dhabi and who was able to escape from his murderous paternal uncle under his maternal uncle’s protection. There remained in Abu Dhabi Shaikh Khalifa bin Zayid and his son Muhammed. Khalifa was older than the other brothers and had been an active war leader in the time of his father Zayid, but had refused to become ruler after his father died and does not appear to have intervened in the trouble between his brothers up to this point. Shaikh Khalifa and his son were both men of importance. The mother of Khalifa was of the shaikhly family of the Manasir tribe and so was the mother of his son. Shaikh Muhammed bin Khalifa began to fear for his own safety as a rival of Saqr and in 1928 he brought in a party of the Manasir, who seized Saqr and held him whilst he was killed by a slave. Men in Abu Dhabi were incensed that their shaikh should be killed by the Manasir, who, though subject to the shaikhs of Abu Dhabi, are not of the Bani Yas tribe and were not, as they said, ahl al-watan, people of the homeland. Because of their shaff, the Abu Dhabi people wished to attack the Manasir. Shaikh Khalifa brought their objections to an end by announcing that it was he who had given the orders for Saqr to be killed. The sons of Sultan were brought back from their retreat in Sharjah and the eldest of them, Shakhbut bin Sultan, became Ruler of Abu Dhabi. The particular group of Manasir who were instrumental in killing Saqr bin Zayid were not connected with Muhammed bin Khalifa more closely than through their tribe. Their leader, known as Bu Halaiba, who is still alive and active, was not a shaikh but became one of the leaders of the Al bi ]Sh-Sha[r section of the Manasir through his own personal qualities. The mothers of Khalifa and his son Muhammed were both from the Al Bu Rahama section of the tribe. It was this same group of Manasir led by Bu Halaiba which was later to help to place the father of the present Ruler of Sharjah in control of that town, and on that occasion they acted for pay. Shakhbut has remained in power in Abu Dhabi from 1928 until 1966. Khalifa bin Zayid is now dead, but his son remains a man of great importance. He appears to have little ambition to rule. The Ruler gives Muhammed bin Khalifa precedence over his own brothers. The three eldest of the sons of Muhammed bin Khalifa are sons of Moza bint Ahmed bin Khalaf of the Al [Utaiba. This family was the richest in Abu Dhabi and they were pearl merchants. Their power was considerable,
182
Shaikhdoms of Eastern Arabia
for they were able to distribute money and provide entertainment and they had a sizeable household. Shakhbut bin Sultan married a girl of the Al [Utaiba but divorced her almost immediately afterwards (it is said that she quarrelled with his mother). The Al [Utaiba succeeded in retaining their position better than most of the pearl merchants when the industry declined. They obtained loans on the security of pearling boats which were to lose their value, but when they provided loans they required date gardens as security, by this means being able to obtain possessions of a considerable area of date gardens at the Buraimi Oasis which had formerly been in the possession of individual pearl fishers. For some reason which I do not know, they became unpopular with the Ruler, and one of them tried to plot with Shaikh Rashid of Dubai that he should seize Buraimi. This member of the Al [Utaiba approached a governor who had been set over part of the Oasis by the Ruler, asking him whether he would agree to let in the people of Dubai. The governor informed the shaikhs and the Al [Utaiba were expelled from Abu Dhabi and their property confiscated. Another wife of Muhammed bin Khalifa is the daughter of Saqr bin Zayid whom he killed, and a third wife is the sister of the Ruler Shakhbut. Two of the sons of Muhammed bin Khalifa by his [Utaiba wife are married to daughters of Shakhbut. One of his daughters by the same wife is married to Sa[id bin Shakhbut, the elder of the Ruler’s two sons, whilst a daughter born to him by the daughter of Saqr is married to Zayid bin Sultan, the Ruler’s brother. Sa[id bin Shakhbut has another wife who is daughter of his father’s brother Khalid. The second wife of Hamdan bin Zayid, who was killed in 1922, was called Shamsa bint Ahmed bin Khalifa of the Sudan tribe. After her husband’s death, she married Hamid bin Buti, the mother’s brother of the Ruler Shakhbut. Two or three years ago, Hamid bin Buti asked his nephew to give him a sum of money to set up business as a contractor and was refused. Thereupon he withdrew to Qatar with his wife and a number of followers; he now lives in Qatar on allowance from the Ruler of Qatar. There are at the present time only twelve adult shaikhs of the ruling family living in Abu Dhabi and of these two are still unmarried, being very young, whilst four of the older shaikhs are brothers. Only three of the shaikhs have daughters old enough to be married, namely Shakhbut bin Sultan, his brother Khalid bin Sultan and his cousin Muhammed bin Khalifa. It will be apparent that the number of possible ways in which these daughters can be married within the ruling family is very limited. By the shari[a, marriage between aunt and nephew or between uncle and niece is forbidden, as is marriage to two sisters at the same
The Shaikhly Families 183
time. Apart from the prohibited degrees of kindred, close kin are preferred in marriage. If we take the marriages of this ruling family, we find first that there is only one brother’s son to whom a daughter of the Ruler can be married, and this marriage has taken place. Two other daughters of the Ruler have been married to sons of the Ruler’s father’s brother’s son. The daughter of the Ruler’s brother is married to the Ruler’s son. The daughters of the Ruler’s father’s brother’s son have been given in marriage to the Ruler’s son, and to two of the Ruler’s brothers. Hence, within this group of close kin all possible marriage ties have been created with the women available outside the prohibited degrees of kinship. If the Ruler were to die, the marriages of his daughters would lead eventually to some distribution of his estate between various lines of his close agnates. From his estate, first of all the fara]id, which are the special portions of the inheritance for particular named dependents, would first be deducted. In the case of his present existing family, there would be two inheritors of fara ]id, his mother who would receive a sixth of the estate and his wife who would receive an eighth of his estate. The mother’s share, on her death, would be inherited by the mother’s children, that is, by the Ruler’s brothers and sister (the sister’s share to be inherited at her death by her children who are of the line of the Ruler’s cousin, Muhammed bin Khalifa). The share of the wife of the Ruler would, at her death, be inherited by her children (her daughters’ shares to be inherited by their children, the share of one of the daughters being therefore inherited in the line of the Ruler’s brother by the children of the brother’s son Muhammed bin Khalifa, and the shares of two of the daughters in their mother’s estate being inherited by their children in the line of Muhammed bin Khalifa). The fara ]id having been deducted from the Ruler’s estate, the remainder is to be inherited by his children, the share of the man being twice the share of the woman. Hence, out of this remainder, the two sons each inherit a quarter whilst the four daughters each inherit an eighth. The shares of the daughters are subsequently to be inherited, at the daughters’ deaths, by the daughters’ children; thus the shares of two of the daughters from their father’s estate will be inherited in the line of the Ruler’s cousin, Muhammed bin Khalifa, by the children they bear to his sons, whilst the share of one of the Ruler’s daughters will be inherited from her by the children she has born to the Ruler’s brother’s son, and therefore in the line of the Ruler’s brother. 20 So long as a man has sons, his brothers do not directly inherit anything from his estate. It is scarcely possible for a ruler to keep the whole business of government in his own hands: he must to some extent rely on the support of the other
184
Shaikhdoms of Eastern Arabia
shaikhs of his own family. There is some significance in the question of which members of his family he must rely on. For purposes of description, we must start an account of the situation at some point in time, but we do so whilst noting that we are, in fact, dealing with a sequence which is constantly proceeding, each new situation in this sequence developing out of a former situation, so that our point of departure is no more than a convenient logical point which we choose in order to show the sequence. In the Eastern Arabian shaikhdoms, there is no system of primogeniture and only a slight tendency towards it. The eldest son is regarded as being senior to his brothers, in the same order of seniority as they are each to the other: age takes precedence. After the death of his father, the brothers will treat the eldest of them as being head of the family; if they are sitting together formally he will have the seat of honour. But if a brother of the dead father is there, then it is he who will, through the seniority of age and generation, take precedence over the eldest son. When a man dies, his sons share equally in his estate. The eldest son does not take the father’s position. He is not head of the group of adult brothers in anything of the same sense in which the dead father was head of the group of sons when he was alive. Thus a family in the shaikhdoms does not have a head in the same sense as applies, or used to apply, in this country when titles and estates were inherited by primogeniture. Members of a family are related to the family as a whole in terms of ancestry and not in terms of their relationship to the living head of the family, the eldest son of the eldest son. The eldest son is, among his brothers, a very distinct case of primus inter pares and no more than that. This is the case both with commoners and with the shaikhs. In the question of who is to succeed to the position of Ruler, the eldest son of the late Ruler has no absolute precedence over his brothers, as he would have in a system where primogeniture was firmly established.21 In the successions to which we have referred in the shaikhdom of Ras Al-Khaimah, Shaikh Muhammed bin Salim was Ruler but, when he was criticized, he resigned in favour of his younger brother, Shaikh Sultan bin Salim. When the latter was driven out by Shaikh Saqr bin Muhammed, Shaikh Muhammed bin Salim, father of Saqr, did not become Ruler again. Shaikh Saqr was not the eldest son of Shaikh Muhammed. Shaikh Saqr offered the rule to his eldest brother, a gesture towards the precedence of age, and the eldest brother refused it. Now, when the Ruler of Ras Al-Khaimah, Shaikh Saqr bin Muhammed, is visited by his father, the
The Shaikhly Families 185
father is given the seat of honour, but his son remains Ruler, giving the father and his brothers small allowances out of state revenues. In the same way, when the Shaikh of Sharjah is visited by his paternal uncle it is the paternal uncle to whom the seat of honour is given. It is, however, the Ruler who rules and both the power and the state revenues are in his hands. The case is much the same with commoners. If a ruler is dealing with an important family, the precedence of honour is given to the eldest member of that family, so that, for example, if two brothers visit a ruler the eldest brother is seated next to him, but the ruler pays practical attention to the most able of the brothers, whether he happens to be the eldest of them or not. Every adult male member of a ruling family is a shaikh. It is significant that there is no really separate title for a ruler to distinguish him from other men of his family who do not rule. The term now used is hakim, but it is little used in conversation and is probably a title used as a translation of the English word ‘ruler’. The term amir is also used sometimes in writing but this has the disadvantage of being in local usage taken to mean an appointed governor (of a village, for example) and not to mean the independent ruler of a state. In Kuwait, the ruler is often referred to in the local language as ash-shaikh al-]ud, the big shaikh, but this is purely colloquial and also slightly vulgar in a pejorative sense. In the Trucial Coast it is common to refer to the ruler by using the word shaikh in the plural and calling the ruler ash-shuyukh, but this again is purely colloquial (the late king was referred to in the same way in Saudi Arabia). People usually speak of the shaikhs, whether of the ruler or of the other members of his family, by using their names. In theory, succession to the position of ruler is elective and any member of the ruling family is qualified by birth to become ruler. Yet there is no system of election and even the matter of acclamation is of minor importance: when Shaikh Mubarak murdered two of his brothers and became Ruler of Kuwait, he is said to have taken the principal seat in his brother’s majlis the next morning, supported by his sons and bedouin whom he had brought in from the desert with him, and to have announced to the Kuwaitis who came into the majlis expecting to find the brother there that his brothers had died the night before and that he had succeeded to the rule, and then to have asked threateningly whether anyone had any comment to make. Those who did object had fled and were not to return. In practice, even when he succeeds to the position peacefully, a shaikh becomes ruler through the strength of his supporters as compared with that of his rivals. There is a practical tendency to primogeniture
186
Shaikhdoms of Eastern Arabia
here because the eldest son of the late ruler, so long as he is an able man, is an easier compromise than anyone else. The closest identification of person and interest – the closest shaff – between men is that which exists between father and son. Next in order of closeness is that between brothers. When a ruler comes to power, he requires the support of a part at least of the ruling family, and those closest to him, his sons if they are adult and then his brothers, are his closest supporters. When a new ruler has no adult sons, his dependence on his brothers is considerable. They help him to overcome rivals, to remain in power and to control the state. There is traditionally no army and no civil service. Each important shaikh has his own retainers whom he leads himself and deploys when necessary in the interests of the state and of the maintenance of order. Any shaikh of the ruling family will intervene in disputes which come about when he is in the vicinity, and he may do so by force. Usually a ruler will allow other members of the ruling family to sit in judgement on cases that occur in places other than the one in which he happens to be or else that he has no time to deal with, though he will usually reserve the most serious cases for himself. In Kuwait, where there is the development of a system of administration, there is a member of the ruling family at the head of every department of the state – Public Security, Police, Courts, Electricity, Public Works, Education, Health, Customs and so on. The only department which is conspicuously lacking in a formal head from among the other members of the ruling family is the Finance Department, where control lies formally with the ruler himself but a good deal of activity is undertaken by his brother. In Abu Dhabi, two of the Ruler’s brothers have been, in the recent times of danger, left in control of the interior. When there was a report that his cousin Dhiyab bin Saqr was planning to come from Saudi Arabia with troops to murder him and take control of Abu Dhabi, the Ruler of Abu Dhabi placed his elder son in charge of the fort which guards access to the island on which the town is built; thus the brothers were placed in charge of the Ruler’s boundaries whilst the son was placed in guard over the Ruler’s life. As time proceeds, the generation of the ruler’s sons and his brothers’ sons grows to maturity. When he has sons who are old enough to be relied on, the ruler is less in need of his brothers, whilst their sons may become something of an embarrassment. They are shaikhs with power and prestige of their own, and through need of money they are apt to take goods without paying for them from the ruler’s subjects and act otherwise in disorderly ways. The ruler’s own sons may be guilty of acting in equally disorderly ways, but in their case the ruler may either
The Shaikhly Families 187
restrain them through his authority as father or else regard their shortcomings indulgently. In the case of his brother’s sons, he is less likely to be indulgent for he cannot remain indifferent to the proliferation of disorder which causes himself to be unpopular among his subjects. On the other hand, it is extremely difficult for him to restrain his brothers’ sons by his own authority. The duty and the right to restrain them rests first and foremost with their fathers. But their fathers, the ruler’s brothers, are probably by this time beginning to resent the way in which they are kept short of money by the ruler and so do not have enough money for themselves and their sons: the income from revenues does not increase at the same rate as does the ruling family. The ruler’s brothers too become critical of the behaviour of the ruler’s sons who are not to be controlled and guided by their uncles more than by their father. The question arises of who is to succeed to the position of the ruler when he dies. If the ruler dies and one of his sons succeeds him, then the process here described goes on over again, with the further complication that there is now a further group of cousins of the new ruler and, as time proceeds, their sons. If a family grows over the generations – and the ruling families have both more opportunity of growing (through marriage of more wives and through leading rather more sheltered lives than the poor) and more need to grow (since the political position of each shaikh is altered by the number of sons whom he can bring to support him) – it must necessarily suffer a diminution in the solidarity of its members as time elapses. With most families, however, the source of income of the family is not a more or less steady one such as is obtained from taxes, nor is that income dependent on political power. If we take the case of a bedouin family of shaikhs, that is, a political family in a full way, when a shaikh’s sons inherit equal shares in his camels then the camels which they inherit in the course of time will increase without respect to which of the sons inherits his father’s primacy. The son who inherits the primacy will only be at an advantage in the division of the proceeds of raids.22 In the case of a ruling family, where most of the income of the family comes from taxation, the case is quite different. In the ruling families, as with other people, the estate of a dead man is divided, with each of his sons receiving an equal share. But this estate cannot include the actual source of income of the ruler, unless the state itself is to be progressively divided over the generations into smaller and smaller groups of subjects. In a sense, it is the ruling family as a whole which rules, in that all its members are shaikhs and any one of them could in theory succeed to the position of ruler, whilst all the members of the
188
Shaikhdoms of Eastern Arabia
family have political responsibilities and authority, deriving this not from the living head of the family but from their ancestors. The state revenues, however, go to the member of the ruling family who has political primacy, and it is he who gives money to the other shaikhs as he thinks fit. They depend on him and so naturally on the influence they are able to exert on him. If we take the position of members of a ruling family in its dynamic aspect, we find that the family remains structurally the same through a repetitive process of internal change. It is not merely over the generations but even in the lifetime of each individual shaikh who is not ruler and not to become ruler, that we find each shaikh’s position tending to deteriorate progressively with the passage of time. The position of the ruler’s brothers is weakened during their lifetime by the coming to maturity of the ruler’s sons, whilst the position of his cousins is weakened by the coming to maturity of the ruler’s sons and his brother’s sons together. It is scarcely possible that in these societies there should be anything but a great deal of rivalry and strife within the ruling families, accompanied usually by a steady process of exclusion of the more distant members of the ruling family and often by open discord and violence, even if we consider the ruling families only in their private aspect as we are doing in this chapter. We shall describe in the next chapter the bearing of this private situation on public affairs. Here we continue the account of the ruling families in their private aspect by turning again to the question of marriage and descent through women. There is a proverbial expression current in Bahrain and the Trucial Coast for the occasion of a man’s divorcing his wife. The man is said to remark: fassakht an [al i, bi-ajib ghairha, ‘I have taken off my sandals and will get some more.’23 Following from this is an expression of the Trucial Coast: an-nasab min ar-riyal (rijl), ‘descent from the foot’. Both expressions devalue the part of the woman in a matrimonial relationship and in her importance to the children of a marriage. The woman is the sandal made for the man’s convenience and to be discarded as and when he chooses. The children of the marriage are his whilst she, the mother, might have been any woman whom the father chose to marry. This is often the way in which men speak of marriage and descent when they are speaking in general terms, and some of them are prepared to speak of their own marriages in the same way. (Most, it should be mentioned, would not undervalue their wives in this way, even when they would undervalue wives in general.) It would, however, be most unusual to come across a father who would regard the marriage of his daughter in the same way. Even though he might insist on his authority
The Shaikhly Families 189
over his daughter and compel her to marry the man he chose for her, even though, which would be more extreme, he was prepared to tolerate her being unhappy in the family into which she had married, still he would regard his daughter’s children as being very closely connected with himself. Much of the same situation applies with regard to a man’s sister’s children. In cases of divorce, the children should, strictly speaking, stay with their mother until about the age of eight years and then be free to choose whether they wish to return to their father. In practice, the children generally go with their mother and, so long as she does not marry again, stay with her, so that if a man divorces the mother of his children he is in danger of losing the children as well as the wife. This applies particularly among the poorer people, for the sons will say that if they do not support their mother and live with her to give her the services which it is difficult for her, in lands where women are segregated, to provide for herself, then she will suffer great hardship. Here there is also a strong tendency for the daughter to go with the father, to look after him and to be eventually married off by him. Two points emerge from this. The first is that there is a considerable divergence between local theory and local practice. The second is that we cannot consider relationships through women without drawing distinctions between a woman as daughter, sister, wife, mother and so on. The first of these points is perhaps the more important, since the division between theory and practice can be noticed in much wider circumstances than these, to the point where it becomes one of the characteristics of the society of the Eastern Arabian shaikhdoms. In this case of descent through woman and in many other cases, we find that men’s general ideas of just principles, which they will expect to be followed by people in general and will feel about most strongly when the neglect of them by others injures themselves, are not in practice regarded as being so plain when the following of them is to their own disadvantage. Here, for instance, a man’s attitude to the rights of his son’s wife’s father in the children of his son’s marriage tends to be quite different from his attitude to his own rights in his daughter’s children. It is in the former case that the importance of the woman is undervalued. The stress which is placed upon the importance of agnation in local theory is not simply greater than is justified by a comparison with practice, it is also a distinct cause of bad relations in the society. The area of overlap between the rights which any man will claim for himself and the rights which he will in theory allow to another is considerable, so that in any conflict between two people each tends to think it just to claim of the other much more than a third, disinterested person would allow either: hence a double area
190
Shaikhdoms of Eastern Arabia
which is filled with resentment. This is perhaps the reason for the great importance attached to peacemaking in the society. The most important distinction to be made between women in their various roles is that between women in their agnatic relationships and women as wives and mothers bearing children to an agnatic group other than their own. In spite of the low valuation set on women in the general local theory, their power is, and is recognized to be, considerable. One of the main causes for the breaking up of marriages is the quarrelling of mother-in-law with daughter-in-law. The daughter-in-law comes to live in a house of which her mother-in-law is mistress, ra [iyat al-bait, and is expected to carry out her domestic duties under her mother-in-law’s instructions and supervision.24 The mother is said to tend to be jealous of a rival to her son’s affections (even in the case of the son’s father) and hostility arises between the mother-in-law and daughter-in-law. Often, when hostility between mother-in-law and daughter-in-law has become serious, the son will take the side of his mother and so come to divorce his wife. But then he marries again, and it is said that in such a situation where the mother-in-law is not prepared to be forbearing there will sooner or later come along a daughter-in-law who is able to overcome the mother’s influence with the son and to take the son away, persuading him to move out of his father’s house and set up house with her alone or with her family. Something of this situation no doubt exists in most societies, based as it is upon a common human weakness, but it seems here to be particularly strong. An example of the distinction between practice and theory is the way in which, in spite of the emphasis on agnation, little girls are given the nickname of umm fulan, mother of so-and-so, using their father’s name and meaning that they are expected to bear a son to be named by the name of their father in memory of him after he has died. Boy’s nicknames are not so but are based usually on some play on their real name, e.g. Mutar becomes Abu Ghaith (both meaning ‘rain’), Saif becomes Abu Hanad (both meaning ‘sword’), whilst Ali becomes Abu Hasan because Hasan was the son of Ali ibn Abi Talib. Marriage is preferred between those who are already related. When a woman marries into her father’s or mother’s kin, or even into a family which has affinal connections with her own, she is already known in the household into which she is to go to live and already has a status there apart from the status which she has as her husband’s wife. She is, in some sense, with her own people as well as being with her husband’s people. Wives, even when they are happily married, will often refuse to go to live abroad with their husbands and away from their own families
The Shaikhly Families 191
because of their greater insecurity when they are separated from their kin in this way, because of the lack of company which they will suffer and also because the custom of the society which segregates women and does not allow them to travel about alone makes it difficult, if they are away, to fulfil their duties to their parents in cases of trouble and illness. Marriage to a man to whom she is clearly related tends then to make for greater harmony between the woman and her husband’s family. As far as the woman is concerned, and this is one of the cares of her own family, she is likely to settle into her married life more easily and happily. As far as the husband’s family is concerned, the woman is less likely to draw her husband away from them. Hence, although we have said that in the case of political marriages between agnates in the shaikhly families the arranging of a marriage to repair rifts which either are already in existence or else are potential is often an unsuccessful remedy for situations which are not, after all, those of marriage, still we must regard inward marriage in the shaikhly families as in others as being a way of confirming ties which already exist and of avoiding the positive disadvantages and dangers to family solidarity which are implied by marriage to strangers. But even in the general category of inward marriage we must make the distinction between marriage of a man to his bint [amm, i.e. his father’s brother’s daughter or other agnate, and his bint khal, i.e. his mother’s brother’s daughter or other agnate of his mother. In both cases the marriages are confirmatory of existing relationships, but it is not, analytically speaking, the same type of relationship which is confirmed in the two cases. Marriage to the bint [amm confirms the agnatic relationship within the whole range of cognatic and affinal relationships, whereas marriage with bint khal confirms and extends the relationship of the man who marries, and of his children to be, with his mother’s agnates. The implications of this situation in the ruling families are obvious. Where members of the group of agnates who are the shaikhs are so liable to be rivals of each other for political and financial reasons, the confirming of the ties with mother’s agnates by smaller groups of fathers, sons and grandsons within the ruling families brings with it potentialities of further strife. If one asks why some family is of particular importance in a state it is considered to be enough of an answer to say that they are akhwal ash-shuyukh, mother’s brothers of the shaikhs. This is not said merely of the actual brothers of a ruler’s mother but also of all the mother’s agnates. There is of course a difference in degree, but it will be said even of a whole tribe that they stand in this position relative to a ruler or an important shaikh.
192
Shaikhdoms of Eastern Arabia
We have seen the causes of the rapid decline of members of a ruling family in importance even in their own lifetimes: the sons of the ruler gradually oust his brothers from their incomes and authority, and the sons of the brothers are in a worse position than their fathers. When these groups of declining shaikhs confirm their relationships with their maternal kin, then the reversals which their fortunes gradually suffer as time goes on affect not only themselves but the maternal kins too. In giving one of his daughters in marriage to a shaikh of the ruling family, the shaikh of a tribe relates himself and his tribe as a whole to that ruling family. At the same time, he relates himself and his tribe more closely to one part of the ruling family, one small agnatic group within it, than to the rest. The relationships of these small agnatic groups with each other within the ruling family involve a great deal of strife, real and potential, because of the steady process of exclusion from wealth and authority of those who are not rulers and their sons. Thus, as far as, say, a tribe is concerned, the two different relationships which are created by the marriage of a daughter of the tribe into the ruling family (particularly where the marriage proves lasting and children are born) are (a) a relationship of the tribe to the ruling family as a whole (which is to be regarded in relation to the politics between the larger agnatic groups within the state) and (b) a relationship of the tribe to a particular group within the ruling family (which is to be regarded in relation to the politics between similar groups within the ruling family). If the marriage connection of a particular small group of shaikhs is to a powerful group among the subjects of the ruling family, where it is important that the relationship of the ruling family as a whole with that group of subjects should be maintained, then the continuance of this small group to confirm its ties with the mother’s agnates tends to buoy up the importance of this small group of shaikhs within the ruling family, reducing the speed of decline; but still the decline continues, and as it goes on then the mothers’ agnates suffer a reduction in the patronage which they are able to obtain from the small group of shaikhs within the ruling family. The mothers’ agnates put pressure upon the small groups to assert themselves more, to seize their rights though by force. Thus, what have been marriages that were politically valuable to a ruler in the past become dangerous for his successors. The ruling family as a whole represents the central pillar of the country to which groups of subjects relate themselves by marriage ties, but this family itself is made up of smaller groups, most of which are constantly moving away from the centre of power and wealth. If we take such groups as tribes as being relatively permanent within the state, there is
The Shaikhly Families 193
in fact no permanent line of authority within the state to which they can relate themselves. The political structure of the states is loose in its nature and almost every detail of it tends to function towards that looseness. We can say of any detail that it is disruptive but this would be to mistake the nature of the whole of which the detail forms a part, for the whole is not a highly integrated unit in spite of the institution of central authority which exists in the form of the ruler. Particular rulers are not secure within their families and their states, and the states are not highly organized, internally united nation-states. Nor is there anything within them up to this present time which could tend to make them so, but rather all seems directed towards preserving a loose and weak organization in which the people can live in great freedom even if without any full measure of security.
6 The Shaikhs and Their People
The following is a story, possibly literary in origin but related in the Trucial Coast, about Dhu ]l-Qarnain, a figure identified with Alexander the Great and referred to in the Koran,1 but one who has become a part of myth rather than history: Dhu ]l-Qarnain, who was willed by God to spend his life as a wanderer, came in his journeys to a town and spent the night there. He remarked to the people of the town: ‘I see that your mosques are built outside the town, away from the houses.’ They replied: ‘That is so in order that we may gain merit by walking out to them at the time of prayer.’ Dhu ]l-Qarnain then said: ‘I see that your graveyard is close to the town.’ The people replied: ‘That is so in order that we may always be mindful of death.’ Then Dhu ]l-Qarnain remarked: ‘I see that your houses here have no doors,’ and the people replied: ‘That is because we have no spite among us and because no one of us looks at another man’s wife.’ Dhu ]l-Qarnain said: ‘I see that there is no rich man among you in the town’, and the people replied: ‘That is because we like to have no poor man and so we divide our money among those who would otherwise be poor.’ Dhu ]l-Qarnain said: ‘I see that you have no shaikhs.’ The people replied: ‘We have no shaikhs because we settle our disputes among ourselves.’ ‘By God,’ said Dhu ]l-Qarnain, ‘had He not obliged me to live for ever a wanderer I would have passed my whole life among you.’ This is a story of an ideal society which inverts much of the structure of the actual world. Each thing that Dhu ]l-Qarnain remarks on is different from what would be found in one of the towns of Eastern Arabia. There, the mosques are in the towns and the graveyards outside them, 194
The Shaikhs and Their People 195
houses have doors and, the better to guard the modesty of the women and the privacy of domestic life, they have no windows on the ground floor save into the interior courtyard, there are extremes of wealth and poverty and the government is in the hands of the shaikhs. The contrast implied in the story is that between life as it might be led in an ideal society where all men lived a life of virtue and life as it is lived among the faults of society as it exists. Various details of life as it exists are seen as being due to imperfections: we find rich men because men are not perfect in alms-giving, we find the segregation of women because men are not perfect in chastity, and, finally, we find the government of shaikhs because men are not perfect in justice. It is the last of these which particularly concerns us here. In considering the relationship which exists between the shaikhs and their subjects, it must first be noted that in almost every respect shaikh and subject are regarded as being similar and comparable. The shaikhs have no religious position that distinguishes them from the rest of mankind. Shaikhs are not, as are some rulers in other societies, the representatives of God, nor do they have any priestly functions. There is no religious ceremony for investing them with their office. They have no ‘divine right’, they are not anointed, they are not regarded as being indispensable for the conduct of the affairs of their people save in the very restricted sense that they reduce the disorder which human imperfection brings about.2 There are no ceremonial emblems such as crown, throne and sceptre that distinguish the ruler from an ordinary man. In recent years, some of the rulers have begun to wear the [agal, head rope, of gold thread on formal occasions, and this too is worn by some of the senior shaikhs other than the rulers in Kuwait and Bahrain, but this is no real emblem of nobility. I have seen the same type of [agal worn as a decoration by old men of rather weak understanding in the Trucial Coast, and there was no feeling that they insulted the shaikhs thereby; these gold [agals are a common sight in Basrah, where they have nothing to do with prerogative. Members of the ruling families dress in exactly the same way as the richer of their subjects. Only in Kuwait is there a slight difference of dress: there, in the outer robe, bisht, a shaikh has an extra strip of gold thread running horizontally from the breast to the hole through which the hand emerges. Among the shaikhs of the bedouin it is notable that a positive effort seems to be made to appear as other members of the tribe. The clothing of bedouin shaikhs is as rough and as dirty as the clothing of their tribesmen, even when they are rich enough to afford better. They eat, too, the same bad food as their tribesmen, with dates
196
Shaikhdoms of Eastern Arabia
of the poorest quality, dry and often filled with insects, and poor, gritty coffee. Such behaviour of bedouin shaikhs can be seen to serve ends which are at once political and economical. They identify themselves in every outward appearance with the people they lead and not, for example, with the town merchants and the shaikhs of the towns, and the avoidance of superior food and clothing leads them out of the difficulty of being obviously rich men from whom their people can expect an endless stream of gifts and entertainment. The manners adopted in addressing a shaikh differ only slightly from those adopted towards commoners. He is addressed more with complimentary phrases than are other men. The most common is Allah yutawwil [umrak, may God lengthen your life, and sometimes a shaikh is addressed by the more humble members of society as [ammi, my paternal uncle, which is a phrase used from the slave to his master, but merchants of a town will often address their ruler by his personal name or by the honorific abu fulan, father of so-and-so, with the name of his son, just as they address each other. Men-at-arms and humbler people, when a shaikh is sitting in his majlis in the open air and they have business with him, will take off their sandals outside the circle immediately in front of the shaikh before approaching him. The taking off of the sandals in the open air when there is no carpeting is a sign of considerable respect, but it still does not differentiate a shaikh entirely from other men: sometimes one sees a man remove his sandals when passing between two people who are sitting talking to each other across a narrow suq. The humbler classes, when they have business with a shaikh, will, if he is sitting, squat before him to tell him of their business. The more important people will be seated at his side. When coffee is served, a shaikh will be given the coffee first, but if there is some guest with him whom he wishes to honour it is not unusual for him to give the coffee to that man. The shaikhs usually expect their guards to come running when they call them. When a shaikh enters or leaves a majlis it is common for all men to stand, but this is commonly done with other people and especially with any sort of superior. Men do not remain standing in the presence of a shaikh when he is sitting. We may say then that a shaikh, whether he be a ruler or a member of a ruling family, has no ritual position and no special ceremonial position. He is the superior of his subjects and their leader. If he is a ruler, his name is mentioned in the Friday prayer at the mosque and he has the right to the religious taxes, zaka (i.e. zakat, but not usually so pronounced), which men are bound to pay (unless it is excused) in order to render their property halal, licit, to them. But he lives and dies a man
The Shaikhs and Their People 197
like the rest of men. When he is buried, his grave is like the grave of any other man, a narrow strip of sand slightly heaped and marked with lumps of stone or with rough pieces of wood for the sake of his close kin who will visit it and pray over it. To pay him special ceremonial respect in life would seem in the society to come near to that most grievous sin of setting up equals to God. The rulers of the Eastern Arabian shaikhdoms are readily accessible to their subjects and to visitors. It is usual for a ruler to hold a majlis each morning and afternoon. In winter the majlis is held indoors. In warm weather it is held in the open air, the ruler sitting on a stone or wooden bench outside his house, usually with some of the other members of his family and men of standing who have come to pay their respects and chat, with a group of his men-at-arms sitting, some in a circle around him and others within call. Men coming in to the majlis first greet the whole majlis and then, when they have taken their place and are sitting, ask the ruler of his health (he does not invariably reply). The majlis is a place of public discussion in a limited sense only. Anyone can visit it (though not everyone does without business to conduct) and people can sit there and talk to each other and intervene in some of the general subjects of discussion on occasion. A high value is set upon modesty, and because of this, conversation between people who are seated next to each other is in subdued tones. Normally men will sit and merely look at what is proceeding and listen to the public parts of the ruler’s conversation. Men who have private business with him will approach him in the majlis and speak to him in low tones, often into his ear, so that it is quite impossible even for one sitting at his side to understand what is being said, and the ruler will often reply in tones equally low. If the ruler calls a man to give him instructions, he will often whisper them into his ear. Hence, though the majlis is public the business of the majlis is easily kept secret. Rulers do not, except in special circumstances, make known what are their plans and proposals and have them discussed by the public, and members of the public do not make suggestions or criticisms in private. Many legal cases are dealt with by the rulers in public, but there is no absolute need for this and decisions are often given privately and carried into effect privately. Especially where dealings are with the important men, privacy is usual. If, for example, one important man brings a complaint against another, the ruler sits and hears the arguments of the two opponents in private and not in public majlis. In Kuwait and Bahrain it is no longer possible for the rulers to be constantly accessible in majlis because of the volume of their work and the requests of one sort and another that they receive. Majlises in these
198
Shaikhdoms of Eastern Arabia
Fig. 6.1 A shaikh in his majlis with his guards. From the author’s photographic collection.
two states are now less frequent and it is not possible for all men to gain admittance. Outside the majlis of any important shaikh in Kuwait one usually finds a number of foreign Arabs and Persians hoping to be admitted so that they can ask for the help and influence of the shaikh. Patronage here has become so important that the older system can no longer be adhered to. The ruler sitting in his majlis is not utterly unlike a merchant sitting in his shop. With the merchant, there are no men-at-arms ready to carry out his orders, but apart from this the two positions are similar. The difference in manners is very slight: it is merely that a merchant will often be sitting at a desk and carrying on with some writing as the conversation proceeds. The custom of visiting is the same for the merchants as for the shaikhs. Men go to salute each other and to talk as
The Shaikhs and Their People 199
well as to conduct business. The merchant sits in an open-fronted shop, carrying on his business and at the same time providing coffee for those who call on him for purely social as well as for business reasons. Where necessary, transactions can be discussed in private within this assembly or elsewhere. Shaikhs passing through the market will often call on their friends among merchants and sit and talk with them, and minor shaikhs, or rulers when they are visiting another state, will often make a custom of sitting regularly in one particular shop where they are accessible to people who may wish to see them. Apart from the ruler’s majlis, everyone tends to be in the suq during the daytime and there is little of sitting at home and entertaining guests there in the towns save at night. Administrators and merchants spend much more of their time in public than happens in Europe, but unlike what we see in Europe there is no strong emphasis on the need for certain sorts of activity, particularly those of judicial and political sorts, to be conducted publicly. We have seen earlier how there is no system of primogeniture in the inheritance of property or in the succession to rule in these shaikhdoms. Now we add that there is no formal way in which rulers are distinguished from other members of their families or from ordinary men. Further, the formal person of a ruler is not distinguished from his personality and actions as an individual. These facts go with the lack of distinction in the society between some kinds of public and private activity: privacy is much insisted on when it is related to women, but such things as justice and politics are conducted, without formal distinction being made, both in private and in public. What is of the family is private and very strictly so. The women are segregated and they are little discussed, and the internal affairs of families are usually successfully concealed. Gossip about them is considered shameful. We have seen how there is a strong feeling that even a killing within a family is considered to be a private affair of that family rather than a public wrong in which the ruler should intervene. But outside the family situation with its intense privacy and secrecy, all business tends to be both semi-public and semi-private at the same time. Save for public prayers, there is little that needs to be done or discussed in a formally public way. These general characteristics of the society have a considerable bearing on the society’s politics. Perhaps the most significant fact is the lack of distinction that can be drawn between the ruler and his office as ruler. The lack of primogeniture, the theoretical election of the best man in the ruling family as ruler, the lack of any special ceremonial attaching to the act of ruling, all of these can be seen to go with the informal and personal nature of rule and the lack of any developed institutions of
200
Shaikhdoms of Eastern Arabia
representation of the people to the ruler, together with the lack of other elements of political life that we could describe as constitutional. The people represent themselves to the ruler; he is accessible to them all and deals with most of them personally, but there are no institutions such as, for example, an independent judicature, whereby the formal person of the ruler can guarantee his subjects against his private vices. There is no kingship whereby the ruler can fully transcend his personal nature and fully transcend also the divisions among the people he governs. In a sense, it is the whole ruling family which rules a state and the whole ruling family which represent the unity of the state. Strife within the ruling family is, like other family matters, largely private, but in sudden eruptions it often expresses not only private strife but strife within the state. Ruling families tend to increase in their numbers and each member of a ruling family is a shaikh. Shaikhs have authority by their own right and not merely authority derived from the head of their family, the ruler. On the other hand, they have no income except from their private activities and from what the ruler chooses to give them from the taxes he receives. They are expected to be generous. They are the most difficult people for the ruler to control. Their particular associates who are under their personal protection cannot be controlled very much more easily. It is said proverbially that the dog of a shaikh is a shaikh: chalb ash-shaikh shaikh. When shaikhs are in need of money it is not uncommon for some of them to use their authority and inaccessible position in order to obtain it. They can buy goods from a merchant and fail to pay him. If the merchant refuses to hand over goods to them it is considered insulting and their men-at-arms may beat him and loot his shop. There are also more subtle ways of obtaining money. Alcohol is formally forbidden in the shaikhdoms. It is smuggled in and it is particularly available in the brothels. A kinsman of one of the ruling families of the Trucial Coast used to send his men-at-arms into the brothels to threaten the prostitutes that they would be taken to judgement for both prostitution and the provision of alcohol unless they surrendered their gold ornaments. The same man would have his men-at-arms kidnap slave women and send them off to the Buraimi Oasis to be sold to Saudi Arabian merchants. Cases, too, of rape and violence by members of the ruling families are not unknown. Even the rulers have not invariably acted in a way which we can consider other than arbitrary. In some of the shaikhdoms of the Trucial Coast, there used to exist an institution known as shufa which was a way for the rulers to extract arbitrary taxes from the merchants. The
The Shaikhs and Their People 201
ruler would announce to a merchant that he intended to pay him a formal visit to honour him on a certain date, and that he would like him to gather certain named friends of his to receive the ruler when he called. When the ruler arrived, he expected not only to be entertained with a meal but also to be given a substantial present of money. Shaikh Mubarak of Kuwait adopted the same custom. His method was to announce to a group of merchants that the state was in need of money and that he required a certain sum from each of them. If they were slow in providing it, he would double the assessment and seize the money by force. Shaikhs, whether they be rulers or not, are not ‘paternal’ figures. They are great men difficult to control or restrain. They have both power and social precedence, and through these they have greater scope for both good and ill than have ordinary men in these societies. They are brought up having their friends among the merchants of the towns and marrying girls from merchant families, but they are also constantly in the company of bedouin men-at-arms, speaking their language and sharing their interests as much as they do those of the merchants. Something of the relationship between shaikhs and their subjects can be gathered from the local story of the luh and the khusaif i. Both of these are sea birds, the luh being a large, black gull which fishes far out to sea, swimming with only its head showing above the water, whilst the khusaif i is a white wader which fishes only in the shallows. I take the story as having an implied political analogy of the luh as the seafaring subject and the khusaif i as the shaikh who stays on land, at once guarding and exploiting his subjects. At the beginning of things, the luhs and the khusaif is all lived together on an island where there were no sandbanks and the shore shelved immediately into deep water. Here, the khusaif i could not fish and so they stayed on the island whilst the luhs went out and brought back fish for them as well as for themselves. The khusaif is were very greedy and the luhs were exhausted with the labour of bringing back fish both for them and for their own chicks. The luhs therefore held a council and came to an agreement with the khusaif is that they would bring them a fixed share of the fish and no more. The agreement continued for some time, but then the khusaif is began to demand more than their share of the fish and threatened that if the luhs did not submit to their demand they would pick out the eyes of the luhs’ chicks whilst the luhs were away fishing. The luhs again made a council to discuss what to do. In number they were twice as many as the khusaif is, but still the khusaif is were more fierce than they so that
202
Shaikhdoms of Eastern Arabia
they dared not fight them. An old luh addressed the council and pointed out how unsuccessful the earlier policy of the luhs had been. He said that if the luhs would follow his instructions he would solve the problem, and this they agreed to do. The old luh said that the luhs must begin by providing as much fish as possible for the khusaif is, and that later he would outwit them. So the luhs set themselves to work particularly hard and brought back a great quantity of fish for the khusaif is, and this they continued to do for a whole week. The khusaif is were surprised at the great amount of fish that was brought to them and asked how it came about. The old luh told them that he and the other luhs had found a wonderful sandbank not very far away where the fish were lying in the water simply waiting to be picked up. He suggested that the khusaif is should come out to the sandbank with the luhs since the only obstacle to providing more fish even than had been brought was the difficulty of carrying it back to the island. He said that if the khusaif is grew tired on the journey the luhs would settle in the water and allow the khusaif is to rest on their backs, whilst on the homeward journey the luhs would actually carry the khusaif is, one luh to one khusaif i. The khusaif is agreed to this proposal, only modifying it with the respect to the homeward journey. About this they said that there must be two luhs to carry each khusaif i. The fact was that they wanted to fill themselves so full of fish that they would be too heavy for one luh to carry one of them. The khusaif i set off with the luhs, leaving five khusaif is behind to be guards on the island. The luhs flew very high, telling the khusaif is that they wished to get a good view of the sandbank where all the fish were. They all flew for some time until they came to a patch of absolutely calm, deep water, and the luhs told the khusaif is that this was the sandbank and that they could now dive straight down and stand on it. The khusaif is dived straight down into the water and were all drowned. Then the luhs returned to the island and told the five remaining khusaif is that their comrades had found so much fish that they had decided to sleep on the sandbank and had sent a message to tell the five to follow them there the next day. The next day four of the five khusaif is went off out to sea with the luhs and were dealt with in the same way as their comrades. The luh discussed whether they should get rid of the fifth khusaif i, but they asked each other: ‘If we kill the last khusaif i, what shall we do at night? We need one khusaif i to keep watch over us, since all that luhs can do at night is to snore and break wind.’ (On the islands at night it is the call of the khusaif i which warns the sleeping luhs of
The Shaikhs and Their People 203
the approach of danger.) So they made an agreement with the last khusaif i that he was to act as guard for them and that they would give him an allowance of fish. The taxes paid to a ruler are considered to be a return for his work and expenses in managing and defending the state. It is said that in the past during wars and raiding when the townsmen owned many camels and were all in possession of their own rifles, and many of them had their own slaves to support them too, the leaders of the tribal sections in the towns would each bring a band of men to fight on the side of the ruler, equipped with camels and rifles. During this century, many Kuwaitis, both merchants and sailors, joined their ruler in defending their town from an attack by the Wahhabis. At the present day, the townsmen have been disarmed whilst the bedouin say that they will only fight for a ruler if they are paid to do so. The spoils of raids were distributed by the rulers among the raiders, the rulers retaining a part for themselves. It is now about ten years since there has been any serious raiding. The situation is changing rapidly, but it is still difficult to know whether to use the present or the past tense for such matters. Formerly, the way of life of the townsmen is said to have been much closer to that of the bedouin than it has become in the last four years. Townsmen owned many camels and went out into the desert with them in the spring. They needed rifles to defend themselves. The bedouin were not at such a disadvantage as they are now in the face of motor transport and the more simple modern weapons such as the Bren gun. The position of the ruling families is now changing even in the Trucial Coast, for they depend now less on their subjects and more on their own retainers and on the support of the Protecting Power. The income from oil royalties and exploration concessions concentrates power in the hands of the rulers. But even in the past, the position of the ruling families was characterized by their power as well as their popular leadership. In the ruling families, power does not only imply authority; it also implies wealth. In relation to raiding in the past and the division of the spoils, we have the ironical proverb: al-[ud li-l-[ud, wa-l-wusti li-l-[ud wa-s-saghir li-l-[ud, wa-l-baj i lak, ‘The great is for the great and the middling is for the great and the small is for the great: what is left is for you.’ Inherited wealth is personal to each shaikh and shaikha. Such and such date gardens, houses, boats, shops or other assets are inherited from the father, mother or other close kin. If the mother is, say, one of few children of a rich merchant, then her inheritance may contribute substantially to the assets of one part of the ruling family. It is customary
204
Shaikhdoms of Eastern Arabia
for the shaikhs to give their daughters in marriage only within their own families, hence there is no corresponding drain on the shaikhs’ assets through inheritance passing to other families through women. Shaikhs, including rulers, and also shaikhas, can acquire wealth through business and investment like other people. They buy and sell town property and date gardens, invest money in trading voyages and smuggling enterprises, go into partnerships in new developments such as cinemas and office buildings, and in general use their capital and sometimes their influence to the best advantage they can. In the traditional system, rulers are in no way debarred from entering into commercial competition in their own states, and it has even been possible for a ruler to insist on a foreign company’s hiring his launches rather than the launches of local merchants, though such behaviour depends on what store a ruler sets by his popularity. Such commercial activities are one side of the acquisition of wealth, in which shaikhs are like other people. But there is another side of acquisition peculiar to the shaikhs alone. This takes the form of confiscation of other people’s property. Every so often a situation arises where, either during a struggle for power among the members of a ruling family or for other reasons, some of the shaikhs or some of the richer people are driven out of the state. When this happens, the property they leave behind them is usually confiscated. It is confiscated by the ruler, but since he may have received indispensable support from some of the other shaikhs and also wishes to consolidate his position, he may distribute some of the property among other members of the ruling family. It should be noted in passing that such a distribution of property can hold awkward implications for the future. If some distinguished members of the public have been driven out and, at later stage, the ruler would like them to return, he knows that he can only persuade them by offering to restore the status quo and give them their property back. He may find that other members of his family, having been given the property for themselves, have a strong interest in opposing his policy of reconciliation. It is common for the merchants to try to avoid political involvement with the shaikhs. As in the story of the town visited by Dhu ]l-Qarnain at the beginning of this chapter, the merchants of the Eastern Arabian towns try hard to settle their disputes among themselves. If two merchants have a disagreement, they prefer if possible to agree on the choice of a third merchant to arbitrate between them. They agree to accept his ruling and take their dispute to him. If one of the parties to the dispute is dissatisfied with the ruling given, he is not legally bound to accept it,
The Shaikhs and Their People 205
but he is under strong social pressure to do so. The arbitrator is unpaid and acts out of goodwill only. To refuse to accept his decision is to dishonour him, and this is added to the dishonour of revoking a given word. In Kuwait particularly, and to a lesser degree in Bahrain, there is a strong solidarity within the merchant community. Kuwait merchants expect each other to carry out contracts which are made only by word of mouth and without witnesses, even when to deny that the contract had been made would be to their immediate financial gain. There is a tradition that a merchant who is in difficulties should be helped by the rest of the merchant community to save him from failure, some merchants giving him money and others not pressing for the payment of their debts. Recent cases of merchants who have been allowed to go bankrupt in Kuwait are said to involve men who were newly set up as merchants and had engaged in sharp practice by refusing to fulfil obligations which they had contracted in the traditional Kuwaiti manner. The importance of mutual assistance and confidence within the merchant communities was of great importance in the past when most of the trade depended on seafaring in small boats which were frequently lost, together with their cargoes, and when there was no other type of insurance. The settling of disputes by arbitration and not by appeal to the shaikhs has several advantages. In the first place it is free of charge. When a dispute involving money is carried to a shaikh, the man who wins the case must pay the shaikh for his trouble. The payment is formally for the guards of the shaikh, who are authorized to compel the defeated party to surrender the debt or property. The charge is invariably ten per cent whether the sum involved is great or small. With the greater sums, it would be most unusual for the guards to receive the whole, though they are given a substantial present out of it, the rest being retained by the shaikh. When a case which involves special technical difficulties is brought to a shaikh to decide, it is quite common for him to send the two disputants to some expert. Where the difficulties are those of the shari[a the case is sent to the qadi and he advises the shaikh on a just decision. Where the difficulties involve the custom of trade or the valuation of goods, the disputants are commonly sent to a merchant, who sends his advice to the shaikh. The shaikh is at liberty to reject the advice if he sees fit. His decision is binding on the disputants and is supported by force. The instruments of force, his menat-arms, profit directly by being used as such: they depend for additions to their wages on asserting force under the authority of the shaikhs. Thus, the retainers are divided from the people of the towns more than
206
Shaikhdoms of Eastern Arabia
are the shaikhs themselves, for in their case the assertion of power is connected with financial profit whereas the shaikhs’ assertion of power is connected with social precedence. The men-at-arms are never able to favour or give honour to the townspeople, they are able only to perform acts which are unpopular with the victims. It is so too with the commoners who are appointed amirs of villages. If, for some fault, they set a man in the stocks, they require a payment from him before he is released, and this payment is their own property. The distinction between the formal and the informal behaviour of shaikhs, the theory and the practice, is of some importance. They are not all alike in character; some are more just and more generous than others. In theory, the shaikh who judges a case is unbiassed and he judges without payment; it is both his duty and his prerogative to sit in judgement. Thus, as far as can be formally managed without kingship he transcends disputes within the society. His use of force does not reduce his honour. It is the instrument of force, the men-at-arms, and not the agent of it, the shaikh, who is paid. In practice, shaikhs do not invariably disregard their personal profit. A further advantage of settling a dispute by arbitration without the intervention of a shaikh is that it is only by avoiding contact with power that a man can be sure of being safe from its abuse. If a shaikh does extract money from a debtor, it is quite possible for him to delay in handing it over to the man who has a right to it, and to keep it for himself. He may also accept bribes. Bribery seems to be more common (it is very difficult to obtain adequate information on such points) among the shaikhs who are not rulers than among the rulers themselves. It is a way in which they can increase their income. Nevertheless, whatever wrongs may be committed by members of ruling families, a ruling family cannot be regarded as being a corporation organized for the sake of holding the people to ransom. They are identified with the communities they govern and form the centres of them. The Eastern Arabian version of the expression that mothers think their ducklings swans is: wuld al-ya[la [and ummu shaikh, ‘The son of the dung-beetle is a shaikh to his mother.’ The ideal existence of shaikh is not attainable by effort: a man must be born a shaikh, he cannot become one, but this does not reduce the part played by the ideal in the popular imagination. It is not uncommon, however, for particularly important men to be referred to by their inferiors with the title shaikh. An example is the three Darwish brothers who are the power behind the throne in Qatar, though they are self-made and of Persian extraction. In the Trucial Coast, boys sometimes play a game of matchbox,
The Shaikhs and Their People 207
throwing it in turn and taking their positions within the group according to the category which each throws. If the matchbox comes down with the side without the label uppermost, the boy who throws this is called ba ]iq, the thief. Similarly, if the side with the label is uppermost the boy is tajir, the merchant, whilst the edge of the matchbox gives shara[, the religious law, and the end, which is most difficult of all to throw, gives shaikh. The point of the game is that the boy who throws shaikh is allowed to slap the one who throws ba]iq, the shara[ recording punishments to come whilst there is no shaikh. Such details, unimportant in themselves, still reflect the very definite place which the shaikhs have in the imaginations of the people. Similarly, if a man is reproached for being domineering the word shaikh will be used about him sarcastically, whilst men who have made themselves great by their wealth, their entertaining and their influence, so that few can stand in their way, are said to be like shaikhs, [abara shuyukh. Though the young shaikhs are not brought up to exercise a great deal of moral restraint – they have enough superiority to be violent even from childhood – yet restraint of manners is firmly instilled into them. From youth, they are expected to show in public the grave demeanour of their seniors. They become accustomed to hearing violent altercations without wishing to intervene in them until the parties are already exhausted and ashamed, and even to listening to complaints about themselves with superior indifference, without the attempts at self-justification which other men of the society tend to make in similar circumstances. On the whole, too, they are brought up to a greater indifference to physical discomfort than are the sons of merchants: the main sport of the shaikhs is hawking in the winter, and this involves going far into the desert for tiring days, rough sleeping and sometimes lack of water. Their main company, apart from their mothers and wives, is their own men-at-arms, who are of a harsher sort than the townsmen and are usually, in fact, bedouin who take harshness and fatigue for granted. Hawking enables the young shaikhs to learn better of the desert and its people, whilst for the older shaikhs it is in part a relaxation from constant cares of exercising authority and in part an opportunity for seeing and influencing the bedouin tribes. The code of manners and the desert pursuits in which the shaikhs are brought up can together be regarded in two ways. From the point of view of the person, the upbringing is of obvious advantage: the code of manners implants in the shaikhs habits of caution and circumspection which as political figures they require, whilst desert pursuits, the company of bedouin and men-at-arms, and indifference to fatigue and hardship
208
Shaikhdoms of Eastern Arabia
are a training for warlike activity. From the point of view of a general structural analysis, we may regard both types of training as involving an important political transcendence. The shaikhs are trained to act as superiors, regarding the excitement and imprudent speech of their inferiors with indulgence or indifference, refusing to be betrayed by excitement into acting as though they were no more than other men and thus abandoning their superiority. By their superiority they transcend the passions of the disputes which it is their business to settle. The training which the shaikhs receive in the desert and amongst its people involves another type of transcendence. In having the training of the desert added to the training of the towns, the shaikhs transcend the differences in the ways of life of the different sections of their subjects. They are no closer to the townsmen than they are to the bedouin; they are not unconsciously more partial to one type of subject than to another. They transcend sectional antipathies. Through this transcendence, the ruling family of a state is able to form the centre around which the people of the state, their subjects, are united. The background of Eastern Arabian society is one of tribalism. Each tribe is a segmentary political group, united as a whole by the loyalty, shaff, which subsists between its members and united in its parts by the closer loyalty, shaff min dun, which subsists between the members of the parts. But we have seen how there is, above the segmentary principle, another principle of impartiality, and that it is this principle which makes it possible for the tribes to settle disputes between themselves without fighting and, when fighting occurs, to make peace with each other again. It is the principle of impartiality too which enables tribes to assimilate strangers. The principle of impartiality is not confined to the shaikhs, but since they are at the head of the tribes, able both to lead and to restrain, impartiality is particularly important in them. In a bedouin tribe, the shaikhs of the tribe as a whole have a dual role. They are leaders of the whole tribe whilst at the same time being involved in the tribe’s internal segmentary structure. Their loyalty with their own tribal section is closer than that with the rest of the tribe. There can be little doubt that it is because of this mixed system – on the one hand segmentary and on the other hand including hereditary leadership and authority – that we find that the principle of shaff, which is here the basis of segmentation, is not quite the same as the principle which guides acephalous segmentary systems. In the Arab system, the shaff of the tribe as a whole exerts a strong pressure towards the peaceful settlement of disputes between groups within the tribe. This is the same pressure as is exerted by the tribal shaikhs. There can be little
The Shaikhs and Their People 209
doubt too that it is the presence of hereditary leaders which makes possible the functioning of impartiality between members of different tribes. In a system where a case of homicide between tribes cannot be peacefully settled, the individual member of a tribe can be said to represent the tribe as a whole in his dealings with other tribes, but where the payment of respect money between the shaikhs of two tribes can dissociate the tribes as wholes from the opposition which a homicide between them brings about, it is the shaikh and he alone who is fully the representative of his tribe. In spite of the very great political freedom of the individual bedouin, he is to some extent under the control of hereditary superiors, and in this sense a shaikh is freer than an individual commoner. Shaikhly leadership among the bedouin modifies the segmentary element in tribal political structure. In the states we find a further modification to this segmentary element. This modification is connected with two main differences which distinguish the town state from the bedouin tribe. The first is a difference in the people, who are much more mixed and fragmented in the town state than in the tribe. The second difference lies in the increase of authority, as distinct from leadership, in the town shaikhs when we compare them with the shaikhs of the bedouin. In the Eastern Arabian shaikhdoms, we can not only see the reasons why the nomadic tribal system should be substantially modified in the town states but we can also observe the historical process whereby the modification comes about. In the desert, the larger groupings of bedouin can be seen to cohere for political reasons rather than for any other. There is little or no economic advantage in the unity of a bedouin tribe, excepting an economic advantage consequent upon a political situation: the bigger and more unified the tribe, the better is it able to defend its own economic assets, its flocks and herds, and to seize the herds of others, and the less it loses by internal insecurity. When there is serious political trouble within a tribe, it is solved by spatial separation; those who are at strife move away from each other into different places. But the towns are not so. The town is an economic concentration, a place of capital and investment. The attraction it offers is that of money, of employment and of economic co-operation. The size of the town depends on economic rather than purely political factors: the town gathers to itself strangers and a mixed population coming to seek prosperity. Townsmen therefore are much less homogeneous than bedouin tribesmen. Whilst being less homogeneous they are also living in much closer contact with each other than are the members of a bedouin tribe.
210
Shaikhdoms of Eastern Arabia
Hence, in the town, through lack of homogeneity in the population, we find a greater intrinsic disunity than in the tribe whilst at the same time, through the much closer and more constant contact of the townsmen with each other, there is a need for greater political unity, for a more determined avoidance of fighting, in the town than there is in the tribe. We can imagine that were tribal loyalty to count for so much within the town as it does in the desert, then any serious quarrel between members of different tribes in the same town would lead to fighting between all the members of the two tribes, and hence the towns would be torn and destroyed by constant civil wars. The difficulty of this situation is overcome by the raising of the power of the shaikhs, so that they become rulers of their people as well as leaders, enforcing a law in which individuals are individually responsible for their acts and where groups are not collectively responsible for the acts of their members, and, moreover, insisting that the use of force should be limited to themselves, the shaikhs, and not permitted to their subjects without their consent. In the desert, in the political unity which constitutes a tribe, the shaikh has only the authority of his prestige and the fighting power of his family. He is the leader of his tribe, not its ruler. In the towns, on the other hand, where the ruling families collect taxes and employ men-at-arms to do their bidding, men-at-arms in personal contract with members of the ruling families, the shaikhs are able to settle disputes by coercion as well as by persuasion, and to punish crimes by fines, beating, imprisonment and execution. They are both leaders and rulers of their people, and it is necessary to their people that they should be so. The right of the shaikhs of the towns, like those of the bedouin, to occupy their position of leadership is based on their traditional primacy. In the case of the bedouin shaikhs, their primacy is primacy in an ethnic group, the tribe, conceived to be united through a common apical ancestor, even though it is recognized that many groups within the tribe are descended originally from strangers. There is at least a core of tribesmen sharing the same agnatic ancestry, and the accretions of strangers are incorporated around this core. In the towns, the primacy of the shaikhs is not so much primacy in an ethnic group as primacy in a fixed place: the town itself. Sometimes the town shaikhs have tribesmen of their own within their territories, but this is not invariably so, and indeed we shall argue that the tendency after settlement of a tribal group under its own shaikhs is towards the elimination of the group with which the shaikhs share a common ancestry, the greater authority of the shaikhs when they have settled and the greater mixture of population which
The Shaikhs and Their People 211
they come to control resulting in a process whereby all who are not shaikhs become equally subjects. In none of the Eastern Arabian shaikhdoms is the ruling family believed to have had its origins in the place which it now rules. Each ruling family is traditionally associated with the state which it rules, but the association is believed to have had a beginning in time and, usually, in comparatively recent time. Originally the ruling families are thought to have shared the wandering tradition of their subjects, and there is no reason why we should suppose that this is not in fact the case. In the only two shaikhdoms where we have reliable historical information about the advent of the ruling families in the first place, that is in Bahrain and Dubai, the ruling families did in fact come from other places and settle together with their followers in the states which they now rule. They began as leaders and became rulers. The traditional history of the other shaikhdoms tells the same story. There is no local explanation of how it has come about that the shaikhs, beginning as leaders of ethnic groups and evolving into rulers of places and the people who live in them, should have come to acquire particular rights in the places, in what roughly speaking we should call real property, in the process of the change. It seems likely that the shaikhs hold these rights as representatives of the communities they govern, noting that these settled communities are no longer conceived of as being ethnic unities. The rights themselves are in water and in land. The shaikhs who rule the states recognize no private rights to water excepting in the case of water that is used for irrigation. Wells in the desert and in the towns are the property of the shaikhs. Only the shaikhs have the right to exclude people from the use of them. The land too is the property of the shaikhs except when it has been acquired from them. In the past it is said to have been customary, when a man came to settle in a town, for the shaikhs of the town to grant him a piece of land on which to build a house and to help him with a gift of materials towards the building of it. This still happens at the present time, but it is not the invariable rule. Conversely, in the past and in the Trucial Coast until very recently, it was not possible for a man to own, except on sufferance, property in a state without living in the state. Membership of the community was a prerequisite of property ownership. If a man left a state and went to live in another place, then his house and the land it was built on and his date garden if he had one were forfeit to the shaikhs, should they choose to take them. The right of the shaikhs to confiscate the property of those who left the state was looked upon as complementary to their duty to grant property to newcomers. The state boundaries are the boundaries of the ruling families and the
212
Shaikhdoms of Eastern Arabia
land is ultimately theirs. This applies to the desert as well as to the towns. No man may assert possession over a patch of unused land by building on it. If he wishes to build a house, he must either buy the land or else ask for it as a free grant. He buys the land from the ruler and the ruler gives him a receipt for the price paid which serves as a title deed to the land and which the owner carefully preserves. Having the title deed of the land, he is at liberty to sell it, just as another man is at liberty to buy, without reference to the ruler. These title deeds are kept with care, for without them it is difficult to assert ownership, particularly if the ground has not been built on. They are not readily entrusted to other people, even for inspection, in case they should be destroyed. A further example of the arbitrary behaviour of some of the shaikhs is the following. A man owned a strip of unused land in one of the Trucial shaikhdoms. A member of the ruling family wished to build on it and asserted that it was unused land and therefore belonged to the ruling family. The shaikh asked to see the title deed which the owner of the land asserted that he had. The owner said he had it but could not lay his hands on it. He told me that he was intentionally keeping it out of the hands of the shaikh because he thought that if he brought it out the shaikh would not dare to build for fear of the deed’s being brought out publicly before the ruler to prove that he did not own the land he had built on. The sanction against the shaikh here, as so frequently, was really of public opprobrium rather than law. The commoner was defending himself by his wits, making sure that any wrong which was meditated would have to be committed publicly. It is the ruling family which has the traditional primacy in the state. The head of the ruling family is the ruler of the state, so that at any time the rights of the ruling family as a whole are vested in one particular member. This member who is to be head of the family and of the state is, in theory, ‘elected’: he has no absolute right to the succession; indeed, we cannot accurately use the word ‘succession’ in this context. It is required that he should be satisfactory to the people as well as satisfactory to his family, for he is to head both. The position of the ruling family relative to the ruler is not defined for the very good reason that the members of the ruling family, like the members of other families, are regarded as owing unlimited moral duties to each other, even though everyone knows that in practice men do not live up to these duties. We have seen that there is almost invariably a great deal of dissension within the ruling families, and we have discussed some of the private causes which bring this about. We can now continue by discussing the bearing of this dissension on the political life of the subjects.
The Shaikhs and Their People 213
There is always a good deal of speculation as to who is to succeed a ruler when he dies, so that public opinion on the subject is largely formed before the event. In theory, a shaikh becomes ruler of a state by ‘election’, but there is no formal system whereby the election is conducted. There is no question of votes: men are not equated to each other. The importance attached to a man’s opinion on the subject depends on the importance of the man. Naturally, therefore, the members of the ruling family are particularly important in the making of the choice. Even here, some are more important than others. The followings of different parties have been created before the ruler dies. A consensus is aimed at, and practically speaking it is unlikely that there should be more than two candidates. This is because the consensus is most easily reached by choosing the eldest son of the late ruler. To this extent, there is a practical tendency towards primogeniture. This shaikh, however, has no absolute right to succeed his father. If he is considered unsuitable for some reason such as youth or extreme unsuitablity of character, then another shaikh can be chosen. But the choosing becomes a matter of factions within the ruling family. The way in which the factions are formed will appear in examples. Setting aside personal choices, which cannot of course be disregarded in any particular situation, the general tendency of the building up of the factions is on the line of closeness in descent. Those who are closer in descent to one of the rivals than they are to the other will support him. If, for example, the rival of the late ruler’s son is the late ruler’s brother, then the later ruler’s son is supported by his brothers whilst the late ruler’s brother is supported by his sons. Those who are equidistant from both rivals may support either, but they will tend to act together and not to divide among themselves over the question. The general principle is illustrated in Figure 6.2. If the strongest rival of C is D, then sons of the late ruler support C and the sons of D support D, whilst the descendants of B act together in support of either C or D. If the strongest rival of C is E, then all the descendants of A support C whilst all the descendants of B support E. This is to say that each descent group within the ruling family tends to try to bring the rule as near to itself as possible. In Kuwait, where on the whole the shaikhs have succeeded in living much more peacefully together than have the shaikhs of other states, an agreement for alternate succession has worked over two generations. When Shaikh Mubarak died in 1915, he was succeeded by his son Jabir Al-Mubarak. Jabir ruled for only two years. When he died it was agreed that he should be succeeded by his brother Salim Al-Mubarak, on condition that Salim should be succeeded by a son of Jabir. Salim was in
214
Shaikhdoms of Eastern Arabia
B+
A+
late ruler +
D
E
F
C Fig. 6.2
The formation of factions in the choice of a ruler.
fact succeeded by Ahmed Al-Jabir (son of Jabir). Ahmed Al-Jabir was succeeded by [Abdullah Al-Salim (son of Salim). The second important body of men who choose the ruler are the leading men from the subjects: the merchants and the heads of tribal sections. These cannot themselves rule, nor have they any reason for acting together save for reasons of public policy. The major candidates are already made by the ruling family, and these secondary electors can support either the one or the other, according to which they prefer. It is by this means that internal rivalry within the ruling family can be made a vehicle for much wider questions, questions not merely of personal advantage but of public policy. Except possibly at the present time in Bahrain, where the British Government has brought into being a new system as a result of what Aristotle called ‘insignificant change’, 3 disputes within the Eastern Arabian shaikhdoms have always been expressed through the internal discord of the ruling families. The lack of solidarity between members of the ruling families has had two effects. It has made the shaikhs subject to popular pressure, since an unpopular ruler would find his supporters drifting away and following a rival (there is not only the question of succession, but also that of
The Shaikhs and Their People 215
the coup d’état). It has also protected the principle of shaikhly rule, since it is much easier to support a rival shaikh from the ruling family against the ruler than to seek to expel the whole of the ruling family. A shaikh can ususally protect his supporters, so long as his rivalry with the ruler has not reached a pitch of open violence. The complex situation of public and private issues in the struggle for power in the state is best shown in some episodes of the present century in the history of Dubai. This is in some ways the most interesting of all the Eastern Arabian shaikhdoms since we know of its history from its very beginning as an independent state. In the history of Dubai we may see enacted the process of change in the ruling family from leadership to authority, the process of change whereby the main tribal groups in the town were reduced in political power and cohesion, to become a little more powerful than other subjects, the reflection of disputes over public policy in the internal struggles of the ruling family and the progressive reduction of the ruling family itself, keeping pace with its growth, through exile and flight. The following is a brief account of the history of Dubai up to the time of the publication of Lorimer’s Gazetteer. Dubai existed in 1799. Its chief, Hazza[ bin Za[al, then a minor, signed the 1820 General Treaty of Peace, but Dubai seems to have been a dependency of Abu Dhabi until 1833. At that date, eight hundred of the Al Bu Falasa section of the Bani Yas seceded from their home in Abu Dhabi after being worsted in a struggle for power there and they set up an independent state at Dubai. Their leaders were [Ubaid bin Sa[id and Maktum bin Buti. [Ubaid, the senior of the two, died in 1836, after which Maktum was sole Ruler. Lorimer says of him ‘ . . . it is probable that peril from outside (i.e. from Abu Dhabi and Sharjah) was not without advantage in strengthening his authority at home’. He was attacked by Abu Dhabi and Sharjah in conjunction but defended himself successfully. He was an active war leader, losing an eye in 1843 whilst fighting the Bani Ghafala. He died in 1852 of smallpox at sea whilst returning from a visit to his ally the Sultan of Muscat. He was succeeded by his brother, Sa[id bin Buti. Sa[id paid a visit to Muscat in the year of his succession, whereupon his late brother’s sons took the opportunity provided by his absence and seized the fort of Dubai. They were, however, expelled by Sa[id bin Rashid, the mother’s brother of Sa[id, who had been left in charge of the town in the ruler’s absence. They sought asylum at Sharjah. According to Lorimer, the ruler of Sharjah also ‘persuaded’ a group of the Al Bu Muhair tribe to leave Dubai and settle at Sharjah at the same time. It seems likely that these were supporters of the worsted shaikhs. In 1852, Sa[id bin Buti was
216
Shaikhdoms of Eastern Arabia
able to ally himself to Abu Dhabi and Umm Al-Qaiwain and so nullify intrigues on the part of the shaikh of Sharjah. Sa[id bin Buti and his surviving brother Buti both died without issue, of smallpox, in 1859. Hashar bin Maktum, son of the penultimate Ruler, returned from Sharjah and became Ruler. When Hashar died in 1886, the succession was claimed by his brother Rashid bin Maktum and by his son Maktum bin Hashar, who according to Lorimer were supported by followers equal in number and power. The ‘elders of the tribe’ peacefully settled the succession on Rashid, who ruled until 1894. On his death, his former rival Maktum bin Hashar succeeded him, imprisoning for five months the sons of his predecessor who disputed the succession, after which they went to Sharjah. Maktum’s sons were very young when he died in 1906, and his cousin Buti became Ruler. Buti died in 1912, to be succeeded by the present Ruler, Sa[id bin Maktum.4 The significant characteristics of the foregoing history can be briefly indicated. The history shows how rapid can be the change from leadership to rule. There can be no doubt that in the first place the eight hundred discontented Al Bu Falasa were led to Dubai from Abu Dhabi and not constrained to go there. In an earlier chapter we have given a number of examples of this old characteristics of tribal movement in the shaikhdoms. This ethnic group, the Al Bu Falasa, united in its own loyalty, had supported the losing side in a struggle for power within the ruling family of Abu Dhabi, and some of its members appear to have been punished. They left Abu Dhabi to form an independent ethnic state by themselves where they could live without the troubles caused by being ruled by outsiders: in the first stage the tribal ideal is dominant. The move of this tribal section to the small fishing village of Dubai took place in 1833. The first struggle for power took place only nineteen years later, on the death of the original Ruler who had led the people from Abu Dhabi. It was a struggle between late ruler’s brother and late ruler’s sons. The losers in the struggle fled to Sharjah, taking some of their supporters with them. (These were of the Al Bu Muhair, a tribe which now has a traditional association of intermarriage with the Al Bu Falasa. It is not known when or from where the Al Bu Muhair came to Dubai in the first place.) The pattern of struggle between ruler’s sons and late brother repeats itself in 1886 and in 1894. Hence, after only nineteen years of government on the tribal principle, the same struggles begin to show themselves in Dubai as caused the original departure of the Al Bu Falasa from Abu Dhabi. We may add here that the same eruptions clearly took place in Kuwait, where at first the Al Sabah, the Al Khalifa and the Jalahima are said to have settled the town together,
The Shaikhs and Their People 217
then the Al Khalifa and the Jalahima both left Kuwait, and later there was a split between the Al Khalifa and the Jalahima. By the beginning of the present century, Dubai had grown from being a fishing village to being a sizeable town. It had attracted a large population of foreigners, mainly Persians and Indians, and of Arabs from other parts of Trucial Oman. The earlier period of power of the present Ruler, Shaikh Sa[id bin Maktum, was peaceful, but it did not remain so, for during this period the pearling industry, the means of livelihood both of the Arabs of the higher classes and of a great proportion of the poorer subjects, collapsed, and at the same time an urban, modernizing movement began to grow. The decline of the pearling industry, which began about 1926 and had reached a stage of ruin in the early 1930s, brought about a great deal of hardship and a great change in the life of the people. The pearl merchants were slow in adapting themselves to the situation. They seem at first to have imagined that they were merely suffering a temporary setback in the markets of the world. They continued to invest in the industry through the loan system, but their pearl fishers were never to repay the loans. The great pearling fortunes were thus, in effect, spent in cushioning the poorer people from the full effects of what was happening. Very few indeed of the families of pearl merchants and boat owners who financed the pearling industry thirty years ago5 are any longer to be counted even among the prosperous families of the Gulf at the present time. The stocks of pearls which they kept lost their value, the debts owing to them could not be recovered and there was the further, social cause that they began by despising any trade or employment other then pearling as being ignoble. When they were forced to recognize that they must seek other sources of income, they found the competition with those already established in other sorts of trade was not easy. Trade in general was reduced, the pearling families lacked the necessary contacts and experience and they were moreover weighed down by the number of their dependents. Their slaves resented the reduction in their standard of living. Some were sold, some were emancipated and some even sold themselves and gave the money to their masters whilst bidding them goodbye. Men’s honour required them to support their followers and to avoid performing servile tasks. Men who tried to set up in trade found themselves frustrated by the applications of their former associates in the pearling industry for endless credit which it was socially difficult to refuse. Resentment and jealousy were felt at the continuing prosperity of the Indian and Persian merchants and traders who, even when they had engaged in the pearl trade, had
218
Shaikhdoms of Eastern Arabia
never committed themselves to investment and loans in the same way as the Arabs and whose ideas of honour and prestige were rather different. The Indians, who were still part of the British Empire, were well enough protected politically, and so action was directed at the Persians. The ruling family of Dubai was made up of the sons and grandsons of three former rulers. The Ruler, Shaikh Sa[id bin Maktum, was one of the three sons of Maktum (ruled 1894–1906). The two other dead rulers whose sons were still alive were Rashid (ruled 1886–94) and Buti (ruled 1906–12). The other rival of the Ruler Sa[id was Shaikh Mani[ bin Rashid. Mani[ was a pearl merchant who had travelled to India and had friends in Kuwait and Bahrain. He was a modernizer. He had his daughter taught to read and write – a most uncommon thing for women in general and an accomplishment lacked by any of the other shaikhly ladies of Dubai even now – and he himself made a map of the pearling banks of the Gulf and had it printed. Mani[ became leader of the reformist movement in Dubai. His contact with Kuwait was not unimportant in this connection since much the same thing was happening there. The party led by Mani[ included all the sons and grandsons of the late Shaikh Rashid bin Buti, and a considerable number of the ex-pearl merchants and boat owners, including most of the Al Bu Falasa, the tribal section which includes the Dubai shaikhs, who remained in Dubai. The demands of the reformist party were as follows: 1. That there should be brought into being a consultative and legislative council, the majlis. 2. That this council should have control of the state revenues, allowing the ruler one quarter for his personal expenses. 3. That a formal court of law, mahkama, should be set up. 4. That a school of modern type should be started. 5. That considerable restrictions should be put on the trading of the Persians in Dubai by increasing the customs duties they paid and limiting the length of time they should spend in the town. The Ruler Shaikh Sa[id, his brothers and their sons were opposed to the majlis party. They claimed that the financial and judicial power of the ruler was being unwarrantably curbed. They had considerable support in the town of Dubai from some of the Arabs there, and many of the Persians were ready to lend them money. Apart from the Persians whose own interests were seriously threatened by Mani[ and his party and therefore inclined to taking financial risks in support of Sa[id, and the shaikhs who divided according to the genealogical pattern which
The Shaikhs and Their People 219
we have described, the other parts of the Dubai community did not divide on any lines which we can strictly define. The greater men chose according to their personal preference or prejudice or what they thought to be their interests, taking their followers with them, whilst many people remained as spectators in the struggle. The struggle quite clearly had a double aspect. On the one side it was concerned with reform and improvement, and we have no reason to doubt that the intention of the party in this direction was quite honest. But on the other hand there remained the aspect of ambition for financial profit and increase of power on the part of the ruling family that was further from the ruler. Motives therefore were considerably mixed. The struggle came to open civil war in 1939. Dubai is on two sides of a creek, and Shaikh Sa[id ruled the one side whilst Shaikh Mani[ controlled the other. Many bedouin, not all from Dubai territory, were attracted to the side of Sa[id by his generosity. Sa[id pawned the customs to a Persian merchant for ready cash and was reduced to such dire straits at one point that he had to sell some of his household goods. Sporadic fighting continued for some months, and then a treaty was made between the opponents under the auspices of the Political Resident in the Persian Gulf. Shaikh Sa[id acceded to the demands of the reformers and attempted to resign in favour of Mani[. Administrative offices were set up in the latter’s side of the town. Then, in 1940, Rashid bin Sa[id, the elder son of the Ruler, arranged a marriage in Mani[’s side of the town and began to visit the family of his prospective bride daily, taking guards with him, and when the suspicions of Mani[ had been lulled by time Rashid told his guards to launch a surprise attack on the opposition. The attack was made and Mani[’s party were taken unaware and defeated. A brother of Mani[ was killed on the jetty, whilst some of the members of the majlis party who were captured were blinded with kabab sticks.6 Mani[ supporters fled from the town. Their property was seized by the brothers of Sa[id. Information about the Kuwait majlis which was put down by force in 1939 is difficult to obtain. The demands of the majlis were exactly the same as those of the Dubai majlis, except in the matter of education which had already begun in Kuwait. The leader of the majlis party (and this is no doubt one of the reasons for the difficulty of inquiry) was the present Ruler of Kuwait. His rival was the then Ruler, Shaikh Ahmed Al-Jabir, the relevant part of the genealogy being as shown in Figure 6.3. The Ruler of Kuwait acceded to the demands of the majlis party, and a legislative and consultative council was set up in June 1938. This body
220
Shaikhdoms of Eastern Arabia
Mubarak (Ruler 1896–1915)
Jabir
Salim
(Ruler 1915–17)
(Ruler 1917–21)
Ahmed
[Abdullah
(Ruler 1921–50)
(present Ruler)*
Fig. 6.3 Rivalry within the ruling family of Kuwait. *The word ‘present’ refers to the time when Peter Lienhardt carried out his fieldwork and wrote up the findings of his research. [Ed.]
was able to set up a court and to reform the traditional law of the sea, giving slightly better conditions to the crews of trading and pearling boats. A number of Persians were deported. But the council itself divided into one faction favouring union with Iraq and the other the continuation of shaikhly government. The former, young Kuwaiti, faction gained control. Thus what had began as a domestic dispute expressed through shaikhly rivalry was carried beyond the shaikhs and the system they represented. The majlis was dissolved in December 1938 and new elections were held. The new majlis drafted a constitution which involved the loosening of ties with Britain and the taking of control over the state revenues out of the hands of the ruler. Propaganda against the shaikhs was broadcast from Iraq and, according to Col. Dickson, the German Government was also active. 7 Oil had recently been discovered. The majlis was forcibly suppressed, with slight loss of life, by the Ruler and his supporters from the shaikhs in March 1939, and a new majlis was set up which had only advisory powers and was ineffective. Shaikh [Abdullah Al-Salim, who had possibly withdrawn his support before the end, left the country for a while but remained powerful, and there appears to have been very little victimization of the pro-majlis party. The state secretary was retired and replaced by his son. Prominent merchants were forgiven.
The Shaikhs and Their People 221
In Dubai, after the defeat of Mani[, there remained of the ruling family the Ruler, his brothers and Hashar and their sons. Sa[id had two sons, Jum[a six and Hashar one. A great deal of power remained with the brothers of the Ruler and particularly with Jum[a. He and his sons exercised most of the control in the part of the town which had been under the control of Mani[, and his sons and their men-at-arms held the forts there. Sa[id attempted to restore good relations with the defeated party, his younger son marrying Mani[’s daughter, but there was no reconciliation. The situation when I first arrived in Dubai in 1954 was complex. Sa[id was an old man and had given most of the business of governing to his elder son Rashid, but in a sense all the shaikhs shared in the government, and even those who were quite young would readily give judgments in the street without consulting the ruler. Jum[a and his sons formed the majority of the shaikhs, and so Jum[a was the main rival of Rashid. Rashid tried to please Jum[a and his sons by giving them money to such a point that his brother Khalifa was jealous. The wives of Rashid and his brother Khalifa did not visit each other. When Jum[a or one of his sons wronged a man, it was common for Khalifa to take action to recover his rights. We see here the way in which the ruler tries to hold the ruling family together by putting more emphasis on the more distant and less reliable ties, by this very policy increasing the enmity between those who are close to him and those who are distant whilst he is trying to retain the support and favour of the distant ones. Rashid was anxious to settle the trouble he had had over his treachery in the civil war and bring home some of the exiles, but in this he was strongly opposed by Jum[a and his sons who had taken their property. Rashid, after some apparent doubts, became anxious to co-operate with the British, whilst Jum[a was opposed to the British and had some connections with Saudi Arabia during the Buraimi dispute. In 1955, Rashid succeeded in expelling Jum[a and most of his sons from Dubai, as has been related in an earlier chapter. When the Al Bu Falasa first came to Dubai a little over a hundred years ago, there were eight hundred of them led there by their own shaikhs. Now, Dubai has grown to be by far the biggest town in the Trucial Coast with a population of thirty thousand or more. The town is still ruled by the shaikhs of the Al Bu Falasa, but of that tribal section itself there remain only a few families in the town, numbering in all a mere thirty or forty souls, a number quite negligible in proportion to the original settlers. This example illustrates the political change which results from the establishment of a state. In the Eastern Arabian shaikhdoms, the state is a political unit of a very simple kind. The states are
222
Shaikhdoms of Eastern Arabia
small in area and in population. With the exception of Bahrain, which has two towns, Muharraq and Manama, which are very close together, each state has one town only, and we have good reason for supposing that this common characteristic is by no means accidental but, on the contrary, is a result partly of ecological conditions and partly of the political structure, two things which are by no means unconnected. The history of what was originally the joint Ras Al-Khaimah and Sharjah shaikhdoms of the Jawasim suggests the way in which the historical process develops, for here, when a second town8 had grown up in the state, the rivalries within the ruling family found their outlet in the assertion of independence of one part of the ruling family in one of the towns. The ruling family divided, according to the pattern we have described, and the result was the division of the state into two states, each headed by its town. After this had occurred, the process of divisions within the ruling families of each of the towns continued as in the other shaikhdoms. The smallness of scale of these town states results in a very simple and direct administrative system. There is no elaboration of offices and functions, and even taxation does not require the development of a civil service. For warfare, the states rely (with the exception of presentday Kuwait) on citizen armies and supporting bedouin tribes, and these are controlled by the shaikhs of the towns and the lesser shaikhs of the bedouin tribes and sections; hence there is no elaboration of a special military organization. The men-at-arms of the shaikhs take part in the wars but they are employed mainly for internal security and for garrisoning the towers from which the towns and villages are defended. These men are under the direct command of the town shaikhs. The number of subjects is not so great that they cannot have access to their rulers. Thus we are dealing with states on a small scale where administration is very slight and where rule can remain simple, direct and personal. There is another aspect too to the simplicity of the Eastern Arabian shaikhdoms. This is the aspect which we have described in an earlier chapter in connection with the very slight degree to which the subjects rely on living in one particular spot to make their living. Very few of the people depend traditionally for their livelihood on landed property: their assets for gaining an income are mobile, with the result that the subjects too are mobile, able to leave a state which dissatisfies them in some way and go to live in another. This general orientation of the society towards personal independence, refusal to tolerate the excessive use of authority and readiness to move away from oppression affects
The Shaikhs and Their People 223
even those who cannot move without considerable financial loss – those who own, for example, date gardens, permanent houses in the towns, and valuable building land in the towns. Because of the readiness of the settled people to remove themselves from one place to another, and the ease with which the movement is usually accomplished, we may describe the political type of the ‘settled’ community in the shaikhdoms to be only minimally different from the political type represented by the bedouin. The bedouin, either in groups or singly, can move from one area to another and from one jurisdiction to another, and they can, among themselves, separate. Hence they are able to maintain a considerable measure of independence wherever they live, both from governments and from each other when they so wish. In these respects, the bedouin differ very considerably indeed from a landed peasantry, since short of disaster or a complete change in their way of life peasants are bound economically to their land and therefore bound politically to place. The difference between the bedouin and the settled people of the shaikhdoms is very much less. These settled people of the shaikhdoms are ‘settled’ only in the sense of not being nomads and of living in permanent settlements. They are not settled in the sense of being, like a peasantry, fully and permanently bound to particular settlements. In this respect, therefore, we may regard the town polities of the shaikhdoms as representing the differences which are implied in the changeover from nomadic life to the mere living in permanent settlements, without the added complications which follow from a necessary attachment to particular places. Even these minimal differences lead to considerable changes in the polity. Local traditions and, in the case of Bahrain and Dubai, more firmly established historical fact, lead us to suppose that the towns began with the arrival by land or sea of wandering tribal groups under tribal leadership. They have all developed on much the same pattern. The tribal section closest to the shaikhs has disappeared or almost disappeared, and sometimes the shaikhs retain no one in their territory with whom they have agnatic connections. Almost invariably, the ruling families have constantly excluded and expelled parts of themselves. Of all the shaikhdoms, Kuwait is the only one in which the ruling family has grown to include a large number of men without expelling most of them. We have a number of historical examples, successful and unsuccessful, of the founding or taking over of settlements by homogeneous groups of tribesmen, and where this has resulted in the successful establishment of a settled community under the descendants of the original tribal shaikhs, the tribe or the tribal section of those shaikhs has been
224
Shaikhdoms of Eastern Arabia
eliminated. We may therefore inquire what it is in the nature of a state of the simplest form on this coast of Arabia which is gradually destructive of the principle of tribalism. For reasons already stated, the kind of town life which we find in the shaikhdoms, life in permanent settlements but with no permanent obligatory attachment of individuals or groups to particular settlements, is logically as close as any settled life can be to the circumstances of bedouin tribal polity. The differences which are inherent in the situation may be listed. The first and perhaps the most important is the constant proximity in which the people of one settlement live. For most of the year, the bedouin spread out over the desert in small groups which are only intermittently in contact with each other. We can say, broadly speaking, that here political and social distance are reflected in spatial distance. Those who are not closely bound together in kinship, members of the same small co-operative group, are at most times out of personal contact with each other. In the towns, even where settlement is on the tribal pattern, the town being divided into quarters in which members of the same tribal section tend to be gathered together, spatial distance cannot reflect political and social distance for most practical purposes. The town is a compact settlement and its inhabitants are in daily contact with each other. There is therefore a need that they should adapt themselves to each other outside situations of kinship much more than there is for the bedouin. The change involved in this adaptation is both social and political. On the social side, we invariably find that settlement is accompanied with the progressive segregation of women, in sharp contrast to the free and usually unveiled condition of the women of the bedouin. Bedouin tents are open, giving little privacy to their occupants, but privacy is not necessary when the tents themselves are cut off from strangers by areas of desert. In the towns, the houses are built to be as private as possible, a wall usually being built inside the doorway in order to make it impossible to see into the courtyard of the house from the street when the door is opened. The women are expected to stay in the house as much as possible. They are expected to avoid almost entirely any social contact with men outside the members of the family. The life of the women in the towns and villages is more secluded than that of the women of the desert, whereas the life of the men is more gregarious. The suq becomes the centre of social life and every man spends part of his day there, sitting and talking with a wide variety of his fellow townsmen and with visitors to the town. In the towns and villages, therefore, we find a division between the social life of the men and that of the
The Shaikhs and Their People 225
women. The women are segregated and each family, men and women, becomes an intensely private unit, whilst at the same time there comes into being a purely male society of wide scope, the society of the suq. Even in this social way, therefore, we can see how concentration of population in the towns is accompanied by a change which stresses the smallest social unit, the family, and the greatest social unit, the total male society of the town, at the expense of the intermediate social groups of the secondary and primary sections of the tribe. We find the same tendency in the political structure. The second difference, which has already been mentioned, is that the town as a commercial centre attracts a mixed population. A bedouin tribe draws to itself accretions from other tribes in the course of time, but these strangers come only occasionally and are incorporated in the tribal system – they are tribesmen originally. The population attracted by the commercial prosperity of the towns is much more mixed, including many of different sect, religion and race. The third difference is that, even though we have stressed the readiness of the Gulf population to move from place to place, when the settled people are compared not with peasants but with bedouin, we find that the townsmen are less free to move than the bedouin for financial reasons. Even though they can move from one place to another without losing their source of income and are often ready to move, still, the owning of permanent houses and land means that movement cannot be accomplished without financial loss. The fourth difference is the growth in the towns of employment and patronage. Until the developments which have come about since the exploitation of oil, we do not find a wage-earning economy in the shaikhdoms. Income is traditionally distributed not through the payment of wages but through profit-sharing, supported by a debt system. The uncertainties of seafaring provide a sufficient reason why this should be so. The social effect of this profit-sharing economy is greater ostensible co-operation between the owner of capital and capital equipment and those who provide the labour. The debt system which accompanied profit-sharing, though it involved what would be regarded in our own society as exploitation, was also a system of patronage and clientage. Each seaman and pearl fisher was in a direct, personal relationship with a superior, the owner of his boat, and could expect from him loans and favours. Because of this, each town contained a large number of patron-client groups, each of these groups of superiors and inferiors being in some senses more closely bound together in mutual dependence and expectation than any part of the group was bound to a similar
226
Shaikhdoms of Eastern Arabia
part of another group. These, so to speak, vertical divisions within the society modified the tendency towards horizontal stratification into social classes based upon wealth and power and the lack of them. We do not find a simple opposition between capitalist class and proletariat; rather do we find the division formed on the model of the jama [a or gum, each of these consisting of one leading family plus all its followers. Such a situation is patently closer to and more compatible with a tribal system than is a system of simple class division between capitalist and proletarian, but it is still a substantial modification of the tribal system. These differences suggest a great deal of what is implied in the modification of the tribal principle of political organization in circumstances of settlement in towns. The history of the shaikhdoms, however, provides us with some knowledge of the actual process whereby the modification comes about. We argue that the living of a large body of people in daily contact with each other renders a simple, segmentary system impossible. A quarrel between two members of different tribes or sections living in the same town cannot be allowed to draw into active hostilities all the members of the respective tribes or sections without completely disrupting the life, and threatening the existence of, the community. Instead, the coercive power of the shaikhs is increased so that order may be preserved. This power is supported by income from taxation, the spending of which is in the hands of the ruler with the results we have described. The increased power of the ruler and of the ruling family as a whole produces a dilemma. The question arises as to whether the power is to be used to transcend the differences within the community or whether it is to be used in favour of those to whom the ruler has special obligations of loyalty, shaff, namely his own family and his own tribal section. This is, simply, a question of whether one man, the ruler, is to dominate the town or whether a large group, the tribal section of the ruler, is to dominate the town through dominating the ruler. The difficulty is increased by the swelling of the town population by foreigners and Arabs from other places. The question is resolved through the internal struggles of the ruling family over who is to be ruler. Those who are dominant in the town – the ruling family and the tribal section of the ruling family – are most closely concerned in the question of which man is to rule. The struggles are frequent, and after each struggle many members of the losing party flee from the state. By this process, the ruling family is constantly reduced in its numbers so that it does not grow over the generations, whilst the tribal section of the ruling family tends to be almost or entirely
The Shaikhs and Their People 227
eliminated. The section which, in the segmentary system of the tribe, enables the shaikhly family to dominate the whole tribe becomes the greatest danger to the principle of impartial transcendence which enables a ruler successfully to rule a town. The minor shaikhs of tribal sections in the town are in a more favourable position for survival since they are the hereditary intermediaries between the ruler and the sections which they themselves lead, but they also tend to be expelled through involving themselves in the internal struggles of the ruling family. In earlier times, some, when dissatisfied, were able to move away with a body of their followers great enough to found a new town, but this, in days of modern governments and frontiers, is no longer possible. When a new town has been founded in the past, the process we have described repeated itself over again there. The economic system in the towns which we have called patronage and clientage can function according to tribe (the richer members of the tribe being economic patrons of their poorer tribesmen) and it has functioned in this way. The system does not, however, depend on the tribal system but can equally well function with a group of followers of various origins as clients of a leading family. Some such families are destroyed through involving themselves in the strife of the ruling family (the Al Ibrahim, at the time the leading merchant family of Kuwait, was destroyed through supporting the sons of the shaikhs whom Mubarak had killed because they were related to them by marriage, and the same came near to happening to the present leading family, the Al Ghanim, in the more recent Kuwait troubles), but when a family is expelled and destroyed in this way its clients can transfer themselves to other families, whereas a tribal section, being formed by common descent, cannot re-form itself once it has been broken. Though the system of patronage and clientage cuts vertically through a class system, nevertheless, once tribalism has been reduced, the merchants of a town begin to form a class and the poor another class. Thus the town, though originally formed on a core of tribal unity, comes slowly, by historical stages, to organize itself into a polity where the main values are total locality on the one hand and family on the other, and within the locality a class system begins to develop.
Editor’s Epilogue
Peter Lienhardt’s study represents, to the best of my knowledge, the first systematic anthropological investigation (conducted over forty years ago) into the political and social conditions of the shaikhdoms of the Trucial States. It is hoped that the present volume will stimulate further anthropological researches into the region that will complement Peter’s research endeavours. However, he left four typed pages of numbered notes which were intended to be incorporated in a conclusion to his book. In these he reflects on some recent developments in the shaikhdoms of the Gulf, and the points raised will be of interest to future researchers on the shaikhdoms of the Trucial States who may find them useful topics for further investigation. I have made a few necessary comments in the form of endnotes but these in no way represent an appraisal of the points enumerated. Peter writes: 1. With oil revenues and the interests of foreign companies a more elaborate and less personal form of government becomes necessary. The shaikhs try to keep senior positions in the family, as in Kuwait.1 From a ruler’s point of view, the members of his family are the people he can trust (like a merchant running a family business) and also their goodwill can be bought by giving them profitable positions. This has happened in Abu Dhabi as it happened before in Kuwait. 2. The major interest of oil companies makes it impossible that the old system of superseding rulers by rebellion and murder should continue. For example, there are claimants to the shaikhdoms of Abu Dhabi now living 2 in Saudi Arabia. If they were to seize power in Abu Dhabi with Saudi Arabia help, as Shakhbut and Zayid have feared they might, the result would be much more than a local political crisis. 228
Editor’s Epilogue
229
3. The position of rulers is strengthened by their power to distribute wealth. They can now also afford to set up private armies on a scale of members and armaments hitherto unheard of.3 4. But the rulers and shaikhs cannot change the moral expectations of their people overnight. All the shaikhs in administrative positions become extremely harassed by the business of administration. They have much less time to spare for people than they had before. They are criticized for being overbearing, vain and incompetent. 5. The oil companies provide a model of labour relations and concern with the welfare of employees derived from Britain and America. They also provide a model of administrative competence. This has been so in the whole of the Gulf, Iraq and Iran. Local people are not aware that such administration depends on the social system of the countries it is taken from and think that, given the money for paying salaries and building offices, it ought to be easy to copy. The shaikhs begin to be regarded as unfit to hold their positions and people travelling to foreign countries, which they can now well afford to, begin to think of the shaikhs as an embarrassment and a disgrace. 6. It is thought that the most important contribution to progress, where local people do not have to compare themselves unfavourably with anyone else in the world, is education.4 Arab teachers are brought in from other countries. The shaikhs prefer people who have a suitable respect for hereditary authority. Abu Dhabi, therefore, brought Palestinian teachers from Jordan, mainly because Jordan has a king. When I knew them, the shaikhs never considered that these teachers came from the same background of anger and disappointment as those who make up Al-Fatah. 7. However much they pay them, the shaikhs treat their foreign teachers as helots. The lessons taught to small children cannot help but communicate attitudes critical of the way the countries are run. 8. What no one in the Trucial States is likely to understand (and not so many people here) is that education is produced in particular societies and then exported from them in something of the same way as motor cars, designed basically for home consumption, are exported by companies like General Motors. There is no possibility whatsoever, in these small-scale societies of the Gulf, of producing an educational system in accordance with the local system of values. The values which make their way in cannot help but be associated with forms of government quite different from those traditional in the Gulf.5
230
Shaikhdoms of Eastern Arabia
9. Administrative jobs fall progressively into the hands of a new class of local people. These people do not come from either the ruling families or the families of merchants because such families have opportunities of making far more money out of trade and land speculation than they could ever do out of civil service salaries. The day-to-day running of affairs, and subsequently the formulation of policy, come to be the business of members of poorer families who have had to work hard at school and university. This is more compatible with ideas of Arab socialism than with Middle Eastern shaikhs and kings. 10. On the other side is military organization. The shaikhs continue to maintain contact with the bedouin and with bedouin values, and to this extent they continue to transcend the diverse values of sections of the populations of their own countries. Most of the military forces, now well-armed, are made up of bedouin. The shaikhs would probably be ready to use them to keep control of their townspeople. If any sort of federation really comes into being, the major military force will be the one that started as the Trucial Oman Levies. This force contains a large number of minor shaikhs from the various states, and some of them have already tried to advance their own political prospects through their military connections. 11. If this force and the modern townsmen are able to agree with each other (which would be much easier in a federation than in separate states) then we may expect the same sort of change as has happened in the Yemen and South Yemen. NB. I shall have to be extraordinarily careful with these speculations, because this will in any case be the first book in English that has tried to say anything about the weakness of the shaikhs in traditional life, though this is what one is told over and over again in the countries. I expect it to make some contribution towards explaining the problems of the Trucial Coast to the people who live there – at least to the younger ones. It will have some influence on future events – how much of course I do not know – but I do not want that influence to be damaging. Written properly it might be constructive. 12. There will be some rather more abstract considerations set out in a few pages of anthropological conclusions. I cannot formulate them until I have the rest of the book as a whole. They will mainly be concerned with how far one can push the ideas of status and contract in analysis, but there are all sorts of other things adumbrated in various places and they seem to me to fit together.6
Appendix On separately typed pages Peter Lienhardt included the terms of the truce which were intended to be included in Chapter 1. In view of their relevance to the contents of that chapter, I include his version of them here. Arabic words and names of people and places have been maintained as they appeared in the text of the truce. The following is the text of the first of the series of truces from which the Trucial Coast of Oman derived its name: Translation of the terms of a truce for six months agreed on by the Chiefs of the Arabian Coast in the presence of Captain Hennell, the Acting Resident in the Persian Gulf, dated Bassadore, the 21st May, 1835. We whose seals are hereunto attached, Vizt. Sultan bin Suggur, Shaik of the Joasmee Tribe, Shackboot, the father and plenipotentiary of Shaik Khuleefa, Chief of the Beniyas, Obed ben Saeed, the Chief of the Boo Felasa tribe and Dubaye, and Rashid ben Humed, the Shaik of Eyman, being fully impressed with a sense of the evils suffered by our subjects and dependents in consequence of their being debarred from carrying on the Pearl Fishery on the Banks during the present state of hostilities among ourselves, and duly appreciating the general advantages that would be derived from the establishment of a Truce during the Fishing Season, do hereby agree to bind ourselves down to observe the following conditions: 1st. That from the 2nd Mohurrum 1251 (or 21st May 1835) there shall be a cessation of hostilities at sea between our respective subjects and dependents, and that from the above date until the 29th Rujub, AH 1251 (21st November 1835) an inviolable truce shall be established, during which period our several claims upon each other shall rest in abeyance. 2nd. That in the event of any one of our subjects or dependents committing an act of aggression by sea upon those of the parties to this engagement we will immediately afford full redress upon the same being brought to our notice. 3rd. That in the event of an act of aggression being committed at sea upon any one of our subjects or dependents who are parties to 231
232
Appendix
the truce, we will not proceed to immediately retaliate, but will inform the Rest. at Bushire, or the Commodore at Bassadore, who will forthwith take the necessary steps for obtaining reparation for the injury inflicted upon its being satisfactorily proved. 4th. That on the 30th Jumadee-ool-Akhir 1251, by the blessing of Providence, we will endeavour to arrange either an extension of this truce, or a firm and lasting peace, but in the event of our not being able to come to a satisfactory arrangement regarding our respective claims among ourselves, we hereby bind ourselves to give notice on or about the above date, to the Resident of Bushire of our intention to renew hostilities after the expiration of the term now fixed upon for this Truce, Vizt. the 29th Rujub 1251. L.S. L.S. L.S. L.S. L.S.
Seal of Sultan bin Suggur Seal of Obed bin Saeed Seal of Rusheed bin Humed ⎫ Seal of Hajee Ahmed ⎬ as witnesses for Shackboot Seal of Moolah Hoosein ⎭ True translation /Signed/ S. Hennell, Assistt. in Charge of the Resy., Pr. Gulf.
Peter Lienhardt made the following important comment with regard to the above agreement: Apart from giving rise to the name Trucial, this truce and its successors represent some of the stages of intervention on the part of the British Government whereby the Trucial Coast came to be isolated for a century from many of the currents of world affairs. The very name is symptomatic, for its use by British officials gave it international currency and the name Trucial States came subsequently to be translated literally into Arabic as Al-Imarat Al-Mutasaliha, but it was not much used locally. The people who lived in the area usually called it ‘the Coast of Oman’, Sahil [Uman.
Notes Inverted commas signify Peter Lienhardt’s orginal footnotes. Editorial additions to these are in square brackets. All other notes are the Editor’s.
Editor’s Preface 1 For a short biography and publications of Peter Lienhardt, see Ahmed Al-Shahi, ‘Peter Lienhardt 1928–1986: Biographical notes and bibliography’ in Journal of the Anthropological Society of Oxford, vol. 27, no. 2, 1996. 2 For Peter’s appointment and resignation, see the reports of HM Political Agent in the Public Record Office, FO 371/File 157018. 3 Public Record Office, FO 371/File 162780, Despatch no. 104, 28 December 1961. 4 “This has been commented on by Professor Ian Cunnison in The Baggara Arabs.” See Preface, ibid. 5 “Since quite strong political feelings are involved over whether it should be called the Persian or the Arabian Gulf, it seems better simply to call it the Gulf.” 6 When Lienhardt was at Cambridge University (1946–9), he read English for Tripos Part I and Arabic and Persian for Part II. 7 Agnus Dei is a part of the Roman Catholic mass set to music and including the words Agnus Dei, ‘Lamb of God’.
1
The Political Complex
1 “I use the word ideology here in a sense broader than it carries in some political writings, to mean the complex of ideas, beliefs and assumptions of value whereby life is interpreted and actions are decided and judged in a community. An alternative expression would be collective representations, but this term of Durkheim’s is still not widely understood outside sociological and anthropological circles and may be taken to imply too little possibility of disagreement about standards in a community and an absence of conscious intellectual activity. In my use of the word, all societies have ideologies and the explicit ideology of various political schools would only be part of the data for studying the ideology of the countries or groups where such schools are powerful.” 2 See Appendix. 3 “a few years ago”: i.e. about the late 1960s, when Lienhardt was writing this book. 4 It must be remembered that Lienhardt gave this information in the late 1960s, and thus the equivalent of this amount nowadays is considerable. 5 “So written in Arabic, but pronounced Jtab.” 6 Lienhardt added, on a separate sheet of paper, some points which he wanted to elaborate on at the end of the section. However, this he did not do and hence I give below these points, which would be of interest for future researchers in the area: 233
234
Notes
1. Truces and treaties: related to possibility of contract beyond status in local terms. British dealing equally with rulers – a special alliance with Muscat passed away with embarrassments. They would not have been dealing equally with European states, i.e., the differences of power were so disproportionate they did not need allies. 2. The British could not try pirates in India: i.e., it was a whole legal system they had in culture contact, not capable of partial manipulation. The local system is not so centralized as to make execution of ‘criminals’ a deterrent. The object of blood compensation is reconciliation, not deterrence. 3. Did the British change the position of rulers by making them more responsible for their subjects? A bit, because rulers became more cautious about claiming followings they knew they could not control. 4. Increase in town wealth, with peace, meant more power for capitalists against shaikhs. Decline of pearl fishing will fit into passage about the change in the balance of power between rulers and people. 5. It took the States a long time to take over a version of the dominant foreign educational system. Had they been left to their own devices in relation to the outside world, they would have needed to do more. Had they been colonized (protectorates) the colonizers would have had to do more. 6. Trucial Scouts made the British arbiters of who was to rule: interfered with the ‘natural’ system of coup d’état. Oil required, or preferred, greater continuity and consistency in government. This relates to forcing shaikhs to stand by agreements in early 19th C. 7 “as represented in J. B. Kelly, Oxford 1968 and J. H. Lorimer 1970.” 8 Lienhardt gives the following list of names of different types of local crafts with their uses: Trading Voyages: Bum (pl. abwam), Kutiya (kutiyat), Bidan (bidana), Balam (balama), Jalbut ( jawalbit), Sanbuk (sanabich), Battil (batatil), [Uwaisi ([uwaisat), Baggara (bagagir), Baghala (baghalat). Pearl Fishing: Sam[a (sim[), Jalbut, Sanbuk, Bidan, Bailai (biyali), Shahuf (shawahif ), Baggara, [Uwaisi. Fishing: Shahuf, Sanbuk, Jalbut, [Uwaisi, Baggara, Sam[a, Shu[li (shawa[i), Huri (hawari), Gharaib ( ghiyarib), Shasha (shashat).
2
Women and Men
1 In the Trucial States and neighbouring Arab countries, this room is referred to as d iwaniya. Many of the houses nowadays have a guest room, but only a category of people keep diwaniya open regularly to receive guests. The implication is that these people occupy important positions in society and it is expected that other people come to see them. 2 al-bait baitak wa-]l masjid adfalak. 3 The Baluchis are a group of tribes who inhabit the province of Baluchistan in Pakistan and also neighbouring areas of Iran, Afghanistan and Punjab (India). Some Baluchis are also to be found in Oman. They speak Baluchi, a language belonging to the Iranian branch of the Indo-European language family. They are Muslims and number about two million people whose main mode of livelihood is nomadism. Since the early periods of the development of the Trucial
Notes 235
4 5 6
7
8 9
States, some Baluchis have been working there in the domestic and construction sectors. al-rayyal afdal min al-hurma bi-darajatain. “Reports of the sayings or actions of the Prophet Muhammed upon which so much of the religious law is based.” For this quotation and the prescribed distribution of shares among potential inheritors, see Arthur J. Arberry, 1964, ‘Women’, verses 11, 12, 34 and 176 (pp. 73–4, 77 and 97). Lienhardt wrote in the margin, as an additional observation, that “women praying in the house – a ‘holy’ place – say bismillah, in the name of God, when entering it”. “bismillah al-Rahman al-Rahim, in the name of Allah, the Merciful and the Compassionate.” Arthur J. Arberry, 1964, ‘The Light’, verse 31 (pp. 355–6). Peter Lienhardt wrote, in a footnote, that the chapter is so entitled because of the passage: God is the Light of the heavens and the earth; the likeness of His Light is as a niche wherein is a lamp (the lamp in a glass, the glass as it were a glittering star) kindled from a Blessed Tree, an olive that is neither of the East nor of the West whose oil wellnigh would shine, even if no fire touched it; Light Upon Light; (God guides to His Light whom He will.) (And God strikes similitudes for men, and God has knowledge of everything.) Ibid., verse 35 (pp. 356–7).
10 Ibid., ‘Women’, verse 34 (pp. 77–8). 11 Ibid., ‘The Light’, verses 25–6 (p. 355). 12 “The word meaning ‘a girl’, bint, also means colloquially ‘virgin’, and the word meaning ‘old woman’, [ajuz (a category a woman enters in her forties, though one might hesitate to call her so in her presence) has a philological connection with inability – in this case inability to bear children.” 13 ra [iyat al-bait, or ja [ida from the classical Arabic word qa [ida, meaning foundation, basis, fundament, support, base. 14 majlis is a shaikh’s court, a bedouin council or a public meeting. 15 Peter Lienhardt wrote in the margin: “more? period?? of qa [ada”. No doubt he wanted to elaborate further on this issue. The qa[ada (meaning in Arabic: to sit down, stay, abide) is similar in implication to the legally prescribed forty days, period of confinement, [idda. 16 Lienhardt left a blank as to whatever caused the blame attached to the pagan Arabs. In the context of his discussion here, I think he was referring to the custom of wa [d, female infanticide. 17 At the time of writing his material on the Trucial States, the Ruler of Dubai was Shaikh Sa[id bin Maktum. He was born probably in 1878 and he ruled from 1912 until his death in 1958. 18 “One hears in Dubai of one poor but extraordinary woman, also described as being courageous like a man, who used to go out fishing in a boat with men. I do not know what her family thought of her, but she was not despised by the general public, whom she astonished.” 19 Shaikha Hussa bint Al-Murr’s date of birth is unknown. She died in AH 1369 (AD 1949–50). My thanks for this information to Dr Paul Auchterlonie, Librarian, University of Exeter.
236
Notes
20 “The local name for which is the Persian word karkhaneh, which means ‘factory’ or ‘workshop’.” 21 From the Arabic: sabba, to insult, cause, provoke, arouse, revile. 22 “On this subject, I naturally speak from hearsay.” 23 See also Chapter 5, note 9. 24 Lienhardt intended to elaborate on what he wrote in the margin: “Status of nasib [affine] in the towns and villages confirms the practicality.” 25 “Unlike the use of the term brother to mean ‘friend’, which can be highly confusing to a visitor.” 26 Lienhardt cited the following parallel but then he crossed it from the text: “in a way much more profound and complex than that in which the members of English suburban golf clubs aspire to being country squires.” 27 Lienhardt’s comment in the margin reads: “Yes, i.e., only son can be agnatic link, not daughter.”
3
The Bedouin
1 Bitter Apple or colocynth (also known as Bitter Cucumber), citrullus colocynthis, a hairy stemmed climbing vine of the gourd family, native to the Mediterranean region. It has rounded yellow or green fruits that have a bitter taste. It is cultivated for the cathartic quality of its fruits. 2 falaj (plural: fuluj) from the Arabic falaj: to split, cleave. In classical Arabic shari [a means the canonical law of Islam, water hole, drinking place and the approach to a water hole. 3 shaff : diaphanous fabric, gauze. In classical Arabic shaffa means to be thin, flimsy; to be transparent, diaphanous. 4 In Arabic, jidd or jadd (plural: ajdad or judud). But in the Trucial Coast it is pronounced yidd. 5 “I.e. Elias, though he is strictly Ilyas, which would make the derivation more awkward than it is when the first syllable is taken to be definite article.” 6 “ya [ayal Mansur”, ya [ayal [Amir. 7 d iya from the classical Arabic wada yad i, to pay blood money. Lienhardt also gave mudda as another term for the d iya. Mudda in classical Arabic means space or length of time. This refers to the period of time given to the offender and his kin to pay the blood money. It is conventional that no further homicide should take place during the mudda. 8 “I do not, any more than do most anthropologists of the present day, subscribe to much of what has been said about function in earlier anthropological writing, but I think it valid to use the word in the present context. If the concept of function, as it has been used in earlier anthropological analysis, is totally and absolutely wrong, then a good deal of the present argument falls to the ground, but I can conceive of no other argument which would make an equally coherent structure out of the ethnographic facts.”
4
Towns and Maritime Activities
1 J. G. Lorimer, Gazetteer, vol. 1 (Historical), p. 763. 2 J. H. Stocqueler, 1832, p. 19.
Notes 237 3 This travel document was translated from the Greek and annotated by W. H. Schoff (1912). 4 The Ruler of Dubai at the time of writing this book was Shaikh Rashid bin Sa[id Maktum. 5 The Ruler of Abu Dhabi was Shaikh Shakhbut bin Sultan, who ruled from 1928 until 1966. 6 Lienhardt’s comment on this reads: “The British Government made some financial contribution.” 7 “Cf. Alan Villiers, 1940, pp. 284–5. In the Trucial States the idea does not seem to be taken seriously, and there are no special precautions of watching the keel day and night as Commander Villiers reports from Kuwait.” 8 “I am grateful to Commander Villiers for letting me see this MS.” 9 J. H. Stocqueler, ibid., pp. 2–3. 10 Ibid, p. 2. 11 jus gentium: “in Roman law, that part of the law which the Romans applied both to themselves and to foreigners. In legal theory it is linked with the idea of a law common to all peoples and dictated by nature. . . . ” See Encyclopædia Britannica, Micropædia, vol. 5, 15th edition, Chicago, London etc., 1974, p. 642. 12 “ghos al-ghaid (Lorimer has al-ghos al-Kabir).” 13 “ghos as-saif (Lorimer has ghos al-bard).” 14 “Lorimer calls this rudud, but people in the Trucial Coast did not speak of this season.” 15 “I think of Trollope’s [1815–82] account, in his Autobiography, of the moneylender who used to visit him at his office when, as a young man, he had contracted a debt of £16, and who eventually extracted over £200 from him.” 16 “Called, in local pronunciation, tajaidama, derived from the word for mate, mujaddam, i.e. muqaddam.” 17 “jallas (sing.).” 18 “The contract was called iltizam or tamassuk, and the Arabic formula was: Aqul anni bin fulan wa-anni qad iltazamt hadha al-kuthr fulus min haqq al-madhkur, la fukakan f i dhimmati illa bi-]l-waf a] wa-]t-tamam.” 19 “sagam, also called rudud.” 20 “Ila man yara min kafat an-nawakhidha bi-anna lana [ala fulan mablagh (kadha) man arad yadummu yusellim al-mablagh.” 21 “[As above] . . . bi-anna ma lana [ala fulan shai] man arad yadummu yadummu. Such slips were called barwa, which is said to be a corruption of bara]a.”
5
The Shaikhly Families
1 The Ruler of Ras Al-Khaimah, when Lienhardt wrote his thesis in 1956–7, was Shaikh Saqr bin Muhammed bin Salim bin Sultan who came to be the Ruler in 1948. 2 I.e. 1948 (see preceding note). 3 “The word bin, son of, is little used in names in Kuwait. Among the more humble people, Ahmed the son of Jabir would be called Ahmed Jabir. For the shaikhs and more important men, al is used as in this example. It is another form of ahl and refers to the family group.”
238
Notes
4 Shaikh Ahmed Jabir Al Sabah ruled from 1921 until his death in 1950. 5 This would be about the mid 1950s when Lienhardt was undertaking his fieldwork in the Trucial Coast. 6 No doubt the situation with regard to the distribution of oil revenues in the Trucial States has changed since Peter wrote this. 7 Lorimer, Gazetteer, vol. 2 B (Geographical and Statistical), p. 1076. 8 Lorimer, loc. cit. 9 “In the Trucial Coast, there is a considerable amount of folklore about the ways in which it is possible, by ritual means, to prevent the consummation of a marriage. This can be done in any of the following six ways: (a) by taking one of the stones that support the cooking pots whilst the marriage feast is being prepared and dropping it down a well. (b) by taking some of the cud of a goat and mixing it in the stew, marag, of the wedding feast. (c) by taking rice from the wedding feast and burying it in the ground. (d) by turning a ring round and round on the finger at the ceremony at which the marriage is contracted, milcha, whilst the religious man who conducts the ceremony is reading the Koran. (e) by breaking a twig or a matchstick over and over again on the same occasion of the milcha. (f) by turning a date stone over and over in the hand whilst the milcha is proceeding. It is said to be for fear of such magic that the milcha ceremony is kept absolutely private for the close male kin of the bride and bridegroom, and also that this is the reason why those present at this ceremony must keep their hands flat on their knees whilst the ceremony continues. That there should be such variety of ways in which it is thought possible to make the bridegroom physically incapable of consummating the marriage, and that these should be taken seriously even though there is relatively little superstition of other sorts, would seem to reflect a considerable fear (and perhaps of incidence) of such inability of the part of the bridegrooms, a result, perhaps, of the segregation of women. The bride and bridegroom do not meet until the wedding night.” 10 The Ruler was Shaikh Shakhbut bin Sultan. 11 “The late Yusuf bin [Isa al-Qina[i, a Kuwaiti, writes in his pamphlet Saf ha min Tar ikh Al-Kuwait of the position of women there that to open the windows of a house is considered very shameful because the voices of the women would be heard. He remarks that in spite of this jealousy over women whereby it is thought that they should be not seen nor heard, Kuwaitis do not honour them and if a man is telling a story to a friend and mentions a woman he will say akramak Allah (‘may God make you more noble’ i.e. excusing himself), which he would not say if he were mentioning an insect to his hearer: op. cit., p. 73.” 12 In the colloquial Arabic of the Gulf the k is sometimes pronounced ch. Thus in keeping with the convention adopted in this book, I have changed Machtum, as given by Lienhardt, to Maktum. These changes apply to names of people only.
Notes 239 13 The situation with regard to the education of women is different now, as young girls nowadays go to schools and pursue higher education. 14 The Ruler of Kuwait was Shaikh [Abdullah Al-Salim Al Sabah. 15 “Before the making of the marriage contract, the father or other wakil, matrimonial guardian, of the girl takes a witness, shahid, with him to consult the girl from outside the room in which she is. The father asks her: wakkaltini amlich bich, ‘Do you authorize me to make a wedding contract for you?’ If the girl is young and it is her first marriage, silence is taken for consent, as-sukut ar-rida]. If she is older, has been married already or if her wakil is someone other than her father, it is necessary that she should reply: muwakkal, ‘You are authorized.’ At the ceremony where the marriage contract is made, milcha, the religious man, mutawwa[, who officiates first asks the witness whether he has been to the girl and whether her father has consulted her: sirt [andaha wa astantaqha abuha?” 16 The present Ruler was Shaikh Shakhbut bin Sultan. 17 “The earliest parts of this genealogy as given by Lorimer are omitted since they deal not with actual people, as Lorimer’s genealogy suggests, but with the four sections of the Al Bu Falah, viz. Al Nahaiyan, Al Sa[dun, Al Muhammed and Al Sultan. The dates refer to the periods during which each ruler was in office.” See Lorimer, op. cit., Historical, Part 3 (Genealogical Trees, Pocket No. 3). 18 See H. R. P. Dickson, 1956, p. 102. 19 “This account is taken from Wilfred Thesiger, 1950.” Text slightly amended. Discrepancies in dating, such as those that may be noticed between versions based on Lorimer and Thesiger, may be attributed to the largely undocumented nature of the region’s pre-twentieth-century history. 20 “This account is slightly simplified. At the deaths of the Ruler’s inheritors there will also be fara]id to deduct, so that their whole estates will not be inherited by their children but only the remainders after the fara]id applicable in the particular cases have been deducted.” 21 “When the late King of Saudi Arabia was alive, his eldest son was made heir apparent. Now that son rules, but his brother, not his son, is heir apparent, the intention clearly being to avoid strife between the two brothers.” 22 “Raids have almost ceased now: I speak of bedouin society in the past, before modern restraints applied.” 23 “Even a joke in this tone is thought rather shameful by many young people of the present day. The complementary attitude of the wife is in the terms: himar yubadil himar, one donkey will take the place of another.” 24 “There is an interesting and accurate description of unhappy relations of a wife with her husband’s family in a book of short stories by the modern Kuwaiti writer Fadil Khalaf, in a story entitled ‘Hanan Umm’.”
6
The Shaikhs and Their People
1 The Koran, ‘The Cave’. 2 “The verse of the Koran (‘Women’, 62): ‘O you who believe, obey God and the Prophet and those of you who are in command’, though a favourite with the late King of Ibn Saud, gives only a very limited prerogative to the commander
240
3 4 5 6 7
8
Notes and the turn of the phrase makes it clear that the commander is no special person as compared with the rest.” J. E. C. Welldon, pp. 348 and 367–8. Lorimer, op. cit., vol. 1 (Historical, Part 1A), pp. 772–4. About the early 1920s. kabab refers to minced or diced meat cooked on a metal or wooden skewer. “H. R. P. Dickson, 1956, p. 430. There is a short account of the Kuwait majlis in Allan Villiers, 1940, pp. 211–13, but both Dickson and Villiers omit any reference to the involvement of important members of the Kuwait ruling family against the ruler. A better short account is given in R. H. Sanger, 1954, pp. 168–9.” “Ras Al-Khaimah was the older of the two, under its earlier name of Julfar. It is referred to in the Calender of State Papers (Colonial, East Indies and Persia 1630–1634), p. 131, and in the sixteenth century was taxed by the Portuguese (see S. B. Miles, 1966), p. 155. Chinese pottery found there by myself suggests that the town may have existed as early as the fourteenth century.”
Editor’s Epilogue 1 This is still the case in all the shaikhdoms of the Trucial States. 2 The word ‘now’ refers to the time when the present material was written, between 1957 and the late 1960s. 3 While the shaikhdoms have established armies since they became independent states, the interest in strengthening their military capabilities and alliances has assumed importance and priority after the Iraqi invasion of Kuwait in 1990. 4 Interest in education and the establishment of educational institutions such as schools, universities and technical institutes are seen as an investment for the future. This sector has grown considerably as educated people have begun to seek employment within their own countries. Also, the educated class think that entering the field of politics requires academic qualifications. 5 Lienhardt has proved to be correct in his assumption: educated people are beginning to demand more liberalization of their political systems. 6 Unfortunately Lienhardt did not write the anthropological conclusions to his research as this book was left incomplete and unpublished at the time of his death.
Select Bibliography Peter Lienhardt discussed in this volume a number of themes such as the state, traditional economy, women and men, bedouin, towns and shaikhs. It has not been feasible to include the vast published material on these themes. Thus the following bibliography includes items cited by Peter and some selected recent publications on the Arab shaikhdoms of the Gulf. Abboud, S. H. ‘Fishing for pearls’ in Middle Eastern Forum, vol. 38, no. 7, 1962. [Abd Al-Rahman, [Abd Allah. Al-imarat f i dhakirat abna]iha; jam[ wa-tahqiq, 2 vols., Ittihad Kuttab wa-Udaba] al-Imarat, Al-Shariqah, 1898–90. Abdullah, Muhammad Morsey. The United Arab Emirates, Croom Helm, London, 1978. Abu-Hakima, Ahmed Mustafa. History of Eastern Arabia 1750–1800: the rise and development of Bahrain and Kuwait, Khyats, Beirut, 1965. — — . The Modern History of Kuwait, Luzac and Co., London, 1983. — — . (ed.). History of Kuwait, 2 vols., Kuwait Government Printing Press, Kuwait, 1967. Abu Saud, Abeer. Qatari Women: Past and present, Longman, London, 1984. Ainsworth, W. F. A Personal Narrative of the Euphrates Expedition, Kegan Paul, London, 1888. Al-Ebraheem, Hassan Ali. Kuwait and the Gulf, Georgetown University, Centre for Contemporary Arab Studies, Washington, DC, 1984. Al-Ghanim, Salwa. The Reign of Mubarak Al-Sabah, Shaikh of Kuwait, 1896–1915, I. B. Turis, London, 1998. Al-Moosa, Abdulrasool and McLachlan, Keith. Immigrant Labour in Kuwait, Croom Helm, London, 1985. Al-Mughni, Haya. Women in Kuwait: the politics of gender, Saqi Books, London, 1993. Al-Nabahani, Shaikh Muhammed b. [Isa. Al-tuhfa al-Nabahaniya f i tar ikh al-Jazira al-[Arabiya, 2nd edition, Al-Mahmudiya Press, Cairo, AH 1347 (AD 1928). Al-Qasimi, Sultan ibn Muhammad. The Myth of Arab Piracy in the Gulf, Routledge, London, 1988. Al-Qina[i, Yusuf b. [Isa. Safha min tarikh al-Kuwait (A page from the history of Kuwait), 4th edition, State of Kuwait Printing Press, Kuwait, 1968 (originally published in Cairo in 1946). Al-Rashid,[Abd Al-[Aziz. Tarikh al-Kuwait (History of Kuwait), Al-Hayat Publishing House, Beirut, 1971. Al-Sar[awi,[Abd Al-[Aziz. Sittat asabi[ f i imarat Abu Dhabi (Six weeks in the Emirate of Abu Dhabi), Kuwait Government Printing Press, Kuwait, c.1961. Al-Shamlan, Saif Marzoog. Min tarikh al-Kuwait (From the history of Kuwait), Nahdat Misr Press, Cairo, 1959. — — . Tarikh al-ghaws [ala al-lu ]lu] f i al-Kuwait wa-]l-Khalij al-[Arabi (History of pearling in Kuwait and the Arabian Gulf ), Kuwait Government Printing Press, Kuwait, 1975. Al-Taher, Isam. Kuwait: the reality, Dorrance, Pittsburgh, Pa., 1995. 241
242
Select Bibliography
Amin, Abdul Amir. British Interests in the Persian Gulf, E. J. Brill, Leiden, 1967. Anthony, John Duke. Arab States of the Lower Gulf: People, politics, petroleum, Middle East Institute, Washington, DC, 1975. Arberry, Arthur. J. The Koran Interpreted, Oxford University Press, London, 1964. Assiri, Abdul-Reda. Kuwait’s Foreign Policy: City-state in world politics, Westview, Boulder, Colorado, 1990. Ayoob, Mohammed. ‘Oil, Arabism and Islam: the Persian Gulf in world politics’ in The Middle East in World Politics, edited by Mohammed Ayoob, Croom Helm, London, 1981. Badger, G. P. The History of the Imams and Seyyids of [Oman, by Salil-ibn-Razik from AD 661–1856, Hakluyt Society, London, 1871. Bahrain Government. Government of Bahrain Annual Reports, 1937–. Balfour-Paul, H. G. ‘Recent developments in the Persian Gulf’ in Journal of the Royal Central Asian Society, vol. 56, no. 1, 1969. Belgrave, Sir Charles D. ‘Pearl diving in Bahrain’ in Journal of the Royal Central Asian Society, vol. 21, no. 3, 1934. — — . ‘The Portuguese in the Bahrain Islands 1521–1602’ in Journal of the Royal Central Asian Society, vol. 22, October, 1935. — — . Personal Column, Hutchinson, London, 1960. — — . The Pirate Coast, Bell, London, 1966; Librairie du Liban, Beirut, 1972. Belgrave, J. H. D. ‘A brief survey of the history of the Bahrain Islands’ in Journal of the Royal Central Asian Society, vol. 39, January, 1952. Bent, T. ‘The Bahrain Islands in the Persian Gulf’ in Proceedings of the Royal Geographical Society, vol. 12, no. 1, 1890. Southern Arabia, Smith, Elder, London, 1900. Bidwell, R. Travellers in Arabia, Hamlyn, London, 1976. — — . (ed.) The Affairs of Kuwait, 1896–1905, 2 vols., Cass, London, 1971. Blandford, L. Oil Sheikhs, Weidenfeld and Nicolson, London, 1976. Bombay Government. The Persian Gulf: Selections from the records of the Bombay Government (Bombay Selection 25), Bombay, 1856. Bosworth, C. E., Burrell, R. M., McLachlan, K. and Savory, R. M. (eds.). The Persian Gulf: a general survey, John Hopkins University Press, Baltimore, Maryland, 1980. Bowen, R. Le Baron. ‘Pearl fisheries of the Persian Gulf’ in The Middle East Journal, vol. 5, no. 2, 1951. Bowen-Jones, H. ‘Agriculture in Bahrain, Kuwait, Qatar and UAE’ in Issues in Development: the Arab Gulf states, edited by May Ziwar-Daftari, MD Research and Services, London, 1980. Brydges, Sir H. J. An Account of the Transactions of H.M. Mission to the court of Persia in the Years 1807–11; and a brief history of the Wahauby, 2 vols. (in one), James Bohn, London, 1834. Burdett, A. (ed.) Records of the Persian Gulf Pearl Fisheries, 4 vols., Archive Editions, London, 1995. Burrell, R. M. The Persian Gulf (Washington DC: Centre for Strategic and International Studies, Georgetown University), Library Press, New York, 1972. — — . ‘Policies of the Arab littoral states in the Persian Gulf Region’ in The Persian Gulf and Indian Ocean in International Politics, edited by Abbas Amiri, Institute for International Political and Economic Studies, Tehran, 1975.
Select Bibliography
243
Busch, B. C. Britain and the Persian Gulf, 1894–1914, University of California Press, Berkeley, 1967. Clements, F. Oman, the Reborn Land, Longman, London, 1980. Codrai, R. The Seven Shaikhdoms: Life in the Trucial States before the federation of the United Arab Emirates, Stacey International, London, 1990. Cook, David C. Kuwait: Miracle of the desert, Grasset and Dunlop, New York, 1970. Crystal, Jill. Oil and Politics in the Gulf: Rulers and merchants in Kuwait and Qatar, Cambridge University Press, Cambridge, 1990. — — . Kuwait: the transformation of an oil state, Westview, Boulder, Colorado, 1992. Cunnison, I. The Baggara Arabs: Power and the lineage in a Sudanese nomad tribe, Clarendon Press, Oxford, 1966. Daniels, J. Abu Dhabi: a portrait, Longman, 1974. Dickson, H. R. P. Kuwait and her Neighbours (edited for publication by Clifford Witting), Allen and Unwin, London, 1956. — — . The Arab of the Desert: a glimpse into badawin life in Kuwait and Sa[udi Arabia, Allen and Unwin, London, 1971. Dickson, Violet. Forty Years in Kuwait, Allen and Unwin, London, 1971. El Mallakh, Ragaei. The Economic Development of the UAE, Croom Helm, London, 1981. El-Solh, Raghid (ed.). The Sultanate of Oman, Ithaca Press, Reading, 1998. Emirate of Abu Dhabi – Documents and Research Bureau. Abu Dhabi: Past and present, Orient Press, Beirut, 1969. — — . Two Glorious Years in the History of the Emirate of Abu Dhabi, Beirut Printing Press, Beirut, n.d. Fakhro, Munira A. Women at Work in the Gulf: a case study of Bahrain, Kegan Paul International, London, 1990. Faroughby, A. The Bahrain Islands 750–1951: a contribution to the study of power politics in the Persian Gulf. An historical, economic and geographical survey, Very, Fisher, New York, 1951. Fenelon, K. The United Arab Emirates: an economic and social survey, Longman, London, 1976. Firth, R. W. Malay Fishermen: Their peasant economy, Routledge and Kegan Paul, London, 1966. Freeth, Zahra Dickson. Kuwait was my Home, Allen and Unwin, London, 1956. — — . New Look at Kuwait, Allen and Unwin, London, 1972. — — . with H. V. F. Winstone, Kuwait: Prospect and reality, Allen and Unwin, London, 1972. — — . with R. B. Lewcock, Traditional Architecture in Kuwait and the Northern Gulf, United Bank of Kuwait, London, 1978. Ghabra, Shafeeq N. Palestinians in Kuwait: the family and the politics of survival, Westview, Boulder, Colorado, 1978. Graham, Helga. Arabian Time Machine: Self-portrait of an oil state, Heinemann, London, 1978. Harrison, P. W. Doctor in Arabia, R. Hale, London, 1943. Hawley, D. The Trucial States, Allen and Unwin, London, 1970. Hay, R. The Persian Gulf States, Middle East Institute, Washington, DC, 1959. Heard-Bey, F. ‘Social change in the Gulf States and Oman’ in Asian Affairs, vol. 59, no. 3, 1972.
244
Select Bibliography
——. From Trucial States to United Arab Emirates: a society in transition, Longman, London, 1982. Henderson, E. This Strange Eventful History: Memoirs of earlier days in the UAE and Oman, Quartet Books, London, New York, 1989 [1988]. Hewins, R. A Golden Dream: the miracle of Kuwait, W. A. Allen, London, 1963. Hill, A. G. ‘The Gulf states: Petroleum and population growth’ in Populations of the Middle East and North Africa, edited by J. I. Clarke and W. B. Fisher, University of London Press, London, 1972. Hopwood, D. (ed.). The Arabian Peninsula: Society and politics, Allen and Unwin, London, 1972. Husain, [Abd Allah Gulum and Ismail, [Izzat Sayyid. Al-zawaj f i al-Kuwait (Marriage in Kuwait), Kuwait Government Printing Press, Kuwait, n.d. Ibrahim Ibrahim. ‘The impact of Arab expatriates on social and political development in the Gulf states’ in The Arab Brain Drain, edited by A. B. Zahlan, Ithaca Press, London, 1981. Ismael, Jacqueline S. Kuwait: Social change in historical perspective, Syracuse University Press, New York, 1982. — — . Kuwait: Dependency and class in a rentier state, University of Florida Press, Gainsville, 1993. Ives, E. A Voyage from England to India, in the year mdccliv, and an historical narrative of the operations of the squadron and army in India, under the command of Vice-Admiral Watson and Colonel Clive in the years 1755, 1756, 1757; including a correspondence between the Vice-Admiral and the Nabob Serajab Dowlab. Also, a journey from Persia to England in 1758 and 1759 by an unusual route, printed for Edward and Charles Dilly, London, 1773. ‘Izz Al-Din, Amin. [Ummal al-Kuwait: Min al-lu]lu] ila al-betrol (Kuwait’s labour force: from pearl fishing to petrol), Government of Kuwait Press, Kuwait, 1958. Izzard, Molly. The Gulf: Arabia’s western approaches, Murray, London, 1979. Joyce, Miriam. The Sultanate of Oman: a twentieth century history, Praegar, Westport, Conn., 1995. — — . Kuwait, 1945–1996: an Anglo-American perspective, Frank Cass, London, 1998. Kajare, Prince Firuz. Le Sultanat d’Oman, A. Pedone, Paris, 1914. Kamal, Safwat. Min ‘ adat wa-taqalid al-zawaj f i al-Kuwait (Some marriage customs and traditions in Kuwait), Ministry of Information, Kuwait Government Printing Press, Kuwait, n.d. Kaye, J. W. The Life and Correspondence of Major-General Sir John Malcolm, C.C.B., Smith, Elder, London, 1856. Kelly, J. B. Sultanate and Imamate in Oman, Oxford University Press (for the Royal Institute of International Affairs), London, 1959. — — . Britain and the Persian Gulf 1795–1880, Oxford University Press, London, 1968. — — . Arabia, the Gulf and the West, Weidenfeld and Nicolson, London, 1980. Khalaf, Fadil. Ahlam al-shabab (Dreams of the youth), Al-Mubarkiya Press, Beirut, c.1953. Khalifa, Ali Mohammed. The United Arab Emirates: Unity in fragmentation, Croom Helm, London, 1979. Khuri, Fadil. Tribe and State in Bahrain: the transformation of social and political authority in an Arab state (Publications of the Centre for Middle Eastern Studies, no. 14), University of Chicago Press, Chicago, 1980.
Select Bibliography
245
Koran, see Arberry. Kumar, R. India and the Persian Gulf Region 1858–1907: a study in British imperial history, Asia Publishing House, London, 1965. Lienhardt, P. A. ‘Village politics in Trucial Oman’ in Man, A Monthly Record of Anthropological Science (RAI), vol. 57, no. 61, 1957. — — . ‘Kuwait: a Changing Country’ in The Listener, vol. 84, no. 1904, 23 September 1965. [This article was originally given as a talk on BBC Third Programme.] — — .‘Some social aspects of the Trucial States’ in The Arabian Peninsula: Society and politics, edited by D. Hopwood, Allen and Unwin, London, 1972. — — . ‘Women behind the veil – the Arab world’ in The Arab World (vol. 17 of ‘Peoples of the Earth’), edited by Ahmed Al-Shahi, Europa Verlag, Arnoldo Mondadori Editore, Verona, 1973. — — . ‘The authority of shaykhs in the Gulf: an essay in nineteenth-century history’ in Arabian Studies II edited by R. B. Serjeant and R. L. Bidwell, C. Hurst and Co. (published for the Middle East Centre, University of Cambridge), London, 1975. — — . ‘British by birth’ in Journal of the Anthropological Society of Oxford, vol. 18, no. 3, 1987. — — . Disorientations: a society in flux. Kuwait in the early 1950’s, edited by Ahmed Al-Shahi, Ithaca Press (Middle Eastern Culture Series, vol. 19), Reading, 1993. — — . (ed. and trans.), The Medicine Man: Swifa Ya Nguvumali, by Hasani bin Ismail, Oxford Library of African Literature, Clarendon Press, Oxford, 1968. Lockhardt, L. ‘Outline of the history of Kuwait’ in Journal of the Royal Central Asian Society, vol. 34, July–October, 1947. Long, D. E. The Persian Gulf: an introduction to its people, politics and economics, Westview Press, Boulder, Colorado, 1976. Longrigg, S. H. Oil in the Middle East: Its discovery and development, Oxford University Press, London, 1954. Longva, Anh Nga. Walls Built on Sand: Migration, exclusion and society in Kuwait, Westview, Boulder, Colorado, 1997. Lorimer, J. G. Gazetteer of the Persian Gulf, 2 vols., Superintendent Government Printing, Calcutta, India, 1915. [This was the edition principally used by Lienhardt. It was subsequently reprinted in 5 vols. plus maps and tables, Gregg International Publishers Ltd., Westmead, Farnborough, Hants., England and Irish University Press, Shannon, Ireland, 1970.] Malcolm, Sir J. The History of Persia from the most early period to the present time, containing an account of the religion, government, usages and character of the inhabitants of that Kingdom, 2 vols., Murray, London, 1815. Mann, C. Abu Dhabi: Birth of an oil sheikhdom, Khayats, Beirut, 1964. Mann, M. The Trucial Oman Scouts: the story of a bedouin force, Michael Russell, Norwich, 1994. Mansfield, P. Kuwait: Vanguard of the Gulf, Hutchinson, London, 1990. Marlow, J. The Persian Gulf in the Twentieth Century, Cresset Press, London, 1962. Melikian, L. H. Jasim: a study in the psychosocial development of a young man in Qatar, Longman, London, 1981. Mertz, R. A. Education and Manpower in the Arabian Gulf, American Friends of the Middle East, Washington, DC, 1972. Miles, S. B. The Countries and Tribes of the Persian Gulf with a new introduction by J. B. Kelly, 2nd edition, 2 vols. (in one), Frank Cass, London, 1966. [Reprint of Harrison, London, 1919].
246
Select Bibliography
Netton, R. (ed.). Arabia and the Gulf: From traditional society into modern states, Croom Helm, London, 1986. Niblock, T. (ed.). Social and Economic Development in the Arab Gulf, Croom Helm, London, 1980. Palgrave, W. G. Personal Narrative of a Year’s Journey Through Central and Eastern Arabia (1862–1863), 7th edition, Macmillan Press, London – now Palgrave – 1873. Pelly, L. ‘Remarks on the tribes, trade and resources around the shore line of the Persian Gulf’ in Transactions of the Bombay Government Geographical Society, vol. 17, 1863. Pelt, M. C. Van. ‘The Sheikhdom of Kuwait’ in Middle East Journal, vol. 4, no. 1, 1950. Peterson, J. E. ‘Tribes and politics in eastern Arabia’ in Middle East Journal, vol. 31, no. 3, 1977. Pridham, B. R. (ed.). The Arab Gulf and the Arab World, Croom Helm, London, 1989. Public Record Office. Files: FO 371/157018, 1961 and FO 371/162780, 1961. Rihani, Ameen. Around the Sea Coasts of Arabia, Constable and Co., London, 1930. Royal Scottish Museum. The Evolving Culture of Kuwait, Royal Scottish Museum, HMSO, Edinburgh, 1985. Rumaihi, Muhammed Ghanim. Beyond Oil: Unity and development in the Gulf, translated from the Arabic (Khalij Laysa Naft) by James Dickens, Al-Saqi Books, London, 1986. Rush, Alan. Al-Sabah; history and genealogy of Kuwait’s ruling family 1752–1987, Ithaca press, London, 1987. Sadik, Muhammad T. and Snavely, William P. Bahrain, Qatar and the United Arab Emirates: Colonial past, present problems, and future prospects, Lexington Books, Toronto, London, 1972. Sandwick, John A. (ed.) The Gulf Cooperation Council: Moderation and stability in an independent world, Westview, Boulder, Colorado, 1987. Sanger, R. H. The Arabian Peninsula, Cornell University Press, New York, 1954. Schoff, W. H. (trans.) The Periplus of the Erythraean Sea: Travel and trade in the Indian Ocean, Longman, Green, New York, 1912. Shea, R. O. The Sand Kings of Oman, being the experiences of an R.A.F. officer in the little known regions of Trucial Oman, Arabia, Methuen, London, 1947. Shiber, Saba G. The Kuwait Urbanization, Kuwait Government Printing Press, Kuwait, 1964. Skeet, I. Muscat and Oman: the end of an era, The Travel Book Club, London, 1975. — — . Oman: Politics and development, Macmillan Press, Basingstoke – now Palgrave – 1992. Soffan, Linda. The Women of the United Arab Emirates, Croom Helm, London, 1980. State of Kuwait. Ministry of Guidance and Information. Kuwait Today: a welfare state (written by Quality Publications Ltd.), Quality Publications Ltd., Nairobi, n.d. — — . Ministry of Information (International Media Information Publication Administration). Kuwait: Facts and figures, 5th edition, Al-Assriya Printing Press, Kuwait, n.d. — — . (On the Occasion of the 50th Anniversary of the United Nations). For History, 3 vols., Amiri Diwan edition, Kuwait, 1995. Vol. 1: Kuwait and Democracy, Human Rights, Social and Economic Development. Vol. 2: Kuwait and Cultural Aspects. Vol. 3: Kuwait and the Iraqi Aggression.
Select Bibliography
247
Stocqueler, J. H. Fifteen months’ pilgrimage through untrodden tracts of Khuzistan and Persia in a journey from India to England, through parts of Turkish Arabia, Persia, Armenia, Russia and Germany, performed in the years 1831–32, Saunders and Otley, London, 1832. Sweet, Louise E. ‘Pirates or Polities? Arab societies of the Persian or Arabian Gulf, 18th century’ in Ethnohistory, vol. 11, 1964. Sykes, Sir P. A History of Persia, Macmillan and Co. – now Palgrave (2nd edition), London, 1921. Taki, Ali Hassan. The Changing Status of the Bedouin Woman, Oriental Press, Bahrain, 1974. Taryam, R. O. The Establishment of the United Arab Emirates, 1980–85, Croom Helm, London and New York, 1987. Thesiger, Wilfred. ‘Travel on the Trucial Coast’ in The Geographical Magazine, 22 (3), July, 1949. — — . ‘Desert borderlands of Oman’ in The Geographical Journal, vol. 116, nos. 4–6, October–December, 1950. Thomas, B. ‘Among some unknown tribes of South Arabia’ in Journal of the Royal Anthropological Institute, vol. 59, pts 1–2, 1929. Trollope, A. An Autobiography (edited by his son, Henry Merivale Trollope), Blackwood and Sons, Edinburgh, 1883. Twitchell, K. S. (with the collaboration of Edward J. Jurji). Saudi Arabia: With an account of the development of its natural resources, Princeton University Press, Princeton, 1947. Villiers, A. J. Sons of Sindbad: an account of sailing with the Arabs in their dhows, in the Red Sea, round the coasts of Arabia, and to Zanzibar and Tanganyika; pearling in the Persian Gulf and the life of the shipmasters and the mariners of Kuwait, Hodder and Stoughton, London, 1940. — — . ‘Dhow-builders of Kuwait’ in Geographical Magazine, 20 (9), 1948. — — . ‘Some aspects of the Arab dhow trade’ in Middle East Journal, vol. 2, no. 4, 1948. Vine, Peter and Casey, Paula. Arab Gold: Heritage of the UAE, Immel Publishing, London, 1989. — — . United Arab Emirates: Profiles of a country’s heritage and modern development, Immel Publishing, London, 1992. Welldon, J. E. C. (trans. with an analysis and critical notes). The Politics of Aristotle, Macmillan Press, London – now Palgrave – 1883. Wellsted, J. R. Travels in Arabia, Murray, London, 1838. — — . Travels to the city of the Caliphs, along the shores of the Persian Gulf and the Mediterranean. Including a voyage to the coast of Arabia, and a tour on the Island of Socotra, H. Colburn, London, 1840. Wilkinson, J. C. The Organization of the Falaj Irrigation System in Oman, University of Oxford, School of Geography (Research Papers, no. 10), Oxford, 1971. — — . Problems of Oasis Development, University of Oxford, School of Geography (Research Papers, no. 20), Oxford, 1978. Wilson, Sir A. T. The Persian Gulf: an historical sketch from the earliest times to the beginning of the twentieth century, Clarendon Press, Oxford, 1928. Winstone, H. V. E. and Freeth, Zahra. Kuwait: Prospect and reality, Allen and Unwin, London, 1972. Wright, D. ‘The changed balance of power in the Persian Gulf’ in Asian Affairs, vol. 60, no. 3, 1973.
248
Select Bibliography
Young, T. Cuyler (ed.). The Middle East Focus: the Persian Gulf (Proceedings of the Twentieth Annual Near East Conference), Princeton University Press, Princeton, New Jersey, 1969. Zahlan, Rosemarie Said. The Origins of the United Arab Emirates: a political and social history of the Trucial states, Macmillan Press, London – now Palgrave – 1978. — — . The Making of the Modern Gulf States: Kuwait, Bahrain, Qatar, The United Arab Emirates and Oman, Unwin Hyman, London, 1989. [A revised edition was published by Ithaca Press, Reading, 1998.]
Index In compiling the Index I have been selective in the number and citation of entries. References to topics such as women, men, desert, bedouin, authority, shaikhs, tribe, politics and Trucial Coast are so frequent in the text that they have not been listed. Only certain Arabic words are given, in italic and with diacritical marks. Groups of many successive page references are given as spans. References that include illustrative material are indicated in bold. [Ed.]
[abara, 171 [Abdullah Al-Salim Al Sabah, Shaikh, xiv, 214, 220, 239 Abu Dhabi, ix–x, xvi, xvii, 9–11, 14, 29, 85, 88–9, 90, 95, 114–22, 124, 130, 167, 171, 174–82, 186, 215, 216, 228, 229, 237 Aden (and Adenese), 127, 151 affinity (and affines), 34, 71, 72, 76, 77, 78, 173, 190, 191, 236 Afghanistan, 234 Africa, xix, 8, 28, 125, 127, 138 [agal, 195 Agnus Dei, xvi agriculture, 19, 25, 80 ahl al-bait, 48 ahl al-watan, 181 Ahmed Al-Jabir Al Sabah, Shaikh, 167, 214, 219, 220 Ahmed bin Dalmuch, 175, 178 Ahmed bin Dhiyab bin Khalifa, Shaikh, 176 Ahmed bin Khalifa, Shaikh, 178, 180 [A]isha bint Muhammed, Shaikha, 172 [A]isha bint Shakhbut, Shaikha, 178 Ajman, xviii, 231 Alexander the Great, 194 Ali ibn Abi Talib, 190 alimony, 55 America, 10, 15, 229 amir, 185, 206 [Amir bin Sa[asa[a, 86, 91 anchorage, 31, 114, 116–17, 139 arbitration (and arbitrators), 6, 8, 13, 204–6, 234
Aristotle, 214 asl wa-]l fasl, al-, 42, 78 [Awamir (tribe), 85, 86, 91, 97, 98, 106, 121 Baggara Arabs (tribe), 233 Bahrain, xi, xii, xvi, xviii, 10, 15, 28, 33, 66, 115, 155, 166, 172, 188, 195, 197, 205, 211, 214, 218, 222, 223 Baluchis, 40, 124, 234 Bani Ghafala (tribe), 215 Bani Hamud (tribe and residential quarter), 121 Bani Qitab (tribe), 12, 14, 90 Bani Yas (tribe), 85–6, 87, 89, 90, 119, 171, 180, 181, 215, 231 barasti, 118 barrenness land, 80, 137 women, 58, 60, 138 Basrah, 8, 28, 125, 137, 138, 195 Bassadore, 231, 232 betrothal ceremony, 69 bi-]l-Khail, Al (a tribal section), 94, 95 bi-]Sh-Sha[r, Al (a tribal section), 94, 181 Bitter Apple, 80, 236 blood compensation (diya), 5, 103–4, 107, 108, 109, 234, 236 boat-owners, 16, 28–31, 131–6, 139, 140–3, 145, 146, 149, 155, 159, 163, 217, 218, 225 Bombay (and Government), 2, 5, 20, 139 bridewealth, 55, 56, 63, 71, 148, 169 249
250
Index
Britain Empire, 218 Foreign Office, xi, 12 Foreign Service, 14 Government, x, 2–9, 11–15, 31, 124, 172, 214, 220, 221, 229, 232, 234, 237 officers, 12 Political Agents and Residents, x, xi, 6, 7, 8, 13, 14, 123, 125 British Bank of the Middle East, 123 British Petroleum Company, 119 Bu Falah, Al (a tribal section), 90, 174, 176, 177, 178, 239 Bu Falasa, Al (a tribal section), 171, 175, 178, 180, 215, 216, 218, 221, 231 Bu Halaiba, Al (a tribal section), 181 Bu Hamir, Al (a tribal section), 94 Bukha, 171 Bu Mindhir, Al (a tribal section), 94 Bu Muhair, Al (a tribal section), 119, 215, 216 Bu Rahama, Al (a tribal section), 94, 95, 181 Buraimi (oasis), 14, 25, 80, 95, 151, 174, 180, 181, 182, 200 disputes, 10, 221 Bu Shamis, Al (a tribal section), 85 Bushire, 232 Buti bin Buti, Shaikh, 216 Cambridge University, viii, xv camels, 25, 26, 38, 39, 76, 81, 86, 92, 97, 114, 115, 120, 168, 187, 203 brands, 90–1 Canute, King, 86 chastity, 70, 108, 195 Chiyabil (a tribal section), 85–6 Christianity (and Christians), xv, 25, 45, 50, 76 cinemas, 10, 123, 204 civil wars, 9, 10, 124, 171, 174, 175, 219 clientage, 16, 31, 225, 227 concubinage (and concubines), 34, 50, 173
contracts, 6, 27, 48, 69, 83, 111, 129, 156, 161, 205, 230, 234, 238, 239 coup d’état, 9, 13, 14, 215, 234 cousinship (and cousins), 23, 26, 41, 42, 84, 90, 103 credit (and creditors), 6, 134, 136, 144, 158, 159, 160, 161, 162, 164, 217 Cunnison, I., 233 Dairah (a residential quarter), 115, 172 Darwish Brothers, 206 date gardens, 56, 81, 121, 137, 143, 168, 169, 182, 203, 204, 211, 223 debts, xv, 6, 29, 30, 133–6, 138, 142–4, 152, 156–61, 164, 168, 205, 206, 217, 225 Dhiyab bin Hilal bin Dhiyab, Shaikh, 176 Dhiyab bin Hilal bin Shakhbut, Shaikh, 176 Dhiyab bin [Isa, Shaikh, 176, 177 Dhiyab bin Khalifa bin Shakhbut, Shaikh, 176, 178 Dhiyab bin Saqr bin Zayid, Shaikh, 176, 178, 186 dhows, xv, 31, 116, 117, 118, 121, 122, 125, 127, 134, 138, 147, 151, 152 Dhu ]l-Qarnain, 194, 204 Dickson, H. R. P., 220, 240 divorce, 19, 21, 43, 45, 50, 54–8, 60–2, 70, 71, 72, 73, 169, 170, 173, 188, 189, 190 Dubai, xvi, xvii, 8, 9, 10, 12, 13, 14, 15, 20, 31, 32, 62, 64, 70, 88, 89, 115, 116, 117, 118, 120–6, 138, 141, 143, 144, 152, 153, 165, 171, 172, 174, 175, 178, 179, 180, 182, 211, 215–19, 221, 223, 231, 235 Durkheim, E., 233 education, xiv, xvii, 9, 12, 30, 38, 43, 44, 64, 124, 154, 229, 234, 239, 240 Egypt, 10, 137
Index Empty Quarter, 25, 91, 97, 121 Euphrates, 138 Europe, xii, 3, 10, 15, 123, 126, 140, 144, 160, 199, 234 Failakah Island, xiv, xv, xvi, 146, 147, 148 Fakhira bint Dhiyab bint Khalifa, Shaikha, 178 falaj, 80, 236 Fao, 168 fara]id, 46, 183, 239 farij, 82, 119, 121 female infanticide, 60, 235 feminism, 64 feud, 75, 76, 78, 88, 91, 92, 102–13 see also warfare fishing (and fishermen), xiv, 29, 67, 86, 117, 120, 121, 122, 125–37, 140, 147, 155, 201, 202 salt fish, 121, 125, 129, 134, 137, 141, 142 traps, 128 French Government, 2 Fujairah, xviii General Treaty of Peace, 2, 215 generosity, 41, 51, 76, 81, 107, 135, 180, 200 Ghanim, Al (family), 227 gold, 25, 122, 142 gossip, 41, 67, 70, 199 Gout bint Shakhbut bin Sultan, Shaikha, 178 guards, 28, 71, 77, 112, 116, 120, 154, 205, 219 see also men-at-arms Gulf Cooperation Council (GCC), xviii Gumzan (a tribal section), 119 Ha]il, 176 hakim, 185 Hamdan bin Hamdan bin Zayid, Shaikh, 174, 176, 178, 179 Hamdan bin Muhammed bin Khalifa, Shaikh, 176, 177, 178 Hamdan bin Rashid, Shaikh, 180 Hamdan bin Sultan bin Hilal, Shaikh, 176
251
Hamdan bin Tahnun bin Shakhbut, Shaikh, 176 Hamdan bin Zayid bin Khalifa, Shaikh, 174, 175, 176, 177, 178, 179, 182 Hamid bin Buti, Shaikh, 180, 182 Hanbali, 47 haram, 51 harim (hareem), 34, 51 hurma, 51, 235 hurmat al-saliha, al, 49, 63 Harib bin Shakhbut bin Dhiyab, Shaikh, 176 Hashar bin Maktum, Shaikh, 216, 221 Hawamil (a tribal section), 85, 86 Hazza[ bin Sultan bin Zayid, Shaikh, 174, 176, 177, 178 Hazza[ bin Za[al, Shaikh, 215 Hazza[ bin Zayid bin Khalifa, Shaikh, 176, 178 Hennell, Captain S., 231 Hilal Al-Mutairi, 37 Hilal bin Dhiyab bin Hilal, Shaikh, 176 Hilal bin Sa[id bin Sultan, Shaikh, 176 Hilal bin Shakhbut bin Dhiyab, Shaikh, 176 homicide, 27, 88, 102–13, 209, 228 honour (sharaf ), 35, 40, 42, 43, 44, 60, 66, 96, 105, 108, 112, 150, 164, 184, 185, 196, 201, 206, 217, 218 Humaid bin Muhammed, Shaikh, 174 Hussa bint Al-Murr, Shaikha, 62, 63, 64, 171, 172, 235 Hussa bint Hamdan bin Hamdan, Shaikha, 178 Hussa bint Muhammed bin Khalifa, Shaikha, 175, 178 Hussa bint Rashid, Shaikha, 178 Ibrahim, Al (family), 227 [idda, 235 illegitimacy, 40 Imarat Al-Mutasaliha, Al-, 232 immigrants (and foreigners), xiii, xiv, xvii, 9, 31, 40, 83, 124, 144, 145, 147, 151, 152, 161, 226, 229
252
Index
India (and Indians), 2, 3, 10, 25, 28, 31, 42, 119, 122, 123, 137, 138, 139, 142, 169, 217, 218, 234 Indian Ocean, 1, 2, 8, 137 inheritance, 45–6 Iraq, 28, 84, 137, 220, 229, 240 irrigation, 80, 211 [Isa Al Khalifa, Shaikh, 172 [Isa bin Khalid, Shaikh, 177 Isfahan, xix Jabal [Ali, 14 Jabir Al-Mubarak Al Sabah, Shaikh, 213, 220 Ja[fur bin Shakhbut bin Dhiyab, Shaikh, 176 Jahrah (oasis), xiv ja[ida, 235 Jalahima (tribe), 216, 217 jama[a, 31, 148, 163, 226 jariya, 173 Jawasim (tribe), 2, 165, 222, 231 jewellery (and ornaments), 39, 48, 49, 53, 55, 169, 170, 200 Jordan, 229 Julfar, 240 see also Ras Al-Khaimah Jum[a bin Maktum, Shaikh, 15, 179, 221 Karun river, 138 Keir, General Sir William Grant, 5 Kelly, J. B., 234 Kenya, 137 Khalaf, Fadil, 239 Khalid bin [Isa bin Khalid, Shaikh, 177 Khalid bin Sultan bin Zayid, Shaikh, 176, 177, 178, 180, 182 Khalifa, Al (family), 216, 217 Khalifa bin Muhammed bin Khalifa, Shaikh, 176, 177, 178 Khalifa bin Sa[id, Shaikh, 174, 221 Khalifa bin Shakhbut bin Dhiyab, Shaikh, 176, 177 Khalifa bin Zayid bin Khalifa, Shaikh, 176, 178, 181 Khalifa bin Zayid bin Sultan, Shaikh, 176, 178 khaluwi, 158, 159
Kharg Island, 141 Khasab, 146 khusaif i, 201–3 kingship, 200, 206 kinship, 21, 22, 26, 34, 41, 46, 58, 67, 75, 76, 77, 78, 94, 103, 110, 111, 224 agnation (and agnates), 20, 21, 23, 27, 42, 43, 58, 75, 76, 78, 84, 93, 94, 96, 103, 108, 183, 189, 190, 191, 192, 223, 236 descent, 26, 27, 41–2, 58, 73, 75, 79, 82–7, 91, 92, 94, 95, 96, 103, 109, 111, 149, 171, 188, 189, 210, 213 kins, xv, 20, 21, 22, 23, 34, 39, 40, 41, 53, 60, 69, 71–9, 84, 85, 86, 88, 93, 95, 96, 97, 103–10, 113, 129, 132, 143, 173, 175, 190, 191, 192, 196, 197, 238 Koran, xvi, 38, 43, 45, 46, 48, 49, 53, 60, 70, 120, 145, 148, 172, 194, 238 Kuwait, ix, xi, xii, xiii, xiv, xvi, xvii, xviii, 9, 10, 15, 16, 17, 28, 37, 38, 115, 117, 123, 124, 125, 131, 136, 137, 138, 141, 142, 143, 144, 146, 147, 159, 160, 166, 167, 168, 171, 173, 185, 186, 195, 197, 198, 205, 213, 216–20, 222, 223, 227, 228, 237, 238, 240 Oil Company, 167 Lamu Island, xix Latifa bint Hamdan bin Zayid, Shaikha, 174, 178, 179 Latifa bint Sa[id bin Tahnun, Shaikha, 178 lawm wa-luzum, 109 Lienhardt, R. G., vii Liwa (oasis), 29, 95, 151 loans, 6, 28, 29, 30, 31, 133–6, 145, 146, 150, 154, 156, 157, 159, 161, 170, 182, 217, 218, 225 London, 10, 122 Lorimer, J. G., 175, 176, 178, 215, 216, 234, 237, 239 luh, 201–2
Index mahkama, 218 Maitha bint Muhammed bin Khalifa, Shaikha, 178 Maitha bint Salimin, Shaikha, 175, 178 majlis, x, 52, 55, 62, 147, 148, 171, 172, 185, 196, 197, 198, 199, 218, 219, 220, 235, 240 Makhazima, Al (a tribal section), 95 Maktum, Al (family), 179 Maktum bin Buti, Shaikh, 215 Maktum bin Hashar, Shaikh, 216 Maktum bin Jum[a, Shaikh, 174 Malabar, 125 Malcolm, Sir John, 2, 19 Malna bint Khalid bin Sultan, Shaikha, 178 Manama, 222 Manani (a tribal section), 121 Manasir (tribe), 85, 88, 89, 90, 94, 95, 106, 175, 178, 181 Mani[ bin Rashid, Shaikh, 153, 174, 218, 219, 221 Marazig (a tribal section), 121 maritime activities, 1, 2, 3, 25, 29, 31, 114–64 markets, 25, 28, 52, 53, 54, 119, 123 marriage, 19, 20, 21, 22, 34, 43, 44, 49, 50, 51, 54–8, 60, 62, 67, 69–73, 77, 78, 93, 169, 170, 171, 173, 182, 183, 187–92, 201, 204 intermarriage, 22, 42, 87, 93, 94, 97, 109 preferred, 21, 39, 71, 72, 94, 183 re-marriage, 45, 57, 58–9, 61, 62 masculinity, 38, 70 Mauritius, 2 Mazari[ (a tribal section), 87, 89 Mecca, 47 men-at-arms, 171, 196, 197, 198, 200, 201, 203, 205, 206, 207, 210, 221, 223 see also guards merchants, 63, 69, 122, 123, 133, 134, 135, 136, 140, 159, 163, 164, 168, 198, 199, 200, 201, 203, 204, 205, 207, 214, 220, 227 Miryam bint Hamdan bin Zayid, Shaikha, 174, 178
253
Miryam bint Muhammed bin Saif, 175, 178 Miryam bint Sa[id bin Zayid, Shaikha, 178 Miryam bint Sultan bin Zayid, Shaikha, 175, 178 modesty, 33, 34, 39, 47, 49, 52, 54, 57, 62, 67, 69, 195, 197 moneylending, 133, 152 monogamy, 50 mosque, 33, 47, 48, 52, 55, 120, 138, 194, 196 Moza bint Ahmed bin Khalifa, Shaikha, 178, 181 Moza bint Hamdan bin Muhammed, Shaikha, 178 Moza bint Muhammed, Shaikha, 178 Moza bint Sa[id, Shaikha, 172 mu[allim, 139 Mubarak Al-Salim Al Sabah, Shaikh, 37, 167, 168, 169, 185, 201, 213, 227 Mubarak bin Muhammed bin Khalifa, Shaikh, 176, 177, 178 mudda, 236 Mudiya bint Sultan bin Mujrin, Shaikha, 179 Muhammed, Al (a tribal section), 239 Muhammed bin Dhiyab bin Khalifa, Shaikh, 176, 178 Muhammed bin Khalid bin Sultan, Shaikh, 177, 178 Muhammed bin Khalifa bin Zayid, Shaikh, 175, 177, 178, 181, 182, 183 Muhammed bin Salim bin Sultan, Shaikh, 165, 184 Muhammed bin Shakhbut bin Dhiyab, Shaikh, 176, 177 Muhammed bin Zayid bin Khalifa, Shaikh, 176 Muharraq, 222 muhtasham, see respect money Mujrin bin Sultan bin Mujrin, Shaikh, 179 Murr bin Huraiz, Shaikh Al-, 171 musa[ada, 135 Musandam Peninsula, 146 Muscat, 2, 28, 137, 139, 146, 215, 234 Mutairi (tribe), 37
254
Index
Nahyan, Al (a tribal section), 239 Napoleon, 2 navigation, 116, 139, 140 oil, ix, xiii, xvi, xix, 7, 9, 10, 11, 12, 14, 32, 117, 118, 123, 160, 167, 203, 220, 225, 228, 229 companies, 13, 14, 15 Oman, xviii, 1, 2, 80, 217, 231, 232, 234 Ommana, 117 Pakistan, 119, 123 Palestinians, xiv, 229 palm houses, 118, 119, 121 patronage, 16, 31, 145, 146, 198, 225, 227 pearls, 25, 120, 132, 134, 137, 157, 158, 217 boats, 6, 117, 120 cultured, 15, 153 diving (and divers), 67, 150–2, 154–5 fishing (and fishers), xvii, 1, 4, 5, 6, 7, 15, 16, 25, 29, 31, 117, 118, 131, 133, 136, 137, 146, 147, 149, 150–62, 163, 217, 225, 231, 234 industry, 28, 30, 64, 125, 167, 217 merchants, 16, 29, 30, 31, 118, 146, 149, 150, 153, 154, 157, 158, 160, 163, 164, 169, 175, 181, 182, 217, 218 trade, 8, 150, 158, 162, 163, 217 peasants, 19, 81, 114, 160, 223 Pemba Island, xix Persia (and Persians), xii, xix, 4, 29, 31, 42, 80, 81, 114, 116, 119, 120, 121, 122, 123, 127, 137, 138, 141, 146, 151, 167, 169, 172, 198, 206, 217, 218, 220, 229, 234, 240 Fars (Province), 85 Perso-Arabs, 119 residential quarter, 121 pilgrimage (and pilgrims), 27, 47, 83 piracy, 2, 4, 5, 234 Pirate Coast, 2 polygyny, 19, 20, 21, 22, 23, 50, 56–8, 187
poverty, 5, 12, 62, 64, 65, 66, 75, 78, 98, 117, 118, 195 prestige, 37, 43, 44, 145, 146, 163, 164, 210, 218 see also status primogeniture, 184, 185, 199, 213 profit (and profit sharing), 16, 29, 30, 77, 117, 128–34, 136, 140–3, 146, 151, 152, 153, 155–6, 158, 160, 161, 169, 170, 206, 225 Prophet Al-Yas, 85 Prophet Muhammed, xv, 44, 45, 76, 86, 173, 235, 239 Prophet Solomon, 152, 153 prostitution (and prostitutes), 40, 42, 53, 64, 71, 200 qanat, 80 Qatar, xii, xviii, 14, 114, 146, 147, 166, 167, 182, 206 Qina[i, Yusuf bin [Isa Al-, 238 Qishm Island, 141 Qubaisat (a tribal section), 119, 178, 180 raiding (and raiders), 26, 27, 76, 87, 97–8, 187, 203, 239 Ramadan, 47, 48, 86, 115, 126, 127 Rams, 116, 121, 141, 143, 144 Ras Al-Khaimah, xviii, 4, 5, 55, 56, 64, 115, 117, 121, 125, 129, 130, 141, 143, 165, 167, 172, 174, 178, 184, 222, 237, 240 Rashid, Al (family), 176 Rashid bin Buti, Shaikh, 218 Rashid bin Hamid, Shaikh, 231 Rashid bin Maktum, Shaikh, 216 Rashid bin Sa[id bin Maktum, Shaikh, 64, 174, 180, 182, 219, 221, 237 Rashid bin Saqr bin Zayid, Shaikh, 176, 178 Rawashid, Al (a tribal section), 119 rayyal, 51, 235 refuge, 6, 106, 107, 109 respectability, 34, 63, 64, 66, 67, 140, 196, 197 respect money, 109, 110, 111, 113, 209 right over people, 11 right over territory, 11, 12
Index rivalry (between shaikhs), 95, 214, 220, 222 Roda bin Sultan bin Mujrin, Shaikh, 179 Royal Air Force (RAF), 123 ruling families, ix, 15, 17, 19, 22, 81, 164, 165, 166, 168, 169, 170, 171, 173, 174, 175, 176, 183, 185, 186, 187, 188, 191, 192, 196, 199, 200, 203, 204, 206, 208, 210–16, 218, 219, 221, 222–30, 234 Rumaithat (a tribal section), 119 sabaha, al-, 169 Sabah, Al (family), 216, 220 sabbab, 67 sadaqa, 93 Sa[dun, Al (a tribal section), 239 Sa[dun bin Sultan bin Hilal, Shaikh, 176 Sa[id bin Buti, Shaikh, 215, 216 Sa[id bin Maktum, Shaikh, 15, 64, 171, 174, 216, 217, 218, 219, 221, 235 Sa[id bin Rashid, Shaikh, 215 Sa[id bin Shakhbut bin Dhiyab, Shaikh, 176 Sa[id bin Shakhbut bin Sultan, Shaikh, 176, 177, 178, 182 Sa[id bin Sultan bin Hilal, Shaikh, 176 Sa[id bin Tahnun bin Shakhbut, Shaikh, 176, 177 Sa[id bin Zayid bin Khalifa, Shaikh, 176, 178 Salama bint Buti, Shaikha, 178, 180 Saliha bint Ahmed bin Dalmuch, 175, 178 Salim Al-Mubarak Al Sabah, Shaikh, 213, 220 Salimin, Al (a tribal section), 95 Salt marshes, 115 San[a bint Roda bin Sultan, Shaikha, 178, 179 Sanger, R. H., 240 Saqr bin Muhammed bin Salim, Shaikh, 165, 174, 184, 237 Saqr bin Tahnun bin Khalifa, Shaikh, 176
255
Saqr bin Zayid bin Khalifa, Shaikh, 175, 176, 177, 178, 180, 181 Saudi Arabia, xviii, 12, 14, 64, 185, 186, 200, 221, 228, 239 Sawalim, (a tribal section), 121 scandal, 35, 41, 43, 63, 88 Schoff, W. H., 237 seafaring (and seafarers), 19, 25, 67, 85, 116, 117, 120, 122, 129, 136–7, 153, 205, 225 seclusion (of women), 21, 33, 34, 35, 37, 38, 39, 41, 52, 59, 62, 67, 71, 191, 195, 199, 224, 225, 238 segmentation kinship, 83, 85, 87, 94, 95, 98–102, 105, 106, 107, 108 political, 19, 83, 84, 96, 98–102, 106, 107, 108, 113, 208, 209, 226, 227 Shaikha bint Muhammed bin Salimin, Shaikha, 175, 178 Shaikha bint Sa[id, Shaikha, 172, 174 Shaikha bint Saqr bin Zayid, 175, 178 Shakhbut bin Dhiyab, Shaikh, 176, 178, 231 Shakhbut bin Sultan bin Zayid, Shaikh, ix, x, xi, xix, 176, 177, 178, 181, 182, 228, 237, 238, 239 Sha[m, 115 shame, 37, 40, 42, 43, 63, 66, 67, 107, 109, 112, 127, 163, 179, 199, 238, 239 Shamsa bint Ahmed bin Khalifa, Shaikha, 178, 182 Shamsa bint Sultan bin Mujrin, Shaikha, 178, 179 Shandaqah (a residential quarter), 115 shar i[a channels, 80 religious laws, 6, 45–6, 47, 63, 83, 136, 144, 145, 167, 169, 172, 182, 183, 205, 207, 235, 236 Sharjah, xvii, 12, 13, 31, 63, 90, 116, 117, 120, 122, 123, 165, 177, 181, 185, 215, 216, 222 Shatt Al-Arab, 115, 137, 168 Shi[a, xvii, 31 Shihuh, 116, 171 shufa, 200
256
Index
sitting-room (diwaniya), 33, 234 slavery (and slaves), 8, 12, 34, 50, 60, 63, 64, 73, 123, 138, 162, 173, 181, 196, 200, 203, 217 emancipation, 8, 64, 123 smuggling, 62, 122, 142, 172, 200, 204 solidarity, 23, 41, 97, 98, 102, 105, 108, 109, 113, 187, 205, 214 Somalia (and Somalis), 127, 151 sorcery, 69 status, 26, 27, 37, 38, 42, 44, 52, 63, 72, 73, 75, 78, 79, 81, 82, 83, 87, 99, 100, 111, 113, 120, 149, 171, 190, 227, 230, 234 see also prestige Stocqueler, J. H., 115, 139, 140 Strait of Hormuz, 171 strangers, 27, 35, 40, 41, 48, 65, 69, 84, 113, 191, 208, 209, 210, 224, 225 structure, 16–18, 23, 24, 99 subay, 51 Sudan (tribe), 119, 121, 178, 182 Sultan, Al (a tribal section), 239 Sultan bin Dhiyab bin Khalifa, Shaikh, 176 Sultan bin Hilal bin Shaikhbut, Shaikh, 176 Sultan bin Salim bin Sultan bin Saqr, Shaikh, 165, 174, 184 Sultan bin Saqr, Shaikh, 125, 165, 231 Sultan bin Shakhbut bin Dhiyab, Shaikh, 176 Sultan bin Shakhbut bin Sultan, Shaikh, xi, 176, 177, 178 Sultan bin Zayid bin Khalifa, Shaikh, 174, 175, 176, 177, 178, 180 Sultan bin Zayid bin Sultan, Shaikh, 178 sunna, 49 Sur, 137 Swahili (language), xix Syria, ix Tahnun bin Mubarak bin Muhammed, Shaikh, 178 Tahnun bin Muhammed bin Khalifa, Shaikh, 176, 177, 178
Tahnun bin Shakhbut bin Dhiyab, Shaikh, 176, 177 Tahnun bin Zayid bin Khalifa, Shaikh, 176, 177, 178 talaba, al-, 169 Tanaig (a residential quarter), 121 Tanzania, 137 Taraiarifa, Al (a tribal section), 95 Taraif, Al Al- (a tribal section), 95 tastahi, 34 Thani bin Jum[a bin Maktum, Shaikh, 179 Thesiger, W., 239 Tigris, 138 towns (and townsmen), 1, 4, 7, 12, 16, 25, 27, 28, 31, 35–41, 45, 52, 61, 67, 75, 76, 77, 80, 81, 82, 83, 114–27, 144, 148, 149, 151, 158, 160, 163, 194, 201–12, 215, 221–7, 230 trade, xix, 1, 2, 4, 5, 6, 8, 25, 28, 29, 31, 62, 63, 67, 77, 83, 116–20, 122, 123, 125, 131, 132, 133, 137, 138, 140, 146, 155, 204, 205, 217, 230 Treaty of Maritime Peace, 5 tribal loyalty (shaff ), 82–4, 87, 88, 90, 91, 93, 111, 181, 186, 208, 210, 216, 226, 236 tribesmen, 20, 27, 40, 41, 82, 85, 87–93, 99, 102, 103, 119, 195, 210, 223, 224, 225, 227 Trollope, A., 237 Trucial Council, xvii, 7, 14 Trucial Oman Levies, 12, 230 Trucial Oman Scouts, 12, 13, 14, 15, 234 Tu[aib, Al Al- (a tribal section), 95 Turkish army, 17 [Ubaid bin Sa[id, Shaikh, 215, 231 Umm Al-Qaiwain, xvii, 216 Umm Sa[id, 14 United Arab Emirates (UAE), xvii, xviii United Arab Union, xvii usury, 167, 168 [Utaiba, Al (a tribal section), 178, 181, 182
Index veiling (and veils), 34–5, 38, 48, 53, 58, 63, 137 villages, 1, 4, 15, 25, 35, 40, 41, 42, 61, 67, 76, 77, 81, 118, 119, 121, 126, 127, 144, 149, 151, 206, 224 Villiers, A., 237, 240 virginity, 69, 70 Wabran, Al (a tribal section), 95 Wahhabis, 2, 3, 11, 203 wajh, 111–13 walad, see subay warfare, 4, 5, 8, 9, 27, 75, 77, 87, 88–9, 93, 94, 97, 108–10, 112, 117, 203, 208, 210, 222 see also feud wells, 26, 39, 40, 41, 92, 93, 97, 114, 211
257
widowers, 51, 59 widows, 51, 59, 60, 61 Wudima bint Sa[id bin Shakhbut, Shaikha, 178 Yas bin [Amir, Al-, 86 Yemen, 91, 230 zabin, 105 zakat, 168, 196 Zanzibar Island, xix, 138 Zayid, Al (a tribal section), 179 Zayid bin Khalifa bin Shakhbut, Shaikh, 85, 176, 177, 178 Zayid bin Saqr bin Zayid, Shaikh, 176, 178 Zayid bin Sultan bin Zayid, Shaikh, xi, 175, 176, 177, 178, 180, 182, 228